testing_trained_models/calculating_test_metrics_on_official_chexpert_validation_set.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "97833658",
   "metadata": {},
   "source": [
    "### Test metrics on official chexpert validation set (test set for this project)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "17f8a53b",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2021-11-28 20:32:47.278150: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.10.1\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "from tensorflow.keras.preprocessing.image import ImageDataGenerator\n",
    "from tensorflow.keras.models import load_model\n",
    "from sklearn.metrics import classification_report, multilabel_confusion_matrix, roc_auc_score, roc_curve\n",
    "\n",
    "# Custom objects required for loading CheXpert multi-label classification models\n",
    "import sys \n",
    "sys.path.append('..')\n",
    "from loss import (\n",
    "    set_binary_crossentropy_weighted_loss\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "212355e6",
   "metadata": {},
   "source": [
    "### 1- Loading test set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "9bd6a5d3",
   "metadata": {},
   "outputs": [],
   "source": [
    "test_df = pd.read_csv(\n",
    "    filepath_or_buffer=\"../labels/train_validation_split_data/test_u-zeroes.csv\",\n",
    "    dtype={  # Setting labels to type np.float32 was necessary for conversion to tf.Tensor object\n",
    "        \"Path\": str,\n",
    "        \"Atelectasis\": np.float32,\n",
    "        \"Cardiomegaly\": np.float32,\n",
    "        \"Consolidation\": np.float32,\n",
    "        \"Edema\": np.float32,\n",
    "        \"Pleural Effusion\": np.float32,\n",
    "        \"Pleural Other\": np.float32,\n",
    "        \"Pneumonia\": np.float32,\n",
    "        \"Pneumothorax\": np.float32,\n",
    "        \"Enlarged Cardiomediastinum\": np.float32,\n",
    "        \"Lung Opacity\": np.float32,\n",
    "        \"Lung Lesion\": np.float32,\n",
    "        \"Fracture\": np.float32,\n",
    "        \"Support Devices\": np.float32,\n",
    "        \"No Finding\": np.float32\n",
    "    }\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "d1386724",
   "metadata": {},
   "outputs": [],
   "source": [
    "list_columns = list(test_df.columns)\n",
    "y_cols = list_columns[1::]  # First column is 'Path' column"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "ecc1616b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['Path',\n",
       " 'Atelectasis',\n",
       " 'Cardiomegaly',\n",
       " 'Consolidation',\n",
       " 'Edema',\n",
       " 'Pleural Effusion',\n",
       " 'Pleural Other',\n",
       " 'Pneumonia',\n",
       " 'Pneumothorax',\n",
       " 'Enlarged Cardiomediastinum',\n",
       " 'Lung Opacity',\n",
       " 'Lung Lesion',\n",
       " 'Fracture',\n",
       " 'Support Devices',\n",
       " 'No Finding']"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "list_columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "8d5b89b2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['Atelectasis',\n",
       " 'Cardiomegaly',\n",
       " 'Consolidation',\n",
       " 'Edema',\n",
       " 'Pleural Effusion',\n",
       " 'Pleural Other',\n",
       " 'Pneumonia',\n",
       " 'Pneumothorax',\n",
       " 'Enlarged Cardiomediastinum',\n",
       " 'Lung Opacity',\n",
       " 'Lung Lesion',\n",
       " 'Fracture',\n",
       " 'Support Devices',\n",
       " 'No Finding']"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_cols"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "07bd7ce1",
   "metadata": {},
   "outputs": [],
   "source": [
    "test_datagen = ImageDataGenerator(\n",
    "    featurewise_center=True,  # Mean and standard deviation values of the training set will be loaded to the object\n",
    "    featurewise_std_normalization=True\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "03ac5d95",
   "metadata": {},
   "outputs": [],
   "source": [
    "training_dataset_mean = np.load(\"../misc/calculated_chexpert_training_dataset_mean_and_std_values/CheXpert_training_set_mean.npy\")\n",
    "training_dataset_std = np.load(\"../misc/calculated_chexpert_training_dataset_mean_and_std_values/CheXpert_training_set_std.npy\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "358f2ccd",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 234 validated image filenames.\n"
     ]
    }
   ],
   "source": [
    "test_datagen.mean = training_dataset_mean\n",
    "test_datagen.std = training_dataset_std\n",
    "\n",
    "test_datagenerator = test_datagen.flow_from_dataframe(\n",
    "    dataframe=test_df,\n",
    "    directory=\"../dataset/\",\n",
    "    x_col='Path',\n",
    "    y_col=y_cols,\n",
    "    weight_col=None,\n",
    "    target_size=(512, 512),\n",
    "    color_mode='grayscale',\n",
    "    class_mode='raw',\n",
    "    batch_size=1,\n",
    "    validate_filenames=True,\n",
    "    shuffle=False\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cec9879c",
   "metadata": {},
   "source": [
    "### 2- Loading models"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "f6ea17ab",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2021-11-28 20:32:48.731289: I tensorflow/compiler/jit/xla_cpu_device.cc:41] Not creating XLA devices, tf_xla_enable_xla_devices not set\n",
      "2021-11-28 20:32:48.731830: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcuda.so.1\n",
      "2021-11-28 20:32:48.780061: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:941] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
      "2021-11-28 20:32:48.780685: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1720] Found device 0 with properties: \n",
      "pciBusID: 0000:01:00.0 name: NVIDIA TITAN RTX computeCapability: 7.5\n",
      "coreClock: 1.77GHz coreCount: 72 deviceMemorySize: 23.65GiB deviceMemoryBandwidth: 625.94GiB/s\n",
      "2021-11-28 20:32:48.780702: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.10.1\n",
      "2021-11-28 20:32:48.781843: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublas.so.10\n",
      "2021-11-28 20:32:48.781870: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublasLt.so.10\n",
      "2021-11-28 20:32:48.782893: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcufft.so.10\n",
      "2021-11-28 20:32:48.783054: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcurand.so.10\n",
      "2021-11-28 20:32:48.784200: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusolver.so.10\n",
