Yoga-Pose-Classification-Using-DL.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import math\n",
    "from tensorflow import keras\n",
    "import numpy as np\n",
    "import pandas as pd \n",
    "import tensorflow as tf\n",
    "import matplotlib.pyplot as plt\n",
    "from livelossplot import PlotLossesKerasTF\n",
    "from tensorflow.keras.preprocessing.image import ImageDataGenerator\n",
    "from tensorflow.keras.models import Sequential \n",
    "from tensorflow.keras.applications import VGG16, Xception\n",
    "from tensorflow.keras.layers import Dense, Conv2D, Flatten, BatchNormalization, Dropout\n",
    "from tensorflow.keras.layers import MaxPooling2D, AveragePooling2D, Activation\n",
    "\n",
    "from tensorflow.keras import optimizers\n",
    "from tensorflow.keras.utils import to_categorical\n",
    "from tensorflow.keras.models import Model\n",
    "\n",
    "from PIL import Image\n",
    "from PIL import ImageFile\n",
    "\n",
    "ImageFile.LOAD_TRUNCATED_IMAGES = True\n",
    "%matplotlib inline\n",
    "%config Completer.use_jedi = False"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Num GPUs Available:  0\n"
     ]
    }
   ],
   "source": [
    "# GPU Configurations\n",
    "\n",
    "print(\"Num GPUs Available: \", len(tf.config.experimental.list_physical_devices('GPU')))\n",
    "gpus = tf.config.experimental.list_physical_devices('GPU')\n",
    "if gpus:\n",
    "  try:\n",
    "    # Currently, memory growth needs to be the same across GPUs\n",
    "    for gpu in gpus:\n",
    "      tf.config.experimental.set_memory_growth(gpu, True)\n",
    "    logical_gpus = tf.config.experimental.list_logical_devices('GPU')\n",
    "    print(len(gpus), \"Physical GPUs,\", len(logical_gpus), \"Logical GPUs\")\n",
    "  except RuntimeError as e:\n",
    "    # Memory growth must be set before GPUs have been initialized\n",
    "    print(e)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "basedir = \"./dataset/\" # here below the train and validation data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "def removeCorruptedImages(path):\n",
    "    for filename in os.listdir(path):\n",
    "        try:\n",
    "            img = Image.open(os.path.join(path,filename))\n",
    "            img.verify() \n",
    "        except (IOError, SyntaxError) as e:\n",
    "            print('Bad file:', filename)\n",
    "            os.remove(os.path.join(path,filename))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Yoga Set 1\n",
    "yoga_train_path = os.path.join(\"yoga_set1\",\"train\") # root for training\n",
    "yoga_test_path = os.path.join(\"yoga_set1\",\"test\") # root for testing\n",
    "\n",
    "# Yoga Set 2\n",
    "# yoga_train_path = os.path.join(\"yoga_set2\",\"train\") # root for training\n",
    "# yoga_test_path = os.path.join(\"yoga_set2\",\"test\") # root for testing\n",
    "\n",
    "traindir = os.path.join(basedir,yoga_train_path) \n",
    "validdir = os.path.join(basedir,yoga_test_path) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "# Yoga Set 1\n",
    "removeCorruptedImages(os.path.join(traindir,'tree'))\n",
    "removeCorruptedImages(os.path.join(traindir,'downdog'))\n",
    "removeCorruptedImages(os.path.join(traindir,'warrior1'))\n",
    "\n",
    "removeCorruptedImages(os.path.join(validdir,'tree'))\n",
    "removeCorruptedImages(os.path.join(validdir,'downdog'))\n",
    "removeCorruptedImages(os.path.join(validdir,'warrior1'))\n",
    "\n",
    "\n",
    "# Yoga Set 2\n",
    "# removeCorruptedImages(os.path.join(traindir,'goddess'))\n",
    "# removeCorruptedImages(os.path.join(traindir,'mountain'))\n",
    "# removeCorruptedImages(os.path.join(traindir,'warrior2'))\n",
    "\n",
    "# removeCorruptedImages(os.path.join(validdir,'goddess'))\n",
    "# removeCorruptedImages(os.path.join(validdir,'mountain'))\n",
    "# removeCorruptedImages(os.path.join(validdir,'warrior2'))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Image Data Generator"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_datagen=ImageDataGenerator(rescale=1./255,shear_range=0.2,zoom_range=0.2,horizontal_flip=True)\n",
    "test_datagen=ImageDataGenerator(rescale=1./255)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "For Training: \n",
      "Found 1184 images belonging to 3 classes.\n",
      "\n",
      "For Testing: \n",
      "Found 170 images belonging to 3 classes.\n"
     ]
    }
   ],
   "source": [
    "batch_size=8\n",
    "print(\"For Training: \")\n",
    "train_datagen = train_datagen.flow_from_directory(\n",
    "                  directory = traindir,\n",
