face_analysis.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "5l-GR-UThGHx"
   },
   "source": [
    "# Transfer learning for face analysis\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "4iZjBcCdjFGl"
   },
   "source": [
    "by Clément Rebut, Matthew Rousseau and Sophie Starck"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "3TWxWy5zv_4W"
   },
   "source": [
    "The goal of this notebook is to present two methods for face comparison: given two faces, our model will be able to tell if it is the same or not. We first explore a simple distance between two set of features, and then a siamese architecture, with a comparative performance analysis."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "7GUu7Boqwuxn"
   },
   "source": [
    "* [Data Management](#dataman)\n",
    "* [Euclidean Distance](#euclidean)\n",
    "* [Siamese Architecture](#siamese)\n",
    "* [Comparative Performance analysis](#analysis)\n",
    "* [Demo](#demo)\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "DL4tMfvkolDh"
   },
   "source": [
    "**Package installations**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "id": "8IRmrk3Igq1g"
   },
   "outputs": [],
   "source": [
    "import tensorflow\n",
    "import keras\n",
    "import cv2\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import sklearn"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "id": "4oULiSHyhgZr",
    "outputId": "fadba8c8-6a5a-4d31-a80a-dc0f1376c287"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Requirement already satisfied: keras_applications in /Users/sophie/opt/anaconda3/lib/python3.8/site-packages (1.0.8)\n",
      "Requirement already satisfied: h5py in /Users/sophie/opt/anaconda3/lib/python3.8/site-packages (from keras_applications) (2.10.0)\n",
      "Requirement already satisfied: numpy>=1.9.1 in /Users/sophie/opt/anaconda3/lib/python3.8/site-packages (from keras_applications) (1.19.4)\n",
      "Requirement already satisfied: six in /Users/sophie/opt/anaconda3/lib/python3.8/site-packages (from h5py->keras_applications) (1.15.0)\n",
      "Requirement already satisfied: keras_vggface in /Users/sophie/opt/anaconda3/lib/python3.8/site-packages (0.6)\n",
      "Requirement already satisfied: pillow in /Users/sophie/opt/anaconda3/lib/python3.8/site-packages (from keras_vggface) (8.0.1)\n",
      "Requirement already satisfied: pyyaml in /Users/sophie/opt/anaconda3/lib/python3.8/site-packages (from keras_vggface) (5.3.1)\n",
      "Requirement already satisfied: numpy>=1.9.1 in /Users/sophie/opt/anaconda3/lib/python3.8/site-packages (from keras_vggface) (1.19.4)\n",
      "Requirement already satisfied: six>=1.9.0 in /Users/sophie/opt/anaconda3/lib/python3.8/site-packages (from keras_vggface) (1.15.0)\n",
      "Requirement already satisfied: keras in /Users/sophie/opt/anaconda3/lib/python3.8/site-packages (from keras_vggface) (2.4.3)\n",
      "Requirement already satisfied: scipy>=0.14 in /Users/sophie/opt/anaconda3/lib/python3.8/site-packages (from keras_vggface) (1.5.2)\n",
      "Requirement already satisfied: h5py in /Users/sophie/opt/anaconda3/lib/python3.8/site-packages (from keras_vggface) (2.10.0)\n"
     ]
    }
   ],
   "source": [
    "!pip install keras_applications \n",
    "!pip install keras_vggface"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "id": "b--ODtXIh5iT"
   },
   "outputs": [],
   "source": [
    "from keras_vggface.vggface import VGGFace\n",
    "import keras_vggface\n",
    "import keras_applications"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "wCNhiAdR4rWC"
   },
   "source": [
    "## Data management <a class=\"anchor\" id=\"dataman\"></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "Y6MOr3CMosUC"
   },
   "source": [
    "### Data loading"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "id": "pVsg9sCziTje",
    "outputId": "7cb7200f-0899-4cff-ead9-1cc636222cab"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['x_0', 'x_1', 'y', 'orig_ids']"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "path = \"../TP2/cacd_vs_img_bgr_preprocessed.npz\"\n",
    "\n",
    "data = np.load(path, allow_pickle=True)\n",
    "data.files"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "id": "oOPmRERJk3Ua"
   },
   "outputs": [],
   "source": [
    "x_0 = data['x_0']\n",
    "x_1 = data['x_1']\n",
    "y = data['y']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "48egzSdRziP4"
   },
   "source": [
    "### Data Visualisation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "id": "O9Oq-xhbCkmU"
   },
   "outputs": [],
   "source": [
    "def to_rgb(img):\n",
    "    return cv2.cvtColor(img, cv2.COLOR_BGR2RGB)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "id": "VKJmoJLckgyH",
    "outputId": "794707d9-14e7-4677-fcca-55fd8ccb2451"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Y = 1\n"
     ]
    }
   ],
   "source": [
    "# positive pair\n",
    "plt.subplot(121)\n",
    "\n",
    "img_float = x_0[100]\n",
    "img_disp = (img_float - img_float.min()) / (img_float.max() - img_float.min())\n",
    "plt.imshow(to_rgb(img_disp))\n",
    "\n",
    "plt.subplot(122)\n",
    "img_float = x_1[100]\n",
    "img_disp = (img_float - img_float.min()) / (img_float.max() - img_float.min())\n",
    "plt.imshow(to_rgb(img_disp))\n",
    "plt.show()\n",
    "print(f\"Y = {y[100]}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "id": "jty2eQWSk--i",
    "outputId": "16cae7d8-716d-4ae2-b438-d7a7114d17b6"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Y = 0\n"
     ]
    }
   ],
   "source": [
    "# negative pair\n",
    "plt.subplot(121)\n",
    "\n",
    "img_float = x_0[200]\n",
    "img_disp = (img_float - img_float.min()) / (img_float.max() - img_float.min())\n",
    "plt.imshow(to_rgb(img_disp))\n",
    "\n",
    "plt.subplot(122)\n",
    "img_float = x_1[200]\n",
    "img_disp = (img_float - img_float.min()) / (img_float.max() - img_float.min())\n",
    "plt.imshow(to_rgb(img_disp))\n",
    "plt.show()\n",
    "print(f\"Y = {y[200]}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "del(img_float)\n",
    "del(img_disp)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "2-fq7x2GoxH8"
   },
   "source": [
    "## Euclidean distance <a class=\"anchor\" id=\"euclidean\"></a>\n",
    "The goal in the practical was to use vggface to compute features on the faces and then compute the distance between the features of two faces to determine if they were the same person or not. We are using a `resnet50` backend because it gave better results."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "id": "851nwOBmlQFN"
   },
   "outputs": [],
   "source": [
    "vggface = VGGFace(model='resnet50', include_top=False, input_shape=(224, 224, 3), pooling='avg')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "PYn2-LwNm_hR"
   },
   "source": [
    "We first compute the features for all faces."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "id": "GS_DOd-dm57y"
   },
   "outputs": [],
   "source": [
    "Y_pred_0 = vggface.predict(x_0[:400])\n",
    "Y_pred_1 = vggface.predict(x_1[:400])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "9Te3VieQnIj4"
   },
   "source": [
    "We then compute the distance between each pair of faces from the extracted features."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "id": "6tcjOgqun_yi"
   },
   "outputs": [],
   "source": [
    "# euclidean distance\n",
    "dist = np.array([np.linalg.norm(Y_pred_1[i]-Y_pred_0[i], ord=2) for i in range(400)])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "del(Y_pred_0)\n",
    "del(Y_pred_1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "P57irE2Ww3ob"
   },
   "source": [
    "We now verify that the two classes can be distiguished from only the distance."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "id": "XDRkPLxNq88r",
