3. Model Inception & DenseNet.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "9b50b325-ffab-4d65-bbbd-2f3938967c71",
   "metadata": {},
   "source": [
    "<div align=\"center\">\n",
    "  <img src=\"Images/image2.jpg\" alt=\"Logo\" width=\"500\">\n",
    "  <h1 style=\"font-size: 40px; text-decoration: underline;\">TRASH IMAGE CLASSIFICATION</h1>\n",
    "</div>\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6a1321b2",
   "metadata": {},
   "source": [
    "### DATA AUGMENTATION AND MODELING"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fce56fdc",
   "metadata": {},
   "source": [
    "### Splitting the data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "12c89fe3",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Copying files: 4291 files [00:15, 269.80 files/s]\n"
     ]
    }
   ],
   "source": [
    "import splitfolders\n",
    "input_folder = 'RealWaste'\n",
    "splitfolders.ratio(input_folder,output = 'RealWaste2',seed = 42,ratio = (.7,.2,.1),group_prefix=None)\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "446ad8b8",
   "metadata": {},
   "source": [
    "Observations\n",
    "\n",
    "Training Set Size:\n",
    "Largest Class: Plastic with 644 images.\n",
    "Smallest Class: Textile Trash with 222 images.\n",
    "\n",
    "Validation Set Size:\n",
    "Largest Class: Glass with 84 images.\n",
    "Smallest Class: Textile Trash with 63 images.\n",
    "\n",
    "Test Set Size:\n",
    "Largest Class: Plastic with 93 images.\n",
    "Smallest Class: Textile Trash with 33 images.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "820013ae",
   "metadata": {},
   "source": [
    "### Data Augmentation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "64d66650",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 3001 images belonging to 8 classes.\n",
      "Found 857 images belonging to 8 classes.\n",
      "Found 433 images belonging to 8 classes.\n"
     ]
    }
   ],
   "source": [
    "#train data generator\n",
    "train_dir = ImageDataGenerator(\n",
    "         rescale=1./255,\n",
    "        rotation_range=45,     \n",
    "        width_shift_range=0.2,   \n",
    "        height_shift_range=0.2,\n",
    "        shear_range=0.2,\n",
    "        zoom_range=0.2,\n",
    "        horizontal_flip=True,\n",
    "        fill_mode='reflect')\n",
    "\n",
    "train_generator = train_dir.flow_from_directory(\n",
    "    'RealWaste2/train',\n",
    "    target_size=(524, 524),\n",
    "    batch_size=32,\n",
    "    class_mode='categorical'\n",
    ")\n",
    "#Val data generator\n",
    "val_dir = ImageDataGenerator(rescale = 1./255)\n",
    "validation_generator = val_dir.flow_from_directory(\n",
    "    'RealWaste2/val',\n",
    "    target_size=(524, 524),\n",
    "    batch_size=32,\n",
    "    class_mode='categorical'\n",
    ")\n",
    "#Test Data Generator\n",
    "test_dir = ImageDataGenerator(rescale=1./255)\n",
    "test_generator = test_dir.flow_from_directory(\n",
    "    'RealWaste2/test',\n",
    "    target_size=(524, 524),\n",
    "    batch_size=32,\n",
    "    class_mode='categorical'\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "b55eb8be",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(32, 524, 524, 3)\n",
      "(32, 8)\n"
     ]
    }
   ],
   "source": [
    "train_images, train_labels = next(train_generator)\n",
    "print(train_images.shape)\n",
    "print(train_labels.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "66f8a5f3",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(32, 524, 524, 3)\n",
      "(32, 8)\n"
     ]
    }
   ],
   "source": [
    "test_images, test_labels = next(test_generator)\n",
    "print(test_images.shape)\n",
    "print(test_labels.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "f33feaab",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(32, 524, 524, 3)\n",
      "(32, 8)\n"
     ]
    }
   ],
   "source": [
    "val_images, val_labels = next(validation_generator)\n",
    "print(val_images.shape)\n",
    "print(val_labels.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cb580b8e",
   "metadata": {},
   "source": [
    "The original image serves as a baseline, while the augmented images show various transformations applied. This includes rotations, shifts, shears, and zooms. The augmentations introduced significant variations, demonstrating how the data augmentation techniques can enrich the dataset by creating different views of the same image."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fdb7e1a6",