    "                  target_size=(300,300),\n",
    "                  batch_size=batch_size,\n",
    "                  shuffle=True,\n",
    "                  color_mode=\"rgb\",\n",
    "                  class_mode='categorical')\n",
    "\n",
    "print(\"\\nFor Testing: \")\n",
    "val_datagen = test_datagen.flow_from_directory(\n",
    "                directory = validdir,\n",
    "                target_size=(300,300),\n",
    "                batch_size=batch_size,\n",
    "                shuffle=False,\n",
    "                color_mode=\"rgb\",\n",
    "                class_mode='categorical')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Loading Pretrained Model\n",
    "\n",
    "# base_model_path2 = \"./pretrained_models/xception_weights_tf_dim_ordering_tf_kernels_notop.h5\"\n",
    "# base_model =  Xception(weights=base_model_path1, include_top=False, input_shape=(300, 300, 3))\n",
    "\n",
    "base_model_path1 = \"./pretrained_models/vgg16_weights_tf_dim_ordering_tf_kernels_notop.h5\"\n",
    "base_model = VGG16(weights=base_model_path1, include_top=False, input_shape=(300, 300, 3))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"model\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "input_1 (InputLayer)         [(None, 300, 300, 3)]     0         \n",
      "_________________________________________________________________\n",
      "block1_conv1 (Conv2D)        (None, 300, 300, 64)      1792      \n",
      "_________________________________________________________________\n",
      "block1_conv2 (Conv2D)        (None, 300, 300, 64)      36928     \n",
      "_________________________________________________________________\n",
      "block1_pool (MaxPooling2D)   (None, 150, 150, 64)      0         \n",
      "_________________________________________________________________\n",
      "block2_conv1 (Conv2D)        (None, 150, 150, 128)     73856     \n",
      "_________________________________________________________________\n",
      "block2_conv2 (Conv2D)        (None, 150, 150, 128)     147584    \n",
      "_________________________________________________________________\n",
      "block2_pool (MaxPooling2D)   (None, 75, 75, 128)       0         \n",
      "_________________________________________________________________\n",
      "block3_conv1 (Conv2D)        (None, 75, 75, 256)       295168    \n",
      "_________________________________________________________________\n",
      "block3_conv2 (Conv2D)        (None, 75, 75, 256)       590080    \n",
      "_________________________________________________________________\n",
      "block3_conv3 (Conv2D)        (None, 75, 75, 256)       590080    \n",
      "_________________________________________________________________\n",
      "block3_pool (MaxPooling2D)   (None, 37, 37, 256)       0         \n",
      "_________________________________________________________________\n",
      "block4_conv1 (Conv2D)        (None, 37, 37, 512)       1180160   \n",
      "_________________________________________________________________\n",
      "block4_conv2 (Conv2D)        (None, 37, 37, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block4_conv3 (Conv2D)        (None, 37, 37, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block4_pool (MaxPooling2D)   (None, 18, 18, 512)       0         \n",
      "_________________________________________________________________\n",
      "block5_conv1 (Conv2D)        (None, 18, 18, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block5_conv2 (Conv2D)        (None, 18, 18, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block5_conv3 (Conv2D)        (None, 18, 18, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block5_pool (MaxPooling2D)   (None, 9, 9, 512)         0         \n",
      "_________________________________________________________________\n",
      "sequential (Sequential)      (None, 3)                 5308931   \n",
      "=================================================================\n",
      "Total params: 20,023,619\n",
      "Trainable params: 20,023,619\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "output_neurons = 3\n",
    "\n",
    "model = Sequential()\n",
    "model.add(Flatten(input_shape=base_model.output_shape[1:]))\n",
    "model.add(Dense(128))\n",
    "model.add(Activation('relu'))\n",
    "model.add(Dropout(0.2))\n",
    "model.add(Dense(output_neurons))\n",
    "model.add(Activation('softmax'))\n",
    "\n",
    "\n",
    "model = Model(inputs=base_model.input, outputs=model(base_model.output))\n",