    "outputId": "653af6dd-de4b-4908-8dfc-57c55cd86104"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "133.90572 86.75831\n"
     ]
    }
   ],
   "source": [
    "print(dist[y[:400]==0].mean(), dist[y[:400]==1].mean())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "mn0wt2Q3xF_R"
   },
   "source": [
    "We can notice that the mean of the two classes is different so we will be able to classify the data to a certain degree.\n",
    "\n",
    "Now We want to display the histograms to check for the distribution of both classes and also their cumulative distribution function."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "id": "IE7R59TPpJVQ",
    "outputId": "601d48d3-ce39-4c96-cd2d-fff67037b0d1"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x360 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(15, 5))\n",
    "plt.subplot(121)\n",
    "# positive matches\n",
    "plt.hist(dist[:200], bins=64, label='Positive')\n",
    "# negative matches\n",
    "plt.hist(dist[200:],bins=64, label='Negative')\n",
    "plt.title(\"Histogram\")\n",
    "plt.legend()\n",
    "\n",
    "plt.subplot(122)\n",
    "# positive matches\n",
    "plt.hist(dist[:200], bins=200, density=True, histtype='step', cumulative=1, label='Positive')\n",
    "# negative matches\n",
    "plt.hist(dist[200:], bins=200, density=True, histtype='step', cumulative=-1, label='Negative')\n",
    "plt.title(\"Cumulative Distribution\")\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "id": "Iq-A6webwyOW"
   },
   "outputs": [],
   "source": [
    "dist = 1 - (dist - min(dist))/(max(dist) - min(dist))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "kCuHNVphnloU"
   },
   "source": [
    "### Computing the threshold"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "id": "e8jLImxsva58",
    "outputId": "f6ede0b7-7e92-4b24-905d-931962c37604"
   },
   "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"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "area under the curve = 0.9920249999999999\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import roc_curve, roc_auc_score\n",
    "\n",
    "fpr, tpr, thresholds = roc_curve(y[:400], dist)\n",
    "plt.plot(fpr, tpr)\n",
    "plt.xlabel('False Positive Rate')\n",
    "plt.ylabel('True Positive Rate')\n",
    "plt.suptitle('Roc Curve')\n",
    "plt.show()\n",
    "print(f\"area under the curve = {roc_auc_score(y[:400], dist)}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "UZo9QVECyT7z"
   },
   "source": [
    "We can notice that the area under the ROC curve is quite high, which is what we want for a good classification result."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "id": "uax4s-Tn5WCG",
    "outputId": "f3c1a779-71e5-436a-afcf-8f6e5285c5cd"
   },
   "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(tpr-fpr)\n",
    "plt.title('True positive and False positive rate difference')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "jy14iJ9Ey18i"
   },
   "source": [
    "We are computing the difference between the true positive and false positive rates to decide which threshold we should use. We wanted to experiment with automatically choosing the best threshold and we came up with this metric from the ROC curve. We could probably weight the rates in order to adapt the threshold based on user preferences."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "id": "X4y0LzYW3J7E",
    "outputId": "7e2687d8-c331-4d89-8203-f5d404924bd2"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "x:4, fpr:0.005, tpr:0.94, threshold:0.4440177083015442\n"
     ]
    },
    {
     "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": [
    "x = np.argmax(tpr-fpr)\n",
    "print(f'x:{x}, fpr:{fpr[x]}, tpr:{tpr[x]}, threshold:{thresholds[x]}')\n",
    "plt.plot(thresholds)\n",
    "plt.title('Thresholds')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "lWdu0w7Xzcxs"
   },
   "source": [
    "Now that we decided which false positive rate we can tolerate, we can select the corresponding threshold that will be used for the classification."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "id": "Bl_zlqfn3Tol",
    "outputId": "de080309-556e-4e91-8d5a-b9da7f198610"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(15, 5))\n",
    "\n",
    "# positive matches\n",
    "plt.hist(dist[:200], bins=64, label='Positive')\n",
    "# negative matches\n",
    "plt.hist(dist[200:],bins=64, label='Negative')\n",
    "\n",
    "plt.axvline(x=thresholds[x], c='r', label='Threshold')\n",
    "plt.legend()\n",
    "plt.title('Histogram distribution with classification threshold')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "76beJJpc0NxU"
   },
   "source": [
    "We can see that the threshold is located between both classes. We can also notice that in this case, the false negative rate might be a bit high."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "PWw-mXXN0j5k"
   },
   "source": [
    "### Classification\n",
    "\n",
    "Now that we know the threshold between both classes, we can apply this threshold to our distances to generate a prediction."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "id": "LwcutoSR5-AZ"
   },
   "outputs": [],
   "source": [
    "dist[dist < thresholds[x]] = 0\n",
    "dist[dist >= thresholds[x]] = 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "id": "k1NE6ZPp6Gu5",
    "outputId": "c04f8c6f-7e54-4911-dd8e-01b66d699043"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9675"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sklearn.metrics.accuracy_score(y[:400], dist)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "del(dist)\n",
    "del(fpr)\n",
    "del(tpr)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "KA1h5IE11Aym"
   },
   "source": [
    "Our accuracy is 96% on the training set, which is quite good considering how simple our method is."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "LfuC4ug4_YQ1"
   },
   "source": [
    "### Prediction on other values\n",
    "We now want to check that our method works on another dataset. Thus we are using a verification dataset and running all the steps necessary to classify each sample:\n",
    "- Compute the features using vggface\n",
    "- Compute the distance between pairs of points\n",
    "- Check the distance against the threshold we computed in the previous section"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "id": "BM0Xpgcq9v9h"
   },
   "outputs": [],
   "source": [
    "Y_test_0 = vggface.predict(x_0[400:800])\n",
    "Y_test_1 = vggface.predict(x_1[400:800])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "id": "FxbS2NBK9z-O"
   },
   "outputs": [],
   "source": [
    "dist_test = [np.linalg.norm(Y_test_0[i]-Y_test_1[i]) for i in range(400)]\n",
    "dist_test = 1 - (dist_test - min(dist_test))/(max(dist_test) - min(dist_test) + 0.0001)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "id": "CLmdUoCQ_4y2"
   },
   "outputs": [],
   "source": [
    "del(Y_test_0)\n",
    "del(Y_test_1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "id": "ymOrvJgq-qAH",
    "outputId": "8b8c5b59-d29e-4f3e-8673-cf2988e8c55b"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(15, 5))\n",
    "\n",
    "# positive matches\n",
    "plt.hist(dist_test[:200], bins=64, label='Positive')\n",
    "# negative matches\n",
    "plt.hist(dist_test[200:],bins=64, label='Negative')\n",
    "\n",
    "plt.axvline(x=thresholds[x], c='r', label='Threshold')\n",
    "plt.legend()\n",
    "plt.title('Histogram distribution with classification threshold')\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "GEXXudlh1z6p"
   },
   "source": [
