Datasets
Models
Downloads: 1
[c4e8fb]:
/
Deep_learning_project_part1 (1) (1).ipynb
2700 lines (2700 with data), 998.1 kB
{
"nbformat": 4,
"nbformat_minor": 0,
"metadata": {
"kernelspec": {
"display_name": "Python 3",
"language": "python",
"name": "python3"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.6.7"
},
"colab": {
"name": "Deep learning project_part1.ipynb",
"provenance": [],
"toc_visible": true
},
"accelerator": "GPU"
},
"cells": [
{
"cell_type": "markdown",
"metadata": {
"id": "C1xp4LPGfzJQ"
},
"source": [
"\r\n",
"## 1- Pneumonia classification using transfer learning with Keras\r\n",
"\r\n",
"\r\n",
"\r\n",
"\r\n",
"\r\n"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "byXtS0nYP9UN"
},
"source": [
"The aim of this project is to use transfer learning to detect pneumonia on XRay images (converted to .jpeg). We are going to test different models (VGG16, VGG19 and InceptionV3) and compare their performance."
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "Zwz_HkCDk6PA"
},
"source": [
"Pneumonia : (From Wikipedia)\r\n",
"\r\n",
"\r\n",
"> is an inflammatory condition of the lung primarily affecting the small air sacs known as alveoli. Symptoms typically include some combination of productive or dry cough, chest pain, fever and difficulty breathing. Diagnosis is often based on symptoms and physical examination. Chest X-rays, blood tests, and culture of the sputum may help confirm the diagnosis.\r\n",
"\r\n"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "nquVJOWJlOpi"
},
"source": [
""
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "49h6ow50fzJb"
},
"source": [
"Please download the dataset from the below url\n",
"\n",
"https://www.kaggle.com/paultimothymooney/chest-xray-pneumonia"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "JeSk1v9GUM9v"
},
"source": [
"## Data collection (from Kaggle)"
]
},
{
"cell_type": "code",
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/",
"height": 0
},
"id": "5jLdOKj7xoWw",
"outputId": "4a73be68-1222-4171-9c4f-1b2fc3f7a246"
},
"source": [
"!pip install kaggle"
],
"execution_count": 2,
"outputs": [
{
"output_type": "stream",
"text": [
"Requirement already satisfied: kaggle in /usr/local/lib/python3.6/dist-packages (1.5.10)\n",
"Requirement already satisfied: tqdm in /usr/local/lib/python3.6/dist-packages (from kaggle) (4.41.1)\n",
"Requirement already satisfied: requests in /usr/local/lib/python3.6/dist-packages (from kaggle) (2.23.0)\n",
"Requirement already satisfied: python-slugify in /usr/local/lib/python3.6/dist-packages (from kaggle) (4.0.1)\n",
"Requirement already satisfied: certifi in /usr/local/lib/python3.6/dist-packages (from kaggle) (2020.12.5)\n",
"Requirement already satisfied: urllib3 in /usr/local/lib/python3.6/dist-packages (from kaggle) (1.24.3)\n",
"Requirement already satisfied: six>=1.10 in /usr/local/lib/python3.6/dist-packages (from kaggle) (1.15.0)\n",
"Requirement already satisfied: python-dateutil in /usr/local/lib/python3.6/dist-packages (from kaggle) (2.8.1)\n",
"Requirement already satisfied: idna<3,>=2.5 in /usr/local/lib/python3.6/dist-packages (from requests->kaggle) (2.10)\n",
"Requirement already satisfied: chardet<4,>=3.0.2 in /usr/local/lib/python3.6/dist-packages (from requests->kaggle) (3.0.4)\n",
"Requirement already satisfied: text-unidecode>=1.3 in /usr/local/lib/python3.6/dist-packages (from python-slugify->kaggle) (1.3)\n"
],
"name": "stdout"
}
]
},
{
"cell_type": "code",
"metadata": {
"colab": {
"resources": {
"http://localhost:8080/nbextensions/google.colab/files.js": {
"data": "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",
"ok": true,
"headers": [
[
"content-type",
"application/javascript"
]
],
"status": 200,
"status_text": ""
}
},
"base_uri": "https://localhost:8080/",
"height": 90
},
"id": "leQMh-cTyQHA",
"outputId": "493ef85b-5b02-4c66-a041-1087484d2beb"
},
"source": [
"from google.colab import files \r\n",
"files.upload()"
],
"execution_count": 3,
"outputs": [
{
"output_type": "display_data",
"data": {
"text/html": [
"\n",
" <input type=\"file\" id=\"files-691132b1-6817-4d31-8e3a-d369cb359ed2\" name=\"files[]\" multiple disabled\n",
" style=\"border:none\" />\n",
" <output id=\"result-691132b1-6817-4d31-8e3a-d369cb359ed2\">\n",
" Upload widget is only available when the cell has been executed in the\n",
" current browser session. Please rerun this cell to enable.\n",
" </output>\n",
" <script src=\"/nbextensions/google.colab/files.js\"></script> "
],
"text/plain": [
"<IPython.core.display.HTML object>"
]
},
"metadata": {
"tags": []
}
},
{
"output_type": "stream",
"text": [
"Saving kaggle.json to kaggle.json\n"
],
"name": "stdout"
},
{
"output_type": "execute_result",
"data": {
"text/plain": [
"{'kaggle.json': b'{\"username\":\"sammdl8\",\"key\":\"afd3cb05f9bd1d4f18a4ab2330c37714\"}'}"
]
},
"metadata": {
"tags": []
},
"execution_count": 3
}
]
},
{
"cell_type": "code",
"metadata": {
"id": "-XnAgJ8lydzq"
},
"source": [
"!mkdir -p ~/.kaggle\r\n",
"!cp kaggle.json ~/.kaggle/\r\n",
"\r\n",
"#change the permission\r\n",
"!chmod 600 ~/.kaggle/kaggle.json"
],
"execution_count": 4,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/",
"height": 0
},
"id": "oHwu4mTPyojF",
"outputId": "ccb6f2c1-4849-4192-9bda-cf283568ba26"
},
"source": [
"!kaggle datasets download -d paultimothymooney/chest-xray-pneumonia"
],
"execution_count": 5,
"outputs": [
{
"output_type": "stream",
"text": [
"Downloading chest-xray-pneumonia.zip to /content\n",
"100% 2.29G/2.29G [00:31<00:00, 26.8MB/s]\n",
"100% 2.29G/2.29G [00:31<00:00, 77.5MB/s]\n"
],
"name": "stdout"
}
]
},
{
"cell_type": "code",
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/",
"height": 0
},
"id": "hx_jxso9yugj",
"outputId": "23f29e1f-ebda-4f3c-a879-15f99b921fa4"
},
"source": [
"from zipfile import ZipFile\r\n",
"file_name= 'chest-xray-pneumonia.zip'\r\n",
"\r\n",
"with ZipFile(file_name,'r') as zip:\r\n",
" zip.extractall()\r\n",
" print('Done')"
],
"execution_count": 6,
"outputs": [
{
"output_type": "stream",
"text": [
"Done\n"
],
"name": "stdout"
}
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "BHWV9P6peaFL"
},
"source": [
"# Data visualization"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "vsuIdavNUWHS"
},
"source": [
"Our dataset contains three folders:\r\n",
"\r\n",
"1. Train\r\n",
"2. Test\r\n",
"3. Validation\r\n",
"\r\n",
"Note that our dataset is replicated (we do not consider \"chest_xray folder\" contained in the dataset folder).\r\n",
"\r\n",
"> We have two classes of data: \r\n",
"\r\n",
"\r\n",
"1. NORMAL: absence of pneumonia\r\n",
"2. PNEUMONIA: presence of pneumonia\r\n",
"\r\n",
"\r\n",
"\r\n",
"\r\n",
"\r\n"
]
},
{
"cell_type": "code",
"metadata": {
"id": "_vFmhjQ21mdl"
},
"source": [
"from glob import glob\r\n",
"# useful for getting number of output classes\r\n",
"folders = glob('/content/chest_xray/train/*')"
],
"execution_count": 7,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"id": "YYKr0X6EfzJc"
},
"source": [
"train_normal_path = '/content/chest_xray/train/NORMAL'\n",
"train_pneumonia_path = '/content/chest_xray/train/PNEUMONIA'\n",
"test_normal_path = '/content/chest_xray/test/NORMAL'\n",
"test_pneumonia_path = '/content/chest_xray/test/PNEUMONIA'"
],
"execution_count": 8,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/",
"height": 0
},
"id": "sK-CcTVCWvhp",
"outputId": "4fb2276c-e27e-4505-a85a-83206b5f403a"
},
"source": [
"import os\r\n",
"\r\n",
"list1 = os.listdir(train_normal_path)\r\n",
"n1 = len(list1)\r\n",
"\r\n",
"list2 = os.listdir(train_pneumonia_path)\r\n",
"n2 = len(list2)\r\n",
"\r\n",
"list3 = os.listdir(test_normal_path)\r\n",
"n3 = len(list3)\r\n",
"\r\n",
"list4 = os.listdir(test_pneumonia_path)\r\n",
"n4 = len(list4)\r\n",
"print(\"Train set: Number of images of \\nNormal cases: \"+str(n1)+\r\n",
" \"\\nPneumonia cases: \"+str(n2)+\r\n",
" \"\\nTotal training set: \"+str(n1+n2)+\r\n",
" \"\\n\\nTest set: Number of images of\\nNormal cases: \"+str(n3)+\r\n",
" \"\\nPneumonia cases: \"+str(n4)+\r\n",
" \"\\nTotal test set: \"+str(n3+n4)\r\n",
")\r\n",
"\r\n"
],
"execution_count": 9,
"outputs": [
{
"output_type": "stream",
"text": [
"Train set: Number of images of \n",
"Normal cases: 1341\n",
"Pneumonia cases: 3875\n",
"Total training set: 5216\n",
"\n",
"Test set: Number of images of\n",
"Normal cases: 234\n",
"Pneumonia cases: 390\n",
"Total test set: 624\n"
],
"name": "stdout"
}
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "DeHtsPAsc3lH"
},
"source": [
"As we see the dataset is unbalanced especially in training set: the number of images representing pneumonia class is approximately 3 times the number of images representing normal class."
