test.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/sharanya/snap/jupyter/common/lib/python3.7/site-packages/joblib/_multiprocessing_helpers.py:45: UserWarning: [Errno 13] Permission denied.  joblib will operate in serial mode\n",
      "  warnings.warn('%s.  joblib will operate in serial mode' % (e,))\n",
      "Using TensorFlow backend.\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "from sklearn.metrics import classification_report, confusion_matrix\n",
    "from keras.models import load_model\n",
    "from keras.preprocessing.image import img_to_array\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "from PIL import Image\n",
    "from keras.utils import np_utils\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "model = load_model('model.h5')\n",
    "\n",
    "x_test = np.load('testx.npy')\n",
    "y_test = np.load('testy.npy')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(587, 512, 512, 3)\n",
      "(587,)\n"
     ]
    }
   ],
   "source": [
    "print(x_test.shape)\n",
    "print(y_test.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "# RESIZING TEST IMAGES TO (300,300)\n",
    "\n",
    "image_height = 300\n",
    "image_width = 300\n",
    "channels = 3\n",
    "dataset = np.ndarray(shape=(len(x_test), image_height, image_width, channels),\n",
    "                     dtype=np.float16)\n",
    "cnt = 0\n",
    "for i in x_test:\n",
    "    im = Image.fromarray(i)\n",
    "    img = im.resize((image_height,image_width))\n",
    "    img.thumbnail((image_height, image_width))\n",
    "    x = img_to_array(img)\n",
    "    x = x.reshape((image_height, image_width, channels))\n",
    "    x = (x - 128.0) / 128.0\n",
    "    dataset[cnt] = x\n",
    "    cnt += 1\n",
    "    \n",
    "np.save('x_test_resized.npy',dataset)\n",
    "x_test = np.load('x_test_resized.npy')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(587, 300, 300, 3)"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x_test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "encoder = LabelEncoder()\n",
    "encoder.fit(y_test)\n",
    "encoded_Y = encoder.transform(y_test)\n",
    "y_test_enc = np_utils.to_categorical(encoded_Y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "# Evaluate on test data\n",
      "587/587 [==============================] - 13s 22ms/step\n",
      "test loss, test acc: [0.2197212950664444, 0.9165247082710266, 0.9165247082710266]\n"
     ]
    }
   ],
   "source": [
    "print('\\n# Evaluate on test data')\n",
    "results = model.evaluate(x_test, y_test_enc, batch_size=128)\n",
    "print('test loss, test acc:', results)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " Generate predictions for test samples\n"
     ]
    }
   ],
   "source": [
    "print('\\n Generate predictions for test samples')\n",
    "y_pred = model.predict(x_test_resized)\n",
    "Y_pred = np.argmax(y_pred, axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Confusion Matrix\n",
      "[[ 33   3  13]\n",
      " [  0 248  21]\n",
      " [  0  12 257]]\n"
     ]
    }
   ],
   "source": [
    "print('Confusion Matrix')\n",
    "print(confusion_matrix(y_test, Y_pred))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Classification Report\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "       covid       1.00      0.67      0.80        49\n",
      "      normal       0.94      0.92      0.93       269\n",
      "   pneumonia       0.88      0.96      0.92       269\n",
      "\n",
      "    accuracy                           0.92       587\n",
      "   macro avg       0.94      0.85      0.89       587\n",
      "weighted avg       0.92      0.92      0.92       587\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print('Classification Report')\n",
    "target_names = ['covid', 'normal', 'pneumonia']\n",
    "print(classification_report(y_test, Y_pred, target_names=target_names))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "encoder = LabelEncoder()\n",
    "encoder.fit(Y_pred)\n",
    "encoded_Y = encoder.transform(Y_pred)\n",
    "y_pred_enc = np_utils.to_categorical(encoded_Y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "ROC-AUC Score:\n",
      "\n",
      "0.8994507929191643\n"
     ]
    },
    {
     "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": [
    "from sklearn.metrics import roc_auc_score\n",
    "print(\"\\nROC-AUC Score:\\n\")\n",
    "print(roc_auc_score(y_test_enc,y_pred_enc))\n",
    "import matplotlib.pyplot as plt\n",
    "import scikitplot as skplt\n",
    "skplt.metrics.plot_roc(y_test, y_pred)\n",
    "plt.show()"
   ]
  },
  {
   "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": "3.7.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}