   "metadata": {},
   "source": [
    "### Building A CNN Model to Classify the Images "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "06a05ee1",
   "metadata": {},
   "source": [
    "Landfills contain a wide range of waste materials, and accurately categorizing them is essential for effective waste management. By leveraging CNNs, the classification process can be automated, significantly improving the efficiency and accuracy of waste diversion strategies such as recycling, composting, and proper disposal. CNNs excel at image analysis and feature extraction, allowing them to capture intricate details and patterns from the waste images in the dataset. The convolutional layers in CNNs are able to identify edges, textures, and other relevant characteristics, enabling the network to learn and leverage these features for accurate waste classification. Pooling layers nested throughout the architecture down-sample produced feature maps, allowing models to generalize object specific characteristics into differing contexts. "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5281a778",
   "metadata": {},
   "source": [
    "\n",
    "\n",
    "### Inception V3 CNN Model\n",
    "\n",
    "Due to the suboptimal performance of our baseline models with custom convolution layers, which failed to meet our predefined success thresholds, we opted to utilize pre-trained architectures. We chose Inception V3 for the following reasons:\n",
    "\n",
    "1. **Proven Performance**: Demonstrated high accuracy across various image classification tasks.\n",
    "2. **Pre-trained Weights**: Benefits from extensive prior training on the ImageNet dataset.\n",
    "3. **Inception Modules**: Simultaneous multiple convolutional operations enhance feature capture.\n",
    "4. **Computational Efficiency**: Balances computational cost and accuracy effectively.\n",
    "5. **Scalability and Flexibility**: Easily fine-tuned and adapted for our specific dataset and requirements.\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "aa1e21ed",
   "metadata": {},
   "outputs": [],
   "source": [
    "from tensorflow.keras.applications import InceptionV3\n",
    "import tensorflow as tf\n",
    "from tensorflow.keras.models import Model, load_model, Sequential\n",
    "from tensorflow.keras.layers import Dense, GlobalAveragePooling2D, Dropout, Conv2D, MaxPooling2D, Flatten\n",
    "from tensorflow.keras.applications import InceptionV3\n",
    "from tensorflow.keras.preprocessing.image import ImageDataGenerator, img_to_array, load_img\n",
    "from tensorflow.keras.optimizers import Adam\n",
    "from tensorflow.keras.callbacks import EarlyStopping, ReduceLROnPlateau\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import os"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "4943a605",
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "\n",
    "class InceptionV3Model:\n",
    "    def __init__(self, input_shape=(224, 224, 3), num_classes=8):\n",
    "        self.input_shape = input_shape\n",
    "        self.num_classes = num_classes\n",
    "        self.model = None\n",
    "        self.history = None\n",
    "\n",
    "    def build_model(self):\n",
    "        # Load the InceptionV3 model pre-trained on ImageNet\n",
    "        base_model = InceptionV3(weights='imagenet', include_top=False, input_shape=self.input_shape)\n",
    "\n",
    "        # Freeze the base model layers\n",
    "        for layer in base_model.layers:\n",
    "            layer.trainable = False\n",
    "\n",
    "        # Add custom layers on top of the base model\n",
    "        x = base_model.output\n",
    "        x = GlobalAveragePooling2D()(x)\n",
    "        x = Dense(512, activation='relu')(x)\n",
    "        x = Dropout(0.5)(x)\n",
    "        x = Dense(256, activation='relu')(x)\n",
    "        x = Dropout(0.5)(x)\n",
    "        predictions = Dense(self.num_classes, activation='softmax')(x)  # 8 categories for classification\n",
    "\n",
    "        # Create the model\n",
    "        self.model = Model(inputs=base_model.input, outputs=predictions)\n",
    "\n",
    "        # Compile the model\n",
    "        self.model.compile(optimizer=Adam(learning_rate=0.001), loss='categorical_crossentropy', metrics=['accuracy'])\n",
