age-sex-ethnicity-detection-using-multi-ouput.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "polish-shower",
   "metadata": {
    "_cell_guid": "b1076dfc-b9ad-4769-8c92-a6c4dae69d19",
    "_uuid": "8f2839f25d086af736a60e9eeb907d3b93b6e0e5",
    "execution": {
     "iopub.execute_input": "2021-05-11T11:18:39.036544Z",
     "iopub.status.busy": "2021-05-11T11:18:39.035824Z",
     "iopub.status.idle": "2021-05-11T11:18:45.826664Z",
     "shell.execute_reply": "2021-05-11T11:18:45.825945Z"
    },
    "papermill": {
     "duration": 6.81294,
     "end_time": "2021-05-11T11:18:45.826877",
     "exception": false,
     "start_time": "2021-05-11T11:18:39.013937",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "import tensorflow as tf\n",
    "\n",
    "import numpy as np\n",
    "import os\n",
    "from kaggle_datasets import KaggleDatasets"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "formed-purple",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-05-11T11:18:45.923692Z",
     "iopub.status.busy": "2021-05-11T11:18:45.922815Z",
     "iopub.status.idle": "2021-05-11T11:18:52.272406Z",
     "shell.execute_reply": "2021-05-11T11:18:52.271255Z"
    },
    "papermill": {
     "duration": 6.435377,
     "end_time": "2021-05-11T11:18:52.272560",
     "exception": false,
     "start_time": "2021-05-11T11:18:45.837183",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Running on TPU  grpc://10.0.0.2:8470\n",
      "REPLICAS:  8\n"
     ]
    }
   ],
   "source": [
    "AUTO = tf.data.experimental.AUTOTUNE\n",
    "\n",
    "try:\n",
    "    tpu = tf.distribute.cluster_resolver.TPUClusterResolver()  \n",
    "    print('Running on TPU ', tpu.master())\n",
    "except ValueError:\n",
    "    tpu = None\n",
    "\n",
    "if tpu:\n",
    "    tf.config.experimental_connect_to_cluster(tpu)\n",
    "    tf.tpu.experimental.initialize_tpu_system(tpu)\n",
    "    strategy = tf.distribute.experimental.TPUStrategy(tpu)\n",
    "else:\n",
    "    strategy = tf.distribute.get_strategy()\n",
    "\n",
    "print(\"REPLICAS: \", strategy.num_replicas_in_sync)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "divine-evening",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-05-11T11:18:52.307638Z",
     "iopub.status.busy": "2021-05-11T11:18:52.300311Z",
     "iopub.status.idle": "2021-05-11T11:18:52.750347Z",
     "shell.execute_reply": "2021-05-11T11:18:52.749668Z"
    },
    "papermill": {
     "duration": 0.466849,
     "end_time": "2021-05-11T11:18:52.750498",
     "exception": false,
     "start_time": "2021-05-11T11:18:52.283649",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "gs://kds-6d3cf3666dde7c685207ed552ff146f9f315e466366bade3da15dd2d\n"
     ]
    }
   ],
   "source": [
    "GCS_DS_PATH = KaggleDatasets().get_gcs_path(\"agedetection\")\n",
    "TRAIN_PATH = GCS_DS_PATH + \"/dataset/dataset\"\n",
    "\n",
    "HEIGHT,WIDTH = 128,128\n",
    "CHANNELS = 3\n",
    "BATCH_SIZE = 8 * strategy.num_replicas_in_sync\n",
    "SEED = 143\n",
    "\n",
    "AUTO = tf.data.experimental.AUTOTUNE\n",
    "print(GCS_DS_PATH)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "crude-jewel",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-05-11T11:18:52.790855Z",
     "iopub.status.busy": "2021-05-11T11:18:52.789778Z",
     "iopub.status.idle": "2021-05-11T11:18:52.792421Z",
     "shell.execute_reply": "2021-05-11T11:18:52.792896Z"
    },
    "papermill": {
     "duration": 0.030765,
     "end_time": "2021-05-11T11:18:52.793078",
     "exception": false,
     "start_time": "2021-05-11T11:18:52.762313",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "def process_img(filepath,label):\n",
    "    image = tf.io.read_file(filepath)\n",
