1023 lines (1022 with data), 255.6 kB

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    }
   ],
   "source": [
    "import os\n",
    "import pickle\n",
    "import random\n",
    "import glob\n",
    "import datetime\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import cv2\n",
    "import pydicom\n",
    "from tqdm import tqdm\n",
    "from joblib import delayed, Parallel\n",
    "import zipfile\n",
    "from pydicom.filebase import DicomBytesIO\n",
    "import sys\n",
    "from PIL import Image\n",
    "import cv2\n",
    "#from focal_loss import sparse_categorical_focal_loss\n",
    "import keras\n",
    "#import tensorflow_addons as tfa\n",
    "from tensorflow.keras.preprocessing.image import ImageDataGenerator\n",
    "from keras.models import model_from_json\n",
    "import tensorflow as tf\n",
    "import keras\n",
    "from keras.models import Sequential, Model\n",
    "from keras.layers import Dense, Conv2D, Flatten, MaxPooling2D, GlobalAveragePooling2D, Dropout\n",
    "from keras.applications.inception_v3 import InceptionV3\n",
    "\n",
    "# importing pyplot and image from matplotlib \n",
    "import matplotlib.pyplot as plt \n",
    "import matplotlib.image as mpimg \n",
    "from keras.optimizers import SGD\n",
    "from keras import backend\n",
    "from keras.models import load_model\n",
    "\n",
    "from keras.preprocessing import image\n",
    "import albumentations as A\n",
    "\n",
    "\n",
    "from sklearn.metrics import classification_report, confusion_matrix, roc_auc_score, roc_curve\n",
    "import tensorflow as tf"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "from tensorflow.keras.losses import Reduction\n",
    "\n",
    "from tensorflow_addons.losses import SigmoidFocalCrossEntropy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "base_url = '/home/ubuntu/kaggle/rsna-intracranial-hemorrhage-detection/'\n",
    "TRAIN_DIR = '/home/ubuntu/kaggle/rsna-intracranial-hemorrhage-detection/stage_2_train/'\n",
    "TEST_DIR = '/home/ubuntu/kaggle/rsna-intracranial-hemorrhage-detection/stage_2_test/'\n",
    "image_dir = '/home/ubuntu/kaggle/rsna-intracranial-hemorrhage-detection/png/train/adjacent-brain-cropped/'\n",
    "save_dir = 'home/ubuntu/kaggle/models/'\n",
    "os.listdir(base_url)\n",
    "\n",
    "def png(image): \n",
    "    return image + '.png'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# learning rate"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "initial_learning_rate = 1e-2\n",
    "first_decay_steps = 1000\n",
    "lr_decayed_fn = (\n",
    "  tf.keras.experimental.CosineDecayRestarts(\n",
    "      initial_learning_rate,\n",
    "      first_decay_steps))\n",
    "opt = tf.keras.optimizers.SGD(learning_rate=lr_decayed_fn, nesterov=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# generator"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_idg = ImageDataGenerator(\n",
    "        featurewise_center=False,  # set input mean to 0 over the dataset\n",
    "        samplewise_center=False,  # set each sample mean to 0\n",
    "        featurewise_std_normalization=False,  # divide inputs by std of the dataset\n",
    "        samplewise_std_normalization=False,  # divide each input by its std\n",
    "        zca_whitening=False,  # apply ZCA whitening\n",
    "        shear_range=0.05,\n",
    "        rotation_range=50, # randomly rotate images in the range (degrees, 0 to 180)\n",
    "        width_shift_range=0.1,  # randomly shift images horizontally (fraction of total width)\n",
    "        height_shift_range=0.1,  # randomly shift images vertically (fraction of total height)\n",
    "        horizontal_flip=True,\n",
    "        rescale=1./255)\n",
    "valid_idg = ImageDataGenerator(rescale=1./255)\n",
    "training_data = pd.read_csv(f'train_0.csv') \n",
    "training_data['Image'] = training_data['Image'].apply(png)\n",
    "\n",
    "validation_data = pd.read_csv(f'valid_0.csv')\n",
    "validation_data['Image'] = validation_data['Image'].apply(png)\n",
    "\n",
    "columns=['any','epidural','intraparenchymal','intraventricular', 'subarachnoid','subdural']\n",
    "\n",
    "#train_data_generator = train_idg.flow_from_dataframe(training_data, directory = image_dir,\n",
    "#                           x_col = \"Image\", y_col = columns,batch_size=64,\n",