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "uF_pZYcsTIFm"
},
"source": [
"We are going to visualize random examples of the two classes: NORMAL and PNEUMONIA"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "ofm3UJCNS_9f"
},
"source": [
"Normal"
]
},
{
"cell_type": "code",
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/",
"height": 257
},
"id": "oqHBNshO2FaJ",
"outputId": "2fdcaf1b-c636-4aa6-b95d-af099cdf0dfe"
},
"source": [
"from fastai.vision import *\r\n",
"img1=open_image('/content/chest_xray/train/NORMAL/IM-0115-0001.jpeg')\r\n",
"img1.resize(torch.Size([img1.shape[0],400, 400]))\r\n",
"\r\n",
"img2=open_image('/content/chest_xray/train/NORMAL/IM-0133-0001.jpeg')\r\n",
"img2.resize(torch.Size([img2.shape[0],400, 400]))\r\n",
"\r\n",
"img3=open_image('/content/chest_xray/train/NORMAL/IM-0288-0001.jpeg')\r\n",
"img3.resize(torch.Size([img3.shape[0],400, 400]))\r\n",
"\r\n",
"img4=open_image('/content/chest_xray/train/NORMAL/IM-0446-0001.jpeg')\r\n",
"img4.resize(torch.Size([img3.shape[0],400, 400]))\r\n",
"\r\n",
"_,axs = plt.subplots(1,4, figsize=(16,8))\r\n",
"\r\n",
"img1.show(ax=axs[0])\r\n",
"img2.show(ax=axs[1])\r\n",
"img3.show(ax=axs[2])\r\n",
"img3.show(ax=axs[3])"
],
"execution_count": 10,
"outputs": [
{
"output_type": "stream",
"text": [
"/usr/local/lib/python3.6/dist-packages/torch/nn/functional.py:3103: UserWarning: The default behavior for interpolate/upsample with float scale_factor changed in 1.6.0 to align with other frameworks/libraries, and now uses scale_factor directly, instead of relying on the computed output size. If you wish to restore the old behavior, please set recompute_scale_factor=True. See the documentation of nn.Upsample for details. \n",
" warnings.warn(\"The default behavior for interpolate/upsample with float scale_factor changed \"\n"
],
"name": "stderr"
},
{
"output_type": "display_data",
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 1152x576 with 4 Axes>"
]
},
"metadata": {
"tags": [],
"needs_background": "light"
}
}
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "xQIDffulTkey"
},
"source": [
"PNEUMONIA"
]
},
{
"cell_type": "code",
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/",
"height": 202
},
"id": "68PxIQ9PTnQV",
"outputId": "98f7b4a1-8ebf-4cd0-9366-aad79dd92e17"
},
"source": [
"from fastai.vision import *\r\n",
"img1=open_image('/content/chest_xray/train/PNEUMONIA/person1003_virus_1685.jpeg')\r\n",
"img1.resize(torch.Size([img1.shape[0],400, 400]))\r\n",
"\r\n",
"img2=open_image('/content/chest_xray/train/PNEUMONIA/person1395_virus_2398.jpeg')\r\n",
"img2.resize(torch.Size([img2.shape[0],400, 400]))\r\n",
"\r\n",
"img3=open_image('/content/chest_xray/train/PNEUMONIA/person1308_bacteria_3288.jpeg')\r\n",
"img3.resize(torch.Size([img3.shape[0],400, 400]))\r\n",
"\r\n",
"img4=open_image('/content/chest_xray/train/PNEUMONIA/person1449_bacteria_3743.jpeg')\r\n",
"img4.resize(torch.Size([img3.shape[0],400, 400]))\r\n",
"\r\n",
"_,axs = plt.subplots(1,4, figsize=(16,8))\r\n",
"\r\n",
"img1.show(ax=axs[0])\r\n",
"img2.show(ax=axs[1])\r\n",
"img3.show(ax=axs[2])\r\n",
"img3.show(ax=axs[3])"
],
"execution_count": 11,
"outputs": [
{
"output_type": "display_data",
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 1152x576 with 4 Axes>"
]
},
"metadata": {
"tags": [],
"needs_background": "light"
}
}
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "btTCwJbugryo"
},
"source": [
"Import libraries"
]
},
{
"cell_type": "code",
"metadata": {
"id": "mKRBFr2CWJbC"
},
"source": [
"# import the libraries as shown below\r\n",
"\r\n",
"from keras.layers import Input, Lambda, Dense, Flatten\r\n",
"from keras.models import Model\r\n",
"#from keras.applications.resnet50 import ResNet50\r\n",
"from keras.applications.vgg16 import VGG16\r\n",
"from keras.applications.vgg16 import preprocess_input\r\n",
"from keras.preprocessing import image\r\n",
"from keras.preprocessing.image import ImageDataGenerator\r\n",
"from keras.models import Sequential\r\n",
"import numpy as np\r\n",
"from glob import glob\r\n",
"import matplotlib.pyplot as plt"
],
"execution_count": 12,
"outputs": []
},
{
"cell_type": "markdown",
"metadata": {
"id": "pGJpxOC7hMK3"
},
"source": [
"Now we are going to implement different models based on famous pretrained models (VGG16, VGG19, InceptionV3) and compare their performance.\r\n",
"\r\n",
"\r\n",
"> Let's begin with VGG16\r\n",
"\r\n"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "QHgt6uqBg-IS"
},
"source": [
"## VGG16"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "vE93X_LiSVY2"
},
"source": [
"The following architecture is taken from this link: https://neurohive.io/en/popular-networks/vgg16/"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "fzvYMV7oSG29"
},
"source": [
""
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "cjHfZn2fhEKD"
},
"source": [
"We are going to use a pretrained model called VGG16. \r\n",
"\r\n",
"\r\n",
"> In fact we use weights of the VGG16 trained on ImageNet dataset. But we remove the fully connected layer of this model because in our case we have only two outputs (normal and pneumonia) and not 22000 categories.\r\n",
"\r\n"
]
},
{
"cell_type": "code",
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/",
"height": 0
},
"id": "rAMLy0MEfzJd",
"outputId": "91d9034d-f5bf-4f81-957a-7c50ac3a8552"
},
"source": [
"# Import the Vgg 16 library as shown below and add preprocessing layer to the front of VGG\n",
"# Here we will be using imagenet weights\n",
"\n",
"IMAGE_SIZE=[224,224]\n",
"vgg = VGG16(input_shape=IMAGE_SIZE + [3], weights='imagenet', include_top=False)\n",
"\n",
"\n"
],
"execution_count": 13,
"outputs": [
{
"output_type": "stream",
"text": [
"Downloading data from https://storage.googleapis.com/tensorflow/keras-applications/vgg16/vgg16_weights_tf_dim_ordering_tf_kernels_notop.h5\n",
"58892288/58889256 [==============================] - 0s 0us/step\n"
],
"name": "stdout"
}
]
},
{
"cell_type": "code",
"metadata": {
"id": "SA_OEW0rfzJf"
},
"source": [
"# we keep existing weights\n",
"for layer in vgg.layers:\n",
" layer.trainable = False"
],
"execution_count": 14,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"id": "PxPws3aQfzJf"
},
"source": [
"# our layers\n",
"x = Flatten()(vgg.output)"
],
"execution_count": 15,
"outputs": []
},
{
"cell_type": "markdown",
"metadata": {
"id": "46HNFlOG3WQN"
},
"source": [
"We use softmax for prediction because we have two classes (normal and pneumonia)"
]
},
{
"cell_type": "code",
"metadata": {
"id": "uCU5htZ0fzJg"
},
"source": [
"prediction = Dense(len(folders), activation='softmax')(x)\n",
"\n",
"# create a model object\n",
"model = Model(inputs=vgg.input, outputs=prediction)"
],
"execution_count": 16,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/",
"height": 0
},
"id": "NFYTcgzKfzJh",
"outputId": "d34f39b3-4553-4dbc-fb61-99a14bc52540"
},
"source": [
"\n",
"# view the structure of the model\n",
"model.summary()\n"
],
"execution_count": 17,
"outputs": [
{
"output_type": "stream",
"text": [
"Model: \"model\"\n",
"_________________________________________________________________\n",
"Layer (type) Output Shape Param # \n",
"=================================================================\n",
"input_1 (InputLayer) [(None, 224, 224, 3)] 0 \n",
"_________________________________________________________________\n",
"block1_conv1 (Conv2D) (None, 224, 224, 64) 1792 \n",
"_________________________________________________________________\n",
"block1_conv2 (Conv2D) (None, 224, 224, 64) 36928 \n",
"_________________________________________________________________\n",
"block1_pool (MaxPooling2D) (None, 112, 112, 64) 0 \n",
"_________________________________________________________________\n",
"block2_conv1 (Conv2D) (None, 112, 112, 128) 73856 \n",
"_________________________________________________________________\n",
"block2_conv2 (Conv2D) (None, 112, 112, 128) 147584 \n",
"_________________________________________________________________\n",
"block2_pool (MaxPooling2D) (None, 56, 56, 128) 0 \n",
"_________________________________________________________________\n",
"block3_conv1 (Conv2D) (None, 56, 56, 256) 295168 \n",
"_________________________________________________________________\n",
"block3_conv2 (Conv2D) (None, 56, 56, 256) 590080 \n",
"_________________________________________________________________\n",
"block3_conv3 (Conv2D) (None, 56, 56, 256) 590080 \n",
"_________________________________________________________________\n",