    "\n",
    "    def train_model(self, train_dir, val_dir, batch_size=32, epochs=10):\n",
    "        # Define image size and batch size\n",
    "        IMAGE_SIZE = self.input_shape[:2]\n",
    "\n",
    "        # Define data generators with reduced image size\n",
    "        train_datagen = ImageDataGenerator(\n",
    "            rescale=1./255,\n",
    "            rotation_range=30,\n",
    "            width_shift_range=0.2,\n",
    "            height_shift_range=0.2,\n",
    "            shear_range=0.2,\n",
    "            zoom_range=0.2,\n",
    "            horizontal_flip=True,\n",
    "            fill_mode='nearest'\n",
    "        )\n",
    "        validation_datagen = ImageDataGenerator(rescale=1./255)\n",
    "\n",
    "        train_generator = train_datagen.flow_from_directory(\n",
    "            train_dir,\n",
    "            target_size=IMAGE_SIZE,\n",
    "            batch_size=batch_size,\n",
    "            class_mode='categorical'\n",
    "        )\n",
    "        validation_generator = validation_datagen.flow_from_directory(\n",
    "            val_dir,\n",
    "            target_size=IMAGE_SIZE,\n",
    "            batch_size=batch_size,\n",
    "            class_mode='categorical'\n",
    "        )\n",
    "\n",
    "        # Define callbacks\n",
    "        early_stopping = EarlyStopping(monitor='val_loss', patience=5, restore_best_weights=True)\n",
    "        reduce_lr = ReduceLROnPlateau(monitor='val_loss', patience=3)\n",
    "\n",
    "        # Train the model\n",
    "        self.history = self.model.fit(\n",
    "            train_generator,\n",
    "            epochs=epochs,\n",
    "            validation_data=validation_generator,\n",
    "            callbacks=[early_stopping, reduce_lr]\n",
    "        )\n",
    "\n",
    "    def evaluate_model(self, test_dir, batch_size=32):\n",
    "        # Define image size\n",
    "        IMAGE_SIZE = self.input_shape[:2]\n",
    "\n",
    "        # Define data generator\n",
    "        test_datagen = ImageDataGenerator(rescale=1./255)\n",
    "\n",
    "        test_generator = test_datagen.flow_from_directory(\n",
    "            test_dir,\n",
    "            target_size=IMAGE_SIZE,\n",
    "            batch_size=batch_size,\n",
    "            class_mode='categorical'\n",
    "        )\n",
    "\n",
    "        # Evaluate the model\n",
    "        test_loss, test_accuracy = self.model.evaluate(test_generator)\n",
    "        print(f'Test Accuracy: {test_accuracy:.2f}')\n",
    "        return test_loss, test_accuracy\n",
    "\n",
    "    def save_model(self, file_path):\n",
    "        self.model.save(file_path)\n",
    "        print(f\"Model saved to {file_path}\")\n",
    "\n",
    "    def plot_training_history(self):\n",
    "        if self.history:\n",
    "            plt.figure(figsize=(12, 6))\n",
    "\n",
    "            # Plot training & validation accuracy values\n",
    "            plt.subplot(1, 2, 1)\n",
    "            plt.plot(self.history.history['accuracy'])\n",
    "            plt.plot(self.history.history['val_accuracy'])\n",
    "            plt.title('Model Accuracy')\n",
    "            plt.ylabel('Accuracy')\n",
    "            plt.xlabel('Epoch')\n",
    "            plt.legend(['Train', 'Validation'], loc='upper left')\n",
    "\n",
    "            # Plot training & validation loss values\n",
    "            plt.subplot(1, 2, 2)\n",
    "            plt.plot(self.history.history['loss'])\n",
    "            plt.plot(self.history.history['val_loss'])\n",
    "            plt.title('Model Loss')\n",
    "            plt.ylabel('Loss')\n",
    "            plt.xlabel('Epoch')\n",
    "            plt.legend(['Train', 'Validation'], loc='upper left')\n",
    "\n",
    "            plt.tight_layout()\n",
    "            plt.show()\n",
    "        else:\n",
    "            print(\"No training history found. Please train the model first.\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "c2e1b9fc",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Instantiate the model\n",
    "inception_model = InceptionV3Model()\n",
    "\n",
    "# Build the model\n",
    "inception_model.build_model()\n",
    "\n",
    "# Define directories\n",
    "train_dir = 'RealWaste2/train'\n",
    "val_dir = 'RealWaste2/val'\n",