    "    image = tf.image.decode_jpeg(image, channels=CHANNELS)\n",
    "    image = tf.image.convert_image_dtype(image, tf.float32) \n",
    "    image = tf.image.resize(image, [HEIGHT,WIDTH])\n",
    "    return image,label\n",
    "\n",
    "def data_augment(image, label):\n",
    "    p_spatial = tf.random.uniform([], 0, 1.0, dtype=tf.float32)\n",
    "    p_rotate = tf.random.uniform([], 0, 1.0, dtype=tf.float32)\n",
    "    p_pixel_1 = tf.random.uniform([], 0, 1.0, dtype=tf.float32)\n",
    "    p_pixel_2 = tf.random.uniform([], 0, 1.0, dtype=tf.float32)\n",
    "    p_pixel_3 = tf.random.uniform([], 0, 1.0, dtype=tf.float32)\n",
    "    p_crop = tf.random.uniform([], 0, 1.0, dtype=tf.float32)\n",
    "            \n",
    "    # Flips\n",
    "    image = tf.image.random_flip_left_right(image)\n",
    "    image = tf.image.random_flip_up_down(image)\n",
    "    if p_spatial > .75:\n",
    "        image = tf.image.transpose(image)\n",
    "        \n",
    "    # Rotates\n",
    "    if p_rotate > .75:\n",
    "        image = tf.image.rot90(image, k=3) \n",
    "    elif p_rotate > .5:\n",
    "        image = tf.image.rot90(image, k=2) \n",
    "    elif p_rotate > .25:\n",
    "        image = tf.image.rot90(image, k=1) \n",
    "        \n",
    "    \n",
    "    if p_pixel_1 >= .4:\n",
    "        image = tf.image.random_saturation(image, lower=.7, upper=1.3)\n",
    "    if p_pixel_2 >= .4:\n",
    "        image = tf.image.random_contrast(image, lower=.8, upper=1.2)\n",
    "    if p_pixel_3 >= .4:\n",
    "        image = tf.image.random_brightness(image, max_delta=.1)\n",
    "        \n",
    "    \n",
    "    if p_crop > .7:\n",
    "        if p_crop > .9:\n",
    "            image = tf.image.central_crop(image, central_fraction=.7)\n",
    "        elif p_crop > .8:\n",
    "            image = tf.image.central_crop(image, central_fraction=.8)\n",
    "        else:\n",
    "            image = tf.image.central_crop(image, central_fraction=.9)\n",
    "    elif p_crop > .4:\n",
    "        crop_size = tf.random.uniform([], int(HEIGHT*.8), HEIGHT, dtype=tf.int32)\n",
    "        image = tf.image.random_crop(image, size=[crop_size, crop_size, CHANNELS])\n",
    "    \n",
    "    image = tf.image.resize(image, [HEIGHT,WIDTH])\n",
    "    return image,label"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "pointed-groove",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-05-11T11:18:52.827503Z",
     "iopub.status.busy": "2021-05-11T11:18:52.826772Z",
     "iopub.status.idle": "2021-05-11T11:18:56.290695Z",
     "shell.execute_reply": "2021-05-11T11:18:56.289968Z"
    },
    "papermill": {
     "duration": 3.487178,
     "end_time": "2021-05-11T11:18:56.290863",
     "exception": false,
     "start_time": "2021-05-11T11:18:52.803685",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "files_ls = tf.io.gfile.glob(TRAIN_PATH + '/*.jpg')\n",
    "filenames = [name.split(\"/\")[-1] for name in files_ls]\n",
    "\n",
    "age_label = np.array(tf.one_hot([int(name.split(\"_\")[0])//10 for name in filenames],depth=12))\n",
    "gender_label = np.array(tf.one_hot([int(name.split(\"_\")[1]) for name in filenames],depth=2))\n",
    "ethnicity_label = np.array(tf.one_hot([int(name.split(\"_\")[2]) for name in filenames],depth=5))\n",
    "\n",
    "labels = np.concatenate((age_label,gender_label,ethnicity_label),axis=1)\n",
    "\n",
    "dataset = tf.data.Dataset.from_tensor_slices((files_ls,labels))\n",
    "dataset = dataset.map(process_img,num_parallel_calls=AUTO).prefetch(AUTO)\n",
    "dataset = dataset.map(data_augment,num_parallel_calls=AUTO).prefetch(AUTO)\n",
    "\n",
    "SPLIT = int(0.8*len(files_ls))\n",
