CNN + Residual NN.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import os\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "import warnings\n",
    "from tqdm.notebook import tqdm\n",
    "warnings.filterwarnings('ignore')\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "9936419be2be462684a9732c2e891706",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/23708 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "BASE_DIR = 'UTKFace/'\n",
    "# Reading all the images present in Base_Dir\n",
    "image_paths = []\n",
    "age_labels = []\n",
    "gender_labels = []\n",
    "\n",
    "for filename in tqdm(os.listdir(BASE_DIR)):\n",
    "    image_path = os.path.join(BASE_DIR, filename)\n",
    "    temp = filename.split('_')\n",
    "    age = int(temp[0])\n",
    "    gender = int(temp[1])\n",
    "    image_paths.append(image_path)\n",
    "    age_labels.append(age)\n",
    "    gender_labels.append(gender)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>image</th>\n",
       "      <th>age</th>\n",
       "      <th>gender</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>UTKFace/100_0_0_20170112213500903.jpg.chip.jpg</td>\n",
       "      <td>100</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>UTKFace/100_0_0_20170112215240346.jpg.chip.jpg</td>\n",
       "      <td>100</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>UTKFace/100_1_0_20170110183726390.jpg.chip.jpg</td>\n",
       "      <td>100</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>UTKFace/100_1_0_20170112213001988.jpg.chip.jpg</td>\n",
       "      <td>100</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>UTKFace/100_1_0_20170112213303693.jpg.chip.jpg</td>\n",
       "      <td>100</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                            image  age  gender\n",
       "0  UTKFace/100_0_0_20170112213500903.jpg.chip.jpg  100       0\n",
       "1  UTKFace/100_0_0_20170112215240346.jpg.chip.jpg  100       0\n",
       "2  UTKFace/100_1_0_20170110183726390.jpg.chip.jpg  100       1\n",
       "3  UTKFace/100_1_0_20170112213001988.jpg.chip.jpg  100       1\n",
       "4  UTKFace/100_1_0_20170112213303693.jpg.chip.jpg  100       1"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Converting to dataframe\n",
    "df = pd.DataFrame()\n",
    "df['image'], df['age'], df['gender'] = image_paths, age_labels, gender_labels\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "gender_dict = {0:'Male', 1:'Female'}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = df.astype({'age':'float32', 'gender': 'int32'})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "train, test = train_test_split(df, test_size=0.85, random_state=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Converting Image to numpy array (extracting feature)\n",
    "from PIL import Image\n",
    "import tensorflow as tf\n",
    "from tensorflow import keras\n",
    "from keras import utils\n",
    "from keras.preprocessing import image\n",
    "x_train = []\n",
    "for file in train.image:\n",
    "    img = keras.utils.load_img(\"\"+file, grayscale=True)\n",
    "    img = img.resize((128,128), Image.ANTIALIAS)\n",
    "    img = np.array(img)\n",
    "    x_train.append(img)\n",
    "\n",
    "x_train = np.array(x_train)\n",
    "\n",
    "x_train = x_train.reshape(len(x_train), 128,128,1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "x_train.shape\n",
    "x_train = x_train/255"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_gender = np.array(train.gender)\n",
    "y_age = np.array(train.age)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# CNN"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "from tensorflow.keras.initializers import random_uniform, glorot_uniform, constant, identity\n",
