{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "papermill": {
     "duration": 0.017047,
     "end_time": "2021-01-13T19:04:02.236885",
     "exception": false,
     "start_time": "2021-01-13T19:04:02.219838",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "# Import Statements"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "_cell_guid": "79c7e3d0-c299-4dcb-8224-4455121ee9b0",
    "_uuid": "d629ff2d2480ee46fbb7e2d37f6b5fab8052498a",
    "execution": {
     "iopub.execute_input": "2021-01-13T19:04:02.274589Z",
     "iopub.status.busy": "2021-01-13T19:04:02.273851Z",
     "iopub.status.idle": "2021-01-13T19:04:10.359394Z",
     "shell.execute_reply": "2021-01-13T19:04:10.358181Z"
    },
    "papermill": {
     "duration": 8.107497,
     "end_time": "2021-01-13T19:04:10.359532",
     "exception": false,
     "start_time": "2021-01-13T19:04:02.252035",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "import os\n",
    "import keras \n",
    "from keras.models import Sequential\n",
    "from keras.layers import Conv2D, MaxPooling2D, Flatten, Dense, Dropout, BatchNormalization\n",
    "from PIL import Image\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.preprocessing import OneHotEncoder "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "papermill": {
     "duration": 0.013964,
     "end_time": "2021-01-13T19:04:10.387998",
     "exception": false,
     "start_time": "2021-01-13T19:04:10.374034",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "# One Hot Encoding the Target Classes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-01-13T19:04:10.426343Z",
     "iopub.status.busy": "2021-01-13T19:04:10.425663Z",
     "iopub.status.idle": "2021-01-13T19:04:10.432273Z",
     "shell.execute_reply": "2021-01-13T19:04:10.431739Z"
    },
    "papermill": {
     "duration": 0.030157,
     "end_time": "2021-01-13T19:04:10.432405",
     "exception": false,
     "start_time": "2021-01-13T19:04:10.402248",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "OneHotEncoder()"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "encoder = OneHotEncoder()\n",
    "encoder.fit([[0], [1]]) \n",
    "\n",
    "# 0 - Tumor\n",
    "# 1 - Normal"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "papermill": {
     "duration": 0.014519,
     "end_time": "2021-01-13T19:04:10.462278",
     "exception": false,
     "start_time": "2021-01-13T19:04:10.447759",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "# Creating 3 Important Lists --\n",
    "1. data list for storing image data in numpy array form\n",
    "2. paths list for storing paths of all images\n",
    "3. result list for storing one hot encoded form of target class whether normal or tumor"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-01-13T19:04:10.501611Z",
     "iopub.status.busy": "2021-01-13T19:04:10.500916Z",
     "iopub.status.idle": "2021-01-13T19:04:11.944664Z",
     "shell.execute_reply": "2021-01-13T19:04:11.942997Z"
    },
    "papermill": {
     "duration": 1.467737,
     "end_time": "2021-01-13T19:04:11.944810",
     "exception": false,
     "start_time": "2021-01-13T19:04:10.477073",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "# This cell updates result list for images with tumor\n",
    "\n",
    "data = []\n",
    "paths = []\n",
    "result = []\n",
    "\n",
    "for r, d, f in os.walk(r'../input/brain-mri-images-for-brain-tumor-detection/yes'):\n",
    "    for file in f:\n",
    "        if '.jpg' in file:\n",
    "            paths.append(os.path.join(r, file))\n",
    "\n",
    "for path in paths:\n",
    "    img = Image.open(path)\n",
    "    img = img.resize((128,128))\n",
    "    img = np.array(img)\n",
    "    if(img.shape == (128,128,3)):\n",
    "        data.append(np.array(img))\n",
    "        result.append(encoder.transform([[0]]).toarray())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-01-13T19:04:11.985224Z",
     "iopub.status.busy": "2021-01-13T19:04:11.984515Z",
     "iopub.status.idle": "2021-01-13T19:04:13.038027Z",
     "shell.execute_reply": "2021-01-13T19:04:13.037399Z"
    },
    "papermill": {
     "duration": 1.078081,
     "end_time": "2021-01-13T19:04:13.038146",
     "exception": false,
