image_sentiment_model.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "id": "hI_poirKudmu"
   },
   "outputs": [],
   "source": [
    "import os\n",
    "os.mkdir(\"train\")\n",
    "os.mkdir(\"train/positive\")\n",
    "os.mkdir(\"train/negative\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "id": "C4SpxEk_I5hz"
   },
   "outputs": [],
   "source": [
    "from numpy.random import seed\n",
    "seed(1)\n",
    "import tensorflow\n",
    "tensorflow.random.set_seed(2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "id": "HFNuGDcsI5h_"
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import os\n",
    "import cv2\n",
    "import tensorflow as tf\n",
    "\n",
    "from tensorflow.keras import optimizers\n",
    "from tensorflow.keras.models import Sequential\n",
    "from tensorflow.keras.layers import Dropout, BatchNormalization, LeakyReLU, Activation\n",
    "from tensorflow.keras.preprocessing.image import ImageDataGenerator, load_img\n",
    "from keras.layers import Conv2D, Dense, BatchNormalization, Activation, Dropout, MaxPooling2D, Flatten\n",
    "from tensorflow.keras.layers import Conv2D, Flatten, MaxPooling2D, Dropout, Activation, Dense\n",
    "from tensorflow.keras.layers import BatchNormalization\n",
    "from tensorflow.keras.optimizers import Adam, RMSprop, SGD\n",
    "from tensorflow.keras import regularizers\n",
    "from tensorflow.keras.callbacks import ModelCheckpoint, CSVLogger, TensorBoard, EarlyStopping, ReduceLROnPlateau\n",
    "import datetime\n",
    "import matplotlib.pyplot as plt\n",
    "from tensorflow.keras.utils import plot_model\n",
    "import math\n",
    "from sklearn.metrics import classification_report\n",
    "from sklearn.metrics import roc_auc_score\n",
    "\n",
    "# import scikitplot\n",
    "import seaborn as sns\n",
    "from matplotlib import pyplot\n",
    "\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "from sklearn.metrics import classification_report\n",
    "\n",
    "\n",
    "from keras.utils import np_utils"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "id": "gkDYtyDFI5iE"
   },
   "outputs": [],
   "source": [
    "# img = cv2.imread(\"/content/train/positive/S042_001_00000018.png\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "id": "HG7DQF-nI5iK"
   },
   "outputs": [],
   "source": [
    "# img.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "id": "N68upOI2I5iL"
   },
   "outputs": [],
   "source": [
    "# data_path = 'D:/sem_4_project_material/train/'\n",
    "data_path = '/content/train'\n",
    "\n",
    "row, col = 48, 48\n",
    "classes = 2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "id": "8GGQzb7VI5iN"
   },
   "outputs": [],
   "source": [
    "\n",
    "def count_image(data_path):\n",
    "    total_images = 0\n",
    "    for directory in os.listdir(data_path):\n",
    "        count =0\n",
    "        for f in os.listdir(data_path + '/'+ directory + '/'):\n",
    "            count +=1\n",
    "            total_images +=1\n",
    "            \n",
    "        print(f\"{directory} has {count} number images\")\n",
    "        \n",
    "    print(f\"Total number images is {total_images}\")\n",
    "    \n",
    "    return total_images"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "uYd72RByI5iO",
    "outputId": "dcc9fe68-4b1e-4ec6-e77e-c6c760996887"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "negative has 431 number images\n",
      "positive has 457 number images\n",
      "Total number images is 888\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "888"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "total_images=count_image(data_path)\n",
    "total_images"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "id": "d1csOcE9I5iQ"
   },
   "outputs": [],
   "source": [
    "EMOTIONS = ['negative','positive']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "id": "6gI0VoxzI5iR"
   },
   "outputs": [],
   "source": [
    "def image_to_array(total_image, path):\n",
    "    image_array = np.empty(shape=(total_image, 48, 48, 3))\n",
    "    image_label = np.empty(shape=(total_image))\n",
    "    img_label = []\n",
    "    label_to_text= {}\n",
    "    \n",
    "    \n",
    "    i = 0\n",
    "    e = 0\n",
    "    \n",
    "    for directory in os.listdir(path):\n",
    "#         print(directory)\n",
    "        if directory in EMOTIONS:\n",
    "            label_to_text[e] = directory\n",
    "            \n",
    "            for f in os.listdir(path + \"/\"+directory + \"/\"):\n",
    "                \n",
    "                \n",
    "                img = cv2.imread(path +  \"/\"+ directory + \"/\" +f)\n",
    "                image_array[i] = img\n",
    "            \n",
    "\n",
    "                image_label[i]= e\n",
    "                i +=1\n",
    "                img_label.append([img,e])\n",
    "\n",
    "            print(f\"All images converted to array of {directory} directory\")\n",
    "            \n",
    "            e +=1\n",
    "            \n",
    "    return image_array, image_label, label_to_text, img_label"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "471VgppRI5iU",
    "outputId": "8549d462-1d5c-47de-e838-4a5349b26639"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "All images converted to array of negative directory\n",
      "All images converted to array of positive directory\n"
     ]
    }
   ],
   "source": [
    "train_image_array, train_image_label, train_label_to_text, img_label = image_to_array(total_images, data_path)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "GxFi4ObEI5iV",
    "outputId": "7bf7de9b-7471-4083-cc8f-6181c762602f"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[[19., 19., 19.],\n",
       "        [22., 22., 22.],\n",
       "        [14., 14., 14.],\n",
       "        ...,\n",
       "        [ 0.,  0.,  0.],\n",
       "        [ 0.,  0.,  0.],\n",
       "        [12., 12., 12.]],\n",
       "\n",
       "       [[18., 18., 18.],\n",
       "        [20., 20., 20.],\n",
       "        [10., 10., 10.],\n",
       "        ...,\n",
       "        [ 0.,  0.,  0.],\n",
       "        [ 0.,  0.,  0.],\n",
       "        [ 4.,  4.,  4.]],\n",
       "\n",
       "       [[17., 17., 17.],\n",
       "        [16., 16., 16.],\n",
       "        [ 3.,  3.,  3.],\n",
       "        ...,\n",
       "        [ 0.,  0.,  0.],\n",
       "        [ 0.,  0.,  0.],\n",
       "        [ 1.,  1.,  1.]],\n",
       "\n",
       "       ...,\n",
       "\n",
       "       [[17., 17., 17.],\n",
       "        [19., 19., 19.],\n",
       "        [20., 20., 20.],\n",
       "        ...,\n",
       "        [35., 35., 35.],\n",
       "        [35., 35., 35.],\n",
       "        [34., 34., 34.]],\n",
       "\n",
       "       [[18., 18., 18.],\n",
       "        [19., 19., 19.],\n",
       "        [20., 20., 20.],\n",
       "        ...,\n",
       "        [34., 34., 34.],\n",
       "        [33., 33., 33.],\n",
       "        [33., 33., 33.]],\n",
       "\n",
       "       [[18., 18., 18.],\n",
       "        [19., 19., 19.],\n",
       "        [22., 22., 22.],\n",
       "        ...,\n",
       "        [36., 36., 36.],\n",
       "        [35., 35., 35.],\n",
       "        [33., 33., 33.]]])"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_image_array[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "r0ofWNYXI5iZ",