      "2021-11-28 20:32:48.784850: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusparse.so.10\n",
      "2021-11-28 20:32:48.787304: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudnn.so.7\n",
      "2021-11-28 20:32:48.787398: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:941] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
      "2021-11-28 20:32:48.787984: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:941] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
      "2021-11-28 20:32:48.788561: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1862] Adding visible gpu devices: 0\n",
      "2021-11-28 20:32:48.788785: I tensorflow/core/platform/cpu_feature_guard.cc:142] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations:  SSE4.1 SSE4.2 AVX AVX2 FMA\n",
      "To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.\n",
      "2021-11-28 20:32:48.789222: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:941] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
      "2021-11-28 20:32:48.789762: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1720] Found device 0 with properties: \n",
      "pciBusID: 0000:01:00.0 name: NVIDIA TITAN RTX computeCapability: 7.5\n",
      "coreClock: 1.77GHz coreCount: 72 deviceMemorySize: 23.65GiB deviceMemoryBandwidth: 625.94GiB/s\n",
      "2021-11-28 20:32:48.789779: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.10.1\n",
      "2021-11-28 20:32:48.789792: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublas.so.10\n",
      "2021-11-28 20:32:48.789804: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublasLt.so.10\n",
      "2021-11-28 20:32:48.789815: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcufft.so.10\n",
      "2021-11-28 20:32:48.789826: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcurand.so.10\n",
      "2021-11-28 20:32:48.789838: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusolver.so.10\n",
      "2021-11-28 20:32:48.789849: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusparse.so.10\n",
      "2021-11-28 20:32:48.789861: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudnn.so.7\n",
      "2021-11-28 20:32:48.789899: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:941] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
      "2021-11-28 20:32:48.790573: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:941] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
      "2021-11-28 20:32:48.791087: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1862] Adding visible gpu devices: 0\n",
      "2021-11-28 20:32:48.791109: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.10.1\n",
      "2021-11-28 20:32:49.157720: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1261] Device interconnect StreamExecutor with strength 1 edge matrix:\n",
      "2021-11-28 20:32:49.157737: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1267]      0 \n",
      "2021-11-28 20:32:49.157741: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1280] 0:   N \n",
      "2021-11-28 20:32:49.157872: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:941] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
      "2021-11-28 20:32:49.158437: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:941] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
      "2021-11-28 20:32:49.159151: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:941] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
      "2021-11-28 20:32:49.159651: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1406] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 21846 MB memory) -> physical GPU (device: 0, name: NVIDIA TITAN RTX, pci bus id: 0000:01:00.0, compute capability: 7.5)\n",
      "2021-11-28 20:32:49.159821: I tensorflow/compiler/jit/xla_gpu_device.cc:99] Not creating XLA devices, tf_xla_enable_xla_devices not set\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:No training configuration found in the save file, so the model was *not* compiled. Compile it manually.\n"
     ]
    }
   ],
   "source": [
    "densenet201_model = load_model(\n",
    "    \"../trained_models/densenet201.h5\",\n",
    "    custom_objects={\n",
    "        \"binary_crossentropy_weighted_loss\": set_binary_crossentropy_weighted_loss\n",
    "    }\n",
    ")\n",
    "\n",
    "inceptionresnetv2_model = load_model(\n",
    "    \"../trained_models/inceptionresnetv2.h5\",\n",
    "    custom_objects={\n",
    "        \"binary_crossentropy_weighted_loss\": set_binary_crossentropy_weighted_loss\n",
    "    }\n",
    ")\n",
    "\n",
    "ensemble_model = load_model(\n",
    "    \"../trained_models/ensemble-model.h5\",\n",
    "    custom_objects={\n",
    "        \"binary_crossentropy_weighted_loss\": set_binary_crossentropy_weighted_loss\n",
    "    }\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "abbae5b4",
   "metadata": {},
   "source": [
    "### 3- Predictions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "9527dcfd",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2021-11-28 20:33:05.183722: I tensorflow/compiler/mlir/mlir_graph_optimization_pass.cc:116] None of the MLIR optimization passes are enabled (registered 2)\n",
      "2021-11-28 20:33:05.202208: I tensorflow/core/platform/profile_utils/cpu_utils.cc:112] CPU Frequency: 3600000000 Hz\n",
      "2021-11-28 20:33:07.238484: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublas.so.10\n",
      "2021-11-28 20:33:07.365945: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudnn.so.7\n",
      "2021-11-28 20:33:08.063795: W tensorflow/stream_executor/gpu/asm_compiler.cc:63] Running ptxas --version returned 256\n",
      "2021-11-28 20:33:08.102789: W tensorflow/stream_executor/gpu/redzone_allocator.cc:314] Internal: ptxas exited with non-zero error code 256, output: \n",
      "Relying on driver to perform ptx compilation. \n",
      "Modify $PATH to customize ptxas location.\n",
      "This message will be only logged once.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "234/234 [==============================] - 51s 19ms/step\n"
     ]
    }
   ],
   "source": [
    "predictions_densenet201_model_probabilities = densenet201_model.predict(\n",
    "    x=test_datagenerator,\n",
    "    steps=len(test_datagenerator),\n",
    "    workers=1,\n",
    "    verbose=1\n",
    ")\n",
    "\n",
    "# \"Sigmoid\" activation for output probabilities\n",