    "We can see from the distance histogram that both classes are still well separated and that the threshold we computed before still is in a reasonnable spot for a proper classification.\n",
    "\n",
    "Now we run the prediction and check the accuracy:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "id": "yyUwILVM_9JR"
   },
   "outputs": [],
   "source": [
    "dist_test[dist_test < thresholds[x]] = 0\n",
    "dist_test[dist_test >= thresholds[x]] = 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "id": "SaY4sHFaABUU",
    "outputId": "5fcb8c8e-b07f-4f08-cd85-e9e6c3d8c2da"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.945"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sklearn.metrics.accuracy_score(y[400:800], dist_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "del(dist_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "Zz6KJI2E2GjN"
   },
   "source": [
    "We obtained an 94% accuracy which is pretty high considering the simplicity of the method. We still have a 6% error rate, that we might want to reduce in a production environment.\n",
    "\n",
    "Therefore, we will explore another classification method based on a siamese architecture in the next section, and we are hoping that it will give better results."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "glf-5B-12jiF"
   },
   "source": [
    "## Siamese architecture <a class=\"anchor\" id=\"siamese\"></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "fHHJdxfo5jn_"
   },
   "source": [
    "We expose, in this section, another method for comparing two faces based on a siamese architecture."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "id": "Myn0V_uiM962"
   },
   "outputs": [],
   "source": [
    "del(data)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "GlDPQrVQ6Y4v"
   },
   "source": [
    "### Dataset creation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "id": "oFOwnKZFgyMa",
    "outputId": "a5536300-d8c7-46c9-f38e-1d6c57afa087"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(-0.51408315, 0.6411949)"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Normalize data\n",
    "x_0 = x_0 / 255.\n",
    "x_1 = x_1 / 255.\n",
    "\n",
    "x_0.min(), x_0.max()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "tPwv65R-jjQz"
   },
   "source": [
    "We want to produce three inputs to the model: the two images and the label, but a classic generator only produces two, traditionnaly one image and one label. So we had to implement a custom generator to produce the type of data we needed to feed to the network, in our case: `(image, image) and label`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "id": "6YW6sIKZIzri"
   },
   "outputs": [],
   "source": [
    "from keras.utils import Sequence\n",
    "\n",
    "class Mygenerator(Sequence):\n",
    "    def __init__(self, x1, x2, y, batch_size):\n",
    "        self.x1, self.x2, self.y = x1, x2, y\n",
    "        self.batch_size = batch_size\n",
    "        self.indices = np.arange(self.x1.shape[0])\n",
    "\n",
    "    def __len__(self):\n",
    "        return len(self.x1) // self.batch_size\n",
    "\n",
    "    def __getitem__(self, idx):\n",
    "        x1 = self.x1[idx * self.batch_size:(idx + 1) * self.batch_size]\n",
    "        x2 = self.x2[idx * self.batch_size:(idx + 1) * self.batch_size]\n",
    "        y = self.y[idx * self.batch_size:(idx + 1) * self.batch_size]\n",
    "\n",
    "        return [np.array(x1), np.array(x2)], np.array(y)\n",
    "    \n",
    "    def on_epoch_end(self):\n",
    "        # random order on the images\n",
    "        np.random.shuffle(self.indices)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "37fW0APzl_e4"
   },
   "source": [
    "Here we split our data into three distinctive sections:\n",
    "- **Training set**: data used for training the model.\n",
    "- **Validation set**: data used at the end of each epoch to check whether it is overfitting or not.\n",
    "- **Test set**: data used to assess the performance of the model after the training."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "id": "ADNGKE4mm9qM"
   },
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "X0, X0_test, X1, X1_test, Y, y_test = train_test_split(x_0, x_1, y, test_size=0.1, random_state=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "id": "feFFDbBukivl"
   },
   "outputs": [],
   "source": [
    "shape = x_0.shape\n",
    "\n",
    "# freeup RAM for google colab\n",
    "del(x_0)\n",
    "del(x_1)\n",
    "del(y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "id": "o-4_m7KUKEko"
   },
   "outputs": [],
   "source": [
    "X0_train, X0_val, X1_train, X1_val, y_train, y_val = train_test_split(X0, X1, Y, test_size=0.33, random_state=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "id": "UdFmjNQ3NCr6"
   },
   "outputs": [],
   "source": [
    "# freeup RAM for google colab\n",
    "del(X0)\n",
    "del(Y)\n",
    "del(X1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "bpVho4fX6f0v"
   },
   "source": [
    "### Model definition"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "id": "W5xNezjBMw_n"
   },
   "outputs": [],
   "source": [
    "from keras import Sequential, Model\n",
    "from keras.layers import Conv2D, MaxPool2D, Dense, Input, Flatten, Lambda, merge, Dropout\n",
    "from sklearn.metrics.pairwise import manhattan_distances\n",
    "import tensorflow as tf\n",
    "import keras.backend as K"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "gI5i8jQmmxFQ"
   },
   "source": [
    "A siamese model is structured as follows: the same network is used to compute features on both images and a manhattan distance is computed over these feature vectors."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "id": "z7Yf62JfBVfo"
   },
   "outputs": [],
   "source": [
    "def siamese_model(vggface, input_shape):\n",
    "    input_left = Input(input_shape)\n",
    "    input_right = Input(input_shape)\n",
    "\n",
    "    # We use the same model for both inputs \n",
    "    model_1 = vggface(input_left, training=False)\n",
    "    model_2 = vggface(input_right, training=False)\n",
    "\n",
    "    L1_layer = Lambda(lambda models: K.abs(models[0] - models[1]))\n",
    "\n",
    "    # Add the distance function to the network\n",
    "    dist_layer = L1_layer([model_1, model_2])\n",
    "    out = Dense(128, activation='sigmoid')(dist_layer)\n",
    "    out = Dense(1, activation='sigmoid')(out)\n",
    "\n",
    "    return Model(inputs=[input_left, input_right], outputs=out)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "id": "NOt4e14YMK0P"
   },
   "outputs": [],
   "source": [
    "from keras.callbacks import EarlyStopping, ModelCheckpoint\n",
    "\n",
    "checkpoint_filepath = 'siamese.h5'\n",
    "# we use checkpoints to save the weights of our model\n",
    "callbacks = [\n",
    "             ModelCheckpoint(\n",
    "                 filepath=checkpoint_filepath,\n",
    "                 save_weights_only=True,\n",
    "                 monitor='val_accuracy',\n",
    "                 mode='max',\n",
    "                 verbose=1,\n",
    "                 save_best_only=True\n",
    "                 ),]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "id": "SbsazTmrSuGz",
    "outputId": "ffbc400e-2d06-495a-8927-b76820f0db48"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Downloading data from https://github.com/rcmalli/keras-vggface/releases/download/v2.0/rcmalli_vggface_tf_notop_vgg16.h5\n",
      "58916864/58909280 [==============================] - 2s 0us/step\n",
      "Model: \"functional_1\"\n",
      "__________________________________________________________________________________________________\n",
      "Layer (type)                    Output Shape         Param #     Connected to                     \n",