    "outputId": "ecd65ac2-9109-41de-8206-0499b556c7da"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(48, 48, 3)"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_image_array[0].shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "id": "aaAZvmcMI5ia"
   },
   "outputs": [],
   "source": [
    "def show_some_image(label_to_text, image_array, image_label):\n",
    "    fig = plt.figure(1,(14, 14))\n",
    "    \n",
    "    idx = 0\n",
    "    \n",
    "    for k in label_to_text:\n",
    "        indices = np.random.choice(np.where(image_label==k)[0], size=1, replace=False)\n",
    "        \n",
    "        sample_images = image_array[indices]\n",
    "        \n",
    "        for img in sample_images:\n",
    "            idx += 1\n",
    "            ax = plt.subplot(7,7,idx)\n",
    "            ax.imshow(img[:,:,0], cmap='gray')\n",
    "            ax.set_xticks([])\n",
    "            ax.set_yticks([])\n",
    "            ax.set_title(label_to_text[k])\n",
    "            plt.tight_layout()\n",
    "#         plt.savefig('chart/ck/ck7.png')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "id": "d9pObjToPqjM"
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 177
    },
    "id": "ocCT7wvsI5id",
    "outputId": "60a63dca-c12e-4a6f-abc2-05820e40d3fd"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1008x1008 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "show_some_image(train_label_to_text, train_image_array, train_image_label)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "id": "nkzaS02AuUbZ"
   },
   "outputs": [],
   "source": [
    "# train_image_label"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "id": "uReOWLBTI5if"
   },
   "outputs": [],
   "source": [
    "train_image_label_cat = np_utils.to_categorical(train_image_label)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "SNCmZj2TI5ih",
    "outputId": "b217bf19-5868-4147-a1f9-de0610e0a3a9"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       ...,\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [0., 1.]], dtype=float32)"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_image_label_cat"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "id": "VfSKABUTPy-i"
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "id": "QACZ55GYI5ij"
   },
   "outputs": [],
   "source": [
    "train_image_array_resize = train_image_array/ 255.0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "w5_cqeJoI5ik",
    "outputId": "8b40776c-f4bb-4838-ddf3-038a1f858e89"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[[0.0745098 , 0.0745098 , 0.0745098 ],\n",
       "        [0.08627451, 0.08627451, 0.08627451],\n",
       "        [0.05490196, 0.05490196, 0.05490196],\n",
       "        ...,\n",
       "        [0.        , 0.        , 0.        ],\n",
       "        [0.        , 0.        , 0.        ],\n",
       "        [0.04705882, 0.04705882, 0.04705882]],\n",
       "\n",
       "       [[0.07058824, 0.07058824, 0.07058824],\n",
       "        [0.07843137, 0.07843137, 0.07843137],\n",
       "        [0.03921569, 0.03921569, 0.03921569],\n",
       "        ...,\n",
       "        [0.        , 0.        , 0.        ],\n",
       "        [0.        , 0.        , 0.        ],\n",
       "        [0.01568627, 0.01568627, 0.01568627]],\n",
       "\n",
       "       [[0.06666667, 0.06666667, 0.06666667],\n",
       "        [0.0627451 , 0.0627451 , 0.0627451 ],\n",
       "        [0.01176471, 0.01176471, 0.01176471],\n",
       "        ...,\n",
       "        [0.        , 0.        , 0.        ],\n",
       "        [0.        , 0.        , 0.        ],\n",
       "        [0.00392157, 0.00392157, 0.00392157]],\n",
       "\n",
       "       ...,\n",
       "\n",
       "       [[0.06666667, 0.06666667, 0.06666667],\n",
       "        [0.0745098 , 0.0745098 , 0.0745098 ],\n",
       "        [0.07843137, 0.07843137, 0.07843137],\n",
       "        ...,\n",
       "        [0.1372549 , 0.1372549 , 0.1372549 ],\n",
       "        [0.1372549 , 0.1372549 , 0.1372549 ],\n",
       "        [0.13333333, 0.13333333, 0.13333333]],\n",
       "\n",
       "       [[0.07058824, 0.07058824, 0.07058824],\n",
       "        [0.0745098 , 0.0745098 , 0.0745098 ],\n",
       "        [0.07843137, 0.07843137, 0.07843137],\n",
       "        ...,\n",
       "        [0.13333333, 0.13333333, 0.13333333],\n",
       "        [0.12941176, 0.12941176, 0.12941176],\n",
       "        [0.12941176, 0.12941176, 0.12941176]],\n",
       "\n",
       "       [[0.07058824, 0.07058824, 0.07058824],\n",
       "        [0.0745098 , 0.0745098 , 0.0745098 ],\n",
       "        [0.08627451, 0.08627451, 0.08627451],\n",
       "        ...,\n",
       "        [0.14117647, 0.14117647, 0.14117647],\n",
       "        [0.1372549 , 0.1372549 , 0.1372549 ],\n",
       "        [0.12941176, 0.12941176, 0.12941176]]])"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_image_array_resize[0]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "Nku4KSmgI5iz"
   },
   "source": [
    "### Splitting the dataset for training and validation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "CXSBC2vcI5jT",
    "outputId": "71ad983e-50cb-41c4-b6d1-217ca76edf79"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((710, 48, 48, 3), (178, 48, 48, 3), (710, 2), (178, 2))"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train, X_valid, y_train, y_valid = train_test_split(train_image_array_resize, train_image_label_cat, stratify = train_image_label_cat, shuffle =True, test_size=0.2, random_state= 42)\n",
    "\n",
    "X_train.shape, X_valid.shape, y_train.shape, y_valid.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "id": "80cEuaVGI5ja"
   },
   "outputs": [],
   "source": [
    "img_width = X_train.shape[1]\n",
    "img_height = X_train.shape[2]\n",
    "img_depth = X_train.shape[3]\n",
    "num_classes = y_train.shape[1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "cl1SR8HuI5jc",
    "outputId": "cc177c73-4d8b-432e-d13e-dde7620f4736"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "48\n",
      "48\n",
      "3\n",
      "2\n"
     ]
    }
   ],
   "source": [
    "print(img_width)\n",
    "print(img_height)\n",
    "print(img_depth)\n",
    "print(num_classes)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "id": "8hv4NLZnI5jf"
   },
   "outputs": [],
   "source": [
    "def build_net(optim):\n",
    "   \n",
    "    net = Sequential()\n",
    "\n",
    "    net.add(\n",
    "        Conv2D(\n",
    "            filters=64,\n",
    "            kernel_size=(5,5),\n",
    "            input_shape=(img_width, img_height, img_depth),\n",
    "            activation='elu',\n",
    "            padding='same',\n",
    "            kernel_initializer='he_normal'\n",
    "        )\n",
    "    )\n",
    "    net.add(BatchNormalization())\n",
    "    net.add(\n",
    "        Conv2D(\n",
    "            filters=64,\n",
    "            kernel_size=(5,5),\n",
    "            activation='elu',\n",
    "            padding='same',\n",
    "            kernel_initializer='he_normal'\n",
    "        )\n",
    "    )\n",
    "    net.add(BatchNormalization())\n",
    "    \n",
    "    net.add(MaxPooling2D(pool_size=(2,2)))\n",
