train.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Collecting fastparquet\n",
      "  Downloading fastparquet-2024.11.0-cp39-cp39-macosx_11_0_arm64.whl.metadata (4.2 kB)\n",
      "Requirement already satisfied: pyarrow in /Users/theerat/Documents/sign-language-translator/.venv/lib/python3.9/site-packages (18.0.0)\n",
      "Requirement already satisfied: pandas>=1.5.0 in /Users/theerat/Documents/sign-language-translator/.venv/lib/python3.9/site-packages (from fastparquet) (2.2.3)\n",
      "Requirement already satisfied: numpy in /Users/theerat/Documents/sign-language-translator/.venv/lib/python3.9/site-packages (from fastparquet) (2.0.2)\n",
      "Collecting cramjam>=2.3 (from fastparquet)\n",
      "  Using cached cramjam-2.9.0-cp39-cp39-macosx_11_0_arm64.whl.metadata (4.9 kB)\n",
      "Requirement already satisfied: fsspec in /Users/theerat/Documents/sign-language-translator/.venv/lib/python3.9/site-packages (from fastparquet) (2024.3.1)\n",
      "Requirement already satisfied: packaging in /Users/theerat/Documents/sign-language-translator/.venv/lib/python3.9/site-packages (from fastparquet) (24.1)\n",
      "Requirement already satisfied: python-dateutil>=2.8.2 in /Users/theerat/Documents/sign-language-translator/.venv/lib/python3.9/site-packages (from pandas>=1.5.0->fastparquet) (2.9.0.post0)\n",
      "Requirement already satisfied: pytz>=2020.1 in /Users/theerat/Documents/sign-language-translator/.venv/lib/python3.9/site-packages (from pandas>=1.5.0->fastparquet) (2024.2)\n",
      "Requirement already satisfied: tzdata>=2022.7 in /Users/theerat/Documents/sign-language-translator/.venv/lib/python3.9/site-packages (from pandas>=1.5.0->fastparquet) (2024.2)\n",
      "Requirement already satisfied: six>=1.5 in /Users/theerat/Documents/sign-language-translator/.venv/lib/python3.9/site-packages (from python-dateutil>=2.8.2->pandas>=1.5.0->fastparquet) (1.16.0)\n",
      "Downloading fastparquet-2024.11.0-cp39-cp39-macosx_11_0_arm64.whl (684 kB)\n",
      "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m684.7/684.7 kB\u001b[0m \u001b[31m7.1 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
      "\u001b[?25hUsing cached cramjam-2.9.0-cp39-cp39-macosx_11_0_arm64.whl (1.9 MB)\n",
      "Installing collected packages: cramjam, fastparquet\n",
      "Successfully installed cramjam-2.9.0 fastparquet-2024.11.0\n",
      "Note: you may need to restart the kernel to use updated packages.\n",
      "Collecting tensorflow-addons\n",
      "  Using cached tensorflow_addons-0.23.0-cp39-cp39-macosx_11_0_arm64.whl.metadata (1.8 kB)\n",
      "Collecting seaborn\n",
      "  Using cached seaborn-0.13.2-py3-none-any.whl.metadata (5.4 kB)\n",
      "Requirement already satisfied: fastparquet in /Users/theerat/Documents/sign-language-translator/.venv/lib/python3.9/site-packages (2024.11.0)\n",
      "Collecting typeguard<3.0.0,>=2.7 (from tensorflow-addons)\n",
      "  Using cached typeguard-2.13.3-py3-none-any.whl.metadata (3.6 kB)\n",
      "Requirement already satisfied: packaging in /Users/theerat/Documents/sign-language-translator/.venv/lib/python3.9/site-packages (from tensorflow-addons) (24.1)\n",
      "Requirement already satisfied: numpy!=1.24.0,>=1.20 in /Users/theerat/Documents/sign-language-translator/.venv/lib/python3.9/site-packages (from seaborn) (2.0.2)\n",
      "Requirement already satisfied: pandas>=1.2 in /Users/theerat/Documents/sign-language-translator/.venv/lib/python3.9/site-packages (from seaborn) (2.2.3)\n",
      "Requirement already satisfied: matplotlib!=3.6.1,>=3.4 in /Users/theerat/Documents/sign-language-translator/.venv/lib/python3.9/site-packages (from seaborn) (3.9.2)\n",
      "Requirement already satisfied: cramjam>=2.3 in /Users/theerat/Documents/sign-language-translator/.venv/lib/python3.9/site-packages (from fastparquet) (2.9.0)\n",
      "Requirement already satisfied: fsspec in /Users/theerat/Documents/sign-language-translator/.venv/lib/python3.9/site-packages (from fastparquet) (2024.3.1)\n",
      "Requirement already satisfied: contourpy>=1.0.1 in /Users/theerat/Documents/sign-language-translator/.venv/lib/python3.9/site-packages (from matplotlib!=3.6.1,>=3.4->seaborn) (1.3.0)\n",
      "Requirement already satisfied: cycler>=0.10 in /Users/theerat/Documents/sign-language-translator/.venv/lib/python3.9/site-packages (from matplotlib!=3.6.1,>=3.4->seaborn) (0.12.1)\n",
      "Requirement already satisfied: fonttools>=4.22.0 in /Users/theerat/Documents/sign-language-translator/.venv/lib/python3.9/site-packages (from matplotlib!=3.6.1,>=3.4->seaborn) (4.54.1)\n",
      "Requirement already satisfied: kiwisolver>=1.3.1 in /Users/theerat/Documents/sign-language-translator/.venv/lib/python3.9/site-packages (from matplotlib!=3.6.1,>=3.4->seaborn) (1.4.7)\n",
      "Requirement already satisfied: pillow>=8 in /Users/theerat/Documents/sign-language-translator/.venv/lib/python3.9/site-packages (from matplotlib!=3.6.1,>=3.4->seaborn) (11.0.0)\n",
      "Requirement already satisfied: pyparsing>=2.3.1 in /Users/theerat/Documents/sign-language-translator/.venv/lib/python3.9/site-packages (from matplotlib!=3.6.1,>=3.4->seaborn) (3.2.0)\n",
      "Requirement already satisfied: python-dateutil>=2.7 in /Users/theerat/Documents/sign-language-translator/.venv/lib/python3.9/site-packages (from matplotlib!=3.6.1,>=3.4->seaborn) (2.9.0.post0)\n",
      "Requirement already satisfied: importlib-resources>=3.2.0 in /Users/theerat/Documents/sign-language-translator/.venv/lib/python3.9/site-packages (from matplotlib!=3.6.1,>=3.4->seaborn) (6.4.5)\n",
      "Requirement already satisfied: pytz>=2020.1 in /Users/theerat/Documents/sign-language-translator/.venv/lib/python3.9/site-packages (from pandas>=1.2->seaborn) (2024.2)\n",
      "Requirement already satisfied: tzdata>=2022.7 in /Users/theerat/Documents/sign-language-translator/.venv/lib/python3.9/site-packages (from pandas>=1.2->seaborn) (2024.2)\n",
      "Requirement already satisfied: zipp>=3.1.0 in /Users/theerat/Documents/sign-language-translator/.venv/lib/python3.9/site-packages (from importlib-resources>=3.2.0->matplotlib!=3.6.1,>=3.4->seaborn) (3.20.2)\n",
      "Requirement already satisfied: six>=1.5 in /Users/theerat/Documents/sign-language-translator/.venv/lib/python3.9/site-packages (from python-dateutil>=2.7->matplotlib!=3.6.1,>=3.4->seaborn) (1.16.0)\n",
      "Using cached tensorflow_addons-0.23.0-cp39-cp39-macosx_11_0_arm64.whl (12.6 MB)\n",
      "Using cached seaborn-0.13.2-py3-none-any.whl (294 kB)\n",
      "Using cached typeguard-2.13.3-py3-none-any.whl (17 kB)\n",
      "Installing collected packages: typeguard, tensorflow-addons, seaborn\n",
      "Successfully installed seaborn-0.13.2 tensorflow-addons-0.23.0 typeguard-2.13.3\n",
      "Note: you may need to restart the kernel to use updated packages.\n"
     ]
    }
   ],
   "source": [
    "# %pip install tensorflow-addons\n",
    "# %pip uninstall tensorflow-addons\n",
    "# %pip install tqdm scikit-learn\n",
    "# %pip install fastparquet pyarrow\n",
    "# %pip install tensorflow-addons seaborn fastparquet"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import tensorflow as tf\n",
    "# import tensorflow_addons as tfa\n",
    "import matplotlib.pyplot as plt\n",
    "import matplotlib as mpl\n",
    "import seaborn as sn\n",
    "\n",
    "from tqdm.notebook import tqdm\n",
    "from sklearn.model_selection import train_test_split, GroupShuffleSplit \n",
    "\n",
    "import glob\n",
    "import sys\n",
    "import os\n",
    "import math\n",
    "import gc\n",
    "import sys\n",
    "import sklearn\n",
    "import scipy\n",
    "import ast\n",
    "import pyarrow\n",
    "import fastparquet\n",
    "\n",
    "# print(f'Tensorflow V{tf.__version__}')\n",
    "# print(f'Keras V{tf.keras.__version__}')\n",
    "# print(f'Python V{sys.version}')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# LSTM"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "from keras import utils"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [],
   "source": [
    "words = os.listdir('/Users/theerat/Dataset/train')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [],
   "source": [
    "words.remove('.DS_Store')\n",
    "# words.remove('ไม่มี')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "all_words = [\n",
    "    'ฉัน','กิน','พูด','อยู่','หนังสือ','คุณ','ดื่ม','ฟัง','อะไร',\n",
    "    'โรงเรียน','เขา','นอน','หัวเราะ','บ้าน','ข้าว','นักเรียน',\n",
    "    'เดิน','ชอบ','ที่ไหน','วันนี้','ไก่ผัดกะเพรา','วิ่ง','ไม่มี','ปลา','ควาย'\n",
    "             ]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['หนังสือ',\n",
       " 'อยู่',\n",
       " 'พูด',\n",
       " 'อะไร',\n",
       " 'ไก่ผัดกะเพรา',\n",
       " 'ปลา',\n",
       " 'ชอบ',\n",
       " 'หัวเราะ',\n",
       " 'ควาย',\n",
       " 'ไม่มี',\n",
       " 'ฟัง',\n",
       " 'ดื่ม',\n",
       " 'คุณ',\n",
       " 'นอน',\n",
       " 'เขา',\n",
       " 'กิน',\n",
       " 'โรงเรียน',\n",
       " 'ฉัน',\n",
       " 'ที่ไหน',\n",
       " 'ข้าว',\n",
       " 'บ้าน',\n",
       " 'นักเรียน',\n",
       " 'วันนี้',\n",
       " 'วิ่ง',\n",
       " 'เดิน']"
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "words"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "25"
      ]
     },
     "execution_count": 62,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(words)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [],
   "source": [
    "words_np = np.array(words)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [],
   "source": [
    "DATA_PATH = os.path.join('/Users/theerat/Dataset/train')\n",
    "\n",
    "sequence_length = 30\n",
    "no_sequences = 30"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'หนังสือ': 0,\n",
       " 'อยู่': 1,\n",
       " 'พูด': 2,\n",
       " 'อะไร': 3,\n",
       " 'ไก่ผัดกะเพรา': 4,\n",
       " 'ปลา': 5,\n",
       " 'ชอบ': 6,\n",
       " 'หัวเราะ': 7,\n",
       " 'ควาย': 8,\n",
       " 'ไม่มี': 9,\n",
       " 'ฟัง': 10,\n",
       " 'ดื่ม': 11,\n",
       " 'คุณ': 12,\n",
       " 'นอน': 13,\n",
       " 'เขา': 14,\n",
       " 'กิน': 15,\n",
       " 'โรงเรียน': 16,\n",
       " 'ฉัน': 17,\n",
       " 'ที่ไหน': 18,\n",
       " 'ข้าว': 19,\n",
       " 'บ้าน': 20,\n",
       " 'นักเรียน': 21,\n",
       " 'วันนี้': 22,\n",
       " 'วิ่ง': 23,\n",
       " 'เดิน': 24}"
      ]
     },
     "execution_count": 66,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "label_map = {label:num for num, label in enumerate(words_np)}\n",
    "label_map"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "อัดไฟล์"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['หนังสือ', 'อยู่', 'พูด', 'อะไร', 'ไก่ผัดกะเพรา', 'ปลา', 'ชอบ',\n",
       "       'หัวเราะ', 'ควาย', 'ไม่มี', 'ฟัง', 'ดื่ม', 'คุณ', 'นอน', 'เขา',\n",
       "       'กิน', 'โรงเรียน', 'ฉัน', 'ที่ไหน', 'ข้าว', 'บ้าน', 'นักเรียน',\n",
       "       'วันนี้', 'วิ่ง', 'เดิน'], dtype='<U12')"
      ]
     },
     "execution_count": 67,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "filtered_words = np.array(list(filter(lambda word: word != '.DS_Store', label_map)))\n",
    "filtered_words"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#use with npy file \n",
    "sequences, labels = [], []\n",
    "for action in words_np:\n",
    "    for sequence in range(no_sequences):\n",
    "        window = []\n",
    "        for frame_num in range(sequence_length):\n",
    "            res = np.load(os.path.join(DATA_PATH, action, str(sequence), \"{}.npy\".format(frame_num)))\n",
    "            window.append(res)\n",
    "        sequences.append(window)\n",
    "        labels.append(label_map[action])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#test\n",
    "csv_path ='/Users/theerat/Documents/tsl/sign-language-translator/Datasets/Landmarked/ควาย/1/0.csv'\n",
    "df = pd.read_csv(csv_path, skiprows=1)\n",
    "frame_data = df.to_numpy()\n",
    "\n",
    "frame_length = frame_data.size\n",
    "\n",
    "print(f\"Frame shape: {frame_data.shape}\")\n",
    "print(f\"Frame length (number of features): {frame_length}\")\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Processing Actions: 100%|██████████| 25/25 [04:30<00:00, 10.84s/it]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Shape of X: (1940, 30, 1404)\n",
      "Shape of y: (1940,)\n"
     ]
    }
   ],
   "source": [
    "import os\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import ast\n",
    "from tqdm import tqdm\n",
    "\n",
    "sequences_train, labels_train = [], []\n",
    "\n",
    "sequence_length = 30\n",
    "no_sequences = 20\n",
    "frame_length = 1404\n",
    "\n",
    "for action in tqdm(words, desc=\"Processing Actions\"):\n",
    "    n_sequences = os.listdir(os.path.join(DATA_PATH, action))\n",
    "\n",
    "    n_sequences = list(filter(lambda seq: seq != '.DS_Store', n_sequences))\n",
    "    for sequence in tqdm(n_sequences, desc=f\"Sequences for {action}\", leave=False):\n",
    "        window = []\n",
    "        for frame_num in range(sequence_length):\n",
    "            csv_path = os.path.join(DATA_PATH, action, sequence, f'{frame_num}.csv')\n",
    "            \n",
    "            res = pd.read_csv(csv_path, skiprows=1).to_numpy()\n",
    "    \n",
    "            res_cleaned = []\n",
    "            for row in res:\n",
    "                for val in row:\n",
    "                    if isinstance(val, str):\n",
    "                        res_cleaned.extend(ast.literal_eval(val))\n",
    "                    elif not np.isnan(val):\n",
    "                        res_cleaned.append(val)\n",
    "            \n",
    "            # NaN = 0\n",
    "            res_cleaned = np.nan_to_num(res_cleaned, nan=0.0)\n",
    "            if len(res_cleaned) < frame_length:\n",
    "                res_cleaned = np.pad(res_cleaned, (0, frame_length - len(res_cleaned)), mode='constant', constant_values=0)\n",
    "            elif len(res_cleaned) > frame_length:\n",
    "                res_cleaned = res_cleaned[:frame_length]\n",
    "            \n",
    "            window.append(res_cleaned)\n",
    "        \n",
    "        if len(window) < sequence_length:\n",
    "            padding_frame = np.zeros((frame_length,))\n",
    "            for _ in range(sequence_length - len(window)):\n",
    "                window.append(padding_frame)\n",
    "        elif len(window) > sequence_length:\n",
    "            window = window[:sequence_length]\n",
    "        \n",
    "        sequences.append(window)\n",
    "        labels.append(label_map[action])\n",
    "\n",
    "X = np.array(sequences)\n",
    "y = np.array(labels)\n",
    "\n",
    "print(f\"Shape of X: {X.shape}\")\n",
    "print(f\"Shape of y: {y.shape}\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train = np.array(sequences)\n",
    "y_train = utils.to_categorical(labels).astype(int)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Processing Actions: 100%|██████████| 25/25 [02:20<00:00,  5.60s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Shape of X: (1940, 30, 1404)\n",
      "Shape of y: (1940,)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    }
   ],
   "source": [
    "import os\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import ast\n",
    "from tqdm import tqdm\n",
    "\n",
    "sequences_test, labels_test = [], []\n",
    "\n",
    "sequence_length = 30\n",
    "# no_sequences = 20\n",
    "frame_length = 1404\n",
    "DATA_PATH_2 = os.path.join('/Users/theerat/Dataset/test')\n",
    "\n",
    "for action in tqdm(words, desc=\"Processing Actions\"):\n",
    "    n_sequences = os.listdir(os.path.join(DATA_PATH_2, action))\n",
    "\n",
    "    n_sequences = list(filter(lambda seq: seq != '.DS_Store', n_sequences))\n",
    "    for sequence in tqdm(n_sequences, desc=f\"Sequences for {action}\", leave=False):\n",
    "        window = []\n",
    "        for frame_num in range(sequence_length):\n",
    "            csv_path = os.path.join(DATA_PATH_2, action, sequence, f'{frame_num}.csv')\n",
    "            \n",
    "            res = pd.read_csv(csv_path, skiprows=1).to_numpy()\n",
    "    \n",
    "            res_cleaned = []\n",
    "            for row in res:\n",
    "                for val in row:\n",
    "                    if isinstance(val, str):\n",
    "                        res_cleaned.extend(ast.literal_eval(val))\n",
    "                    elif not np.isnan(val):\n",
    "                        res_cleaned.append(val)\n",
    "            \n",
    "            # NaN = 0\n",
    "            res_cleaned = np.nan_to_num(res_cleaned, nan=0.0)\n",
    "            if len(res_cleaned) < frame_length:\n",
    "                res_cleaned = np.pad(res_cleaned, (0, frame_length - len(res_cleaned)), mode='constant', constant_values=0)\n",
    "            elif len(res_cleaned) > frame_length:\n",
    "                res_cleaned = res_cleaned[:frame_length]\n",
    "            \n",
    "            window.append(res_cleaned)\n",
    "        \n",
    "        if len(window) < sequence_length:\n",
    "            padding_frame = np.zeros((frame_length,))\n",
    "            for _ in range(sequence_length - len(window)):\n",
    "                window.append(padding_frame)\n",
    "        elif len(window) > sequence_length:\n",
    "            window = window[:sequence_length]\n",
    "        \n",
    "        sequences_test.append(window)\n",
    "        labels_test.append(label_map[action])\n",
    "\n",
    "X_test = np.array(sequences_test)\n",
    "y_test = utils.to_categorical(labels_test).astype(int)\n",
    "\n",
    "print(f\"Shape of X: {X.shape}\")\n",
    "print(f\"Shape of y: {y.shape}\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {},
   "outputs": [],
   "source": [
    "np.save('s_train.npy', X_train)\n",
    "np.save('l_train.npy', y_train)\n",
    "np.save('s_test.npy', X_test)\n",
    "np.save('l_test.npy', y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {},
   "outputs": [],
   "source": [
    "sequences2, labels2 = np.load('s_train.npy'), np.load('l_train.npy')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>sequence</th>\n",
       "      <th>label</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>[[0.5058546662330627, 0.3687451481819153, -0.5...</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>[[0.5072938799858093, 0.3691006600856781, -0.6...</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>[[0.5099446773529053, 0.3753768801689148, -0.6...</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>[[0.508685827255249, 0.37606343626976013, -0.6...</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>[[0.5110201239585876, 0.3756888210773468, -0.6...</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1935</th>\n",
       "      <td>[[0.5248710513114929, 0.39305877685546875, -0....</td>\n",
       "      <td>24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1936</th>\n",
       "      <td>[[0.5181424617767334, 0.3898189067840576, -0.5...</td>\n",
       "      <td>24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1937</th>\n",
       "      <td>[[0.5432792901992798, 0.2515603005886078, -1.5...</td>\n",
       "      <td>24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1938</th>\n",
       "      <td>[[0.5269699096679688, 0.3915901184082031, -0.5...</td>\n",
       "      <td>24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1939</th>\n",
       "      <td>[[0.48171404004096985, 0.4502109885215759, -0....</td>\n",
       "      <td>24</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1940 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                               sequence  label\n",
       "0     [[0.5058546662330627, 0.3687451481819153, -0.5...      0\n",
       "1     [[0.5072938799858093, 0.3691006600856781, -0.6...      0\n",
       "2     [[0.5099446773529053, 0.3753768801689148, -0.6...      0\n",
       "3     [[0.508685827255249, 0.37606343626976013, -0.6...      0\n",
       "4     [[0.5110201239585876, 0.3756888210773468, -0.6...      0\n",
       "...                                                 ...    ...\n",
       "1935  [[0.5248710513114929, 0.39305877685546875, -0....     24\n",
       "1936  [[0.5181424617767334, 0.3898189067840576, -0.5...     24\n",
       "1937  [[0.5432792901992798, 0.2515603005886078, -1.5...     24\n",
       "1938  [[0.5269699096679688, 0.3915901184082031, -0.5...     24\n",
       "1939  [[0.48171404004096985, 0.4502109885215759, -0....     24\n",
       "\n",
       "[1940 rows x 2 columns]"
      ]
     },
     "execution_count": 82,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.DataFrame({'sequence':sequences, 'label':labels})\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.20,random_state=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((1940, 30, 1404), (259, 30, 1404), (1940, 25), (259, 25))"
      ]
     },
     "execution_count": 83,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train.shape, X_test.shape, y_train.shape, y_test.shape    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 0,  0,  0, ..., 24, 24, 24])"
      ]
     },
     "execution_count": 85,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_train_sparse = np.argmax(y_train, axis=-1)\n",
    "y_train_sparse"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "__New Model__"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {},
   "outputs": [],
   "source": [
    "from keras import models, layers, callbacks\n",
    "from keras import callbacks, Model, Input"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {},
   "outputs": [],
   "source": [
    "# include early stopping and reducelr\n",
    "def get_callbacks():\n",
    "    return [\n",
    "        callbacks.EarlyStopping(\n",
    "            monitor=\"val_accuracy\", patience=10, restore_best_weights=True\n",
    "        ),\n",
    "        callbacks.ReduceLROnPlateau(\n",
    "            monitor=\"val_accuracy\", factor=0.5, patience=3\n",
    "        ),\n",
    "    ]\n",
    "\n",
    "\n",
    "# a single dense block followed by a normalization block and relu activation\n",
    "def dense_block(units, name):\n",
    "    fc = layers.Dense(units)\n",
    "    norm = layers.LayerNormalization()\n",
    "    act = layers.Activation(\"relu\")\n",
    "    drop = layers.Dropout(0.1)\n",
    "    return lambda x: drop(act(norm(fc(x))))\n",
    "\n",
    "\n",
    "# the lstm block with the final dense block for the classification\n",
    "def classifier(lstm_units):\n",
    "    lstm = layers.LSTM(lstm_units)\n",
    "    out = layers.Dense(25, activation=\"softmax\")\n",
    "    return lambda x: out(lstm(x))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\">Model: \"functional_2\"</span>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1mModel: \"functional_2\"\u001b[0m\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓\n",
       "┃<span style=\"font-weight: bold\"> Layer (type)                    </span>┃<span style=\"font-weight: bold\"> Output Shape           </span>┃<span style=\"font-weight: bold\">       Param # </span>┃\n",
       "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩\n",
       "│ input_layer_2 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">InputLayer</span>)      │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">30</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1404</span>)       │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dense_6 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>)                 │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">30</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span>)        │       <span style=\"color: #00af00; text-decoration-color: #00af00\">719,360</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ layer_normalization_4           │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">30</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span>)        │         <span style=\"color: #00af00; text-decoration-color: #00af00\">1,024</span> │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">LayerNormalization</span>)            │                        │               │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ activation_4 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)       │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">30</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span>)        │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dropout_4 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dropout</span>)             │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">30</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span>)        │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dense_7 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>)                 │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">30</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)        │       <span style=\"color: #00af00; text-decoration-color: #00af00\">131,328</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ layer_normalization_5           │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">30</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)        │           <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span> │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">LayerNormalization</span>)            │                        │               │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ activation_5 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)       │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">30</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)        │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dropout_5 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dropout</span>)             │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">30</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)        │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ lstm_2 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">LSTM</span>)                   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">500</span>)            │     <span style=\"color: #00af00; text-decoration-color: #00af00\">1,514,000</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dense_8 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>)                 │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">25</span>)             │        <span style=\"color: #00af00; text-decoration-color: #00af00\">12,525</span> │\n",
       "└─────────────────────────────────┴────────────────────────┴───────────────┘\n",
       "</pre>\n"
      ],
      "text/plain": [
       "┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓\n",
       "┃\u001b[1m \u001b[0m\u001b[1mLayer (type)                   \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mOutput Shape          \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m      Param #\u001b[0m\u001b[1m \u001b[0m┃\n",
       "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩\n",
       "│ input_layer_2 (\u001b[38;5;33mInputLayer\u001b[0m)      │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m30\u001b[0m, \u001b[38;5;34m1404\u001b[0m)       │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dense_6 (\u001b[38;5;33mDense\u001b[0m)                 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m30\u001b[0m, \u001b[38;5;34m512\u001b[0m)        │       \u001b[38;5;34m719,360\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ layer_normalization_4           │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m30\u001b[0m, \u001b[38;5;34m512\u001b[0m)        │         \u001b[38;5;34m1,024\u001b[0m │\n",
       "│ (\u001b[38;5;33mLayerNormalization\u001b[0m)            │                        │               │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ activation_4 (\u001b[38;5;33mActivation\u001b[0m)       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m30\u001b[0m, \u001b[38;5;34m512\u001b[0m)        │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dropout_4 (\u001b[38;5;33mDropout\u001b[0m)             │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m30\u001b[0m, \u001b[38;5;34m512\u001b[0m)        │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dense_7 (\u001b[38;5;33mDense\u001b[0m)                 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m30\u001b[0m, \u001b[38;5;34m256\u001b[0m)        │       \u001b[38;5;34m131,328\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ layer_normalization_5           │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m30\u001b[0m, \u001b[38;5;34m256\u001b[0m)        │           \u001b[38;5;34m512\u001b[0m │\n",
       "│ (\u001b[38;5;33mLayerNormalization\u001b[0m)            │                        │               │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ activation_5 (\u001b[38;5;33mActivation\u001b[0m)       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m30\u001b[0m, \u001b[38;5;34m256\u001b[0m)        │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dropout_5 (\u001b[38;5;33mDropout\u001b[0m)             │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m30\u001b[0m, \u001b[38;5;34m256\u001b[0m)        │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ lstm_2 (\u001b[38;5;33mLSTM\u001b[0m)                   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m500\u001b[0m)            │     \u001b[38;5;34m1,514,000\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dense_8 (\u001b[38;5;33mDense\u001b[0m)                 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m25\u001b[0m)             │        \u001b[38;5;34m12,525\u001b[0m │\n",
       "└─────────────────────────────────┴────────────────────────┴───────────────┘\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Total params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">2,378,749</span> (9.07 MB)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Total params: \u001b[0m\u001b[38;5;34m2,378,749\u001b[0m (9.07 MB)\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">2,378,749</span> (9.07 MB)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m2,378,749\u001b[0m (9.07 MB)\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Non-trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> (0.00 B)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Non-trainable params: \u001b[0m\u001b[38;5;34m0\u001b[0m (0.00 B)\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import tensorflow as tf\n",
    "from keras import layers, Model, Input, callbacks,optimizers\n",
    "\n",
    "encoder_units = [512, 256]\n",
    "lstm_units = 500  \n",
    "\n",
    "inputs = Input(shape=(30, 1404))\n",
    "\n",
    "x = inputs\n",
    "for i, n in enumerate(encoder_units):\n",
    "    x = dense_block(n, f\"encoder_{i}\")(x)\n",
    "\n",
    "out = classifier(lstm_units)(x)\n",
    "\n",
    "model = Model(inputs=inputs, outputs=out)\n",
    "model.compile(optimizer=optimizers.Adam(), loss=\"categorical_crossentropy\", metrics=[\"accuracy\"])\n",
    "callbacks_list = get_callbacks()\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<PIL.Image.Image image mode=RGBA size=1561x2078>"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import visualkeras\n",
    "from PIL import ImageFont\n",
    "# font = ImageFont.truetype(\"SukhumvitSet-Medium.ttf\", 9)\n",
    "visualkeras.layered_view(model, legend=True,spacing=100)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/100\n",
      "\u001b[1m49/49\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m17s\u001b[0m 281ms/step - accuracy: 0.1072 - loss: 3.0355 - val_accuracy: 0.3196 - val_loss: 1.7563 - learning_rate: 0.0010\n",
      "Epoch 2/100\n",
      "\u001b[1m49/49\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m15s\u001b[0m 296ms/step - accuracy: 0.4417 - loss: 1.5304 - val_accuracy: 0.3969 - val_loss: 1.3146 - learning_rate: 0.0010\n",
      "Epoch 3/100\n",
      "\u001b[1m49/49\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m14s\u001b[0m 282ms/step - accuracy: 0.5110 - loss: 1.2411 - val_accuracy: 0.5979 - val_loss: 1.0759 - learning_rate: 0.0010\n",
      "Epoch 4/100\n",
      "\u001b[1m49/49\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m14s\u001b[0m 285ms/step - accuracy: 0.6675 - loss: 0.9063 - val_accuracy: 0.5799 - val_loss: 0.9784 - learning_rate: 0.0010\n",
      "Epoch 5/100\n",
      "\u001b[1m49/49\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m15s\u001b[0m 310ms/step - accuracy: 0.7057 - loss: 0.7319 - val_accuracy: 0.4948 - val_loss: 1.3837 - learning_rate: 0.0010\n",
      "Epoch 6/100\n",