    "predictions_densenet201_model = (predictions_densenet201_model_probabilities >= 0.5).astype(np.uint8)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "b4e80513",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "234/234 [==============================] - 23s 25ms/step\n"
     ]
    }
   ],
   "source": [
    "predictions_inceptionresnetv2_model_probabilities = inceptionresnetv2_model.predict(\n",
    "    x=test_datagenerator,\n",
    "    steps=len(test_datagenerator),\n",
    "    workers=1,\n",
    "    verbose=1\n",
    ")\n",
    "\n",
    "# \"Sigmoid\" activation for output probabilities\n",
    "predictions_inceptionresnetv2_model = (predictions_inceptionresnetv2_model_probabilities >= 0.5).astype(np.uint8)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "e168c00b",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "234/234 [==============================] - 14s 45ms/step\n"
     ]
    }
   ],
   "source": [
    "predictions_ensemble_model_probabilities = ensemble_model.predict(\n",
    "    x=test_datagenerator,\n",
    "    steps=len(test_datagenerator),\n",
    "    workers=1,\n",
    "    verbose=1\n",
    ")\n",
    "\n",
    "# \"Sigmoid\" activation for output probabilities\n",
    "predictions_ensemble_model = (predictions_ensemble_model_probabilities >= 0.5).astype(np.uint8)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d28bee6b",
   "metadata": {},
   "source": [
    "### All CheXpert labels (all 14 labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "1aa0790d",
   "metadata": {},
   "outputs": [],
   "source": [
    "labels = y_cols"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "83b77052",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['Atelectasis',\n",
       " 'Cardiomegaly',\n",
       " 'Consolidation',\n",
       " 'Edema',\n",
       " 'Pleural Effusion',\n",
       " 'Pleural Other',\n",
       " 'Pneumonia',\n",
       " 'Pneumothorax',\n",
       " 'Enlarged Cardiomediastinum',\n",
       " 'Lung Opacity',\n",
       " 'Lung Lesion',\n",
       " 'Fracture',\n",
       " 'Support Devices',\n",
       " 'No Finding']"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "labels"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "666856d5",
   "metadata": {},
   "source": [
    "#### 1- DenseNet-201 model"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "433555d5",
   "metadata": {},
   "source": [
    "##### 1.1 Classification Report"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "9d010a91",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "DenseNet-201:\n",
      "\n",
      "                            precision    recall  f1-score   support\n",
      "\n",
      "               Atelectasis     0.5429    0.4750    0.5067        80\n",
      "              Cardiomegaly     0.6136    0.3971    0.4821        68\n",
      "             Consolidation     0.3860    0.6667    0.4889        33\n",
      "                     Edema     0.5556    0.6667    0.6061        45\n",
      "          Pleural Effusion     0.7424    0.7313    0.7368        67\n",
      "             Pleural Other     0.0169    1.0000    0.0333         1\n",
      "                 Pneumonia     0.1042    0.6250    0.1786         8\n",
      "              Pneumothorax     0.1290    0.5000    0.2051         8\n",
      "Enlarged Cardiomediastinum     0.5694    0.3761    0.4530       109\n",
      "              Lung Opacity     0.8519    0.7302    0.7863       126\n",
      "               Lung Lesion     0.0000    0.0000    0.0000         1\n",
      "                  Fracture     0.0000    0.0000    0.0000         0\n",
      "           Support Devices     0.8738    0.8411    0.8571       107\n",
      "                No Finding     0.3571    0.9211    0.5147        38\n",
      "\n",
      "                 micro avg     0.4474    0.6281    0.5226       691\n",
      "                 macro avg     0.4102    0.5664    0.4178       691\n",
      "              weighted avg     0.6527    0.6281    0.6207       691\n",
      "               samples avg     0.4499    0.6518    0.4880       691\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/tamer/anaconda3/envs/chexpert/lib/python3.7/site-packages/sklearn/metrics/_classification.py:1248: UndefinedMetricWarning: Recall and F-score are ill-defined and being set to 0.0 in labels with no true samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n",
      "/home/tamer/anaconda3/envs/chexpert/lib/python3.7/site-packages/sklearn/metrics/_classification.py:1248: 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": [
    "report_densenet201_model = classification_report(\n",
    "    y_true=test_datagenerator.labels,\n",
    "    y_pred=predictions_densenet201_model,\n",
    "    digits=4,\n",
    "    target_names=labels\n",
    ")\n",
    "\n",
    "print(\"DenseNet-201:\\n\")\n",
    "print(report_densenet201_model)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f6d10e42",
   "metadata": {},
   "source": [
    "##### 1.2 Confusion Matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "6eb166be",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "DenseNet-201:\n",
      "\n",
      "\n",
      "Atelectasis:\n",
      "\n",
      "[[122  32]\n",
      " [ 42  38]]\n",
      "------------\n",
      "Cardiomegaly:\n",
      "\n",
      "[[149  17]\n",
      " [ 41  27]]\n",
      "------------\n",
      "Consolidation:\n",
      "\n",
      "[[166  35]\n",
      " [ 11  22]]\n",
      "------------\n",
      "Edema:\n",
      "\n",
      "[[165  24]\n",
      " [ 15  30]]\n",
      "------------\n",
      "Pleural Effusion:\n",
      "\n",
      "[[150  17]\n",
      " [ 18  49]]\n",
      "------------\n",
      "Pleural Other:\n",
      "\n",
      "[[175  58]\n",
      " [  0   1]]\n",
      "------------\n",
      "Pneumonia:\n",
      "\n",
      "[[183  43]\n",
      " [  3   5]]\n",
      "------------\n",
      "Pneumothorax:\n",
      "\n",
      "[[199  27]\n",
      " [  4   4]]\n",
      "------------\n",
      "Enlarged Cardiomediastinum:\n",
      "\n",
      "[[94 31]\n",
      " [68 41]]\n",
      "------------\n",
      "Lung Opacity:\n",
      "\n",
      "[[92 16]\n",
      " [34 92]]\n",
      "------------\n",
      "Lung Lesion:\n",
      "\n",
      "[[151  82]\n",
      " [  1   0]]\n",
      "------------\n",
      "Fracture:\n",
      "\n",
      "[[156  78]\n",
      " [  0   0]]\n",
      "------------\n",
      "Support Devices:\n",
      "\n",
      "[[114  13]\n",
      " [ 17  90]]\n",
      "------------\n",
      "No Finding:\n",
      "\n",
      "[[133  63]\n",
      " [  3  35]]\n",