    "    net.add(Dropout(0.2))\n",
    "\n",
    "    net.add(\n",
    "        Conv2D(\n",
    "            filters=128,\n",
    "            kernel_size=(3,3),\n",
    "            activation='elu',\n",
    "            padding='same',\n",
    "            kernel_initializer='he_normal'\n",
    "        )\n",
    "    )\n",
    "    net.add(BatchNormalization())\n",
    "    net.add(\n",
    "        Conv2D(\n",
    "            filters=128,\n",
    "            kernel_size=(3,3),\n",
    "            activation='elu',\n",
    "            padding='same',\n",
    "            kernel_initializer='he_normal'\n",
    "        )\n",
    "    )\n",
    "    net.add(BatchNormalization())\n",
    "    \n",
    "    net.add(MaxPooling2D(pool_size=(2,2), ))\n",
    "    net.add(Dropout(0.2))\n",
    "\n",
    "    net.add(\n",
    "        Conv2D(\n",
    "            filters=256,\n",
    "            kernel_size=(3,3),\n",
    "            activation='elu',\n",
    "            padding='same',\n",
    "            kernel_initializer='he_normal'\n",
    "        )\n",
    "    )\n",
    "    net.add(BatchNormalization())\n",
    "    net.add(\n",
    "        Conv2D(\n",
    "            filters=256,\n",
    "            kernel_size=(3,3),\n",
    "            activation='elu',\n",
    "            padding='same',\n",
    "            kernel_initializer='he_normal'\n",
    "        )\n",
    "    )\n",
    "    net.add(BatchNormalization())\n",
    "    \n",
    "    net.add(MaxPooling2D(pool_size=(2,2) ))\n",
    "    net.add(Dropout(0.2))\n",
    "\n",
    "    net.add(Flatten())\n",
    "        \n",
    "    net.add(\n",
    "        Dense(\n",
    "            128,\n",
    "            activation='elu',\n",
    "            kernel_initializer='he_normal'\n",
    "        )\n",
    "    )\n",
    "    net.add(BatchNormalization())\n",
    "    \n",
    "    net.add(Dropout(0.2))\n",
    "    \n",
    "    net.add(\n",
    "        Dense(\n",
    "            num_classes,\n",
    "            activation='sigmoid'\n",
    "        )\n",
    "    )\n",
    "    \n",
    "    net.compile(\n",
    "        loss='binary_crossentropy',\n",
    "        optimizer=optim,\n",
    "        metrics=['accuracy']\n",
    "    )\n",
    "    \n",
    "    net.summary()\n",
    "    \n",
    "    return net"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "id": "qpOZdue9I5jh"
   },
   "outputs": [],
   "source": [
    "\n",
    "early_stopping = EarlyStopping(\n",
    "    monitor='val_accuracy',\n",
    "    min_delta=0.00005,\n",
    "    patience=30,\n",
    "    verbose=1,\n",
    "    restore_best_weights=True,\n",
    ")\n",
    "\n",
    "lr_scheduler = ReduceLROnPlateau(\n",
    "    monitor='val_accuracy',\n",
    "    factor=0.5,\n",
    "    patience=7,\n",
    "    min_lr=1e-7,\n",
    "    verbose=1,\n",
    ")\n",
    "\n",
    "callbacks = [\n",
    "    early_stopping,\n",
    "    lr_scheduler,\n",
    "]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "id": "hMxnwadoI5ji"
   },
   "outputs": [],
   "source": [
    "# As the data in hand is less as compared to the task so ImageDataGenerator is good to go.\n",
    "train_datagen = ImageDataGenerator(\n",
    "    rotation_range=15,\n",
    "    width_shift_range=0.15,\n",
    "    height_shift_range=0.15,\n",
    "    shear_range=0.15,\n",
    "    zoom_range=0.15,\n",
    "    horizontal_flip=True,\n",
    ")\n",
    "train_datagen.fit(X_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "Req_rKAMI5jj",
    "outputId": "1e855ddc-146e-4f6b-a21c-44871ed3833a",
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " conv2d (Conv2D)             (None, 48, 48, 64)        4864      \n",
      "                                                                 \n",
      " batch_normalization (BatchN  (None, 48, 48, 64)       256       \n",
      " ormalization)                                                   \n",
      "                                                                 \n",
      " conv2d_1 (Conv2D)           (None, 48, 48, 64)        102464    \n",
      "                                                                 \n",
      " batch_normalization_1 (Batc  (None, 48, 48, 64)       256       \n",
      " hNormalization)                                                 \n",
      "                                                                 \n",
      " max_pooling2d (MaxPooling2D  (None, 24, 24, 64)       0         \n",
      " )                                                               \n",
      "                                                                 \n",
      " dropout (Dropout)           (None, 24, 24, 64)        0         \n",
      "                                                                 \n",
      " conv2d_2 (Conv2D)           (None, 24, 24, 128)       73856     \n",
      "                                                                 \n",
      " batch_normalization_2 (Batc  (None, 24, 24, 128)      512       \n",
      " hNormalization)                                                 \n",
      "                                                                 \n",
      " conv2d_3 (Conv2D)           (None, 24, 24, 128)       147584    \n",
      "                                                                 \n",
      " batch_normalization_3 (Batc  (None, 24, 24, 128)      512       \n",
      " hNormalization)                                                 \n",
      "                                                                 \n",
      " max_pooling2d_1 (MaxPooling  (None, 12, 12, 128)      0         \n",
      " 2D)                                                             \n",
      "                                                                 \n",
      " dropout_1 (Dropout)         (None, 12, 12, 128)       0         \n",
      "                                                                 \n",
      " conv2d_4 (Conv2D)           (None, 12, 12, 256)       295168    \n",
      "                                                                 \n",
      " batch_normalization_4 (Batc  (None, 12, 12, 256)      1024      \n",
      " hNormalization)                                                 \n",
      "                                                                 \n",
      " conv2d_5 (Conv2D)           (None, 12, 12, 256)       590080    \n",
      "                                                                 \n",
      " batch_normalization_5 (Batc  (None, 12, 12, 256)      1024      \n",
      " hNormalization)                                                 \n",
      "                                                                 \n",
      " max_pooling2d_2 (MaxPooling  (None, 6, 6, 256)        0         \n",
      " 2D)                                                             \n",
      "                                                                 \n",
      " dropout_2 (Dropout)         (None, 6, 6, 256)         0         \n",
      "                                                                 \n",
      " flatten (Flatten)           (None, 9216)              0         \n",
      "                                                                 \n",
      " dense (Dense)               (None, 128)               1179776   \n",