    "#                           class_mode=\"raw\", target_size=(224,224), shuffle = True)\n",
    "#valid_data_generator  = valid_idg.flow_from_dataframe(validation_data, directory = image_dir,\n",
    "#                        x_col = \"Image\", y_col = columns,batch_size=64,\n",
    "#                        class_mode = \"raw\",target_size=(224,224), shuffle = False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# undersampling"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "def undersample(dataframe,steps,batch_size):\n",
    "        part = np.int(steps/3 * batch_size)\n",
    "        zero_ids = np.random.choice(dataframe.loc[dataframe[\"any\"] == 0].index.values, size=2*part, replace=False)\n",
    "        hot_ids = np.random.choice(dataframe.loc[dataframe[\"any\"] == 1].index.values, size=1*part, replace=False)\n",
    "        data_ids = list(set(zero_ids).union(hot_ids))\n",
    "        np.random.shuffle(data_ids)\n",
    "        return data_ids\n",
    "        "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "257598\n"
     ]
    }
   ],
   "source": [
    "train_indices = undersample(training_data, 8050,32)\n",
    "print(len(train_indices))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "64320\n"
     ]
    }
   ],
   "source": [
    "valid_indices = undersample(validation_data, 2010,32)\n",
    "print(len(valid_indices))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "l = training_data[training_data.index.isin(train_indices)]\n",
    "m = validation_data[validation_data.index.isin(valid_indices)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "z = l[['any','epidural','intraparenchymal','intraventricular', 'subarachnoid','subdural']]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f87d59e16d8>"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "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": [
    "import seaborn as sns\n",
    "sns.countplot(z.iloc[:,1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.0    169177\n",
       "Name: epidural, dtype: int64"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "q = z.iloc[:,1]\n",
    "q[q==0].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 257598 validated image filenames.\n",
      "Found 64320 validated image filenames.\n"
     ]
    }
   ],
   "source": [
    "train_under_generator = train_idg.flow_from_dataframe(l, directory = image_dir,\n",
    "                           x_col = \"Image\", y_col = columns,batch_size=32,\n",
    "                           class_mode=\"raw\", target_size=(224,224), shuffle = True)\n",
    "valid_under_generator  = valid_idg.flow_from_dataframe(m, directory = image_dir,\n",
    "                        x_col = \"Image\", y_col = columns,batch_size=32,\n",
    "                        class_mode = \"raw\",target_size=(224,224), shuffle = False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"vgg19\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "input_1 (InputLayer)         [(None, 224, 224, 3)]     0         \n",
      "_________________________________________________________________\n",
      "block1_conv1 (Conv2D)        (None, 224, 224, 64)      1792      \n",
      "_________________________________________________________________\n",
      "block1_conv2 (Conv2D)        (None, 224, 224, 64)      36928     \n",
      "_________________________________________________________________\n",
      "block1_pool (MaxPooling2D)   (None, 112, 112, 64)      0         \n",
      "_________________________________________________________________\n",
      "block2_conv1 (Conv2D)        (None, 112, 112, 128)     73856     \n",
      "_________________________________________________________________\n",
      "block2_conv2 (Conv2D)        (None, 112, 112, 128)     147584    \n",
      "_________________________________________________________________\n",
      "block2_pool (MaxPooling2D)   (None, 56, 56, 128)       0         \n",
      "_________________________________________________________________\n",
      "block3_conv1 (Conv2D)        (None, 56, 56, 256)       295168    \n",
      "_________________________________________________________________\n",
      "block3_conv2 (Conv2D)        (None, 56, 56, 256)       590080    \n",
