84-78 nasnet.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "1ae1d4f5",
   "metadata": {},
   "outputs": [],
   "source": [
    "import tensorflow as tf\n",
    "from tensorflow.keras.preprocessing.image import ImageDataGenerator\n",
    "import numpy as np\n",
    "from sklearn.metrics import classification_report, confusion_matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "8a0b45e7",
   "metadata": {},
   "outputs": [],
   "source": [
    "data_dir='D:\\\\Msc GU\\\\Msc 4th Sem\\\\Project\\\\DataSet\\\\teadataset'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "b9b5e329",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 711 images belonging to 8 classes.\n",
      "Found 174 images belonging to 8 classes.\n"
     ]
    }
   ],
   "source": [
    "# Set up data augmentation\n",
    "datagen = ImageDataGenerator(\n",
    "    rescale=1./255,\n",
    "    rotation_range=20,\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",
    "    validation_split=0.2)\n",
    "\n",
    "# Load in the dataset\n",
    "train_data = datagen.flow_from_directory(\n",
    "    data_dir,\n",
    "    target_size=(224, 224),\n",
    "    batch_size=32,\n",
    "    class_mode='categorical',\n",
    "    subset='training')\n",
    "\n",
    "val_data = datagen.flow_from_directory(\n",
    "    data_dir,\n",
    "    target_size=(224, 224),\n",
    "    batch_size=32,\n",
    "    class_mode='categorical',\n",
    "    subset='validation')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "1663fc9c",
   "metadata": {},
   "outputs": [],
   "source": [
    "batch_size=32"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "e89fa29d",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:`lr` is deprecated, please use `learning_rate` instead, or use the legacy optimizer, e.g.,tf.keras.optimizers.legacy.Adam.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/50\n",
      "22/22 [==============================] - 234s 9s/step - loss: 1.8455 - accuracy: 0.2916 - val_loss: 1.2862 - val_accuracy: 0.5437\n",
      "Epoch 2/50\n",
      "22/22 [==============================] - 184s 9s/step - loss: 1.1826 - accuracy: 0.5611 - val_loss: 1.1004 - val_accuracy: 0.5938\n",
      "Epoch 3/50\n",
      "22/22 [==============================] - 126s 6s/step - loss: 1.0331 - accuracy: 0.5994 - val_loss: 0.9507 - val_accuracy: 0.6562\n",
      "Epoch 4/50\n",
      "22/22 [==============================] - 91s 4s/step - loss: 0.8908 - accuracy: 0.6524 - val_loss: 0.9061 - val_accuracy: 0.6187\n",
      "Epoch 5/50\n",
      "22/22 [==============================] - 92s 4s/step - loss: 0.7932 - accuracy: 0.6878 - val_loss: 0.9026 - val_accuracy: 0.6062\n",
      "Epoch 6/50\n",
      "22/22 [==============================] - 132s 6s/step - loss: 0.7569 - accuracy: 0.7261 - val_loss: 0.8816 - val_accuracy: 0.6187\n",
      "Epoch 7/50\n",
      "22/22 [==============================] - 180s 8s/step - loss: 0.6843 - accuracy: 0.7349 - val_loss: 0.9354 - val_accuracy: 0.6500\n",
      "Epoch 8/50\n",
      "22/22 [==============================] - 163s 8s/step - loss: 0.7327 - accuracy: 0.7054 - val_loss: 0.9399 - val_accuracy: 0.6062\n",
      "Epoch 9/50\n",
      "22/22 [==============================] - 92s 4s/step - loss: 0.6486 - accuracy: 0.7526 - val_loss: 0.8735 - val_accuracy: 0.6500\n",
      "Epoch 10/50\n",
      "22/22 [==============================] - 97s 4s/step - loss: 0.6391 - accuracy: 0.7658 - val_loss: 0.9640 - val_accuracy: 0.6125\n",