    "train_ds = dataset.take(SPLIT)\n",
    "val_ds = dataset.skip(SPLIT)\n",
    "\n",
    "train_ds = train_ds.cache().repeat().shuffle(2048).batch(BATCH_SIZE).prefetch(AUTO)\n",
    "val_ds = val_ds.cache().repeat().batch(BATCH_SIZE).prefetch(AUTO)\n",
    "\n",
    "STEPS_PER_EPOCH  = SPLIT//BATCH_SIZE\n",
    "VALID_STEPS = (len(files_ls)-SPLIT)//BATCH_SIZE"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "precise-peripheral",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-05-11T11:18:56.323914Z",
     "iopub.status.busy": "2021-05-11T11:18:56.322910Z",
     "iopub.status.idle": "2021-05-11T11:18:58.929018Z",
     "shell.execute_reply": "2021-05-11T11:18:58.928463Z"
    },
    "papermill": {
     "duration": 2.627672,
     "end_time": "2021-05-11T11:18:58.929167",
     "exception": false,
     "start_time": "2021-05-11T11:18:56.301495",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Downloading data from https://storage.googleapis.com/tensorflow/keras-applications/mobilenet_v2/mobilenet_v2_weights_tf_dim_ordering_tf_kernels_1.0_128_no_top.h5\n",
      "9412608/9406464 [==============================] - 0s 0us/step\n",
      "Model: \"model\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "input_2 (InputLayer)         [(None, 128, 128, 3)]     0         \n",
      "_________________________________________________________________\n",
      "mobilenetv2_1.00_128 (Functi (None, 4, 4, 1280)        2257984   \n",
      "_________________________________________________________________\n",
      "global_average_pooling2d (Gl (None, 1280)              0         \n",
      "_________________________________________________________________\n",
      "dense (Dense)                (None, 19)                24339     \n",
      "=================================================================\n",
      "Total params: 2,282,323\n",
      "Trainable params: 2,248,211\n",
      "Non-trainable params: 34,112\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "def create_model():\n",
    "    \n",
    "    pretrained = tf.keras.applications.MobileNetV2(include_top=False,weights=\"imagenet\",input_shape = (HEIGHT,WIDTH,CHANNELS))\n",
    "    \n",
    "    inputs = tf.keras.Input(shape = (HEIGHT,WIDTH,CHANNELS))\n",
    "    x = pretrained(inputs)\n",
    "    x = tf.keras.layers.GlobalAveragePooling2D()(x)\n",
    "    \n",
    "    outputs = tf.keras.layers.Dense(19,activation=\"sigmoid\")(x)\n",
    "\n",
    "    \n",
    "    model = tf.keras.Model(inputs,outputs)\n",
    "    return model\n",
    "\n",
    "model = create_model()\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "utility-occurrence",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-05-11T11:18:58.961962Z",
     "iopub.status.busy": "2021-05-11T11:18:58.961229Z",
     "iopub.status.idle": "2021-05-11T11:18:58.965925Z",
     "shell.execute_reply": "2021-05-11T11:18:58.965313Z"
    },
    "papermill": {
     "duration": 0.023204,
     "end_time": "2021-05-11T11:18:58.966130",
     "exception": false,
     "start_time": "2021-05-11T11:18:58.942926",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "def compile_model(model, lr=0.001):\n",
    "    \n",
    "    optimizer = tf.keras.optimizers.Adam(lr=lr)\n",
    "    \n",
    "    loss = tf.keras.losses.BinaryCrossentropy()\n",
    "    metrics = [\n",
    "        tf.keras.metrics.BinaryAccuracy(name='acc')\n",
    "    ]\n",
    "\n",
    "    model.compile(optimizer=optimizer, loss=loss, metrics=metrics)\n",
    "\n",
    "    return model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "further-dayton",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-05-11T11:18:58.995324Z",
     "iopub.status.busy": "2021-05-11T11:18:58.994601Z",