      "_________________________________________________________________\n",
      "block3_conv3 (Conv2D)        (None, 56, 56, 256)       590080    \n",
      "_________________________________________________________________\n",
      "block3_conv4 (Conv2D)        (None, 56, 56, 256)       590080    \n",
      "_________________________________________________________________\n",
      "block3_pool (MaxPooling2D)   (None, 28, 28, 256)       0         \n",
      "_________________________________________________________________\n",
      "block4_conv1 (Conv2D)        (None, 28, 28, 512)       1180160   \n",
      "_________________________________________________________________\n",
      "block4_conv2 (Conv2D)        (None, 28, 28, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block4_conv3 (Conv2D)        (None, 28, 28, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block4_conv4 (Conv2D)        (None, 28, 28, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block4_pool (MaxPooling2D)   (None, 14, 14, 512)       0         \n",
      "_________________________________________________________________\n",
      "block5_conv1 (Conv2D)        (None, 14, 14, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block5_conv2 (Conv2D)        (None, 14, 14, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block5_conv3 (Conv2D)        (None, 14, 14, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block5_conv4 (Conv2D)        (None, 14, 14, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block5_pool (MaxPooling2D)   (None, 7, 7, 512)         0         \n",
      "=================================================================\n",
      "Total params: 20,024,384\n",
      "Trainable params: 20,024,384\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "from tensorflow.keras.applications.vgg19 import VGG19\n",
    "\n",
    "\n",
    "base_model = VGG19(weights='imagenet', include_top=False, input_shape=(224,224,3))\n",
    "\n",
    "base_model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model loaded.\n",
      "Model: \"functional_6\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "input_6 (InputLayer)         [(None, 224, 224, 3)]     0         \n",
      "_________________________________________________________________\n",
      "block1_conv1 (Conv2D)        (None, 224, 224, 64)      1792      \n",
      "_________________________________________________________________\n",
      "block1_conv2 (Conv2D)        (None, 224, 224, 64)      36928     \n",
      "_________________________________________________________________\n",
      "block1_pool (MaxPooling2D)   (None, 112, 112, 64)      0         \n",
      "_________________________________________________________________\n",
      "block2_conv1 (Conv2D)        (None, 112, 112, 128)     73856     \n",
      "_________________________________________________________________\n",
      "block2_conv2 (Conv2D)        (None, 112, 112, 128)     147584    \n",
      "_________________________________________________________________\n",
      "block2_pool (MaxPooling2D)   (None, 56, 56, 128)       0         \n",
      "_________________________________________________________________\n",
      "block3_conv1 (Conv2D)        (None, 56, 56, 256)       295168    \n",
      "_________________________________________________________________\n",
      "block3_conv2 (Conv2D)        (None, 56, 56, 256)       590080    \n",
      "_________________________________________________________________\n",
      "block3_conv3 (Conv2D)        (None, 56, 56, 256)       590080    \n",
      "_________________________________________________________________\n",
      "block3_conv4 (Conv2D)        (None, 56, 56, 256)       590080    \n",
      "_________________________________________________________________\n",
      "block3_pool (MaxPooling2D)   (None, 28, 28, 256)       0         \n",
      "_________________________________________________________________\n",
      "block4_conv1 (Conv2D)        (None, 28, 28, 512)       1180160   \n",
      "_________________________________________________________________\n",