      "------------\n"
     ]
    }
   ],
   "source": [
    "confusion_matrix_densenet201 = multilabel_confusion_matrix(\n",
    "    y_true=test_datagenerator.labels,\n",
    "    y_pred=predictions_densenet201_model\n",
    ")\n",
    "\n",
    "print(\"DenseNet-201:\\n\\n\")\n",
    "for i in range(len(labels)):\n",
    "    print(f\"{labels[i]}:\\n\")\n",
    "    print(confusion_matrix_densenet201[i, :, :])\n",
    "    print(\"------------\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4b3ba7e3",
   "metadata": {},
   "source": [
    "##### 1.3 Receiver Operating Characteristic curve (ROC Curve)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "80c4777f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Error in generating ROC curve for Fracture. Dataset lacks enough examples.\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 936x936 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "auc_roc_values = []\n",
    "\n",
    "for i in range(len(labels)):\n",
    "    try:\n",
    "        y_true = test_datagenerator.labels[:, i]\n",
    "        y_pred = predictions_densenet201_model_probabilities[:, i]\n",
    "        \n",
    "        auc_roc = roc_auc_score(y_true=y_true, y_score=y_pred)\n",
    "        auc_roc_values.append(auc_roc)\n",
    "        \n",
    "        false_positive_rates, true_positive_rates, _ = roc_curve(y_true=y_true, y_score=y_pred)\n",
    "\n",
    "        plt.figure(1, figsize=(13, 13))\n",
    "        plt.plot([0, 1], [0, 1], 'k--')\n",
    "        plt.plot(\n",
    "            false_positive_rates,\n",
    "            true_positive_rates,\n",
    "            label=labels[i] + \" (\" + str(round(auc_roc, 4)) + \")\"\n",
    "        )\n",
    "        plt.xlabel('False positive rate')\n",
    "        plt.ylabel('True positive rate')\n",
    "        plt.title('ROC curve - DenseNet-201 model')\n",
    "        plt.legend(loc='best')\n",
    "    except:\n",
    "        print(\n",
    "            f\"Error in generating ROC curve for {labels[i]}. \"\n",
    "            f\"Dataset lacks enough examples.\"\n",
    "        )\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "63cfd16a",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[0.7436688311688313,\n",
       " 0.7431785967399007,\n",
       " 0.8599427106889792,\n",
       " 0.883480305702528,\n",
       " 0.9154526767360801,\n",
       " 0.8068669527896996,\n",
       " 0.6980088495575222,\n",
       " 0.7721238938053097,\n",
       " 0.5705688073394496,\n",
       " 0.8900646678424455,\n",
       " 0.15879828326180256,\n",
       " 0.9351681507101333,\n",
       " 0.84828141783029]"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "auc_roc_values"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b6017c6e",
   "metadata": {},
   "source": [
    "#### 2- Inception-ResNet-V2 model"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "23d701d2",
   "metadata": {},
   "source": [
    "##### 2.1 Classification Report"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "a91cf5b6",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Inception-ResNet-V2:\n",
      "\n",
      "                            precision    recall  f1-score   support\n",
      "\n",
      "               Atelectasis     0.6190    0.4875    0.5455        80\n",
      "              Cardiomegaly     0.6809    0.4706    0.5565        68\n",
      "             Consolidation     0.4242    0.8485    0.5657        33\n",
      "                     Edema     0.6182    0.7556    0.6800        45\n",
      "          Pleural Effusion     0.7895    0.6716    0.7258        67\n",
      "             Pleural Other     0.0164    1.0000    0.0323         1\n",
      "                 Pneumonia     0.0455    0.2500    0.0769         8\n",
      "              Pneumothorax     0.0962    0.6250    0.1667         8\n",
      "Enlarged Cardiomediastinum     0.5233    0.4128    0.4615       109\n",
      "              Lung Opacity     0.8818    0.7698    0.8220       126\n",
      "               Lung Lesion     0.0000    0.0000    0.0000         1\n",
      "                  Fracture     0.0000    0.0000    0.0000         0\n",
      "           Support Devices     0.8901    0.7570    0.8182       107\n",
      "                No Finding     0.3431    0.9211    0.5000        38\n",
      "\n",
      "                 micro avg     0.4498    0.6425    0.5292       691\n",
      "                 macro avg     0.4234    0.5693    0.4251       691\n",
      "              weighted avg     0.6774    0.6425    0.6393       691\n",
      "               samples avg     0.4403    0.6511    0.4784       691\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/tamer/anaconda3/envs/chexpert/lib/python3.7/site-packages/sklearn/metrics/_classification.py:1248: UndefinedMetricWarning: Recall and F-score are ill-defined and being set to 0.0 in labels with no true samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n",
      "/home/tamer/anaconda3/envs/chexpert/lib/python3.7/site-packages/sklearn/metrics/_classification.py:1248: 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": [
    "report_inceptionresnetv2_model = classification_report(\n",
    "    y_true=test_datagenerator.labels,\n",
    "    y_pred=predictions_inceptionresnetv2_model,\n",
    "    digits=4,\n",
    "    target_names=labels\n",
    ")\n",
    "\n",
    "print(\"Inception-ResNet-V2:\\n\")\n",
    "print(report_inceptionresnetv2_model)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "29a86366",
   "metadata": {},
   "source": [
    "##### 2.2 Confusion Matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "aa121429",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Inception-ResNet-V2:\n",
      "\n",
      "\n",
      "Atelectasis:\n",
      "[[130  24]\n",
      " [ 41  39]]\n",
      "------------\n",
      "Cardiomegaly:\n",
      "[[151  15]\n",
      " [ 36  32]]\n",
      "------------\n",
      "Consolidation:\n",
      "[[163  38]\n",
      " [  5  28]]\n",
      "------------\n",
      "Edema:\n",
      "[[168  21]\n",
      " [ 11  34]]\n",
      "------------\n",
      "Pleural Effusion:\n",
      "[[155  12]\n",
      " [ 22  45]]\n",
      "------------\n",
      "Pleural Other:\n",
      "[[173  60]\n",
      " [  0   1]]\n",
      "------------\n",
      "Pneumonia:\n",
      "[[184  42]\n",
      " [  6   2]]\n",
      "------------\n",
      "Pneumothorax:\n",
      "[[179  47]\n",
      " [  3   5]]\n",
      "------------\n",
      "Enlarged Cardiomediastinum:\n",