     "iopub.status.idle": "2021-05-11T11:18:59.001336Z",
     "shell.execute_reply": "2021-05-11T11:18:59.001810Z"
    },
    "papermill": {
     "duration": 0.023063,
     "end_time": "2021-05-11T11:18:59.001990",
     "exception": false,
     "start_time": "2021-05-11T11:18:58.978927",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "def create_callbacks():\n",
    "    checkpoint = tf.keras.callbacks.ModelCheckpoint(\n",
    "        filepath=\"./best_model.h5\",\n",
    "        save_weights_only=False,\n",
    "        monitor='val_acc',\n",
    "        mode='max',\n",
    "        save_best_only=True,\n",
    "        save_freq='epoch',\n",
    "        verbose = 1\n",
    "    )\n",
    "    \n",
    "    reducelr = tf.keras.callbacks.ReduceLROnPlateau(\n",
    "        monitor= 'val_acc',\n",
    "        mode='max',\n",
    "        factor=0.5,\n",
    "        patience=3,\n",
    "        verbose=1\n",
    "    )\n",
    "\n",
    "    earlystop = tf.keras.callbacks.EarlyStopping(\n",
    "        monitor= 'val_acc',\n",
    "        mode='max',\n",
    "        patience=10, \n",
    "        verbose=1\n",
    "    )\n",
    "    \n",
    "    callbacks = [checkpoint, reducelr, earlystop]\n",
    "    return callbacks"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "after-hamburg",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-05-11T11:18:59.031930Z",
     "iopub.status.busy": "2021-05-11T11:18:59.031273Z",
     "iopub.status.idle": "2021-05-11T11:44:42.128818Z",
     "shell.execute_reply": "2021-05-11T11:44:42.128126Z"
    },
    "papermill": {
     "duration": 1543.113422,
     "end_time": "2021-05-11T11:44:42.128977",
     "exception": false,
     "start_time": "2021-05-11T11:18:59.015555",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/40\n",
      "296/296 [==============================] - 1364s 4s/step - loss: 0.2430 - acc: 0.8909 - val_loss: 1.5170 - val_acc: 0.8798\n",
      "\n",
      "Epoch 00001: val_acc improved from -inf to 0.87980, saving model to ./best_model.h5\n",
      "Epoch 2/40\n",
      "296/296 [==============================] - 11s 37ms/step - loss: 0.1890 - acc: 0.9217 - val_loss: 1.9510 - val_acc: 0.8421\n",
      "\n",
      "Epoch 00002: val_acc did not improve from 0.87980\n",
      "Epoch 3/40\n",
      "296/296 [==============================] - 11s 36ms/step - loss: 0.1725 - acc: 0.9301 - val_loss: 1.6752 - val_acc: 0.8184\n",
      "\n",
      "Epoch 00003: val_acc did not improve from 0.87980\n",
      "Epoch 4/40\n",
      "296/296 [==============================] - 11s 37ms/step - loss: 0.1626 - acc: 0.9340 - val_loss: 1.6726 - val_acc: 0.8590\n",
      "\n",
      "Epoch 00004: val_acc did not improve from 0.87980\n",
      "\n",
      "Epoch 00004: ReduceLROnPlateau reducing learning rate to 0.0005000000237487257.\n",
      "Epoch 5/40\n",
      "296/296 [==============================] - 11s 37ms/step - loss: 0.1466 - acc: 0.9417 - val_loss: 0.9702 - val_acc: 0.8870\n",
      "\n",
      "Epoch 00005: val_acc improved from 0.87980 to 0.88705, saving model to ./best_model.h5\n",
      "Epoch 6/40\n",
      "296/296 [==============================] - 11s 37ms/step - loss: 0.1321 - acc: 0.9481 - val_loss: 0.9197 - val_acc: 0.8927\n",
      "\n",
      "Epoch 00006: val_acc improved from 0.88705 to 0.89269, saving model to ./best_model.h5\n",
      "Epoch 7/40\n",
      "296/296 [==============================] - 11s 37ms/step - loss: 0.1235 - acc: 0.9512 - val_loss: 0.9451 - val_acc: 0.8920\n",
      "\n",
      "Epoch 00007: val_acc did not improve from 0.89269\n",
      "Epoch 8/40\n",
      "296/296 [==============================] - 11s 36ms/step - loss: 0.1178 - acc: 0.9540 - val_loss: 0.8663 - val_acc: 0.8877\n",