      "Epoch 11/50\n",
      "22/22 [==============================] - 100s 5s/step - loss: 0.5920 - accuracy: 0.7658 - val_loss: 0.9090 - val_accuracy: 0.6250\n",
      "Epoch 12/50\n",
      "22/22 [==============================] - 99s 5s/step - loss: 0.6385 - accuracy: 0.7511 - val_loss: 0.9670 - val_accuracy: 0.6187\n",
      "Epoch 13/50\n",
      "22/22 [==============================] - 104s 5s/step - loss: 0.6179 - accuracy: 0.7658 - val_loss: 0.9546 - val_accuracy: 0.6187\n",
      "Epoch 14/50\n",
      "22/22 [==============================] - 123s 6s/step - loss: 0.6329 - accuracy: 0.7673 - val_loss: 0.8424 - val_accuracy: 0.6938\n",
      "Epoch 15/50\n",
      "22/22 [==============================] - 99s 4s/step - loss: 0.5705 - accuracy: 0.7732 - val_loss: 0.8413 - val_accuracy: 0.6625\n",
      "Epoch 16/50\n",
      "22/22 [==============================] - 98s 4s/step - loss: 0.6059 - accuracy: 0.7555 - val_loss: 0.7770 - val_accuracy: 0.6812\n",
      "Epoch 17/50\n",
      "22/22 [==============================] - 100s 5s/step - loss: 0.5456 - accuracy: 0.7912 - val_loss: 0.7485 - val_accuracy: 0.7188\n",
      "Epoch 18/50\n",
      "22/22 [==============================] - 98s 4s/step - loss: 0.4922 - accuracy: 0.8130 - val_loss: 0.7944 - val_accuracy: 0.6687\n",
      "Epoch 19/50\n",
      "22/22 [==============================] - 104s 5s/step - loss: 0.5207 - accuracy: 0.8144 - val_loss: 0.9114 - val_accuracy: 0.6562\n",
      "Epoch 20/50\n",
      "22/22 [==============================] - 111s 5s/step - loss: 0.5029 - accuracy: 0.8189 - val_loss: 0.8040 - val_accuracy: 0.7000\n",
      "Epoch 21/50\n",
      "22/22 [==============================] - 105s 5s/step - loss: 0.5104 - accuracy: 0.8041 - val_loss: 0.8215 - val_accuracy: 0.6625\n",
      "Epoch 22/50\n",
      "22/22 [==============================] - 116s 5s/step - loss: 0.4523 - accuracy: 0.8351 - val_loss: 0.8037 - val_accuracy: 0.6875\n",
      "Epoch 23/50\n",
      "22/22 [==============================] - 110s 5s/step - loss: 0.4400 - accuracy: 0.8306 - val_loss: 0.8080 - val_accuracy: 0.6625\n",
      "Epoch 24/50\n",
      "22/22 [==============================] - 100s 5s/step - loss: 0.4939 - accuracy: 0.7894 - val_loss: 0.8177 - val_accuracy: 0.6750\n",
      "Epoch 25/50\n",
      "22/22 [==============================] - 99s 4s/step - loss: 0.4717 - accuracy: 0.8041 - val_loss: 0.6994 - val_accuracy: 0.7000\n",
      "Epoch 26/50\n",
      "22/22 [==============================] - 104s 5s/step - loss: 0.4652 - accuracy: 0.7894 - val_loss: 0.8584 - val_accuracy: 0.6625\n",
      "Epoch 27/50\n",
      "22/22 [==============================] - 100s 5s/step - loss: 0.4959 - accuracy: 0.8056 - val_loss: 0.8357 - val_accuracy: 0.6938\n",
      "Epoch 28/50\n",
      "22/22 [==============================] - 103s 5s/step - loss: 0.4235 - accuracy: 0.8324 - val_loss: 0.7990 - val_accuracy: 0.6625\n",
      "Epoch 29/50\n",
      "22/22 [==============================] - 100s 5s/step - loss: 0.4503 - accuracy: 0.8218 - val_loss: 0.7280 - val_accuracy: 0.6938\n",
      "Epoch 30/50\n",
      "22/22 [==============================] - 100s 5s/step - loss: 0.4337 - accuracy: 0.8306 - val_loss: 0.8453 - val_accuracy: 0.6500\n",
      "Epoch 31/50\n",
      "22/22 [==============================] - 101s 5s/step - loss: 0.4098 - accuracy: 0.8439 - val_loss: 1.0157 - val_accuracy: 0.6313\n",