      "\u001b[1m49/49\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m14s\u001b[0m 289ms/step - accuracy: 0.6448 - loss: 0.9208 - val_accuracy: 0.5361 - val_loss: 1.0636 - learning_rate: 0.0010\n",
      "Epoch 7/100\n",
      "\u001b[1m49/49\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m15s\u001b[0m 307ms/step - accuracy: 0.7501 - loss: 0.6067 - val_accuracy: 0.6856 - val_loss: 0.6201 - learning_rate: 5.0000e-04\n",
      "Epoch 8/100\n",
      "\u001b[1m49/49\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m15s\u001b[0m 299ms/step - accuracy: 0.7934 - loss: 0.5305 - val_accuracy: 0.7603 - val_loss: 0.6628 - learning_rate: 5.0000e-04\n",
      "Epoch 9/100\n",
      "\u001b[1m49/49\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m19s\u001b[0m 388ms/step - accuracy: 0.8545 - loss: 0.3802 - val_accuracy: 0.7990 - val_loss: 0.4861 - learning_rate: 5.0000e-04\n",
      "Epoch 10/100\n",
      "\u001b[1m49/49\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m16s\u001b[0m 336ms/step - accuracy: 0.8712 - loss: 0.3492 - val_accuracy: 0.8505 - val_loss: 0.4360 - learning_rate: 5.0000e-04\n",
      "Epoch 11/100\n",
      "\u001b[1m49/49\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m15s\u001b[0m 310ms/step - accuracy: 0.7651 - loss: 0.6540 - val_accuracy: 0.7629 - val_loss: 0.5485 - learning_rate: 5.0000e-04\n",
      "Epoch 12/100\n",
      "\u001b[1m49/49\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m17s\u001b[0m 341ms/step - accuracy: 0.8704 - loss: 0.3736 - val_accuracy: 0.8840 - val_loss: 0.4187 - learning_rate: 5.0000e-04\n",
      "Epoch 13/100\n",
      "\u001b[1m49/49\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m15s\u001b[0m 308ms/step - accuracy: 0.8915 - loss: 0.3154 - val_accuracy: 0.7191 - val_loss: 0.7756 - learning_rate: 5.0000e-04\n",
      "Epoch 14/100\n",
      "\u001b[1m49/49\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m15s\u001b[0m 304ms/step - accuracy: 0.8513 - loss: 0.4028 - val_accuracy: 0.7887 - val_loss: 0.4181 - learning_rate: 5.0000e-04\n",
      "Epoch 15/100\n",
      "\u001b[1m49/49\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m15s\u001b[0m 316ms/step - accuracy: 0.8757 - loss: 0.3537 - val_accuracy: 0.8995 - val_loss: 0.3143 - learning_rate: 5.0000e-04\n",
      "Epoch 16/100\n",
      "\u001b[1m49/49\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m15s\u001b[0m 298ms/step - accuracy: 0.8358 - loss: 0.4561 - val_accuracy: 0.7474 - val_loss: 0.5566 - learning_rate: 5.0000e-04\n",
      "Epoch 17/100\n",
      "\u001b[1m49/49\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m14s\u001b[0m 287ms/step - accuracy: 0.9020 - loss: 0.3074 - val_accuracy: 0.8402 - val_loss: 0.3573 - learning_rate: 5.0000e-04\n",
      "Epoch 18/100\n",
      "\u001b[1m49/49\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m14s\u001b[0m 289ms/step - accuracy: 0.9323 - loss: 0.2116 - val_accuracy: 0.8505 - val_loss: 0.3503 - learning_rate: 5.0000e-04\n",
      "Epoch 19/100\n",
      "\u001b[1m49/49\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m14s\u001b[0m 296ms/step - accuracy: 0.9596 - loss: 0.1555 - val_accuracy: 0.9536 - val_loss: 0.2016 - learning_rate: 2.5000e-04\n",
      "Epoch 20/100\n",
      "\u001b[1m49/49\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m15s\u001b[0m 309ms/step - accuracy: 0.9658 - loss: 0.1356 - val_accuracy: 0.9407 - val_loss: 0.1968 - learning_rate: 2.5000e-04\n",
      "Epoch 21/100\n",
      "\u001b[1m49/49\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m14s\u001b[0m 294ms/step - accuracy: 0.9698 - loss: 0.1125 - val_accuracy: 0.8892 - val_loss: 0.3052 - learning_rate: 2.5000e-04\n",
      "Epoch 22/100\n",
      "\u001b[1m49/49\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m15s\u001b[0m 308ms/step - accuracy: 0.9555 - loss: 0.1526 - val_accuracy: 0.9639 - val_loss: 0.1458 - learning_rate: 2.5000e-04\n",
      "Epoch 23/100\n",
      "\u001b[1m49/49\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m14s\u001b[0m 290ms/step - accuracy: 0.9199 - loss: 0.2103 - val_accuracy: 0.8531 - val_loss: 0.3064 - learning_rate: 2.5000e-04\n",
      "Epoch 24/100\n",
      "\u001b[1m49/49\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m14s\u001b[0m 290ms/step - accuracy: 0.9293 - loss: 0.1752 - val_accuracy: 0.7552 - val_loss: 0.4961 - learning_rate: 2.5000e-04\n",
      "Epoch 25/100\n",
      "\u001b[1m49/49\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m14s\u001b[0m 289ms/step - accuracy: 0.9357 - loss: 0.2045 - val_accuracy: 0.9253 - val_loss: 0.2203 - learning_rate: 2.5000e-04\n",
      "Epoch 26/100\n",
      "\u001b[1m49/49\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m14s\u001b[0m 293ms/step - accuracy: 0.9423 - loss: 0.1595 - val_accuracy: 0.9665 - val_loss: 0.1338 - learning_rate: 1.2500e-04\n",
      "Epoch 27/100\n",
      "\u001b[1m49/49\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m14s\u001b[0m 295ms/step - accuracy: 0.9748 - loss: 0.0835 - val_accuracy: 0.9639 - val_loss: 0.1042 - learning_rate: 1.2500e-04\n",
      "Epoch 28/100\n",
      "\u001b[1m49/49\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m14s\u001b[0m 291ms/step - accuracy: 0.9751 - loss: 0.0729 - val_accuracy: 0.9665 - val_loss: 0.0956 - learning_rate: 1.2500e-04\n",
      "Epoch 29/100\n",
      "\u001b[1m49/49\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m15s\u001b[0m 295ms/step - accuracy: 0.9810 - loss: 0.0585 - val_accuracy: 0.9716 - val_loss: 0.1081 - learning_rate: 1.2500e-04\n",
      "Epoch 30/100\n",
      "\u001b[1m49/49\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m14s\u001b[0m 286ms/step - accuracy: 0.9792 - loss: 0.0650 - val_accuracy: 0.9794 - val_loss: 0.0877 - learning_rate: 1.2500e-04\n",
      "Epoch 31/100\n",
      "\u001b[1m49/49\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m14s\u001b[0m 292ms/step - accuracy: 0.9857 - loss: 0.0577 - val_accuracy: 0.9742 - val_loss: 0.0980 - learning_rate: 1.2500e-04\n",
      "Epoch 32/100\n",
      "\u001b[1m49/49\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m14s\u001b[0m 290ms/step - accuracy: 0.9842 - loss: 0.0601 - val_accuracy: 0.9897 - val_loss: 0.0615 - learning_rate: 1.2500e-04\n",
      "Epoch 33/100\n",
      "\u001b[1m49/49\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m14s\u001b[0m 295ms/step - accuracy: 0.9759 - loss: 0.0783 - val_accuracy: 0.9716 - val_loss: 0.0856 - learning_rate: 1.2500e-04\n",
      "Epoch 34/100\n",
      "\u001b[1m49/49\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m15s\u001b[0m 297ms/step - accuracy: 0.9757 - loss: 0.0690 - val_accuracy: 0.9820 - val_loss: 0.0640 - learning_rate: 1.2500e-04\n",
      "Epoch 35/100\n",
      "\u001b[1m49/49\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m15s\u001b[0m 294ms/step - accuracy: 0.9861 - loss: 0.0544 - val_accuracy: 0.9716 - val_loss: 0.0679 - learning_rate: 1.2500e-04\n",
      "Epoch 36/100\n",
      "\u001b[1m49/49\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m17s\u001b[0m 338ms/step - accuracy: 0.9828 - loss: 0.0521 - val_accuracy: 0.9845 - val_loss: 0.0544 - learning_rate: 6.2500e-05\n",
      "Epoch 37/100\n",
      "\u001b[1m49/49\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m17s\u001b[0m 354ms/step - accuracy: 0.9934 - loss: 0.0349 - val_accuracy: 0.9768 - val_loss: 0.0682 - learning_rate: 6.2500e-05\n",
      "Epoch 38/100\n",
      "\u001b[1m49/49\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m17s\u001b[0m 335ms/step - accuracy: 0.9817 - loss: 0.0559 - val_accuracy: 0.9871 - val_loss: 0.0453 - learning_rate: 6.2500e-05\n",
      "Epoch 39/100\n",
      "\u001b[1m49/49\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m17s\u001b[0m 347ms/step - accuracy: 0.9862 - loss: 0.0481 - val_accuracy: 0.9923 - val_loss: 0.0428 - learning_rate: 3.1250e-05\n",
      "Epoch 40/100\n",
      "\u001b[1m49/49\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m16s\u001b[0m 331ms/step - accuracy: 0.9916 - loss: 0.0363 - val_accuracy: 0.9897 - val_loss: 0.0468 - learning_rate: 3.1250e-05\n",
      "Epoch 41/100\n",
      "\u001b[1m49/49\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m16s\u001b[0m 329ms/step - accuracy: 0.9926 - loss: 0.0344 - val_accuracy: 0.9923 - val_loss: 0.0368 - learning_rate: 3.1250e-05\n",
      "Epoch 42/100\n",
      "\u001b[1m49/49\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m17s\u001b[0m 343ms/step - accuracy: 0.9952 - loss: 0.0269 - val_accuracy: 0.9923 - val_loss: 0.0364 - learning_rate: 3.1250e-05\n",
      "Epoch 43/100\n",
      "\u001b[1m49/49\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m16s\u001b[0m 317ms/step - accuracy: 0.9963 - loss: 0.0257 - val_accuracy: 0.9923 - val_loss: 0.0380 - learning_rate: 1.5625e-05\n",
      "Epoch 44/100\n",
      "\u001b[1m49/49\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m15s\u001b[0m 309ms/step - accuracy: 0.9921 - loss: 0.0287 - val_accuracy: 0.9923 - val_loss: 0.0344 - learning_rate: 1.5625e-05\n",
      "Epoch 45/100\n",
      "\u001b[1m49/49\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m16s\u001b[0m 326ms/step - accuracy: 0.9965 - loss: 0.0208 - val_accuracy: 0.9923 - val_loss: 0.0332 - learning_rate: 1.5625e-05\n",
      "Epoch 46/100\n",
      "\u001b[1m49/49\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m15s\u001b[0m 310ms/step - accuracy: 0.9921 - loss: 0.0290 - val_accuracy: 0.9923 - val_loss: 0.0349 - learning_rate: 7.8125e-06\n",
      "Epoch 47/100\n",
      "\u001b[1m49/49\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m15s\u001b[0m 309ms/step - accuracy: 0.9926 - loss: 0.0331 - val_accuracy: 0.9923 - val_loss: 0.0333 - learning_rate: 7.8125e-06\n",
      "Epoch 48/100\n",
      "\u001b[1m49/49\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m15s\u001b[0m 315ms/step - accuracy: 0.9946 - loss: 0.0257 - val_accuracy: 0.9923 - val_loss: 0.0332 - learning_rate: 7.8125e-06\n",
      "Epoch 49/100\n",
      "\u001b[1m49/49\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m15s\u001b[0m 314ms/step - accuracy: 0.9949 - loss: 0.0244 - val_accuracy: 0.9923 - val_loss: 0.0326 - learning_rate: 3.9063e-06\n"
     ]
    }
   ],
   "source": [
    "history = model.fit(\n",
    "                    X_train,\n",
    "                    y_train,\n",
    "                    epochs=100,\n",
    "                    batch_size=32,\n",
    "                    validation_split=0.2,\n",
    "                    callbacks=callbacks_list,\n",
    "                    verbose=1\n",
    "                    )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 172ms/step - accuracy: 0.0380 - loss: 3.4773 - val_accuracy: 0.0167 - val_loss: 3.2237\n",
      "Epoch 2/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 168ms/step - accuracy: 0.1125 - loss: 2.9988 - val_accuracy: 0.1500 - val_loss: 2.5614\n",
      "Epoch 3/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 307ms/step - accuracy: 0.2298 - loss: 2.2898 - val_accuracy: 0.2750 - val_loss: 1.8654\n",
      "Epoch 4/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 294ms/step - accuracy: 0.3034 - loss: 1.9575 - val_accuracy: 0.3000 - val_loss: 1.6400\n",
      "Epoch 5/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 297ms/step - accuracy: 0.3504 - loss: 1.6448 - val_accuracy: 0.4667 - val_loss: 1.2230\n",
      "Epoch 6/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 298ms/step - accuracy: 0.4290 - loss: 1.3807 - val_accuracy: 0.4333 - val_loss: 1.3254\n",
      "Epoch 7/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 296ms/step - accuracy: 0.4506 - loss: 1.3364 - val_accuracy: 0.4333 - val_loss: 1.3211\n",
      "Epoch 8/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 293ms/step - accuracy: 0.4468 - loss: 1.3390 - val_accuracy: 0.5417 - val_loss: 1.1390\n",
      "Epoch 9/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 324ms/step - accuracy: 0.5777 - loss: 1.1485 - val_accuracy: 0.6417 - val_loss: 0.9854\n",
      "Epoch 10/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 289ms/step - accuracy: 0.5624 - loss: 1.0765 - val_accuracy: 0.6000 - val_loss: 0.9324\n",
      "Epoch 11/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 298ms/step - accuracy: 0.5831 - loss: 1.0146 - val_accuracy: 0.6500 - val_loss: 0.9056\n",
      "Epoch 12/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 304ms/step - accuracy: 0.6603 - loss: 0.9033 - val_accuracy: 0.6167 - val_loss: 0.8868\n",
      "Epoch 13/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 302ms/step - accuracy: 0.6301 - loss: 0.8589 - val_accuracy: 0.6583 - val_loss: 0.7510\n",
      "Epoch 14/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 292ms/step - accuracy: 0.5433 - loss: 1.0545 - val_accuracy: 0.5750 - val_loss: 0.9066\n",
      "Epoch 15/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 301ms/step - accuracy: 0.6339 - loss: 0.8935 - val_accuracy: 0.7333 - val_loss: 0.7234\n",
      "Epoch 16/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 296ms/step - accuracy: 0.7369 - loss: 0.6985 - val_accuracy: 0.7000 - val_loss: 0.8275\n",
      "Epoch 17/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 291ms/step - accuracy: 0.6633 - loss: 0.7993 - val_accuracy: 0.6750 - val_loss: 0.7831\n",
      "Epoch 18/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 308ms/step - accuracy: 0.5664 - loss: 1.0019 - val_accuracy: 0.6750 - val_loss: 0.8049\n",
      "Epoch 19/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 298ms/step - accuracy: 0.5979 - loss: 0.9269 - val_accuracy: 0.7250 - val_loss: 0.7361\n",
      "Epoch 20/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 303ms/step - accuracy: 0.7084 - loss: 0.7388 - val_accuracy: 0.7333 - val_loss: 0.6979\n",
      "Epoch 21/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 302ms/step - accuracy: 0.7604 - loss: 0.6244 - val_accuracy: 0.7750 - val_loss: 0.5831\n",
      "Epoch 22/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 308ms/step - accuracy: 0.7462 - loss: 0.6259 - val_accuracy: 0.7333 - val_loss: 0.5547\n",
      "Epoch 23/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 290ms/step - accuracy: 0.7787 - loss: 0.5699 - val_accuracy: 0.5833 - val_loss: 0.9645\n",
      "Epoch 24/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 305ms/step - accuracy: 0.5954 - loss: 0.9230 - val_accuracy: 0.7000 - val_loss: 0.6806\n",
      "Epoch 25/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 297ms/step - accuracy: 0.7907 - loss: 0.5967 - val_accuracy: 0.7250 - val_loss: 0.6607\n",
      "Epoch 26/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 296ms/step - accuracy: 0.7558 - loss: 0.5995 - val_accuracy: 0.7667 - val_loss: 0.5581\n",
      "Epoch 27/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 300ms/step - accuracy: 0.7727 - loss: 0.5554 - val_accuracy: 0.7583 - val_loss: 0.5782\n",
      "Epoch 28/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 303ms/step - accuracy: 0.8393 - loss: 0.4648 - val_accuracy: 0.7917 - val_loss: 0.5086\n",
      "Epoch 29/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 297ms/step - accuracy: 0.7674 - loss: 0.5263 - val_accuracy: 0.8500 - val_loss: 0.4574\n",
      "Epoch 30/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 294ms/step - accuracy: 0.8140 - loss: 0.4863 - val_accuracy: 0.7750 - val_loss: 0.5982\n",
      "Epoch 31/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 291ms/step - accuracy: 0.7926 - loss: 0.4888 - val_accuracy: 0.8083 - val_loss: 0.5297\n",
      "Epoch 32/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 303ms/step - accuracy: 0.7964 - loss: 0.5624 - val_accuracy: 0.8083 - val_loss: 0.5669\n",
      "Epoch 33/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 332ms/step - accuracy: 0.7699 - loss: 0.6080 - val_accuracy: 0.7333 - val_loss: 0.6606\n",
      "Epoch 34/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 325ms/step - accuracy: 0.8069 - loss: 0.4917 - val_accuracy: 0.8333 - val_loss: 0.4480\n",
      "Epoch 35/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 315ms/step - accuracy: 0.8623 - loss: 0.4146 - val_accuracy: 0.8000 - val_loss: 0.5204\n",
      "Epoch 36/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 317ms/step - accuracy: 0.7929 - loss: 0.4887 - val_accuracy: 0.6167 - val_loss: 0.9522\n",
      "Epoch 37/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 308ms/step - accuracy: 0.7502 - loss: 0.6155 - val_accuracy: 0.8000 - val_loss: 0.4915\n",
      "Epoch 38/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 317ms/step - accuracy: 0.8463 - loss: 0.4118 - val_accuracy: 0.7667 - val_loss: 0.5784\n",
      "Epoch 39/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 354ms/step - accuracy: 0.8029 - loss: 0.4841 - val_accuracy: 0.8250 - val_loss: 0.6167\n",
      "Epoch 40/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 323ms/step - accuracy: 0.7973 - loss: 0.4642 - val_accuracy: 0.8000 - val_loss: 0.5282\n",
      "Epoch 41/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 321ms/step - accuracy: 0.8667 - loss: 0.4036 - val_accuracy: 0.8417 - val_loss: 0.4427\n",
      "Epoch 42/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 319ms/step - accuracy: 0.8245 - loss: 0.4423 - val_accuracy: 0.6250 - val_loss: 1.0590\n",
      "Epoch 43/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 313ms/step - accuracy: 0.6620 - loss: 0.8835 - val_accuracy: 0.7750 - val_loss: 0.6532\n",
      "Epoch 44/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 306ms/step - accuracy: 0.7942 - loss: 0.5519 - val_accuracy: 0.8417 - val_loss: 0.4929\n",
      "Epoch 45/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 303ms/step - accuracy: 0.8685 - loss: 0.3942 - val_accuracy: 0.8583 - val_loss: 0.4321\n",
      "Epoch 46/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 343ms/step - accuracy: 0.8779 - loss: 0.3183 - val_accuracy: 0.8500 - val_loss: 0.3763\n",
      "Epoch 47/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 296ms/step - accuracy: 0.8909 - loss: 0.2852 - val_accuracy: 0.8917 - val_loss: 0.3403\n",
      "Epoch 48/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 301ms/step - accuracy: 0.9266 - loss: 0.2443 - val_accuracy: 0.8000 - val_loss: 0.5723\n",
      "Epoch 49/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 302ms/step - accuracy: 0.8453 - loss: 0.4121 - val_accuracy: 0.7000 - val_loss: 0.7351\n",
      "Epoch 50/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 305ms/step - accuracy: 0.7726 - loss: 0.5292 - val_accuracy: 0.7833 - val_loss: 0.6448\n",
      "Epoch 51/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 295ms/step - accuracy: 0.8504 - loss: 0.4105 - val_accuracy: 0.8750 - val_loss: 0.3752\n",
      "Epoch 52/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 295ms/step - accuracy: 0.9028 - loss: 0.3389 - val_accuracy: 0.8667 - val_loss: 0.4526\n",
      "Epoch 53/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 300ms/step - accuracy: 0.9192 - loss: 0.2589 - val_accuracy: 0.9083 - val_loss: 0.3540\n",
      "Epoch 54/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 300ms/step - accuracy: 0.9225 - loss: 0.2383 - val_accuracy: 0.8167 - val_loss: 0.4480\n",
      "Epoch 55/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 304ms/step - accuracy: 0.8981 - loss: 0.2743 - val_accuracy: 0.7917 - val_loss: 0.5874\n",
      "Epoch 56/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 315ms/step - accuracy: 0.8082 - loss: 0.4771 - val_accuracy: 0.8083 - val_loss: 0.6214\n",
      "Epoch 57/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 307ms/step - accuracy: 0.7456 - loss: 0.5626 - val_accuracy: 0.7583 - val_loss: 0.6627\n",
      "Epoch 58/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 301ms/step - accuracy: 0.8260 - loss: 0.4931 - val_accuracy: 0.8250 - val_loss: 0.4609\n",
      "Epoch 59/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 288ms/step - accuracy: 0.9168 - loss: 0.2693 - val_accuracy: 0.8750 - val_loss: 0.3305\n",
      "Epoch 60/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 302ms/step - accuracy: 0.9009 - loss: 0.2637 - val_accuracy: 0.8583 - val_loss: 0.4048\n",
      "Epoch 61/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 308ms/step - accuracy: 0.9118 - loss: 0.2682 - val_accuracy: 0.8667 - val_loss: 0.3653\n",
      "Epoch 62/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 303ms/step - accuracy: 0.9020 - loss: 0.2187 - val_accuracy: 0.8000 - val_loss: 0.5623\n",
      "Epoch 63/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 306ms/step - accuracy: 0.7548 - loss: 0.6401 - val_accuracy: 0.4833 - val_loss: 1.9414\n",
      "Epoch 64/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 341ms/step - accuracy: 0.6869 - loss: 0.8388 - val_accuracy: 0.8167 - val_loss: 0.5699\n",
      "Epoch 65/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 298ms/step - accuracy: 0.8688 - loss: 0.4312 - val_accuracy: 0.8417 - val_loss: 0.4411\n",
      "Epoch 66/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 341ms/step - accuracy: 0.9081 - loss: 0.2631 - val_accuracy: 0.8500 - val_loss: 0.4270\n",
      "Epoch 67/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 482ms/step - accuracy: 0.9430 - loss: 0.1909 - val_accuracy: 0.8667 - val_loss: 0.4180\n",
      "Epoch 68/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 376ms/step - accuracy: 0.9123 - loss: 0.2405 - val_accuracy: 0.8917 - val_loss: 0.3144\n",
      "Epoch 69/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 307ms/step - accuracy: 0.8883 - loss: 0.2609 - val_accuracy: 0.8667 - val_loss: 0.4401\n",
      "Epoch 70/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 316ms/step - accuracy: 0.9092 - loss: 0.2478 - val_accuracy: 0.8083 - val_loss: 0.5203\n",
      "Epoch 71/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 318ms/step - accuracy: 0.8840 - loss: 0.3439 - val_accuracy: 0.8833 - val_loss: 0.4055\n",
      "Epoch 72/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 299ms/step - accuracy: 0.8900 - loss: 0.2606 - val_accuracy: 0.8667 - val_loss: 0.4289\n",
      "Epoch 73/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 296ms/step - accuracy: 0.9490 - loss: 0.1751 - val_accuracy: 0.9167 - val_loss: 0.3391\n",
      "Epoch 74/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 300ms/step - accuracy: 0.9339 - loss: 0.1817 - val_accuracy: 0.9083 - val_loss: 0.2861\n",
      "Epoch 75/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 300ms/step - accuracy: 0.9477 - loss: 0.1391 - val_accuracy: 0.8417 - val_loss: 0.5025\n",
      "Epoch 76/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 311ms/step - accuracy: 0.9534 - loss: 0.1578 - val_accuracy: 0.8000 - val_loss: 0.5326\n",
      "Epoch 77/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 313ms/step - accuracy: 0.9327 - loss: 0.1907 - val_accuracy: 0.8833 - val_loss: 0.3559\n",
      "Epoch 78/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 294ms/step - accuracy: 0.9416 - loss: 0.1498 - val_accuracy: 0.8917 - val_loss: 0.4081\n",
      "Epoch 79/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 285ms/step - accuracy: 0.9395 - loss: 0.1634 - val_accuracy: 0.8583 - val_loss: 0.4231\n",
      "Epoch 80/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 311ms/step - accuracy: 0.9386 - loss: 0.1809 - val_accuracy: 0.9250 - val_loss: 0.2928\n",
      "Epoch 81/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 300ms/step - accuracy: 0.9331 - loss: 0.1667 - val_accuracy: 0.7667 - val_loss: 0.6240\n",
      "Epoch 82/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 296ms/step - accuracy: 0.8309 - loss: 0.3965 - val_accuracy: 0.7167 - val_loss: 0.8564\n",
      "Epoch 83/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 298ms/step - accuracy: 0.7695 - loss: 0.6275 - val_accuracy: 0.8167 - val_loss: 0.5809\n",
      "Epoch 84/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 296ms/step - accuracy: 0.8976 - loss: 0.3579 - val_accuracy: 0.8167 - val_loss: 0.4088\n",
      "Epoch 85/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 308ms/step - accuracy: 0.8665 - loss: 0.2874 - val_accuracy: 0.8917 - val_loss: 0.3360\n",
      "Epoch 86/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 297ms/step - accuracy: 0.9286 - loss: 0.1706 - val_accuracy: 0.8917 - val_loss: 0.3917\n",
      "Epoch 87/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 293ms/step - accuracy: 0.9645 - loss: 0.1409 - val_accuracy: 0.9083 - val_loss: 0.2709\n",
      "Epoch 88/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 300ms/step - accuracy: 0.9626 - loss: 0.1285 - val_accuracy: 0.9333 - val_loss: 0.2736\n",
      "Epoch 89/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 312ms/step - accuracy: 0.9653 - loss: 0.1304 - val_accuracy: 0.9000 - val_loss: 0.3444\n",
      "Epoch 90/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 300ms/step - accuracy: 0.9144 - loss: 0.1841 - val_accuracy: 0.8417 - val_loss: 0.5063\n",
      "Epoch 91/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 300ms/step - accuracy: 0.8854 - loss: 0.2550 - val_accuracy: 0.8417 - val_loss: 0.4047\n",
      "Epoch 92/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 292ms/step - accuracy: 0.9514 - loss: 0.1424 - val_accuracy: 0.9250 - val_loss: 0.3029\n",
      "Epoch 93/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 299ms/step - accuracy: 0.9733 - loss: 0.1122 - val_accuracy: 0.9167 - val_loss: 0.3393\n",
      "Epoch 94/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 300ms/step - accuracy: 0.9492 - loss: 0.1122 - val_accuracy: 0.9417 - val_loss: 0.2622\n",
      "Epoch 95/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 298ms/step - accuracy: 0.8963 - loss: 0.2642 - val_accuracy: 0.7583 - val_loss: 0.6071\n",
      "Epoch 96/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 295ms/step - accuracy: 0.8757 - loss: 0.2742 - val_accuracy: 0.7333 - val_loss: 0.8492\n",
      "Epoch 97/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 298ms/step - accuracy: 0.9120 - loss: 0.2454 - val_accuracy: 0.8667 - val_loss: 0.3984\n",
      "Epoch 98/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 302ms/step - accuracy: 0.9461 - loss: 0.1538 - val_accuracy: 0.9250 - val_loss: 0.2855\n",
      "Epoch 99/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 299ms/step - accuracy: 0.9671 - loss: 0.1263 - val_accuracy: 0.9000 - val_loss: 0.3662\n",
      "Epoch 100/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 292ms/step - accuracy: 0.9328 - loss: 0.1588 - val_accuracy: 0.8333 - val_loss: 0.3993\n",
      "Epoch 101/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 296ms/step - accuracy: 0.9163 - loss: 0.2235 - val_accuracy: 0.8917 - val_loss: 0.4121\n",
      "Epoch 102/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 314ms/step - accuracy: 0.7230 - loss: 1.0635 - val_accuracy: 0.5750 - val_loss: 1.4006\n",
      "Epoch 103/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 315ms/step - accuracy: 0.6993 - loss: 0.7849 - val_accuracy: 0.7750 - val_loss: 0.6611\n",
      "Epoch 104/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 294ms/step - accuracy: 0.8987 - loss: 0.3380 - val_accuracy: 0.8667 - val_loss: 0.4814\n",
      "Epoch 105/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 301ms/step - accuracy: 0.9469 - loss: 0.2217 - val_accuracy: 0.9083 - val_loss: 0.3557\n",
      "Epoch 106/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 292ms/step - accuracy: 0.9556 - loss: 0.1639 - val_accuracy: 0.8833 - val_loss: 0.3736\n",
      "Epoch 107/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 302ms/step - accuracy: 0.8969 - loss: 0.2141 - val_accuracy: 0.9083 - val_loss: 0.3205\n",
      "Epoch 108/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 310ms/step - accuracy: 0.9563 - loss: 0.1289 - val_accuracy: 0.8500 - val_loss: 0.4579\n",
      "Epoch 109/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 311ms/step - accuracy: 0.9650 - loss: 0.1207 - val_accuracy: 0.8500 - val_loss: 0.4611\n",
      "Epoch 110/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 310ms/step - accuracy: 0.9336 - loss: 0.1908 - val_accuracy: 0.9083 - val_loss: 0.3312\n",
      "Epoch 111/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 295ms/step - accuracy: 0.9661 - loss: 0.1297 - val_accuracy: 0.8500 - val_loss: 0.5268\n",
      "Epoch 112/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 302ms/step - accuracy: 0.9043 - loss: 0.2176 - val_accuracy: 0.9000 - val_loss: 0.3521\n",
      "Epoch 113/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 303ms/step - accuracy: 0.8688 - loss: 0.4057 - val_accuracy: 0.7917 - val_loss: 0.6372\n",
      "Epoch 114/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 292ms/step - accuracy: 0.8616 - loss: 0.3337 - val_accuracy: 0.8667 - val_loss: 0.4748\n",
      "Epoch 115/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 291ms/step - accuracy: 0.8972 - loss: 0.2296 - val_accuracy: 0.9000 - val_loss: 0.4543\n",
      "Epoch 116/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 304ms/step - accuracy: 0.9537 - loss: 0.1637 - val_accuracy: 0.8917 - val_loss: 0.3229\n",
      "Epoch 117/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 315ms/step - accuracy: 0.9633 - loss: 0.1570 - val_accuracy: 0.8167 - val_loss: 0.4700\n",
      "Epoch 118/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 301ms/step - accuracy: 0.9152 - loss: 0.2394 - val_accuracy: 0.8500 - val_loss: 0.5202\n",
      "Epoch 119/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 298ms/step - accuracy: 0.9378 - loss: 0.1731 - val_accuracy: 0.7833 - val_loss: 0.8993\n",
      "Epoch 120/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 297ms/step - accuracy: 0.8970 - loss: 0.3255 - val_accuracy: 0.8833 - val_loss: 0.4408\n",
      "Epoch 121/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 309ms/step - accuracy: 0.9540 - loss: 0.1239 - val_accuracy: 0.8917 - val_loss: 0.3416\n",
      "Epoch 122/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 293ms/step - accuracy: 0.9755 - loss: 0.0937 - val_accuracy: 0.9083 - val_loss: 0.3301\n",
      "Epoch 123/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 296ms/step - accuracy: 0.9765 - loss: 0.0808 - val_accuracy: 0.8833 - val_loss: 0.3252\n",
      "Epoch 124/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 299ms/step - accuracy: 0.9818 - loss: 0.0741 - val_accuracy: 0.9333 - val_loss: 0.3268\n",
      "Epoch 125/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 321ms/step - accuracy: 0.9820 - loss: 0.0756 - val_accuracy: 0.9167 - val_loss: 0.3616\n",
      "Epoch 126/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 299ms/step - accuracy: 0.9781 - loss: 0.0726 - val_accuracy: 0.8917 - val_loss: 0.3342\n",
      "Epoch 127/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 310ms/step - accuracy: 0.9541 - loss: 0.1285 - val_accuracy: 0.8833 - val_loss: 0.4166\n",
      "Epoch 128/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 297ms/step - accuracy: 0.9443 - loss: 0.1485 - val_accuracy: 0.8083 - val_loss: 0.5350\n",
      "Epoch 129/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 305ms/step - accuracy: 0.9129 - loss: 0.1906 - val_accuracy: 0.8500 - val_loss: 0.4025\n",
      "Epoch 130/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 305ms/step - accuracy: 0.9506 - loss: 0.1300 - val_accuracy: 0.8750 - val_loss: 0.4173\n",
      "Epoch 131/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 311ms/step - accuracy: 0.9542 - loss: 0.1701 - val_accuracy: 0.9250 - val_loss: 0.3026\n",
      "Epoch 132/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 301ms/step - accuracy: 0.9440 - loss: 0.1352 - val_accuracy: 0.9167 - val_loss: 0.3106\n",
      "Epoch 133/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 297ms/step - accuracy: 0.9684 - loss: 0.0826 - val_accuracy: 0.9083 - val_loss: 0.3162\n",
      "Epoch 134/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 319ms/step - accuracy: 0.9330 - loss: 0.1664 - val_accuracy: 0.9417 - val_loss: 0.2627\n",
      "Epoch 135/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 324ms/step - accuracy: 0.9856 - loss: 0.0535 - val_accuracy: 0.9250 - val_loss: 0.3138\n",
      "Epoch 136/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 408ms/step - accuracy: 0.9897 - loss: 0.0436 - val_accuracy: 0.9417 - val_loss: 0.2692\n",
      "Epoch 137/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 357ms/step - accuracy: 0.9843 - loss: 0.0486 - val_accuracy: 0.8750 - val_loss: 0.3594\n",
      "Epoch 138/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 429ms/step - accuracy: 0.9578 - loss: 0.1080 - val_accuracy: 0.9417 - val_loss: 0.2838\n",
      "Epoch 139/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 412ms/step - accuracy: 0.9918 - loss: 0.0482 - val_accuracy: 0.9250 - val_loss: 0.3034\n",