      "==================================================================================================\n",
      "input_3 (InputLayer)            [(None, 224, 224, 3) 0                                            \n",
      "__________________________________________________________________________________________________\n",
      "input_4 (InputLayer)            [(None, 224, 224, 3) 0                                            \n",
      "__________________________________________________________________________________________________\n",
      "vggface_vgg16 (Functional)      (None, 512)          14714688    input_3[0][0]                    \n",
      "                                                                 input_4[0][0]                    \n",
      "__________________________________________________________________________________________________\n",
      "lambda (Lambda)                 (None, 512)          0           vggface_vgg16[0][0]              \n",
      "                                                                 vggface_vgg16[1][0]              \n",
      "__________________________________________________________________________________________________\n",
      "dense (Dense)                   (None, 128)          65664       lambda[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "dense_1 (Dense)                 (None, 1)            129         dense[0][0]                      \n",
      "==================================================================================================\n",
      "Total params: 14,780,481\n",
      "Trainable params: 65,793\n",
      "Non-trainable params: 14,714,688\n",
      "__________________________________________________________________________________________________\n",
      "None\n"
     ]
    }
   ],
   "source": [
    "from keras.optimizers import Adam, RMSprop, Adagrad, Adadelta, Nadam, SGD\n",
    "\n",
    "input_shape = shape[1:]\n",
    "vggface = VGGFace(model='vgg16', include_top=False, input_shape=(224, 224, 3), pooling='avg')\n",
    "\n",
    "# We freeze the layers from the base model\n",
    "vggface.trainable = False\n",
    "\n",
    "EPOCHS = 100\n",
    "BATCH_SIZE = 32\n",
    "LR = 0.0001\n",
    "\n",
    "model = siamese_model(vggface, input_shape)\n",
    "\n",
    "print(model.summary())\n",
    "\n",
    "optimizer = Adam(lr=LR, decay=LR/EPOCHS)\n",
    "\n",
    "model.compile(loss=\"binary_crossentropy\",\n",
    "              optimizer= optimizer,\n",
    "              metrics=['accuracy'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "egHs8ikorKIy"
   },
   "source": [
    "### Model training"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "id": "QZsJASuN0ZbT",
    "outputId": "64b6ee6c-cd83-49f7-f812-f76c61c20b7c"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.7695 - accuracy: 0.4929\n",
      "Epoch 00001: val_accuracy improved from -inf to 0.51436, saving model to siamese.h5\n",
      "75/75 [==============================] - 15s 202ms/step - loss: 0.7695 - accuracy: 0.4929 - val_loss: 0.7219 - val_accuracy: 0.5144\n",
      "Epoch 2/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.7171 - accuracy: 0.4929\n",
      "Epoch 00002: val_accuracy did not improve from 0.51436\n",
      "75/75 [==============================] - 15s 194ms/step - loss: 0.7171 - accuracy: 0.4929 - val_loss: 0.7029 - val_accuracy: 0.5144\n",
      "Epoch 3/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.7015 - accuracy: 0.4929\n",
      "Epoch 00003: val_accuracy did not improve from 0.51436\n",
      "75/75 [==============================] - 14s 193ms/step - loss: 0.7015 - accuracy: 0.4929 - val_loss: 0.6949 - val_accuracy: 0.5144\n",
      "Epoch 4/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.6927 - accuracy: 0.4929\n",
      "Epoch 00004: val_accuracy did not improve from 0.51436\n",
      "75/75 [==============================] - 14s 192ms/step - loss: 0.6927 - accuracy: 0.4929 - val_loss: 0.6876 - val_accuracy: 0.5144\n",
      "Epoch 5/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.6848 - accuracy: 0.5183\n",
      "Epoch 00005: val_accuracy improved from 0.51436 to 0.60557, saving model to siamese.h5\n",
      "75/75 [==============================] - 15s 197ms/step - loss: 0.6848 - accuracy: 0.5183 - val_loss: 0.6800 - val_accuracy: 0.6056\n",
      "Epoch 6/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.6771 - accuracy: 0.7808\n",
      "Epoch 00006: val_accuracy improved from 0.60557 to 0.84882, saving model to siamese.h5\n",
      "75/75 [==============================] - 15s 193ms/step - loss: 0.6771 - accuracy: 0.7808 - val_loss: 0.6727 - val_accuracy: 0.8488\n",
      "Epoch 7/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.6696 - accuracy: 0.7700\n",
      "Epoch 00007: val_accuracy did not improve from 0.84882\n",
      "75/75 [==============================] - 14s 192ms/step - loss: 0.6696 - accuracy: 0.7700 - val_loss: 0.6650 - val_accuracy: 0.8319\n",
      "Epoch 8/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.6622 - accuracy: 0.8329\n",
      "Epoch 00008: val_accuracy improved from 0.84882 to 0.89527, saving model to siamese.h5\n",
      "75/75 [==============================] - 15s 195ms/step - loss: 0.6622 - accuracy: 0.8329 - val_loss: 0.6582 - val_accuracy: 0.8953\n",
      "Epoch 9/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.6551 - accuracy: 0.8592\n",
      "Epoch 00009: val_accuracy improved from 0.89527 to 0.89696, saving model to siamese.h5\n",
      "75/75 [==============================] - 15s 197ms/step - loss: 0.6551 - accuracy: 0.8592 - val_loss: 0.6511 - val_accuracy: 0.8970\n",
      "Epoch 10/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.6480 - accuracy: 0.9008\n",
      "Epoch 00010: val_accuracy improved from 0.89696 to 0.90625, saving model to siamese.h5\n",
      "75/75 [==============================] - 15s 194ms/step - loss: 0.6480 - accuracy: 0.9008 - val_loss: 0.6443 - val_accuracy: 0.9062\n",
      "Epoch 11/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.6411 - accuracy: 0.9092\n",
      "Epoch 00011: val_accuracy improved from 0.90625 to 0.90794, saving model to siamese.h5\n",
      "75/75 [==============================] - 14s 193ms/step - loss: 0.6411 - accuracy: 0.9092 - val_loss: 0.6375 - val_accuracy: 0.9079\n",
      "Epoch 12/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.6342 - accuracy: 0.9108\n",
      "Epoch 00012: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 15s 194ms/step - loss: 0.6342 - accuracy: 0.9108 - val_loss: 0.6310 - val_accuracy: 0.8995\n",
      "Epoch 13/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.6275 - accuracy: 0.9096\n",
      "Epoch 00013: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 14s 192ms/step - loss: 0.6275 - accuracy: 0.9096 - val_loss: 0.6241 - val_accuracy: 0.9071\n",
      "Epoch 14/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.6209 - accuracy: 0.9142\n",
      "Epoch 00014: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 14s 192ms/step - loss: 0.6209 - accuracy: 0.9142 - val_loss: 0.6174 - val_accuracy: 0.9062\n",
      "Epoch 15/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.6145 - accuracy: 0.9054\n",
      "Epoch 00015: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 14s 191ms/step - loss: 0.6145 - accuracy: 0.9054 - val_loss: 0.6113 - val_accuracy: 0.9020\n",
      "Epoch 16/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.6080 - accuracy: 0.9071\n",
      "Epoch 00016: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 14s 193ms/step - loss: 0.6080 - accuracy: 0.9071 - val_loss: 0.6053 - val_accuracy: 0.8953\n",
      "Epoch 17/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.6016 - accuracy: 0.9117\n",