      "[[84 41]\n",
      " [64 45]]\n",
      "------------\n",
      "Lung Opacity:\n",
      "[[95 13]\n",
      " [29 97]]\n",
      "------------\n",
      "Lung Lesion:\n",
      "[[150  83]\n",
      " [  1   0]]\n",
      "------------\n",
      "Fracture:\n",
      "[[164  70]\n",
      " [  0   0]]\n",
      "------------\n",
      "Support Devices:\n",
      "[[117  10]\n",
      " [ 26  81]]\n",
      "------------\n",
      "No Finding:\n",
      "[[129  67]\n",
      " [  3  35]]\n",
      "------------\n"
     ]
    }
   ],
   "source": [
    "confusion_matrix_inceptionresnetv2 = multilabel_confusion_matrix(\n",
    "    y_true=test_datagenerator.labels,\n",
    "    y_pred=predictions_inceptionresnetv2_model\n",
    ")\n",
    "\n",
    "print(\"Inception-ResNet-V2:\\n\\n\")\n",
    "for i in range(len(labels)):\n",
    "    print(f\"{labels[i]}:\")\n",
    "    print(confusion_matrix_inceptionresnetv2[i, :, :])\n",
    "    print(\"------------\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d2c5c5ac",
   "metadata": {},
   "source": [
    "##### 2.3 Receiver Operating Characteristic curve (ROC Curve)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "b4a460fb",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Error in generating ROC curve for Fracture. Dataset lacks enough examples.\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 936x936 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "auc_roc_values = []\n",
    "\n",
    "for i in range(len(labels)):\n",
    "    try:\n",
    "        y_true = test_datagenerator.labels[:, i]\n",
    "        y_pred = predictions_inceptionresnetv2_model_probabilities[:, i]\n",
    "        \n",
    "        auc_roc = roc_auc_score(y_true=y_true, y_score=y_pred)\n",
    "        auc_roc_values.append(auc_roc)\n",
    "        \n",
    "        false_positive_rates, true_positive_rates, _ = roc_curve(y_true=y_true, y_score=y_pred)\n",
    "\n",
    "        plt.figure(1, figsize=(13, 13))\n",
    "        plt.plot([0, 1], [0, 1], 'k--')\n",
    "        plt.plot(\n",
    "            false_positive_rates,\n",
    "            true_positive_rates,\n",
    "            label=labels[i] + \" (\" + str(round(auc_roc, 4)) + \")\"\n",
    "        )\n",
    "        plt.xlabel('False positive rate')\n",
    "        plt.ylabel('True positive rate')\n",
    "        plt.title('ROC curve - Inception-ResNet-V2 model')\n",
    "        plt.legend(loc='best')\n",
    "    except:\n",
    "        print(\n",
    "            f\"Error in generating ROC curve for {labels[i]}. \"\n",
    "            f\"Dataset lacks enough examples.\"\n",
    "        )\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "160c271e",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[0.744724025974026,\n",
       " 0.7822466335931963,\n",
       " 0.9057741595055028,\n",
       " 0.9039388594944151,\n",
       " 0.9066047010456698,\n",
       " 0.7811158798283262,\n",
       " 0.6366150442477876,\n",
       " 0.8125,\n",
       " 0.5373211009174312,\n",
       " 0.9159318048206937,\n",
       " 0.16309012875536477,\n",
       " 0.9241298108764442,\n",
       " 0.8692266380236305]"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "auc_roc_values"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "355267c8",
   "metadata": {},
   "source": [
    "#### 3- Ensemble model"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "94e03c6a",
   "metadata": {},
   "source": [
    "##### 3.1 Classification Report"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "a81fe731",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                            precision    recall  f1-score   support\n",
      "\n",
      "               Atelectasis     0.5634    0.5000    0.5298        80\n",
      "              Cardiomegaly     0.6667    0.4118    0.5091        68\n",
      "             Consolidation     0.4032    0.7576    0.5263        33\n",
      "                     Edema     0.6327    0.6889    0.6596        45\n",
      "          Pleural Effusion     0.7966    0.7015    0.7460        67\n",
      "             Pleural Other     0.0182    1.0000    0.0357         1\n",
      "                 Pneumonia     0.0769    0.3750    0.1277         8\n",
      "              Pneumothorax     0.1111    0.5000    0.1818         8\n",
      "Enlarged Cardiomediastinum     0.5062    0.3761    0.4316       109\n",
      "              Lung Opacity     0.8919    0.7857    0.8354       126\n",
      "               Lung Lesion     0.0000    0.0000    0.0000         1\n",
      "                  Fracture     0.0000    0.0000    0.0000         0\n",
      "           Support Devices     0.8788    0.8131    0.8447       107\n",
      "                No Finding     0.3684    0.9211    0.5263        38\n",
      "\n",
      "                 micro avg     0.4632    0.6382    0.5368       691\n",
      "                 macro avg     0.4224    0.5593    0.4253       691\n",
      "              weighted avg     0.6695    0.6382    0.6356       691\n",
      "               samples avg     0.4670    0.6547    0.4971       691\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/tamer/anaconda3/envs/chexpert/lib/python3.7/site-packages/sklearn/metrics/_classification.py:1248: UndefinedMetricWarning: Recall and F-score are ill-defined and being set to 0.0 in labels with no true samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n",
      "/home/tamer/anaconda3/envs/chexpert/lib/python3.7/site-packages/sklearn/metrics/_classification.py:1248: 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": [
    "report_ensemble_model = classification_report(\n",
    "    y_true=test_datagenerator.labels,\n",
    "    y_pred=predictions_ensemble_model,\n",
    "    digits=4,\n",
    "    target_names=labels\n",
    ")\n",
    "\n",
    "print(report_ensemble_model)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "918f9e30",
   "metadata": {},
   "source": [
    "##### 3.2 Confusion Matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "0b33da83",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Ensemble:\n",
      "\n",
      "\n",
      "Atelectasis:\n",
      "[[123  31]\n",
      " [ 40  40]]\n",
      "------------\n",
      "Cardiomegaly:\n",
      "[[152  14]\n",
      " [ 40  28]]\n",
      "------------\n",
      "Consolidation:\n",
      "[[164  37]\n",
      " [  8  25]]\n",
      "------------\n",
      "Edema:\n",
      "[[171  18]\n",
      " [ 14  31]]\n",
      "------------\n",
      "Pleural Effusion:\n",
