4. WD + LSTM.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# A Novel Deep Learning Approach for Wind Power Forecasting Based on WD-LSTM Model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Import all the needed packages"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Basic Data Science Packages\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "\n",
    "#TensorFlow packages required for building CNN + LSTM model\n",
    "import tensorflow as tf\n",
    "from tensorflow import keras\n",
    "from tensorflow.keras.layers import BatchNormalization, Input, Dense, LSTM, Reshape\n",
    "from tensorflow.keras.models import Sequential\n",
    "\n",
    "#Plotting graph\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "#Plotting Keras Model\n",
    "from tensorflow.keras.utils import plot_model\n",
    "\n",
    "#Wavelet Decomposition package\n",
    "import pywt\n",
    "\n",
    "#Normalization of data\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "\n",
    "#Utility Module for computing and displaying metrics\n",
    "from utility_functions import metrics\n",
    "\n",
    "import warnings\n",
    "warnings.filterwarnings(\"ignore\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Performing Wavelet Decomposition"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "db3 = pywt.Wavelet('db3')\n",
    "scaler = StandardScaler()\n",
    "def normalizeAndDecompose(df_input):\n",
    "    df = pd.DataFrame(df_input)\n",
    "    normalized_df =  scaler.fit_transform(df)\n",
    "    decomposed_df = pywt.wavedec2(normalized_df, db3)\n",
    "    normalized_decomposed_df = pd.DataFrame(np.concatenate(decomposed_df))\n",
    "    for index, name in enumerate(df.columns):\n",
    "            df[name] = normalized_decomposed_df[index].values\n",
    "    \n",
    "    return df if isinstance(df_input, pd.DataFrame) else df.squeeze()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "def inverse_normalizeAndDecompose(df_input):\n",
    "    df = pd.DataFrame(df_input)\n",
    "    return np.concatenate(pywt.wavedec2(scaler.inverse_transform(df), db3))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Import the data and split into Train, Test and Validation\n",
    "- Fetch the first 70% records as train and the rest 15% each as validation and test\n",
    "    - 8711 * 70% = 6097 records from the start (Train)\n",
    "    - 6097 + 1307 = 6098 to 7404 records as (Validation)\n",
    "    - 7404 + 1307 = 7405 to 8711 records as (Test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "thesis_dataset = pd.read_csv('data/final_thesis_dataset.csv',parse_dates=[0], index_col=0)\n",
    "\n",
    "#Extracting dataset into train, validation and test sets\n",
    "train = thesis_dataset[:6097]\n",
    "valid = thesis_dataset[6097:7404]\n",
    "test = thesis_dataset[7404:]\n",
    "\n",
    "#Extracting solar and wind X, y columns per set\n",
    "X_solar_train = normalizeAndDecompose(train[['SWTDN', 'SWGDN', 'T']])\n",
    "y_solar_train = normalizeAndDecompose(train['DE_solar_generation_actual'])\n",
    "X_solar_valid = normalizeAndDecompose(valid[['SWTDN', 'SWGDN', 'T']])\n",
    "y_solar_valid = normalizeAndDecompose(valid['DE_solar_generation_actual'])\n",
    "X_solar_test = normalizeAndDecompose(test[['SWTDN', 'SWGDN', 'T']])\n",
    "y_solar_test = normalizeAndDecompose(test['DE_solar_generation_actual'])\n",
    "\n",
    "X_wind_train = normalizeAndDecompose(train[['v1', 'v2', 'v_50m', 'z0']])\n",
    "y_wind_train = normalizeAndDecompose(train['DE_wind_generation_actual'])\n",
    "X_wind_valid = normalizeAndDecompose(valid[['v1', 'v2', 'v_50m', 'z0']])\n",
    "y_wind_valid = normalizeAndDecompose(valid['DE_wind_generation_actual'])\n",
    "X_wind_test = normalizeAndDecompose(test[['v1', 'v2', 'v_50m', 'z0']])\n",
    "y_wind_test = normalizeAndDecompose(test['DE_wind_generation_actual'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Building the Model using LSTM layer and Fully Connected Layer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "reshape (Reshape)            (None, 3, 1)              0         \n",
      "_________________________________________________________________\n",
      "lstm (LSTM)                  (None, 3, 64)             16896     \n",
      "_________________________________________________________________\n",
      "lstm_1 (LSTM)                (None, 64)                33024     \n",
      "_________________________________________________________________\n",
      "dense (Dense)                (None, 64)                4160      \n",
      "_________________________________________________________________\n",
      "batch_normalization (BatchNo (None, 64)                256       \n",