      "Epoch 140/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 389ms/step - accuracy: 0.9870 - loss: 0.0593 - val_accuracy: 0.8833 - val_loss: 0.3558\n",
      "Epoch 141/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 421ms/step - accuracy: 0.9490 - loss: 0.1704 - val_accuracy: 0.8833 - val_loss: 0.4456\n",
      "Epoch 142/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 402ms/step - accuracy: 0.9199 - loss: 0.2226 - val_accuracy: 0.8250 - val_loss: 0.6314\n",
      "Epoch 143/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 397ms/step - accuracy: 0.9315 - loss: 0.1952 - val_accuracy: 0.8750 - val_loss: 0.4210\n",
      "Epoch 144/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 351ms/step - accuracy: 0.9251 - loss: 0.2521 - val_accuracy: 0.8750 - val_loss: 0.4003\n",
      "Epoch 145/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 353ms/step - accuracy: 0.9607 - loss: 0.1227 - val_accuracy: 0.8750 - val_loss: 0.4408\n",
      "Epoch 146/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 337ms/step - accuracy: 0.8946 - loss: 0.2275 - val_accuracy: 0.9167 - val_loss: 0.3388\n",
      "Epoch 147/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 331ms/step - accuracy: 0.9623 - loss: 0.1324 - val_accuracy: 0.9250 - val_loss: 0.3296\n",
      "Epoch 148/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 358ms/step - accuracy: 0.9659 - loss: 0.1399 - val_accuracy: 0.8750 - val_loss: 0.4240\n",
      "Epoch 149/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 361ms/step - accuracy: 0.9364 - loss: 0.1782 - val_accuracy: 0.8667 - val_loss: 0.3560\n",
      "Epoch 150/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 326ms/step - accuracy: 0.9386 - loss: 0.1419 - val_accuracy: 0.8167 - val_loss: 0.6525\n",
      "Epoch 151/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 333ms/step - accuracy: 0.9796 - loss: 0.1055 - val_accuracy: 0.8917 - val_loss: 0.4229\n",
      "Epoch 152/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 334ms/step - accuracy: 0.9689 - loss: 0.1084 - val_accuracy: 0.8583 - val_loss: 0.5385\n",
      "Epoch 153/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 366ms/step - accuracy: 0.9386 - loss: 0.1722 - val_accuracy: 0.9250 - val_loss: 0.3381\n",
      "Epoch 154/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 345ms/step - accuracy: 0.9569 - loss: 0.1215 - val_accuracy: 0.8833 - val_loss: 0.4678\n",
      "Epoch 155/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 348ms/step - accuracy: 0.9206 - loss: 0.1793 - val_accuracy: 0.8917 - val_loss: 0.4525\n",
      "Epoch 156/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 349ms/step - accuracy: 0.9491 - loss: 0.1221 - val_accuracy: 0.9250 - val_loss: 0.3467\n",
      "Epoch 157/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 352ms/step - accuracy: 0.9763 - loss: 0.0900 - val_accuracy: 0.9333 - val_loss: 0.2861\n",
      "Epoch 158/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 382ms/step - accuracy: 0.9505 - loss: 0.1307 - val_accuracy: 0.8417 - val_loss: 0.5818\n",
      "Epoch 159/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 334ms/step - accuracy: 0.7851 - loss: 0.6358 - val_accuracy: 0.7583 - val_loss: 0.7587\n",
      "Epoch 160/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 313ms/step - accuracy: 0.8176 - loss: 0.5035 - val_accuracy: 0.8417 - val_loss: 0.6146\n",
      "Epoch 161/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 310ms/step - accuracy: 0.9391 - loss: 0.2107 - val_accuracy: 0.9333 - val_loss: 0.3432\n",
      "Epoch 162/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 333ms/step - accuracy: 0.9486 - loss: 0.1480 - val_accuracy: 0.9250 - val_loss: 0.3173\n",
      "Epoch 163/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 333ms/step - accuracy: 0.9770 - loss: 0.0934 - val_accuracy: 0.9000 - val_loss: 0.3891\n",
      "Epoch 164/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 349ms/step - accuracy: 0.9850 - loss: 0.0871 - val_accuracy: 0.9250 - val_loss: 0.3286\n",
      "Epoch 165/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 334ms/step - accuracy: 0.9744 - loss: 0.0806 - val_accuracy: 0.8917 - val_loss: 0.3459\n",
      "Epoch 166/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 306ms/step - accuracy: 0.9629 - loss: 0.0849 - val_accuracy: 0.9250 - val_loss: 0.3555\n",
      "Epoch 167/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 305ms/step - accuracy: 0.9866 - loss: 0.0773 - val_accuracy: 0.9250 - val_loss: 0.3551\n",
      "Epoch 168/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 310ms/step - accuracy: 0.9580 - loss: 0.1291 - val_accuracy: 0.9083 - val_loss: 0.4282\n",
      "Epoch 169/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 304ms/step - accuracy: 0.9236 - loss: 0.2226 - val_accuracy: 0.9250 - val_loss: 0.3497\n",
      "Epoch 170/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 300ms/step - accuracy: 0.9586 - loss: 0.1191 - val_accuracy: 0.8917 - val_loss: 0.4446\n",
      "Epoch 171/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 305ms/step - accuracy: 0.9651 - loss: 0.1301 - val_accuracy: 0.9167 - val_loss: 0.3572\n",
      "Epoch 172/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 307ms/step - accuracy: 0.9788 - loss: 0.0820 - val_accuracy: 0.8917 - val_loss: 0.4402\n",
      "Epoch 173/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 303ms/step - accuracy: 0.9850 - loss: 0.0570 - val_accuracy: 0.9000 - val_loss: 0.4034\n",
      "Epoch 174/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 304ms/step - accuracy: 0.9652 - loss: 0.1087 - val_accuracy: 0.9167 - val_loss: 0.3869\n",
      "Epoch 175/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 302ms/step - accuracy: 0.9886 - loss: 0.0473 - val_accuracy: 0.9250 - val_loss: 0.3612\n",
      "Epoch 176/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 306ms/step - accuracy: 0.9669 - loss: 0.1157 - val_accuracy: 0.9250 - val_loss: 0.3071\n",
      "Epoch 177/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 300ms/step - accuracy: 0.9816 - loss: 0.0587 - val_accuracy: 0.9333 - val_loss: 0.3001\n",
      "Epoch 178/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 323ms/step - accuracy: 0.9823 - loss: 0.0727 - val_accuracy: 0.8917 - val_loss: 0.4200\n",
      "Epoch 179/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 311ms/step - accuracy: 0.9825 - loss: 0.0636 - val_accuracy: 0.8583 - val_loss: 0.4722\n",
      "Epoch 180/180\n",
      "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 305ms/step - accuracy: 0.9544 - loss: 0.1033 - val_accuracy: 0.8167 - val_loss: 0.7065\n"
     ]
    }
   ],
   "source": [
    "history = model.fit(\n",
    "                    X_train,\n",
    "                    y_train,\n",
    "                    epochs=180,\n",
    "                    batch_size=32,\n",
    "                    validation_split=0.2,\n",
    "                    verbose=1\n",
    "                    )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x600 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# Plot training & validation accuracy\n",
    "plt.figure(figsize=(12, 6))\n",
    "\n",
    "# Accuracy plot\n",
    "plt.subplot(1, 2, 1)\n",
    "plt.plot(history.history['accuracy'], label='Training Accuracy')\n",
    "plt.plot(history.history['val_accuracy'], label='Validation Accuracy')\n",
    "plt.title('Model Accuracy')\n",
    "plt.xlabel('Epochs')\n",
    "plt.ylabel('Accuracy')\n",
    "plt.legend(loc='upper left')\n",
    "\n",
    "# Loss plot\n",
    "plt.subplot(1, 2, 2)\n",
    "plt.plot(history.history['loss'], label='Training Loss')\n",
    "plt.plot(history.history['val_loss'], label='Validation Loss')\n",
    "plt.title('Model Loss')\n",
    "plt.xlabel('Epochs')\n",
    "plt.ylabel('Loss')\n",
    "plt.legend(loc='upper left')\n",
    "\n",
    "# Show the plots\n",
    "plt.tight_layout()\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:You are saving your model as an HDF5 file via `model.save()` or `keras.saving.save_model(model)`. This file format is considered legacy. We recommend using instead the native Keras format, e.g. `model.save('my_model.keras')` or `keras.saving.save_model(model, 'my_model.keras')`. \n"
     ]
    }
   ],
   "source": [
    "model.save('model_25_2.h5')\n",
    "model.save_weights('weights_25_2.weights.h5')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[1m9/9\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 63ms/step\n"
     ]
    }
   ],
   "source": [
    "labels_predict = model.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Decoding\n",
    "labels_predict = np.argmax(labels_predict , axis=1)\n",
    "labels_test_normal = np.argmax(y_test , axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((259,), (259,))"
      ]
     },
     "execution_count": 96,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "labels_test_normal.shape , labels_predict.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  1,  1,  1,  1,  1,  1,  1,\n",
       "        1,  1,  1,  2,  2,  2,  2,  2,  2,  2,  2,  2,  2,  3,  3,  3,  3,\n",
       "        3,  3,  3,  3,  3,  3,  4,  4,  4,  4,  4,  4,  4,  4,  4,  4,  5,\n",
       "        5,  5,  5,  5,  5,  5,  5,  5,  5, 19,  6,  6,  6,  6,  6,  6,  6,\n",
       "        6,  6,  6,  7,  7,  7,  7,  7,  7,  7,  7,  7,  7,  8,  8,  8,  8,\n",
       "        8,  8,  8,  8,  8,  8,  9,  9,  9,  9,  9,  9,  9,  9,  9,  9, 10,\n",
       "       10, 10, 10, 10, 10, 10, 10, 10, 10, 11, 11, 11, 11, 11, 11, 11, 11,\n",
       "       11, 11, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 13, 13, 13, 13, 13,\n",
       "       13, 13, 13, 13, 13, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14,\n",
       "       14, 14, 15, 15, 15, 15, 15, 15, 15, 15, 15, 14, 16, 16, 16, 16, 16,\n",
       "       16, 16, 16, 16, 16, 16, 16, 16, 17, 17, 17, 17, 17, 17, 17, 17, 17,\n",
       "       17, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 19, 19, 19, 19, 19, 19,\n",
       "       19, 19, 19, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 21, 21, 21,\n",
       "       21, 21, 21, 21, 21, 21, 21, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22,\n",
       "       22, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 24, 24, 24, 24, 24, 24,\n",
       "       24, 24, 24, 24])"
      ]
     },
     "execution_count": 97,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "labels_predict"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy Score is :  0.9884169884169884\n"
     ]
    }
   ],
   "source": [
    "#Accuracy Score\n",
    "from sklearn.metrics import accuracy_score\n",
    "AccScore = accuracy_score(labels_predict, labels_test_normal)\n",
    "print('Accuracy Score is : ', AccScore)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 2000x1900 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import seaborn as sns\n",
    "from sklearn.metrics import confusion_matrix\n",
    "import matplotlib.pyplot as plt\n",
    "import matplotlib\n",
    "from matplotlib import rc\n",
    "\n",
    "# ตั้งค่า Thai Font (ตัวอย่างใช้ฟอนต์ TH Sarabun New)\n",
    "thai_font_path = \"/Users/theerat/Documents/tsl/sign-language-translator/Datasets/SukhumvitSet-Medium.ttf\"  # หรือพาธที่ตรงกับระบบของคุณ\n",
    "font_prop = matplotlib.font_manager.FontProperties(fname=thai_font_path)\n",
    "matplotlib.rcParams['font.family'] = font_prop.get_name()\n",
    "\n",
    "class_labels = words\n",
    "\n",
    "\n",
    "cm = confusion_matrix(labels_test_normal, labels_predict)\n",
    "\n",
    "fig, ax = plt.subplots(figsize=(20, 19))\n",
    "sns.heatmap(cm, annot=True, fmt='g', ax=ax, cmap=\"Blues\")\n",
    "\n",
    "ax.set_xlabel('Predicted labels', fontsize=14)\n",
    "ax.set_ylabel('True labels', fontsize=14)\n",
    "ax.set_title('TSL Confusion Matrix', fontsize=16)\n",
    "\n",
    "ax.set_xticklabels(class_labels, rotation=0, fontsize=12)\n",
    "ax.set_yticklabels(class_labels, rotation=0, fontsize=12)\n",
    "\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 100,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[10  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0\n",
      "   0]\n",
      " [ 0 10  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0\n",
      "   0]\n",
      " [ 0  0 10  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0\n",
      "   0]\n",
      " [ 0  0  0 10  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0\n",
      "   0]\n",
      " [ 0  0  0  0 10  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0\n",
      "   0]\n",
      " [ 0  0  0  0  0 10  0  0  0  0  0  0  0  0  0  0  0  0  0  1  0  0  0  0\n",
      "   0]\n",
      " [ 0  0  0  0  0  0 10  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0\n",
      "   0]\n",
      " [ 0  0  0  0  0  0  0 10  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0\n",
      "   0]\n",
      " [ 0  0  0  0  0  0  0  0 10  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0\n",
      "   0]\n",
      " [ 0  0  0  0  0  0  0  0  0 10  0  0  0  0  0  0  0  0  0  0  0  0  0  0\n",
      "   0]\n",
      " [ 0  0  0  0  0  0  0  0  0  0 10  0  0  0  0  0  0  0  0  0  0  0  0  0\n",
      "   0]\n",
      " [ 0  0  0  0  0  0  0  0  0  0  0 10  0  0  0  0  0  0  0  0  0  0  0  0\n",
      "   0]\n",
      " [ 0  0  0  0  0  0  0  0  0  0  0  0 10  0  0  0  0  0  0  0  0  0  0  0\n",
      "   0]\n",
      " [ 0  0  0  0  0  0  0  0  0  0  0  0  0 10  0  0  0  0  0  0  0  0  0  0\n",
      "   0]\n",
      " [ 0  0  0  0  0  0  0  0  0  0  0  0  0  0 14  0  0  0  0  0  0  0  0  0\n",
      "   0]\n",
      " [ 0  0  0  0  0  0  0  0  0  0  0  0  0  0  1  9  0  0  0  0  0  0  0  0\n",
      "   0]\n",
      " [ 0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0 13  0  0  0  0  0  0  0\n",
      "   0]\n",
      " [ 0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0 10  0  0  0  0  0  0\n",
      "   0]\n",
      " [ 0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0 10  0  0  0  0  0\n",
      "   0]\n",
      " [ 0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  9  1  0  0  0\n",
      "   0]\n",
      " [ 0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0 10  0  0  0\n",
      "   0]\n",
      " [ 0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0 10  0  0\n",
      "   0]\n",
      " [ 0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0 11  0\n",
      "   0]\n",
      " [ 0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0 10\n",
      "   0]\n",
      " [ 0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0\n",
      "  10]]\n"
     ]
    }
   ],
   "source": [
    "print(cm)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 101,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Classification Report is : \n",
      "               precision    recall  f1-score   support\n",
      "\n",
      "           0       1.00      1.00      1.00        10\n",
      "           1       1.00      1.00      1.00        10\n",
      "           2       1.00      1.00      1.00        10\n",
      "           3       1.00      1.00      1.00        10\n",
      "           4       1.00      1.00      1.00        10\n",
      "           5       1.00      0.91      0.95        11\n",
      "           6       1.00      1.00      1.00        10\n",
      "           7       1.00      1.00      1.00        10\n",
      "           8       1.00      1.00      1.00        10\n",
      "           9       1.00      1.00      1.00        10\n",
      "          10       1.00      1.00      1.00        10\n",
      "          11       1.00      1.00      1.00        10\n",
      "          12       1.00      1.00      1.00        10\n",
      "          13       1.00      1.00      1.00        10\n",
      "          14       0.93      1.00      0.97        14\n",
      "          15       1.00      0.90      0.95        10\n",
      "          16       1.00      1.00      1.00        13\n",
      "          17       1.00      1.00      1.00        10\n",
      "          18       1.00      1.00      1.00        10\n",
      "          19       0.90      0.90      0.90        10\n",
      "          20       0.91      1.00      0.95        10\n",
      "          21       1.00      1.00      1.00        10\n",
      "          22       1.00      1.00      1.00        11\n",
      "          23       1.00      1.00      1.00        10\n",
      "          24       1.00      1.00      1.00        10\n",
      "\n",
      "    accuracy                           0.99       259\n",
      "   macro avg       0.99      0.99      0.99       259\n",
      "weighted avg       0.99      0.99      0.99       259\n",
      "\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import classification_report\n",
    "\n",
    "ClassificationReport = classification_report(labels_test_normal,labels_predict)\n",
    "print('Classification Report is : \\n', ClassificationReport)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'หนังสือ': 0,\n",
       " 'อยู่': 1,\n",
       " 'พูด': 2,\n",
       " 'ไก่ผัดกะเพรา': 3,\n",
       " 'ชอบ': 4,\n",
       " 'หัวเราะ': 5,\n",
       " 'ควาย': 6,\n",
       " 'ฟัง': 7,\n",
       " 'ดื่ม': 8,\n",
       " 'คุณ': 9,\n",
       " 'นอน': 10,\n",
       " 'เขา': 11,\n",
       " 'กิน': 12,\n",
       " 'โรงเรียน': 13,\n",
       " 'ฉัน': 14,\n",
       " 'นักเรียน': 15,\n",
       " 'วิ่ง': 16,\n",
       " 'เดิน': 17}"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "label_map"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#คำที่มีปัญหา\n",
    "error_words = ['หัวเราะ','นอน','โรงเรียน','ฉัน']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "19"
      ]
     },
     "execution_count": 70,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(cm)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# __Test__"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 235,
   "metadata": {},
   "outputs": [],
   "source": [
    "from scipy import stats\n",
    "import cv2"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# __Transformer__"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# MatplotLib Global Settings\n",
    "mpl.rcParams.update(mpl.rcParamsDefault)\n",
    "mpl.rcParams['xtick.labelsize'] = 16\n",
    "mpl.rcParams['ytick.labelsize'] = 16\n",
    "mpl.rcParams['axes.labelsize'] = 18\n",
    "mpl.rcParams['axes.titlesize'] = 24"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "# If True, processing data from scratch\n",
    "# If False, loads preprocessed data\n",
    "PREPROCESS_DATA = False\n",
    "TRAIN_MODEL = True\n",
    "# True: use 10% of participants as validation set\n",
    "# False: use all data for training -> gives better LB result\n",
    "USE_VAL = False\n",
    "\n",
    "N_ROWS = 543\n",
    "N_DIMS = 3\n",
    "DIM_NAMES = ['x', 'y', 'z']\n",
    "SEED = 42\n",
    "NUM_CLASSES = 100\n",
    "IS_INTERACTIVE = True\n",
    "VERBOSE = 1 if IS_INTERACTIVE else 2\n",
    "\n",
    "INPUT_SIZE = 64\n",
    "\n",
    "BATCH_ALL_SIGNS_N = 4\n",
    "BATCH_SIZE = 32\n",
    "N_EPOCHS = 100\n",
    "LR_MAX = 1e-3\n",
    "N_WARMUP_EPOCHS = 0\n",
    "WD_RATIO = 0.05\n",
    "MASK_VAL = 4237"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "def print_shape_dtype(l, names):\n",
    "    for e, n in zip(l, names):\n",
    "        print(f'{n} shape: {e.shape}, dtype: {e.dtype}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "#for csv file\n",
    "import os\n",
    "import pandas as pd\n",
    "\n",
    "folder_path = '/Users/theerat/Downloads/บ้าน-2'\n",
    "data = []\n",
    "\n",
    "for subfolder in os.listdir(folder_path):\n",
    "    subfolder_path = os.path.join(folder_path, subfolder)\n",
    "    \n",
    "    if os.path.isdir(subfolder_path):\n",
    "        for filename in os.listdir(subfolder_path):\n",
    "            if filename.endswith('.csv'):\n",
    "                file_path = os.path.join(subfolder_path, filename)\n",
    "                sequence_id = int(filename.split('_')[1].split('.')[0])  # ได้ sequence_id จากชื่อไฟล์\n",
    "                participant_id = 70\n",
    "                sign = 'บ้าน'\n",
    "\n",
    "                # เพิ่มข้อมูลลงใน list\n",
    "                data.append({\n",
    "                    'path': file_path,\n",
    "                    'participant_id': participant_id,\n",
    "                    'sequence_id': sequence_id,\n",
    "                    'sign': sign\n",
    "                })\n",
    "\n",
    "# สร้าง DataFrame จากข้อมูลที่เก็บรวบรวม\n",
    "df = pd.DataFrame(data)\n",
    "\n",
    "df = df.sort_values(by='sequence_id')\n",
    "\n",
    "# บันทึกเป็นไฟล์ CSV\n",
    "df.to_csv('output.csv', index=False)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "\n",
    "folder_path = '/Users/theerat/Downloads/MP_DATA3/คุย'\n",
    "data = []\n",
    "participant_id = 0\n",
    "\n",
    "# อ่านโฟลเดอร์ทั้งหมดภายใน\n",
    "for subfolder in os.listdir(folder_path):\n",
    "    subfolder_path = os.path.join(folder_path, subfolder)\n",
    "    \n",
    "    if os.path.isdir(subfolder_path):\n",
    "        # อ่านไฟล์ .npy ภายในแต่ละโฟลเดอร์\n",
    "        for filename in os.listdir(subfolder_path):\n",
    "            if filename.endswith('.npy'):\n",
    "                file_path = os.path.join(subfolder_path, filename)\n",
    "                sequence_id = int(filename.split('.')[0])  # ได้ sequence_id จากชื่อไฟล์\n",
    "                sign = 'ก'\n",
    "\n",
    "                # เพิ่มข้อมูลลงใน list\n",
    "                data.append({\n",
    "                    'path': file_path,\n",
    "                    'participant_id': participant_id,\n",
    "                    'sequence_id': sequence_id,\n",
    "                    'sign': sign\n",
    "                })\n",
    "                participant_id += 1\n",
    "\n",
    "\n",
    "# สร้าง DataFrame จากข้อมูลที่เก็บรวบรวม\n",
    "df = pd.DataFrame(data)\n",
    "\n",
    "df = df.sort_values(by='sequence_id')\n",
    "# บันทึกเป็นไฟล์ CSV\n",
    "df.to_csv('outputnpy.csv', index=False)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>path</th>\n",
       "      <th>participant_id</th>\n",
       "      <th>sequence_id</th>\n",
       "      <th>sign</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>/Users/theerat/Downloads/MP_DATA3/คุย/17/0.npy</td>\n",
       "      <td>76</td>\n",
       "      <td>0</td>\n",
       "      <td>ก</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>/Users/theerat/Downloads/MP_DATA3/คุย/15/0.npy</td>\n",
       "      <td>166</td>\n",
       "      <td>0</td>\n",
       "      <td>ก</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>/Users/theerat/Downloads/MP_DATA3/คุย/3/0.npy</td>\n",
       "      <td>196</td>\n",
       "      <td>0</td>\n",
       "      <td>ก</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>/Users/theerat/Downloads/MP_DATA3/คุย/4/0.npy</td>\n",
       "      <td>136</td>\n",
       "      <td>0</td>\n",
       "      <td>ก</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>/Users/theerat/Downloads/MP_DATA3/คุย/2/0.npy</td>\n",
       "      <td>226</td>\n",
       "      <td>0</td>\n",
       "      <td>ก</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>235</th>\n",
       "      <td>/Users/theerat/Downloads/MP_DATA3/คุย/4/29.npy</td>\n",
       "      <td>140</td>\n",
       "      <td>29</td>\n",
       "      <td>ก</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>236</th>\n",
       "      <td>/Users/theerat/Downloads/MP_DATA3/คุย/29/29.npy</td>\n",
       "      <td>50</td>\n",
       "      <td>29</td>\n",
       "      <td>ก</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>237</th>\n",
       "      <td>/Users/theerat/Downloads/MP_DATA3/คุย/3/29.npy</td>\n",
       "      <td>200</td>\n",
       "      <td>29</td>\n",
       "      <td>ก</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>238</th>\n",
       "      <td>/Users/theerat/Downloads/MP_DATA3/คุย/15/29.npy</td>\n",
       "      <td>170</td>\n",
       "      <td>29</td>\n",
       "      <td>ก</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>239</th>\n",
       "      <td>/Users/theerat/Downloads/MP_DATA3/คุย/18/29.npy</td>\n",
       "      <td>20</td>\n",
       "      <td>29</td>\n",
       "      <td>ก</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>240 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                path  participant_id  \\\n",
       "0     /Users/theerat/Downloads/MP_DATA3/คุย/17/0.npy              76   \n",
       "1     /Users/theerat/Downloads/MP_DATA3/คุย/15/0.npy             166   \n",
       "2      /Users/theerat/Downloads/MP_DATA3/คุย/3/0.npy             196   \n",
       "3      /Users/theerat/Downloads/MP_DATA3/คุย/4/0.npy             136   \n",
       "4      /Users/theerat/Downloads/MP_DATA3/คุย/2/0.npy             226   \n",
       "..                                               ...             ...   \n",
       "235   /Users/theerat/Downloads/MP_DATA3/คุย/4/29.npy             140   \n",
       "236  /Users/theerat/Downloads/MP_DATA3/คุย/29/29.npy              50   \n",
       "237   /Users/theerat/Downloads/MP_DATA3/คุย/3/29.npy             200   \n",
       "238  /Users/theerat/Downloads/MP_DATA3/คุย/15/29.npy             170   \n",
       "239  /Users/theerat/Downloads/MP_DATA3/คุย/18/29.npy              20   \n",
       "\n",
       "     sequence_id sign  \n",
       "0              0    ก  \n",
       "1              0    ก  \n",
       "2              0    ก  \n",
       "3              0    ก  \n",
       "4              0    ก  \n",
       "..           ...  ...  \n",
       "235           29    ก  \n",
       "236           29    ก  \n",
       "237           29    ก  \n",
       "238           29    ก  \n",
       "239           29    ก  \n",
       "\n",
       "[240 rows x 4 columns]"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.read_csv('outputnpy.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "word_list = { \"สวัสดี\": 0,\n",
    " \"ตก\": 1,\n",
    " \"รอ\": 2,\n",
    " \"กลับ\": 3,\n",
    " \"ขอบคุณ\": 4,\n",
    " \"ตัด\": 5,\n",
    " \"ลง\": 6,\n",
    " \"ขึ้น\": 7,\n",
    " \"เฝ้า\": 8,\n",
    " \"คุย\": 9,\n",
    " \"ช่วย\": 10,\n",
    " \"เชื่อ\": 11,\n",
    " \"ฟัง\": 12,\n",
    " \"มอง\": 13,\n",
    " \"พา\": 14,\n",
    " \"ชวน หลีกภัย\": 15,\n",
    " \"ทักษิณ ชินวัตร\": 16,\n",
    " \"ขนม\": 17,\n",
    " \"พิธา ลิ้มเจริญรัตน์\": 18,\n",
    " \"ศิริกัญญา ตันสกุล\": 19,\n",
    " \"แบก\": 20,\n",
    " \"อนุทิน ชาญวีรกูล\": 21,\n",
    " \"รังสิมันต์ โรม\": 22,\n",
    " \"พีระพันธุ์ สาลีรัฐวิภาค\": 23,\n",
    " \"คุณหญิงสุดารัตน์ เกยุราพันธุ์\": 24,\n",
    " \"พลเอกอนุพงษ์ เผ่าจินดา\": 25,\n",
    " \"สุวัจน์ ลิปตพัลลภ\": 26,\n",
    " \"กรณ์ จาติกวณิช\": 27,\n",
    " \"วราวุธ ศิลปะอาชา\": 28,\n",
    " \"พล.ต.อ.เสรีพิศุทธ์ เตมียเวสวราวุธ ศิลปะอาชา\": 29,\n",
    " \"ศักดิ์สยาม ชิดชอบ\": 30,\n",
    " \"ชาดา ไทยเศรษฐ์\": 31,\n",
    " \"สุชัชวีร์ สุวรรณสวัสดิ์\": 32,\n",
    " \"จุรินทร์ ลักษณวิศิษฏ์\": 33,\n",
    " \"ไตรรงค์ สุวรรณคีรี\": 34,\n",
    " \"พลเอกประยุทธ์ จันทร์โอชา\": 35,\n",
    " \"นฤมล ภิญโญสินวัฒน์\": 36,\n",
    " \"ธรรมนัส พรหมเผ่า\": 37,\n",
    " \"ชัยวุฒิ ธนาคมานุสรณ์\": 38,\n",
    " \"ไพบูลย์ นิติตะวัน\": 39,\n",
    " \"พลเอกประวิตร วงษ์สุวรรณ\": 40,\n",
    " \"พริษฐ์ วัชรสินธุ\": 41,\n",
    " \"ยิ่งลักษณ์ ชินวัตร\": 42,\n",
    " \"ณัฐวุฒิ ใสยเกื้อ\": 43,\n",
    " \"นพ.ชลน่าน ศรีแก้ว\": 44,\n",
    " \"เศรษฐา ทวีสิน\": 45,\n",
    " \"แพทองธาร ชินวัตร\": 46,\n",
    " \"อะไร\": 47,\n",
    " \"สมัคร\": 48,\n",
    " \"กระโดด\": 49,\n",
    " \"ยก\": 50,\n",
    " \"ชน\": 51,\n",
    " \"ผ่าน\": 52,\n",
    " \"แนะนำ\": 53,\n",
    " \"จำได้\": 54,\n",
    " \"ปลูก\": 55,\n",
    " \"รวม\": 56,\n",
    " \"ทำหาย\": 57,\n",
    " \"เจอ\": 58,\n",
    " \"หาย\": 59,\n",
    " \"พนักงานขาย\": 60,\n",
    " \"นักเขียนโปรแกรม\": 61,\n",
    " \"พ่อครัว\": 62,\n",
    " \"เจ้าหน้าที่ตำรวจ\": 63,\n",
    " \"เนื้อลูกแกะ\": 64,\n",
    " \"เนื้อหมู\": 65,\n",
    " \"คุณสบายดีไหม\": 66,\n",
    " \"ดูเหมือน\": 67,\n",
    " \"วาง\": 68,\n",
    " \"อยู่\": 69,\n",
    " \"ก\": 70,\n",
    " \"ข\": 71,\n",
    " \"ค\": 72,\n",
    " \"ฆ\": 73,\n",
    " \"ต\": 74,\n",
    " \"ถ\": 75,\n",
    " \"ง\": 76,\n",
    " \"จ\": 77,\n",
    " \"ฉ\": 78,\n",
    " \"ช\": 79,\n",
    " \"ซ\": 80,\n",
    " \"ฏ\": 81,\n",
    " \"ฐ\": 82,\n",
    " \"ฑ\": 83,\n",
    " \"ฒ\": 84,\n",
    " \"ณ\": 85,\n",
    " \"ด\": 86,\n",
    " \"ท\": 87,\n",
    " \"ธ\": 88,\n",
    " \"น\": 89,\n",
    " \"บ\": 90,\n",
    " \"ป\": 91,\n",
    " \"ผ\": 92,\n",
    " \"ฝ\": 93,\n",
    " \"พ\": 94,\n",
    " \"ฟ\": 95,\n",
    " \"ภ\": 96,\n",
    " \"ม\": 97,\n",
    " \"ย\": 98,\n",
    " \"ร\": 99,\n",
    " \"ล\": 100,\n",
    " \"ว\": 101,\n",
    " \"ศ\": 102,\n",
    " \"ษ\": 103,\n",
    " \"ส\": 104,\n",
    " \"ห\": 105,\n",
    " \"ฬ\": 106,\n",
    " \"อ\": 107,\n",
    " \"ฮ\": 108,\n",
    " \"ญ\": 109,\n",
    " \"ฎ\": 110\n",
    "}\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี\n",
      "/Users/theerat/Downloads/MP_DATA3/ตก\n",
      "/Users/theerat/Downloads/MP_DATA3/รอ\n",
      "/Users/theerat/Downloads/MP_DATA3/กลับ\n",
      "/Users/theerat/Downloads/MP_DATA3/ขอบคุณ\n",
      "/Users/theerat/Downloads/MP_DATA3/ตัด\n",
      "/Users/theerat/Downloads/MP_DATA3/ลง\n",
      "/Users/theerat/Downloads/MP_DATA3/ขึ้น\n",
      "/Users/theerat/Downloads/MP_DATA3/เฝ้า\n",
      "/Users/theerat/Downloads/MP_DATA3/คุย\n",
      "/Users/theerat/Downloads/MP_DATA3/ช่วย\n",
      "/Users/theerat/Downloads/MP_DATA3/เชื่อ\n",
      "/Users/theerat/Downloads/MP_DATA3/ฟัง\n",
      "/Users/theerat/Downloads/MP_DATA3/มอง\n",
      "/Users/theerat/Downloads/MP_DATA3/พา\n",
      "/Users/theerat/Downloads/MP_DATA3/ชวน หลีกภัย\n",
      "/Users/theerat/Downloads/MP_DATA3/ทักษิณ ชินวัตร\n",
      "/Users/theerat/Downloads/MP_DATA3/ขนม\n",
      "/Users/theerat/Downloads/MP_DATA3/พิธา ลิ้มเจริญรัตน์\n",
      "/Users/theerat/Downloads/MP_DATA3/ศิริกัญญา ตันสกุล\n",
      "/Users/theerat/Downloads/MP_DATA3/แบก\n",
      "/Users/theerat/Downloads/MP_DATA3/อนุทิน ชาญวีรกูล\n",
      "/Users/theerat/Downloads/MP_DATA3/รังสิมันต์ โรม\n",
      "/Users/theerat/Downloads/MP_DATA3/พีระพันธุ์ สาลีรัฐวิภาค\n",
      "/Users/theerat/Downloads/MP_DATA3/คุณหญิงสุดารัตน์ เกยุราพันธุ์\n",
      "/Users/theerat/Downloads/MP_DATA3/พลเอกอนุพงษ์ เผ่าจินดา\n",
      "/Users/theerat/Downloads/MP_DATA3/สุวัจน์ ลิปตพัลลภ\n",
      "/Users/theerat/Downloads/MP_DATA3/กรณ์ จาติกวณิช\n",
      "/Users/theerat/Downloads/MP_DATA3/วราวุธ ศิลปะอาชา\n",
      "/Users/theerat/Downloads/MP_DATA3/พล.ต.อ.เสรีพิศุทธ์ เตมียเวสวราวุธ ศิลปะอาชา\n",
      "/Users/theerat/Downloads/MP_DATA3/ศักดิ์สยาม ชิดชอบ\n",
      "/Users/theerat/Downloads/MP_DATA3/ชาดา ไทยเศรษฐ์\n",
      "/Users/theerat/Downloads/MP_DATA3/สุชัชวีร์ สุวรรณสวัสดิ์\n",
      "/Users/theerat/Downloads/MP_DATA3/จุรินทร์ ลักษณวิศิษฏ์\n",
      "/Users/theerat/Downloads/MP_DATA3/ไตรรงค์ สุวรรณคีรี\n",
      "/Users/theerat/Downloads/MP_DATA3/พลเอกประยุทธ์ จันทร์โอชา\n",
      "/Users/theerat/Downloads/MP_DATA3/นฤมล ภิญโญสินวัฒน์\n",
      "/Users/theerat/Downloads/MP_DATA3/ธรรมนัส พรหมเผ่า\n",
      "/Users/theerat/Downloads/MP_DATA3/ชัยวุฒิ ธนาคมานุสรณ์\n",
      "/Users/theerat/Downloads/MP_DATA3/ไพบูลย์ นิติตะวัน\n",
      "/Users/theerat/Downloads/MP_DATA3/พลเอกประวิตร วงษ์สุวรรณ\n",
      "/Users/theerat/Downloads/MP_DATA3/พริษฐ์ วัชรสินธุ\n",
      "/Users/theerat/Downloads/MP_DATA3/ยิ่งลักษณ์ ชินวัตร\n",
      "/Users/theerat/Downloads/MP_DATA3/ณัฐวุฒิ ใสยเกื้อ\n",
      "/Users/theerat/Downloads/MP_DATA3/นพ.ชลน่าน ศรีแก้ว\n",
      "/Users/theerat/Downloads/MP_DATA3/เศรษฐา ทวีสิน\n",
      "/Users/theerat/Downloads/MP_DATA3/แพทองธาร ชินวัตร\n",
      "/Users/theerat/Downloads/MP_DATA3/อะไร\n",
      "/Users/theerat/Downloads/MP_DATA3/สมัคร\n",
      "/Users/theerat/Downloads/MP_DATA3/กระโดด\n",
      "/Users/theerat/Downloads/MP_DATA3/ยก\n",
      "/Users/theerat/Downloads/MP_DATA3/ชน\n",
      "/Users/theerat/Downloads/MP_DATA3/ผ่าน\n",
      "/Users/theerat/Downloads/MP_DATA3/แนะนำ\n",
      "/Users/theerat/Downloads/MP_DATA3/จำได้\n",
      "/Users/theerat/Downloads/MP_DATA3/ปลูก\n",
      "/Users/theerat/Downloads/MP_DATA3/รวม\n",
      "/Users/theerat/Downloads/MP_DATA3/ทำหาย\n",
      "/Users/theerat/Downloads/MP_DATA3/เจอ\n",
      "/Users/theerat/Downloads/MP_DATA3/หาย\n",
      "/Users/theerat/Downloads/MP_DATA3/พนักงานขาย\n",
      "/Users/theerat/Downloads/MP_DATA3/นักเขียนโปรแกรม\n",
      "/Users/theerat/Downloads/MP_DATA3/พ่อครัว\n",
      "/Users/theerat/Downloads/MP_DATA3/เจ้าหน้าที่ตำรวจ\n",
      "/Users/theerat/Downloads/MP_DATA3/เนื้อลูกแกะ\n",