      "[[155  12]\n",
      " [ 20  47]]\n",
      "------------\n",
      "Pleural Other:\n",
      "[[179  54]\n",
      " [  0   1]]\n",
      "------------\n",
      "Pneumonia:\n",
      "[[190  36]\n",
      " [  5   3]]\n",
      "------------\n",
      "Pneumothorax:\n",
      "[[194  32]\n",
      " [  4   4]]\n",
      "------------\n",
      "Enlarged Cardiomediastinum:\n",
      "[[85 40]\n",
      " [68 41]]\n",
      "------------\n",
      "Lung Opacity:\n",
      "[[96 12]\n",
      " [27 99]]\n",
      "------------\n",
      "Lung Lesion:\n",
      "[[151  82]\n",
      " [  1   0]]\n",
      "------------\n",
      "Fracture:\n",
      "[[163  71]\n",
      " [  0   0]]\n",
      "------------\n",
      "Support Devices:\n",
      "[[115  12]\n",
      " [ 20  87]]\n",
      "------------\n",
      "No Finding:\n",
      "[[136  60]\n",
      " [  3  35]]\n",
      "------------\n"
     ]
    }
   ],
   "source": [
    "confusion_matrix_ensemble = multilabel_confusion_matrix(\n",
    "    y_true=test_datagenerator.labels,\n",
    "    y_pred=predictions_ensemble_model\n",
    ")\n",
    "\n",
    "print(\"Ensemble:\\n\\n\")\n",
    "for i in range(len(labels)):\n",
    "    print(f\"{labels[i]}:\")\n",
    "    print(confusion_matrix_ensemble[i, :, :])\n",
    "    print(\"------------\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "45cd1451",
   "metadata": {},
   "source": [
    "##### 3.3 Receiver Operating Characteristic curve (ROC Curve)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "4758be26",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Error in generating ROC curve for Fracture. Dataset lacks enough examples.\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 936x936 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "auc_roc_values = []\n",
    "\n",
    "for i in range(len(labels)):\n",
    "    try:\n",
    "        y_true = test_datagenerator.labels[:, i]\n",
    "        y_pred = predictions_ensemble_model_probabilities[:, i]\n",
    "        \n",
    "        auc_roc = roc_auc_score(y_true=y_true, y_score=y_pred)\n",
    "        auc_roc_values.append(auc_roc)\n",
    "        \n",
    "        false_positive_rates, true_positive_rates, _ = roc_curve(y_true=y_true, y_score=y_pred)\n",
    "\n",
    "        plt.figure(1, figsize=(13, 13))\n",
    "        plt.plot([0, 1], [0, 1], 'k--')\n",
    "        plt.plot(\n",
    "            false_positive_rates,\n",
    "            true_positive_rates,\n",
    "            label=labels[i] + \" (\" + str(round(auc_roc, 4)) + \")\"\n",
    "        )\n",
    "        plt.xlabel('False positive rate')\n",
    "        plt.ylabel('True positive rate')\n",
    "        plt.title('ROC curve - Ensemble model')\n",
    "        plt.legend(loc='best')\n",
    "    except:\n",
    "        print(\n",
    "            f\"Error in generating ROC curve for {labels[i]}. \"\n",
    "            f\"Dataset lacks enough examples.\"\n",
    "        )\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "9d012042",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[0.750487012987013,\n",
       " 0.776665485471297,\n",
       " 0.8955223880597015,\n",
       " 0.8998236331569665,\n",
       " 0.9192957368844401,\n",
       " 0.8197424892703863,\n",
       " 0.6858407079646018,\n",
       " 0.8014380530973452,\n",
       " 0.5542752293577982,\n",
       " 0.9139476778365666,\n",
       " 0.12446351931330468,\n",
       " 0.9384796526602399,\n",
       " 0.8649301825993555]"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "auc_roc_values"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "37a5edff",
   "metadata": {},
   "source": [
    "### CheXpert competition labels (first 5 labels only)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "d923c447",
   "metadata": {},
   "outputs": [],
   "source": [
    "# The dataset was modified as to make the first five labels the chexpert competition labels\n",
    "labels = y_cols[:5]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "7ccd24d7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['Atelectasis', 'Cardiomegaly', 'Consolidation', 'Edema', 'Pleural Effusion']"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "labels"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "103c59e2",
   "metadata": {},
   "source": [
    "#### 1- DenseNet-201 model"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "73e6d38c",
   "metadata": {},
   "source": [
    "##### 1.1 Classification Report"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "7afd0744",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                  precision    recall  f1-score   support\n",
      "\n",
      "     Atelectasis     0.5429    0.4750    0.5067        80\n",
      "    Cardiomegaly     0.6136    0.3971    0.4821        68\n",
      "   Consolidation     0.3860    0.6667    0.4889        33\n",
      "           Edema     0.5556    0.6667    0.6061        45\n",
      "Pleural Effusion     0.7424    0.7313    0.7368        67\n",
      "\n",
      "       micro avg     0.5704    0.5666    0.5685       293\n",
      "       macro avg     0.5681    0.5873    0.5641       293\n",
      "    weighted avg     0.5892    0.5666    0.5669       293\n",
      "     samples avg     0.3386    0.3129    0.3043       293\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/tamer/anaconda3/envs/chexpert/lib/python3.7/site-packages/sklearn/metrics/_classification.py:1248: 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/tamer/anaconda3/envs/chexpert/lib/python3.7/site-packages/sklearn/metrics/_classification.py:1248: 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": [
    "report_densenet201_model = classification_report(\n",
    "    y_true=test_datagenerator.labels[:, :5],\n",
    "    y_pred=predictions_densenet201_model[:, :5],\n",
    "    digits=4,\n",
    "    target_names=labels\n",
    ")\n",
    "\n",
    "print(report_densenet201_model)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "385c77cd",
   "metadata": {},
   "source": [
    "##### 1.2 Confusion Matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "048e994c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "DenseNet-201\n",
      "\n",
      "\n",
      "Atelectasis:\n",
      "[[122  32]\n",
      " [ 42  38]]\n",
      "------------\n",
      "Cardiomegaly:\n",
      "[[149  17]\n",
      " [ 41  27]]\n",
      "------------\n",
      "Consolidation:\n",
      "[[166  35]\n",
      " [ 11  22]]\n",
      "------------\n",
      "Edema:\n",