    "from tensorflow.keras.layers import Dropout, Input, Add, Dense, Activation, BatchNormalization, Flatten, Conv2D, MaxPooling2D, GlobalMaxPooling2D\n",
    "from tensorflow.keras.models import Model, load_model\n",
    "input_size = (128,128,1)\n",
    "inputs = Input((input_size))\n",
    "X = Conv2D(64, (3, 3), activation='relu', kernel_initializer = glorot_uniform(seed=0))(inputs)\n",
    "X = BatchNormalization(axis = 3)(X)\n",
    "X = MaxPooling2D((3, 3))(X)\n",
    "\n",
    "X = Conv2D(128, (3, 3), activation='relu')(X)\n",
    "X = MaxPooling2D((2, 2), strides=(2, 2))(X)\n",
    "\n",
    "X = Conv2D(256, (3, 3), activation='relu')(X)\n",
    "X = MaxPooling2D((2, 2))(X)\n",
    "\n",
    "X = Flatten()(X)\n",
    "\n",
    "dense_1 = Dense(256, activation='relu')(X)\n",
    "dense_2 = Dense(256, activation='relu' )(X)\n",
    "dense_3 = Dense(128, activation='relu' )(dense_2)\n",
    "dropout_1 = Dropout(0.4)(dense_1)\n",
    "dropout_2 = Dropout(0.4)(dense_3)\n",
    "output_1 = Dense(1,activation='sigmoid', name='gender_output')(dropout_1)\n",
    "output_2 = Dense(1, activation='relu', name='age_output')(dropout_2)\n",
    "\n",
    "model = Model(inputs=[inputs], outputs=[output_1,output_2])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"model\"\n",
      "__________________________________________________________________________________________________\n",
      " Layer (type)                   Output Shape         Param #     Connected to                     \n",
      "==================================================================================================\n",
      " input_1 (InputLayer)           [(None, 128, 128, 1  0           []                               \n",
      "                                )]                                                                \n",
      "                                                                                                  \n",
      " conv2d (Conv2D)                (None, 126, 126, 64  640         ['input_1[0][0]']                \n",
      "                                )                                                                 \n",
      "                                                                                                  \n",
      " batch_normalization (BatchNorm  (None, 126, 126, 64  256        ['conv2d[0][0]']                 \n",
      " alization)                     )                                                                 \n",
      "                                                                                                  \n",
      " max_pooling2d (MaxPooling2D)   (None, 42, 42, 64)   0           ['batch_normalization[0][0]']    \n",
      "                                                                                                  \n",
      " conv2d_1 (Conv2D)              (None, 40, 40, 128)  73856       ['max_pooling2d[0][0]']          \n",
      "                                                                                                  \n",
      " max_pooling2d_1 (MaxPooling2D)  (None, 20, 20, 128)  0          ['conv2d_1[0][0]']               \n",
      "                                                                                                  \n",
      " conv2d_2 (Conv2D)              (None, 18, 18, 256)  295168      ['max_pooling2d_1[0][0]']        \n",
      "                                                                                                  \n",
      " max_pooling2d_2 (MaxPooling2D)  (None, 9, 9, 256)   0           ['conv2d_2[0][0]']               \n",
      "                                                                                                  \n",
      " flatten (Flatten)              (None, 20736)        0           ['max_pooling2d_2[0][0]']        \n",
      "                                                                                                  \n",