      "/Users/theerat/Downloads/MP_DATA3/เนื้อหมู\n",
      "/Users/theerat/Downloads/MP_DATA3/คุณสบายดีไหม\n",
      "/Users/theerat/Downloads/MP_DATA3/ดูเหมือน\n",
      "/Users/theerat/Downloads/MP_DATA3/วาง\n",
      "/Users/theerat/Downloads/MP_DATA3/อยู่\n",
      "/Users/theerat/Downloads/MP_DATA3/ก\n",
      "/Users/theerat/Downloads/MP_DATA3/ข\n",
      "/Users/theerat/Downloads/MP_DATA3/ค\n",
      "/Users/theerat/Downloads/MP_DATA3/ฆ\n",
      "/Users/theerat/Downloads/MP_DATA3/ต\n",
      "/Users/theerat/Downloads/MP_DATA3/ถ\n",
      "/Users/theerat/Downloads/MP_DATA3/ง\n",
      "/Users/theerat/Downloads/MP_DATA3/จ\n",
      "/Users/theerat/Downloads/MP_DATA3/ฉ\n",
      "/Users/theerat/Downloads/MP_DATA3/ช\n",
      "/Users/theerat/Downloads/MP_DATA3/ซ\n",
      "/Users/theerat/Downloads/MP_DATA3/ฏ\n",
      "/Users/theerat/Downloads/MP_DATA3/ฐ\n",
      "/Users/theerat/Downloads/MP_DATA3/ฑ\n",
      "/Users/theerat/Downloads/MP_DATA3/ฒ\n",
      "/Users/theerat/Downloads/MP_DATA3/ณ\n",
      "/Users/theerat/Downloads/MP_DATA3/ด\n",
      "/Users/theerat/Downloads/MP_DATA3/ท\n",
      "/Users/theerat/Downloads/MP_DATA3/ธ\n",
      "/Users/theerat/Downloads/MP_DATA3/น\n",
      "/Users/theerat/Downloads/MP_DATA3/บ\n",
      "/Users/theerat/Downloads/MP_DATA3/ป\n",
      "/Users/theerat/Downloads/MP_DATA3/ผ\n",
      "/Users/theerat/Downloads/MP_DATA3/ฝ\n",
      "/Users/theerat/Downloads/MP_DATA3/พ\n",
      "/Users/theerat/Downloads/MP_DATA3/ฟ\n",
      "/Users/theerat/Downloads/MP_DATA3/ภ\n",
      "/Users/theerat/Downloads/MP_DATA3/ม\n",
      "/Users/theerat/Downloads/MP_DATA3/ย\n",
      "/Users/theerat/Downloads/MP_DATA3/ร\n",
      "/Users/theerat/Downloads/MP_DATA3/ล\n",
      "/Users/theerat/Downloads/MP_DATA3/ว\n",
      "/Users/theerat/Downloads/MP_DATA3/ศ\n",
      "/Users/theerat/Downloads/MP_DATA3/ษ\n",
      "/Users/theerat/Downloads/MP_DATA3/ส\n",
      "/Users/theerat/Downloads/MP_DATA3/ห\n",
      "/Users/theerat/Downloads/MP_DATA3/ฬ\n",
      "/Users/theerat/Downloads/MP_DATA3/อ\n",
      "/Users/theerat/Downloads/MP_DATA3/ฮ\n",
      "/Users/theerat/Downloads/MP_DATA3/ญ\n",
      "/Users/theerat/Downloads/MP_DATA3/ฎ\n"
     ]
    }
   ],
   "source": [
    "\n",
    "# กำหนด path ของโฟลเดอร์ที่ต้องการอ่าน\n",
    "base_path = '/Users/theerat/Downloads/MP_DATA3'\n",
    "# สร้างลิสต์เพื่อเก็บข้อมูล\n",
    "data = []\n",
    "\n",
    "\n",
    "# อ่านข้อมูลจากโฟลเดอร์และไฟล์\n",
    "for folder_name, participant_id in word_list.items():\n",
    "    folder_path = os.path.join(base_path, folder_name)\n",
    "    print(folder_path)\n",
    "    for subfolder in os.listdir(folder_path):\n",
    "        subfolder_path = os.path.join(folder_path, subfolder)\n",
    "        if os.path.isdir(subfolder_path):\n",
    "            for i in range(30):  # อ่านไฟล์จาก 0 ถึง 29\n",
    "                file_path = os.path.join(subfolder_path, f\"{i}.npy\")\n",
    "                try:\n",
    "                    # โหลดข้อมูลจากไฟล์ .npy\n",
    "                    data_array = np.load(file_path)\n",
    "                    sign = os.path.basename(folder_path)  # ชื่อโฟลเดอร์เป็น sign\n",
    "                    for sequence_id, value in enumerate(data_array):\n",
    "                        # เพิ่มข้อมูลลงในลิสต์\n",
    "                        data.append({\n",
    "                            'path': file_path,\n",
    "                            'participant_id': participant_id,\n",
    "                            'sequence_id': sequence_id,\n",
    "                            'sign': sign\n",
    "                        })\n",
    "                except FileNotFoundError:\n",
    "                    # ข้ามไฟล์ที่ไม่พบ\n",
    "                    pass\n",
    "\n",
    "# สร้าง DataFrame จากลิสต์ข้อมูล\n",
    "df = pd.DataFrame(data)\n",
    "df.to_csv('outputnpy.csv', index=False)\n",
    "# แสดงข้อมูล\n",
    "print(df)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>path</th>\n",
       "      <th>participant_id</th>\n",
       "      <th>sequence_id</th>\n",
       "      <th>sign</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>/Users/theerat/Downloads/MP_DATA3/สวัสดี/20/20...</td>\n",
       "      <td>0</td>\n",
       "      <td>20</td>\n",
       "      <td>สวัสดี</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>/Users/theerat/Downloads/MP_DATA3/สวัสดี/20/21...</td>\n",
       "      <td>0</td>\n",
       "      <td>21</td>\n",
       "      <td>สวัสดี</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>/Users/theerat/Downloads/MP_DATA3/สวัสดี/20/23...</td>\n",
       "      <td>0</td>\n",
       "      <td>23</td>\n",
       "      <td>สวัสดี</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>/Users/theerat/Downloads/MP_DATA3/สวัสดี/20/22...</td>\n",
       "      <td>0</td>\n",
       "      <td>22</td>\n",
       "      <td>สวัสดี</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>/Users/theerat/Downloads/MP_DATA3/สวัสดี/20/26...</td>\n",
       "      <td>0</td>\n",
       "      <td>26</td>\n",
       "      <td>สวัสดี</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36414</th>\n",
       "      <td>/Users/theerat/Downloads/MP_DATA3/ฎ/25/17.npy</td>\n",
       "      <td>110</td>\n",
       "      <td>17</td>\n",
       "      <td>ฎ</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36415</th>\n",
       "      <td>/Users/theerat/Downloads/MP_DATA3/ฎ/25/13.npy</td>\n",
       "      <td>110</td>\n",
       "      <td>13</td>\n",
       "      <td>ฎ</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36416</th>\n",
       "      <td>/Users/theerat/Downloads/MP_DATA3/ฎ/25/12.npy</td>\n",
       "      <td>110</td>\n",
       "      <td>12</td>\n",
       "      <td>ฎ</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36417</th>\n",
       "      <td>/Users/theerat/Downloads/MP_DATA3/ฎ/25/10.npy</td>\n",
       "      <td>110</td>\n",
       "      <td>10</td>\n",
       "      <td>ฎ</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36418</th>\n",
       "      <td>/Users/theerat/Downloads/MP_DATA3/ฎ/25/11.npy</td>\n",
       "      <td>110</td>\n",
       "      <td>11</td>\n",
       "      <td>ฎ</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>36419 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                    path  participant_id  \\\n",
       "0      /Users/theerat/Downloads/MP_DATA3/สวัสดี/20/20...               0   \n",
       "1      /Users/theerat/Downloads/MP_DATA3/สวัสดี/20/21...               0   \n",
       "2      /Users/theerat/Downloads/MP_DATA3/สวัสดี/20/23...               0   \n",
       "3      /Users/theerat/Downloads/MP_DATA3/สวัสดี/20/22...               0   \n",
       "4      /Users/theerat/Downloads/MP_DATA3/สวัสดี/20/26...               0   \n",
       "...                                                  ...             ...   \n",
       "36414      /Users/theerat/Downloads/MP_DATA3/ฎ/25/17.npy             110   \n",
       "36415      /Users/theerat/Downloads/MP_DATA3/ฎ/25/13.npy             110   \n",
       "36416      /Users/theerat/Downloads/MP_DATA3/ฎ/25/12.npy             110   \n",
       "36417      /Users/theerat/Downloads/MP_DATA3/ฎ/25/10.npy             110   \n",
       "36418      /Users/theerat/Downloads/MP_DATA3/ฎ/25/11.npy             110   \n",
       "\n",
       "      sequence_id    sign  \n",
       "0              20  สวัสดี  \n",
       "1              21  สวัสดี  \n",
       "2              23  สวัสดี  \n",
       "3              22  สวัสดี  \n",
       "4              26  สวัสดี  \n",
       "...           ...     ...  \n",
       "36414          17       ฎ  \n",
       "36415          13       ฎ  \n",
       "36416          12       ฎ  \n",
       "36417          10       ฎ  \n",
       "36418          11       ฎ  \n",
       "\n",
       "[36419 rows x 4 columns]"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "base_path = '/Users/theerat/Downloads/MP_DATA3'\n",
    "data = []\n",
    "id = 0\n",
    "for folder_name, participant_id in word_list.items():\n",
    "    folder_path = os.path.join(base_path, folder_name)\n",
    "    for subfolder in os.listdir(folder_path):\n",
    "        if subfolder == '.DS_Store':\n",
    "                continue\n",
    "        subfolder_path = os.path.join(folder_path, subfolder)\n",
    "        sign = os.path.basename(folder_path)\n",
    "        \n",
    "        for i in os.listdir(subfolder_path):\n",
    "            if i == '.DS_Store':\n",
    "                continue\n",
    "            npy_path = os.path.join(subfolder_path, i)\n",
    "            sequence_id = os.path.splitext(i)[0]\n",
    "            data.append({\n",
    "                'path': npy_path,\n",
    "                'participant_id': participant_id,\n",
    "                'sequence_id': sequence_id,\n",
    "                'sign': sign\n",
    "            })\n",
    "            \n",
    "            \n",
    "    id += 1\n",
    "df = pd.DataFrame(data)\n",
    "df_sorted = df.sort_values(by=['participant_id', 'sequence_id']).reset_index(drop=True)\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "new_df = pd.read_csv('train.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>index</th>\n",
       "      <th>path</th>\n",
       "      <th>participant_id</th>\n",
       "      <th>sequence_id</th>\n",
       "      <th>sign</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2389</td>\n",
       "      <td>สวัสดี\\22.parquet</td>\n",
       "      <td>22</td>\n",
       "      <td>0</td>\n",
       "      <td>สวัสดี</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2377</td>\n",
       "      <td>สวัสดี\\1.parquet</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>สวัสดี</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2378</td>\n",
       "      <td>สวัสดี\\10.parquet</td>\n",
       "      <td>10</td>\n",
       "      <td>0</td>\n",
       "      <td>สวัสดี</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2379</td>\n",
       "      <td>สวัสดี\\11.parquet</td>\n",
       "      <td>11</td>\n",
       "      <td>0</td>\n",
       "      <td>สวัสดี</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2380</td>\n",
       "      <td>สวัสดี\\12.parquet</td>\n",
       "      <td>12</td>\n",
       "      <td>0</td>\n",
       "      <td>สวัสดี</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2992</th>\n",
       "      <td>723</td>\n",
       "      <td>ฎ\\4.parquet</td>\n",
       "      <td>4</td>\n",
       "      <td>110</td>\n",
       "      <td>ฎ</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2993</th>\n",
       "      <td>724</td>\n",
       "      <td>ฎ\\5.parquet</td>\n",
       "      <td>5</td>\n",
       "      <td>110</td>\n",
       "      <td>ฎ</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2994</th>\n",
       "      <td>725</td>\n",
       "      <td>ฎ\\6.parquet</td>\n",
       "      <td>6</td>\n",
       "      <td>110</td>\n",
       "      <td>ฎ</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2995</th>\n",
       "      <td>727</td>\n",
       "      <td>ฎ\\8.parquet</td>\n",
       "      <td>8</td>\n",
       "      <td>110</td>\n",
       "      <td>ฎ</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2996</th>\n",
       "      <td>728</td>\n",
       "      <td>ฎ\\9.parquet</td>\n",
       "      <td>9</td>\n",
       "      <td>110</td>\n",
       "      <td>ฎ</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>2997 rows × 5 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      index               path  participant_id  sequence_id    sign\n",
       "0      2389  สวัสดี\\22.parquet              22            0  สวัสดี\n",
       "1      2377   สวัสดี\\1.parquet               1            0  สวัสดี\n",
       "2      2378  สวัสดี\\10.parquet              10            0  สวัสดี\n",
       "3      2379  สวัสดี\\11.parquet              11            0  สวัสดี\n",
       "4      2380  สวัสดี\\12.parquet              12            0  สวัสดี\n",
       "...     ...                ...             ...          ...     ...\n",
       "2992    723        ฎ\\4.parquet               4          110       ฎ\n",
       "2993    724        ฎ\\5.parquet               5          110       ฎ\n",
       "2994    725        ฎ\\6.parquet               6          110       ฎ\n",
       "2995    727        ฎ\\8.parquet               8          110       ฎ\n",
       "2996    728        ฎ\\9.parquet               9          110       ฎ\n",
       "\n",
       "[2997 rows x 5 columns]"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = new_df.sort_values(by=['sequence_id']).reset_index(drop=False)\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "ename": "KeyboardInterrupt",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
      "Cell \u001b[0;32mIn[49], line 24\u001b[0m\n\u001b[1;32m     19\u001b[0m                 \u001b[38;5;28;01mcontinue\u001b[39;00m\n\u001b[1;32m     20\u001b[0m             \u001b[38;5;28;01mfor\u001b[39;00m seq_id, value \u001b[38;5;129;01min\u001b[39;00m \u001b[38;5;28menumerate\u001b[39m(np_data):\n\u001b[1;32m     21\u001b[0m                 data\u001b[38;5;241m.\u001b[39mappend({\n\u001b[1;32m     22\u001b[0m                         \u001b[38;5;124m'\u001b[39m\u001b[38;5;124mpath\u001b[39m\u001b[38;5;124m'\u001b[39m: file_path,\n\u001b[1;32m     23\u001b[0m                         \u001b[38;5;124m'\u001b[39m\u001b[38;5;124mparticipant_id\u001b[39m\u001b[38;5;124m'\u001b[39m: participant_id,\n\u001b[0;32m---> 24\u001b[0m                         \u001b[38;5;124m'\u001b[39m\u001b[38;5;124msequence_id\u001b[39m\u001b[38;5;124m'\u001b[39m: \u001b[43mseq_id\u001b[49m,\n\u001b[1;32m     25\u001b[0m                         \u001b[38;5;124m'\u001b[39m\u001b[38;5;124msign\u001b[39m\u001b[38;5;124m'\u001b[39m: sign\n\u001b[1;32m     26\u001b[0m                     })\n\u001b[1;32m     28\u001b[0m         \u001b[38;5;66;03m#อ่านค่า npy แต่ละไฟล์ละนำมาเก็บเป็น parquet file ตามแต่ละ subfolder\u001b[39;00m\n\u001b[1;32m     32\u001b[0m \u001b[38;5;28mid\u001b[39m \u001b[38;5;241m+\u001b[39m\u001b[38;5;241m=\u001b[39m \u001b[38;5;241m1\u001b[39m\n",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m: "
     ]
    }
   ],
   "source": [
    "base_path = '/Users/theerat/Downloads/MP_DATA3'\n",
    "data = []\n",
    "id = 0\n",
    "for folder_name, participant_id in word_list.items():\n",
    "    folder_path = os.path.join(base_path, folder_name)\n",
    "    for subfolder in os.listdir(folder_path):\n",
    "        if subfolder == '.DS_Store':\n",
    "                continue\n",
    "        subfolder_path = os.path.join(folder_path, subfolder)\n",
    "        sign = os.path.basename(folder_path)\n",
    "        \n",
    "        for filename in os.listdir(subfolder_path):\n",
    "            if filename.endswith('.npy') and filename != '.DS_Store':\n",
    "                file_path = os.path.join(subfolder_path, filename)\n",
    "                try:\n",
    "                    np_data = np.load(file_path)\n",
    "                except Exception as e :\n",
    "                    print('Error')\n",
    "                    continue\n",
    "                for seq_id, value in enumerate(np_data):\n",
    "                    data.append({\n",
    "                            'path': file_path,\n",
    "                            'participant_id': participant_id,\n",
    "                            'sequence_id': seq_id,\n",
    "                            'sign': sign\n",
    "                        })\n",
    "                    \n",
    "            #อ่านค่า npy แต่ละไฟล์ละนำมาเก็บเป็น parquet file ตามแต่ละ subfolder\n",
    "        \n",
    "\n",
    "            \n",
    "    id += 1\n",
    "df = pd.DataFrame(data)\n",
    "df_sorted = df.sort_values(by=['participant_id', 'sequence_id']).reset_index(drop=True)\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [],
   "source": [
    "def create_landmark_rows(landmarks, frame, index, type_label):\n",
    "    for landmark_index, (x, y, z) in enumerate(landmarks):\n",
    "        new_row = {\n",
    "            'frame': frame,\n",
    "            'row_id': f'{folder_name}-{index}',\n",
    "            'type': type_label,\n",
    "            'landmark_index': landmark_index,\n",
    "            'x': x,\n",
    "            'y': y,\n",
    "            'z': z\n",
    "        }\n",
    "    yield new_row"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/20/20.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     NaN             NaN NaN   \n",
      "1      ß?                     Qö?             NaN NaN   \n",
      "2       ?                     ~Ô?             NaN NaN   \n",
      "3      Ø?                     NaN             NaN NaN   \n",
      "4      Ð?                     Út¿             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/20/20.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/20/21.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1     ?Á¿                     NaN             NaN  NaN   \n",
      "2     \u0013ö?                      \u0003@             e\u0004@  NaN   \n",
      "3       ?                     NaN             NaN  NaN   \n",
      "4     dá?                     NaN             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/20/21.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/20/23.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1     BÔ?                     $Ó?              õ>  NaN   \n",
      "2     ¡¿                     NaN             NaN  NaN   \n",
      "3     ¡Õ?                     ¼Õ?             NaN  NaN   \n",
      "4     êà?                     à?             Qà?   Ø?   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/20/23.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/20/22.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1      Ò?                     NaN             NaN  NaN   \n",
      "2     fì?                     NaN             NaN  NaN   \n",
      "3      Ù?                     NaN             NaN  NaN   \n",
      "4     <Ò?                     NaN             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/20/22.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/20/26.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False   'shape': (1662   )  \\\n",
      "0     NaN                     NaN              NaN NaN   \n",
      "1     \u0005â¿                     \u001dþ?              NaN NaN   \n",
      "2     á?                     NaN              NaN NaN   \n",
      "3     yØ?                     NaN              NaN NaN   \n",
      "4      Ö?                       ¿              NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/20/26.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/20/27.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     Bá¿             NaN  NaN   \n",
      "1      à?                     NaN             NaN  NaN   \n",
      "2      ?                     NaN             NaN  NaN   \n",
      "3      á?                     NaN             NaN  NaN   \n",
      "4      ¿                     NaN             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/20/27.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/20/25.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     NaN             NaN NaN   \n",
      "1      ö¿                     NaN             NaN NaN   \n",
      "2     \u0002\u0006?                      ¿             NaN NaN   \n",
      "3     LÒ?                     NaN             NaN NaN   \n",
      "4     Æc¿                     NaN             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/20/25.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/20/19.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     NaN             NaN NaN   \n",
      "1     RÑ?                     þï?             NaN NaN   \n",
      "2      à?                     õ¿             NaN NaN   \n",
      "3     Lö?                     NaN             NaN NaN   \n",
      "4      D?                     NaN             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/20/19.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/20/18.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                      õ?             Ë¿  Õ?   \n",
      "1      Í?                     NaN             NaN  NaN   \n",
      "2      {¿                     êà?             NaN  NaN   \n",
      "3     :¿                     NaN             NaN  NaN   \n",
      "4      á?                      á?             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/20/18.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/20/24.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                      Ñ?              è?  ç?   \n",
      "1     ÇØ?                     NaN             NaN  NaN   \n",
      "2      ¿                     k?              Ó?   ?   \n",
      "3      Ò?                     NaN             NaN  NaN   \n",
      "4     \u001e?                      á?             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/20/24.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/20/5.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     NaN             NaN NaN   \n",
      "1     NaN                     NaN             NaN NaN   \n",
      "2     »á?                     NaN             NaN NaN   \n",
      "3     Ãá?                      Õ?              u? NaN   \n",
      "4      ¿                     \u001eØ?             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/20/5.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/20/4.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                      Á¿             NaN  NaN   \n",
      "1     \u0004\u0004@                      â?             NaN  NaN   \n",
      "2     NaN                     ¬á?             NaN  NaN   \n",
      "3      à?                     NaN             NaN  NaN   \n",
      "4      à?                     {Ø?             ÂÔ?   ¿   \n",
      "\n",
      "   }                                                           \n",
      "0                                                NaN           \n",
      "1                                                NaN           \n",
      "2                                                NaN           \n",
      "3                                                NaN           \n",
      "4                                                 á?           \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/20/4.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/20/6.npy\n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/20/6.npy: Error tokenizing data. C error: Expected 5 fields in line 22, saw 6\n",
      "\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/20/7.npy\n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/20/7.npy: Error tokenizing data. C error: Expected 5 fields in line 24, saw 8\n",
      "\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/20/3.npy\n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/20/3.npy: Error tokenizing data. C error: Buffer overflow caught - possible malformed input file.\n",
      "\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/20/2.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     NaN             NaN NaN   \n",
      "1     µÑ?                     þï?             NaN NaN   \n",
      "2     ÞÔ?                      Â?             NaN NaN   \n",
      "3     ^é?                     NaN             NaN NaN   \n",
      "4      ~¿                     NaN             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/20/2.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/20/0.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1      ?                     NaN             NaN  NaN   \n",
      "2      V?                     NaN             NaN  NaN   \n",
      "3      Ø?                     Ká?             NaN  NaN   \n",
      "4      Õ?                     NaN             NaN  NaN   \n",
      "\n",
      "   }                                                           \n",
      "0                                                NaN           \n",
      "1                                                NaN           \n",
      "2                                                NaN           \n",
      "3                                                NaN           \n",
      "4                                                NaN           \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/20/0.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/20/1.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     NaN             NaN NaN   \n",
      "1      õ?                      å¿             NaN NaN   \n",
      "2     Úð>                     NaN             NaN NaN   \n",
      "3     &â?                     NaN             NaN NaN   \n",
      "4     {?                     NaN             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/20/1.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/20/9.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                      Ó?             B? NaN   \n",
      "1      þ?                     NaN             NaN NaN   \n",
      "2     ¿                     NaN             NaN NaN   \n",
      "3     Þ?                     âß?             NaN NaN   \n",
      "4     \u0014á?                     \u0006á?             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/20/9.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/20/8.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     NaN             NaN NaN   \n",
      "1     FÑ?                     NaN             NaN NaN   \n",
      "2      ë¿                     NaN             NaN NaN   \n",
      "3       ?                     NaN             NaN NaN   \n",
      "4     á?                     Ü¿             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/20/8.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/20/29.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                      þ>             NaN NaN   \n",
      "1      à?                     ¿             NaN NaN   \n",
      "2      V¿                      á?             NaN NaN   \n",
      "3      ß?                      á?             ?? NaN   \n",
      "4      k¿                      ?             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/20/29.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/20/15.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1      ý¿                      Û?             NaN  NaN   \n",
      "2      \u0004@                     Ñ£¿             NaN  NaN   \n",
      "3     Cá?                     NaN             NaN  NaN   \n",
      "4     ·Ò?                     NaN             NaN  NaN   \n",
      "\n",
      "   }                                                           \n",
      "0                                                NaN           \n",
      "1                                                NaN           \n",
      "2                                                NaN           \n",
      "3                                                NaN           \n",
      "4                                                NaN           \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/20/15.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/20/14.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     üï?              Ø?   ô?   \n",
      "1      á?                     NaN             NaN  NaN   \n",
      "2      á?                     NaN             NaN  NaN   \n",
      "3     ¿                     NaN             NaN  NaN   \n",
      "4      Ø?                     NaN             NaN  NaN   \n",
      "\n",
      "   }                                                           \n",
      "0                                                 Ê?           \n",
      "1                                                NaN           \n",
      "2                                                NaN           \n",
      "3                                                NaN           \n",
      "4                                                NaN           \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/20/14.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/20/28.npy\n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/20/28.npy: Error tokenizing data. C error: Expected 5 fields in line 9, saw 6\n",
      "\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/20/16.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     ê¿             NaN NaN   \n",
      "1    ?                     NaN             NaN NaN   \n",
      "2     mu?                     NaN             NaN NaN   \n",
      "3     ¤Õ?                      á?             Sá? NaN   \n",
      "4      à?                     NaN             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/20/16.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/20/17.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     NaN             NaN NaN   \n",
      "1     êà?                     \u0019¿             NaN NaN   \n",
      "2      ¿                     NaN             NaN NaN   \n",
      "3     tà?                     NaN             NaN NaN   \n",
      "4      o?                     NaN             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/20/17.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/20/13.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     Aã?              ? NaN   \n",
      "1     ¥ ?                      Ñ?             NaN NaN   \n",
      "2      Þ?                     NaN             NaN NaN   \n",
      "3     7Ñ?                     tà?              ß? NaN   \n",
      "4     Y¢?                     NaN             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/20/13.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/20/12.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1      ã¿                     NaN             NaN  NaN   \n",
      "2     øè?                     ïÀ?             NaN  NaN   \n",
      "3      Ù?                     NaN             NaN  NaN   \n",
      "4      S?                     Và?             NaN  NaN   \n",
      "\n",
      "   }                                                           \n",
      "0                                                NaN           \n",
      "1                                                NaN           \n",
      "2                                                NaN           \n",
      "3                                                NaN           \n",
      "4                                                NaN           \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/20/12.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/20/10.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1      õ?                     NaN             NaN  NaN   \n",
      "2      Ô?                     8\u0004@             NaN  NaN   \n",
      "3       ?                     NaN             NaN  NaN   \n",
      "4      £¿                     Hà?             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/20/10.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/20/11.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     NaN             NaN NaN   \n",
      "1      ä?                     NaN             NaN NaN   \n",
      "2      ç?                     NaN             NaN NaN   \n",
      "3      á?                      Ð?             NaN NaN   \n",
      "4     ÛÙ?                     NaN             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/20/11.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/7/20.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     NaN             NaN NaN   \n",
      "1      ã?                      Ø?             NaN NaN   \n",
      "2      ï?                      á?              Ò? NaN   \n",
      "3     HÒ?                     NaN             NaN NaN   \n",
      "4      e¿                     NaN             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/7/20.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/7/21.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     NaN             NaN NaN   \n",
      "1     <à?                     NaN             NaN NaN   \n",
      "2      á?                      ¿             NaN NaN   \n",
      "3      p¿                     NaN             NaN NaN   \n",
      "4     \u0004á?                     NaN             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/7/21.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/7/23.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1     è?                     NaN             NaN  NaN   \n",