      "[[165  24]\n",
      " [ 15  30]]\n",
      "------------\n",
      "Pleural Effusion:\n",
      "[[150  17]\n",
      " [ 18  49]]\n",
      "------------\n"
     ]
    }
   ],
   "source": [
    "confusion_matrix_densenet201 = multilabel_confusion_matrix(\n",
    "    y_true=test_datagenerator.labels[:, :5],\n",
    "    y_pred=predictions_densenet201_model[:, :5]\n",
    ")\n",
    "\n",
    "print(\"DenseNet-201\\n\\n\")\n",
    "for i in range(len(labels)):\n",
    "    print(f\"{labels[i]}:\")\n",
    "    print(confusion_matrix_densenet201[i, :, :])\n",
    "    print(\"------------\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0ef4a2a3",
   "metadata": {},
   "source": [
    "##### 1.3 Receiver Operating Characteristic curve (ROC Curve)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "bc267670",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 936x936 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "auc_roc_values = []\n",
    "\n",
    "for i in range(len(labels)):\n",
    "    try:\n",
    "        y_true = test_datagenerator.labels[:, i]\n",
    "        y_pred = predictions_densenet201_model_probabilities[:, i]\n",
    "        \n",
    "        auc_roc = roc_auc_score(y_true=y_true, y_score=y_pred)\n",
    "        auc_roc_values.append(auc_roc)\n",
    "        \n",
    "        false_positive_rates, true_positive_rates, _ = roc_curve(y_true=y_true, y_score=y_pred)\n",
    "\n",
    "        plt.figure(1, figsize=(13, 13))\n",
    "        plt.plot([0, 1], [0, 1], 'k--')\n",
    "        plt.plot(\n",
    "            false_positive_rates,\n",
    "            true_positive_rates,\n",
    "            label=labels[i] + \" (\" + str(round(auc_roc, 4)) + \")\"\n",
    "        )\n",
    "        plt.xlabel('False positive rate')\n",
    "        plt.ylabel('True positive rate')\n",
    "        plt.title('ROC curve - DenseNet-201 model')\n",
    "        plt.legend(loc='best')\n",
    "    except:\n",
    "        print(\n",
    "            f\"Error in generating ROC curve for {labels[i]}. \"\n",
    "            f\"Dataset lacks enough examples.\"\n",
    "        )\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "79b4c060",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[0.7436688311688313,\n",
       " 0.7431785967399007,\n",
       " 0.8599427106889792,\n",
       " 0.883480305702528,\n",
       " 0.9154526767360801]"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "auc_roc_values"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6aa905ed",
   "metadata": {},
   "source": [
    "#### 2- Inception-ResNet-V2 model"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "caefb6a7",
   "metadata": {},
   "source": [
    "##### 2.1 Classification Report"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "7afd4937",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                            precision    recall  f1-score   support\n",
      "\n",
      "               Atelectasis     0.5634    0.5000    0.5298        80\n",
      "              Cardiomegaly     0.6667    0.4118    0.5091        68\n",
      "             Consolidation     0.4032    0.7576    0.5263        33\n",
      "                     Edema     0.6327    0.6889    0.6596        45\n",
      "          Pleural Effusion     0.7966    0.7015    0.7460        67\n",
      "             Pleural Other     0.0182    1.0000    0.0357         1\n",
      "                 Pneumonia     0.0769    0.3750    0.1277         8\n",
      "              Pneumothorax     0.1111    0.5000    0.1818         8\n",
      "Enlarged Cardiomediastinum     0.5062    0.3761    0.4316       109\n",
      "              Lung Opacity     0.8919    0.7857    0.8354       126\n",
      "               Lung Lesion     0.0000    0.0000    0.0000         1\n",
      "                  Fracture     0.0000    0.0000    0.0000         0\n",
      "           Support Devices     0.8788    0.8131    0.8447       107\n",
      "                No Finding     0.3684    0.9211    0.5263        38\n",
      "\n",
      "                 micro avg     0.4632    0.6382    0.5368       691\n",
      "                 macro avg     0.4224    0.5593    0.4253       691\n",
      "              weighted avg     0.6695    0.6382    0.6356       691\n",
      "               samples avg     0.4670    0.6547    0.4971       691\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/tamer/anaconda3/envs/chexpert/lib/python3.7/site-packages/sklearn/metrics/_classification.py:1248: 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/tamer/anaconda3/envs/chexpert/lib/python3.7/site-packages/sklearn/metrics/_classification.py:1248: 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": [
    "report_inceptionresnetv2_model = classification_report(\n",
    "    y_true=test_datagenerator.labels[:, :5],\n",
    "    y_pred=predictions_inceptionresnetv2_model[:, :5],\n",
    "    digits=4,\n",
    "    target_names=labels\n",
    ")\n",
    "\n",
    "print(report_ensemble_model)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "88f62a53",
   "metadata": {},
   "source": [
    "##### 2.2 Confusion Matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "52d34d11",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Inception-ResNet-V2:\n",
      "\n",
      "\n",
      "Atelectasis:\n",
      "[[130  24]\n",
      " [ 41  39]]\n",
      "------------\n",
      "Cardiomegaly:\n",
      "[[151  15]\n",
      " [ 36  32]]\n",
      "------------\n",
      "Consolidation:\n",
      "[[163  38]\n",
      " [  5  28]]\n",
      "------------\n",
      "Edema:\n",
      "[[168  21]\n",
      " [ 11  34]]\n",
      "------------\n",
      "Pleural Effusion:\n",
      "[[155  12]\n",
      " [ 22  45]]\n",
      "------------\n"
     ]
    }
   ],
   "source": [
    "confusion_matrix_inceptionresnetv2 = multilabel_confusion_matrix(\n",
    "    y_true=test_datagenerator.labels[:, :5],\n",
    "    y_pred=predictions_inceptionresnetv2_model[:, :5]\n",
    ")\n",
    "\n",
    "print(\"Inception-ResNet-V2:\\n\\n\")\n",
    "for i in range(len(labels)):\n",
    "    print(f\"{labels[i]}:\")\n",
    "    print(confusion_matrix_inceptionresnetv2[i, :, :])\n",
    "    print(\"------------\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "30e75a92",
   "metadata": {},
   "source": [
    "##### 2.3 Receiver Operating Characteristic curve (ROC Curve)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "7f27bef5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 936x936 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "auc_roc_values = []\n",