      "_________________________________________________________________\n",
      "dense_1 (Dense)              (None, 1)                 65        \n",
      "=================================================================\n",
      "Total params: 54,401\n",
      "Trainable params: 54,273\n",
      "Non-trainable params: 128\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "solar_model = Sequential()\n",
    "#Mapping to the input size\n",
    "solar_model.add(Input(shape=(3,)))\n",
    "#Reshaping for the RNN Layer to perform\n",
    "solar_model.add(Reshape((-1,1)))\n",
    "#Creating 2 LSTMs with 64 units and relu Activation\n",
    "solar_model.add(LSTM(64, activation = 'relu', input_shape = X_solar_train.shape, return_sequences=True))\n",
    "solar_model.add(LSTM(64, activation = 'relu', return_sequences=False))\n",
    "#Attaching a Dense Layer of the same unit size\n",
    "solar_model.add(Dense(64))\n",
    "#Batch Normalization on the output from the dense layer\n",
    "solar_model.add(BatchNormalization())\n",
    "#Extracting single output using Dense of 1\n",
    "solar_model.add(Dense(1))\n",
    "solar_model.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Compiling the model by passing adam optimizer, loss=MAE"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/15\n",
      "191/191 [==============================] - 4s 10ms/step - loss: 0.3000 - val_loss: 0.6827\n",
      "Epoch 2/15\n",
      "191/191 [==============================] - 1s 3ms/step - loss: 0.2280 - val_loss: 0.5863\n",
      "Epoch 3/15\n",
      "191/191 [==============================] - 1s 3ms/step - loss: 0.2128 - val_loss: 0.3630\n",
      "Epoch 4/15\n",
      "191/191 [==============================] - 1s 3ms/step - loss: 0.2082 - val_loss: 0.5580\n",
      "Epoch 5/15\n",
      "191/191 [==============================] - 1s 3ms/step - loss: 0.2083 - val_loss: 0.3872\n",
      "Epoch 6/15\n",
      "191/191 [==============================] - 1s 3ms/step - loss: 0.2115 - val_loss: 0.1734\n",
      "Epoch 7/15\n",
      "191/191 [==============================] - 1s 3ms/step - loss: 0.1903 - val_loss: 0.1445\n",
      "Epoch 8/15\n",
      "191/191 [==============================] - 1s 3ms/step - loss: 0.2081 - val_loss: 0.4142\n",
      "Epoch 9/15\n",
      "191/191 [==============================] - 1s 3ms/step - loss: 0.1917 - val_loss: 0.1617\n",
      "Epoch 10/15\n",
      "191/191 [==============================] - 1s 3ms/step - loss: 0.1823 - val_loss: 0.1580\n",
      "Epoch 11/15\n",
      "191/191 [==============================] - 1s 3ms/step - loss: 0.1970 - val_loss: 0.2420\n",
      "Epoch 12/15\n",
      "191/191 [==============================] - 1s 3ms/step - loss: 0.1830 - val_loss: 0.1968\n",
      "Epoch 13/15\n",
      "191/191 [==============================] - 1s 3ms/step - loss: 0.1872 - val_loss: 0.1590\n",
      "Epoch 14/15\n",
      "191/191 [==============================] - 1s 3ms/step - loss: 0.1850 - val_loss: 0.3086\n",
      "Epoch 15/15\n",
      "191/191 [==============================] - 1s 3ms/step - loss: 0.1877 - val_loss: 0.2182\n"
     ]
    }
   ],
   "source": [
    "solar_model.compile(optimizer = 'adam', loss = 'mean_absolute_error')\n",
    "history = solar_model.fit(X_solar_train, y_solar_train, epochs = 15, validation_data = (X_solar_valid, y_solar_valid))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "41/41 [==============================] - 0s 1ms/step - loss: 0.1969\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "0.19693244993686676"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "solar_model.evaluate(X_solar_test, y_solar_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Plotting the Keras model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "plot_model(\n",
    "    solar_model, to_file='figs/4-model.png', show_shapes=True, show_dtype=False,\n",
    "    show_layer_names=True, rankdir='TB', expand_nested=False, dpi=96\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Plotting Model's Loss Metrics over Train and Validation sets for Solar"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(history.history['loss'])\n",
    "plt.plot(history.history['val_loss'])\n",
    "plt.title('model loss')\n",
    "plt.ylabel('loss')\n",
    "plt.xlabel('epoch')\n",
    "plt.legend(['train', 'validation'], loc='upper left')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Plotting Actual vs Predictions on the same plot"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_solar_pred = solar_model.predict(X_solar_train)\n",
    "val_solar_pred = solar_model.predict(X_solar_valid)\n",