     "start_time": "2021-01-13T19:04:11.960065",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "# This cell updates result list for images without tumor\n",
    "\n",
    "paths = []\n",
    "for r, d, f in os.walk(r\"../input/brain-mri-images-for-brain-tumor-detection/no\"):\n",
    "    for file in f:\n",
    "        if '.jpg' in file:\n",
    "            paths.append(os.path.join(r, file))\n",
    "\n",
    "for path in paths:\n",
    "    img = Image.open(path)\n",
    "    img = img.resize((128,128))\n",
    "    img = np.array(img)\n",
    "    if(img.shape == (128,128,3)):\n",
    "        data.append(np.array(img))\n",
    "        result.append(encoder.transform([[1]]).toarray())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-01-13T19:04:13.073556Z",
     "iopub.status.busy": "2021-01-13T19:04:13.072843Z",
     "iopub.status.idle": "2021-01-13T19:04:13.086117Z",
     "shell.execute_reply": "2021-01-13T19:04:13.085426Z"
    },
    "papermill": {
     "duration": 0.031647,
     "end_time": "2021-01-13T19:04:13.086256",
     "exception": false,
     "start_time": "2021-01-13T19:04:13.054609",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(139, 128, 128, 3)"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data = np.array(data)\n",
    "data.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-01-13T19:04:13.123027Z",
     "iopub.status.busy": "2021-01-13T19:04:13.122015Z",
     "iopub.status.idle": "2021-01-13T19:04:13.126144Z",
     "shell.execute_reply": "2021-01-13T19:04:13.126733Z"
    },
    "papermill": {
     "duration": 0.025052,
     "end_time": "2021-01-13T19:04:13.126878",
     "exception": false,
     "start_time": "2021-01-13T19:04:13.101826",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "result = np.array(result)\n",
    "result = result.reshape(139,2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "papermill": {
     "duration": 0.015536,
     "end_time": "2021-01-13T19:04:13.158356",
     "exception": false,
     "start_time": "2021-01-13T19:04:13.142820",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "# Splitting the Data into Training & Testing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-01-13T19:04:13.196149Z",
     "iopub.status.busy": "2021-01-13T19:04:13.195442Z",
     "iopub.status.idle": "2021-01-13T19:04:13.200535Z",
     "shell.execute_reply": "2021-01-13T19:04:13.201076Z"
    },
    "papermill": {
     "duration": 0.027054,
     "end_time": "2021-01-13T19:04:13.201267",
     "exception": false,
     "start_time": "2021-01-13T19:04:13.174213",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "x_train,x_test,y_train,y_test = train_test_split(data, result, test_size=0.2, shuffle=True, random_state=0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "papermill": {
     "duration": 0.015584,
     "end_time": "2021-01-13T19:04:13.232970",
     "exception": false,
     "start_time": "2021-01-13T19:04:13.217386",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "# Model Building"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "papermill": {
     "duration": 0.015538,
     "end_time": "2021-01-13T19:04:13.264319",
     "exception": false,
     "start_time": "2021-01-13T19:04:13.248781",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "Batch normalization is a technique for training very deep neural networks that standardizes the inputs to a layer for each mini-batch. This has the effect of stabilizing the learning process and dramatically reducing the number of training epochs required to train deep networks."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-01-13T19:04:13.299474Z",
     "iopub.status.busy": "2021-01-13T19:04:13.298760Z",
     "iopub.status.idle": "2021-01-13T19:04:13.898376Z",
     "shell.execute_reply": "2021-01-13T19:04:13.897760Z"
    },
    "papermill": {
     "duration": 0.618409,
     "end_time": "2021-01-13T19:04:13.898494",
     "exception": false,
     "start_time": "2021-01-13T19:04:13.280085",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "conv2d (Conv2D)              (None, 128, 128, 32)      416       \n",
      "_________________________________________________________________\n",