    "\n",
    "for i in range(len(labels)):\n",
    "    try:\n",
    "        y_true = test_datagenerator.labels[:, i]\n",
    "        y_pred = predictions_inceptionresnetv2_model_probabilities[:, i]\n",
    "        \n",
    "        auc_roc = roc_auc_score(y_true=y_true, y_score=y_pred)\n",
    "        auc_roc_values.append(auc_roc)\n",
    "        \n",
    "        false_positive_rates, true_positive_rates, _ = roc_curve(y_true=y_true, y_score=y_pred)\n",
    "\n",
    "        plt.figure(1, figsize=(13, 13))\n",
    "        plt.plot([0, 1], [0, 1], 'k--')\n",
    "        plt.plot(\n",
    "            false_positive_rates,\n",
    "            true_positive_rates,\n",
    "            label=labels[i] + \" (\" + str(round(auc_roc, 4)) + \")\"\n",
    "        )\n",
    "        plt.xlabel('False positive rate')\n",
    "        plt.ylabel('True positive rate')\n",
    "        plt.title('ROC curve - Inception-ResNet-V2 model')\n",
    "        plt.legend(loc='best')\n",
    "    except:\n",
    "        print(\n",
    "            f\"Error in generating ROC curve for {labels[i]}. \"\n",
    "            f\"Dataset lacks enough examples.\"\n",
    "        )\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "239aea4e",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[0.744724025974026,\n",
       " 0.7822466335931963,\n",
       " 0.9057741595055028,\n",
       " 0.9039388594944151,\n",
       " 0.9066047010456698]"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "auc_roc_values"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a38d601e",
   "metadata": {},
   "source": [
    "#### 3- Ensemble model"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cc56b11f",
   "metadata": {},
   "source": [
    "##### 3.1 Classification Report"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "0310d664",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                  precision    recall  f1-score   support\n",
      "\n",
      "     Atelectasis     0.5634    0.5000    0.5298        80\n",
      "    Cardiomegaly     0.6667    0.4118    0.5091        68\n",
      "   Consolidation     0.4032    0.7576    0.5263        33\n",
      "           Edema     0.6327    0.6889    0.6596        45\n",
      "Pleural Effusion     0.7966    0.7015    0.7460        67\n",
      "\n",
      "       micro avg     0.6042    0.5836    0.5938       293\n",
      "       macro avg     0.6125    0.6119    0.5942       293\n",
      "    weighted avg     0.6333    0.5836    0.5940       293\n",
      "     samples avg     0.3484    0.3250    0.3160       293\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/tamer/anaconda3/envs/chexpert/lib/python3.7/site-packages/sklearn/metrics/_classification.py:1248: 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/tamer/anaconda3/envs/chexpert/lib/python3.7/site-packages/sklearn/metrics/_classification.py:1248: 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": [
    "report_ensemble_model = classification_report(\n",
    "    y_true=test_datagenerator.labels[:, :5],\n",
    "    y_pred=predictions_ensemble_model[:, :5],\n",
    "    digits=4,\n",
    "    target_names=labels\n",
    ")\n",
    "\n",
    "print(report_ensemble_model)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "570abf83",
   "metadata": {},
   "source": [
    "##### 3.2 Confusion Matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "7ed89ed5",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Ensemble:\n",
      "\n",
      "\n",
      "Atelectasis:\n",
      "[[123  31]\n",
      " [ 40  40]]\n",
      "------------\n",
      "Cardiomegaly:\n",
      "[[152  14]\n",
      " [ 40  28]]\n",
      "------------\n",
      "Consolidation:\n",
      "[[164  37]\n",
      " [  8  25]]\n",
      "------------\n",
      "Edema:\n",
      "[[171  18]\n",
      " [ 14  31]]\n",
      "------------\n",
      "Pleural Effusion:\n",
      "[[155  12]\n",
      " [ 20  47]]\n",
      "------------\n"
     ]
    }
   ],
   "source": [
    "confusion_matrix_ensemble = multilabel_confusion_matrix(\n",
    "    y_true=test_datagenerator.labels[:, :5],\n",
    "    y_pred=predictions_ensemble_model[:, :5]\n",
    ")\n",
    "\n",
    "\n",
    "print(\"Ensemble:\\n\\n\")\n",
    "for i in range(len(labels)):\n",
    "    print(f\"{labels[i]}:\")\n",
    "    print(confusion_matrix_ensemble[i, :, :])\n",
    "    print(\"------------\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b1d008f2",
   "metadata": {},
   "source": [
    "##### 3.3 Receiver Operating Characteristic curve (ROC Curve)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "f8ff1c5c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 936x936 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "auc_roc_values = []\n",
    "\n",
    "for i in range(len(labels)):\n",
    "    try:\n",
    "        y_true = test_datagenerator.labels[:, i]\n",
    "        y_pred = predictions_ensemble_model_probabilities[:, i]\n",
    "        \n",
    "        auc_roc = roc_auc_score(y_true=y_true, y_score=y_pred)\n",
    "        auc_roc_values.append(auc_roc)\n",
    "        \n",
    "        false_positive_rates, true_positive_rates, _ = roc_curve(y_true=y_true, y_score=y_pred)\n",
    "\n",
    "        plt.figure(1, figsize=(13, 13))\n",
    "        plt.plot([0, 1], [0, 1], 'k--')\n",
    "        plt.plot(\n",
    "            false_positive_rates,\n",
    "            true_positive_rates,\n",
    "            label=labels[i] + \" (\" + str(round(auc_roc, 4)) + \")\"\n",
    "        )\n",
    "        plt.xlabel('False positive rate')\n",
    "        plt.ylabel('True positive rate')\n",
    "        plt.title('ROC curve - Ensemble model')\n",
    "        plt.legend(loc='best')\n",
    "    except:\n",
    "        print(\n",
    "            f\"Error in generating ROC curve for {labels[i]}. \"\n",
    "            f\"Dataset lacks enough examples.\"\n",
    "        )\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "c8a30302",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[0.750487012987013,\n",
       " 0.776665485471297,\n",
       " 0.8955223880597015,\n",
       " 0.8998236331569665,\n",
       " 0.9192957368844401]"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "auc_roc_values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "474ed616",
   "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": "3.7.11"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}