      "block4_conv2 (Conv2D)        (None, 28, 28, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block4_conv3 (Conv2D)        (None, 28, 28, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block4_conv4 (Conv2D)        (None, 28, 28, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block4_pool (MaxPooling2D)   (None, 14, 14, 512)       0         \n",
      "_________________________________________________________________\n",
      "block5_conv1 (Conv2D)        (None, 14, 14, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block5_conv2 (Conv2D)        (None, 14, 14, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block5_conv3 (Conv2D)        (None, 14, 14, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block5_conv4 (Conv2D)        (None, 14, 14, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block5_pool (MaxPooling2D)   (None, 7, 7, 512)         0         \n",
      "_________________________________________________________________\n",
      "global_average_pooling2d_4 ( (None, 512)               0         \n",
      "_________________________________________________________________\n",
      "dense_8 (Dense)              (None, 1152)              590976    \n",
      "_________________________________________________________________\n",
      "dense_9 (Dense)              (None, 6)                 6918      \n",
      "=================================================================\n",
      "Total params: 20,622,278\n",
      "Trainable params: 20,622,278\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "from tensorflow.keras.applications.vgg19 import VGG19\n",
    "import tensorflow as tf\n",
    "from tensorflow.keras.models import Model,Sequential\n",
    "from tensorflow.keras.layers import Dense, GlobalAveragePooling2D, Dropout\n",
    "\n",
    "METRICS = [\n",
    "      tf.keras.metrics.TruePositives(name='tp'),\n",
    "      tf.keras.metrics.FalsePositives(name='fp'),\n",
    "      tf.keras.metrics.TrueNegatives(name='tn'),\n",
    "      tf.keras.metrics.FalseNegatives(name='fn'), \n",
    "      tf.keras.metrics.BinaryAccuracy(name='accuracy'),\n",
    "      tf.keras.metrics.Precision(name='precision'),\n",
    "      tf.keras.metrics.Recall(name='recall'),\n",
    "      tf.keras.metrics.AUC(name='auc')\n",
    "      \n",
    "]\n",
    "\n",
    "\n",
    "# create the base pre-trained model\n",
    "base_model = VGG19(weights='imagenet', include_top=False, input_shape=(224,224,3))\n",
    "print('Model loaded.')\n",
    "\n",
    "\n",
    "\n",
    "# add a global spatial average pooling layer\n",
    "x = base_model.output\n",
    "\n",
    "x = GlobalAveragePooling2D()(x)\n",
    "net = Dense(1152, activation='elu')(x)\n",
    "net = Dense(6, activation='sigmoid')(net)\n",
    "\n",
    "\n",
    "# this is the model we will train\n",
    "model = Model(inputs=base_model.input, outputs=net)\n",
    "\n",
    "\n",
    "\n",
    "# first: train only the top layers (which were randomly initialized)\n",
    "# i.e. freeze all convolutional InceptionV3 layers\n",
    "#for layer in base_model.layers:\n",
    "#    layer.trainable = False\n",
    "\n",
    "\n",
    "\n",
    "# compile the model (should be done *after* setting layers to non-trainable)\n",
    "model.compile(opt, loss='binary_crossentropy', metrics=METRICS)\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:`period` argument is deprecated. Please use `save_freq` to specify the frequency in number of batches seen.\n"
     ]
    }
   ],
   "source": [
    "from keras import backend as K\n",
    "\n",
    "from tensorflow.keras.callbacks import ModelCheckpoint\n",
    "\n",
    "\n",
    "checkpoint = tf.keras.callbacks.ModelCheckpoint('/kaggle/models/vgg19_{epoch:08d}.h5', period=1,mode= 'auto',save_best_only=True) \n",
    "\n",
    "learning_rate_reduction = tf.keras.callbacks.ReduceLROnPlateau(monitor='val_acc', \n",
    "                                            patience=2, \n",
    "                                            verbose=1, \n",
    "                                            factor=0.5, \n",
    "                                            min_lr=0.00001)\n",
    "\n",
    "callback_list = [checkpoint]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "tf.config.experimental_run_functions_eagerly(True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{0: 1.0, 1: 2.0}"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "class_weight = {0:1.0,1:2.0}\n",