      " dense_1 (Dense)                (None, 256)          5308672     ['flatten[0][0]']                \n",
      "                                                                                                  \n",
      " dense (Dense)                  (None, 256)          5308672     ['flatten[0][0]']                \n",
      "                                                                                                  \n",
      " dense_2 (Dense)                (None, 128)          32896       ['dense_1[0][0]']                \n",
      "                                                                                                  \n",
      " dropout (Dropout)              (None, 256)          0           ['dense[0][0]']                  \n",
      "                                                                                                  \n",
      " dropout_1 (Dropout)            (None, 128)          0           ['dense_2[0][0]']                \n",
      "                                                                                                  \n",
      " gender_output (Dense)          (None, 1)            257         ['dropout[0][0]']                \n",
      "                                                                                                  \n",
      " age_output (Dense)             (None, 1)            129         ['dropout_1[0][0]']              \n",
      "                                                                                                  \n",
      "==================================================================================================\n",
      "Total params: 11,020,546\n",
      "Trainable params: 11,020,418\n",
      "Non-trainable params: 128\n",
      "__________________________________________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model.compile(loss=['binary_crossentropy','mae'], optimizer='adam', metrics=['accuracy'])\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Training"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/20\n",
      "320/320 [==============================] - 186s 567ms/step - loss: 17.1335 - gender_output_loss: 0.7343 - age_output_loss: 16.3991 - gender_output_accuracy: 0.6384 - age_output_accuracy: 0.0491 - val_loss: 26.7346 - val_gender_output_loss: 0.6295 - val_age_output_loss: 26.1051 - val_gender_output_accuracy: 0.6629 - val_age_output_accuracy: 0.0562\n",
      "Epoch 2/20\n",
      "320/320 [==============================] - 195s 610ms/step - loss: 13.3193 - gender_output_loss: 0.5331 - age_output_loss: 12.7862 - gender_output_accuracy: 0.7513 - age_output_accuracy: 0.0359 - val_loss: 12.4263 - val_gender_output_loss: 0.5126 - val_age_output_loss: 11.9136 - val_gender_output_accuracy: 0.7444 - val_age_output_accuracy: 0.0478\n",
      "Epoch 3/20\n",
      "320/320 [==============================] - 161s 502ms/step - loss: 11.4878 - gender_output_loss: 0.4737 - age_output_loss: 11.0141 - gender_output_accuracy: 0.7756 - age_output_accuracy: 0.0253 - val_loss: 13.6182 - val_gender_output_loss: 0.4383 - val_age_output_loss: 13.1799 - val_gender_output_accuracy: 0.8371 - val_age_output_accuracy: 0.0197\n",
      "Epoch 4/20\n",
      "320/320 [==============================] - 162s 505ms/step - loss: 10.5514 - gender_output_loss: 0.4124 - age_output_loss: 10.1389 - gender_output_accuracy: 0.8156 - age_output_accuracy: 0.0134 - val_loss: 11.7182 - val_gender_output_loss: 0.4164 - val_age_output_loss: 11.3019 - val_gender_output_accuracy: 0.7837 - val_age_output_accuracy: 0.0253\n",
      "Epoch 5/20\n",
      "320/320 [==============================] - 154s 481ms/step - loss: 9.9265 - gender_output_loss: 0.3822 - age_output_loss: 9.5443 - gender_output_accuracy: 0.8231 - age_output_accuracy: 0.0116 - val_loss: 14.9592 - val_gender_output_loss: 0.3484 - val_age_output_loss: 14.6108 - val_gender_output_accuracy: 0.8427 - val_age_output_accuracy: 0.0056\n",