"block3_pool (MaxPooling2D) (None, 28, 28, 256) 0 \n",
"_________________________________________________________________\n",
"block4_conv1 (Conv2D) (None, 28, 28, 512) 1180160 \n",
"_________________________________________________________________\n",
"block4_conv2 (Conv2D) (None, 28, 28, 512) 2359808 \n",
"_________________________________________________________________\n",
"block4_conv3 (Conv2D) (None, 28, 28, 512) 2359808 \n",
"_________________________________________________________________\n",
"block4_pool (MaxPooling2D) (None, 14, 14, 512) 0 \n",
"_________________________________________________________________\n",
"block5_conv1 (Conv2D) (None, 14, 14, 512) 2359808 \n",
"_________________________________________________________________\n",
"block5_conv2 (Conv2D) (None, 14, 14, 512) 2359808 \n",
"_________________________________________________________________\n",
"block5_conv3 (Conv2D) (None, 14, 14, 512) 2359808 \n",
"_________________________________________________________________\n",
"block5_pool (MaxPooling2D) (None, 7, 7, 512) 0 \n",
"_________________________________________________________________\n",
"flatten (Flatten) (None, 25088) 0 \n",
"_________________________________________________________________\n",
"dense (Dense) (None, 2) 50178 \n",
"=================================================================\n",
"Total params: 14,764,866\n",
"Trainable params: 50,178\n",
"Non-trainable params: 14,714,688\n",
"_________________________________________________________________\n"
],
"name": "stdout"
}
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "BBrcDRcJ3oe9"
},
"source": [
"We create a data generator with Keras for image augmentation. Data augmentation is a crucial step to boost our model's aptitude to generalize.\r\n",
"\r\n"
]
},
{
"cell_type": "code",
"metadata": {
"id": "idjBCZAO_RqA"
},
"source": [
"# Use the Image Data Generator to import the images from the dataset\r\n",
"from keras.preprocessing.image import ImageDataGenerator\r\n",
"\r\n",
"train_datagen = ImageDataGenerator(rescale = 1./255,\r\n",
" shear_range = 0.2,\r\n",
" zoom_range = 0.2,\r\n",
" horizontal_flip = True)\r\n",
"\r\n",
"test_datagen = ImageDataGenerator(rescale = 1./255)"
],
"execution_count": 18,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/",
"height": 0
},
"id": "gNiXWRvo_Tc3",
"outputId": "968f89b9-27f2-4082-c752-846ef9f7ffc5"
},
"source": [
"# Make sure you provide the same target size as initialied for the image size\r\n",
"training_set = train_datagen.flow_from_directory('/content/chest_xray/train',\r\n",
" target_size = (224, 224),\r\n",
" batch_size = 32,\r\n",
" class_mode = 'categorical')"
],
"execution_count": 19,
"outputs": [
{
"output_type": "stream",
"text": [
"Found 5216 images belonging to 2 classes.\n"
],
"name": "stdout"
}
]
},
{
"cell_type": "code",
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/",
"height": 0
},
"id": "4m2e48ZC_ZAm",
"outputId": "b256f685-b6b6-4827-ae69-6c5312de5589"
},
"source": [
"test_set = test_datagen.flow_from_directory('/content/chest_xray/test',\r\n",
" target_size = (224, 224),\r\n",
" batch_size = 32,\r\n",
" class_mode = 'categorical')"
],
"execution_count": 20,
"outputs": [
{
"output_type": "stream",
"text": [
"Found 624 images belonging to 2 classes.\n"
],
"name": "stdout"
}
]
},
{
"cell_type": "code",
"metadata": {
"id": "GGYLNjw6fzJh"
},
"source": [
"# tell the model what cost and optimization method to use\n",
"model.compile(\n",
" loss='categorical_crossentropy',\n",
" optimizer='adam',\n",
" metrics=['accuracy']\n",
")\n"
],
"execution_count": 21,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/",
"height": 0
},
"id": "vu9S1tpz_4nr",
"outputId": "a66520c6-9221-4d84-efdb-e147f4bc87ad"
},
"source": [
"# fit the model\r\n",
"\r\n",
"r = model.fit_generator(\r\n",
" training_set,\r\n",
" validation_data=test_set,\r\n",
" epochs=4,\r\n",
" steps_per_epoch=len(training_set),\r\n",
" validation_steps=len(test_set)\r\n",
")"
],
"execution_count": 22,
"outputs": [
{
"output_type": "stream",
"text": [
"/usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/engine/training.py:1844: UserWarning: `Model.fit_generator` is deprecated and will be removed in a future version. Please use `Model.fit`, which supports generators.\n",
" warnings.warn('`Model.fit_generator` is deprecated and '\n"
],
"name": "stderr"
},
{
"output_type": "stream",
"text": [
"Epoch 1/4\n",
"163/163 [==============================] - 111s 629ms/step - loss: 0.5487 - accuracy: 0.8261 - val_loss: 0.2359 - val_accuracy: 0.9119\n",
"Epoch 2/4\n",
"163/163 [==============================] - 102s 624ms/step - loss: 0.1151 - accuracy: 0.9592 - val_loss: 0.2767 - val_accuracy: 0.9071\n",
"Epoch 3/4\n",
"163/163 [==============================] - 102s 624ms/step - loss: 0.0866 - accuracy: 0.9676 - val_loss: 0.2765 - val_accuracy: 0.9006\n",
"Epoch 4/4\n",
"163/163 [==============================] - 102s 623ms/step - loss: 0.0842 - accuracy: 0.9725 - val_loss: 0.2963 - val_accuracy: 0.9054\n"
],
"name": "stdout"
}
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "Rr6EBzgK82TE"
},
"source": [
"After training the model in four epochs we achieve the following results:\r\n",
"\r\n",
"\r\n",
"* training accuracy = 0,9658\r\n",
"* validation accyracy = 0,9183\r\n",
"\r\n",
"Do not forget that our training set is unbalanced : the number of PNEUMONIA images is approximately 3 times larger than the NORMAL class. So by balancing our training set the model could perform much better."
]
},
{
"cell_type": "code",
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/",
"height": 543
},
"id": "-eLeozZnfzJk",
"outputId": "210e4a91-d18c-461f-e9b9-fba8e0b15e34"
},
"source": [
"# plot the loss\n",
"plt.plot(r.history['loss'], label='train loss')\n",
"plt.plot(r.history['val_loss'], label='val loss')\n",
"plt.legend()\n",
"plt.show()\n",
"plt.savefig('LossVal_loss')\n",
"\n",
"# plot the accuracy\n",
"acc_train = r.history['accuracy']\n",
"acc_val = r.history['val_accuracy']\n",
"epochs = range(1,5)\n",
"plt.plot(epochs, acc_train, 'g', label='Training accuracy')\n",
"plt.plot(epochs, acc_val, 'b', label='validation accuracy')\n",
"plt.title('Training and Validation accuracy')\n",
"plt.xlabel('Epochs')\n",
"plt.ylabel('Accuracy')\n",
"plt.legend()\n",
"plt.show()"
],
"execution_count": 23,
"outputs": [
{
"output_type": "display_data",
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAAAXoAAAD4CAYAAADiry33AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nO3deXxU9b3/8dcne0ICCSECJiCLyC5BA3JFROuGooAruLRavfqzrVW0tpe6Vqu3VqtFq1btvfZhe6XghsWqpWpBtHUhaNh3FAmyJGFLgOzf3x8zJENIyAQmOTOT9/PxyCNz5nzPzOcw4X3OfL9nMeccIiISvWK8LkBERFqXgl5EJMop6EVEopyCXkQkyinoRUSiXJzXBTTUpUsX16tXL6/LEBGJKIsWLSp2zmU1Ni/sgr5Xr17k5+d7XYaISEQxs41NzVPXjYhIlFPQi4hEuaCC3szGmdlqM1tnZtMamX+zmS01swIz+9jMBgXM+7l/udVmdl4oixcRkeY120dvZrHAM8A5QCGw0MzmOOdWBDSb4Zx7zt9+AvAEMM4f+FOAwcCxwPtmdoJzrqYlRVZVVVFYWEh5eXlLFpMASUlJ5OTkEB8f73UpItLGghmMHQmsc85tADCzmcBEoC7onXN7Atp3AA5cQGciMNM5VwF8ZWbr/K/3SUuKLCwsJC0tjV69emFmLVlUAOccJSUlFBYW0rt3b6/LEZE2FkzXTTawKWC60P/cQczsR2a2HngUuLWFy95kZvlmll9UVHRIAeXl5WRmZirkj5CZkZmZqW9EIu1UyAZjnXPPOOf6Av8F3NPCZV9wzuU55/Kysho9DFQhf5T07yfSfgXTdbMZ6BEwneN/rikzgd8f4bIiIu3L/l2wbTlsWwax8ZB3fcjfIpigXwj0M7Pe+EJ6CnBVYAMz6+ecW+ufHA8ceDwHmGFmT+AbjO0HfB6KwtvSrl27mDFjBj/84Q9bvOwFF1zAjBkzSE9PD6r9L37xC1JTU7nzzjtb/F4iEsZqa2DHV7BtqS/Yty7zhfvugN7tnBHeBL1zrtrMbgHmArHAi8655Wb2IJDvnJsD3GJmZwNVwE7gWv+yy83sFXwDt9XAj1p6xE042LVrF88++2yjQV9dXU1cXNP/jO+8805rliYi4ah8T/1e+lZ/sG9fAVX7fPMtFrr0gx4jfcHebSh0HQJp3VqlnKAugeCcewd4p8Fz9wU8vu0wyz4MPHykBYaDadOmsX79enJzcznnnHMYP3489957LxkZGaxatYo1a9YwadIkNm3aRHl5Obfddhs33XQTUH9Jh7KyMs4//3xOO+00/v3vf5Odnc1f//pXkpOTm3zfgoICbr75Zvbt20ffvn158cUXycjI4KmnnuK5554jLi6OQYMGMXPmTD788ENuu833MZgZCxYsIC0trU3+fUTardpa2PlVQKgv8+2x7/qmvk1Sui/IT7oWug6GbkMgayDEJ7VZmWF3rZvmPPDWclZ8u6f5hi0w6NiO3H/R4CbnP/LIIyxbtoyCggIA5s+fzxdffMGyZcvqDld88cUX6dy5M/v372fEiBFceumlZGZmHvQ6a9eu5S9/+Qt/+MMfuOKKK3j99de55pprmnzf733ve/zud79j7Nix3HfffTzwwANMnz6dRx55hK+++orExER27doFwG9+8xueeeYZRo8eTVlZGUlJbfdHJNIuVJTCthW+IN+6rH4vvbLMN99iIPN4yD7ZF+rdhvqCvWM2eHwwRMQFfbgYOXLkQcekP/XUU8yePRuATZs2sXbt2kOCvnfv3uTm5gJw8skn8/XXXzf5+rt372bXrl2MHTsWgGuvvZbLL78cgBNPPJGrr76aSZMmMWnSJABGjx7NHXfcwdVXX80ll1xCTk5OyNZVpF2prYVdGxt0vSyDnV/Xt0ns5Nszz73a97vrYN9eekKKZ2UfTsQF/eH2vNtShw4d6h7Pnz+f999/n08++YSUlBTOOOOMRo9ZT0xMrHscGxvL/v37j+i93377bRYsWMBbb73Fww8/zNKlS5k2bRrjx4/nnXfeYfTo0cydO5cBAwYc0euLtBuVew/dS9+2HCpL/Q0MMvtC92GQe40/1IdApxzP99JbIuKC3gtpaWmUlpY2OX/37t1kZGSQkpLCqlWr+PTTT4/6PTt16kRGRgYfffQRY8aM4c9//jNjx46ltraWTZs2ceaZZ3Laaacxc+ZMysrKKCkpYejQoQwdOpSFCxeyatUqBb3IAc75+s0b7qXv+Iq6E/kTO/r2zIdNqQ/0YwZCQofDvnQkUNAHITMzk9GjRzNkyBDOP/98xo8ff9D8cePG8dxzzzFw4ED69+/PqFGjQvK+L730Ut1gbJ8+ffjjH/9ITU0N11xzDbt378Y5x6233kp6ejr33nsv8+bNIyYmhsGDB3P++eeHpAaRiFO5D7avPHQvvWJ3fZvOfXyhfmJAqKf3jKi99JYw51zzrdpQXl6ea3jjkZUrVzJw4ECPKooe+neUqOIc7C70B/nS+uPSS9ZTt5eekOoL9K5DAvbSB0FiqqeltwYzW+Scy2tsnvboRST8Ve3376UvC9hLXwblu+rbZPTyBfmQywL20o+DGN12Q0EvIuHDOdjzbSN76evA1fraxHeAroNg8MUH76UndfS29jCmoBcRb1SVQ9GqgL10/8/+nfVt0ntC16EwaFJ9qGf01l56CynoRaR1OQelWw/dSy9eCweuiBKX7NtLHzih/kSjroMhqZO3tUcJBb2IhE51BRStPnQvfV9JfZtOPXx75gMu9O+lD4XOvSEm1ru6o5yCXkSOTOm2+iCv20tfA7XVvvlxSb7j0PtfcPBeenKGt3W3Qwr6VpKamkpZWVnQz4uErepKX4AHXolx2zLYG3A3uI7Zvr30E8bV76Vn9tVeephQ0ItIvbKiQ6+XXrQaaqt882MT4ZgB0O+8+mu8dB0CKZ29rVsOS0EfhGnTptGjRw9+9KMfAfU3B7n55puZOHEiO3fupKqqioceeoiJEycG9ZrOOX72s5/x7rvvYmbcc889TJ48mS1btjB58mT27NlDdXU1v//97zn11FO54YYbyM/Px8y4/vrruf3221tzlUOnct/Bl2ytO/PQGkw39VxzbYJ5ndZs07BGr+tp8FxTZ3rWVPkGQxvupZdtq2+T1t0X4sefXX+99MzjIVaxEWki7xN7d5rvDzOUug2F8x9pcvbkyZOZOnVqXdC/8sorzJ07l6SkJGbPnk3Hjh0pLi5m1KhRTJgwIaj7s77xxhsUFBSwePFiiouLGTFiBKeffjozZszgvPPO4+6776ampoZ9+/ZRUFDA5s2bWbZsGUDdpYnDlnPwzadQ8DIsfzPgAlESHoy6M0cBYhMgqz/0PStgL30odMhs8hUkskRe0Htg+PDhbN++nW+//ZaioiIyMjLo0aMHVVVV3HXXXSxYsICYmBg2b97Mtm3b6Nat+bvEfPzxx1x55ZXExsbStWtXxo4dy8KFCxkxYgTXX389VVVVTJo0idzcXPr06cOGDRv48Y9/zPjx4zn33HPbYK2PwK5NsHgmLJ4BOzb4TmwZfDH0OcPfV+sPl8Yuu1H3XGNtGj7Xlm0a1ud1Pe6gX0f8Ohbj60PvOsR3p6PYeCR6RV7QH2bPuzVdfvnlvPbaa2zdupXJkycD8PLLL1NUVMSiRYuIj4+nV69ejV6euCVOP/10FixYwNtvv811113HHXfcwfe+9z0WL17M3Llzee6553jllVd48cUXQ7FaR69yH6z6m2/vfcOHgINeY+D0n/qOiY7Ca4qIRJrIC3qPTJ48mRtvvJHi4mI+/PBDwHd54mOOOYb4+HjmzZvHxo0bg369MWPG8Pzzz3PttdeyY8cOFixYwGOPPcbGjRvJycnhxhtvpKKigi+++IILLriAhIQELr30Uvr373/Yu1K1Cedg02e+cF8229c1k34cnDHNd4nXjF7e1iciB1HQB2nw4MGUlpaSnZ1N9+7dAbj66qu56KKLGDp0KHl5eS26/vvFF1/MJ598wrBhwzAzHn30Ubp168ZLL73EY489Rnx8PKmpqfzpT39i8+bNfP/736e21netj1/96letso7N2rUJlsyEgsCumUmQexX0PFWnpYuEKV2muB05on/Hprpmcq9S14xIGNFliqVlnINNn/uPmpkNFXt8F5ca+1+Qe6W6ZkQijIJe6u0u9B01UzADdqyH+BTfVQNzr4LjRqtrRiRCRUzQO+eCOj5dGtdkF13lPlj1tr9rZj7g4LjTYMxPYNAESExryzJFpBVERNAnJSVRUlJCZmamwv4IOOcoKSkhKSnpwBNQuBC+/L/6rplO/q6ZYVN8VxIUkagREUGfk5NDYWEhRUVFzTeWRiUlJZHTMRY+etzXNVOyzt81MxFyr1bXjEgUi4igj4+Pp3dv7WUekar9vq6Zj1+G9fPwdc2MhtNu94W8umZEol5EBL20kHNQmA8F/+c7oalit79r5mf+rpk+XlcoIm1IQR9N9nxbf9RMydqArpmrfAOs6poRaZeiJuj3lFfxy7dW8P/G9uH4Y9pRd8SBrpmCGbBhHrha31mqo2/znbWqrhmRdi9qgr68soYPVm1nxZY9zP7haBLionjvta5r5mVY9oa/a6YHjLnT1zWT2dfrCkUkjERN0B/TMYlHLhnKTX9exOPvrebn50fhJRMads3EJdd3zfQao64ZEWlU1AQ9wLmDu3HlyJ68sGADY0/I4tS+Xbwu6ehVlcNqf9fM+n/6u2b+w9c1M2giJHX0ukIRCXNRFfQA9144kM82lPCTVxbz99tOp1NKBN5QwTnYvMjfNfM6lO+Gjjm+s1WHXamuGRFpkagL+pSEOKZPyeWSZ//NXW8u5ekrh0fO2bR7ttRfBrh4jb9rZoK/a+Z0dc2IyBGJuqAHODEnndvPOYHH5q7mO/2P4dKTc7wuqWlV5bD6HX/XzAf1XTMTfue7oJi6ZkTkKEVl0APcPLYvH64u4v45yxnZuzM9Oqd4XVI952DzF/6umdf8XTPZcNodvr13dc2ISAgF1RdgZuPMbLWZrTOzaY3Mv8PMVpjZEjP7wMyOC5hXY2YF/p85oSz+cGJjjCcmD8OAqbMKqK6pbau3btqeLfDxdHjmFPif7/iCvt958N03YepSOOtehbyIhFyze/RmFgs8A5wDFAILzWyOc25FQLMvgTzn3D4z+wHwKDDZP2+/cy43xHUHJScjhV9OGsLUWQU8O389t57Vr+2LqCqHNe/6umbWve/rmukxCi56yndCU1Kntq9JRNqVYLpuRgLrnHMbAMxsJjARqAt659y8gPafAh7fvbrepOHZzFu9nSc/WMuYfl0Y3jOj9d/UOfj2C1+4L30NynfVd80MuxK6HN/6NYiI+AUT9NnApoDpQuCUw7S/AXg3YDrJzPKBauAR59ybDRcws5uAmwB69uwZREkt8+DEIeR/vZOpswp4+9YxpCa20tBE6VZYMssX8EWrIC4JBl7k63fvPRZiYlvnfUVEDiOkiWdm1wB5wNiAp49zzm02sz7AP81sqXNufeByzrkXgBfAd3PwUNYE0Ck5nieuGMaUP3zKg28t59HLhoXuxasr6o+aqeuaOQUuehIGX6yuGRHxXDBBvxnoETCd43/uIGZ2NnA3MNY5V3HgeefcZv/vDWY2HxgOrG+4fGs7pU8mPxjbl2fnr+c7A45h3JDuR/5izsG3X/q7Zl4N6Jq5HYZdpa4ZEQkrwQT9QqCfmfXGF/BTgKsCG5jZcOB5YJxzbnvA8xnAPudchZl1AUbjG6j1xNSzT+CjtcVMe2MpuT0y6NYpqWUvULotoGtmpbpmRCQiNBv0zrlqM7sFmAvEAi8655ab2YNAvnNuDvAYkAq86j8L9Rvn3ARgIPC8mdXiO5TzkQZH67SphLgYpk/J5cKnPubOVxfzp+tHEhPTzFmz1RWwOvComRrIGQkXTvd1zSSnt03xIiJHyJwLeZf4UcnLy3P5+fmt+h4zPvuGu2Yv5Z7xA/nPMY3cbck52FJQ3zWzfyekHeu7BHDuVdDFg8M0RUQOw8wWOefyGpsXtWfGHs6VI3swb/V2Hv37ak7t24VBx/ovM1C6DZa+4gv47St8XTMDLvSFe58z1DUjIhGpXe7RA5SUVTDuyY84Jhlmn1NKwtKZsPa9+q6Z3KvUNSMiEUN79A05R+aelczu9SYd1swm4fUySOsOo2/1HTWTdYLXFYqIhEz7Cvqy7bDkQNfMcnJiE1naeQy3bTuZG8Zfz9gB3byuUEQk5KI/6KsrYc3ffeG+9h++rpnsPLjwtzD4EvrFpbH16Y+58/Vl/P22DDJTE72uWEQkpKIz6J2DLYsDjprZAand4NQf+/res/rXNU0Cpk8ezqRn/sW0N5bywndPjpwblYiIBCG6gr6sqP6Epu3LITYRBoyH3Kt9R83ENr66g47tyM/G9eeht1cyc+EmrhwZ+uvtiIh4JXqCfscGeHoE1Fb7umbGPwFDLoHk4K5Wef3o3sxfXcSDb63glN6d6ZOV2soFi4i0jei5CWlGbzjrPvjhZ3DjBzDihqBDHiAmxvjN5cNIjI9h6qwCqsLhRiUiIiEQPUFvBqNvg2MGHPFLdOuUxK8uHsqSwt1Mf39NCIsTEfFO9AR9iJw/tDuXn5zDs/PX8/lXO7wuR0TkqCnoG3H/hMH07JzC7bMK2FNe5XU5IiJHRUHfiNTEOH47OZete8q5781lXpcjInJUFPRNOKlnBrd+px9vFnzLXwsOuc+KiEjEUNAfxo/O7MvJx2Vwz+xlFO7c53U5IiJHREF/GHGxMfz2ilwccMesxdTUhteVPkVEgqGgb0bPzBR+MWEwn3+9g+c+bPNb3YqIHDUFfRAuPSmb8UO789v31rCkcJfX5YiItIiCPghmxsMXDyErLZGpMwvYV1ntdUkiIkFT0AcpPSWBx68Yxlcle3no7ZVelyMiEjQFfQuc2rcLN43pw4zPvuEfy7d6XY6ISFAU9C10x7knMKh7R6a9sZTtpeVelyMi0iwFfQslxsXy1JW57K2o5qevLiHcbq4uItKQgv4IHH9MGnePH8iHa4p46d9fe12OiMhhKeiP0HdHHceZ/bP473dXsWZbqdfliIg0SUF/hMyMRy8bRlpiHLf+5Usqqmu8LklEpFEK+qOQlZbIo5edyKqtpfxm7mqvyxERaZSC/iidNbAr14zqyR8++oqP1xZ7XY6IyCEU9CFw9wWD6JvVgZ+8WsDOvZVelyMichAFfQgkJ8Ty5JTh7NhbyV2zl+qQSxEJKwr6EBmS3Yk7zunPu8u28uqiQq/LERGpo6APoZtO78OoPp15YM5yNpbs9bocERFAQR9SsTHGE1fkEhtj3DazgKqaWq9LEhFR0IfasenJPHzxUAo27eJ3/1zndTkiIgr61nDRsGO5ZHg2T/9zLYs27vC6HBFp5xT0reSBiYM5Nj2ZqbMKKC2v8rocEWnHFPStJC0pnumTc9m8cz+/mLPC63JEpB1T0LeivF6dueXM43n9i0LeXrLF63JEpJ0KKujNbJyZrTazdWY2rZH5d5jZCjNbYmYfmNlxAfOuNbO1/p9rQ1l8JPjxWf0Y1iOdu2YvZcvu/V6XIyLtULNBb2axwDPA+cAg4EozG9Sg2ZdAnnPuROA14FH/sp2B+4FTgJHA/WaWEbryw198bAxPTs6lqqaWO2YtprZWZ82KSNsKZo9+JLDOObfBOVcJzAQmBjZwzs1zzu3zT34K5Pgfnwe855zb4ZzbCbwHjAtN6ZGjV5cO3H/RID7ZUMIfPtrgdTki0s4EE/TZwKaA6UL/c025AXi3Jcua2U1mlm9m+UVFRUGUFHmuyOvBeYO78pt/rGbZ5t1elyMi7UhIB2PN7BogD3isJcs5515wzuU55/KysrJCWVLYMDMeueREMlISmDqrgP2VulGJiLSNYIJ+M9AjYDrH/9xBzOxs4G5ggnOuoiXLthcZHRJ4/IphrNtexq/eXel1OSLSTgQT9AuBfmbW28wSgCnAnMAGZjYceB5fyG8PmDUXONfMMvyDsOf6n2u3xvTL4obTevOnTzYyb9X25hcQETlKzQa9c64auAVfQK8EXnHOLTezB81sgr/ZY0Aq8KqZFZjZHP+yO4Bf4ttYLAQe9D/Xrv30vP4M6JbGT19bTHFZRfMLiIgcBQu3m2Tk5eW5/Px8r8todau3lnLR0x9z2vFd+N9r8zAzr0sSkQhmZoucc3mNzdOZsR7p3y2NaeMG8M9V2/m/z77xuhwRiWIKeg9dd2ovxvTrwsNvr2Dd9jKvyxGRKKWg91BMjPH45cNIjo9l6qwvqazWjUpEJPQU9B47pmMSv770RJZt3sMT763xuhwRiUIK+jBw7uBuXDmyB88vWM8n60u8LkdEooyCPkzce+EgemV24CevFLB7n25UIiKho6APEykJcUyfnMv20grufnMp4XbYq4hELgV9GBnWI52pZ/fjb0u2MPvLdnulCBEJMQV9mPnBGcczolcG9/11OZt27Gt+ARGRZijow0xsjPHEFbkYcPusAqprdMiliBwdBX0Y6tE5hV9OGkL+xp38fv56r8sRkQinoA9Tk4ZnM2HYsUz/YC0Fm3Z5XY6IRDAFfRj75aQhdOuYxNSZX7K3otrrckQkQinow1in5Hgev2IYG3fs48G3VnhdjohEKAV9mBvVJ5Obx/ZlVv4m/r5sq9fliEgEUtBHgNvPPoGh2Z2Y9sYStu0p97ocEYkwCvoIkBAXw/QpuVRU1XLnq4uprdVZsyISPAV9hOiblco9Fw7ko7XF/PHfX3tdjohEEAV9BLlqZE/OHtiVX7+7ipVb9nhdjohECAV9BDEzfn3pUDomxzN1ZgHlVTVelyQiEUBBH2EyUxN57PITWb2tlF//fZXX5YhIBFDQR6Az+x/Ddaf24o//+poFa4q8LkdEwpyCPkJNO38AJ3RN5SevLmbH3kqvyxGRMKagj1BJ8bFMnzyc3fuqmPb6Et2oRESapKCPYIOO7chPz+vPP1ZsY9bCTV6XIyJhSkEf4W44rTejj8/kgbdWsKGozOtyRCQMKegjXEyM8ZvLh5EQF8Ptswqo0o1KRKQBBX0U6N4pmV9dMpTFhbt58v21XpcjImFGQR8lLhjanctPzuHZ+etY+PUOr8sRkTCioI8i908YTI/OKUydWcCe8iqvyxGRMKGgjyKpiXH8dnIuW/eUc/9fl3tdjoiECQV9lDmpZwY//s7xzP5yM3MWf+t1OSISBhT0UeiWM4/npJ7p3D17KZt37fe6HBHxmII+CsXFxjB98nBqax23zyqgRjcqEWnXFPRRqmdmCg9MHMLnX+3g+QXrvS5HRDykoI9il56Uzfih3XniH2tYWrjb63JExCMK+ihmZjx88RC6pCZy26wv2V+pG5WItEcK+iiXnpLAE1cM46vivTz09gqvyxERDwQV9GY2zsxWm9k6M5vWyPzTzewLM6s2s8sazKsxswL/z5xQFS7BO/X4Ltw4pg8vf/YN76/Y5nU5ItLGmg16M4sFngHOBwYBV5rZoAbNvgGuA2Y08hL7nXO5/p8JR1mvHKGfnHsCg7p35GevL2F7abnX5YhIGwpmj34ksM45t8E5VwnMBCYGNnDOfe2cWwLo0olhKjEulien5LK3opqfvqoblYi0J8EEfTYQeFeLQv9zwUoys3wz+9TMJjXWwMxu8rfJLyrSPVBbS7+uadw9fiAfriniT59s9LocEWkjbTEYe5xzLg+4CphuZn0bNnDOveCcy3PO5WVlZbVBSe3Xd0cdx5n9s/jvd1aydlup1+WISBsIJug3Az0CpnP8zwXFObfZ/3sDMB8Y3oL6JMTMjEcvG0ZqYhy3ziygolqHXIpEu2CCfiHQz8x6m1kCMAUI6ugZM8sws0T/4y7AaEDH+HksKy2RRy87kZVb9vD4P9Z4XY6ItLJmg945Vw3cAswFVgKvOOeWm9mDZjYBwMxGmFkhcDnwvJkduEbuQCDfzBYD84BHnHMK+jBw1sCuXH1KT15YsIF/rSv2uhwRaUUWbkdf5OXlufz8fK/LaBf2V9Yw/ncfsa+ihr9PHUN6SoLXJYnIETKzRf7x0EPozNh2LDkhlqemDKdkbwV3zV6qQy5FopSCvp0bkt2JO87pzztLt/LaokKvyxGRVqCgF246vQ+n9O7ML+YsZ2PJXq/LEZEQU9ALsTHGE5NziYkxbp9VQHWNTnAWiSYKegEgOz2Zhy8eyhff7OLpeeu8LkdEQkhBL3UmDDuWi4dn89QHa1m0cafX5YhIiCjo5SAPTBzMsenJ3D6rgLKKaq/LEZEQUNDLQTomxTN9ci6FO/fxiznLm19ARMKegl4OkderMz8683heW1TIO0u3eF2OiBwlBb006taz+jGsRzo/f2MpW3bv97ocETkKCnppVHxsDNMn51JVU8udry6mtlZnzYpEKgW9NKl3lw7cd+Eg/rWuhP/9+CuvyxGRI6Sgl8OaPKIH5w3uyqNzV7H8291elyMiR0BBL4dlZjxyyYlkpCRw28wCyqt0oxKRSKOgl2ZldEjg8SuGsW57Gb96Z6XX5YhICynoJShj+mVx/ejevPTJRuat3u51OSLSAgp6CdrPxvVnQLc0fvrqEorLKrwuR0SCpKCXoCXFxzJ9Si57yquY9voS3ahEJEIo6KVFBnTryH+NG8D7K7cz4/NvvC5HRIKgoJcW+/6pvRjTrwu//NsK1m0v87ocEWmGgl5aLCbGePzyYSTHxzJ11pdUVutGJSLhTEEvR+SYjkk8cumJLNu8h9++v8brckTkMBT0csTOG9yNKSN68NyH6/l0Q4nX5YhIExT0clTuvXAQx3VO4Y5ZBezeX+V1OSLSCAW9HJUOiXFMnzKcbaUV3PPmMh1yKRKGFPRy1HJ7pHP72f14a/G3vFmw2etyRKQBBb2ExA/OOJ4RvTK4783lbNqxz+tyRCSAgl5CIjbGeOKKXADueKWAGt2oRCRsKOglZHp0TuHBSYNZ+PVOfj9/ndfliIifgl5CalJuNhcNO5bp769l8aZdXpcjIijoJcTMjIcmDeGYtESmzipgb0W11yWJtHsKegm5TsnxPDE5l69L9vLQ2yu8Lkek3VPQS6sY1SeTm5XccRAAAApbSURBVMf25S+fb+KV/E1s21NOVY2uiSPihTivC5DodfvZJ/Dx2mJ+9tqSuucyUuLpkpro+0lLpEtqgn864ZDnE+NiPaxeJHoo6KXVJMTFMOPGU/jXuhKKyyrqf0orKS6rYGnhLorLKilroh8/LSmOrLrw920IMjvUP+6Smuibn5ZASoL+lEWaov8d0qrSkuIZN6TbYduUV9X4NwKVFJcGbBDKKikqq6C4tILVW0v5V1lJk9fTSUmIPeibQWZqIlmpCf5vB4n189ISSUuMw8xaY3VFwpKCXjyXFB9LTkYKORkpzbatrK5lx17fN4IDG4HissqDvjFsLNnHoo072bGvksYuvZMQF+P/pnBgo3Bot9GBbxKdkuOJidFGQSJbUEFvZuOAJ4FY4H+cc480mH86MB04EZjinHstYN61wD3+yYeccy+FonBpnxLiYujWKYlunZKabVtT6+o2Cg27jYr83xi27C5n6ebdlOytbPRs3rgYq9sQZKYevBEI7ELqkppI5w4JxGqjIGGo2aA3s1jgGeAcoBBYaGZznHOBx819A1wH3Nlg2c7A/UAe4IBF/mV3hqZ8kabFxhhZaYlkpSU227a21rF7f9VBG4GG3UjFZRWs21ZKcVkllY0cQRRj0LlDwiHjCIHdRgc2EpmpCcTH6qA3aRvB7NGPBNY55zYAmNlMYCJQF/TOua/98xr+9Z8HvOec2+Gf/x4wDvjLUVcuEkIxMUZGhwQyOiTQr2vaYds65yitqD6026i0gqKA6S++2UlxaSX7q2oafZ30lHgyOyTUdRkFdic1PCopKV5HIMmRCybos4FNAdOFwClBvn5jy2Y3bGRmNwE3AfTs2TPIlxbxhpnRMSmejknx9Mlqvv2+ymqKS/0Dyw26kA78rPh2D8VlFZSWN34EUmpiXIONwMHfGLICplMSYjXYLAcJi8FY59wLwAsAeXl5uuyhRJWUhDh6ZsbRM7P5webyqhpK9jZy9FHA9LqiMj77qoKd+xo/AikpPqbJjUBgN9KxnZJJTtA3hfYgmKDfDPQImM7xPxeMzcAZDZadH+SyIu1OUnws2enJZKcnN9u2qsZ3BFL9RqCyrgvpwHThzn0UbNrFjr0VNHbl6M4dEureLycjmewM3+PsjGRy0lPomKxDUaNBMEG/EOhnZr3xBfcU4KogX38u8N9mluGfPhf4eYurFJFDxMfG0LVjEl07BncE0s59lXXdRkVl5Xy7q5zCnfvZvGs/a7eXMn/NdsqrDh5mS02Mqwv+hr9z0pPpkpqow08jQLNB75yrNrNb8IV2LPCic265mT0I5Dvn5pjZCGA2kAFcZGYPOOcGO+d2mNkv8W0sAB48MDArIm0nNsbqum5o4vw153yHo27etZ/NO/fXbQQO/M7/egd7GowhJMTFcGynpPoNQHqKbyPgn+7WKUlHF4UBC7ebOefl5bn8/HyvyxCRRpSWV9VtCOo2CAHTRaUVB7WPMejWManBN4KUgA2DxglCxcwWOefyGpsXFoOxIhIZ0pLiGdAtngHdOjY6v7yqhi27y/3Bv++gDUH+xp28tWTLISemZXZIOCj4D+oeykihU3J8W6xaVFPQi0jIJMXH0rtLB3p36dDo/OqaWraVVhy0ITjQPbR6Wyn/XLWdiuqDxwnSEuMaHSM48DsrNVEDxs1Q0ItIm4mLjQk4qqjzIfOdc5TsrTyoayhwnGBhE+MEdd8GGtkYdO+URFw7HydQ0ItI2DCrHzQe1iO90TaB4wSFDcYKPli1neKyxscJcjJSmvxmEO1nHivoRSSiBDNO8O2u/Y0OGn/+1Q627ik/ZJygS2rCwRuA9GSyM1Lqnov0cQIFvYhElaT4WPpkpdInK7XR+YcbJ1i1pZQPVjY/TpDT4OihLqkJYT1OoKAXkXYlmHGC4rLKgG8EB28MPv96xyHXJEo8ME7QxNFD3Tp6O06goBcRCWBWf3nr3CbGCfaUV/nC/0DXUED30MqVh44TxMaY73yCRsYHcjKSObaVxwkU9CIiLdQxKZ6O3eMZ2L3pcYKGYwQHfjc9TpDIqD6defqqk0Jer4JeRCTEkuJj6ZuVSt/DjBNs3VN+yIagc4eEVqlHQS8i0sbiYmOCvk9yKLTvswhERNoBBb2ISJRT0IuIRDkFvYhIlFPQi4hEOQW9iEiUU9CLiEQ5Bb2ISJQLu3vGmlkRsPEoXqILUByicrwULesBWpdwFS3rEi3rAUe3Lsc557IamxF2QX+0zCy/qRvkRpJoWQ/QuoSraFmXaFkPaL11UdeNiEiUU9CLiES5aAz6F7wuIESiZT1A6xKuomVdomU9oJXWJer66EVE5GDRuEcvIiIBFPQiIlEuIoPezMaZ2WozW2dm0xqZn2hms/zzPzOzXm1fZXCCWJfrzKzIzAr8P//pRZ3NMbMXzWy7mS1rYr6Z2VP+9VxiZqG/X1qIBLEuZ5jZ7oDP5L62rjEYZtbDzOaZ2QozW25mtzXSJiI+lyDXJVI+lyQz+9zMFvvX5YFG2oQ2w5xzEfUDxALrgT5AArAYGNSgzQ+B5/yPpwCzvK77KNblOuBpr2sNYl1OB04CljUx/wLgXcCAUcBnXtd8FOtyBvA3r+sMYj26Ayf5H6cBaxr5+4qIzyXIdYmUz8WAVP/jeOAzYFSDNiHNsEjcox8JrHPObXDOVQIzgYkN2kwEXvI/fg04y8ysDWsMVjDrEhGccwuAHYdpMhH4k/P5FEg3s+5tU13LBLEuEcE5t8U594X/cSmwEshu0CwiPpcg1yUi+P+ty/yT8f6fhkfFhDTDIjHos4FNAdOFHPqB17VxzlUDu4HMNqmuZYJZF4BL/V+rXzOzHm1TWsgFu66R4j/8X73fNbPBXhfTHP9X/+H49h4DRdzncph1gQj5XMws1swKgO3Ae865Jj+XUGRYJAZ9e/MW0Ms5dyLwHvVbefHOF/iuKzIM+B3wpsf1HJaZpQKvA1Odc3u8rudoNLMuEfO5OOdqnHO5QA4w0syGtOb7RWLQbwYC92pz/M812sbM4oBOQEmbVNcyza6Lc67EOVfhn/wf4OQ2qi3UgvncIoJzbs+Br97OuXeAeDPr4nFZjTKzeHzB+LJz7o1GmkTM59LcukTS53KAc24XMA8Y12BWSDMsEoN+IdDPzHqbWQK+gYo5DdrMAa71P74M+Kfzj2qEmWbXpUF/6QR8fZORaA7wPf9RHqOA3c65LV4XdSTMrNuB/lIzG4nv/1HY7Uj4a/xfYKVz7okmmkXE5xLMukTQ55JlZun+x8nAOcCqBs1CmmFxR7qgV5xz1WZ2CzAX31ErLzrnlpvZg0C+c24Ovj+IP5vZOnyDalO8q7hpQa7LrWY2AajGty7XeVbwYZjZX/Ad9dDFzAqB+/ENMuGcew54B98RHuuAfcD3vam0eUGsy2XAD8ysGtgPTAnTHYnRwHeBpf7+YIC7gJ4QcZ9LMOsSKZ9Ld+AlM4vFtzF6xTn3t9bMMF0CQUQkykVi142IiLSAgl5EJMop6EVEopyCXkQkyinoRUSinIJeRCTKKehFRKLc/weQCGiBk8TkCwAAAABJRU5ErkJggg==\n",
"text/plain": [
"<Figure size 432x288 with 1 Axes>"
]
},
"metadata": {
"tags": [],
"needs_background": "light"
}
},
{
"output_type": "display_data",
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 432x288 with 1 Axes>"
]
},
"metadata": {
"tags": [],
"needs_background": "light"
}
}
]
},
{
"cell_type": "code",
"metadata": {
"id": "G0brbE1ZfzJk"
},
"source": [
"# save it as a h5 file\n",
"\n",
"import tensorflow as tf\n",
"\n",
"from keras.models import load_model\n",
"\n",
"model.save('model_vgg16.h5')"
],
"execution_count": 24,
"outputs": []
},
{
"cell_type": "markdown",
"metadata": {
"id": "KTUnzHN491Yf"
},
"source": [
"After saving our model we upload it in order to diagnose new patients."
]
},
{
"cell_type": "code",
"metadata": {
"id": "kckKTOe0LC-Z"
},
"source": [
"from keras.models import load_model\r\n",
"from keras.preprocessing import image\r\n",
"from keras.applications.vgg16 import preprocess_input\r\n",
"import numpy as np"
],
"execution_count": 25,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"id": "21GJ2fM2LQc0"
},
"source": [
"model=load_model('model_vgg16.h5')"
],
"execution_count": 26,
"outputs": []
},
{
"cell_type": "markdown",
"metadata": {
"id": "IEuhi7xZ7mqo"
},
"source": [
"Now we are going to test the validation of our model. For that we consider our validation set which contains 16 images: 8 NORMAL and 8 PNEUMONIA.\r\n",
"\r\n",
"\r\n",
"> For example we take one image from the class PNEUMONIA.\r\n",
"\r\n"
]
},
{
"cell_type": "code",
"metadata": {
"id": "mGb7vj0TL6AW"
},
"source": [
"img=image.load_img('/content/chest_xray/val/PNEUMONIA/person1947_bacteria_4876.jpeg',target_size=(224,224))"
],
"execution_count": 27,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/",
"height": 0
},
"id": "8C8Ll1DFMTWl",
"outputId": "3f9ec8e2-f3fa-4367-b468-2433e46334b5"
},
"source": [
"x=image.img_to_array(img)\r\n",
"x=np.expand_dims(x,axis=0)\r\n",
"img_data=preprocess_input(x)\r\n",
"classes=model.predict(img_data)\r\n",
"classes"
],
"execution_count": 28,
"outputs": [
{
"output_type": "execute_result",
"data": {
"text/plain": [
"array([[0., 1.]], dtype=float32)"
]
},
"metadata": {
"tags": []
},
"execution_count": 28
}
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "yj1hYJ3l8M2k"
},
"source": [
"As we can see above the prediction is correct. This model can assist non specialist doctors to diagnose pneumonia disease.\r\n",
"\r\n",
"\r\n",
"\r\n"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "WLcvYLVM_nrf"
},
"source": [
"## VGG19"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "9ppzUzEUTXP-"
},
"source": [
"The following architecture is taken from: https://saicharanars.medium.com/building-vgg19-with-keras-f516101c24cf"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "Uw2Ae3u9S8wQ"
},
"source": [
""
]
},
{
"cell_type": "code",
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/",
"height": 0
},
"id": "cbvw7oPI_sg9",
"outputId": "dda30c84-609e-4b4d-dfb2-33f3e5ac0bc1"
},
"source": [
"from keras.applications.vgg19 import VGG19\r\n",
"IMAGE_SIZE=[224,224]\r\n",
"vgg = VGG19(input_shape=IMAGE_SIZE + [3], weights='imagenet', include_top=False)"
],
"execution_count": 29,
"outputs": [
{
"output_type": "stream",
"text": [
"Downloading data from https://storage.googleapis.com/tensorflow/keras-applications/vgg19/vgg19_weights_tf_dim_ordering_tf_kernels_notop.h5\n",
"80142336/80134624 [==============================] - 0s 0us/step\n"
],
"name": "stdout"
}
]
},
{
"cell_type": "code",
"metadata": {
"id": "1o7nqXn9_87x"
},
"source": [
"# we keep existing weights\r\n",
"for layer in vgg.layers:\r\n",
" layer.trainable = False"
],
"execution_count": 30,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"id": "L4ZgnniEADQq"
},
"source": [
"# our layers\r\n",
"x = Flatten()(vgg.output)"
],
"execution_count": 31,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"id": "Wv3EAuZtAIEd"
},
"source": [
"prediction = Dense(len(folders), activation='softmax')(x)\r\n",
"\r\n",
"# create a model object\r\n",
"model = Model(inputs=vgg.input, outputs=prediction)"
],
"execution_count": 32,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/",
"height": 0
},
"id": "wTjb5Gz7AMjU",
"outputId": "13e18504-65ae-4095-c86b-c067d6e6ab6a"
},
"source": [
"# view the structure of the model\r\n",
"model.summary()"
],
"execution_count": 33,
"outputs": [
{
"output_type": "stream",
"text": [
"Model: \"model_1\"\n",
"_________________________________________________________________\n",
"Layer (type) Output Shape Param # \n",
