DeepBungaBatchNormalization.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "light-angle",
   "metadata": {},
   "outputs": [],
   "source": [
    "from tensorflow.compat.v1 import ConfigProto\n",
    "from tensorflow.compat.v1 import InteractiveSession\n",
    "\n",
    "config = ConfigProto()\n",
    "config.gpu_options.per_process_gpu_memory_fraction = 0.5\n",
    "config.gpu_options.allow_growth = True\n",
    "session = InteractiveSession(config=config)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "decent-portland",
   "metadata": {},
   "outputs": [],
   "source": [
    "# import the libraries as shown below\n",
    "\n",
    "from tensorflow.keras.layers import Input, Lambda, Dense, Flatten\n",
    "from tensorflow.keras.models import Model\n",
    "from tensorflow.keras.preprocessing import image\n",
    "from tensorflow.keras.preprocessing.image import ImageDataGenerator,load_img\n",
    "from tensorflow.keras.models import Sequential\n",
    "from keras.optimizers import SGD, RMSprop, Adam\n",
    "import numpy as np\n",
    "from glob import glob\n",
    "import matplotlib.pyplot as plt\n",
    "import tensorflow as tf"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "banned-thumbnail",
   "metadata": {},
   "outputs": [],
   "source": [
    "# re-size all the images to this\n",
    "IMAGE_SIZE = [224, 224]\n",
    "\n",
    "train_path = 'flowers Dataset/train'\n",
    "valid_path = 'flowers Dataset/val'\n",
    "test_path = 'flowers Dataset/test'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "strategic-vitamin",
   "metadata": {},
   "outputs": [],
   "source": [
    "# useful for getting number of output classes\n",
    "folders = glob('flowers Dataset/train/*')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "sticky-huntington",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['flowers Dataset/train\\\\daisy',\n",
       " 'flowers Dataset/train\\\\dandelion',\n",
       " 'flowers Dataset/train\\\\rose',\n",
       " 'flowers Dataset/train\\\\sunflower',\n",
       " 'flowers Dataset/train\\\\tulip']"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "folders"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "thirty-arbor",
   "metadata": {},
   "outputs": [],
   "source": [
    "from tensorflow.keras.layers import BatchNormalization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "cordless-trial",
   "metadata": {},
   "outputs": [],
   "source": [
    "model = tf.keras.models.Sequential([tf.keras.layers.Conv2D(32,(3,3),activation = 'relu',input_shape = (224,224,3)),\n",
    "                                    tf.keras.layers.MaxPool2D(2,2),\n",
    "                                    tf.keras.layers.BatchNormalization(),\n",
    "                                    #\n",
    "                                    tf.keras.layers.Conv2D(64,(3,3),activation = 'relu'),\n",
    "                                    tf.keras.layers.MaxPool2D(2,2),\n",
    "                                    tf.keras.layers.BatchNormalization(),\n",
    "                                    #\n",
    "                                    tf.keras.layers.Conv2D(64,(3,3),activation = 'relu'),\n",
    "                                    tf.keras.layers.MaxPool2D(2,2),\n",
    "                                    tf.keras.layers.BatchNormalization(),\n",
    "                                    ##\n",
    "                                    tf.keras.layers.Flatten(),\n",
    "                                    ##\n",
    "                                    tf.keras.layers.Dense(128,activation = 'relu'),\n",
    "                                    tf.keras.layers.BatchNormalization(),\n",
    "                                    ##\n",
    "                                    tf.keras.layers.Dense(5,activation = 'softmax'),\n",
    "                                    ])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "ignored-reach",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "conv2d (Conv2D)              (None, 222, 222, 32)      896       \n",
      "_________________________________________________________________\n",
      "max_pooling2d (MaxPooling2D) (None, 111, 111, 32)      0         \n",
      "_________________________________________________________________\n",
      "batch_normalization (BatchNo (None, 111, 111, 32)      128       \n",
      "_________________________________________________________________\n",
      "conv2d_1 (Conv2D)            (None, 109, 109, 64)      18496     \n",
      "_________________________________________________________________\n",
      "max_pooling2d_1 (MaxPooling2 (None, 54, 54, 64)        0         \n",