      "                                                                 \n",
      " batch_normalization_6 (Batc  (None, 128)              512       \n",
      " hNormalization)                                                 \n",
      "                                                                 \n",
      " dropout_3 (Dropout)         (None, 128)               0         \n",
      "                                                                 \n",
      " dense_1 (Dense)             (None, 2)                 258       \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 2,398,146\n",
      "Trainable params: 2,396,098\n",
      "Non-trainable params: 2,048\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "batch_size = 40 #batch size of 32 performs the best.\n",
    "epochs = 60\n",
    "optims = optimizers.Adam(0.001)\n",
    "\n",
    "# I tried both `Nadam` and `Adam`, the difference in results is not different but I finally went with Adam as it is more popular.\n",
    "model = build_net(optims) \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "28ne24v_I5jl",
    "outputId": "9ab62644-487a-41ed-d916-8d9fc1027a65",
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " conv2d (Conv2D)             (None, 48, 48, 64)        4864      \n",
      "                                                                 \n",
      " batch_normalization (BatchN  (None, 48, 48, 64)       256       \n",
      " ormalization)                                                   \n",
      "                                                                 \n",
      " conv2d_1 (Conv2D)           (None, 48, 48, 64)        102464    \n",
      "                                                                 \n",
      " batch_normalization_1 (Batc  (None, 48, 48, 64)       256       \n",
      " hNormalization)                                                 \n",
      "                                                                 \n",
      " max_pooling2d (MaxPooling2D  (None, 24, 24, 64)       0         \n",
      " )                                                               \n",
      "                                                                 \n",
      " dropout (Dropout)           (None, 24, 24, 64)        0         \n",
      "                                                                 \n",
      " conv2d_2 (Conv2D)           (None, 24, 24, 128)       73856     \n",
      "                                                                 \n",
      " batch_normalization_2 (Batc  (None, 24, 24, 128)      512       \n",
      " hNormalization)                                                 \n",
      "                                                                 \n",
      " conv2d_3 (Conv2D)           (None, 24, 24, 128)       147584    \n",
      "                                                                 \n",
      " batch_normalization_3 (Batc  (None, 24, 24, 128)      512       \n",
      " hNormalization)                                                 \n",
      "                                                                 \n",
      " max_pooling2d_1 (MaxPooling  (None, 12, 12, 128)      0         \n",
      " 2D)                                                             \n",
      "                                                                 \n",
      " dropout_1 (Dropout)         (None, 12, 12, 128)       0         \n",
      "                                                                 \n",
      " conv2d_4 (Conv2D)           (None, 12, 12, 256)       295168    \n",
      "                                                                 \n",
      " batch_normalization_4 (Batc  (None, 12, 12, 256)      1024      \n",
      " hNormalization)                                                 \n",
      "                                                                 \n",
      " conv2d_5 (Conv2D)           (None, 12, 12, 256)       590080    \n",
      "                                                                 \n",
      " batch_normalization_5 (Batc  (None, 12, 12, 256)      1024      \n",
      " hNormalization)                                                 \n",
      "                                                                 \n",
      " max_pooling2d_2 (MaxPooling  (None, 6, 6, 256)        0         \n",
      " 2D)                                                             \n",
      "                                                                 \n",
      " dropout_2 (Dropout)         (None, 6, 6, 256)         0         \n",
      "                                                                 \n",
      " flatten (Flatten)           (None, 9216)              0         \n",
      "                                                                 \n",
      " dense (Dense)               (None, 128)               1179776   \n",
      "                                                                 \n",
      " batch_normalization_6 (Batc  (None, 128)              512       \n",
      " hNormalization)                                                 \n",
      "                                                                 \n",
      " dropout_3 (Dropout)         (None, 128)               0         \n",
      "                                                                 \n",
      " dense_1 (Dense)             (None, 2)                 258       \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 2,398,146\n",
      "Trainable params: 2,396,098\n",
      "Non-trainable params: 2,048\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "stringlist = []\n",
    "model.summary(print_fn=lambda x: stringlist.append(x))\n",
    "short_model_summary = \"\\n\".join(stringlist)\n",
    "print(short_model_summary)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 1000
    },
    "id": "bBb6Mif8I5jn",
    "outputId": "a86f1494-c8c8-43e2-da32-2ccaab35fa10"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from keras.utils.vis_utils import plot_model\n",
    "plot_model(model, to_file='model_plot.png', show_shapes=True, show_layer_names=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "_da500SlI5jo",
    "outputId": "2ae7862f-9b2d-4e52-fd6e-fc21693dfae3"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/60\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.7/dist-packages/ipykernel_launcher.py:7: UserWarning: `Model.fit_generator` is deprecated and will be removed in a future version. Please use `Model.fit`, which supports generators.\n",
      "  import sys\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "17/17 [==============================] - 15s 116ms/step - loss: 0.7236 - accuracy: 0.6662 - val_loss: 7.4895 - val_accuracy: 0.5169 - lr: 0.0010\n",
      "Epoch 2/60\n",
      "17/17 [==============================] - 1s 61ms/step - loss: 0.5137 - accuracy: 0.7775 - val_loss: 3.3522 - val_accuracy: 0.5730 - lr: 0.0010\n",
      "Epoch 3/60\n",
      "17/17 [==============================] - 1s 78ms/step - loss: 0.3936 - accuracy: 0.8225 - val_loss: 2.4322 - val_accuracy: 0.6067 - lr: 0.0010\n",
      "Epoch 4/60\n",
      "17/17 [==============================] - 1s 56ms/step - loss: 0.3195 - accuracy: 0.8690 - val_loss: 5.1012 - val_accuracy: 0.5730 - lr: 0.0010\n",
      "Epoch 5/60\n",