      "Epoch 00017: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 14s 193ms/step - loss: 0.6016 - accuracy: 0.9117 - val_loss: 0.5988 - val_accuracy: 0.8961\n",
      "Epoch 18/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.5953 - accuracy: 0.9129\n",
      "Epoch 00018: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 14s 192ms/step - loss: 0.5953 - accuracy: 0.9129 - val_loss: 0.5927 - val_accuracy: 0.8944\n",
      "Epoch 19/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.5895 - accuracy: 0.8975\n",
      "Epoch 00019: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 14s 193ms/step - loss: 0.5895 - accuracy: 0.8975 - val_loss: 0.5857 - val_accuracy: 0.9037\n",
      "Epoch 20/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.5830 - accuracy: 0.9083\n",
      "Epoch 00020: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 14s 192ms/step - loss: 0.5830 - accuracy: 0.9083 - val_loss: 0.5802 - val_accuracy: 0.8961\n",
      "Epoch 21/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.5769 - accuracy: 0.9125\n",
      "Epoch 00021: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 15s 194ms/step - loss: 0.5769 - accuracy: 0.9125 - val_loss: 0.5745 - val_accuracy: 0.8944\n",
      "Epoch 22/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.5709 - accuracy: 0.9104\n",
      "Epoch 00022: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 14s 192ms/step - loss: 0.5709 - accuracy: 0.9104 - val_loss: 0.5687 - val_accuracy: 0.8927\n",
      "Epoch 23/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.5648 - accuracy: 0.9137\n",
      "Epoch 00023: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 14s 191ms/step - loss: 0.5648 - accuracy: 0.9137 - val_loss: 0.5624 - val_accuracy: 0.8961\n",
      "Epoch 24/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.5590 - accuracy: 0.9133\n",
      "Epoch 00024: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 14s 191ms/step - loss: 0.5590 - accuracy: 0.9133 - val_loss: 0.5569 - val_accuracy: 0.8919\n",
      "Epoch 25/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.5531 - accuracy: 0.9117\n",
      "Epoch 00025: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 15s 195ms/step - loss: 0.5531 - accuracy: 0.9117 - val_loss: 0.5512 - val_accuracy: 0.8927\n",
      "Epoch 26/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.5473 - accuracy: 0.9108\n",
      "Epoch 00026: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 14s 192ms/step - loss: 0.5473 - accuracy: 0.9108 - val_loss: 0.5460 - val_accuracy: 0.8919\n",
      "Epoch 27/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.5417 - accuracy: 0.9117\n",
      "Epoch 00027: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 14s 192ms/step - loss: 0.5417 - accuracy: 0.9117 - val_loss: 0.5399 - val_accuracy: 0.8936\n",
      "Epoch 28/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.5358 - accuracy: 0.9121\n",
      "Epoch 00028: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 14s 192ms/step - loss: 0.5358 - accuracy: 0.9121 - val_loss: 0.5342 - val_accuracy: 0.8927\n",
      "Epoch 29/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.5302 - accuracy: 0.9133\n",
      "Epoch 00029: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 15s 195ms/step - loss: 0.5302 - accuracy: 0.9133 - val_loss: 0.5283 - val_accuracy: 0.8961\n",
      "Epoch 30/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.5248 - accuracy: 0.9142\n",
      "Epoch 00030: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 14s 192ms/step - loss: 0.5248 - accuracy: 0.9142 - val_loss: 0.5233 - val_accuracy: 0.8927\n",
      "Epoch 31/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.5192 - accuracy: 0.9104\n",
      "Epoch 00031: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 14s 190ms/step - loss: 0.5192 - accuracy: 0.9104 - val_loss: 0.5177 - val_accuracy: 0.8944\n",
      "Epoch 32/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.5137 - accuracy: 0.9121\n",
      "Epoch 00032: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 14s 192ms/step - loss: 0.5137 - accuracy: 0.9121 - val_loss: 0.5129 - val_accuracy: 0.8944\n",
      "Epoch 33/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.5082 - accuracy: 0.9117\n",
      "Epoch 00033: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 15s 194ms/step - loss: 0.5082 - accuracy: 0.9117 - val_loss: 0.5072 - val_accuracy: 0.8936\n",
      "Epoch 34/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.5029 - accuracy: 0.9121\n",
      "Epoch 00034: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 15s 193ms/step - loss: 0.5029 - accuracy: 0.9121 - val_loss: 0.5025 - val_accuracy: 0.8944\n",
      "Epoch 35/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.4976 - accuracy: 0.9125\n",
      "Epoch 00035: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 14s 191ms/step - loss: 0.4976 - accuracy: 0.9125 - val_loss: 0.4969 - val_accuracy: 0.8927\n",
      "Epoch 36/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.4923 - accuracy: 0.9146\n",
      "Epoch 00036: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 14s 193ms/step - loss: 0.4923 - accuracy: 0.9146 - val_loss: 0.4918 - val_accuracy: 0.8936\n",
      "Epoch 37/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.4871 - accuracy: 0.9129\n",
      "Epoch 00037: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 14s 192ms/step - loss: 0.4871 - accuracy: 0.9129 - val_loss: 0.4866 - val_accuracy: 0.8953\n",
      "Epoch 38/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.4819 - accuracy: 0.9133\n",
      "Epoch 00038: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 14s 192ms/step - loss: 0.4819 - accuracy: 0.9133 - val_loss: 0.4817 - val_accuracy: 0.8936\n",
      "Epoch 39/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.4768 - accuracy: 0.9137\n",
      "Epoch 00039: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 15s 194ms/step - loss: 0.4768 - accuracy: 0.9137 - val_loss: 0.4767 - val_accuracy: 0.8953\n",
      "Epoch 40/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.4718 - accuracy: 0.9146\n",
      "Epoch 00040: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 14s 190ms/step - loss: 0.4718 - accuracy: 0.9146 - val_loss: 0.4717 - val_accuracy: 0.8953\n",
      "Epoch 41/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.4671 - accuracy: 0.9142\n",
      "Epoch 00041: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 14s 192ms/step - loss: 0.4671 - accuracy: 0.9142 - val_loss: 0.4667 - val_accuracy: 0.8961\n",
      "Epoch 42/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.4620 - accuracy: 0.9142\n",
      "Epoch 00042: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 14s 192ms/step - loss: 0.4620 - accuracy: 0.9142 - val_loss: 0.4622 - val_accuracy: 0.8953\n",
      "Epoch 43/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.4572 - accuracy: 0.9142\n",
      "Epoch 00043: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 14s 191ms/step - loss: 0.4572 - accuracy: 0.9142 - val_loss: 0.4572 - val_accuracy: 0.8970\n",
      "Epoch 44/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.4523 - accuracy: 0.9129\n",
      "Epoch 00044: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 14s 192ms/step - loss: 0.4523 - accuracy: 0.9129 - val_loss: 0.4528 - val_accuracy: 0.8953\n",
      "Epoch 45/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.4476 - accuracy: 0.9150\n",
      "Epoch 00045: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 14s 193ms/step - loss: 0.4476 - accuracy: 0.9150 - val_loss: 0.4488 - val_accuracy: 0.8944\n",
      "Epoch 46/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.4429 - accuracy: 0.9158\n",