      "2       ?                     NaN             NaN  NaN   \n",
      "3       ?                     NaN             NaN  NaN   \n",
      "4     \u0003Ó?                     Û¤¿             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/7/23.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/7/22.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     ð\u0003@             NaN NaN   \n",
      "1      ¿                     NaN             NaN NaN   \n",
      "2      á?                     NaN             NaN NaN   \n",
      "3      á?                      ¿             NaN NaN   \n",
      "4     \u001b?                     NaN             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/7/22.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/7/26.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                      Ó?             NaN  NaN   \n",
      "1      è?                     à?              £?   á?   \n",
      "2      ?                      á?             1ß?  NaN   \n",
      "3      ã?                     NaN             NaN  NaN   \n",
      "4      â?                     NaN             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/7/26.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/7/27.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1      ö?                      ï?             Àô?  ý?   \n",
      "2      \u0004@                      \u0006?              á?  NaN   \n",
      "3      ?                     NaN             NaN  NaN   \n",
      "4      ¿                     NaN             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/7/27.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/7/25.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                      á?             NaN  NaN   \n",
      "1     ç?                     NaN             NaN  NaN   \n",
      "2      Î?                     NaN             NaN  NaN   \n",
      "3     é\u0003@                     NaN             NaN  NaN   \n",
      "4      á?                     NaN             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/7/25.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/7/19.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1     nÑ?                     NaN             NaN  NaN   \n",
      "2     fÓ?                     NaN             NaN  NaN   \n",
      "3     ^â?                     NaN             NaN  NaN   \n",
      "4     2Ø?                     èî?             X\u0011?  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/7/19.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/7/18.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     Çö¿             NaN NaN   \n",
      "1     ¬ô?                      Û?              ß? NaN   \n",
      "2     ß?                     NaN             NaN NaN   \n",
      "3     /u?                     ¤à?             NaN NaN   \n",
      "4      à?                      |?             Và? NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/7/18.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/7/24.npy\n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/7/24.npy: Error tokenizing data. C error: Buffer overflow caught - possible malformed input file.\n",
      "\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/7/5.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     NaN             NaN NaN   \n",
      "1     £Ð¿                     NaN             NaN NaN   \n",
      "2      Ò?                     NaN             NaN NaN   \n",
      "3     IÜ?                     Iå?             NaN NaN   \n",
      "4     «Ø?                     NaN             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/7/5.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/7/4.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     þï?             NaN NaN   \n",
      "1      Ó¿                     NaN             NaN NaN   \n",
      "2     ÓÖ?                     NaN             NaN NaN   \n",
      "3       ?                     NaN             NaN NaN   \n",
      "4      Ò?                     NaN             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/7/4.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/7/6.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1     Eá?                     NaN             NaN  NaN   \n",
      "2     ÆØ?                     ?              á?   Ù?   \n",
      "3      ß?                     NaN             NaN  NaN   \n",
      "4      á?                     NaN             NaN  NaN   \n",
      "\n",
      "   }                                                           \n",
      "0                                                NaN           \n",
      "1                                                NaN           \n",
      "2                                                YÕ?           \n",
      "3                                                NaN           \n",
      "4                                                NaN           \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/7/6.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/7/7.npy\n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/7/7.npy: Error tokenizing data. C error: Expected 5 fields in line 13, saw 6\n",
      "\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/7/3.npy\n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/7/3.npy: Error tokenizing data. C error: Buffer overflow caught - possible malformed input file.\n",
      "\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/7/2.npy\n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/7/2.npy: Error tokenizing data. C error: EOF inside string starting at row 17\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/7/0.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     õá¿             NaN NaN   \n",
      "1      Ô?                     NaN             NaN NaN   \n",
      "2       ?                      Ñ?             fÎ? NaN   \n",
      "3     ¿                      T?             NaN NaN   \n",
      "4      Ð?                     NaN             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/7/0.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/7/1.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                      Ê?             NaN  NaN   \n",
      "1     ëå?                     2Ö?              Â¿  \u000fà?   \n",
      "2     Ç?                     £?             NaN  NaN   \n",
      "3      ¿                     NaN             NaN  NaN   \n",
      "4     sá?                      A¿             NaN  NaN   \n",
      "\n",
      "   }                                                           \n",
      "0                                                NaN           \n",
      "1                                                NaN           \n",
      "2                                                NaN           \n",
      "3                                                NaN           \n",
      "4                                                NaN           \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/7/1.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/7/9.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     \u001bæ¿             NaN  NaN   \n",
      "1     þï?                     NaN             NaN  NaN   \n",
      "2      Ì?                     NaN             NaN  NaN   \n",
      "3     Ëõ?                     NaN             NaN  NaN   \n",
      "4     Ö\u0003@                     NaN             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/7/9.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/7/8.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     Sî?             NaN  NaN   \n",
      "1     ?                      Ó?             NaN  NaN   \n",
      "2     1¿                     NaN             NaN  NaN   \n",
      "3     Ms¿                     NaN             NaN  NaN   \n",
      "4     ãß?                     NaN             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/7/8.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/7/29.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                      Ñ?             NaN  NaN   \n",
      "1     Pá¿                     NaN             NaN  NaN   \n",
      "2      ß?                     NaN             NaN  NaN   \n",
      "3     ÀÑ?                     8Ò?             \u001fç?  NaN   \n",
      "4      \u0004@                      á?              á?   Õ?   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/7/29.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/7/15.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1     þâ¿                      ä?             NaN  NaN   \n",
      "2      ô?                     NaN             NaN  NaN   \n",
      "3      Ó?                     [£¿              Ò?  l¿   \n",
      "4      [?                     NaN             NaN  NaN   \n",
      "\n",
      "   }                                                           \n",
      "0                                                NaN           \n",
      "1                                                NaN           \n",
      "2                                                NaN           \n",
      "3                                                NaN           \n",
      "4                                                NaN           \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/7/15.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/7/14.npy\n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/7/14.npy: Error tokenizing data. C error: Expected 5 fields in line 12, saw 8\n",
      "\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/7/28.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     NaN             NaN NaN   \n",
      "1      ß?                     NaN             NaN NaN   \n",
      "2     NaN                      à?             \u0001ã? NaN   \n",
      "3      \u0004@                     NaN             NaN NaN   \n",
      "4      Ô?                      á?             ZY¿ NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/7/28.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/7/16.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     þï?             NaN NaN   \n",
      "1     hâ?                      ä?             NaN NaN   \n",
      "2      Å¿                     NaN             NaN NaN   \n",
      "3     ÿÓ?                     NaN             NaN NaN   \n",
      "4     Ú\u0003@                     NaN             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/7/16.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/7/17.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1      å¿                     ²Ñ?              Ø?   à?   \n",
      "2      á?                     NaN             NaN  NaN   \n",
      "3     à?                     NaN             NaN  NaN   \n",
      "4      Ð?                      ¿             NaN  NaN   \n",
      "\n",
      "   }                                                           \n",
      "0                                                NaN           \n",
      "1                                                NaN           \n",
      "2                                                NaN           \n",
      "3                                                NaN           \n",
      "4                                                NaN           \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/7/17.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/7/13.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     NaN             NaN NaN   \n",
      "1     Øå¿                     NaN             NaN NaN   \n",
      "2     µã¿                     ²Û?             NaN NaN   \n",
      "3       ?                     NaN             NaN NaN   \n",
      "4      ä?                     NaN             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/7/13.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/7/12.npy\n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/7/12.npy: Error tokenizing data. C error: Buffer overflow caught - possible malformed input file.\n",
      "\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/7/10.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     NaN             NaN NaN   \n",
      "1       ?                     NaN             NaN NaN   \n",
      "2      Ö¿                     Ò?             NaN NaN   \n",
      "3      ¿                     NaN             NaN NaN   \n",
      "4      á?                     NaN             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/7/10.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/7/11.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     >â?              ã?  NaN   \n",
      "1     VÔ?                     NaN             NaN  NaN   \n",
      "2       ?                     ae¿             NaN  NaN   \n",
      "3     YÑ?                      Ò?              Ù?  NaN   \n",
      "4      ?                     NaN             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/7/11.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/17/20.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1      é?                     NaN             NaN  NaN   \n",
      "2      Ô?                     NaN             NaN  NaN   \n",
      "3      ¾¿                     NaN             NaN  NaN   \n",
      "4      ß?                      ¿             NaN  NaN   \n",
      "\n",
      "   }                                                           \n",
      "0                                                NaN           \n",
      "1                                                NaN           \n",
      "2                                                NaN           \n",
      "3                                                NaN           \n",
      "4                                                NaN           \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/17/20.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/17/21.npy\n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/17/21.npy: Error tokenizing data. C error: Expected 5 fields in line 8, saw 6\n",
      "\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/17/23.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                      Ñ?             NaN  NaN   \n",
      "1       ?                      á?              ß?  NaN   \n",
      "2      Ø?                     NaN             NaN  NaN   \n",
      "3     ÛÝ?                     NaN             NaN  NaN   \n",
      "4      ß?                     NaN             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/17/23.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/17/22.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     þï?             üï?  NaN   \n",
      "1      Ò?                     NaN             NaN  NaN   \n",
      "2      ¿                     ºô?             NaN  NaN   \n",
      "3      9?                      á?              ¿  x£¿   \n",
      "4     ¿                      ¿             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/17/22.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/17/26.npy\n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/17/26.npy: Error tokenizing data. C error: EOF inside string starting at row 25\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/17/27.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1      ¥¿                     èà?             NaN  NaN   \n",
      "2      Ð?                     NaN             NaN  NaN   \n",
      "3      à?                     NaN             NaN  NaN   \n",
      "4     ¼à?                     ø¿             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/17/27.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/17/25.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                      è?             NaN  NaN   \n",
      "1     Gá?                     NaN             NaN  NaN   \n",
      "2     G?                     NaN             NaN  NaN   \n",
      "3     ZÞ?                     NaN             NaN  NaN   \n",
      "4     \u0014?                     NaN             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/17/25.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/17/19.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     NaN             NaN NaN   \n",
      "1     NaN                     Ñß?              Ô? NaN   \n",
      "2      Õ?                      \\?             NaN NaN   \n",
      "3     Ð?                     NaN             NaN NaN   \n",
      "4     ÷Ý?                     NaN             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/17/19.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/17/18.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     NaN             NaN NaN   \n",
      "1      Ñ?                     \u0018Ñ?             NaN NaN   \n",
      "2      è?                     NaN             NaN NaN   \n",
      "3     p\u0001?                     NaN             NaN NaN   \n",
      "4      á?                     NaN             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/17/18.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/17/24.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                      ?             NaN NaN   \n",
      "1     ÿÒ?                      ?             NaN NaN   \n",
      "2     Rá?                      ?              Ò? NaN   \n",
      "3     á¡¿                      C?             NaN NaN   \n",
      "4      ¿                     NaN             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/17/24.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/17/5.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     NaN             NaN NaN   \n",
      "1     ä?                     NaN             NaN NaN   \n",
      "2     Té?                     NaN             NaN NaN   \n",
      "3      Õ?                      ?              ? NaN   \n",
      "4       ?                     NaN             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/17/5.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/17/4.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                      ß?             NaN  NaN   \n",
      "1      õ?                     NaN             NaN  NaN   \n",
      "2      Ë?                      à?             NaN  NaN   \n",
      "3      u?                     NaN             NaN  NaN   \n",
      "4     à?                     NaN             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/17/4.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/17/6.npy\n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/17/6.npy: Error tokenizing data. C error: Expected 5 fields in line 24, saw 6\n",
      "\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/17/7.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1      Ñ?                     V\u000e?             âÇ?  NaN   \n",
      "2     ]Ð?                     ¬Ð?             NaN  NaN   \n",
      "3     NaN                     NaN             NaN  NaN   \n",
      "4      Õ?                     NaN             NaN  NaN   \n",
      "\n",
      "   }                                                           \n",
      "0                                                NaN           \n",
      "1                                                NaN           \n",
      "2                                                NaN           \n",
      "3                                                NaN           \n",
      "4                                                NaN           \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/17/7.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/17/3.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     NaN             NaN NaN   \n",
      "1      Ò¿                     NaN             NaN NaN   \n",
      "2      î?                     NaN             NaN NaN   \n",
      "3      \u0003@                     NaN             NaN NaN   \n",
      "4      ¿                     NaN             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/17/3.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/17/2.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     \u0010ß?             NaN NaN   \n",
      "1     ç?                     NaN             NaN NaN   \n",
      "2      ù¿                     \u0012á?             NaN NaN   \n",
      "3     \u0003Ø?                     NaN             NaN NaN   \n",
      "4     Ö?                     NaN             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/17/2.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/17/0.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     NaN             NaN NaN   \n",
      "1      Ñ?                     Sà?             NaN NaN   \n",
      "2     ×?                     NaN             NaN NaN   \n",
      "3      Õ¿                     NaN             NaN NaN   \n",
      "4     \u0003Î?                     NaN             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/17/0.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/17/1.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     NaN             NaN NaN   \n",
      "1     üï?                     NaN             NaN NaN   \n",
      "2      Õ?                     NaN             NaN NaN   \n",
      "3       ?                     NaN             NaN NaN   \n",
      "4      á?                     NaN             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/17/1.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/17/9.npy\n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/17/9.npy: Error tokenizing data. C error: Expected 5 fields in line 23, saw 6\n",
      "\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/17/8.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     \u0007\u0016¿             NaN  NaN   \n",
      "1      ×?                     NaN             NaN  NaN   \n",
      "2     ?Ô?                     NaN             NaN  NaN   \n",
      "3      Ð?                     NaN             NaN  NaN   \n",
      "4      Ò?                     ~t?             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/17/8.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/17/29.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1     ûâ?                     NaN             NaN  NaN   \n",
      "2     Ûâ¿                     NaN             NaN  NaN   \n",
      "3     þï?                      ¸¿              è?  NaN   \n",
      "4      Å?                     NaN             NaN  NaN   \n",
      "\n",
      "   }                                                           \n",
      "0                                                NaN           \n",
      "1                                                NaN           \n",
      "2                                                NaN           \n",
      "3                                                NaN           \n",
      "4                                                NaN           \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/17/29.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/17/15.npy\n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/17/15.npy: Error tokenizing data. C error: Expected 5 fields in line 15, saw 6\n",
      "\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/17/14.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1     ~ù¿                      é?             o\u0003@  NaN   \n",
      "2      Ö?                     NaN             NaN  NaN   \n",
      "3     ïÓ?                     NaN             NaN  NaN   \n",
      "4     [Ý?                      Þ?             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/17/14.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/17/28.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     þï?             ÇÑ?  NaN   \n",
      "1      á?                     +Þ?             NaN  NaN   \n",
      "2      ¿                     «à?             NaN  NaN   \n",
      "3      Þ?                     NaN             NaN  NaN   \n",
      "4      á?                      Õ?             øÝ?  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/17/28.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/17/16.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     NaN             NaN NaN   \n",
      "1      ç¿                     NaN             NaN NaN   \n",
      "2     iÑ?                     NaN             NaN NaN   \n",
      "3      Ø?                     NaN             NaN NaN   \n",
      "4     úÚ?                     NaN             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/17/16.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/17/17.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1      Ø?                     NaN             NaN  NaN   \n",
      "2     Iï?                     ø¿              \u0002?  NaN   \n",
      "3       ?                      á?             .á?  NaN   \n",
      "4     öÞ?                     \u0010Ø?             °?   Ê?   \n",
      "\n",
      "   }                                                           \n",
      "0                                                NaN           \n",
      "1                                                NaN           \n",
      "2                                                NaN           \n",
      "3                                                NaN           \n",
      "4                                                NaN           \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/17/17.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/17/13.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1     \u0019Ó?                     \u0015j¿             NaN  NaN   \n",
      "2      ¿                     NaN             NaN  NaN   \n",
      "3     \u001c¿                      Ð?             eÝ?   Ý?   \n",
      "4     L2¿                     NaN             NaN  NaN   \n",
      "\n",
      "   }                                                           \n",
      "0                                                NaN           \n",
      "1                                                NaN           \n",
      "2                                                NaN           \n",
      "3                                                \u0006Õ?           \n",
      "4                                                NaN           \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/17/13.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/17/12.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1     ÆÒ¿                     NaN             NaN  NaN   \n",
      "2     ç?                     ·Õ?             %Ñ¿  à?   \n",
      "3     \u0015Ò?                     NaN             NaN  NaN   \n",
      "4     à?                     Ä ?             \u000fÞ?  NaN   \n",
      "\n",
      "   }                                                           \n",
      "0                                                NaN           \n",
      "1                                                NaN           \n",
      "2                                                NaN           \n",
      "3                                                NaN           \n",
      "4                                                NaN           \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/17/12.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/17/10.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     NaN             NaN NaN   \n",
      "1     %á?                     NaN             NaN NaN   \n",
      "2      Ñ?                     NaN             NaN NaN   \n",
      "3     \u0016Ö?                     NaN             NaN NaN   \n",
      "4      ~?                     15?             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/17/10.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/17/11.npy\n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/17/11.npy: Error tokenizing data. C error: Expected 5 fields in line 30, saw 7\n",
      "\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/10/20.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     \u000fÑ?             NaN  NaN   \n",
      "1     \u0001Î¿                      è?             NaN  ý?   \n",
      "2     ¤¿                     NaN             NaN  NaN   \n",
      "3      á?                     NaN             NaN  NaN   \n",
      "4      á?                      m¿              Þ?  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/10/20.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/10/21.npy\n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/10/21.npy: Error tokenizing data. C error: Expected 5 fields in line 3, saw 6\n",
      "\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/10/23.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     NaN             NaN NaN   \n",
      "1       ?                     NaN             NaN NaN   \n",
      "2      Ô¿                      ß?             NaN NaN   \n",
      "3     ñ¿                     NaN             NaN NaN   \n",
      "4      ¿                     NaN             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/10/23.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/10/22.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     NaN             NaN NaN   \n",
      "1     üï?                      Ó?             ÍÝ? NaN   \n",
      "2     :à?                     NaN             NaN NaN   \n",
      "3     _Ú?                     Sí?             NaN NaN   \n",
      "4      Õ?                     W¿             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/10/22.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/10/26.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     Bß?              Ï?  NaN   \n",
      "1     ~Ø?                     NaN             NaN  NaN   \n",
      "2     W­?                     NaN             NaN  NaN   \n",
      "3      ß?                     NaN             NaN  NaN   \n",
      "4     º7¿                     NaN             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/10/26.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/10/27.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                      Ô?             °á¿  NaN   \n",
      "1      ã?                     NaN             NaN  NaN   \n",
      "2      ß?                     NaN             NaN  NaN   \n",
      "3      Ù?                     NaN             NaN  NaN   \n",
      "4      ×?                     \u0017Ü?             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/10/27.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/10/25.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                      Î?              Ó¿ NaN   \n",
      "1      á?                     NaN             NaN NaN   \n",
      "2     à?                     NaN             NaN NaN   \n",
      "3     Uà?                     8Ï?             NaN NaN   \n",
      "4     ÎÒ?                     NaN             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/10/25.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/10/19.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1      Ñ?                     mà?              Õ?   Ý?   \n",
      "2      á?                     NaN             NaN  NaN   \n",
      "3     ÌÈ?                     H¿              Ñ?   Ò?   \n",
      "4      ?                     NaN             NaN  NaN   \n",
      "\n",
      "   }                                                           \n",
      "0                                                NaN           \n",
      "1                                                \fÐ?           \n",
      "2                                                NaN           \n",
      "3                                                NaN           \n",
      "4                                                NaN           \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/10/19.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/10/18.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1      â?                     NaN             NaN  NaN   \n",
      "2      ã?                     Ó?              À¿   Õ¿   \n",
      "3     ³ô?                     s¿             _\u0003@  NaN   \n",
      "4      \u0004@                     NaN             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/10/18.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/10/24.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1      Ñ?                     NaN             NaN  NaN   \n",
      "2     ?Ø?                     NaN             NaN  NaN   \n",