      "17/17 [==============================] - 1s 71ms/step - loss: 0.2935 - accuracy: 0.8662 - val_loss: 0.1901 - val_accuracy: 0.9551 - lr: 0.0010\n",
      "Epoch 6/60\n",
      "17/17 [==============================] - 1s 59ms/step - loss: 0.2675 - accuracy: 0.8873 - val_loss: 0.1870 - val_accuracy: 0.9382 - lr: 0.0010\n",
      "Epoch 7/60\n",
      "17/17 [==============================] - 1s 69ms/step - loss: 0.2465 - accuracy: 0.8930 - val_loss: 0.2757 - val_accuracy: 0.9045 - lr: 0.0010\n",
      "Epoch 8/60\n",
      "17/17 [==============================] - 1s 55ms/step - loss: 0.2042 - accuracy: 0.9155 - val_loss: 0.1912 - val_accuracy: 0.9438 - lr: 0.0010\n",
      "Epoch 9/60\n",
      "17/17 [==============================] - 1s 70ms/step - loss: 0.1641 - accuracy: 0.9437 - val_loss: 0.2545 - val_accuracy: 0.9045 - lr: 0.0010\n",
      "Epoch 10/60\n",
      "17/17 [==============================] - 1s 56ms/step - loss: 0.1731 - accuracy: 0.9310 - val_loss: 0.1011 - val_accuracy: 0.9551 - lr: 0.0010\n",
      "Epoch 11/60\n",
      "17/17 [==============================] - 1s 70ms/step - loss: 0.1416 - accuracy: 0.9507 - val_loss: 0.1148 - val_accuracy: 0.9438 - lr: 0.0010\n",
      "Epoch 12/60\n",
      "18/17 [==============================] - ETA: 0s - loss: 0.1436 - accuracy: 0.9408\n",
      "Epoch 12: ReduceLROnPlateau reducing learning rate to 0.0005000000237487257.\n",
      "17/17 [==============================] - 1s 57ms/step - loss: 0.1436 - accuracy: 0.9408 - val_loss: 0.0912 - val_accuracy: 0.9551 - lr: 0.0010\n",
      "Epoch 13/60\n",
      "17/17 [==============================] - 1s 69ms/step - loss: 0.1218 - accuracy: 0.9535 - val_loss: 0.1766 - val_accuracy: 0.9494 - lr: 5.0000e-04\n",
      "Epoch 14/60\n",
      "17/17 [==============================] - 1s 61ms/step - loss: 0.1020 - accuracy: 0.9662 - val_loss: 0.1213 - val_accuracy: 0.9663 - lr: 5.0000e-04\n",
      "Epoch 15/60\n",
      "17/17 [==============================] - 1s 67ms/step - loss: 0.0863 - accuracy: 0.9732 - val_loss: 0.1933 - val_accuracy: 0.9551 - lr: 5.0000e-04\n",
      "Epoch 16/60\n",
      "17/17 [==============================] - 1s 61ms/step - loss: 0.0813 - accuracy: 0.9690 - val_loss: 0.1922 - val_accuracy: 0.9438 - lr: 5.0000e-04\n",
      "Epoch 17/60\n",
      "17/17 [==============================] - 1s 57ms/step - loss: 0.0846 - accuracy: 0.9775 - val_loss: 0.1138 - val_accuracy: 0.9551 - lr: 5.0000e-04\n",
      "Epoch 18/60\n",
      "17/17 [==============================] - 1s 62ms/step - loss: 0.0801 - accuracy: 0.9704 - val_loss: 0.1405 - val_accuracy: 0.9607 - lr: 5.0000e-04\n",
      "Epoch 19/60\n",
      "17/17 [==============================] - 1s 70ms/step - loss: 0.0613 - accuracy: 0.9831 - val_loss: 0.1021 - val_accuracy: 0.9775 - lr: 5.0000e-04\n",
      "Epoch 20/60\n",
      "17/17 [==============================] - 1s 56ms/step - loss: 0.0807 - accuracy: 0.9662 - val_loss: 0.1342 - val_accuracy: 0.9551 - lr: 5.0000e-04\n",
      "Epoch 21/60\n",
      "17/17 [==============================] - 1s 59ms/step - loss: 0.0929 - accuracy: 0.9662 - val_loss: 0.1821 - val_accuracy: 0.9607 - lr: 5.0000e-04\n",
      "Epoch 22/60\n",
      "17/17 [==============================] - 1s 70ms/step - loss: 0.0767 - accuracy: 0.9732 - val_loss: 0.1187 - val_accuracy: 0.9551 - lr: 5.0000e-04\n",
      "Epoch 23/60\n",
      "17/17 [==============================] - 1s 72ms/step - loss: 0.0700 - accuracy: 0.9803 - val_loss: 0.0895 - val_accuracy: 0.9663 - lr: 5.0000e-04\n",
      "Epoch 24/60\n",
      "17/17 [==============================] - 1s 56ms/step - loss: 0.0597 - accuracy: 0.9831 - val_loss: 0.0526 - val_accuracy: 0.9775 - lr: 5.0000e-04\n",
      "Epoch 25/60\n",
      "17/17 [==============================] - 1s 58ms/step - loss: 0.0561 - accuracy: 0.9803 - val_loss: 0.0881 - val_accuracy: 0.9663 - lr: 5.0000e-04\n",
      "Epoch 26/60\n",
      "18/17 [==============================] - ETA: 0s - loss: 0.0524 - accuracy: 0.9845\n",
      "Epoch 26: ReduceLROnPlateau reducing learning rate to 0.0002500000118743628.\n",
      "17/17 [==============================] - 1s 70ms/step - loss: 0.0524 - accuracy: 0.9845 - val_loss: 0.0865 - val_accuracy: 0.9663 - lr: 5.0000e-04\n",
      "Epoch 27/60\n",
      "17/17 [==============================] - 1s 59ms/step - loss: 0.0374 - accuracy: 0.9887 - val_loss: 0.0904 - val_accuracy: 0.9719 - lr: 2.5000e-04\n",
      "Epoch 28/60\n",
      "17/17 [==============================] - 1s 68ms/step - loss: 0.0424 - accuracy: 0.9831 - val_loss: 0.1099 - val_accuracy: 0.9607 - lr: 2.5000e-04\n",
      "Epoch 29/60\n",
      "17/17 [==============================] - 1s 57ms/step - loss: 0.0344 - accuracy: 0.9915 - val_loss: 0.0977 - val_accuracy: 0.9607 - lr: 2.5000e-04\n",
      "Epoch 30/60\n",
      "17/17 [==============================] - 1s 62ms/step - loss: 0.0349 - accuracy: 0.9859 - val_loss: 0.1179 - val_accuracy: 0.9663 - lr: 2.5000e-04\n",
      "Epoch 31/60\n",
      "17/17 [==============================] - 1s 70ms/step - loss: 0.0342 - accuracy: 0.9901 - val_loss: 0.1220 - val_accuracy: 0.9663 - lr: 2.5000e-04\n",
      "Epoch 32/60\n",
      "17/17 [==============================] - 1s 73ms/step - loss: 0.0291 - accuracy: 0.9915 - val_loss: 0.1471 - val_accuracy: 0.9719 - lr: 2.5000e-04\n",
      "Epoch 33/60\n",
      "18/17 [==============================] - ETA: 0s - loss: 0.0370 - accuracy: 0.9873\n",
      "Epoch 33: ReduceLROnPlateau reducing learning rate to 0.0001250000059371814.\n",
      "17/17 [==============================] - 1s 58ms/step - loss: 0.0370 - accuracy: 0.9873 - val_loss: 0.1386 - val_accuracy: 0.9719 - lr: 2.5000e-04\n",
      "Epoch 34/60\n",
      "17/17 [==============================] - 1s 67ms/step - loss: 0.0315 - accuracy: 0.9901 - val_loss: 0.1607 - val_accuracy: 0.9607 - lr: 1.2500e-04\n",
      "Epoch 35/60\n",
      "17/17 [==============================] - 1s 54ms/step - loss: 0.0246 - accuracy: 0.9901 - val_loss: 0.1601 - val_accuracy: 0.9607 - lr: 1.2500e-04\n",
      "Epoch 36/60\n",
      "17/17 [==============================] - 1s 61ms/step - loss: 0.0338 - accuracy: 0.9845 - val_loss: 0.1345 - val_accuracy: 0.9663 - lr: 1.2500e-04\n",
      "Epoch 37/60\n",
      "17/17 [==============================] - 1s 72ms/step - loss: 0.0326 - accuracy: 0.9887 - val_loss: 0.1376 - val_accuracy: 0.9607 - lr: 1.2500e-04\n",
      "Epoch 38/60\n",
      "17/17 [==============================] - 1s 74ms/step - loss: 0.0436 - accuracy: 0.9887 - val_loss: 0.1272 - val_accuracy: 0.9607 - lr: 1.2500e-04\n",
      "Epoch 39/60\n",
      "17/17 [==============================] - 1s 69ms/step - loss: 0.0234 - accuracy: 0.9915 - val_loss: 0.1091 - val_accuracy: 0.9663 - lr: 1.2500e-04\n",
      "Epoch 40/60\n",
      "17/17 [===========================>..] - ETA: 0s - loss: 0.0232 - accuracy: 0.9955\n",
      "Epoch 40: ReduceLROnPlateau reducing learning rate to 6.25000029685907e-05.\n",
      "17/17 [==============================] - 1s 57ms/step - loss: 0.0252 - accuracy: 0.9944 - val_loss: 0.1154 - val_accuracy: 0.9663 - lr: 1.2500e-04\n",
      "Epoch 41/60\n",
      "17/17 [==============================] - 1s 58ms/step - loss: 0.0378 - accuracy: 0.9901 - val_loss: 0.1282 - val_accuracy: 0.9719 - lr: 6.2500e-05\n",
      "Epoch 42/60\n",