"=================================================================\n",
"input_2 (InputLayer) [(None, 224, 224, 3)] 0 \n",
"_________________________________________________________________\n",
"block1_conv1 (Conv2D) (None, 224, 224, 64) 1792 \n",
"_________________________________________________________________\n",
"block1_conv2 (Conv2D) (None, 224, 224, 64) 36928 \n",
"_________________________________________________________________\n",
"block1_pool (MaxPooling2D) (None, 112, 112, 64) 0 \n",
"_________________________________________________________________\n",
"block2_conv1 (Conv2D) (None, 112, 112, 128) 73856 \n",
"_________________________________________________________________\n",
"block2_conv2 (Conv2D) (None, 112, 112, 128) 147584 \n",
"_________________________________________________________________\n",
"block2_pool (MaxPooling2D) (None, 56, 56, 128) 0 \n",
"_________________________________________________________________\n",
"block3_conv1 (Conv2D) (None, 56, 56, 256) 295168 \n",
"_________________________________________________________________\n",
"block3_conv2 (Conv2D) (None, 56, 56, 256) 590080 \n",
"_________________________________________________________________\n",
"block3_conv3 (Conv2D) (None, 56, 56, 256) 590080 \n",
"_________________________________________________________________\n",
"block3_conv4 (Conv2D) (None, 56, 56, 256) 590080 \n",
"_________________________________________________________________\n",
"block3_pool (MaxPooling2D) (None, 28, 28, 256) 0 \n",
"_________________________________________________________________\n",
"block4_conv1 (Conv2D) (None, 28, 28, 512) 1180160 \n",
"_________________________________________________________________\n",
"block4_conv2 (Conv2D) (None, 28, 28, 512) 2359808 \n",
"_________________________________________________________________\n",
"block4_conv3 (Conv2D) (None, 28, 28, 512) 2359808 \n",
"_________________________________________________________________\n",
"block4_conv4 (Conv2D) (None, 28, 28, 512) 2359808 \n",
"_________________________________________________________________\n",
"block4_pool (MaxPooling2D) (None, 14, 14, 512) 0 \n",
"_________________________________________________________________\n",
"block5_conv1 (Conv2D) (None, 14, 14, 512) 2359808 \n",
"_________________________________________________________________\n",
"block5_conv2 (Conv2D) (None, 14, 14, 512) 2359808 \n",
"_________________________________________________________________\n",
"block5_conv3 (Conv2D) (None, 14, 14, 512) 2359808 \n",
"_________________________________________________________________\n",
"block5_conv4 (Conv2D) (None, 14, 14, 512) 2359808 \n",
"_________________________________________________________________\n",
"block5_pool (MaxPooling2D) (None, 7, 7, 512) 0 \n",
"_________________________________________________________________\n",
"flatten_1 (Flatten) (None, 25088) 0 \n",
"_________________________________________________________________\n",
"dense_1 (Dense) (None, 2) 50178 \n",
"=================================================================\n",
"Total params: 20,074,562\n",
"Trainable params: 50,178\n",
"Non-trainable params: 20,024,384\n",
"_________________________________________________________________\n"
],
"name": "stdout"
}
]
},
{
"cell_type": "code",
"metadata": {
"id": "xeKze9IWAXLT"
},
"source": [
"# Use the Image Data Generator to import the images from the dataset\r\n",
"from keras.preprocessing.image import ImageDataGenerator\r\n",
"\r\n",
"train_datagen = ImageDataGenerator(rescale = 1./255,\r\n",
" shear_range = 0.2,\r\n",
" zoom_range = 0.2,\r\n",
" horizontal_flip = True)\r\n",
"\r\n",
"test_datagen = ImageDataGenerator(rescale = 1./255)"
],
"execution_count": 34,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/",
"height": 0
},
"id": "Zud2fG6iAceB",
"outputId": "ee23d527-dfe3-4e28-ebc3-548a19e2b725"
},
"source": [
"# Make sure you provide the same target size as initialied for the image size\r\n",
"training_set = train_datagen.flow_from_directory('/content/chest_xray/train',\r\n",
" target_size = (224, 224),\r\n",
" batch_size = 32,\r\n",
" class_mode = 'categorical')"
],
"execution_count": 35,
"outputs": [
{
"output_type": "stream",
"text": [
"Found 5216 images belonging to 2 classes.\n"
],
"name": "stdout"
}
]
},
{
"cell_type": "code",
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/",
"height": 0
},
"id": "0YT6c4uNAeus",
"outputId": "802e50cb-4528-47fa-e661-1bbe14edac97"
},
"source": [
"test_set = test_datagen.flow_from_directory('/content/chest_xray/test',\r\n",
" target_size = (224, 224),\r\n",
" batch_size = 32,\r\n",
" class_mode = 'categorical')"
],
"execution_count": 36,
"outputs": [
{
"output_type": "stream",
"text": [
"Found 624 images belonging to 2 classes.\n"
],
"name": "stdout"
}
]
},
{
"cell_type": "code",
"metadata": {
"id": "rCncY7zVAlCv"
},
"source": [
"# tell the model what cost and optimization method to use\r\n",
"model.compile(\r\n",
" loss='categorical_crossentropy',\r\n",
" optimizer='adam',\r\n",
" metrics=['accuracy']\r\n",
")"
],
"execution_count": 37,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/",
"height": 0
},
"id": "E0LZ8hcHAwAu",
"outputId": "78ac1a20-b858-4571-e2ce-5c7e1ca6380a"
},
"source": [
"r = model.fit_generator(\r\n",
" training_set,\r\n",
" validation_data=test_set,\r\n",
" epochs=4,\r\n",
" steps_per_epoch=len(training_set),\r\n",
" validation_steps=len(test_set)\r\n",
")"
],
"execution_count": 38,
"outputs": [
{
"output_type": "stream",
"text": [
"/usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/engine/training.py:1844: UserWarning: `Model.fit_generator` is deprecated and will be removed in a future version. Please use `Model.fit`, which supports generators.\n",
" warnings.warn('`Model.fit_generator` is deprecated and '\n"
],
"name": "stderr"
},
{
"output_type": "stream",
"text": [
"Epoch 1/4\n",
"163/163 [==============================] - 105s 642ms/step - loss: 0.6380 - accuracy: 0.8245 - val_loss: 0.4249 - val_accuracy: 0.8413\n",
"Epoch 2/4\n",
"163/163 [==============================] - 104s 636ms/step - loss: 0.1313 - accuracy: 0.9440 - val_loss: 0.2794 - val_accuracy: 0.9038\n",
"Epoch 3/4\n",
"163/163 [==============================] - 104s 636ms/step - loss: 0.1336 - accuracy: 0.9518 - val_loss: 0.3817 - val_accuracy: 0.8766\n",
"Epoch 4/4\n",
"163/163 [==============================] - 104s 636ms/step - loss: 0.0992 - accuracy: 0.9633 - val_loss: 0.3168 - val_accuracy: 0.9103\n"
],
"name": "stdout"
}
]
},
{
"cell_type": "code",
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/",
"height": 0
},
"id": "SlitKTm5znbR",
"outputId": "d92eb0a8-5161-4e34-ac2e-f92c2fe40373"
},
"source": [
"(619+612+611+614)/4"
],
"execution_count": 39,
"outputs": [
{
"output_type": "execute_result",
"data": {
"text/plain": [
"614.0"
]
},
"metadata": {
"tags": []
},
"execution_count": 39
}
]
},
{
"cell_type": "code",
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/",
"height": 543
},
"id": "c59b_-pJA5Kb",
"outputId": "cbdb7a1d-e395-4c80-9233-7f3da0af0b5a"
},
"source": [
"# plot the loss\r\n",
"plt.plot(r.history['loss'], label='train loss')\r\n",
"plt.plot(r.history['val_loss'], label='val loss')\r\n",
"plt.legend()\r\n",
"plt.show()\r\n",
"plt.savefig('LossVal_loss')\r\n",
"\r\n",
"# plot the accuracy\r\n",
"acc_train = r.history['accuracy']\r\n",
"acc_val = r.history['val_accuracy']\r\n",
"epochs = range(1,5)\r\n",
"plt.plot(epochs, acc_train, 'g', label='Training accuracy')\r\n",
"plt.plot(epochs, acc_val, 'b', label='validation accuracy')\r\n",
"plt.title('Training and Validation accuracy')\r\n",
"plt.xlabel('Epochs')\r\n",
"plt.ylabel('Accuracy')\r\n",
"plt.legend()\r\n",
"plt.show()"
],
"execution_count": 40,
"outputs": [
{
"output_type": "display_data",
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 432x288 with 1 Axes>"
]
},
"metadata": {
"tags": [],
"needs_background": "light"
}
},
{
"output_type": "display_data",
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 432x288 with 1 Axes>"
]
},
"metadata": {
"tags": [],
"needs_background": "light"
}
}
]
},
{
"cell_type": "code",
"metadata": {
"id": "wnhv5yZ1A5_Q"
},
"source": [
"# save it as a h5 file\r\n",
"\r\n",
"import tensorflow as tf\r\n",
"\r\n",
"from keras.models import load_model\r\n",
"\r\n",
"model.save('model_vgg19.h5')"
],
"execution_count": 41,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"id": "ZHjc2SVCA-z1"
},
"source": [
"from keras.models import load_model\r\n",
"from keras.preprocessing import image\r\n",