      "3      ç?                     NaN             NaN  NaN   \n",
      "4     L\u0001?                     NaN             NaN  NaN   \n",
      "\n",
      "   }                                                           \n",
      "0                                                NaN           \n",
      "1                                                NaN           \n",
      "2                                                NaN           \n",
      "3                                                NaN           \n",
      "4                                                NaN           \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/10/24.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/10/5.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1     Dâ?                     ¥ã¿             »ó?  NaN   \n",
      "2     NaN                      \u0017?             NaN  NaN   \n",
      "3      Ó?                     fá?             NaN  NaN   \n",
      "4      à?                     NaN             NaN  NaN   \n",
      "\n",
      "   }                                                           \n",
      "0                                                NaN           \n",
      "1                                                NaN           \n",
      "2                                                NaN           \n",
      "3                                                NaN           \n",
      "4                                                NaN           \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/10/5.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/10/4.npy\n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/10/4.npy: Error tokenizing data. C error: Expected 5 fields in line 6, saw 7\n",
      "\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/10/6.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False   'shape': (1662   )  \\\n",
      "0     NaN                     þï?              NaN NaN   \n",
      "1     ãè?                     NaN              NaN NaN   \n",
      "2     ²Ð?                     NaN              NaN NaN   \n",
      "3      ¿                     NaN              NaN NaN   \n",
      "4      É?                     NaN              NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/10/6.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/10/7.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     +ö?             NaN  NaN   \n",
      "1     ?                      Õ?              á?   á?   \n",
      "2      à?                      Õ?             NaN  NaN   \n",
      "3     Ñ¿                     NaN             NaN  NaN   \n",
      "4      á?                     NaN             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/10/7.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/10/3.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                      Ñ?             NaN NaN   \n",
      "1     Ù¿                     NaN             NaN NaN   \n",
      "2     üï?                     NaN             NaN NaN   \n",
      "3     tå¿                     NaN             NaN NaN   \n",
      "4     ßã?                      ô?             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/10/3.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/10/2.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     PÔ?             NaN  NaN   \n",
      "1      Û?                     NaN             NaN  NaN   \n",
      "2     B¤¿                     ¼\u0003@             NaN  NaN   \n",
      "3      à?                     NaN             NaN  NaN   \n",
      "4      ×?                      o¿              Ï?  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/10/2.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/10/0.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     NaN             NaN NaN   \n",
      "1      ×?                     ?             NaN NaN   \n",
      "2      à?                     NaN             NaN NaN   \n",
      "3     gà?                     }Ø?             NaN NaN   \n",
      "4     d¿                     NaN             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/10/0.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/10/1.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1      â?                     NaN             NaN  NaN   \n",
      "2     .ì?                     äÓ?             ¶¿  6¿   \n",
      "3      «?                     NaN             NaN  NaN   \n",
      "4     ¿                     ¸à?             à?  NaN   \n",
      "\n",
      "   }                                                           \n",
      "0                                                NaN           \n",
      "1                                                NaN           \n",
      "2                                                 Ñ?           \n",
      "3                                                NaN           \n",
      "4                                                NaN           \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/10/1.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/10/9.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False   'shape': (1662   )  \\\n",
      "0     NaN                     NaN              NaN NaN   \n",
      "1      ×?                      á?              NaN NaN   \n",
      "2     4¿                     NaN              NaN NaN   \n",
      "3     Þ?                     NaN              NaN NaN   \n",
      "4     à?                     NaN              NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/10/9.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/10/8.npy\n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/10/8.npy: Error tokenizing data. C error: Buffer overflow caught - possible malformed input file.\n",
      "\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/10/29.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1      ç?                     U\u0003@             NaN  NaN   \n",
      "2      Ò?                      ?             NaN  NaN   \n",
      "3     &Þ?                     NaN             NaN  NaN   \n",
      "4     Å¿                     NaN             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/10/29.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/10/15.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False   'shape': (1662   )  \\\n",
      "0     NaN                     ùô?              NaN NaN   \n",
      "1     7Û?                     NaN              NaN NaN   \n",
      "2      ?                     NaN              NaN NaN   \n",
      "3     NaN                      Ø?              NaN NaN   \n",
      "4     ~Þ?                     ®à?              NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/10/15.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/10/14.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     NaN             NaN NaN   \n",
      "1      É?                     NaN             NaN NaN   \n",
      "2     £ä?                     »?             NaN NaN   \n",
      "3      à?                     NaN             NaN NaN   \n",
      "4      Þ?                     NaN             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/10/14.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/10/28.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1      ß?                     NaN             NaN  NaN   \n",
      "2      ³¿                     NaN             NaN  NaN   \n",
      "3     V¿                     NaN             NaN  NaN   \n",
      "4      F?                     NaN             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/10/28.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/10/16.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     ±ë?             NaN  NaN   \n",
      "1      Ó?                      Ï?             NaN  NaN   \n",
      "2      Õ?                     NaN             NaN  NaN   \n",
      "3     Vß?                     NaN             NaN  NaN   \n",
      "4      á?                     NaN             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/10/16.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/10/17.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                      Ñ?             NaN NaN   \n",
      "1      È¿                     NaN             NaN NaN   \n",
      "2     \u0013à?                     NaN             NaN NaN   \n",
      "3     \\Ð?                     NaN             NaN NaN   \n",
      "4     à?                      Õ?             s¿ NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/10/17.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/10/13.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1      Ú?                      Ü?             NaN  NaN   \n",
      "2      Ê?                      à?             NaN  NaN   \n",
      "3      à?                     -¢?             NaN  NaN   \n",
      "4      Ñ?                      ?             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/10/13.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/10/12.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                      í¿             SÖ?  NaN   \n",
      "1     Â¿                     NaN             NaN  NaN   \n",
      "2     à?                     NaN             NaN  NaN   \n",
      "3     ÇØ?                     NaN             NaN  NaN   \n",
      "4      Ù?                     NaN             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/10/12.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/10/10.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1     1Å?                      á?             NaN  NaN   \n",
      "2      ¿                     NaN             NaN  NaN   \n",
      "3      c¿                     NaN             NaN  NaN   \n",
      "4     Ø?                     °°?            \\tá?  \u0016¿   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/10/10.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/10/11.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1      ¢¿                     NaN             NaN  NaN   \n",
      "2      ¿                     \u0003?             NaN  NaN   \n",
      "3      Q?                     NaN             NaN  NaN   \n",
      "4     Rm¿                     ¿             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/10/11.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/19/20.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     Bã?             NaN  NaN   \n",
      "1     4ì?                     NaN             NaN  NaN   \n",
      "2     ð\u0002@                     NaN             NaN  NaN   \n",
      "3       ?                     Rá?             NaN  NaN   \n",
      "4     \u0010á?                     NaN             NaN  NaN   \n",
      "\n",
      "   }                                                           \n",
      "0                                                NaN           \n",
      "1                                                NaN           \n",
      "2                                                NaN           \n",
      "3                                                NaN           \n",
      "4                                                NaN           \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/19/20.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/19/21.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1     ÷Ñ?                      û¿              ¿  NaN   \n",
      "2     ]á?                       ?             NaN  NaN   \n",
      "3     >Õ?                     \u001d~¿             NaN  NaN   \n",
      "4     G¤?                     rÒ?             Ö?   x¿   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/19/21.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/19/23.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     NaN             NaN NaN   \n",
      "1     ýï?                     NaN             NaN NaN   \n",
      "2      ã?                      Á¿             NaN NaN   \n",
      "3      Ò?                     NaN             NaN NaN   \n",
      "4     ËÜ?                     NaN             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/19/23.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/19/22.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1       ?                     NaN             NaN  NaN   \n",
      "2     á?                     NaN             NaN  NaN   \n",
      "3     2à?                      Î?              á?   á?   \n",
      "4      V?                     NaN             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/19/22.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/19/26.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                      à?             ¿è?  2Ò?   \n",
      "1     å{¿                     NaN             NaN  NaN   \n",
      "2     |¿                     NaN             NaN  NaN   \n",
      "3      |¿                     NaN             NaN  NaN   \n",
      "4      ¿                      Õ?              Ð?  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/19/26.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/19/27.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     V\u0003@             NaN NaN   \n",
      "1     ×?                     NaN             NaN NaN   \n",
      "2     ËÕ?                     NaN             NaN NaN   \n",
      "3      Ð?                     NaN             NaN NaN   \n",
      "4      ?                     À~?             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/19/27.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/19/25.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                      Ò?             NaN  NaN   \n",
      "1     Xß?                     NaN             NaN  NaN   \n",
      "2     #\u0019?                      Ù?             NaN  NaN   \n",
      "3     \u000e\u0010?                     NaN             NaN  NaN   \n",
      "4     IÊ?                     NaN             NaN  NaN   \n",
      "\n",
      "   }                                                           \n",
      "0                                                NaN           \n",
      "1                                                NaN           \n",
      "2                                                NaN           \n",
      "3                                                NaN           \n",
      "4                                                NaN           \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/19/25.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/19/19.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     kï?             NaN  NaN   \n",
      "1      ì?                     NaN             NaN  NaN   \n",
      "2       ?                     z?             NaN  NaN   \n",
      "3      à?                     NaN             NaN  NaN   \n",
      "4     Ò?                     NaN             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/19/19.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/19/18.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     #Ô?              \u0004@ NaN   \n",
      "1       ?                     NaN             NaN NaN   \n",
      "2      ¿                      ¿             NaN NaN   \n",
      "3      Ð?                     NaN             NaN NaN   \n",
      "4      Ò?                     NaN             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/19/18.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/19/24.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                      â¿              Ò? NaN   \n",
      "1      ä?                     NaN             NaN NaN   \n",
      "2      ¢¿                     NaN             NaN NaN   \n",
      "3     ¿                      á?             NaN NaN   \n",
      "4     \u0018á?                     ÄØ?             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/19/24.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/19/5.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1      Õ¿                     NaN             NaN  NaN   \n",
      "2     jÓ?                     NaN             NaN  NaN   \n",
      "3      ê?                     \u0007¿               ?  ^Ñ?   \n",
      "4     (ß?                     NaN             NaN  NaN   \n",
      "\n",
      "   }                                                           \n",
      "0                                                NaN           \n",
      "1                                                NaN           \n",
      "2                                                NaN           \n",
      "3                                                Qj¿           \n",
      "4                                                NaN           \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/19/5.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/19/4.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     NaN             NaN NaN   \n",
      "1      ä?                      é?             NaN NaN   \n",
      "2     ¹Ö?                     NaN             NaN NaN   \n",
      "3     µÓ?                     NaN             NaN NaN   \n",
      "4      ?                     yß?             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/19/4.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/19/6.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1     rç¿                      ×?             NaN  NaN   \n",
      "2     Ø?                     NaN             NaN  NaN   \n",
      "3      ?                     NaN             NaN  NaN   \n",
      "4     9y?                     NaN             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/19/6.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/19/7.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     NaN             NaN NaN   \n",
      "1      Þ?                     %Ñ?             NaN NaN   \n",
      "2      Ø?                     ób¿             NaN NaN   \n",
      "3     Õ±?                     NaN             NaN NaN   \n",
      "4      Ù?                      Þ?             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/19/7.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/19/3.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1     ]õ?                     ¹?             Ïö?  NaN   \n",
      "2     ¡v?                     NaN             NaN  NaN   \n",
      "3      \u0004@                     NaN             NaN  NaN   \n",
      "4       ?                     NaN             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/19/3.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/19/2.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1      x¿                     NaN             NaN  NaN   \n",
      "2      ä?                      À¿             NaN  NaN   \n",
      "3     ¯Ô?                     NaN             NaN  NaN   \n",
      "4      à?                      Ð?             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/19/2.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/19/0.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                      è?             \u001b\u0004@  NaN   \n",
      "1       ?                     NaN             NaN  NaN   \n",
      "2      ¿                      ¿             NaN  NaN   \n",
      "3      ¿                       ?              Ø?  NaN   \n",
      "4      ?                     NaN             NaN  NaN   \n",
      "\n",
      "   }                                                           \n",
      "0                                                NaN           \n",
      "1                                                NaN           \n",
      "2                                                NaN           \n",
      "3                                                NaN           \n",
      "4                                                NaN           \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/19/0.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/19/1.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                      à?             Á?  NaN   \n",
      "1     bå¿                     ¹ä?             NaN  NaN   \n",
      "2     \u0005ä?                     NaN             NaN  NaN   \n",
      "3       ?                     lÑ?             NaN  NaN   \n",
      "4      ?                      Ð?             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/19/1.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/19/9.npy\n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/19/9.npy: Error tokenizing data. C error: Expected 5 fields in line 6, saw 6\n",
      "\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/19/8.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1     ýï?                      É¿             NaN  NaN   \n",
      "2      »¿                     NaN             NaN  NaN   \n",
      "3      \u0003@                     NaN             NaN  NaN   \n",
      "4     ¯÷>                     NaN             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/19/8.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/19/29.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1      ã?                     NaN             NaN  NaN   \n",
      "2     \u001bà?                     NaN             NaN  NaN   \n",
      "3      Ø?                     ëà?              à?  NaN   \n",
      "4     Ø?                     NaN             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/19/29.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/19/15.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     ðß?             \u0010û¿  NaN   \n",
      "1      ¿                     NaN             NaN  NaN   \n",
      "2      Ö?                     NaN             NaN  NaN   \n",
      "3      ®?                     !á?             NaN  NaN   \n",
      "4     â¿                     0Ù?             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/19/15.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/19/14.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1     Uâ?                      õ?             *Ð?   ¿   \n",
      "2      ß?                     NaN             NaN  NaN   \n",
      "3      ¿                      á?              á?  NaN   \n",
      "4     Þ?                     A?              ~¿  \u0002Ú?   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/19/14.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/19/28.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1      Ù?                     NaN             NaN  NaN   \n",
      "2      î?                     NaN             NaN  NaN   \n",
      "3       ?                     NaN             NaN  NaN   \n",
      "4     X\u0003@                     NaN             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/19/28.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/19/16.npy\n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/19/16.npy: Error tokenizing data. C error: Expected 5 fields in line 14, saw 7\n",
      "\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/19/17.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False   'shape': (1662   )  \\\n",
      "0     NaN                     NaN              NaN NaN   \n",
      "1     ?á¿                     NaN              NaN NaN   \n",
      "2     ý¾¿                     NaN              NaN NaN   \n",
      "3      Ö?                     NaN              NaN NaN   \n",
      "4      Ø?                     NaN              NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/19/17.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/19/13.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1     õÑ?                      Ð¿             NaN  NaN   \n",
      "2      Þ?                      à?             NaN  NaN   \n",
      "3      Ù?                     NaN             NaN  NaN   \n",
      "4     t¿                     NaN             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/19/13.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/19/12.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     ëé?             NaN  NaN   \n",
      "1      Ô?                      Þ?             NaN  NaN   \n",
      "2     #à?                     NaN             NaN  NaN   \n",
      "3     óH¿                     NaN             NaN  NaN   \n",
      "4      Í?                     NaN             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/19/12.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/19/10.npy\n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/19/10.npy: Error tokenizing data. C error: Expected 5 fields in line 18, saw 6\n",
      "\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/19/11.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                      Ó¿             NaN NaN   \n",
      "1      ¸¿                     NaN             NaN NaN   \n",
      "2     ¾û¿                     NaN             NaN NaN   \n",
      "3      \u001c?                     \u0003¿             NaN NaN   \n",
      "4     @Ó?                     NaN             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/19/11.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/21/20.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     NaN             NaN NaN   \n",
      "1      ä¿                     NaN             NaN NaN   \n",
      "2     ÿï?                     NaN             NaN NaN   \n",
      "3     ®ç?                     NaN             NaN NaN   \n",
      "4      ö?                      Á¿             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/21/20.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/21/21.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     NaN             NaN NaN   \n",
      "1     NaN                      â?             NaN NaN   \n",
      "2     µà?                     NaN             NaN NaN   \n",
      "3     ©Þ?                      w¿             NaN NaN   \n",
      "4     ®à?                      Þ?             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/21/21.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/21/23.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     NaN             NaN NaN   \n",
      "1     \u0002|?                      Ò?             NaN NaN   \n",
      "2     Ùà?                     \u0017á?             NaN NaN   \n",
      "3      Ð?                      ?             NaN NaN   \n",
      "4     hp¿                     \u0002r¿             ¿Ñ? NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/21/23.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/21/22.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1      á?                     NaN             NaN  NaN   \n",
      "2      á?                     NaN             NaN  NaN   \n",
      "3      Ð?                     NaN             NaN  NaN   \n",
      "4      à?                     NaN             NaN  NaN   \n",
      "\n",
      "   }                                                           \n",
      "0                                                NaN           \n",
      "1                                                NaN           \n",
      "2                                                NaN           \n",
      "3                                                NaN           \n",
      "4                                                NaN           \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/21/22.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/21/26.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1     $ö?                     &¿             NaN  NaN   \n",
      "2      ¿                      Ø?             há?  NaN   \n",
      "3      w?                     NaN             NaN  NaN   \n",
      "4     KÓ?                      Ð?              ?  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/21/26.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/21/27.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                      Æ¿             NaN  NaN   \n",
      "1     Óô?                     àù¿             NaN  NaN   \n",
      "2      ×?                     NaN             NaN  NaN   \n",
      "3      Ó?                      ¿             NaN  NaN   \n",
      "4     c¿                     NaN             NaN  NaN   \n",
      "\n",
      "   }                                                           \n",
      "0                                                NaN           \n",
      "1                                                NaN           \n",
      "2                                                NaN           \n",
      "3                                                NaN           \n",
      "4                                                NaN           \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/21/27.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/21/25.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                      ã?             ä? NaN   \n",
      "1     <Ö¿                     \u0010ú¿              Ø¿ NaN   \n",
      "2      þ?                     {Ô?             NaN NaN   \n",
      "3     ®Ó?                     NaN             NaN NaN   \n",
      "4    Ñ?                      á?             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/21/25.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/21/19.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     äá¿              ?  NaN   \n",
      "1      þ>                     NaN             NaN  NaN   \n",
      "2     ^á?                     NaN             NaN  NaN   \n",
      "3     \u0006Ñ?                     NaN             NaN  NaN   \n",
      "4     \u000e\u0010¿                     ´­?              Ø?  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/21/19.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/21/18.npy\n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/21/18.npy: Error tokenizing data. C error: Expected 5 fields in line 14, saw 6\n",
      "\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/21/24.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     NaN             NaN NaN   \n",
      "1     äÒ?                     NaN             NaN NaN   \n",
      "2      ö?                     NaN             NaN NaN   \n",
      "3     )Û?                     NaN             NaN NaN   \n",
      "4     PÝ?                     l?             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/21/24.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/21/5.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     ÿï?             NaN NaN   \n",
      "1      Ù?                      ÷¿              Ý? NaN   \n",
      "2      Ó?                     NaN             NaN NaN   \n",
      "3      á?                     ù¿              \\? NaN   \n",
      "4     2Ø?                     NaN             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/21/5.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/21/4.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1      ù¿                       ?             NaN  NaN   \n",
      "2       ?                     NaN             NaN  NaN   \n",
      "3     7Ý?                     NaN             NaN  NaN   \n",
      "4      à?                     NaN             NaN  NaN   \n",
      "\n",
      "   }                                                           \n",
      "0                                                NaN           \n",
      "1                                                NaN           \n",
      "2                                                NaN           \n",
      "3                                                NaN           \n",
      "4                                                NaN           \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/21/4.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/21/6.npy\n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/21/6.npy: Error tokenizing data. C error: Buffer overflow caught - possible malformed input file.\n",
      "\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/21/7.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                      ö?              Ñ? NaN   \n",
      "1     A?                     NaN             NaN NaN   \n",
      "2     Já?                     NaN             NaN NaN   \n",
      "3     |á?                     NaN             NaN NaN   \n",
      "4     [á?                     NaN             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/21/7.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/21/3.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     ÿï?             NaN  NaN   \n",