      "conv2d_1 (Conv2D)            (None, 128, 128, 32)      4128      \n",
      "_________________________________________________________________\n",
      "batch_normalization (BatchNo (None, 128, 128, 32)      128       \n",
      "_________________________________________________________________\n",
      "max_pooling2d (MaxPooling2D) (None, 64, 64, 32)        0         \n",
      "_________________________________________________________________\n",
      "dropout (Dropout)            (None, 64, 64, 32)        0         \n",
      "_________________________________________________________________\n",
      "conv2d_2 (Conv2D)            (None, 64, 64, 64)        8256      \n",
      "_________________________________________________________________\n",
      "conv2d_3 (Conv2D)            (None, 64, 64, 64)        16448     \n",
      "_________________________________________________________________\n",
      "batch_normalization_1 (Batch (None, 64, 64, 64)        256       \n",
      "_________________________________________________________________\n",
      "max_pooling2d_1 (MaxPooling2 (None, 32, 32, 64)        0         \n",
      "_________________________________________________________________\n",
      "dropout_1 (Dropout)          (None, 32, 32, 64)        0         \n",
      "_________________________________________________________________\n",
      "flatten (Flatten)            (None, 65536)             0         \n",
      "_________________________________________________________________\n",
      "dense (Dense)                (None, 512)               33554944  \n",
      "_________________________________________________________________\n",
      "dropout_2 (Dropout)          (None, 512)               0         \n",
      "_________________________________________________________________\n",
      "dense_1 (Dense)              (None, 2)                 1026      \n",
      "=================================================================\n",
      "Total params: 33,585,602\n",
      "Trainable params: 33,585,410\n",
      "Non-trainable params: 192\n",
      "_________________________________________________________________\n",
      "None\n"
     ]
    }
   ],
   "source": [
    "model = Sequential()\n",
    "\n",
    "model.add(Conv2D(32, kernel_size=(2, 2), input_shape=(128, 128, 3), padding = 'Same'))\n",
    "model.add(Conv2D(32, kernel_size=(2, 2),  activation ='relu', padding = 'Same'))\n",
    "\n",
    "\n",
    "model.add(BatchNormalization())\n",
    "model.add(MaxPooling2D(pool_size=(2, 2)))\n",
    "model.add(Dropout(0.25))\n",
    "\n",
    "model.add(Conv2D(64, kernel_size = (2,2), activation ='relu', padding = 'Same'))\n",
    "model.add(Conv2D(64, kernel_size = (2,2), activation ='relu', padding = 'Same'))\n",
    "\n",
    "model.add(BatchNormalization())\n",
    "model.add(MaxPooling2D(pool_size=(2,2), strides=(2,2)))\n",
    "model.add(Dropout(0.25))\n",
    "\n",
    "model.add(Flatten())\n",
    "\n",
    "model.add(Dense(512, activation='relu'))\n",
    "model.add(Dropout(0.5))\n",
    "model.add(Dense(2, activation='softmax'))\n",
    "\n",
    "model.compile(loss = \"categorical_crossentropy\", optimizer='Adamax')\n",
    "print(model.summary())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-01-13T19:04:13.938797Z",
     "iopub.status.busy": "2021-01-13T19:04:13.938095Z",
     "iopub.status.idle": "2021-01-13T19:04:13.942425Z",
     "shell.execute_reply": "2021-01-13T19:04:13.941846Z"
    },
    "papermill": {
     "duration": 0.02587,
     "end_time": "2021-01-13T19:04:13.942537",
     "exception": false,
     "start_time": "2021-01-13T19:04:13.916667",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(111, 2)"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-01-13T19:04:13.983118Z",
     "iopub.status.busy": "2021-01-13T19:04:13.982398Z",
     "iopub.status.idle": "2021-01-13T19:06:37.972111Z",
     "shell.execute_reply": "2021-01-13T19:06:37.972721Z"
    },
    "papermill": {
     "duration": 144.013421,
     "end_time": "2021-01-13T19:06:37.972915",
     "exception": false,
     "start_time": "2021-01-13T19:04:13.959494",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/30\n",