      "17/17 [==============================] - 1s 58ms/step - loss: 0.0272 - accuracy: 0.9887 - val_loss: 0.1252 - val_accuracy: 0.9719 - lr: 6.2500e-05\n",
      "Epoch 43/60\n",
      "17/17 [==============================] - 1s 71ms/step - loss: 0.0354 - accuracy: 0.9859 - val_loss: 0.1219 - val_accuracy: 0.9719 - lr: 6.2500e-05\n",
      "Epoch 44/60\n",
      "17/17 [==============================] - 1s 72ms/step - loss: 0.0219 - accuracy: 0.9944 - val_loss: 0.1193 - val_accuracy: 0.9719 - lr: 6.2500e-05\n",
      "Epoch 45/60\n",
      "17/17 [==============================] - 1s 71ms/step - loss: 0.0298 - accuracy: 0.9958 - val_loss: 0.1211 - val_accuracy: 0.9663 - lr: 6.2500e-05\n",
      "Epoch 46/60\n",
      "17/17 [==============================] - 1s 72ms/step - loss: 0.0271 - accuracy: 0.9930 - val_loss: 0.1162 - val_accuracy: 0.9719 - lr: 6.2500e-05\n",
      "Epoch 47/60\n",
      "18/17 [==============================] - ETA: 0s - loss: 0.0179 - accuracy: 0.9972\n",
      "Epoch 47: ReduceLROnPlateau reducing learning rate to 3.125000148429535e-05.\n",
      "17/17 [==============================] - 1s 58ms/step - loss: 0.0179 - accuracy: 0.9972 - val_loss: 0.1157 - val_accuracy: 0.9719 - lr: 6.2500e-05\n",
      "Epoch 48/60\n",
      "17/17 [==============================] - 1s 71ms/step - loss: 0.0190 - accuracy: 0.9972 - val_loss: 0.1171 - val_accuracy: 0.9719 - lr: 3.1250e-05\n",
      "Epoch 49/60\n",
      "18/17 [==============================] - ETA: 0s - loss: 0.0223 - accuracy: 0.9944Restoring model weights from the end of the best epoch: 19.\n",
      "17/17 [==============================] - 1s 61ms/step - loss: 0.0223 - accuracy: 0.9944 - val_loss: 0.1215 - val_accuracy: 0.9719 - lr: 3.1250e-05\n",
      "Epoch 49: early stopping\n"
     ]
    }
   ],
   "source": [
    "history = model.fit_generator(\n",
    "    train_datagen.flow(X_train, y_train, batch_size=batch_size),\n",
    "    validation_data=(X_valid, y_valid),\n",
    "    steps_per_epoch=len(X_train) / batch_size,\n",
    "    epochs=epochs,\n",
    "    callbacks=callbacks,\n",
    "    use_multiprocessing=True\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "id": "Bij0jFIJI5jr"
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 513
    },
    "id": "uCSXEdr7I5jr",
    "outputId": "728b409a-f30d-48bb-d26f-d6ef4a201cad"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(8,8))\n",
    "plt.plot(history.history['accuracy'])\n",
    "plt.plot(history.history['val_accuracy'])\n",
    "plt.title('Model Accuracy')\n",
    "plt.xlabel('Epochs')\n",
    "plt.ylabel('Accuracy')\n",
    "plt.legend(['train', 'test'], loc='upper left')\n",
    "# plt.savefig(\"chart/ck/acc_ck_arr_class_7.png\")\n",
    "# plt.savefig(\"chart/ck/acc_ck_arr_class_7.eps\",format='eps',bbox_inches='tight')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 513
    },
    "id": "8_WBjOK_I5js",
    "outputId": "34b944fe-2576-441d-de39-7816cb1f46fd"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(8,8))\n",
    "plt.plot(history.history['loss'])\n",
    "plt.plot(history.history['val_loss'])\n",
    "plt.title('model Loss')\n",
    "plt.xlabel('Epochs')\n",
    "plt.ylabel('Loss')\n",
    "plt.legend(['train', 'test'], loc='upper left')\n",
    "# pyplot.savefig(\"chart/ck/loss_ck_arr_class_7.png\")\n",
    "# pyplot.savefig(\"chart/ck/loss_ck_arr_class_7.eps\",format='eps',bbox_inches='tight')\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "9bmwDEikNBIy",
    "outputId": "cdba518f-6da2-480b-caee-44ca942158c7"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "23/23 [==============================] - 0s 14ms/step - loss: 0.0284 - accuracy: 0.9887\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[0.02842910960316658, 0.9887323975563049]"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.evaluate(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "id": "RudY6ze4WoaO"
   },
   "outputs": [],
   "source": [
    "train_pred = model.predict(X_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "Et4c2mQtwBxF",
    "outputId": "dd8082c2-27d3-4324-a3b7-026f1d880c08"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.argmax(train_pred[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "id": "nxugUHF6wBrf"
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "7HtwV5RSb3i_",
    "outputId": "fb3930cd-8699-4804-d597-48a29b0ba495"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0]"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# classifying the predicted categories\n",
    "train_pred_conv = []\n",
    "for i in train_pred:\n",
    "    train_pred_conv.append(float(np.argmax(i)))\n",
    "\n",
    "train_pred_conv"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "JXNRUNdJbK1J",
    "outputId": "5befc146-ad38-498b-8b37-49ee679a4714"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.0"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_pred_conv[2]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "Juo2rAf1bN3r",
    "outputId": "262b1c17-7bdc-4c57-f0e7-2ca925f85542"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([1., 0.], dtype=float32)"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_train[2]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "M14-WU31wmqs",
    "outputId": "78cd62ad-de57-4bb3-a040-bd39042ec04c"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0]"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_train_conv = []\n",
    "for i in y_train:\n",
    "  y_train_conv.append(float(np.argmax(i)))\n",
    "\n",
    "y_train_conv"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "dRZ-q06CWvmI",
    "outputId": "06ef5442-e0f6-4683-c9c3-2b138b6ea4ba"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<tf.Tensor: shape=(2, 2), dtype=int32, numpy=\n",
       "array([[344,   1],\n",
       "       [  7, 358]], dtype=int32)>"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tf.math.confusion_matrix(y_train_conv, train_pred_conv)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "jDtnLH2t0Z0N",
    "outputId": "24fdb9b4-b8c4-43d8-ddfc-ebd4ab85eb27"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "         0.0       0.98      1.00      0.99       345\n",
      "         1.0       1.00      0.98      0.99       365\n",
      "\n",
      "    accuracy                           0.99       710\n",
      "   macro avg       0.99      0.99      0.99       710\n",
      "weighted avg       0.99      0.99      0.99       710\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(classification_report(y_train_conv, train_pred_conv))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "PcUduH-N1pPT",
    "outputId": "f6d9ac5b-5ef6-4205-8a97-9955c244cf5d"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "auc score:  0.9889616835417908\n"
     ]
    }
   ],
   "source": [
    "auc = roc_auc_score(y_train_conv, train_pred_conv)\n",
    "print(\"auc score: \",auc)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {
    "id": "6A1mUa89XhL1"
   },
   "outputs": [],
   "source": [
    "valid_pred = model.predict(X_valid)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "pmeZGFjleo7n",
    "outputId": "cb1a33e9-3792-4d6f-c688-cd76b8a97d1b"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0]"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# classifying the predicted categories\n",