      "_________________________________________________________________\n",
      "batch_normalization_1 (Batch (None, 54, 54, 64)        256       \n",
      "_________________________________________________________________\n",
      "conv2d_2 (Conv2D)            (None, 52, 52, 64)        36928     \n",
      "_________________________________________________________________\n",
      "max_pooling2d_2 (MaxPooling2 (None, 26, 26, 64)        0         \n",
      "_________________________________________________________________\n",
      "batch_normalization_2 (Batch (None, 26, 26, 64)        256       \n",
      "_________________________________________________________________\n",
      "flatten (Flatten)            (None, 43264)             0         \n",
      "_________________________________________________________________\n",
      "dense (Dense)                (None, 128)               5537920   \n",
      "_________________________________________________________________\n",
      "batch_normalization_3 (Batch (None, 128)               512       \n",
      "_________________________________________________________________\n",
      "dense_1 (Dense)              (None, 5)                 645       \n",
      "=================================================================\n",
      "Total params: 5,596,037\n",
      "Trainable params: 5,595,461\n",
      "Non-trainable params: 576\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "guilty-blood",
   "metadata": {},
   "outputs": [],
   "source": [
    "# tell the model what cost and optimization method to use\n",
    "model.compile(\n",
    "  loss='categorical_crossentropy',\n",
    "  optimizer='adam',\n",
    "  metrics=['accuracy']\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "proper-portsmouth",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Use the Image Data Generator to import the images from the dataset\n",
    "from tensorflow.keras.preprocessing.image import ImageDataGenerator\n",
    "\n",
    "train_datagen = ImageDataGenerator(rescale = 1./255,\n",
    "                                   shear_range = 0.2,\n",
    "                                   zoom_range = 0.2,\n",
    "                                   horizontal_flip = True,\n",
    "                                   ) \n",
    "test_datagen = ImageDataGenerator(rescale = 1./255)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "ignored-engagement",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 2593 images belonging to 5 classes.\n"
     ]
    }
   ],
   "source": [
    "train_generator = train_datagen.flow_from_directory('flowers Dataset/train',\n",
    "                                                 target_size = (224, 224),\n",
    "                                                 batch_size = 32,\n",
    "                                                 class_mode = 'categorical',\n",
    "                                                 ) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "compact-subject",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 860 images belonging to 5 classes.\n"
     ]
    }
   ],
   "source": [
    "valid_generator = train_datagen.flow_from_directory('flowers Dataset/val',\n",
    "                                            target_size = (224, 224),\n",
    "                                            batch_size = 32,\n",
    "                                            class_mode = 'categorical',\n",
    "                                            ) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "cooperative-representation",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 870 images belonging to 5 classes.\n"
     ]
    }
   ],
   "source": [
    "test_generator = test_datagen.flow_from_directory('flowers Dataset/test',\n",
    "                                            target_size = (224, 224),\n",
    "                                            batch_size = 1,\n",
    "                                            class_mode = 'categorical',\n",
    "                                            shuffle = False,\n",
    "                                            )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "hidden-morris",
   "metadata": {},
   "outputs": [],
   "source": [
    "from tensorflow.keras.callbacks import EarlyStopping"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "protected-anniversary",
   "metadata": {},
   "outputs": [],
   "source": [
    "earlystopper=tf.keras.callbacks.EarlyStopping(\n",
    "    monitor='val_loss', min_delta=0, patience=5, verbose=1,\n",