      "Epoch 32/50\n",
      "22/22 [==============================] - 99s 5s/step - loss: 0.4294 - accuracy: 0.8203 - val_loss: 0.7282 - val_accuracy: 0.7063\n",
      "Epoch 33/50\n",
      "22/22 [==============================] - 100s 5s/step - loss: 0.4636 - accuracy: 0.8100 - val_loss: 0.8691 - val_accuracy: 0.6250\n",
      "Epoch 34/50\n",
      "22/22 [==============================] - 100s 5s/step - loss: 0.4170 - accuracy: 0.8351 - val_loss: 0.7807 - val_accuracy: 0.7063\n",
      "Epoch 35/50\n",
      "22/22 [==============================] - 102s 5s/step - loss: 0.4097 - accuracy: 0.8218 - val_loss: 0.9097 - val_accuracy: 0.6125\n",
      "Epoch 36/50\n",
      "22/22 [==============================] - 102s 5s/step - loss: 0.4224 - accuracy: 0.8277 - val_loss: 0.7185 - val_accuracy: 0.7125\n",
      "Epoch 37/50\n",
      "22/22 [==============================] - 98s 4s/step - loss: 0.3849 - accuracy: 0.8483 - val_loss: 0.8861 - val_accuracy: 0.6500\n",
      "Epoch 38/50\n",
      "22/22 [==============================] - 104s 5s/step - loss: 0.3433 - accuracy: 0.8736 - val_loss: 0.9116 - val_accuracy: 0.6562\n",
      "Epoch 39/50\n",
      "22/22 [==============================] - 104s 5s/step - loss: 0.4566 - accuracy: 0.8196 - val_loss: 0.8026 - val_accuracy: 0.6812\n",
      "Epoch 40/50\n",
      "22/22 [==============================] - 102s 5s/step - loss: 0.4408 - accuracy: 0.8380 - val_loss: 0.7644 - val_accuracy: 0.6750\n",
      "Epoch 41/50\n",
      "22/22 [==============================] - 104s 5s/step - loss: 0.4088 - accuracy: 0.8483 - val_loss: 0.8017 - val_accuracy: 0.7125\n",
      "Epoch 42/50\n",
      "22/22 [==============================] - 105s 5s/step - loss: 0.3639 - accuracy: 0.8689 - val_loss: 0.6729 - val_accuracy: 0.7500\n",
      "Epoch 43/50\n",
      "22/22 [==============================] - 101s 5s/step - loss: 0.4090 - accuracy: 0.8247 - val_loss: 0.9509 - val_accuracy: 0.6438\n",
      "Epoch 44/50\n",
      "22/22 [==============================] - 99s 5s/step - loss: 0.3998 - accuracy: 0.8409 - val_loss: 0.8845 - val_accuracy: 0.6875\n",
      "Epoch 45/50\n",
      "22/22 [==============================] - 99s 5s/step - loss: 0.4040 - accuracy: 0.8454 - val_loss: 0.8018 - val_accuracy: 0.6875\n",
      "Epoch 46/50\n",
      "22/22 [==============================] - 101s 5s/step - loss: 0.3285 - accuracy: 0.8792 - val_loss: 0.8830 - val_accuracy: 0.6625\n",
      "Epoch 47/50\n",
      "22/22 [==============================] - 99s 5s/step - loss: 0.3412 - accuracy: 0.8733 - val_loss: 0.7942 - val_accuracy: 0.6938\n",
      "Epoch 48/50\n",
      "22/22 [==============================] - 103s 5s/step - loss: 0.3474 - accuracy: 0.8679 - val_loss: 0.8923 - val_accuracy: 0.6625\n",
      "Epoch 49/50\n",
      "22/22 [==============================] - 100s 5s/step - loss: 0.3538 - accuracy: 0.8616 - val_loss: 0.8879 - val_accuracy: 0.6750\n",
      "Epoch 50/50\n",
      "22/22 [==============================] - 98s 4s/step - loss: 0.3564 - accuracy: 0.8733 - val_loss: 0.8093 - val_accuracy: 0.7000\n"
     ]
    }
   ],
   "source": [
    "base_model = tf.keras.applications.NASNetMobile(\n",
    "    input_shape=(224, 224, 3),\n",
    "    include_top=False)\n",
    "\n",
    "# Freeze all but the last few layers of the base model\n",
    "for layer in base_model.layers[:-10]:\n",