    "test_dir = 'RealWaste2/test'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "1ea76c54",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 3001 images belonging to 8 classes.\n",
      "Found 857 images belonging to 8 classes.\n",
      "Epoch 1/10\n",
      "\u001b[1m94/94\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m222s\u001b[0m 2s/step - accuracy: 0.2494 - loss: 2.2943 - val_accuracy: 0.6441 - val_loss: 1.1271 - learning_rate: 0.0010\n",
      "Epoch 2/10\n",
      "\u001b[1m94/94\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m212s\u001b[0m 2s/step - accuracy: 0.5108 - loss: 1.3434 - val_accuracy: 0.6639 - val_loss: 0.9757 - learning_rate: 0.0010\n",
      "Epoch 3/10\n",
      "\u001b[1m94/94\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m243s\u001b[0m 3s/step - accuracy: 0.5841 - loss: 1.1524 - val_accuracy: 0.6826 - val_loss: 0.8570 - learning_rate: 0.0010\n",
      "Epoch 4/10\n",
      "\u001b[1m94/94\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m242s\u001b[0m 3s/step - accuracy: 0.6183 - loss: 1.0890 - val_accuracy: 0.7141 - val_loss: 0.8067 - learning_rate: 0.0010\n",
      "Epoch 5/10\n",
      "\u001b[1m94/94\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m244s\u001b[0m 3s/step - accuracy: 0.6665 - loss: 0.9635 - val_accuracy: 0.7083 - val_loss: 0.7924 - learning_rate: 0.0010\n",
      "Epoch 6/10\n",
      "\u001b[1m94/94\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m242s\u001b[0m 3s/step - accuracy: 0.6703 - loss: 0.9744 - val_accuracy: 0.7340 - val_loss: 0.7470 - learning_rate: 0.0010\n",
      "Epoch 7/10\n",
      "\u001b[1m94/94\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m240s\u001b[0m 3s/step - accuracy: 0.6700 - loss: 0.9375 - val_accuracy: 0.7223 - val_loss: 0.7564 - learning_rate: 0.0010\n",
      "Epoch 8/10\n",
      "\u001b[1m94/94\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m243s\u001b[0m 3s/step - accuracy: 0.6850 - loss: 0.9117 - val_accuracy: 0.7445 - val_loss: 0.7023 - learning_rate: 0.0010\n",
      "Epoch 9/10\n",
      "\u001b[1m94/94\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m245s\u001b[0m 3s/step - accuracy: 0.6893 - loss: 0.8658 - val_accuracy: 0.7596 - val_loss: 0.6919 - learning_rate: 0.0010\n",
      "Epoch 10/10\n",
      "\u001b[1m94/94\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m240s\u001b[0m 3s/step - accuracy: 0.6840 - loss: 0.8673 - val_accuracy: 0.7211 - val_loss: 0.7129 - learning_rate: 0.0010\n"
     ]
    }
   ],
   "source": [
    "# Train the model\n",
    "inception_model.train_model(train_dir, val_dir, batch_size=32, epochs=10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "b467f4d8",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 433 images belonging to 8 classes.\n",
      "\u001b[1m14/14\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m16s\u001b[0m 1s/step - accuracy: 0.7150 - loss: 0.7252\n",
      "Test Accuracy: 0.74\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "(0.7187150716781616, 0.7436489462852478)"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Evaluate the model\n",
    "inception_model.evaluate_model(test_dir, batch_size=32)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "f5da01b3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x600 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot the training history\n",
    "inception_model.plot_training_history()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "3c524ff9",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:You are saving your model as an HDF5 file via `model.save()` or `keras.saving.save_model(model)`. This file format is considered legacy. We recommend using instead the native Keras format, e.g. `model.save('my_model.keras')` or `keras.saving.save_model(model, 'my_model.keras')`. \n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model saved to waste_classification_inception_model.h5\n"
     ]
    }
   ],
   "source": [
    "# Save the model\n",
    "inception_model.save_model('waste_classification_inception_model.h5')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "id": "033ed1c4",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:Compiled the loaded model, but the compiled metrics have yet to be built. `model.compile_metrics` will be empty until you train or evaluate the model.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 433 images belonging to 8 classes.\n",