      "Epoch 00046: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 15s 193ms/step - loss: 0.4429 - accuracy: 0.9158 - val_loss: 0.4440 - val_accuracy: 0.8970\n",
      "Epoch 47/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.4384 - accuracy: 0.9137\n",
      "Epoch 00047: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 14s 192ms/step - loss: 0.4384 - accuracy: 0.9137 - val_loss: 0.4395 - val_accuracy: 0.8978\n",
      "Epoch 48/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.4337 - accuracy: 0.9146\n",
      "Epoch 00048: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 14s 192ms/step - loss: 0.4337 - accuracy: 0.9146 - val_loss: 0.4346 - val_accuracy: 0.8978\n",
      "Epoch 49/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.4293 - accuracy: 0.9150\n",
      "Epoch 00049: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 14s 193ms/step - loss: 0.4293 - accuracy: 0.9150 - val_loss: 0.4307 - val_accuracy: 0.8986\n",
      "Epoch 50/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.4248 - accuracy: 0.9158\n",
      "Epoch 00050: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 14s 193ms/step - loss: 0.4248 - accuracy: 0.9158 - val_loss: 0.4265 - val_accuracy: 0.8986\n",
      "Epoch 51/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.4205 - accuracy: 0.9150\n",
      "Epoch 00051: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 14s 191ms/step - loss: 0.4205 - accuracy: 0.9150 - val_loss: 0.4226 - val_accuracy: 0.8953\n",
      "Epoch 52/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.4161 - accuracy: 0.9171\n",
      "Epoch 00052: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 14s 193ms/step - loss: 0.4161 - accuracy: 0.9171 - val_loss: 0.4176 - val_accuracy: 0.8978\n",
      "Epoch 53/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.4118 - accuracy: 0.9171\n",
      "Epoch 00053: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 14s 191ms/step - loss: 0.4118 - accuracy: 0.9171 - val_loss: 0.4138 - val_accuracy: 0.8978\n",
      "Epoch 54/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.4078 - accuracy: 0.9171\n",
      "Epoch 00054: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 15s 194ms/step - loss: 0.4078 - accuracy: 0.9171 - val_loss: 0.4107 - val_accuracy: 0.8927\n",
      "Epoch 55/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.4038 - accuracy: 0.9167\n",
      "Epoch 00055: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 14s 192ms/step - loss: 0.4038 - accuracy: 0.9167 - val_loss: 0.4066 - val_accuracy: 0.8953\n",
      "Epoch 56/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.3995 - accuracy: 0.9158\n",
      "Epoch 00056: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 14s 191ms/step - loss: 0.3995 - accuracy: 0.9158 - val_loss: 0.4019 - val_accuracy: 0.8978\n",
      "Epoch 57/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.3954 - accuracy: 0.9162\n",
      "Epoch 00057: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 14s 191ms/step - loss: 0.3954 - accuracy: 0.9162 - val_loss: 0.3979 - val_accuracy: 0.8986\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 58/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.3914 - accuracy: 0.9171\n",
      "Epoch 00058: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 15s 194ms/step - loss: 0.3914 - accuracy: 0.9171 - val_loss: 0.3943 - val_accuracy: 0.8986\n",
      "Epoch 59/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.3875 - accuracy: 0.9171\n",
      "Epoch 00059: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 14s 192ms/step - loss: 0.3875 - accuracy: 0.9171 - val_loss: 0.3903 - val_accuracy: 0.8986\n",
      "Epoch 60/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.3837 - accuracy: 0.9171\n",
      "Epoch 00060: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 14s 191ms/step - loss: 0.3837 - accuracy: 0.9171 - val_loss: 0.3869 - val_accuracy: 0.8986\n",
      "Epoch 61/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.3799 - accuracy: 0.9167\n",
      "Epoch 00061: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 14s 192ms/step - loss: 0.3799 - accuracy: 0.9167 - val_loss: 0.3833 - val_accuracy: 0.8986\n",
      "Epoch 62/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.3762 - accuracy: 0.9192\n",
      "Epoch 00062: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 15s 196ms/step - loss: 0.3762 - accuracy: 0.9192 - val_loss: 0.3794 - val_accuracy: 0.8995\n",
      "Epoch 63/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.3724 - accuracy: 0.9183\n",
      "Epoch 00063: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 14s 191ms/step - loss: 0.3724 - accuracy: 0.9183 - val_loss: 0.3760 - val_accuracy: 0.8995\n",
      "Epoch 64/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.3689 - accuracy: 0.9179\n",
      "Epoch 00064: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 14s 191ms/step - loss: 0.3689 - accuracy: 0.9179 - val_loss: 0.3730 - val_accuracy: 0.8986\n",
      "Epoch 65/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.3653 - accuracy: 0.9187\n",
      "Epoch 00065: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 14s 192ms/step - loss: 0.3653 - accuracy: 0.9187 - val_loss: 0.3694 - val_accuracy: 0.9003\n",
      "Epoch 66/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.3618 - accuracy: 0.9187\n",
      "Epoch 00066: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 15s 194ms/step - loss: 0.3618 - accuracy: 0.9187 - val_loss: 0.3663 - val_accuracy: 0.9003\n",
      "Epoch 67/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.3584 - accuracy: 0.9187\n",
      "Epoch 00067: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 14s 192ms/step - loss: 0.3584 - accuracy: 0.9187 - val_loss: 0.3629 - val_accuracy: 0.9012\n",
      "Epoch 68/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.3550 - accuracy: 0.9208\n",
      "Epoch 00068: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 14s 192ms/step - loss: 0.3550 - accuracy: 0.9208 - val_loss: 0.3592 - val_accuracy: 0.9020\n",
      "Epoch 69/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.3517 - accuracy: 0.9192\n",
      "Epoch 00069: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 14s 191ms/step - loss: 0.3517 - accuracy: 0.9192 - val_loss: 0.3566 - val_accuracy: 0.9012\n",
      "Epoch 70/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.3484 - accuracy: 0.9196\n",
      "Epoch 00070: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 15s 194ms/step - loss: 0.3484 - accuracy: 0.9196 - val_loss: 0.3534 - val_accuracy: 0.9029\n",
      "Epoch 71/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.3452 - accuracy: 0.9200\n",
      "Epoch 00071: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 14s 192ms/step - loss: 0.3452 - accuracy: 0.9200 - val_loss: 0.3505 - val_accuracy: 0.9020\n",
      "Epoch 72/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.3421 - accuracy: 0.9204\n",
      "Epoch 00072: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 14s 193ms/step - loss: 0.3421 - accuracy: 0.9204 - val_loss: 0.3475 - val_accuracy: 0.9020\n",
      "Epoch 73/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.3389 - accuracy: 0.9221\n",
      "Epoch 00073: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 14s 191ms/step - loss: 0.3389 - accuracy: 0.9221 - val_loss: 0.3442 - val_accuracy: 0.9029\n",
      "Epoch 74/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.3358 - accuracy: 0.9212\n",
      "Epoch 00074: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 15s 194ms/step - loss: 0.3358 - accuracy: 0.9212 - val_loss: 0.3418 - val_accuracy: 0.9020\n",
      "Epoch 75/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.3328 - accuracy: 0.9229\n",