    "    layer.trainable = False\n",
    "\n",
    "# Add some layers on top of the pre-trained model\n",
    "x = base_model.output\n",
    "x = tf.keras.layers.GlobalAveragePooling2D()(x)\n",
    "x = tf.keras.layers.Dense(512, activation='relu')(x)\n",
    "x = tf.keras.layers.Dropout(0.5)(x)\n",
    "x = tf.keras.layers.Dense(8, activation='softmax')(x)\n",
    "\n",
    "# Create the final model\n",
    "model = tf.keras.models.Model(inputs=base_model.input, outputs=x)\n",
    "\n",
    "# Compile the model\n",
    "model.compile(optimizer=tf.keras.optimizers.Adam(lr=1e-4),\n",
    "              loss='categorical_crossentropy',\n",
    "              metrics=['accuracy'])\n",
    "\n",
    "# Train the model with data augmentation and early stopping\n",
    "history = model.fit(\n",
    "    train_data,\n",
    "    steps_per_epoch=train_data.n// batch_size,\n",
    "    epochs=50,\n",
    "    validation_steps=val_data.n// batch_size,\n",
    "    validation_data=val_data)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "d92ef2e5",
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "\n",
    "def plot_loss(history):\n",
    "    # Plot training and validation loss\n",
    "    plt.plot(history.history['loss'])\n",
    "    plt.plot(history.history['val_loss'])\n",
    "    plt.title('Model Loss')\n",
    "    plt.ylabel('Loss')\n",
    "    plt.xlabel('Epoch')\n",
    "    plt.legend(['Train', 'Test'], loc='upper right')\n",
    "    plt.show()\n",
    "\n",
    "def plot_accuracy(history):\n",
    "    # Plot training and validation accuracy\n",
    "    plt.plot(history.history['accuracy'])\n",
    "    plt.plot(history.history['val_accuracy'])\n",
    "    plt.title('Model Accuracy')\n",
    "    plt.ylabel('Accuracy')\n",
    "    plt.xlabel('Epoch')\n",
    "    plt.legend(['Train', 'Test'], loc='lower right')\n",
    "    plt.show()\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "4c8b7069",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 528 images belonging to 8 classes.\n"
     ]
    }
   ],
   "source": [
    "datagen = ImageDataGenerator(\n",
    "    rescale=1./255,\n",
    "    rotation_range=20,\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",
    "    validation_split=0.6)\n",
    "\n",
    "test_data = datagen.flow_from_directory(\n",
    "    data_dir,\n",
    "    target_size=(224, 224),\n",
    "    batch_size= 32,\n",
    "    shuffle=True,\n",
    "    class_mode='categorical',\n",
    "    subset='validation')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "e26f4ad2",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "17/17 [==============================] - 58s 3s/step - loss: 0.4465 - accuracy: 0.8333\n",
      "Test Loss: 0.44654977321624756\n",
      "Test Accuracy: 0.8333333134651184\n"
     ]
    }
   ],
   "source": [
    "# Evaluate the model on the test set\n",
    "loss, accuracy = model.evaluate(test_data)\n",
    "\n",
    "# Print the results\n",
    "print('Test Loss:', loss)\n",
    "print('Test Accuracy:', accuracy)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "84a2be23",
   "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"
    },
    {
     "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": [
    "plot_loss(history)\n",
    "plot_accuracy(history)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "436a2387",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}