      "\u001b[1m14/14\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m57s\u001b[0m 2s/step\n",
      "Weighted F1 Score: 0.74\n",
      "                     precision    recall  f1-score   support\n",
      "\n",
      "      Food Organics       0.86      0.74      0.79        42\n",
      "              Glass       0.80      0.76      0.78        42\n",
      "              Metal       0.77      0.73      0.75        79\n",
      "Miscellaneous Trash       0.50      0.46      0.48        50\n",
      "              Paper       0.91      0.78      0.84        50\n",
      "            Plastic       0.72      0.82      0.76        93\n",
      "      Textile Trash       0.61      0.76      0.68        33\n",
      "         Vegetation       0.83      0.86      0.84        44\n",
      "\n",
      "           accuracy                           0.74       433\n",
      "          macro avg       0.75      0.74      0.74       433\n",
      "       weighted avg       0.75      0.74      0.74       433\n",
      "\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "from sklearn.metrics import classification_report, f1_score\n",
    "\n",
    "# Load the trained model\n",
    "model = load_model('waste_classification_inception_model.h5')\n",
    "\n",
    "# Define image size and batch size\n",
    "IMAGE_SIZE = (224, 224)\n",
    "\n",
    "# Define data generator\n",
    "test_datagen = ImageDataGenerator(rescale=1./255)\n",
    "\n",
    "# Define test directory\n",
    "test_dir = 'RealWaste2/test'\n",
    "\n",
    "test_generator = test_datagen.flow_from_directory(\n",
    "    test_dir,\n",
    "    target_size=IMAGE_SIZE,\n",
    "    batch_size=32,\n",
    "    class_mode='categorical',\n",
    "    shuffle=False\n",
    ")\n",
    "\n",
    "# Make predictions\n",
    "y_pred = model.predict(test_generator)\n",
    "y_pred_classes = np.argmax(y_pred, axis=1)\n",
    "\n",
    "# True labels\n",
    "y_true = test_generator.classes\n",
    "\n",
    "# Compute F1 score\n",
    "f1 = f1_score(y_true, y_pred_classes, average='weighted')\n",
    "print(f'Weighted F1 Score: {f1:.2f}')\n",
    "\n",
    "# Print classification report\n",
    "report = classification_report(y_true, y_pred_classes, target_names=test_generator.class_indices.keys())\n",
    "print(report)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "675faeb2",
   "metadata": {},
   "source": [
    "### Observations\n",
    "The InceptionV3 model, using transfer learning, provided a substantial improvement over the custom baseline and augmented CNN models. The higher accuracy and lower loss metrics indicate better performance in classifying waste categories in the RealWaste dataset. The advanced architecture of InceptionV3, combined with its pre-trained weights, enabled more efficient feature extraction and improved model performance, making it a more suitable choice for the waste classification task."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "id": "fafaf154",
   "metadata": {},
   "outputs": [],
   "source": [
    "from tensorflow.keras.callbacks import EarlyStopping, ReduceLROnPlateau, LearningRateScheduler\n",
    "\n",
    "class DenseNet121Model:\n",
    "    def __init__(self, input_shape=(224, 224, 3), num_classes=8):\n",
    "        self.input_shape = input_shape\n",
    "        self.num_classes = num_classes\n",
    "        self.model = None\n",
    "        self.history = None\n",
    "\n",
    "    def build_model(self):\n",
    "        # Load the DenseNet121 model pre-trained on ImageNet\n",
    "        base_model = DenseNet121(weights='imagenet', include_top=False, input_shape=self.input_shape)\n",
    "\n",
    "        # Freeze the base model layers\n",
    "        for layer in base_model.layers:\n",
    "            layer.trainable = False\n",
    "\n",
    "        # Add custom layers on top of the base model\n",
    "        x = base_model.output\n",
    "        x = GlobalAveragePooling2D()(x)\n",
    "        x = Dense(512, activation='relu')(x)\n",
    "        x = Dropout(0.5)(x)\n",
    "        x = Dense(256, activation='relu')(x)\n",
    "        x = Dropout(0.5)(x)\n",
    "        predictions = Dense(self.num_classes, activation='softmax')(x)  # 8 categories for classification\n",
    "\n",
    "        # Create the model\n",
    "        self.model = Model(inputs=base_model.input, outputs=predictions)\n",