      "1     pÑ?                     ³Ó?             NaN  NaN   \n",
      "2      ø¿                       ?             NaN  NaN   \n",
      "3     ½ä?                     NaN             NaN  NaN   \n",
      "4       ?                     NaN             NaN  NaN   \n",
      "\n",
      "   }                                                           \n",
      "0                                                NaN           \n",
      "1                                                NaN           \n",
      "2                                                NaN           \n",
      "3                                                NaN           \n",
      "4                                                NaN           \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/21/3.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/21/2.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1      à¿                     NaN             NaN  NaN   \n",
      "2     ¼¿                     NaN             NaN  NaN   \n",
      "3      á?                     NaN             NaN  NaN   \n",
      "4     ßà?                     NaN             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/21/2.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/21/0.npy\n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/21/0.npy: Error tokenizing data. C error: Expected 5 fields in line 31, saw 6\n",
      "\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/21/1.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                      ß?             IÀ¿  Éå?   \n",
      "1     óô?                     NaN             NaN  NaN   \n",
      "2       ?                     NaN             NaN  NaN   \n",
      "3     \u001fÒ?                     NaN             NaN  NaN   \n",
      "4     C?                     NaN             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/21/1.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/21/9.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1     þï?                     NaN             NaN  NaN   \n",
      "2      è?                     NaN             NaN  NaN   \n",
      "3      Î¿                     NaN             NaN  NaN   \n",
      "4     kå?                       ?             Å×?  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/21/9.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/21/8.npy\n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/21/8.npy: Error tokenizing data. C error: Expected 5 fields in line 19, saw 6\n",
      "\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/21/29.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1       ?                     NaN             NaN  NaN   \n",
      "2     ?Ò?                      Ó?             NaN  NaN   \n",
      "3      Ð?                     ùÔ?             NaN  NaN   \n",
      "4      }?                     NaN             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/21/29.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/21/15.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1     Èô?                      Û?             NaN  NaN   \n",
      "2       ?                     NaN             NaN  NaN   \n",
      "3      ¿                     NaN             NaN  NaN   \n",
      "4     Eà?                     NaN             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/21/15.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/21/14.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                      â?             NaN  NaN   \n",
      "1     xÃ¿                     NaN             NaN  NaN   \n",
      "2      \u000f?                     NaN             NaN  NaN   \n",
      "3     \u0017á?                     \u0011?             NaN  NaN   \n",
      "4      Ó?                     NaN             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/21/14.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/21/28.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     GÑ?              î?  $û¿   \n",
      "1     \u0012£¿                     NaN             NaN  NaN   \n",
      "2     ká?                     NaN             NaN  NaN   \n",
      "3      Ò?                     NaN             NaN  NaN   \n",
      "4     µØ?                      Ø?              á?   á?   \n",
      "\n",
      "   }                                                           \n",
      "0                                                 þ?           \n",
      "1                                                NaN           \n",
      "2                                                NaN           \n",
      "3                                                NaN           \n",
      "4                                                NaN           \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/21/28.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/21/16.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1     næ¿                     â?             NaN  NaN   \n",
      "2     Ú«¿                      R?             NaN  NaN   \n",
      "3      ß?                     rÞ?             NaN  NaN   \n",
      "4     D?                     NaN             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/21/16.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/21/17.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                      é¿             NaN NaN   \n",
      "1      ?                     NaN             NaN NaN   \n",
      "2       ?                     NaN             NaN NaN   \n",
      "3      á?                      ¿              Þ? NaN   \n",
      "4     Ò?                     NaN             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/21/17.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/21/13.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                      ¶¿             NaN  NaN   \n",
      "1       ?                     K¢¿             NaN  NaN   \n",
      "2      ¿                      Ñ?             NaN  NaN   \n",
      "3      w¿                     ¥¿             .r¿  ÄØ?   \n",
      "4      ?                     NaN             NaN  NaN   \n",
      "\n",
      "   }                                                           \n",
      "0                                                NaN           \n",
      "1                                                NaN           \n",
      "2                                                NaN           \n",
      "3                                                 Î?           \n",
      "4                                                NaN           \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/21/13.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/21/12.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1      ç?                     5ñ>             NaN  NaN   \n",
      "2      á?                     NaN             NaN  NaN   \n",
      "3     1¿                      Ø?             °?  cØ?   \n",
      "4     \u001aÔ?                     NaN             NaN  NaN   \n",
      "\n",
      "   }                                                           \n",
      "0                                                NaN           \n",
      "1                                                NaN           \n",
      "2                                                NaN           \n",
      "3                                                 Õ?           \n",
      "4                                                NaN           \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/21/12.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/21/10.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                      å?             NaN  NaN   \n",
      "1      õ?                       ?             NaN  NaN   \n",
      "2      ã?                     NaN             NaN  NaN   \n",
      "3     \u0003@                     \\Õ?              á?  NaN   \n",
      "4     Ï}?                     NaN             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/21/10.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/21/11.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     NaN             NaN NaN   \n",
      "1     ¤Ð¿                     NaN             NaN NaN   \n",
      "2      ¯¿                     NaN             NaN NaN   \n",
      "3      á?                      ¿             NaN NaN   \n",
      "4      Þ?                     NaN             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/21/11.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/12/20.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     NaN             NaN NaN   \n",
      "1      Ø?                     NaN             NaN NaN   \n",
      "2     úé?                      Ð?             NaN NaN   \n",
      "3      Ñ?                     ªÒ?             m¿ NaN   \n",
      "4     ëà?                     NaN             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/12/20.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/12/21.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1      Ø?                     NaN             NaN  NaN   \n",
      "2     Ò¿                     NaN             NaN  NaN   \n",
      "3     \u0018Ù?                     NaN             NaN  NaN   \n",
      "4     x\u001e?                     NaN             NaN  NaN   \n",
      "\n",
      "   }                                                           \n",
      "0                                                NaN           \n",
      "1                                                NaN           \n",
      "2                                                NaN           \n",
      "3                                                NaN           \n",
      "4                                                NaN           \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/12/21.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/12/23.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     NaN             NaN NaN   \n",
      "1     §ß?                     NaN             NaN NaN   \n",
      "2     \u0018ë?                     NaN             NaN NaN   \n",
      "3      Ø?                     NaN             NaN NaN   \n",
      "4      Û?                      á?             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/12/23.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/12/22.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     NaN             NaN NaN   \n",
      "1      Ø?                      ×?             NaN NaN   \n",
      "2      ë?                     NaN             NaN NaN   \n",
      "3     á\u0002@                     NaN             NaN NaN   \n",
      "4      Û?                     NaN             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/12/22.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/12/26.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1     3ã¿                     NaN             NaN  NaN   \n",
      "2     2Ñ?                     NaN             NaN  NaN   \n",
      "3      á?                     \u0006á?             NaN  NaN   \n",
      "4      Ò?                     NaN             NaN  NaN   \n",
      "\n",
      "   }                                                           \n",
      "0                                                NaN           \n",
      "1                                                NaN           \n",
      "2                                                NaN           \n",
      "3                                                NaN           \n",
      "4                                                NaN           \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/12/26.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/12/27.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                      å¿             NaN NaN   \n",
      "1     Vß?                     zÒ?             NaN NaN   \n",
      "2       ?                      Þ?              á? NaN   \n",
      "3      m?                     NaN             NaN NaN   \n",
      "4      á?                      Þ?             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/12/27.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/12/25.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1     °è?                     NaN             NaN  NaN   \n",
      "2      Ò?                     NaN             NaN  NaN   \n",
      "3     \u0002ß?                     NaN             NaN  NaN   \n",
      "4     \u0012á?                      Ñ?              Ø?  ·\u001c?   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/12/25.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/12/19.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1     ã¿                     NaN             NaN  NaN   \n",
      "2      é?                     NaN             NaN  NaN   \n",
      "3      ®¿                      Ò¿              á?  NaN   \n",
      "4     \u001fà?                     NaN             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/12/19.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/12/18.npy\n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/12/18.npy: Error tokenizing data. C error: Expected 5 fields in line 16, saw 6\n",
      "\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/12/24.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1     Çä¿                      á?             /á?  NaN   \n",
      "2      ×?                     NaN             NaN  NaN   \n",
      "3     ìÞ?                     NaN             NaN  NaN   \n",
      "4     û¿                     ºØ?             NaN  NaN   \n",
      "\n",
      "   }                                                           \n",
      "0                                                NaN           \n",
      "1                                                NaN           \n",
      "2                                                NaN           \n",
      "3                                                NaN           \n",
      "4                                                NaN           \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/12/24.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/12/5.npy\n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/12/5.npy: Error tokenizing data. C error: Buffer overflow caught - possible malformed input file.\n",
      "\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/12/4.npy\n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/12/4.npy: Error tokenizing data. C error: Buffer overflow caught - possible malformed input file.\n",
      "\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/12/6.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     üï?             NaN NaN   \n",
      "1      ã?                     NaN             NaN NaN   \n",
      "2     þï?                     NaN             NaN NaN   \n",
      "3      õ?                     NaN             NaN NaN   \n",
      "4       ?                     NaN             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/12/6.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/12/7.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     zä?              ¿  }Þ?   \n",
      "1      Ö?                      µ?             NaN  NaN   \n",
      "2     §à?                     NaN             NaN  NaN   \n",
      "3      ­?                     NaN             NaN  NaN   \n",
      "4     kÔ?                     ¡Ú?             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/12/7.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/12/3.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                      Û?             NaN  NaN   \n",
      "1     ª£¿                     NaN             NaN  NaN   \n",
      "2     \u001aá?                     ðÔ?              Ò?  Q?   \n",
      "3     WÞ?                     NaN             NaN  NaN   \n",
      "4     \u000bá?                     NaN             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/12/3.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/12/2.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     »¿             NaN  NaN   \n",
      "1      Ø?                      Û?             NaN  NaN   \n",
      "2     ?                     NaN             NaN  NaN   \n",
      "3     \u001bß?                     NaN             NaN  NaN   \n",
      "4      Þ?                      Þ?             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/12/2.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/12/0.npy\n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/12/0.npy: Error tokenizing data. C error: Buffer overflow caught - possible malformed input file.\n",
      "\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/12/1.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     Íå¿             NaN NaN   \n",
      "1     æâ?                      Ö?             NaN NaN   \n",
      "2     ¯\u0018?                     \u001e¿             NaN NaN   \n",
      "3     tà?                      ?             NaN NaN   \n",
      "4     Ý¿                      n?             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/12/1.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/12/9.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     NaN             NaN NaN   \n",
      "1     P¼?                      Õ?             NaN NaN   \n",
      "2     ÖÍ?                      ¿             NaN NaN   \n",
      "3     /ß?                     d×?             NaN NaN   \n",
      "4     fÔ?                     NaN             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/12/9.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/12/8.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1      Ö?                      ¿             NaN  NaN   \n",
      "2     µÖ?                      á?              á?  NaN   \n",
      "3     ×?                      ?             NaN  NaN   \n",
      "4      à?                     NaN             NaN  NaN   \n",
      "\n",
      "   }                                                           \n",
      "0                                                NaN           \n",
      "1                                                NaN           \n",
      "2                                                NaN           \n",
      "3                                                NaN           \n",
      "4                                                NaN           \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/12/8.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/12/29.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     Ù?              Û?  NaN   \n",
      "1      Ò?                     yà?             NaN  NaN   \n",
      "2      ß?                     NaN             NaN  NaN   \n",
      "3     Þ?                     G`?             NaN  NaN   \n",
      "4      ¿                      Ø?             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/12/29.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/12/15.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     êï?             NaN NaN   \n",
      "1      Ñ?                     NaN             NaN NaN   \n",
      "2     ~ä?                      Ò?             NaN NaN   \n",
      "3      Û?                     NaN             NaN NaN   \n",
      "4      Ò?                     NaN             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/12/15.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/12/14.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1     \u0011ã?                     NaN             NaN  NaN   \n",
      "2      Ñ?                     4Ú?             NaN  NaN   \n",
      "3     Ñ?                      à?             NaN  NaN   \n",
      "4     ²à?                     r¿             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/12/14.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/12/28.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     NaN             NaN NaN   \n",
      "1      é?                     NaN             NaN NaN   \n",
      "2      Û?                     NaN             NaN NaN   \n",
      "3       ?                     NaN             NaN NaN   \n",
      "4     ;á?                     NaN             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/12/28.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/12/16.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1     ùï?                     NaN             NaN  NaN   \n",
      "2     \u0011ç?                     NaN             NaN  NaN   \n",
      "3      á?                     NaN             NaN  NaN   \n",
      "4     \u0019?                     NaN             NaN  NaN   \n",
      "\n",
      "   }                                                           \n",
      "0                                                NaN           \n",
      "1                                                NaN           \n",
      "2                                                NaN           \n",
      "3                                                NaN           \n",
      "4                                                NaN           \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/12/16.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/12/17.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                      Ó?             \fÜ? NaN   \n",
      "1      \u0013?                     üà?             NaN NaN   \n",
      "2      ?                     ìà?             NaN NaN   \n",
      "3      á?                     NaN             NaN NaN   \n",
      "4      Ó?                      Ù?             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/12/17.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/12/13.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     NaN             NaN NaN   \n",
      "1      õ?                     NaN             NaN NaN   \n",
      "2      Ø?                     NaN             NaN NaN   \n",
      "3     -á?                     NaN             NaN NaN   \n",
      "4      ß?                     NaN             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/12/13.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/12/12.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1      Ó?                      ­?             4?  NaN   \n",
      "2      à?                     !?              Ø?  NaN   \n",
      "3      Ð?                     NaN             NaN  NaN   \n",
      "4      á?                     NaN             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/12/12.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/12/10.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                      à?             Û¿   è?   \n",
      "1      ¿                     NaN             NaN  NaN   \n",
      "2     ÒÞ?                     NaN             NaN  NaN   \n",
      "3     OÙ?                     NaN             NaN  NaN   \n",
      "4      ×?                     NaN             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/12/10.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/12/11.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                      À?             NaN  NaN   \n",
      "1      Ñ?                      ¢¿             NaN  NaN   \n",
      "2      ¬?                     NaN             NaN  NaN   \n",
      "3      ß?                     NaN             NaN  NaN   \n",
      "4     Ýw?                      à?             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/12/11.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/13/20.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False   'shape': (1662   )  \\\n",
      "0     NaN                     NaN              NaN NaN   \n",
      "1      Ç?                     NaN              NaN NaN   \n",
      "2     Ðæ?                      Ö?              NaN NaN   \n",
      "3      Ò?                     NaN              NaN NaN   \n",
      "4     ïÐ?                     NaN              NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/13/20.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/13/21.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     NaN             NaN NaN   \n",
      "1      ä¿                     NaN             NaN NaN   \n",
      "2      ä¿                     NaN             NaN NaN   \n",
      "3      ä¿                     oÙ?             NaN NaN   \n",
      "4     KØ?                     NaN             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/13/21.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/13/23.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1     Û?                     NaN             NaN  NaN   \n",
      "2      á?                      á?             NaN  NaN   \n",
      "3     Ü¿                     NaN             NaN  NaN   \n",
      "4     yà?                     NaN             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/13/23.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/13/22.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     üï?             ç? NaN   \n",
      "1     3á?                     NaN             NaN NaN   \n",
      "2      Ò?                     NaN             NaN NaN   \n",
      "3      ß?                     NaN             NaN NaN   \n",
      "4      ?                     NaN             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/13/22.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/13/26.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                      â?              ô?  NaN   \n",
      "1      ×?                     NaN             NaN  NaN   \n",
      "2     Õõ?                     NaN             NaN  NaN   \n",
      "3      ¿                     NaN             NaN  NaN   \n",
      "4      ¿                     NaN             NaN  NaN   \n",
      "\n",
      "   }                                                           \n",
      "0                                                NaN           \n",
      "1                                                NaN           \n",
      "2                                                NaN           \n",
      "3                                                NaN           \n",
      "4                                                NaN           \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/13/26.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/13/27.npy\n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/13/27.npy: Error tokenizing data. C error: EOF inside string starting at row 20\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/13/25.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     NaN             NaN NaN   \n",
      "1     ýï?                      Þ?             NaN NaN   \n",
      "2     SÆ?                      È?             NaN NaN   \n",
      "3     \u0019á?                      á?             NaN NaN   \n",
      "4      ß?                     Ü ?             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/13/25.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/13/19.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     NaN             NaN NaN   \n",
      "1     XÓ?                      á?             NaN NaN   \n",
      "2     vÙ?                     NaN             NaN NaN   \n",
      "3     ¿¯?                     NaN             NaN NaN   \n",
      "4     ?                     _à?             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/13/19.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/13/18.npy\n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/13/18.npy: Error tokenizing data. C error: Expected 5 fields in line 8, saw 6\n",
      "\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/13/24.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     NaN             NaN NaN   \n",
      "1     üï?                     NaN             NaN NaN   \n",
      "2     ûÎ?                     NaN             NaN NaN   \n",
      "3     )à?                     'ß?             NaN NaN   \n",
      "4     ?Ì?                     NaN             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/13/24.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/13/5.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1     íï?                     NaN             NaN  NaN   \n",
      "2      ä?                     NaN             NaN  NaN   \n",
      "3     q¿                     ?ã?             NaN  NaN   \n",
      "4     Ká?                     NaN             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/13/5.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/13/4.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1     ±Ô?                     NaN             NaN  NaN   \n",
      "2      Þ?                     NaN             NaN  NaN   \n",
      "3      à?                     NaN             NaN  NaN   \n",
      "4      á?                     à?             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/13/4.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/13/6.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1     ­è?                     NaN             NaN  NaN   \n",
      "2      ·?                      á?             NaN  NaN   \n",
      "3     Mà?                     a?              ß?  NaN   \n",
      "4      u¿                     NaN             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/13/6.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/13/7.npy\n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/13/7.npy: Error tokenizing data. C error: Expected 5 fields in line 7, saw 6\n",
      "\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/13/3.npy\n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/13/3.npy: Error tokenizing data. C error: Expected 5 fields in line 5, saw 7\n",
      "\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/13/2.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     ²z?             NaN  NaN   \n",
      "1      ¿                     NaN             NaN  NaN   \n",
      "2     `à?                     NaN             NaN  NaN   \n",
      "3      Ò?                      á?             NaN  NaN   \n",
      "4      Ð?                     NaN             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/13/2.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/13/0.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     NaN             NaN NaN   \n",
      "1     \u0005Ö?                      Û?             NaN NaN   \n",
      "2      \u0004@                     NaN             NaN NaN   \n",
      "3      Ò?                     ò¿              Ð? NaN   \n",
      "4      Ø?                      V?             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/13/0.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/13/1.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     NaN             NaN NaN   \n",
      "1     Õã?                      Ô?             NaN NaN   \n",
      "2      ô?                      ¢¿             NaN NaN   \n",
      "3      ?                      Ó?             à? NaN   \n",
      "4     Nà?                     NaN             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/13/1.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/13/9.npy\n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/13/9.npy: Error tokenizing data. C error: Buffer overflow caught - possible malformed input file.\n",
      "\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/13/8.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1      ï?                      å?             NaN  NaN   \n",
      "2      û¿                     NaN             NaN  NaN   \n",
      "3      ä?                     NaN             NaN  NaN   \n",
      "4      ¿                     öÖ?              Õ?  NaN   \n",
      "\n",
      "   }                                                           \n",
      "0                                                NaN           \n",
      "1                                                NaN           \n",
      "2                                                NaN           \n",
      "3                                                NaN           \n",
      "4                                                NaN           \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/13/8.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/13/29.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     \u0014î?             NaN  NaN   \n",
      "1      ?                     NaN             NaN  NaN   \n",