    "class_weight"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/20\n",
      "1173/1173 [==============================] - 697s 595ms/step - loss: 0.2919 - tp: 6761.0000 - fp: 3858.0000 - tn: 191565.0000 - fn: 23032.0000 - accuracy: 0.8806 - precision: 0.6367 - recall: 0.2269 - auc: 0.8424 - val_loss: 0.2777 - val_tp: 2237.0000 - val_fp: 1183.0000 - val_tn: 48163.0000 - val_fn: 5249.0000 - val_accuracy: 0.8868 - val_precision: 0.6541 - val_recall: 0.2988 - val_auc: 0.8668\n",
      "Epoch 2/20\n",
      "1173/1173 [==============================] - 669s 570ms/step - loss: 0.2493 - tp: 10704.0000 - fp: 4243.0000 - tn: 191540.0000 - fn: 18729.0000 - accuracy: 0.8980 - precision: 0.7161 - recall: 0.3637 - auc: 0.8906 - val_loss: 0.2236 - val_tp: 2848.0000 - val_fp: 621.0000 - val_tn: 48725.0000 - val_fn: 4638.0000 - val_accuracy: 0.9075 - val_precision: 0.8210 - val_recall: 0.3804 - val_auc: 0.9189\n",
      "Epoch 3/20\n",
      "1173/1173 [==============================] - 652s 556ms/step - loss: 0.2154 - tp: 14398.0000 - fp: 4250.0000 - tn: 191429.0000 - fn: 15139.0000 - accuracy: 0.9139 - precision: 0.7721 - recall: 0.4875 - auc: 0.9198 - val_loss: 0.1940 - val_tp: 3987.0000 - val_fp: 875.0000 - val_tn: 48471.0000 - val_fn: 3499.0000 - val_accuracy: 0.9230 - val_precision: 0.8200 - val_recall: 0.5326 - val_auc: 0.9395\n",
      "Epoch 5/20\n",
      "1173/1173 [==============================] - 644s 549ms/step - loss: 0.1969 - tp: 16148.0000 - fp: 4148.0000 - tn: 191608.0000 - fn: 13300.0000 - accuracy: 0.9225 - precision: 0.7956 - recall: 0.5484 - auc: 0.9333 - val_loss: 0.1848 - val_tp: 4151.0000 - val_fp: 837.0000 - val_tn: 48509.0000 - val_fn: 3335.0000 - val_accuracy: 0.9266 - val_precision: 0.8322 - val_recall: 0.5545 - val_auc: 0.9451\n",
      "Epoch 6/20\n",
      "1173/1173 [==============================] - 641s 547ms/step - loss: 0.1904 - tp: 16946.0000 - fp: 4295.0000 - tn: 191186.0000 - fn: 12789.0000 - accuracy: 0.9241 - precision: 0.7978 - recall: 0.5699 - auc: 0.9391 - val_loss: 0.1959 - val_tp: 4679.0000 - val_fp: 1386.0000 - val_tn: 47960.0000 - val_fn: 2807.0000 - val_accuracy: 0.9262 - val_precision: 0.7715 - val_recall: 0.6250 - val_auc: 0.9415\n",
      "Epoch 7/20\n",
      "1173/1173 [==============================] - 639s 545ms/step - loss: 0.2022 - tp: 16387.0000 - fp: 4267.0000 - tn: 190767.0000 - fn: 13795.0000 - accuracy: 0.9198 - precision: 0.7934 - recall: 0.5429 - auc: 0.9313 - val_loss: 0.1972 - val_tp: 4920.0000 - val_fp: 1831.0000 - val_tn: 47515.0000 - val_fn: 2566.0000 - val_accuracy: 0.9226 - val_precision: 0.7288 - val_recall: 0.6572 - val_auc: 0.9411\n",
      "Epoch 8/20\n",
      "1173/1173 [==============================] - 640s 546ms/step - loss: 0.1893 - tp: 16951.0000 - fp: 4180.0000 - tn: 191339.0000 - fn: 12746.0000 - accuracy: 0.9248 - precision: 0.8022 - recall: 0.5708 - auc: 0.9394 - val_loss: 0.1731 - val_tp: 4207.0000 - val_fp: 680.0000 - val_tn: 48666.0000 - val_fn: 3279.0000 - val_accuracy: 0.9303 - val_precision: 0.8609 - val_recall: 0.5620 - val_auc: 0.9526\n",
      "Epoch 9/20\n",
      "1173/1173 [==============================] - 630s 537ms/step - loss: 0.1804 - tp: 17891.0000 - fp: 4102.0000 - tn: 191323.0000 - fn: 11900.0000 - accuracy: 0.9289 - precision: 0.8135 - recall: 0.6006 - auc: 0.9455 - val_loss: 0.1666 - val_tp: 4428.0000 - val_fp: 703.0000 - val_tn: 48643.0000 - val_fn: 3058.0000 - val_accuracy: 0.9338 - val_precision: 0.8630 - val_recall: 0.5915 - val_auc: 0.9565\n",
      "Epoch 10/20\n",
      "1173/1173 [==============================] - 627s 535ms/step - loss: 0.1707 - tp: 18517.0000 - fp: 4044.0000 - tn: 191608.0000 - fn: 11047.0000 - accuracy: 0.9330 - precision: 0.8208 - recall: 0.6263 - auc: 0.9514 - val_loss: 0.1651 - val_tp: 4490.0000 - val_fp: 706.0000 - val_tn: 48640.0000 - val_fn: 2996.0000 - val_accuracy: 0.9349 - val_precision: 0.8641 - val_recall: 0.5998 - val_auc: 0.9561\n",
      "Epoch 11/20\n",
      "1173/1173 [==============================] - 625s 533ms/step - loss: 0.1663 - tp: 18947.0000 - fp: 3893.0000 - tn: 191696.0000 - fn: 10668.0000 - accuracy: 0.9353 - precision: 0.8296 - recall: 0.6398 - auc: 0.9541 - val_loss: 0.1664 - val_tp: 4283.0000 - val_fp: 627.0000 - val_tn: 48719.0000 - val_fn: 3203.0000 - val_accuracy: 0.9326 - val_precision: 0.8723 - val_recall: 0.5721 - val_auc: 0.9579\n",