      "\n",
      "Epoch 00008: val_acc did not improve from 0.89269\n",
      "Epoch 9/40\n",
      "296/296 [==============================] - 11s 36ms/step - loss: 0.1097 - acc: 0.9574 - val_loss: 0.6429 - val_acc: 0.8879\n",
      "\n",
      "Epoch 00009: val_acc did not improve from 0.89269\n",
      "\n",
      "Epoch 00009: ReduceLROnPlateau reducing learning rate to 0.0002500000118743628.\n",
      "Epoch 10/40\n",
      "296/296 [==============================] - 11s 36ms/step - loss: 0.1003 - acc: 0.9616 - val_loss: 0.6137 - val_acc: 0.8817\n",
      "\n",
      "Epoch 00010: val_acc did not improve from 0.89269\n",
      "Epoch 11/40\n",
      "296/296 [==============================] - 10s 35ms/step - loss: 0.0873 - acc: 0.9673 - val_loss: 0.5914 - val_acc: 0.8828\n",
      "\n",
      "Epoch 00011: val_acc did not improve from 0.89269\n",
      "Epoch 12/40\n",
      "296/296 [==============================] - 11s 36ms/step - loss: 0.0802 - acc: 0.9698 - val_loss: 0.5962 - val_acc: 0.8897\n",
      "\n",
      "Epoch 00012: val_acc did not improve from 0.89269\n",
      "\n",
      "Epoch 00012: ReduceLROnPlateau reducing learning rate to 0.0001250000059371814.\n",
      "Epoch 13/40\n",
      "296/296 [==============================] - 11s 36ms/step - loss: 0.0789 - acc: 0.9699 - val_loss: 0.6462 - val_acc: 0.8814\n",
      "\n",
      "Epoch 00013: val_acc did not improve from 0.89269\n",
      "Epoch 14/40\n",
      "296/296 [==============================] - 11s 36ms/step - loss: 0.0732 - acc: 0.9719 - val_loss: 0.6665 - val_acc: 0.8861\n",
      "\n",
      "Epoch 00014: val_acc did not improve from 0.89269\n",
      "Epoch 15/40\n",
      "296/296 [==============================] - 11s 37ms/step - loss: 0.0702 - acc: 0.9728 - val_loss: 0.6536 - val_acc: 0.8872\n",
      "\n",
      "Epoch 00015: val_acc did not improve from 0.89269\n",
      "\n",
      "Epoch 00015: ReduceLROnPlateau reducing learning rate to 6.25000029685907e-05.\n",
      "Epoch 16/40\n",
      "296/296 [==============================] - 11s 36ms/step - loss: 0.0761 - acc: 0.9689 - val_loss: 0.6088 - val_acc: 0.8898\n",
      "\n",
      "Epoch 00016: val_acc did not improve from 0.89269\n",
      "Epoch 00016: early stopping\n"
     ]
    }
   ],
   "source": [
    "EPOCHS= 40\n",
    "VERBOSE =1\n",
    "\n",
    "tf.keras.backend.clear_session()\n",
    "\n",
    "with strategy.scope():\n",
    "    \n",
    "    model = create_model()\n",
    "    model = compile_model(model, lr=0.001)\n",
    "   \n",
    "    callbacks = create_callbacks()\n",
    "    \n",
    "    history = model.fit(\n",
    "                        train_ds,\n",
    "                        epochs=EPOCHS,\n",
    "                        callbacks=callbacks,\n",
    "                        validation_data = val_ds,\n",
    "                        verbose=VERBOSE,\n",
    "                        steps_per_epoch = STEPS_PER_EPOCH,\n",
    "                        validation_steps=VALID_STEPS\n",
    "                       )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "distinct-insert",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-05-11T11:44:43.795613Z",
     "iopub.status.busy": "2021-05-11T11:44:43.794897Z",
     "iopub.status.idle": "2021-05-11T11:44:53.209324Z",
     "shell.execute_reply": "2021-05-11T11:44:53.208740Z"
    },
    "papermill": {
     "duration": 10.2758,
     "end_time": "2021-05-11T11:44:53.209477",
     "exception": false,
     "start_time": "2021-05-11T11:44:42.933677",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Collecting mtcnn\r\n",
      "  Downloading mtcnn-0.1.0-py3-none-any.whl (2.3 MB)\r\n",