    "\n",
    "\n",
    "optimizers = tf.keras.optimizers.SGD(learning_rate=0.0001, momentum=0.9, nesterov=True)\n",
    "losss = tf.keras.losses.CategoricalCrossentropy(label_smoothing=0.2, from_logits=True)\n",
    "model.compile(loss=losss,\n",
    "             optimizer=optimizers,\n",
    "              metrics=['accuracy'])\n",
    "\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Callbacks\n",
    "\n",
    "class ConvolutionCallback(tf.keras.callbacks.Callback):\n",
    "        def on_epoch_end(self,epoch,logs={}):\n",
    "            if(logs.get('accuracy')>=0.97 and logs.get('val_accuracy') >=0.92):\n",
    "                print(\"Reached greater than 97.0% accuracy so cancelling training!\")\n",
    "                self.model.stop_training = True\n",
    "                \n",
    "reduce_lr = keras.callbacks.ReduceLROnPlateau(monitor='val_loss', factor=0.1,\n",
    "                              patience=2, min_lr=0.001, mode='auto')\n",
    "checkpoint = keras.callbacks.ModelCheckpoint(\"./checkpoints_models/pose_classification_model_weights2.h5\", monitor='val_accuracy',\n",
    "                             save_weights_only=True, mode='max', verbose=1)\n",
    "\n",
    "convolutionCallback = ConvolutionCallback()\n",
    "callbacks = [PlotLossesKerasTF(), checkpoint,reduce_lr, convolutionCallback]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Training"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "accuracy\n",
      "\ttraining         \t (min:    0.757, max:    0.973, cur:    0.973)\n",
      "\tvalidation       \t (min:    0.988, max:    1.000, cur:    1.000)\n",
      "Loss\n",
      "\ttraining         \t (min:    0.565, max:    0.812, cur:    0.565)\n",
      "\tvalidation       \t (min:    0.519, max:    0.553, cur:    0.519)\n",
      "\n",
      "Epoch 00004: saving model to ./checkpoints_models\\pose_classification_model_weights2.h5\n",
      "Reached greater than 97.0% accuracy so cancelling training!\n"
     ]
    }
   ],
   "source": [
    "# epoch_for_model2 = 5\n",
    "BATCH_SIZE = 16\n",
    "\n",
    "history = model.fit(train_datagen,\n",
    "                    epochs=4,\n",
    "                    validation_data = val_datagen,\n",
    "                    callbacks=callbacks\n",
    "                    )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(0)\n",
    "plt.plot(history.history['loss'],'g', label=\"Loss\")\n",
    "plt.plot(history.history['val_loss'],'b',label=\"Validation Loss\")\n",
    "plt.plot(history.history['accuracy'],'r', label=\"Accuracy\")\n",
    "plt.plot(history.history['val_accuracy'],'black', label=\"Validation Accuracy\")\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Saving Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "yoga_set1_model_save_path = \"./final_models/H5_Models/Yoga_Set1.h5\"\n",
    "# yoga_set2_model_save_path = \"./final_models/H5_Models/Yoga_Set2.h5\"\n",
    "model.save(yoga_set1_model_save_path)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"model\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "input_1 (InputLayer)         [(None, 300, 300, 3)]     0         \n",
      "_________________________________________________________________\n",
      "block1_conv1 (Conv2D)        (None, 300, 300, 64)      1792      \n",
      "_________________________________________________________________\n",
      "block1_conv2 (Conv2D)        (None, 300, 300, 64)      36928     \n",
      "_________________________________________________________________\n",
      "block1_pool (MaxPooling2D)   (None, 150, 150, 64)      0         \n",
      "_________________________________________________________________\n",
      "block2_conv1 (Conv2D)        (None, 150, 150, 128)     73856     \n",
      "_________________________________________________________________\n",
      "block2_conv2 (Conv2D)        (None, 150, 150, 128)     147584    \n",
      "_________________________________________________________________\n",
      "block2_pool (MaxPooling2D)   (None, 75, 75, 128)       0         \n",
      "_________________________________________________________________\n",
      "block3_conv1 (Conv2D)        (None, 75, 75, 256)       295168    \n",
      "_________________________________________________________________\n",
      "block3_conv2 (Conv2D)        (None, 75, 75, 256)       590080    \n",
      "_________________________________________________________________\n",
      "block3_conv3 (Conv2D)        (None, 75, 75, 256)       590080    \n",