      "Epoch 12/20\n",
      "1173/1173 [==============================] - 622s 530ms/step - loss: 0.1613 - tp: 19212.0000 - fp: 3870.0000 - tn: 191838.0000 - fn: 10296.0000 - accuracy: 0.9371 - precision: 0.8323 - recall: 0.6511 - auc: 0.9565 - val_loss: 0.1584 - val_tp: 4549.0000 - val_fp: 643.0000 - val_tn: 48703.0000 - val_fn: 2937.0000 - val_accuracy: 0.9370 - val_precision: 0.8762 - val_recall: 0.6077 - val_auc: 0.9610\n",
      "Epoch 13/20\n",
      "1173/1173 [==============================] - 621s 529ms/step - loss: 0.1662 - tp: 18890.0000 - fp: 3871.0000 - tn: 191666.0000 - fn: 10789.0000 - accuracy: 0.9349 - precision: 0.8299 - recall: 0.6365 - auc: 0.9536 - val_loss: 0.1702 - val_tp: 4126.0000 - val_fp: 548.0000 - val_tn: 48798.0000 - val_fn: 3360.0000 - val_accuracy: 0.9312 - val_precision: 0.8828 - val_recall: 0.5512 - val_auc: 0.9561\n",
      "Epoch 14/20\n",
      "1173/1173 [==============================] - 619s 528ms/step - loss: 0.1756 - tp: 18430.0000 - fp: 4014.0000 - tn: 191352.0000 - fn: 11420.0000 - accuracy: 0.9315 - precision: 0.8212 - recall: 0.6174 - auc: 0.9482 - val_loss: 0.1837 - val_tp: 3467.0000 - val_fp: 298.0000 - val_tn: 49048.0000 - val_fn: 4019.0000 - val_accuracy: 0.9240 - val_precision: 0.9208 - val_recall: 0.4631 - val_auc: 0.9556\n",
      "Epoch 15/20\n",
      "1173/1173 [==============================] - 618s 526ms/step - loss: 0.1697 - tp: 18605.0000 - fp: 3954.0000 - tn: 191687.0000 - fn: 10970.0000 - accuracy: 0.9337 - precision: 0.8247 - recall: 0.6291 - auc: 0.9515 - val_loss: 0.1694 - val_tp: 5286.0000 - val_fp: 1564.0000 - val_tn: 47782.0000 - val_fn: 2200.0000 - val_accuracy: 0.9338 - val_precision: 0.7717 - val_recall: 0.7061 - val_auc: 0.9558\n",
      "Epoch 16/20\n",
      "1173/1173 [==============================] - 616s 525ms/step - loss: 0.1653 - tp: 19077.0000 - fp: 3865.0000 - tn: 191664.0000 - fn: 10610.0000 - accuracy: 0.9357 - precision: 0.8315 - recall: 0.6426 - auc: 0.9542 - val_loss: 0.1558 - val_tp: 4644.0000 - val_fp: 666.0000 - val_tn: 48680.0000 - val_fn: 2842.0000 - val_accuracy: 0.9383 - val_precision: 0.8746 - val_recall: 0.6204 - val_auc: 0.9623\n",
      "Epoch 17/20\n",
      "1173/1173 [==============================] - 616s 525ms/step - loss: 0.1622 - tp: 19496.0000 - fp: 3790.0000 - tn: 191623.0000 - fn: 10307.0000 - accuracy: 0.9374 - precision: 0.8372 - recall: 0.6542 - auc: 0.9563 - val_loss: 0.1521 - val_tp: 5144.0000 - val_fp: 1045.0000 - val_tn: 48301.0000 - val_fn: 2342.0000 - val_accuracy: 0.9404 - val_precision: 0.8312 - val_recall: 0.6871 - val_auc: 0.9628\n",
      "Epoch 18/20\n",
      "1173/1173 [==============================] - 615s 525ms/step - loss: 0.1581 - tp: 19601.0000 - fp: 3775.0000 - tn: 191843.0000 - fn: 9997.0000 - accuracy: 0.9389 - precision: 0.8385 - recall: 0.6622 - auc: 0.9584 - val_loss: 0.1510 - val_tp: 5129.0000 - val_fp: 973.0000 - val_tn: 48373.0000 - val_fn: 2357.0000 - val_accuracy: 0.9414 - val_precision: 0.8405 - val_recall: 0.6851 - val_auc: 0.9637\n",
      "Epoch 19/20\n",
      "1173/1173 [==============================] - 615s 525ms/step - loss: 0.1531 - tp: 19905.0000 - fp: 3709.0000 - tn: 191990.0000 - fn: 9600.0000 - accuracy: 0.9409 - precision: 0.8429 - recall: 0.6746 - auc: 0.9609 - val_loss: 0.1494 - val_tp: 4919.0000 - val_fp: 749.0000 - val_tn: 48597.0000 - val_fn: 2567.0000 - val_accuracy: 0.9417 - val_precision: 0.8679 - val_recall: 0.6571 - val_auc: 0.9637\n",
      "Epoch 20/20\n",
      "1173/1173 [==============================] - 615s 525ms/step - loss: 0.1530 - tp: 20110.0000 - fp: 3754.0000 - tn: 191784.0000 - fn: 9568.0000 - accuracy: 0.9408 - precision: 0.8427 - recall: 0.6776 - auc: 0.9612 - val_loss: 0.1445 - val_tp: 5020.0000 - val_fp: 744.0000 - val_tn: 48602.0000 - val_fn: 2466.0000 - val_accuracy: 0.9435 - val_precision: 0.8709 - val_recall: 0.6706 - val_auc: 0.9670\n"
     ]
    }
   ],
   "source": [
    "\n",
    "\n",
    "\n",
    "num_epochs = 20\n",
    "\n",
    "batch_size = 512\n",
    "training_steps = len(training_data) // batch_size\n",
    "validation_step = len(validation_data) // batch_size\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "# FIT THE MODEL\n",
    "history = model.fit(train_under_generator,\n",