      "\u001b[K     |████████████████████████████████| 2.3 MB 2.9 MB/s \r\n",
      "\u001b[?25hRequirement already satisfied: keras>=2.0.0 in /opt/conda/lib/python3.7/site-packages (from mtcnn) (2.4.3)\r\n",
      "Requirement already satisfied: opencv-python>=4.1.0 in /opt/conda/lib/python3.7/site-packages (from mtcnn) (4.5.1.48)\r\n",
      "Requirement already satisfied: h5py in /opt/conda/lib/python3.7/site-packages (from keras>=2.0.0->mtcnn) (2.10.0)\r\n",
      "Requirement already satisfied: pyyaml in /opt/conda/lib/python3.7/site-packages (from keras>=2.0.0->mtcnn) (5.3.1)\r\n",
      "Requirement already satisfied: numpy>=1.9.1 in /opt/conda/lib/python3.7/site-packages (from keras>=2.0.0->mtcnn) (1.19.5)\r\n",
      "Requirement already satisfied: scipy>=0.14 in /opt/conda/lib/python3.7/site-packages (from keras>=2.0.0->mtcnn) (1.5.4)\r\n",
      "Requirement already satisfied: six in /opt/conda/lib/python3.7/site-packages (from h5py->keras>=2.0.0->mtcnn) (1.15.0)\r\n",
      "Installing collected packages: mtcnn\r\n",
      "Successfully installed mtcnn-0.1.0\r\n"
     ]
    }
   ],
   "source": [
    "!pip install mtcnn"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "dominican-messenger",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-05-11T11:44:54.820230Z",
     "iopub.status.busy": "2021-05-11T11:44:54.819519Z",
     "iopub.status.idle": "2021-05-11T11:45:00.484275Z",
     "shell.execute_reply": "2021-05-11T11:45:00.484737Z"
    },
    "papermill": {
     "duration": 6.475245,
     "end_time": "2021-05-11T11:45:00.484933",
     "exception": false,
     "start_time": "2021-05-11T11:44:54.009688",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Age is between 21 and 30\n",
      "Gender is Male\n",
      "Ethnicity is Black\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from mtcnn import MTCNN\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "TEST_PATH = \"../input/agedetection/predict.jpg\"\n",
    "\n",
    "gender_dict={\n",
    "    0:\"Male\",\n",
    "    1:\"Female\"\n",
    "}\n",
    "\n",
    "ethnicity_dict = {\n",
    "  0:\"White\",\n",
    "  1:\"Black\",\n",
    "  2:\"Asian\",\n",
    "  3:\"Indian\",\n",
    "  4:\"Others\"\n",
    "}\n",
    "\n",
    "img = tf.keras.preprocessing.image.load_img(TEST_PATH,target_size=(HEIGHT,WIDTH))\n",
    "img = tf.keras.preprocessing.image.img_to_array(img)\n",
    "\n",
    "detector = MTCNN()\n",
    "results = detector.detect_faces(img)\n",
    "\n",
    "if len(results) == 0:\n",
    "    print(\"No Face Detected, Try again\")\n",
    "else:\n",
    "    x, y, width, height = results[0]['box']\n",
    "    face = img[y:y+height, x:x+width]\n",
    "    face = tf.image.resize(face, [HEIGHT,WIDTH])\n",
    "    face = tf.expand_dims(face,0)\n",
    "    pred = model.predict(face)\n",
    "    age = np.argmax(pred[0][:11])\n",
    "    print(f\"Age is between {age*10+1} and {age*10+10}\")\n",
    "    gender = gender_dict[np.argmax(pred[0][12:14])]\n",
    "    print(f\"Gender is {gender}\")\n",
    "    ethnicity = ethnicity_dict[np.argmax(pred[0][14:])]\n",
    "    print(f\"Ethnicity is {ethnicity}\")\n",
    "    \n",
    "plt.imshow(face[0]/255.)\n",
    "plt.axis(\"off\")\n",
    "plt.title(\"Cropped Image\")\n",
    "plt.show()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.10"
  },
  "papermill": {
   "default_parameters": {},
   "duration": 1594.517018,
   "end_time": "2021-05-11T11:45:04.430983",
   "environment_variables": {},
   "exception": null,
   "input_path": "__notebook__.ipynb",
   "output_path": "__notebook__.ipynb",
   "parameters": {},
   "start_time": "2021-05-11T11:18:29.913965",
   "version": "2.3.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}