      "3/3 [==============================] - 3s 1s/step - loss: 79.3326 - val_loss: 42.5277\n",
      "Epoch 2/30\n",
      "3/3 [==============================] - 7s 2s/step - loss: 17.7606 - val_loss: 70.0688\n",
      "Epoch 3/30\n",
      "3/3 [==============================] - 7s 2s/step - loss: 9.4687 - val_loss: 14.7581\n",
      "Epoch 4/30\n",
      "3/3 [==============================] - 3s 993ms/step - loss: 4.5975 - val_loss: 23.5701\n",
      "Epoch 5/30\n",
      "3/3 [==============================] - 3s 1s/step - loss: 3.9477 - val_loss: 16.4190\n",
      "Epoch 6/30\n",
      "3/3 [==============================] - 3s 933ms/step - loss: 1.3051 - val_loss: 4.8896\n",
      "Epoch 7/30\n",
      "3/3 [==============================] - 3s 932ms/step - loss: 1.0897 - val_loss: 1.0343\n",
      "Epoch 8/30\n",
      "3/3 [==============================] - 3s 954ms/step - loss: 0.8229 - val_loss: 0.9565\n",
      "Epoch 9/30\n",
      "3/3 [==============================] - 3s 924ms/step - loss: 0.8666 - val_loss: 2.8124\n",
      "Epoch 10/30\n",
      "3/3 [==============================] - 3s 956ms/step - loss: 0.6506 - val_loss: 4.4251\n",
      "Epoch 11/30\n",
      "3/3 [==============================] - 3s 922ms/step - loss: 0.1757 - val_loss: 4.8350\n",
      "Epoch 12/30\n",
      "3/3 [==============================] - 3s 1s/step - loss: 0.2227 - val_loss: 4.0335\n",
      "Epoch 13/30\n",
      "3/3 [==============================] - 3s 953ms/step - loss: 0.1199 - val_loss: 3.2715\n",
      "Epoch 14/30\n",
      "3/3 [==============================] - 3s 921ms/step - loss: 0.0681 - val_loss: 2.7098\n",
      "Epoch 15/30\n",
      "3/3 [==============================] - 3s 923ms/step - loss: 0.0685 - val_loss: 2.3377\n",
      "Epoch 16/30\n",
      "3/3 [==============================] - 3s 966ms/step - loss: 0.0570 - val_loss: 2.0321\n",
      "Epoch 17/30\n",
      "3/3 [==============================] - 3s 960ms/step - loss: 0.0945 - val_loss: 1.7880\n",
      "Epoch 18/30\n",
      "3/3 [==============================] - 3s 975ms/step - loss: 0.0337 - val_loss: 1.6038\n",
      "Epoch 19/30\n",
      "3/3 [==============================] - 3s 919ms/step - loss: 0.0818 - val_loss: 1.4881\n",
      "Epoch 20/30\n",
      "3/3 [==============================] - 3s 962ms/step - loss: 0.0600 - val_loss: 1.3945\n",
      "Epoch 21/30\n",
      "3/3 [==============================] - 3s 938ms/step - loss: 0.0184 - val_loss: 1.3546\n",
      "Epoch 22/30\n",
      "3/3 [==============================] - 3s 935ms/step - loss: 0.0137 - val_loss: 1.3589\n",
      "Epoch 23/30\n",
      "3/3 [==============================] - 3s 942ms/step - loss: 0.0241 - val_loss: 1.3308\n",
      "Epoch 24/30\n",
      "3/3 [==============================] - 3s 907ms/step - loss: 0.0117 - val_loss: 1.3155\n",
      "Epoch 25/30\n",
      "3/3 [==============================] - 3s 947ms/step - loss: 0.0809 - val_loss: 1.3080\n",
      "Epoch 26/30\n",
      "3/3 [==============================] - 3s 908ms/step - loss: 0.0297 - val_loss: 1.3482\n",
      "Epoch 27/30\n",
      "3/3 [==============================] - 3s 999ms/step - loss: 0.0259 - val_loss: 1.3704\n",
      "Epoch 28/30\n",
      "3/3 [==============================] - 3s 1s/step - loss: 0.0069 - val_loss: 1.3574\n",
      "Epoch 29/30\n",
      "3/3 [==============================] - 3s 939ms/step - loss: 0.0425 - val_loss: 1.3168\n",
      "Epoch 30/30\n",
      "3/3 [==============================] - 3s 952ms/step - loss: 0.0259 - val_loss: 1.2111\n"
     ]
    }
   ],
   "source": [
    "history = model.fit(x_train, y_train, epochs = 30, batch_size = 40, verbose = 1,validation_data = (x_test, y_test))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "papermill": {
     "duration": 0.068349,
     "end_time": "2021-01-13T19:06:38.108121",
     "exception": false,
     "start_time": "2021-01-13T19:06:38.039772",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "# Plotting Losses"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-01-13T19:06:38.247633Z",
     "iopub.status.busy": "2021-01-13T19:06:38.246928Z",
     "iopub.status.idle": "2021-01-13T19:06:38.495973Z",