      "Epoch 6/20\n",
      "320/320 [==============================] - 151s 472ms/step - loss: 9.2857 - gender_output_loss: 0.3498 - age_output_loss: 8.9359 - gender_output_accuracy: 0.8431 - age_output_accuracy: 0.0097 - val_loss: 11.8749 - val_gender_output_loss: 0.3423 - val_age_output_loss: 11.5326 - val_gender_output_accuracy: 0.8174 - val_age_output_accuracy: 0.0028\n",
      "Epoch 7/20\n",
      "320/320 [==============================] - 167s 523ms/step - loss: 8.6475 - gender_output_loss: 0.3533 - age_output_loss: 8.2942 - gender_output_accuracy: 0.8397 - age_output_accuracy: 0.0091 - val_loss: 9.5829 - val_gender_output_loss: 0.3381 - val_age_output_loss: 9.2447 - val_gender_output_accuracy: 0.8624 - val_age_output_accuracy: 0.0084\n",
      "Epoch 8/20\n",
      "320/320 [==============================] - 169s 529ms/step - loss: 8.5848 - gender_output_loss: 0.3156 - age_output_loss: 8.2693 - gender_output_accuracy: 0.8634 - age_output_accuracy: 0.0069 - val_loss: 9.9212 - val_gender_output_loss: 0.3207 - val_age_output_loss: 9.6005 - val_gender_output_accuracy: 0.8652 - val_age_output_accuracy: 0.0056\n",
      "Epoch 9/20\n",
      "320/320 [==============================] - 157s 491ms/step - loss: 8.1210 - gender_output_loss: 0.2903 - age_output_loss: 7.8306 - gender_output_accuracy: 0.8737 - age_output_accuracy: 0.0059 - val_loss: 9.5268 - val_gender_output_loss: 0.2850 - val_age_output_loss: 9.2419 - val_gender_output_accuracy: 0.8624 - val_age_output_accuracy: 0.0000e+00\n",
      "Epoch 10/20\n",
      "320/320 [==============================] - 180s 563ms/step - loss: 7.8195 - gender_output_loss: 0.2674 - age_output_loss: 7.5520 - gender_output_accuracy: 0.8788 - age_output_accuracy: 0.0053 - val_loss: 9.2417 - val_gender_output_loss: 0.2858 - val_age_output_loss: 8.9559 - val_gender_output_accuracy: 0.8876 - val_age_output_accuracy: 0.0084\n",
      "Epoch 11/20\n",
      "320/320 [==============================] - 189s 589ms/step - loss: 7.3901 - gender_output_loss: 0.2460 - age_output_loss: 7.1440 - gender_output_accuracy: 0.8897 - age_output_accuracy: 0.0059 - val_loss: 9.8679 - val_gender_output_loss: 0.3006 - val_age_output_loss: 9.5673 - val_gender_output_accuracy: 0.8652 - val_age_output_accuracy: 0.0028\n",
      "Epoch 12/20\n",
      "320/320 [==============================] - 168s 523ms/step - loss: 7.1240 - gender_output_loss: 0.2351 - age_output_loss: 6.8889 - gender_output_accuracy: 0.8922 - age_output_accuracy: 0.0028 - val_loss: 8.6068 - val_gender_output_loss: 0.3721 - val_age_output_loss: 8.2347 - val_gender_output_accuracy: 0.8567 - val_age_output_accuracy: 0.0112\n",
      "Epoch 13/20\n",
      "320/320 [==============================] - 160s 499ms/step - loss: 6.9261 - gender_output_loss: 0.2179 - age_output_loss: 6.7082 - gender_output_accuracy: 0.8997 - age_output_accuracy: 0.0084 - val_loss: 12.0187 - val_gender_output_loss: 0.4161 - val_age_output_loss: 11.6026 - val_gender_output_accuracy: 0.8090 - val_age_output_accuracy: 0.0028\n",
      "Epoch 14/20\n",
      "320/320 [==============================] - 147s 461ms/step - loss: 6.7698 - gender_output_loss: 0.2009 - age_output_loss: 6.5688 - gender_output_accuracy: 0.9125 - age_output_accuracy: 0.0028 - val_loss: 8.9334 - val_gender_output_loss: 0.3267 - val_age_output_loss: 8.6067 - val_gender_output_accuracy: 0.8567 - val_age_output_accuracy: 0.0000e+00\n",
      "Epoch 15/20\n",
      "320/320 [==============================] - 149s 467ms/step - loss: 6.7374 - gender_output_loss: 0.1827 - age_output_loss: 6.5547 - gender_output_accuracy: 0.9175 - age_output_accuracy: 0.0044 - val_loss: 13.8342 - val_gender_output_loss: 0.3291 - val_age_output_loss: 13.5051 - val_gender_output_accuracy: 0.8680 - val_age_output_accuracy: 0.0028\n",