      "2      ^¿                     NaN             NaN  NaN   \n",
      "3      ®?                     NaN             NaN  NaN   \n",
      "4      Ø?                      ¿             âà?  ]Ó?   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/13/29.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/13/15.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     nÔ?             NaN NaN   \n",
      "1     Ñ?                     NaN             NaN NaN   \n",
      "2     ßû¿                     Ú?             NaN NaN   \n",
      "3      ¿                     NaN             NaN NaN   \n",
      "4      Ø?                     NaN             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/13/15.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/13/14.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1     qæ?                     NaN             NaN  NaN   \n",
      "2      Ü?                     NaN             NaN  NaN   \n",
      "3     H¿                     NaN             NaN  NaN   \n",
      "4      á?                     Ì×?             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/13/14.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/13/28.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     \u001bã?             NaN NaN   \n",
      "1      ×?                      ¿             NaN NaN   \n",
      "2      ¿                     NaN             NaN NaN   \n",
      "3      j¿                     NaN             NaN NaN   \n",
      "4      ?                     ¦×?             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/13/28.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/13/16.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1      á?                     NaN             NaN  NaN   \n",
      "2     '¿                     ÞÙ?             NaN  NaN   \n",
      "3      Ñ?                      à?             ÎÐ?   ¬?   \n",
      "4     \u0004à?                     \u0010Ï?              ?  NaN   \n",
      "\n",
      "   }                                                           \n",
      "0                                                NaN           \n",
      "1                                                NaN           \n",
      "2                                                NaN           \n",
      "3                                                NaN           \n",
      "4                                                NaN           \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/13/16.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/13/17.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1      Ø?                     NaN             NaN  NaN   \n",
      "2      Ø?                     \u001eÐ¿             êé?   ä?   \n",
      "3      \u0002?                      ?              á?   á?   \n",
      "4      Ñ?                     NaN             NaN  NaN   \n",
      "\n",
      "   }                                                           \n",
      "0                                                NaN           \n",
      "1                                                NaN           \n",
      "2                                                NaN           \n",
      "3                                                NaN           \n",
      "4                                                NaN           \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/13/17.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/13/13.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                      ù¿             NaN NaN   \n",
      "1      \u0004@                     NaN             NaN NaN   \n",
      "2      á?                     NaN             NaN NaN   \n",
      "3      á?                      Õ?             NaN NaN   \n",
      "4      ?                     NaN             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/13/13.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/13/12.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     Õ?             NaN  NaN   \n",
      "1      Á¿                     NaN             NaN  NaN   \n",
      "2     Gã?                     NaN             NaN  NaN   \n",
      "3      á?                     NaN             NaN  NaN   \n",
      "4     ]¿                      Õ?             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/13/12.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/13/10.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False   'shape': (1662   )  \\\n",
      "0     NaN                     NaN              NaN NaN   \n",
      "1      ç?                     NaN              NaN NaN   \n",
      "2     ìÓ¿                     NaN              NaN NaN   \n",
      "3      Ù?                     NaN              NaN NaN   \n",
      "4     «|?                     NaN              NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/13/10.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/13/11.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                      Ñ?              æ¿ NaN   \n",
      "1      Ö?                      á?              á? NaN   \n",
      "2      à?                     ]ß?             NaN NaN   \n",
      "3      ¿                     NaN             NaN NaN   \n",
      "4     Í¨?                     NaN             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/13/11.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/14/20.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1     ýï?                     NaN             NaN  NaN   \n",
      "2     ßã¿                     NaN             NaN  NaN   \n",
      "3      à?                      Å?             NaN  NaN   \n",
      "4       ?                     NaN             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/14/20.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/14/21.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     NaN             NaN NaN   \n",
      "1     úá?                     NaN             NaN NaN   \n",
      "2      é?                     ß?             &à? NaN   \n",
      "3     ?                     NaN             NaN NaN   \n",
      "4     H?                     Zà?             Ð¿ NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/14/21.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/14/23.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     ±Ð¿             NaN  NaN   \n",
      "1     fÛ?                     NaN             NaN  NaN   \n",
      "2       ?                     NaN             NaN  NaN   \n",
      "3      ¿                     NaN             NaN  NaN   \n",
      "4      á?                      ?             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/14/23.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/14/22.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     '¿             NaN NaN   \n",
      "1      á?                     NaN             NaN NaN   \n",
      "2     à?                     NaN             NaN NaN   \n",
      "3      Ö?                     NaN             NaN NaN   \n",
      "4      Ù?                     NaN             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/14/22.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/14/26.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1      ã?                     NaN             NaN  NaN   \n",
      "2     JÝ?                     NaN             NaN  NaN   \n",
      "3      ¿                     NaN             NaN  NaN   \n",
      "4      á?                      ¥?             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/14/26.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/14/27.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1     \u001fì?                     4å?             P\u0003@   Ý?   \n",
      "2     Zà?                     NaN             NaN  NaN   \n",
      "3     mÝ?                     NaN             NaN  NaN   \n",
      "4     `à?                     ØØ?             NaN  NaN   \n",
      "\n",
      "   }                                                           \n",
      "0                                                NaN           \n",
      "1                                                 r?           \n",
      "2                                                NaN           \n",
      "3                                                NaN           \n",
      "4                                                NaN           \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/14/27.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/14/25.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     NaN             NaN NaN   \n",
      "1     üï?                     NaN             NaN NaN   \n",
      "2     ÆÝ?                     NaN             NaN NaN   \n",
      "3     Ç?                     NaN             NaN NaN   \n",
      "4      ¿                     NaN             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/14/25.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/14/19.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                      Ë?             NaN  NaN   \n",
      "1      Ð?                      à?             NaN  NaN   \n",
      "2     2Þ?                     NaN             NaN  NaN   \n",
      "3     fÐ?                      ¡?             NaN  NaN   \n",
      "4     à?                     NaN             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/14/19.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/14/18.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     NaN             NaN NaN   \n",
      "1     bà?                     NaN             NaN NaN   \n",
      "2      Ó?                     NaN             NaN NaN   \n",
      "3     çu¿                     NaN             NaN NaN   \n",
      "4     NaN                     NaN             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/14/18.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/14/24.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                      Û?             NaN  NaN   \n",
      "1       ?                     NaN             NaN  NaN   \n",
      "2      Õ?                     NaN             NaN  NaN   \n",
      "3      Ö?                      à?             «Ø?  NaN   \n",
      "4     `Ì?                     NaN             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/14/24.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/14/5.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                      á¿             NaN  NaN   \n",
      "1      Õ¿                     NaN             NaN  NaN   \n",
      "2     wá?                     NaN             NaN  NaN   \n",
      "3     wá?                     NaN             NaN  NaN   \n",
      "4      Ì?                     NaN             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/14/5.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/14/4.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1      A?                     NaN             NaN  NaN   \n",
      "2      ä?                     NaN             NaN  NaN   \n",
      "3       ?                     NaN             NaN  NaN   \n",
      "4      Ô?                     NaN             NaN  NaN   \n",
      "\n",
      "   }                                                           \n",
      "0                                                NaN           \n",
      "1                                                NaN           \n",
      "2                                                NaN           \n",
      "3                                                NaN           \n",
      "4                                                NaN           \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/14/4.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/14/6.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                      Ø?             NaN NaN   \n",
      "1     Ñ?                     NaN             NaN NaN   \n",
      "2      ×?                     NaN             NaN NaN   \n",
      "3     ­à?                      Þ?              à? NaN   \n",
      "4      Ò?                     NaN             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/14/6.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/14/7.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     NaN             NaN NaN   \n",
      "1      æ¿                     NaN             NaN NaN   \n",
      "2     æØ?                     NaN             NaN NaN   \n",
      "3      Ó?                     NaN             NaN NaN   \n",
      "4     úÖ?                     NaN             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/14/7.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/14/3.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                      \u0003@             NaN NaN   \n",
      "1      Ô?                     NaN             NaN NaN   \n",
      "2      á?                     x×?             Ä¿ NaN   \n",
      "3     ;Õ?                     NaN             NaN NaN   \n",
      "4      à?                     NaN             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/14/3.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/14/2.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1      Ð¿                     NaN             NaN  NaN   \n",
      "2      ü¿                     NaN             NaN  NaN   \n",
      "3      Õ?                     NaN             NaN  NaN   \n",
      "4     \fà?                     NaN             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/14/2.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/14/0.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     NaN             NaN NaN   \n",
      "1      è?                     NaN             NaN NaN   \n",
      "2     BË¿                     NaN             NaN NaN   \n",
      "3     2¿                     NaN             NaN NaN   \n",
      "4     9á?                      ß?             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/14/0.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/14/1.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     0à?             ¸ä?  NaN   \n",
      "1     |?                     NaN             NaN  NaN   \n",
      "2      ?                      Ò?             Ï?  NaN   \n",
      "3     ç×?                      j?             NaN  NaN   \n",
      "4     @à?                     NaN             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/14/1.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/14/9.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     `å?             NaN NaN   \n",
      "1      â?                     Yá?             NaN NaN   \n",
      "2      à?                      Ô?             NaN NaN   \n",
      "3     pà?                     NaN             NaN NaN   \n",
      "4      á?                     NaN             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/14/9.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/14/8.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     mÕ¿             NaN NaN   \n",
      "1      ÷?                     ]õ?             \u000f\u0004@ NaN   \n",
      "2      ß?                      [¿             NaN NaN   \n",
      "3      ×?                     NaN             NaN NaN   \n",
      "4     ià?                     NaN             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/14/8.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/14/29.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     Ô?             NaN  NaN   \n",
      "1     ã?                     NaN             NaN  NaN   \n",
      "2      Ï?                     NaN             NaN  NaN   \n",
      "3      Ü?                     |Ø?             NaN  NaN   \n",
      "4      Õ?                      ¿             3à?   à?   \n",
      "\n",
      "   }                                                           \n",
      "0                                                NaN           \n",
      "1                                                NaN           \n",
      "2                                                NaN           \n",
      "3                                                NaN           \n",
      "4                                                 Ó?           \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/14/29.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/14/15.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1     þï?                     NaN             NaN  NaN   \n",
      "2      Ö?                     \u00067?             NaN  NaN   \n",
      "3      á?                     NaN             NaN  NaN   \n",
      "4      ¿                     NaN             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/14/15.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/14/14.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     NaN             NaN NaN   \n",
      "1      â?                     NaN             NaN NaN   \n",
      "2      à?                     ëÔ?             NaN NaN   \n",
      "3     è?                     NaN             NaN NaN   \n",
      "4     ßû¿                     NaN             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/14/14.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/14/28.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     Úâ?             NaN NaN   \n",
      "1     Óì?                     NaN             NaN NaN   \n",
      "2     \u0006á?                     NaN             NaN NaN   \n",
      "3     PÑ?                      ®?             NaN NaN   \n",
      "4     \u0003?                     ¶Ý?             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/14/28.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/14/16.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1      Ñ?                     Ôý?             NaN  NaN   \n",
      "2     Å?                     NaN             NaN  NaN   \n",
      "3     Â?                      ¿             NaN  NaN   \n",
      "4     \u0001?                     NaN             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/14/16.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/14/17.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                      é?             NaN  NaN   \n",
      "1     ÷>                     Ü¿             NaN  NaN   \n",
      "2     ¯¿                     Þ?              Ó?  NaN   \n",
      "3       ¿                     NaN             NaN  NaN   \n",
      "4      à?                      à?             NaN  NaN   \n",
      "\n",
      "   }                                                           \n",
      "0                                                NaN           \n",
      "1                                                NaN           \n",
      "2                                                NaN           \n",
      "3                                                NaN           \n",
      "4                                                NaN           \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/14/17.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/14/13.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     NaN             NaN  NaN   \n",
      "1      Ú?                     ù¿             NaN  NaN   \n",
      "2      á?                     NaN             NaN  NaN   \n",
      "3      á?                     NaN             NaN  NaN   \n",
      "4     ÒÐ?                     NaN             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/14/13.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/14/12.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                     ÂÐ?             NaN  NaN   \n",
      "1      Ô?                      å¿             NaN  NaN   \n",
      "2      é?                     NaN             NaN  NaN   \n",
      "3      õ?                      ¿             NaN  NaN   \n",
      "4     q¢¿                      à?              {?  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/14/12.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/14/10.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662    )  \\\n",
      "0     NaN                      ö?             NaN  NaN   \n",
      "1     ^á?                     NaN             NaN  NaN   \n",
      "2      ­?                      ×?              ¿  K¿   \n",
      "3      á?                     èÞ?             NaN  NaN   \n",
      "4     úÒ?                       ?             NaN  NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/14/10.npy: 'Body'\n",
      "/Users/theerat/Downloads/MP_DATA3/สวัสดี/14/11.npy\n",
      "Columns: Index(['NUMPY\u0001', ' 'fortran_order': False', ' 'shape': (1662', ')',\n",
      "       ' }                                                         '],\n",
      "      dtype='object')\n",
      "First rows:\n",
      "   NUMPY\u0001  'fortran_order': False  'shape': (1662   )  \\\n",
      "0     NaN                     NaN             NaN NaN   \n",
      "1     eá?                     NaN             NaN NaN   \n",
      "2     è×¿                     NaN             NaN NaN   \n",
      "3      é?                     NaN             NaN NaN   \n",
      "4     Gº?                     NaN             NaN NaN   \n",
      "\n",
      "    }                                                           \n",
      "0                                                NaN            \n",
      "1                                                NaN            \n",
      "2                                                NaN            \n",
      "3                                                NaN            \n",
      "4                                                NaN            \n",
      "Error processing file /Users/theerat/Downloads/MP_DATA3/สวัสดี/14/11.npy: 'Body'\n",
      "Processing completed. Final folder processed: 14\n"
     ]
    }
   ],
   "source": [
    "import os\n",
    "import pandas as pd\n",
    "import ast\n",
    "\n",
    "base_path = '/Users/theerat/Downloads/MP_DATA3/สวัสดี'\n",
    "output_path = 'outputnpy.parqute'\n",
    "\n",
    "combined_df = pd.DataFrame()\n",
    "\n",
    "for folder in os.listdir(base_path):\n",
    "    if folder != '.DS_Store':\n",
    "        folder_path = os.path.join(base_path, folder)\n",
    "        \n",
    "        for file in os.listdir(folder_path):\n",
    "            if file == '.DS_Store':\n",
    "                continue\n",
    "            file_path = os.path.join(folder_path, file)\n",
    "            print(file_path)\n",
    "            \n",
    "            try:\n",
    "                df = pd.read_csv(file_path, encoding='ISO-8859-1')\n",
    "                \n",
    "                print(\"Columns:\", df.columns)\n",
    "                print(\"First rows:\\n\", df.head())\n",
    "\n",
    "                # Strip whitespace from column names\n",
    "                df.columns = df.columns.str.strip()\n",
    "\n",
    "                new_rows = []  # List to hold new rows\n",
    "                \n",
    "                for index, row in df.iterrows():\n",
    "                    body_landmarks = ast.literal_eval(row['Body'])\n",
    "                    right_landmarks = ast.literal_eval(row['Right'])\n",
    "                    left_landmarks = ast.literal_eval(row['Left'])\n",
    "\n",
    "                    new_rows.extend(create_landmark_rows(body_landmarks, i, index, 'body'))\n",
    "                    new_rows.extend(create_landmark_rows(right_landmarks, i, index, 'right'))\n",
    "                    new_rows.extend(create_landmark_rows(left_landmarks, i, index, 'left'))\n",
    "\n",
    "                # Append the new rows to combined_df\n",
    "                combined_df = pd.concat([combined_df, pd.DataFrame(new_rows)], ignore_index=True)\n",
    "            \n",
    "            except Exception as e:\n",
    "                print(f\"Error processing file {file_path}: {e}\")\n",
    "\n",
    "# Save combined_df to the output path if needed\n",
    "combined_df.to_parquet(output_path)\n",
    "\n",
    "print(\"Processing completed. Final folder processed:\", folder)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {},
   "outputs": [
    {
     "ename": "FileNotFoundError",
     "evalue": "[Errno 2] No such file or directory: 'outputnpy.parquet'",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mFileNotFoundError\u001b[0m                         Traceback (most recent call last)",
      "Cell \u001b[0;32mIn[82], line 1\u001b[0m\n\u001b[0;32m----> 1\u001b[0m df \u001b[38;5;241m=\u001b[39m \u001b[43mpd\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mread_parquet\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;124;43m'\u001b[39;49m\u001b[38;5;124;43moutputnpy.parquet\u001b[39;49m\u001b[38;5;124;43m'\u001b[39;49m\u001b[43m)\u001b[49m\n",
      "File \u001b[0;32m~/Documents/sign-language-translator/.venv/lib/python3.9/site-packages/pandas/io/parquet.py:667\u001b[0m, in \u001b[0;36mread_parquet\u001b[0;34m(path, engine, columns, storage_options, use_nullable_dtypes, dtype_backend, filesystem, filters, **kwargs)\u001b[0m\n\u001b[1;32m    664\u001b[0m     use_nullable_dtypes \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;01mFalse\u001b[39;00m\n\u001b[1;32m    665\u001b[0m check_dtype_backend(dtype_backend)\n\u001b[0;32m--> 667\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mimpl\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mread\u001b[49m\u001b[43m(\u001b[49m\n\u001b[1;32m    668\u001b[0m \u001b[43m    \u001b[49m\u001b[43mpath\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    669\u001b[0m \u001b[43m    \u001b[49m\u001b[43mcolumns\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mcolumns\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    670\u001b[0m \u001b[43m    \u001b[49m\u001b[43mfilters\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mfilters\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    671\u001b[0m \u001b[43m    \u001b[49m\u001b[43mstorage_options\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mstorage_options\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    672\u001b[0m \u001b[43m    \u001b[49m\u001b[43muse_nullable_dtypes\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43muse_nullable_dtypes\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    673\u001b[0m \u001b[43m    \u001b[49m\u001b[43mdtype_backend\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mdtype_backend\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    674\u001b[0m \u001b[43m    \u001b[49m\u001b[43mfilesystem\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mfilesystem\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    675\u001b[0m \u001b[43m    \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mkwargs\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    676\u001b[0m \u001b[43m\u001b[49m\u001b[43m)\u001b[49m\n",
      "File \u001b[0;32m~/Documents/sign-language-translator/.venv/lib/python3.9/site-packages/pandas/io/parquet.py:267\u001b[0m, in \u001b[0;36mPyArrowImpl.read\u001b[0;34m(self, path, columns, filters, use_nullable_dtypes, dtype_backend, storage_options, filesystem, **kwargs)\u001b[0m\n\u001b[1;32m    264\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m manager \u001b[38;5;241m==\u001b[39m \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124marray\u001b[39m\u001b[38;5;124m\"\u001b[39m:\n\u001b[1;32m    265\u001b[0m     to_pandas_kwargs[\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124msplit_blocks\u001b[39m\u001b[38;5;124m\"\u001b[39m] \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;01mTrue\u001b[39;00m  \u001b[38;5;66;03m# type: ignore[assignment]\u001b[39;00m\n\u001b[0;32m--> 267\u001b[0m path_or_handle, handles, filesystem \u001b[38;5;241m=\u001b[39m \u001b[43m_get_path_or_handle\u001b[49m\u001b[43m(\u001b[49m\n\u001b[1;32m    268\u001b[0m \u001b[43m    \u001b[49m\u001b[43mpath\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    269\u001b[0m \u001b[43m    \u001b[49m\u001b[43mfilesystem\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    270\u001b[0m \u001b[43m    \u001b[49m\u001b[43mstorage_options\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mstorage_options\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    271\u001b[0m \u001b[43m    \u001b[49m\u001b[43mmode\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[38;5;124;43mrb\u001b[39;49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[43m,\u001b[49m\n\u001b[1;32m    272\u001b[0m \u001b[43m\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    273\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[1;32m    274\u001b[0m     pa_table \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mapi\u001b[38;5;241m.\u001b[39mparquet\u001b[38;5;241m.\u001b[39mread_table(\n\u001b[1;32m    275\u001b[0m         path_or_handle,\n\u001b[1;32m    276\u001b[0m         columns\u001b[38;5;241m=\u001b[39mcolumns,\n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m    279\u001b[0m         \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39mkwargs,\n\u001b[1;32m    280\u001b[0m     )\n",
      "File \u001b[0;32m~/Documents/sign-language-translator/.venv/lib/python3.9/site-packages/pandas/io/parquet.py:140\u001b[0m, in \u001b[0;36m_get_path_or_handle\u001b[0;34m(path, fs, storage_options, mode, is_dir)\u001b[0m\n\u001b[1;32m    130\u001b[0m handles \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;01mNone\u001b[39;00m\n\u001b[1;32m    131\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m (\n\u001b[1;32m    132\u001b[0m     \u001b[38;5;129;01mnot\u001b[39;00m fs\n\u001b[1;32m    133\u001b[0m     \u001b[38;5;129;01mand\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m is_dir\n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m    138\u001b[0m     \u001b[38;5;66;03m# fsspec resources can also point to directories\u001b[39;00m\n\u001b[1;32m    139\u001b[0m     \u001b[38;5;66;03m# this branch is used for example when reading from non-fsspec URLs\u001b[39;00m\n\u001b[0;32m--> 140\u001b[0m     handles \u001b[38;5;241m=\u001b[39m \u001b[43mget_handle\u001b[49m\u001b[43m(\u001b[49m\n\u001b[1;32m    141\u001b[0m \u001b[43m        \u001b[49m\u001b[43mpath_or_handle\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mmode\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mis_text\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;28;43;01mFalse\u001b[39;49;00m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mstorage_options\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mstorage_options\u001b[49m\n\u001b[1;32m    142\u001b[0m \u001b[43m    \u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    143\u001b[0m     fs \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;01mNone\u001b[39;00m\n\u001b[1;32m    144\u001b[0m     path_or_handle \u001b[38;5;241m=\u001b[39m handles\u001b[38;5;241m.\u001b[39mhandle\n",
      "File \u001b[0;32m~/Documents/sign-language-translator/.venv/lib/python3.9/site-packages/pandas/io/common.py:882\u001b[0m, in \u001b[0;36mget_handle\u001b[0;34m(path_or_buf, mode, encoding, compression, memory_map, is_text, errors, storage_options)\u001b[0m\n\u001b[1;32m    873\u001b[0m         handle \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mopen\u001b[39m(\n\u001b[1;32m    874\u001b[0m             handle,\n\u001b[1;32m    875\u001b[0m             ioargs\u001b[38;5;241m.\u001b[39mmode,\n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m    878\u001b[0m             newline\u001b[38;5;241m=\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124m\"\u001b[39m,\n\u001b[1;32m    879\u001b[0m         )\n\u001b[1;32m    880\u001b[0m     \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[1;32m    881\u001b[0m         \u001b[38;5;66;03m# Binary mode\u001b[39;00m\n\u001b[0;32m--> 882\u001b[0m         handle \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mopen\u001b[39;49m\u001b[43m(\u001b[49m\u001b[43mhandle\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mioargs\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mmode\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    883\u001b[0m     handles\u001b[38;5;241m.\u001b[39mappend(handle)\n\u001b[1;32m    885\u001b[0m \u001b[38;5;66;03m# Convert BytesIO or file objects passed with an encoding\u001b[39;00m\n",
      "\u001b[0;31mFileNotFoundError\u001b[0m: [Errno 2] No such file or directory: 'outputnpy.parquet'"
     ]
    }
   ],
   "source": [
    "df = pd.read_parquet('outputnpy.parquet')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": ".venv",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}