    "    mode='auto', baseline=None, restore_best_weights=False\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "optional-equivalent",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\DELL\\anaconda3\\envs\\marzharenv\\lib\\site-packages\\tensorflow\\python\\keras\\engine\\training.py:1940: UserWarning: `Model.fit_generator` is deprecated and will be removed in a future version. Please use `Model.fit`, which supports generators.\n",
      "  warnings.warn('`Model.fit_generator` is deprecated and '\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/20\n",
      "82/82 [==============================] - 71s 689ms/step - loss: 1.3778 - accuracy: 0.5199 - val_loss: 2.8271 - val_accuracy: 0.2430\n",
      "Epoch 2/20\n",
      "82/82 [==============================] - 33s 399ms/step - loss: 0.9846 - accuracy: 0.6101 - val_loss: 3.0805 - val_accuracy: 0.2430\n",
      "Epoch 3/20\n",
      "82/82 [==============================] - 32s 395ms/step - loss: 0.8634 - accuracy: 0.6683 - val_loss: 2.1656 - val_accuracy: 0.2442\n",
      "Epoch 4/20\n",
      "82/82 [==============================] - 32s 394ms/step - loss: 0.8183 - accuracy: 0.6838 - val_loss: 2.2734 - val_accuracy: 0.2721\n",
      "Epoch 5/20\n",
      "82/82 [==============================] - 32s 393ms/step - loss: 0.7377 - accuracy: 0.7200 - val_loss: 1.5290 - val_accuracy: 0.3570\n",
      "Epoch 6/20\n",
      "82/82 [==============================] - 32s 396ms/step - loss: 0.6560 - accuracy: 0.7648 - val_loss: 1.3906 - val_accuracy: 0.4733\n",
      "Epoch 7/20\n",
      "82/82 [==============================] - 33s 398ms/step - loss: 0.6177 - accuracy: 0.7713 - val_loss: 1.0921 - val_accuracy: 0.5919\n",
      "Epoch 8/20\n",
      "82/82 [==============================] - 33s 398ms/step - loss: 0.5702 - accuracy: 0.7883 - val_loss: 0.9460 - val_accuracy: 0.6651\n",
      "Epoch 9/20\n",
      "82/82 [==============================] - 33s 409ms/step - loss: 0.5066 - accuracy: 0.8118 - val_loss: 0.8591 - val_accuracy: 0.6837\n",
      "Epoch 10/20\n",
      "82/82 [==============================] - 33s 398ms/step - loss: 0.4745 - accuracy: 0.8307 - val_loss: 1.0248 - val_accuracy: 0.6349\n",
      "Epoch 11/20\n",
      "82/82 [==============================] - 34s 410ms/step - loss: 0.4179 - accuracy: 0.8538 - val_loss: 0.9487 - val_accuracy: 0.6802\n",
      "Epoch 12/20\n",
      "82/82 [==============================] - 35s 425ms/step - loss: 0.3806 - accuracy: 0.8696 - val_loss: 1.0103 - val_accuracy: 0.6674\n",
      "Epoch 13/20\n",
      "82/82 [==============================] - 35s 423ms/step - loss: 0.3701 - accuracy: 0.8739 - val_loss: 1.3866 - val_accuracy: 0.5581\n",
      "Epoch 14/20\n",
      "35/82 [===========>..................] - ETA: 14s - loss: 0.3244 - accuracy: 0.8852"
     ]
    }
   ],
   "source": [
    "# fit the model\n",
    "# Run the cell. It will take some time to execute\n",
    "r = model.fit_generator(\n",
    "  train_generator,\n",
    "  validation_data=valid_generator,\n",
    "  epochs=20,\n",
    "  steps_per_epoch=len(train_generator),\n",
    "  validation_steps=len(valid_generator),\n",
    "  callbacks = [earlystopper]\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "hawaiian-highway",
   "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 the loss\n",
    "plt.plot(r.history['loss'], label='train loss')\n",
    "plt.plot(r.history['val_loss'], label='val loss')\n",
    "plt.ylabel('Loss')\n",
    "plt.xlabel('Epoch')\n",
    "plt.legend()\n",
    "plt.show()\n",
    "\n",
    "# plot the accuracy\n",
    "plt.plot(r.history['accuracy'], label='train acc')\n",
    "plt.plot(r.history['val_accuracy'], label='val acc')\n",
    "plt.ylabel('Accuracy')\n",
    "plt.xlabel('Epoch')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "crude-creek",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "870/870 - 9s - loss: 1.0313 - accuracy: 0.6874\n",
      "\n",
      "Test accuracy: 0.6873562932014465\n",
      "\n",
      "Test loss: 1.0312526226043701\n"
     ]
    }
   ],
   "source": [
    "test_loss, test_acc = model.evaluate(test_generator, verbose=2)\n",
    "print('\\nTest accuracy:', test_acc)\n",
    "print('\\nTest loss:', test_loss)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "returning-cabin",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import classification_report, confusion_matrix, roc_curve, auc, roc_auc_score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "optional-divide",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import LabelBinarizer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "expressed-weight",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "870/870 - 2s\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.72      0.52      0.60       155\n",
      "           1       0.69      0.83      0.75       211\n",