    "valid_pred_conv = []\n",
    "for i in valid_pred:\n",
    "  valid_pred_conv.append(float(np.argmax(i)))\n",
    "\n",
    "valid_pred_conv"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {
    "id": "YtOhO1p6xJ7N"
   },
   "outputs": [],
   "source": [
    "y_valid_conv = []\n",
    "for i in y_valid:\n",
    "  y_valid_conv.append(float(np.argmax(i)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "W4Np8ey3Xg5o",
    "outputId": "eca174d5-7349-4dcd-a451-7b969c6d2047"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<tf.Tensor: shape=(2, 2), dtype=int32, numpy=\n",
       "array([[86,  0],\n",
       "       [ 4, 88]], dtype=int32)>"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tf.math.confusion_matrix(y_valid_conv, valid_pred_conv)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "xFrVlTejNBDL",
    "outputId": "43c3d8b8-8ee3-4e4e-9431-e630f5911a97"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "         0.0       0.96      1.00      0.98        86\n",
      "         1.0       1.00      0.96      0.98        92\n",
      "\n",
      "    accuracy                           0.98       178\n",
      "   macro avg       0.98      0.98      0.98       178\n",
      "weighted avg       0.98      0.98      0.98       178\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(classification_report(y_valid_conv, valid_pred_conv))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "vefw5T56OD7C",
    "outputId": "03ee92c3-174e-41ec-8b56-829e7d9ce4d7"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "auc score:  0.9782608695652174\n"
     ]
    }
   ],
   "source": [
    "auc = roc_auc_score(y_valid_conv, valid_pred_conv)\n",
    "print(\"auc score: \",auc)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {
    "id": "R_egIRQ3ODmD"
   },
   "outputs": [],
   "source": [
    "test_path = '/content/test'\n",
    "\n",
    "row, col = 48, 48\n",
    "classes = 2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {
    "id": "OKNDwVFGxaAg"
   },
   "outputs": [],
   "source": [
    "os.mkdir(\"test\")\n",
    "os.mkdir(\"test/negative\")\n",
    "os.mkdir(\"test/positive\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "hNtgbyNwODDi",
    "outputId": "50c3886c-9cbf-4b7b-fa47-40cd00b73c56"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "negative has 20 number images\n",
      "positive has 20 number images\n",
      "Total number images is 40\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "40"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "total_test_images=count_image(test_path)\n",
    "total_test_images"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "h8fSuroUPfAW",
    "outputId": "44c85485-255e-4e10-8534-e1f0711025a4"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "All images converted to array of negative directory\n",
      "All images converted to array of positive directory\n"
     ]
    }
   ],
   "source": [
    "test_image_array, test_image_label, test_label_to_text, img_label = image_to_array(total_test_images, test_path)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {
    "id": "JwyH4J5bQkxh"
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 177
    },
    "id": "VX6cdGRePe7k",
    "outputId": "6a48a5e0-6722-4936-c287-b8a1b8e452d5"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1008x1008 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "show_some_image(test_label_to_text, test_image_array, test_image_label)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "p3VXxc4MR2u_",
    "outputId": "6034e3d6-1e1c-43e7-c340-09656aaa72eb"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "40"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(test_image_label)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {
    "id": "JeD0Qlm0R2pI"
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {
    "id": "ZH8ZpbopPerc"
   },
   "outputs": [],
   "source": [
    "test_image_label_cat = np_utils.to_categorical(test_image_label)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {
    "id": "usCy0OS-Pem6"
   },
   "outputs": [],
   "source": [
    "test_image_array_resize = test_image_array/ 255.0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "agkLqr_cPeSU",
    "outputId": "6323bb88-f21e-45fb-91b5-87bc18ee9121"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([1., 0.], dtype=float32)"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_image_label_cat[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "CdWgGa6XN9w7",
    "outputId": "e7fc211b-5ff1-464e-a960-b548ca85a9ce"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2/2 [==============================] - 0s 188ms/step - loss: 0.0584 - accuracy: 0.9750\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[0.05837564542889595, 0.9750000238418579]"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.evaluate(test_image_array_resize, test_image_label_cat)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {
    "id": "lyWy_6SoX3fN"
   },
   "outputs": [],
   "source": [
    "test_pred = model.predict(test_image_array_resize)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "hNcLz361zOCq",
    "outputId": "b0791822-7638-4d67-9966-768480485f78"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.9986443 , 0.00154089], dtype=float32)"
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_pred[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "qH3Am5-hy9qX",
    "outputId": "1883547a-7d5a-4ab2-f67d-b72c147d61c1"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0]"
      ]
     },
     "execution_count": 62,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_pred_conv = []\n",
    "for i in test_pred:\n",
    "  test_pred_conv.append(float(np.argmax(i)))\n",
    "\n",
    "test_pred_conv"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {
    "id": "KxOtixc4fHnM"
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "TY_b8K8OYDxT",
    "outputId": "5ced4007-3773-40af-fe24-16f08b0667b8"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<tf.Tensor: shape=(2, 2), dtype=int32, numpy=\n",
       "array([[19,  1],\n",
       "       [ 0, 20]], dtype=int32)>"
      ]
     },
     "execution_count": 63,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tf.math.confusion_matrix(test_image_label, test_pred_conv)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "PCW4F6L3zbb8",
    "outputId": "f2a82395-9fde-4c88-c1f1-a8f4e76e0bfc"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "         0.0       1.00      0.95      0.97        20\n",
      "         1.0       0.95      1.00      0.98        20\n",
      "\n",
      "    accuracy                           0.97        40\n",
      "   macro avg       0.98      0.97      0.97        40\n",
      "weighted avg       0.98      0.97      0.97        40\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(classification_report(test_image_label, test_pred_conv))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "ENu0cX2p2qUH",
    "outputId": "a0270a0b-d66f-4b3b-e619-f6e971413053"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "auc score:  0.975\n"
     ]
    }
   ],
   "source": [
    "auc = roc_auc_score(test_image_label, test_pred_conv)\n",
    "print(\"auc score: \",auc)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "ksbeIbUQUPSa"
   },
   "source": [
    "Using Fer data "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {
    "id": "hTK2ixlzT-OS"
   },
   "outputs": [],
   "source": [
    "test_path = '/content/test2'\n",
    "\n",
    "row, col = 48, 48\n",
    "classes = 2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "QDKQMGCSUbia"
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {
    "id": "hdisAaxrUqmx"
   },
   "outputs": [],
   "source": [
    "os.mkdir(\"test2\")\n",
    "os.mkdir(\"test2/negative\")\n",
    "os.mkdir(\"test2/positive\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "eP2TD9uTUqm1",
    "outputId": "790c4114-e2a2-4823-882f-a6874d5aebb8"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "negative has 84 number images\n",
      "positive has 83 number images\n",
      "Total number images is 167\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "167"
      ]
     },
     "execution_count": 68,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "total_test_images=count_image(test_path)\n",
    "total_test_images"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "XqGGyB7QUqm3",
    "outputId": "09af7b61-6cba-4b65-baa2-1b02552f86a5"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "All images converted to array of negative directory\n",
      "All images converted to array of positive directory\n"
     ]
    }
   ],
   "source": [
    "test_image_array, test_image_label, test_label_to_text, img_label = image_to_array(total_test_images, test_path)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {
    "id": "dgSyDYSzUqm4"
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 177
    },
    "id": "NVIWVULUUqm4",
    "outputId": "43c96cb1-6edd-4729-c4fb-0990e88a16b6"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1008x1008 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "show_some_image(test_label_to_text, test_image_array, test_image_label)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "26qimycWUqm6",
    "outputId": "5b0a3bae-9d74-4aea-a7c0-74c04d8b8a43"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "167"
      ]
     },
     "execution_count": 71,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(test_image_label)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {
    "id": "OuImUO6fUqm7"
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {
    "id": "HZsD8KnfUqm7"
   },
   "outputs": [],
   "source": [
    "test_image_label_cat = np_utils.to_categorical(test_image_label)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {
    "id": "ERwhZ2gTUqm8"
   },
   "outputs": [],
   "source": [
    "test_image_array_resize = test_image_array/ 255.0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "-irAMjQqUqm9",
    "outputId": "556088f1-9a9e-4c47-93b7-5ee46654421c"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([1., 0.], dtype=float32)"
      ]
     },
     "execution_count": 74,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_image_label_cat[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "moi_R6c4Uqm-",
    "outputId": "7cece63e-9fcc-4a59-9c58-6bbb6b82ec43"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "6/6 [==============================] - 0s 48ms/step - loss: 0.9027 - accuracy: 0.7545\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[0.902657151222229, 0.7544910311698914]"
      ]
     },
     "execution_count": 75,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.evaluate(test_image_array_resize, test_image_label_cat)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {
    "id": "SwpgXeQ4Uqm_"
   },
   "outputs": [],
   "source": [
    "test_pred = model.predict(test_image_array_resize)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "D_SFZMvgUqm_",
    "outputId": "f05165e1-287c-4c7b-cc54-93bde7438466"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([9.9960619e-01, 3.8873186e-04], dtype=float32)"
      ]
     },
     "execution_count": 77,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_pred[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "sV3B0_CYUqnA",
    "outputId": "e9e5fcd9-805e-461b-c4d0-d57cb8dff148"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 0.0,\n",
       " 0.0]"
      ]
     },
     "execution_count": 78,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_pred_conv = []\n",
    "for i in test_pred:\n",
    "  test_pred_conv.append(float(np.argmax(i)))\n",
    "\n",
    "test_pred_conv"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {
    "id": "gbJnOv6KUqnB"
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "dmID7aVfUqnC",
    "outputId": "03b8a0a3-546f-436d-f1c2-d1a8b8871003"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<tf.Tensor: shape=(2, 2), dtype=int32, numpy=\n",
       "array([[70, 14],\n",
       "       [27, 56]], dtype=int32)>"
      ]
     },
     "execution_count": 79,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tf.math.confusion_matrix(test_image_label, test_pred_conv)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "07W_HyxRUqnC",
    "outputId": "ddc07eab-b3c3-4484-d745-dbc9713fe3e5"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "         0.0       0.72      0.83      0.77        84\n",
      "         1.0       0.80      0.67      0.73        83\n",
      "\n",
      "    accuracy                           0.75       167\n",
      "   macro avg       0.76      0.75      0.75       167\n",
      "weighted avg       0.76      0.75      0.75       167\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(classification_report(test_image_label, test_pred_conv))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "4xkCkSfpUqnE",
    "outputId": "8582dba2-6f1f-4d7a-8952-594f780e4dbe"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "auc score:  0.7540160642570282\n"
     ]
    }
   ],
   "source": [
    "auc = roc_auc_score(test_image_label, test_pred_conv)\n",
    "print(\"auc score: \",auc)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {
    "id": "BsNeF-ChUbSz"
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "3FY5VXYBI5ju"
   },
   "source": [
    "### storing the model in json"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {
    "id": "2HfiTPsmzn_2"
   },
   "outputs": [],
   "source": [
    "os.mkdir(\"model\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "Jgm4iz9FI5jv",
    "outputId": "31ba75dd-449e-41f8-b961-ca67aee7d010"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "6/6 - 0s - loss: 0.1021 - accuracy: 0.9775 - 76ms/epoch - 13ms/step\n"
     ]
    }
   ],
   "source": [
    "from keras.models import model_from_json\n",
    "\n",
    "# serialize model to json\n",
    "json_model = model.to_json()\n",
    "\n",
    "#save the model architecture to JSON file\n",
    "with open('model/cnn_model3.json', 'w') as json_file:\n",
    "    json_file.write(json_model)\n",
    "\n",
    "#saving the weights of the model\n",
    "model.save_weights('model/cnn_weights3.h5')\n",
    "\n",
    "#Model loss and accuracy\n",
    "loss,acc = model.evaluate(X_valid, y_valid, verbose=2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "EWPfqkJUI5j3"
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "-91R0FoqI5j3"
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "accelerator": "GPU",
  "colab": {
   "collapsed_sections": [],
   "name": "colab_image_model3.ipynb",
   "provenance": []
  },
  "gpuClass": "standard",
  "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": 1
}