"from keras.applications.vgg19 import preprocess_input\r\n",
"import numpy as np"
],
"execution_count": 42,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"id": "XO9Ar59EBEQe"
},
"source": [
"model=load_model('model_vgg19.h5')"
],
"execution_count": 43,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"id": "dzl3dyZ-BTki"
},
"source": [
"img=image.load_img('/content/chest_xray/val/PNEUMONIA/person1947_bacteria_4876.jpeg',target_size=(224,224))"
],
"execution_count": 44,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/",
"height": 0
},
"id": "tEsWDHP3BZEv",
"outputId": "7cdac1c4-09fa-4f63-fd21-45f4e50bba92"
},
"source": [
"x=image.img_to_array(img)\r\n",
"x=np.expand_dims(x,axis=0)\r\n",
"img_data=preprocess_input(x)\r\n",
"classes=model.predict(img_data)\r\n",
"classes"
],
"execution_count": 45,
"outputs": [
{
"output_type": "execute_result",
"data": {
"text/plain": [
"array([[0.0016, 0.9984]], dtype=float32)"
]
},
"metadata": {
"tags": []
},
"execution_count": 45
}
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "tmqQCbtoDh7u"
},
"source": [
"## InceptionV3"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "xYnx21BDTi0Q"
},
"source": [
"The following architecture is taken from: https://paperswithcode.com/method/inception-v3#"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "TgYo0kZIUSb8"
},
"source": [
""
]
},
{
"cell_type": "code",
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/",
"height": 0
},
"id": "qHF7qSgwDuWJ",
"outputId": "dc59df7a-857a-4074-b0e0-5e2f17f4a7f6"
},
"source": [
"from keras.applications.inception_v3 import InceptionV3\r\n",
"IMAGE_SIZE=[224,224]\r\n",
"vgg = InceptionV3(input_shape=IMAGE_SIZE + [3], weights='imagenet', include_top=False)"
],
"execution_count": 46,
"outputs": [
{
"output_type": "stream",
"text": [
"Downloading data from https://storage.googleapis.com/tensorflow/keras-applications/inception_v3/inception_v3_weights_tf_dim_ordering_tf_kernels_notop.h5\n",
"87916544/87910968 [==============================] - 1s 0us/step\n"
],
"name": "stdout"
}
]
},
{
"cell_type": "code",
"metadata": {
"id": "SOkv8MqHF1F9"
},
"source": [
"# we keep existing weights\r\n",
"for layer in vgg.layers:\r\n",
" layer.trainable = False"
],
"execution_count": 47,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"id": "oqhOnpIzF9vx"
},
"source": [
"# our layers\r\n",
"x = Flatten()(vgg.output)"
],
"execution_count": 48,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"id": "DX_Q9eX-GDY8"
},
"source": [
"prediction = Dense(len(folders), activation='softmax')(x)\r\n",
"\r\n",
"# create a model object\r\n",
"model = Model(inputs=vgg.input, outputs=prediction)"
],
"execution_count": 49,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/",
"height": 0
},
"id": "ebp5F2PfGIuj",
"outputId": "0003b7fc-dc31-4004-cc6b-df4bc2d6ff2f"
},
"source": [
"# view the structure of the model\r\n",
"model.summary()"
],
"execution_count": 50,
"outputs": [
{
"output_type": "stream",
"text": [
"Model: \"model_2\"\n",
"__________________________________________________________________________________________________\n",
"Layer (type) Output Shape Param # Connected to \n",
"==================================================================================================\n",
"input_3 (InputLayer) [(None, 224, 224, 3) 0 \n",
"__________________________________________________________________________________________________\n",
"conv2d (Conv2D) (None, 111, 111, 32) 864 input_3[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization (BatchNorma (None, 111, 111, 32) 96 conv2d[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation (Activation) (None, 111, 111, 32) 0 batch_normalization[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_1 (Conv2D) (None, 109, 109, 32) 9216 activation[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_1 (BatchNor (None, 109, 109, 32) 96 conv2d_1[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_1 (Activation) (None, 109, 109, 32) 0 batch_normalization_1[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_2 (Conv2D) (None, 109, 109, 64) 18432 activation_1[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_2 (BatchNor (None, 109, 109, 64) 192 conv2d_2[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_2 (Activation) (None, 109, 109, 64) 0 batch_normalization_2[0][0] \n",
"__________________________________________________________________________________________________\n",
"max_pooling2d (MaxPooling2D) (None, 54, 54, 64) 0 activation_2[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_3 (Conv2D) (None, 54, 54, 80) 5120 max_pooling2d[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_3 (BatchNor (None, 54, 54, 80) 240 conv2d_3[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_3 (Activation) (None, 54, 54, 80) 0 batch_normalization_3[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_4 (Conv2D) (None, 52, 52, 192) 138240 activation_3[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_4 (BatchNor (None, 52, 52, 192) 576 conv2d_4[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_4 (Activation) (None, 52, 52, 192) 0 batch_normalization_4[0][0] \n",
"__________________________________________________________________________________________________\n",
"max_pooling2d_1 (MaxPooling2D) (None, 25, 25, 192) 0 activation_4[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_8 (Conv2D) (None, 25, 25, 64) 12288 max_pooling2d_1[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_8 (BatchNor (None, 25, 25, 64) 192 conv2d_8[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_8 (Activation) (None, 25, 25, 64) 0 batch_normalization_8[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_6 (Conv2D) (None, 25, 25, 48) 9216 max_pooling2d_1[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_9 (Conv2D) (None, 25, 25, 96) 55296 activation_8[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_6 (BatchNor (None, 25, 25, 48) 144 conv2d_6[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_9 (BatchNor (None, 25, 25, 96) 288 conv2d_9[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_6 (Activation) (None, 25, 25, 48) 0 batch_normalization_6[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_9 (Activation) (None, 25, 25, 96) 0 batch_normalization_9[0][0] \n",
"__________________________________________________________________________________________________\n",
"average_pooling2d (AveragePooli (None, 25, 25, 192) 0 max_pooling2d_1[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_5 (Conv2D) (None, 25, 25, 64) 12288 max_pooling2d_1[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_7 (Conv2D) (None, 25, 25, 64) 76800 activation_6[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_10 (Conv2D) (None, 25, 25, 96) 82944 activation_9[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_11 (Conv2D) (None, 25, 25, 32) 6144 average_pooling2d[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_5 (BatchNor (None, 25, 25, 64) 192 conv2d_5[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_7 (BatchNor (None, 25, 25, 64) 192 conv2d_7[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_10 (BatchNo (None, 25, 25, 96) 288 conv2d_10[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_11 (BatchNo (None, 25, 25, 32) 96 conv2d_11[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_5 (Activation) (None, 25, 25, 64) 0 batch_normalization_5[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_7 (Activation) (None, 25, 25, 64) 0 batch_normalization_7[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_10 (Activation) (None, 25, 25, 96) 0 batch_normalization_10[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_11 (Activation) (None, 25, 25, 32) 0 batch_normalization_11[0][0] \n",
"__________________________________________________________________________________________________\n",
"mixed0 (Concatenate) (None, 25, 25, 256) 0 activation_5[0][0] \n",
" activation_7[0][0] \n",
" activation_10[0][0] \n",
" activation_11[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_15 (Conv2D) (None, 25, 25, 64) 16384 mixed0[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_15 (BatchNo (None, 25, 25, 64) 192 conv2d_15[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_15 (Activation) (None, 25, 25, 64) 0 batch_normalization_15[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_13 (Conv2D) (None, 25, 25, 48) 12288 mixed0[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_16 (Conv2D) (None, 25, 25, 96) 55296 activation_15[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_13 (BatchNo (None, 25, 25, 48) 144 conv2d_13[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_16 (BatchNo (None, 25, 25, 96) 288 conv2d_16[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_13 (Activation) (None, 25, 25, 48) 0 batch_normalization_13[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_16 (Activation) (None, 25, 25, 96) 0 batch_normalization_16[0][0] \n",