    "test_solar_pred = solar_model.predict(X_solar_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "df = thesis_dataset[['DE_solar_generation_actual']]\n",
    "\n",
    "train_df = train[['DE_solar_generation_actual']]\n",
    "train_df.loc[:, 'DE_solar_generation_actual'] = inverse_normalizeAndDecompose(train_solar_pred)\n",
    "\n",
    "valid_df = valid[['DE_solar_generation_actual']]\n",
    "valid_df.loc[:, 'DE_solar_generation_actual'] = inverse_normalizeAndDecompose(val_solar_pred)\n",
    "\n",
    "test_df = test[['DE_solar_generation_actual']]\n",
    "test_df.loc[:, 'DE_solar_generation_actual'] = inverse_normalizeAndDecompose(test_solar_pred)\n",
    "\n",
    "# Plot all predictions\n",
    "inversetransform, =plt.plot(df, label = 'Actual')\n",
    "train_solar_predx, =plt.plot(train_df, color='orange', label = 'Train')\n",
    "val_solar_predx, =plt.plot(valid_df, color='green', label = 'Validation')\n",
    "test_solar_predx, =plt.plot(test_df, color='black', label = 'Test')\n",
    "plt.xlabel('Date')\n",
    "plt.ylabel('Solar Output')\n",
    "plt.legend(loc=\"upper right\")\n",
    "plt.title(\"Predicted vs. Actual Solar Generation\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Regression Metrics for Solar"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "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>MAE</th>\n",
       "      <th>MSE</th>\n",
       "      <th>RMSE</th>\n",
       "      <th>RMSLE</th>\n",
       "      <th>R2</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>model</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>wd-lstm</th>\n",
       "      <td>11949.779085</td>\n",
       "      <td>14463.850273</td>\n",
       "      <td>120.265749</td>\n",
       "      <td>4.789704</td>\n",
       "      <td>-36.716601</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                  MAE           MSE        RMSE     RMSLE         R2\n",
       "model                                                               \n",
       "wd-lstm  11949.779085  14463.850273  120.265749  4.789704 -36.716601"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "metrics('wd-lstm', test[['DE_solar_generation_actual']], inverse_normalizeAndDecompose(test_solar_pred))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Modelling and Fitting the model for Wind Dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_1\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "reshape_1 (Reshape)          (None, 4, 1)              0         \n",
      "_________________________________________________________________\n",
      "lstm_2 (LSTM)                (None, 4, 64)             16896     \n",
      "_________________________________________________________________\n",
      "lstm_3 (LSTM)                (None, 64)                33024     \n",
      "_________________________________________________________________\n",
      "dense_2 (Dense)              (None, 64)                4160      \n",
      "_________________________________________________________________\n",
      "batch_normalization_1 (Batch (None, 64)                256       \n",
      "_________________________________________________________________\n",
      "dense_3 (Dense)              (None, 1)                 65        \n",
      "=================================================================\n",
      "Total params: 54,401\n",
      "Trainable params: 54,273\n",
      "Non-trainable params: 128\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "wind_model = Sequential()\n",
    "wind_model.add(Input(shape=(4,)))\n",
    "wind_model.add(Reshape((-1,1)))\n",
    "#Creating 2 LSTMs with 64 units and relu Activation\n",
    "wind_model.add(LSTM(64, activation = 'relu', input_shape = X_wind_train.shape, return_sequences=True))\n",
    "wind_model.add(LSTM(64, activation = 'relu', return_sequences=False))\n",
    "#Attaching a Dense Layer of the same unit size\n",
    "wind_model.add(Dense(64))\n",
    "#Batch Normalization on the output from the dense layer\n",
    "wind_model.add(BatchNormalization())\n",
    "#Extracting single output using Dense of 1\n",
    "wind_model.add(Dense(1))\n",
    "wind_model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/15\n",
      "191/191 [==============================] - 3s 5ms/step - loss: 0.2409 - val_loss: 0.8782\n",
      "Epoch 2/15\n",
      "191/191 [==============================] - 1s 4ms/step - loss: 0.1434 - val_loss: 0.6207\n",
      "Epoch 3/15\n",
      "191/191 [==============================] - 1s 4ms/step - loss: 0.1202 - val_loss: 0.4070\n",
      "Epoch 4/15\n",
      "191/191 [==============================] - 1s 4ms/step - loss: 0.1018 - val_loss: 0.1223\n",
      "Epoch 5/15\n",