    "            epochs=num_epochs,\n",
    "            steps_per_epoch=training_steps,\n",
    "            callbacks=callback_list,\n",
    "            class_weight=class_weight,\n",
    "            validation_data=valid_under_generator,\n",
    "            validation_steps= validation_step\n",
    "                   ) \n",
    "\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "tf.keras.backend.clear_session()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Evalution"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From <ipython-input-35-b13240840368>:1: Model.evaluate_generator (from tensorflow.python.keras.engine.training) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Please use Model.evaluate, which supports generators.\n",
      "[0.15216833353042603, 33639.0, 5408.0, 329759.0, 17114.0, 0.9416413307189941, 0.8615002632141113, 0.662798285484314, 0.961879312992096]\n"
     ]
    }
   ],
   "source": [
    "valid_predict =  model.evaluate_generator(valid_under_generator)\n",
    "print(valid_predict)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['loss', 'tp', 'fp', 'tn', 'fn', 'accuracy', 'precision', 'recall', 'auc']"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.metrics_names"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "---------------\n",
      "\n",
      "validation data **loss** value = 0.15216833353042603\n",
      "\n",
      "---------------\n",
      "\n",
      "validation data **true positive** value =  33639.0\n",
      "\n",
      "---------------\n",
      "\n",
      "validation data **false positive** value = 5408.0\n",
      "\n",
      "---------------\n",
      "\n",
      "validation data **true negative** value = 329759.0\n",
      "\n",
      "---------------\n",
      "\n",
      "validation data **false negative** value = 17114.0\n",
      "\n",
      "---------------\n",
      "\n",
      "validation data **accuracy** value =  0.9416413307189941\n",
      "\n",
      "---------------\n",
      "\n",
      "validation data **precision** value = 0.8615002632141113\n",
      "\n",
      "---------------\n",
      "\n",
      "validation data **recall* value = 0.662798285484314\n",
      "\n",
      "---------------\n",
      "\n",
      "validation data **AUC* value = 0.961879312992096\n",
      "\n",
      "---------------\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print('\\n---------------\\n')\n",
    "print('validation data **loss** value =', valid_predict[0])\n",
    "print('\\n---------------\\n')\n",
    "print('validation data **true positive** value = ', valid_predict[1])\n",
    "print('\\n---------------\\n')\n",
    "print('validation data **false positive** value =', valid_predict[2])\n",
    "print('\\n---------------\\n')\n",
    "print('validation data **true negative** value =', valid_predict[3])\n",
    "print('\\n---------------\\n')\n",
    "print('validation data **false negative** value =', valid_predict[4])\n",
    "print('\\n---------------\\n')\n",
    "print('validation data **accuracy** value = ', valid_predict[5])\n",
    "print('\\n---------------\\n')\n",
    "print('validation data **precision** value =', valid_predict[6])\n",
    "print('\\n---------------\\n')\n",
    "print('validation data **recall* value =', valid_predict[7])\n",
    "print('\\n---------------\\n')\n",
    "print('validation data **AUC* value =', valid_predict[8])\n",
    "print('\\n---------------\\n')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_true = m[['any','epidural','intraparenchymal','intraventricular', 'subarachnoid','subdural']].reset_index(drop=True)\n",
    "\n",
    "#Y_pred = model.predict_generator(valid_under_generator)\n",
    "preds = np.where(Y_pred < 0.5, 0, 1)\n",
    "\n",
    "\n",
    "\n",
    "#val = 0.25\n",
    "\n",
    "#Y_pred[Y_pred>=val]=1\n",
    "#Y_pred[Y_pred<val]=0"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# classification"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Classification Report\n",
      "                  precision    recall  f1-score   support\n",
      "\n",
      "             any       0.90      0.80      0.85     21440\n",
      "        epidural       0.00      0.00      0.00       572\n",
      "intraparenchymal       0.88      0.68      0.76      6912\n",