      "_________________________________________________________________\n",
      "block3_pool (MaxPooling2D)   (None, 37, 37, 256)       0         \n",
      "_________________________________________________________________\n",
      "block4_conv1 (Conv2D)        (None, 37, 37, 512)       1180160   \n",
      "_________________________________________________________________\n",
      "block4_conv2 (Conv2D)        (None, 37, 37, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block4_conv3 (Conv2D)        (None, 37, 37, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block4_pool (MaxPooling2D)   (None, 18, 18, 512)       0         \n",
      "_________________________________________________________________\n",
      "block5_conv1 (Conv2D)        (None, 18, 18, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block5_conv2 (Conv2D)        (None, 18, 18, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block5_conv3 (Conv2D)        (None, 18, 18, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block5_pool (MaxPooling2D)   (None, 9, 9, 512)         0         \n",
      "_________________________________________________________________\n",
      "sequential (Sequential)      (None, 3)                 5308931   \n",
      "=================================================================\n",
      "Total params: 20,023,619\n",
      "Trainable params: 20,023,619\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "loaded_model = tf.keras.models.load_model('./final_models/H5_Models/Yoga_Set1.h5')\n",
    "loaded_model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: C:\\Users\\abhis\\AppData\\Local\\Temp\\tmp8f50qwf_\\assets\n",
      "File size: 76.393 Megabytes\n"
     ]
    }
   ],
   "source": [
    "## TFLite Import\n",
    "\n",
    "tf_lite_converter = tf.lite.TFLiteConverter.from_keras_model(model)\n",
    "tflite_model = tf_lite_converter.convert()\n",
    "\n",
    "yoga_final_model_name = \"Yoga_Set1_TFLite_Model.tflite\"\n",
    "# yoga_final_model_name = \"Yoga_Set2_TFLite_Model.tflite\"\n",
    "\n",
    "tflite_model_name = yoga_final_model_name\n",
    "open(tflite_model_name,\"wb\").write(tflite_model)\n",
    "\n",
    "def get_file_size(file_path):\n",
    "    size = os.path.getsize(file_path)\n",
    "    return size\n",
    "\n",
    "def convert_bytes(size, unit=None):\n",
    "    if unit == \"KB\":\n",
    "        return print('File size: ' + str(round(size / 1024, 3)) + ' Kilobytes')\n",
    "    elif unit == \"MB\":\n",
    "        return print('File size: ' + str(round(size / (1024 * 1024), 3)) + ' Megabytes')\n",
    "    else:\n",
    "        return print('File size: ' + str(size) + ' bytes')\n",
    "    \n",
    "convert_bytes(get_file_size(tflite_model_name),\"MB\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Testing "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "22/22 [==============================] - 42s 2s/step - loss: 0.5194 - accuracy: 1.0000\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[0.5194194316864014, 1.0]"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.evaluate(val_datagen)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "yoga_labels = {0:\"downdog\",1:\"tree\",2:\"warrior1\"}\n",
    "# yoga_labels = {0:\"goddess\",1:\"mountain\",2:\"warrior2\"}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Enter Image Name (from 1-15) : 11\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"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Class Predictions:  [[1.9731702e-23 1.0000000e+00 5.1614697e-20]]\n",
      "\n",
      "Prediction is:  tree\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "from keras.preprocessing import image\n",
    "import matplotlib.image as mpimg\n",
    "\n",
    "# predicting images\n",
    "path = input(\"Enter Image Name (from 1-15) : \")\n",
    "path = \"./dataset/final_test/yoga_set1/\" + path + \".jpg\"\n",
    "\n",
    "img = image.load_img(path, target_size=(300, 300))\n",
    "x = image.img_to_array(img)\n",
    "# print(x/255)\n",
    "x = np.expand_dims(x, axis=0)\n",
    "\n",
    "images = np.vstack([x])\n",
    "classes = loaded_model.predict(images, batch_size=10)\n",
    "\n",
    "plt.axis(\"Off\")\n",
    "img = mpimg.imread(path)\n",
    "plt.imshow(img)\n",
    "plt.show()\n",
    "\n",
    "print(\"Class Predictions: \",classes)\n",
    "pred_index = np.argmax(classes[0])\n",
    "print(\"\\nPrediction is: \", yoga_labels[pred_index])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "tensor",
   "language": "python",
   "name": "tensor"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}