     "shell.execute_reply": "2021-01-13T19:06:38.495448Z"
    },
    "papermill": {
     "duration": 0.31992,
     "end_time": "2021-01-13T19:06:38.496105",
     "exception": false,
     "start_time": "2021-01-13T19:06:38.176185",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(history.history['loss'])\n",
    "plt.plot(history.history['val_loss'])\n",
    "plt.title('Model Loss')\n",
    "plt.ylabel('Loss')\n",
    "plt.xlabel('Epoch')\n",
    "plt.legend(['Test', 'Validation'], loc='upper right')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "papermill": {
     "duration": 0.068951,
     "end_time": "2021-01-13T19:06:38.633552",
     "exception": false,
     "start_time": "2021-01-13T19:06:38.564601",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "# Just Checking the Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-01-13T19:06:38.778565Z",
     "iopub.status.busy": "2021-01-13T19:06:38.777465Z",
     "iopub.status.idle": "2021-01-13T19:06:38.789382Z",
     "shell.execute_reply": "2021-01-13T19:06:38.790323Z"
    },
    "papermill": {
     "duration": 0.087731,
     "end_time": "2021-01-13T19:06:38.790575",
     "exception": false,
     "start_time": "2021-01-13T19:06:38.702844",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "def names(number):\n",
    "    if number==0:\n",
    "        return 'Its a Tumor'\n",
    "    else:\n",
    "        return 'No, Its not a tumor'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-01-13T19:06:38.973252Z",
     "iopub.status.busy": "2021-01-13T19:06:38.972171Z",
     "iopub.status.idle": "2021-01-13T19:06:39.325364Z",
     "shell.execute_reply": "2021-01-13T19:06:39.325840Z"
    },
    "papermill": {
     "duration": 0.434388,
     "end_time": "2021-01-13T19:06:39.325993",
     "exception": false,
     "start_time": "2021-01-13T19:06:38.891605",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "99.97681975364685% Confidence This Is No, Its not a tumor\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from matplotlib.pyplot import imshow\n",
    "img = Image.open(r\"../input/brain-mri-images-for-brain-tumor-detection/no/N17.jpg\")\n",
    "x = np.array(img.resize((128,128)))\n",
    "x = x.reshape(1,128,128,3)\n",
    "res = model.predict_on_batch(x)\n",
    "classification = np.where(res == np.amax(res))[1][0]\n",
    "imshow(img)\n",
    "print(str(res[0][classification]*100) + '% Confidence This Is ' + names(classification))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-01-13T19:06:39.475664Z",
     "iopub.status.busy": "2021-01-13T19:06:39.474672Z",
     "iopub.status.idle": "2021-01-13T19:06:39.688377Z",
     "shell.execute_reply": "2021-01-13T19:06:39.688857Z"
    },
    "papermill": {
     "duration": 0.290742,
     "end_time": "2021-01-13T19:06:39.689010",
     "exception": false,
     "start_time": "2021-01-13T19:06:39.398268",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "97.33012318611145% Confidence This Is A Its a Tumor\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from matplotlib.pyplot import imshow\n",
    "img = Image.open(r\"../input/brain-mri-images-for-brain-tumor-detection/yes/Y3.jpg\")\n",
    "x = np.array(img.resize((128,128)))\n",
    "x = x.reshape(1,128,128,3)\n",
    "res = model.predict_on_batch(x)\n",
    "classification = np.where(res == np.amax(res))[1][0]\n",
    "imshow(img)\n",
    "print(str(res[0][classification]*100) + '% Confidence This Is A ' + names(classification))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "papermill": {
     "duration": 0.074409,
     "end_time": "2021-01-13T19:06:39.836265",
     "exception": false,
     "start_time": "2021-01-13T19:06:39.761856",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "# Thats It !!"
   ]
  }
 ],
 "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": "3.7.6"
  },
  "papermill": {
   "duration": 163.388087,
   "end_time": "2021-01-13T19:06:40.018519",
   "environment_variables": {},
   "exception": null,
   "input_path": "__notebook__.ipynb",
   "output_path": "__notebook__.ipynb",
   "parameters": {},
   "start_time": "2021-01-13T19:03:56.630432",
   "version": "2.1.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}