      "           2       0.72      0.55      0.63       158\n",
      "           3       0.64      0.90      0.75       148\n",
      "           4       0.70      0.61      0.65       198\n",
      "\n",
      "    accuracy                           0.69       870\n",
      "   macro avg       0.69      0.68      0.68       870\n",
      "weighted avg       0.69      0.69      0.68       870\n",
      "\n"
     ]
    }
   ],
   "source": [
    "test_generator.reset()\n",
    "# get the ground truth of your data. \n",
    "y_true = test_generator.classes \n",
    "\n",
    "# predict the probability distribution of the data\n",
    "predictions = model.predict(test_generator, verbose=2)\n",
    "\n",
    "# get the class with highest probability for each sample\n",
    "y_pred = np.argmax(predictions, axis=1)\n",
    "\n",
    "# get the classification report\n",
    "print(classification_report(y_true, y_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "announced-welsh",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[ 81  38   6  22   8]\n",
      " [ 11 176   2  20   2]\n",
      " [ 10  10  87  12  39]\n",
      " [  1  12   0 133   2]\n",
      " [ 10  20  25  22 121]]\n"
     ]
    }
   ],
   "source": [
    "print(confusion_matrix(y_true, y_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "hungarian-george",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['daisy', 'dandelion', 'rose', 'sunflower', 'tulip']"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "list(train_generator.class_indices.keys())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "special-junior",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ROC AUC score: 0.8023440910034466\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "all_labels = ['daisy', 'dandelion', 'rose', 'sunflower', 'tulip']\n",
    "# set plot figure size\n",
    "fig, c_ax = plt.subplots(1,1, figsize = (12, 8))\n",
    "\n",
    "def multiclass_roc_auc_score(y_true, y_pred, average=\"macro\"):\n",
    "    lb = LabelBinarizer()\n",
    "    lb.fit(y_true)\n",
    "    y_true = lb.transform(y_true)\n",
    "    y_pred = lb.transform(y_pred)\n",
    "\n",
    "    for (idx, c_label) in enumerate(all_labels):\n",
    "        fpr, tpr, thresholds = roc_curve(y_true[:,idx].astype(int), y_pred[:,idx])\n",
    "        c_ax.plot(fpr, tpr, label = '%s (AUC:%0.2f)'  % (c_label, auc(fpr, tpr)))\n",
    "    c_ax.plot(fpr, fpr, 'b-', label = 'Random Guessing')\n",
    "    return roc_auc_score(y_true, y_pred, average=average)\n",
    "\n",
    "\n",
    "print('ROC AUC score:', multiclass_roc_auc_score(y_true, y_pred))\n",
    "\n",
    "c_ax.legend()\n",
    "c_ax.set_xlabel('False Positive Rate')\n",
    "c_ax.set_ylabel('True Positive Rate')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "loaded-conspiracy",
   "metadata": {},
   "outputs": [],
   "source": [
    "# save it as a h5 file\n",
    "\n",
    "from tensorflow.keras.models import load_model\n",
    "\n",
    "model.save('Bunga weight/Deepbatchnorm_bunga.h5')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "assumed-hammer",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "test accuracy : 68.73563218390805\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import seaborn as sn\n",
    "import tensorflow as tf\n",
    "\n",
    "model = tf.keras.models.load_model('Bunga weight/Deepbatchnorm_bunga.h5')\n",
    "filenames = test_generator.filenames\n",
    "nb_samples = len(test_generator)\n",
    "y_prob=[]\n",
    "y_act=[]\n",
    "test_generator.reset()\n",
    "for _ in range(nb_samples):\n",
    "   X_test,Y_test = test_generator.next()\n",
    "   y_prob.append(model.predict(X_test))\n",
    "   y_act.append(Y_test)\n",
    "\n",
    "predicted_class = [list(train_generator.class_indices.keys())[i.argmax()] for i in y_prob]\n",
    "actual_class = [list(train_generator.class_indices.keys())[i.argmax()] for i in y_act]\n",
    "\n",
    "out_df = pd.DataFrame(np.vstack([predicted_class,actual_class]).T,columns=['predicted_class','actual_class'])\n",
    "confusion_matrix = pd.crosstab(out_df['actual_class'],out_df['predicted_class'], rownames=['Actual'], colnames=['Predicted'])\n",
    "\n",
    "sn.heatmap(confusion_matrix,cmap='Blues', annot=True,fmt='d')\n",
    "plt.show\n",
    "print('test accuracy : {}'.format((np.diagonal(confusion_matrix).sum()/confusion_matrix.sum().sum()*100)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "enabling-minutes",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python [conda env:marzharenv] *",
   "language": "python",
   "name": "conda-env-marzharenv-py"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}