      "intraventricular       0.80      0.74      0.77      5093\n",
      "    subarachnoid       0.78      0.47      0.59      7218\n",
      "        subdural       0.82      0.48      0.61      9518\n",
      "\n",
      "       micro avg       0.86      0.66      0.75     50753\n",
      "       macro avg       0.70      0.53      0.60     50753\n",
      "    weighted avg       0.85      0.66      0.74     50753\n",
      "     samples avg       0.25      0.22      0.23     50753\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ubuntu/anaconda3/envs/tensorflow2_p36/lib/python3.6/site-packages/sklearn/metrics/_classification.py:1272: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n",
      "/home/ubuntu/anaconda3/envs/tensorflow2_p36/lib/python3.6/site-packages/sklearn/metrics/_classification.py:1272: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in samples with no predicted labels. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n",
      "/home/ubuntu/anaconda3/envs/tensorflow2_p36/lib/python3.6/site-packages/sklearn/metrics/_classification.py:1272: UndefinedMetricWarning: Recall and F-score are ill-defined and being set to 0.0 in samples with no true labels. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n"
     ]
    }
   ],
   "source": [
    "print('Classification Report')\n",
    "target_names = ['any','epidural','intraparenchymal','intraventricular', 'subarachnoid','subdural']\n",
    "print(classification_report(y_true, preds, target_names=target_names))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1008x576 with 6 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import seaborn as sns\n",
    "from sklearn.metrics import multilabel_confusion_matrix\n",
    "# Creating multilabel confusion matrix\n",
    "confusion = multilabel_confusion_matrix(y_true, preds)\n",
    "mlb= ['any','epidural','intraparenchymal','intraventricular', 'subarachnoid','subdural']\n",
    "# Plot confusion matrix \n",
    "fig = plt.figure(figsize = (14, 8))\n",
    "for i, (label, matrix) in enumerate(zip(mlb, confusion)):\n",
    "    plt.subplot(f'23{i+1}')\n",
    "    labels = [f'not_{label}', label]\n",
    "    sns.heatmap(matrix, annot = True, square = True, fmt = 'd', cbar = False, cmap = 'Blues', \n",
    "                xticklabels = labels, yticklabels = labels, linecolor = 'black', linewidth = 1)\n",
    "    plt.title(labels[0])\n",
    "\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.7555278784156534\n"
     ]
    }
   ],
   "source": [
    "auc = roc_auc_score(y_true, preds)\n",
    "print(auc)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# accuarcy and loss"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_training(H):\n",
    "    # construct a plot that plots and saves the training history\n",
    "    with plt.xkcd():\n",
    "        plt.figure(figsize = (10,10))\n",
    "        plt.plot(H.epoch,H.history[\"accuracy\"], label=\"train_acc\")\n",
    "        plt.plot(H.epoch,H.history[\"val_accuracy\"], label=\"val_acc\")\n",
    "        plt.title(\"Training Accuracy\")\n",
    "        plt.xlabel(\"Epoch #\")\n",
    "        plt.ylabel(\"Accuracy\")\n",
    "        plt.legend(loc=\"lower left\")\n",
    "        plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "findfont: Font family ['xkcd', 'xkcd Script', 'Humor Sans', 'Comic Sans MS'] not found. Falling back to DejaVu Sans.\n",
      "findfont: Font family ['xkcd', 'xkcd Script', 'Humor Sans', 'Comic Sans MS'] not found. Falling back to DejaVu Sans.\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_training(history)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_training(H):\n",
    "    # construct a plot that plots and saves the training history\n",
    "    with plt.xkcd():\n",
    "        plt.figure(figsize = (10,10))\n",
    "        plt.plot(H.epoch,H.history[\"loss\"], label=\"train_loss\")\n",
    "        plt.plot(H.epoch,H.history[\"val_loss\"], label=\"val_loss\")\n",
    "        plt.title(\"Training Loss\")\n",
    "        plt.xlabel(\"Epoch #\")\n",
    "        plt.ylabel(\"Loss\")\n",
    "        plt.legend(loc=\"lower left\")\n",
    "        plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_training(history)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}