    "\n",
    "        # Compile the model\n",
    "        self.model.compile(optimizer=Adam(learning_rate=0.001), loss='categorical_crossentropy', metrics=['accuracy'])\n",
    "\n",
    "    def fine_tune_model(self, fine_tune_at):\n",
    "        # Unfreeze the top layers of the model\n",
    "        for layer in self.model.layers[:fine_tune_at]:\n",
    "            layer.trainable = False\n",
    "        for layer in self.model.layers[fine_tune_at:]:\n",
    "            layer.trainable = True\n",
    "\n",
    "        # Compile the model with a lower learning rate for fine-tuning\n",
    "        self.model.compile(optimizer=Adam(learning_rate=1e-5), loss='categorical_crossentropy', metrics=['accuracy'])\n",
    "\n",
    "    def train_model(self, train_dir, val_dir, batch_size=32, epochs=20):\n",
    "        # Define image size and batch size\n",
    "        IMAGE_SIZE = self.input_shape[:2]\n",
    "\n",
    "        # Define data generators with reduced image size\n",
    "        train_datagen = ImageDataGenerator(\n",
    "            rescale=1./255,\n",
    "            rotation_range=30,\n",
    "            width_shift_range=0.2,\n",
    "            height_shift_range=0.2,\n",
    "            shear_range=0.2,\n",
    "            zoom_range=0.2,\n",
    "            horizontal_flip=True,\n",
    "            fill_mode='nearest'\n",
    "        )\n",
    "        validation_datagen = ImageDataGenerator(rescale=1./255)\n",
    "\n",
    "        train_generator = train_datagen.flow_from_directory(\n",
    "            train_dir,\n",
    "            target_size=IMAGE_SIZE,\n",
    "            batch_size=batch_size,\n",
    "            class_mode='categorical'\n",
    "        )\n",
    "        validation_generator = validation_datagen.flow_from_directory(\n",
    "            val_dir,\n",
    "            target_size=IMAGE_SIZE,\n",
    "            batch_size=batch_size,\n",
    "            class_mode='categorical'\n",
    "        )\n",
    "\n",
    "        # Define callbacks\n",
    "        early_stopping = EarlyStopping(monitor='val_loss', patience=5, restore_best_weights=True)\n",
    "        reduce_lr = ReduceLROnPlateau(monitor='val_loss', patience=3)\n",
    "        lr_scheduler = LearningRateScheduler(lambda epoch: 1e-3 * 10**(-epoch / 20))\n",
    "\n",
    "        # Train the model\n",
    "        self.history = self.model.fit(\n",
    "            train_generator,\n",
    "            epochs=epochs,\n",
    "            validation_data=validation_generator,\n",
    "            callbacks=[early_stopping, reduce_lr, lr_scheduler]\n",
    "        )\n",
    "\n",
    "    def evaluate_model(self, test_dir, batch_size=32):\n",
    "        # Define image size\n",
    "        IMAGE_SIZE = self.input_shape[:2]\n",
    "\n",
    "        # Define data generator\n",
    "        test_datagen = ImageDataGenerator(rescale=1./255)\n",
    "\n",
    "        test_generator = test_datagen.flow_from_directory(\n",
    "            test_dir,\n",
    "            target_size=IMAGE_SIZE,\n",
    "            batch_size=batch_size,\n",
    "            class_mode='categorical'\n",
    "        )\n",
    "\n",
    "        # Evaluate the model\n",
    "        test_loss, test_accuracy = self.model.evaluate(test_generator)\n",
    "        print(f'Test Accuracy: {test_accuracy:.2f}')\n",
    "        return test_loss, test_accuracy\n",
    "\n",
    "    def save_model(self, file_path):\n",
    "        self.model.save(file_path)\n",
    "        print(f\"Model saved to {file_path}\")\n",
    "\n",
    "    def plot_training_history(self):\n",
    "        if self.history:\n",
    "            plt.figure(figsize=(12, 6))\n",
    "\n",
    "            # Plot training & validation accuracy values\n",
    "            plt.subplot(1, 2, 1)\n",
    "            plt.plot(self.history.history['accuracy'])\n",
    "            plt.plot(self.history.history['val_accuracy'])\n",
    "            plt.title('Model Accuracy')\n",
    "            plt.ylabel('Accuracy')\n",
    "            plt.xlabel('Epoch')\n",
    "            plt.legend(['Train', 'Validation'], loc='upper left')\n",
    "\n",
    "            # Plot training & validation loss values\n",
    "            plt.subplot(1, 2, 2)\n",
    "            plt.plot(self.history.history['loss'])\n",