      "191/191 [==============================] - 1s 4ms/step - loss: 0.1156 - val_loss: 0.1788\n",
      "Epoch 6/15\n",
      "191/191 [==============================] - 1s 4ms/step - loss: 0.0956 - val_loss: 0.1833\n",
      "Epoch 7/15\n",
      "191/191 [==============================] - 1s 4ms/step - loss: 0.1069 - val_loss: 0.1135\n",
      "Epoch 8/15\n",
      "191/191 [==============================] - 1s 3ms/step - loss: 0.1024 - val_loss: 0.0886\n",
      "Epoch 9/15\n",
      "191/191 [==============================] - 1s 4ms/step - loss: 0.0973 - val_loss: 0.0909\n",
      "Epoch 10/15\n",
      "191/191 [==============================] - 1s 4ms/step - loss: 0.0948 - val_loss: 0.0773\n",
      "Epoch 11/15\n",
      "191/191 [==============================] - 1s 4ms/step - loss: 0.0993 - val_loss: 0.1104\n",
      "Epoch 12/15\n",
      "191/191 [==============================] - 1s 4ms/step - loss: 0.0999 - val_loss: 0.1131\n",
      "Epoch 13/15\n",
      "191/191 [==============================] - 1s 4ms/step - loss: 0.0978 - val_loss: 0.0826\n",
      "Epoch 14/15\n",
      "191/191 [==============================] - 1s 4ms/step - loss: 0.0960 - val_loss: 0.2141\n",
      "Epoch 15/15\n",
      "191/191 [==============================] - 1s 3ms/step - loss: 0.0917 - val_loss: 0.1395\n"
     ]
    }
   ],
   "source": [
    "wind_model.compile(optimizer = 'adam', loss = 'mse')\n",
    "history = wind_model.fit(X_wind_train, y_wind_train, validation_data=(X_wind_valid, y_wind_valid), epochs=15)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "41/41 [==============================] - 0s 1ms/step - loss: 0.1257\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "0.12566658854484558"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "wind_model.evaluate(X_wind_test, y_wind_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Model's Loss for Wind Dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(history.history['loss'])\n",
    "plt.plot(history.history['val_loss'])\n",
    "plt.title('model loss')\n",
    "plt.ylabel('loss')\n",
    "plt.xlabel('epoch')\n",
    "plt.legend(['train', 'validation'], loc='upper left')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Plotting Actual vs Predicted for Wind on same graph"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_wind_pred = wind_model.predict(X_wind_train)\n",
    "val_wind_pred = wind_model.predict(X_wind_valid)\n",
    "test_wind_pred = wind_model.predict(X_wind_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "df = thesis_dataset[['DE_wind_generation_actual']]\n",
    "\n",
    "train_df = train[['DE_wind_generation_actual']]\n",
    "train_df.loc[:, 'DE_wind_generation_actual'] = inverse_normalizeAndDecompose(train_wind_pred)\n",
    "\n",
    "valid_df = valid[['DE_wind_generation_actual']]\n",
    "valid_df.loc[:, 'DE_wind_generation_actual'] = inverse_normalizeAndDecompose(val_wind_pred)\n",
    "\n",
    "test_df = test[['DE_wind_generation_actual']]\n",
    "test_df.loc[:, 'DE_wind_generation_actual'] = inverse_normalizeAndDecompose(test_wind_pred)\n",
    "\n",
    "# Plot all predictions\n",
    "inversetransform, =plt.plot(df, label = 'Actual')\n",
    "train_wind_predx, =plt.plot(train_df, color='orange', label = 'Train')\n",
    "val_wind_predx, =plt.plot(valid_df, color='green', label = 'Validation')\n",
    "test_wind_predx, =plt.plot(test_df, color='black', label = 'Test')\n",
    "plt.xlabel('Date')\n",
    "plt.ylabel('Wind Output')\n",
    "plt.legend(loc=\"upper right\")\n",
    "plt.title(\"Predicted vs. Actual Wind Generation\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Regression Metrics for Wind"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "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>MAE</th>\n",
       "      <th>MSE</th>\n",
       "      <th>RMSE</th>\n",
       "      <th>RMSLE</th>\n",
       "      <th>R2</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>model</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>wd-lstm</th>\n",
       "      <td>2519.725587</td>\n",
       "      <td>3064.454737</td>\n",
       "      <td>55.357517</td>\n",
       "      <td>4.013812</td>\n",
       "      <td>0.874333</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 MAE          MSE       RMSE     RMSLE        R2\n",
       "model                                                           \n",
       "wd-lstm  2519.725587  3064.454737  55.357517  4.013812  0.874333"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "metrics('wd-lstm', test[['DE_wind_generation_actual']], inverse_normalizeAndDecompose(test_wind_pred))"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.8.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}