      "Epoch 00075: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 14s 192ms/step - loss: 0.3328 - accuracy: 0.9229 - val_loss: 0.3391 - val_accuracy: 0.9020\n",
      "Epoch 76/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.3299 - accuracy: 0.9212\n",
      "Epoch 00076: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 14s 192ms/step - loss: 0.3299 - accuracy: 0.9212 - val_loss: 0.3367 - val_accuracy: 0.9012\n",
      "Epoch 77/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.3269 - accuracy: 0.9225\n",
      "Epoch 00077: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 14s 192ms/step - loss: 0.3269 - accuracy: 0.9225 - val_loss: 0.3332 - val_accuracy: 0.9037\n",
      "Epoch 78/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.3241 - accuracy: 0.9208\n",
      "Epoch 00078: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 15s 194ms/step - loss: 0.3241 - accuracy: 0.9208 - val_loss: 0.3307 - val_accuracy: 0.9037\n",
      "Epoch 79/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.3212 - accuracy: 0.9233\n",
      "Epoch 00079: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 14s 192ms/step - loss: 0.3212 - accuracy: 0.9233 - val_loss: 0.3276 - val_accuracy: 0.9062\n",
      "Epoch 80/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.3185 - accuracy: 0.9246\n",
      "Epoch 00080: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 14s 192ms/step - loss: 0.3185 - accuracy: 0.9246 - val_loss: 0.3255 - val_accuracy: 0.9046\n",
      "Epoch 81/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.3157 - accuracy: 0.9225\n",
      "Epoch 00081: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 14s 191ms/step - loss: 0.3157 - accuracy: 0.9225 - val_loss: 0.3226 - val_accuracy: 0.9037\n",
      "Epoch 82/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.3133 - accuracy: 0.9242\n",
      "Epoch 00082: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 14s 193ms/step - loss: 0.3133 - accuracy: 0.9242 - val_loss: 0.3209 - val_accuracy: 0.9054\n",
      "Epoch 83/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.3106 - accuracy: 0.9225\n",
      "Epoch 00083: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 15s 193ms/step - loss: 0.3106 - accuracy: 0.9225 - val_loss: 0.3179 - val_accuracy: 0.9062\n",
      "Epoch 84/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.3080 - accuracy: 0.9233\n",
      "Epoch 00084: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 14s 191ms/step - loss: 0.3080 - accuracy: 0.9233 - val_loss: 0.3157 - val_accuracy: 0.9054\n",
      "Epoch 85/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.3054 - accuracy: 0.9242\n",
      "Epoch 00085: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 14s 191ms/step - loss: 0.3054 - accuracy: 0.9242 - val_loss: 0.3132 - val_accuracy: 0.9062\n",
      "Epoch 86/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.3030 - accuracy: 0.9250\n",
      "Epoch 00086: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 14s 192ms/step - loss: 0.3030 - accuracy: 0.9250 - val_loss: 0.3108 - val_accuracy: 0.9062\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 87/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.3007 - accuracy: 0.9246\n",
      "Epoch 00087: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 14s 193ms/step - loss: 0.3007 - accuracy: 0.9246 - val_loss: 0.3087 - val_accuracy: 0.9062\n",
      "Epoch 88/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.2981 - accuracy: 0.9258\n",
      "Epoch 00088: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 14s 192ms/step - loss: 0.2981 - accuracy: 0.9258 - val_loss: 0.3062 - val_accuracy: 0.9071\n",
      "Epoch 89/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.2957 - accuracy: 0.9254\n",
      "Epoch 00089: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 14s 192ms/step - loss: 0.2957 - accuracy: 0.9254 - val_loss: 0.3044 - val_accuracy: 0.9071\n",
      "Epoch 90/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.2937 - accuracy: 0.9250\n",
      "Epoch 00090: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 14s 191ms/step - loss: 0.2937 - accuracy: 0.9250 - val_loss: 0.3021 - val_accuracy: 0.9062\n",
      "Epoch 91/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.2911 - accuracy: 0.9262\n",
      "Epoch 00091: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 15s 195ms/step - loss: 0.2911 - accuracy: 0.9262 - val_loss: 0.3004 - val_accuracy: 0.9079\n",
      "Epoch 92/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.2889 - accuracy: 0.9267\n",
      "Epoch 00092: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 14s 190ms/step - loss: 0.2889 - accuracy: 0.9267 - val_loss: 0.2980 - val_accuracy: 0.9071\n",
      "Epoch 93/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.2868 - accuracy: 0.9254\n",
      "Epoch 00093: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 14s 192ms/step - loss: 0.2868 - accuracy: 0.9254 - val_loss: 0.2960 - val_accuracy: 0.9071\n",
      "Epoch 94/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.2845 - accuracy: 0.9271\n",
      "Epoch 00094: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 14s 190ms/step - loss: 0.2845 - accuracy: 0.9271 - val_loss: 0.2943 - val_accuracy: 0.9079\n",
      "Epoch 95/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.2824 - accuracy: 0.9267\n",
      "Epoch 00095: val_accuracy did not improve from 0.90794\n",
      "75/75 [==============================] - 15s 193ms/step - loss: 0.2824 - accuracy: 0.9267 - val_loss: 0.2924 - val_accuracy: 0.9079\n",
      "Epoch 96/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.2803 - accuracy: 0.9258\n",
      "Epoch 00096: val_accuracy improved from 0.90794 to 0.90878, saving model to siamese.h5\n",
      "75/75 [==============================] - 15s 194ms/step - loss: 0.2803 - accuracy: 0.9258 - val_loss: 0.2903 - val_accuracy: 0.9088\n",
      "Epoch 97/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.2784 - accuracy: 0.9271\n",
      "Epoch 00097: val_accuracy did not improve from 0.90878\n",
      "75/75 [==============================] - 14s 192ms/step - loss: 0.2784 - accuracy: 0.9271 - val_loss: 0.2886 - val_accuracy: 0.9079\n",
      "Epoch 98/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.2763 - accuracy: 0.9267\n",
      "Epoch 00098: val_accuracy did not improve from 0.90878\n",
      "75/75 [==============================] - 14s 192ms/step - loss: 0.2763 - accuracy: 0.9267 - val_loss: 0.2868 - val_accuracy: 0.9079\n",
      "Epoch 99/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.2743 - accuracy: 0.9271\n",
      "Epoch 00099: val_accuracy did not improve from 0.90878\n",
      "75/75 [==============================] - 15s 193ms/step - loss: 0.2743 - accuracy: 0.9271 - val_loss: 0.2850 - val_accuracy: 0.9088\n",
      "Epoch 100/100\n",
      "75/75 [==============================] - ETA: 0s - loss: 0.2726 - accuracy: 0.9271\n",
      "Epoch 00100: val_accuracy did not improve from 0.90878\n",
      "75/75 [==============================] - 15s 194ms/step - loss: 0.2726 - accuracy: 0.9271 - val_loss: 0.2834 - val_accuracy: 0.9079\n"
     ]
    }
   ],
   "source": [
    "train_generator = Mygenerator(X0_train, X1_train, y_train, BATCH_SIZE)\n",
    "val_generator = Mygenerator(X0_val, X1_val, y_val, BATCH_SIZE)\n",
    "\n",
    "siamese_history = model.fit(\n",
    "     train_generator,\n",
    "     validation_data=val_generator,\n",
    "     epochs=EPOCHS,\n",
    "     callbacks=callbacks,\n",
    " )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {
    "id": "N-0ds9ekqjz5",
    "outputId": "73d364a2-578a-48f7-fe30-6ff6be698726"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dict_keys(['loss', 'accuracy', 'val_loss', 'val_accuracy'])"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "siamese_history.history.keys()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {
    "id": "XD5M7Rubrbh8",
    "outputId": "07d5377e-8d49-4eb5-fbd7-8883a6020776"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x360 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(20, 5))\n",
    "plt.subplot(121)\n",
    "plt.plot(siamese_history.history['loss'], label='loss')\n",
    "plt.plot(siamese_history.history['val_loss'], label='val_loss')\n",
    "plt.legend()\n",
    "plt.title(\"Loss\")\n",
    "plt.subplot(122)\n",
    "plt.plot(siamese_history.history['accuracy'], label='accuracy')\n",
    "plt.plot(siamese_history.history['val_accuracy'], label='val_accuracy')\n",
    "plt.legend()\n",
    "plt.title(\"Accuracy\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "3zXS0sCxrcYF"
   },
   "source": [
    "If we look at the training curves here we can see that there is no overfitting (the curves follow pretty much the same trajectory) and the accuracy curve converges pretty fast but seems to get stuck right below 92%."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {
    "id": "wx8IyDMtVtqP"
   },
   "outputs": [],
   "source": [
    "K.clear_session()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "URgAQNv_sN4M"
   },
   "source": [
    "### Model testing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {
    "id": "UbNFToE1X6Ae",
    "outputId": "8924b93e-4f94-4007-9c4c-cf999a46a629"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "13/13 [==============================] - 2s 122ms/step\n"
     ]
    }
   ],
   "source": [
    "y_pred = model.predict([X0_test, X1_test],\n",
    "                      batch_size=BATCH_SIZE,\n",
    "                      verbose=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {
    "id": "ShwW6buXt4Yl"
   },
   "outputs": [],
   "source": [
    "y_pred[y_pred>0.5] = 1\n",
    "y_pred[y_pred<=0.5] = 0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {
    "id": "25v3xlfcuFvK",
    "outputId": "c493f742-98f6-411e-fb62-8515b3533758"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "13/13 [==============================] - 1s 115ms/step - loss: 0.2533 - accuracy: 0.9400\n"
     ]
    }
   ],
   "source": [
    "from keras.metrics import Accuracy\n",
    "\n",
    "results = model.evaluate([X0_test, X1_test], y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {
    "id": "SUUFgmUHvrjw",
    "outputId": "1dc30ecd-41b0-449c-93fb-b1e8802d1bad"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "test loss, test accuracy: [0.25330838561058044, 0.9399999976158142]\n"
     ]
    }
   ],
   "source": [
    "print(\"test loss, test accuracy:\", results)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "kHHc7iHkrp8N"
   },
   "source": [
    "We tried different approaches to improve the performances of our model:\n",
    "- **Data augmentation**: we introduced a horizontal flip, thinking it might be interesting with face orientation and asymetry. Unfortunately, we lacked ressources (RAM) on this part and we were not able to integrate this...\n",
    "- **Fine tuning**: We tried using a fine tuning method to obtain a better accuracy, but we only observed overfitting and it did not improve the overall results so we abandoned this lead.\n",
    "- **Improve feature quality**: We added `Conv2d` layers before the distance layer in the model, to increase the number of features and maybe alter them a bit. This method did not prove to be effective: we did not surpass 70% so we abandoned the idea.\n",
    "\n",
    "Overall our model seems to attain highest accuracy with the simplest architecure."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "yFoQhoehvv_y"
   },
   "source": [
    "## Comparative performance analysis <a class=\"anchor\" id=\"analysis\"></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "AMWVwGlb6-CA"
   },
   "source": [
    "We expose here the performaces of both methods on the same dataset.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "Jzl5HIFD7TyS"
   },
   "source": [
    "### Euclidean Distance"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {
    "id": "BFM8EvUfxNyt"
   },
   "outputs": [],
   "source": [
    "vggface = VGGFace(model='resnet50', include_top=False, input_shape=(224, 224, 3), pooling='avg')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {
    "id": "6K8BnRZcxoUK"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 2.35 s, sys: 395 ms, total: 2.75 s\n",
      "Wall time: 2.72 s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "\n",
    "Y_pred_0 = vggface.predict(X0_test*255)\n",
    "Y_pred_1 = vggface.predict(X1_test*255)\n",
    "dist = np.array([np.linalg.norm(Y_pred_1[i]-Y_pred_0[i], ord=2) for i in range(400)])\n",
    "\n",
    "dist = 1 - (dist - min(dist))/(max(dist) - min(dist))\n",
    "\n",
    "dist[dist < thresholds[x]] = 0\n",
    "dist[dist >= thresholds[x]] = 1\n",
    "acc_dist = sklearn.metrics.accuracy_score(y_test, dist)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "CRyPFOtH7YtY"
   },
   "source": [
    "### Siamese model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {
    "id": "JmjAtJDzyTTL"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 1.68 s, sys: 316 ms, total: 2 s\n",
      "Wall time: 1.89 s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "\n",
    "y_pred = model.predict([X0_test, X1_test])\n",
    "y_pred[y_pred>0.5] = 1\n",
    "y_pred[y_pred<=0.5] = 0\n",
    "\n",
    "acc_siamese= sklearn.metrics.accuracy_score(y_test, np.squeeze(y_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {
    "id": "pK4mht_mzp9Y",
    "outputId": "d0deceea-a8ee-4677-ab39-d5b1e0aa06f6"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy with euclidean distance: 0.945\n",
      "Accuracy with siamese architecture: 0.94\n"
     ]
    }
   ],
   "source": [
    "print(f\"Accuracy with euclidean distance: {acc_dist}\")\n",
    "print(f\"Accuracy with siamese architecture: {acc_siamese}\")\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "0CfxA7Y17gMs"
   },
   "source": [
    "If we look at the accuracies there is not a big difference between the two methods. Given the fact that the performances are almost the same, there is no simple answer as to, which one is best.\n",
    "- If the application is time critical (realtime face detection), the siamese architecture might be best  because the inference is faster.\n",
    "- In the case of a \"short deadline\" for example, the first one is best because it is easier to setup, and there is no training time."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Demo <a class=\"anchor\" id=\"demo\"></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We implemented a simple app to demonstrate our siamese network. This app detects a face on a videostream and uses the siamese model to predict whether or not it is the same as a reference image. The reference image can be chosen from the computer filesystem or captured directly on the videostream.  \n",
    "This type of software could be used for a face recognition problem, to unlock computers or phones for instance.\n",
    "\n",
    "The source code is located at this address: https://github.com/mattrouss/face-verification/tree/main it can be downloaded and executed from there."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from skimage.io import imread\n",
    "plt.figure(figsize=(15, 10))\n",
    "im = imread(\"../input/example/Screenshot 2020-12-26 at 18.48.24.png\")\n",
    "plt.imshow(im)\n",
    "plt.suptitle(\"Example of our face verification app\")\n",
    "plt.axis('off')\n",
    "plt.show()"
   ]
  }
 ],
 "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": "3.8.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}