    "            plt.plot(self.history.history['val_loss'])\n",
    "            plt.title('Model Loss')\n",
    "            plt.ylabel('Loss')\n",
    "            plt.xlabel('Epoch')\n",
    "            plt.legend(['Train', 'Validation'], loc='upper left')\n",
    "\n",
    "            plt.tight_layout()\n",
    "            plt.show()\n",
    "        else:\n",
    "            print(\"No training history found. Please train the model first.\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "id": "2735da9e",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Instantiate the model\n",
    "densenet_model = DenseNet121Model()\n",
    "\n",
    "# Build the model\n",
    "densenet_model.build_model()\n",
    "\n",
    "# Define directories\n",
    "train_dir = 'RealWaste2/train'\n",
    "val_dir = 'RealWaste2/val'\n",
    "test_dir = 'RealWaste2/test'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "id": "a0d39817",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 3001 images belonging to 8 classes.\n",
      "Found 857 images belonging to 8 classes.\n",
      "Epoch 1/10\n",
      "\u001b[1m94/94\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m317s\u001b[0m 3s/step - accuracy: 0.3127 - loss: 2.0209 - val_accuracy: 0.6056 - val_loss: 1.0717 - learning_rate: 0.0010\n",
      "Epoch 2/10\n",
      "\u001b[1m94/94\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m340s\u001b[0m 4s/step - accuracy: 0.5737 - loss: 1.1610 - val_accuracy: 0.7281 - val_loss: 0.7555 - learning_rate: 8.9125e-04\n",
      "Epoch 3/10\n",
      "\u001b[1m94/94\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m324s\u001b[0m 3s/step - accuracy: 0.6476 - loss: 0.9636 - val_accuracy: 0.7585 - val_loss: 0.6742 - learning_rate: 7.9433e-04\n",
      "Epoch 4/10\n",
      "\u001b[1m94/94\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m326s\u001b[0m 3s/step - accuracy: 0.6719 - loss: 0.8944 - val_accuracy: 0.7620 - val_loss: 0.6513 - learning_rate: 7.0795e-04\n",
      "Epoch 5/10\n",
      "\u001b[1m94/94\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m333s\u001b[0m 3s/step - accuracy: 0.7218 - loss: 0.7913 - val_accuracy: 0.7433 - val_loss: 0.6833 - learning_rate: 6.3096e-04\n",
      "Epoch 6/10\n",
      "\u001b[1m94/94\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m332s\u001b[0m 3s/step - accuracy: 0.7433 - loss: 0.7284 - val_accuracy: 0.7760 - val_loss: 0.6326 - learning_rate: 5.6234e-04\n",
      "Epoch 7/10\n",
      "\u001b[1m94/94\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m336s\u001b[0m 4s/step - accuracy: 0.7612 - loss: 0.6912 - val_accuracy: 0.7713 - val_loss: 0.6406 - learning_rate: 5.0119e-04\n",
      "Epoch 8/10\n",
      "\u001b[1m94/94\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m336s\u001b[0m 4s/step - accuracy: 0.7798 - loss: 0.6390 - val_accuracy: 0.7876 - val_loss: 0.5812 - learning_rate: 4.4668e-04\n",
      "Epoch 9/10\n",
      "\u001b[1m94/94\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m331s\u001b[0m 3s/step - accuracy: 0.7840 - loss: 0.6010 - val_accuracy: 0.8051 - val_loss: 0.5657 - learning_rate: 3.9811e-04\n",
      "Epoch 10/10\n",
      "\u001b[1m94/94\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m337s\u001b[0m 4s/step - accuracy: 0.8002 - loss: 0.5685 - val_accuracy: 0.8121 - val_loss: 0.5635 - learning_rate: 3.5481e-04\n"
     ]
    }
   ],
   "source": [
    "# Train the model\n",
    "densenet_model.train_model(train_dir, val_dir, batch_size=32, epochs=10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "id": "9a851838",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 433 images belonging to 8 classes.\n",
      "\u001b[1m14/14\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m41s\u001b[0m 3s/step - accuracy: 0.7543 - loss: 0.6930\n",
      "Test Accuracy: 0.79\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "(0.6264948844909668, 0.7875288724899292)"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Evaluate the model\n",
    "densenet_model.evaluate_model(test_dir, batch_size=32)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "id": "e7bbedbb",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x600 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot the training history\n",