      "Epoch 16/20\n",
      "320/320 [==============================] - 151s 472ms/step - loss: 6.5870 - gender_output_loss: 0.1808 - age_output_loss: 6.4062 - gender_output_accuracy: 0.9241 - age_output_accuracy: 0.0050 - val_loss: 8.6738 - val_gender_output_loss: 0.3203 - val_age_output_loss: 8.3535 - val_gender_output_accuracy: 0.8848 - val_age_output_accuracy: 0.0112\n",
      "Epoch 17/20\n",
      "320/320 [==============================] - 148s 463ms/step - loss: 6.0922 - gender_output_loss: 0.1685 - age_output_loss: 5.9237 - gender_output_accuracy: 0.9216 - age_output_accuracy: 0.0047 - val_loss: 8.6750 - val_gender_output_loss: 0.3669 - val_age_output_loss: 8.3081 - val_gender_output_accuracy: 0.8596 - val_age_output_accuracy: 0.0056\n",
      "Epoch 18/20\n",
      "320/320 [==============================] - 150s 468ms/step - loss: 5.9408 - gender_output_loss: 0.1479 - age_output_loss: 5.7929 - gender_output_accuracy: 0.9347 - age_output_accuracy: 0.0059 - val_loss: 9.1811 - val_gender_output_loss: 0.3085 - val_age_output_loss: 8.8726 - val_gender_output_accuracy: 0.8904 - val_age_output_accuracy: 0.0112\n",
      "Epoch 19/20\n",
      "320/320 [==============================] - 145s 453ms/step - loss: 5.9273 - gender_output_loss: 0.1611 - age_output_loss: 5.7663 - gender_output_accuracy: 0.9362 - age_output_accuracy: 0.0041 - val_loss: 8.7536 - val_gender_output_loss: 0.3654 - val_age_output_loss: 8.3883 - val_gender_output_accuracy: 0.8680 - val_age_output_accuracy: 0.0084\n",
      "Epoch 20/20\n",
      "320/320 [==============================] - 156s 487ms/step - loss: 5.8989 - gender_output_loss: 0.1427 - age_output_loss: 5.7562 - gender_output_accuracy: 0.9384 - age_output_accuracy: 0.0081 - val_loss: 9.1644 - val_gender_output_loss: 0.3235 - val_age_output_loss: 8.8409 - val_gender_output_accuracy: 0.8708 - val_age_output_accuracy: 0.0084\n"
     ]
    }
   ],
   "source": [
    "history = model.fit(x=x_train, y=[y_gender, y_age], batch_size = 10, epochs=20, validation_split= 0.1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Visualization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plotting the Gender Losses\n",
    "plt.plot(history.history['gender_output_loss'])\n",
    "plt.plot(history.history['val_gender_output_loss'])\n",
    "plt.title('Gender loss')\n",
    "plt.ylabel('loss')\n",
    "plt.xlabel('epoch')\n",
    "plt.legend(['train', 'validation'], loc='upper left')\n",
    "plt.show()\n",
    "\n",
    "# Plotting the Age Losses\n",
    "plt.plot(history.history['age_output_loss'])\n",
    "plt.plot(history.history['val_age_output_loss'])\n",
    "plt.title('Age loss')\n",
    "plt.ylabel('loss')\n",
    "plt.xlabel('epoch')\n",
    "plt.legend(['train', 'validation'], loc='upper left')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Predicting \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Original: Gender =  Male  Age =  50.0\n",
      "1/1 [==============================] - 0s 56ms/step\n",
      "Prediction: Gender =  Male  Age =  50\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x24103b01bb0>"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "index=20\n",
    "print(\"Original: Gender = \", gender_dict[y_gender[index]],\" Age = \", y_age[index])\n",
    "\n",
    "pred = model.predict(x_train[index].reshape(1, 128, 128, 1))\n",
    "pred_gender = gender_dict[round(pred[0][0][0])] \n",
    "pred_age = round(pred[1][0][0])\n",
    "\n",
    "print(\"Prediction: Gender = \", pred_gender,\" Age = \", pred_age)\n",
    "plt.imshow(x_train[index].reshape(128,128))"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.12"
  },
  "vscode": {
   "interpreter": {
    "hash": "dc07d24e2f18896857f0b2a651fe84ba40ce7b297e58d8804a308c8039f752a6"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}