    "densenet_model.plot_training_history()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "id": "f53705c3",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:You are saving your model as an HDF5 file via `model.save()` or `keras.saving.save_model(model)`. This file format is considered legacy. We recommend using instead the native Keras format, e.g. `model.save('my_model.keras')` or `keras.saving.save_model(model, 'my_model.keras')`. \n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model saved to waste_classification_densenet_model.h5\n"
     ]
    }
   ],
   "source": [
    "# Save the model\n",
    "densenet_model.save_model('waste_classification_densenet_model.h5')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "id": "fb70798b",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:Compiled the loaded model, but the compiled metrics have yet to be built. `model.compile_metrics` will be empty until you train or evaluate the model.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 433 images belonging to 8 classes.\n",
      "\u001b[1m14/14\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m74s\u001b[0m 4s/step\n",
      "Weighted F1 Score: 0.78\n",
      "                     precision    recall  f1-score   support\n",
      "\n",
      "      Food Organics       0.84      0.90      0.87        42\n",
      "              Glass       0.60      0.69      0.64        42\n",
      "              Metal       0.75      0.90      0.82        79\n",
      "Miscellaneous Trash       0.72      0.46      0.56        50\n",
      "              Paper       0.86      0.96      0.91        50\n",
      "            Plastic       0.81      0.76      0.78        93\n",
      "      Textile Trash       0.87      0.79      0.83        33\n",
      "         Vegetation       0.90      0.80      0.84        44\n",
      "\n",
      "           accuracy                           0.79       433\n",
      "          macro avg       0.79      0.78      0.78       433\n",
      "       weighted avg       0.79      0.79      0.78       433\n",
      "\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import classification_report, f1_score\n",
    "from tensorflow.keras.models import load_model\n",
    "\n",
    "# Load the trained model\n",
    "model = load_model('waste_classification_densenet_model.h5')\n",
    "\n",
    "# Define image size and batch size\n",
    "IMAGE_SIZE = (224, 224)\n",
    "BATCH_SIZE = 32\n",
    "\n",
    "# Define data generator\n",
    "test_datagen = ImageDataGenerator(rescale=1./255)\n",
    "\n",
    "# Define test directory\n",
    "test_dir = 'RealWaste2/test'\n",
    "\n",
    "test_generator = test_datagen.flow_from_directory(\n",
    "    test_dir,\n",
    "    target_size=IMAGE_SIZE,\n",
    "    batch_size=BATCH_SIZE,\n",
    "    class_mode='categorical',\n",
    "    shuffle=False\n",
    ")\n",
    "\n",
    "# Make predictions\n",
    "y_pred = model.predict(test_generator)\n",
    "y_pred_classes = np.argmax(y_pred, axis=1)\n",
    "\n",
    "# True labels\n",
    "y_true = test_generator.classes\n",
    "\n",
    "# Compute F1 score\n",
    "f1 = f1_score(y_true, y_pred_classes, average='weighted')\n",
    "print(f'Weighted F1 Score: {f1:.2f}')\n",
    "\n",
    "# Print classification report\n",
    "report = classification_report(y_true, y_pred_classes, target_names=test_generator.class_indices.keys())\n",
    "print(report)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5046926c",
   "metadata": {},
   "source": [
    "### DenseNet121 Model Performance\n",
    "- **Test Accuracy**: 0.79\n",
    "- **Weighted F1 Score**: 0.78\n",
    "- **Training and Validation**: The DenseNet121 model demonstrated stable convergence and improved accuracy over the training epochs. The model's regularization techniques, such as dropout layers, contributed to its robustness and ability to generalize well to the test data.\n",
    "\n",
    "\n",
    "While both the DenseNet121 and Inception V3 models significantly outperformed the baseline custom convolutional models, the DenseNet121 model demonstrated superior performance in terms of both test accuracy and weighted F1 score. The DenseNet121 model however falls short of our target 80% accuracy threshold and exhibits class imbalance in minority classes, which leads us to try out more effecient architectures.\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "learn-env",
   "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": "3.11.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}