"__________________________________________________________________________________________________\n",
"average_pooling2d_1 (AveragePoo (None, 25, 25, 256) 0 mixed0[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_12 (Conv2D) (None, 25, 25, 64) 16384 mixed0[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_14 (Conv2D) (None, 25, 25, 64) 76800 activation_13[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_17 (Conv2D) (None, 25, 25, 96) 82944 activation_16[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_18 (Conv2D) (None, 25, 25, 64) 16384 average_pooling2d_1[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_12 (BatchNo (None, 25, 25, 64) 192 conv2d_12[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_14 (BatchNo (None, 25, 25, 64) 192 conv2d_14[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_17 (BatchNo (None, 25, 25, 96) 288 conv2d_17[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_18 (BatchNo (None, 25, 25, 64) 192 conv2d_18[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_12 (Activation) (None, 25, 25, 64) 0 batch_normalization_12[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_14 (Activation) (None, 25, 25, 64) 0 batch_normalization_14[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_17 (Activation) (None, 25, 25, 96) 0 batch_normalization_17[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_18 (Activation) (None, 25, 25, 64) 0 batch_normalization_18[0][0] \n",
"__________________________________________________________________________________________________\n",
"mixed1 (Concatenate) (None, 25, 25, 288) 0 activation_12[0][0] \n",
" activation_14[0][0] \n",
" activation_17[0][0] \n",
" activation_18[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_22 (Conv2D) (None, 25, 25, 64) 18432 mixed1[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_22 (BatchNo (None, 25, 25, 64) 192 conv2d_22[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_22 (Activation) (None, 25, 25, 64) 0 batch_normalization_22[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_20 (Conv2D) (None, 25, 25, 48) 13824 mixed1[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_23 (Conv2D) (None, 25, 25, 96) 55296 activation_22[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_20 (BatchNo (None, 25, 25, 48) 144 conv2d_20[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_23 (BatchNo (None, 25, 25, 96) 288 conv2d_23[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_20 (Activation) (None, 25, 25, 48) 0 batch_normalization_20[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_23 (Activation) (None, 25, 25, 96) 0 batch_normalization_23[0][0] \n",
"__________________________________________________________________________________________________\n",
"average_pooling2d_2 (AveragePoo (None, 25, 25, 288) 0 mixed1[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_19 (Conv2D) (None, 25, 25, 64) 18432 mixed1[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_21 (Conv2D) (None, 25, 25, 64) 76800 activation_20[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_24 (Conv2D) (None, 25, 25, 96) 82944 activation_23[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_25 (Conv2D) (None, 25, 25, 64) 18432 average_pooling2d_2[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_19 (BatchNo (None, 25, 25, 64) 192 conv2d_19[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_21 (BatchNo (None, 25, 25, 64) 192 conv2d_21[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_24 (BatchNo (None, 25, 25, 96) 288 conv2d_24[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_25 (BatchNo (None, 25, 25, 64) 192 conv2d_25[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_19 (Activation) (None, 25, 25, 64) 0 batch_normalization_19[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_21 (Activation) (None, 25, 25, 64) 0 batch_normalization_21[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_24 (Activation) (None, 25, 25, 96) 0 batch_normalization_24[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_25 (Activation) (None, 25, 25, 64) 0 batch_normalization_25[0][0] \n",
"__________________________________________________________________________________________________\n",
"mixed2 (Concatenate) (None, 25, 25, 288) 0 activation_19[0][0] \n",
" activation_21[0][0] \n",
" activation_24[0][0] \n",
" activation_25[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_27 (Conv2D) (None, 25, 25, 64) 18432 mixed2[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_27 (BatchNo (None, 25, 25, 64) 192 conv2d_27[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_27 (Activation) (None, 25, 25, 64) 0 batch_normalization_27[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_28 (Conv2D) (None, 25, 25, 96) 55296 activation_27[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_28 (BatchNo (None, 25, 25, 96) 288 conv2d_28[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_28 (Activation) (None, 25, 25, 96) 0 batch_normalization_28[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_26 (Conv2D) (None, 12, 12, 384) 995328 mixed2[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_29 (Conv2D) (None, 12, 12, 96) 82944 activation_28[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_26 (BatchNo (None, 12, 12, 384) 1152 conv2d_26[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_29 (BatchNo (None, 12, 12, 96) 288 conv2d_29[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_26 (Activation) (None, 12, 12, 384) 0 batch_normalization_26[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_29 (Activation) (None, 12, 12, 96) 0 batch_normalization_29[0][0] \n",
"__________________________________________________________________________________________________\n",
"max_pooling2d_2 (MaxPooling2D) (None, 12, 12, 288) 0 mixed2[0][0] \n",
"__________________________________________________________________________________________________\n",
"mixed3 (Concatenate) (None, 12, 12, 768) 0 activation_26[0][0] \n",
" activation_29[0][0] \n",
" max_pooling2d_2[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_34 (Conv2D) (None, 12, 12, 128) 98304 mixed3[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_34 (BatchNo (None, 12, 12, 128) 384 conv2d_34[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_34 (Activation) (None, 12, 12, 128) 0 batch_normalization_34[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_35 (Conv2D) (None, 12, 12, 128) 114688 activation_34[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_35 (BatchNo (None, 12, 12, 128) 384 conv2d_35[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_35 (Activation) (None, 12, 12, 128) 0 batch_normalization_35[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_31 (Conv2D) (None, 12, 12, 128) 98304 mixed3[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_36 (Conv2D) (None, 12, 12, 128) 114688 activation_35[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_31 (BatchNo (None, 12, 12, 128) 384 conv2d_31[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_36 (BatchNo (None, 12, 12, 128) 384 conv2d_36[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_31 (Activation) (None, 12, 12, 128) 0 batch_normalization_31[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_36 (Activation) (None, 12, 12, 128) 0 batch_normalization_36[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_32 (Conv2D) (None, 12, 12, 128) 114688 activation_31[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_37 (Conv2D) (None, 12, 12, 128) 114688 activation_36[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_32 (BatchNo (None, 12, 12, 128) 384 conv2d_32[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_37 (BatchNo (None, 12, 12, 128) 384 conv2d_37[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_32 (Activation) (None, 12, 12, 128) 0 batch_normalization_32[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_37 (Activation) (None, 12, 12, 128) 0 batch_normalization_37[0][0] \n",
"__________________________________________________________________________________________________\n",
"average_pooling2d_3 (AveragePoo (None, 12, 12, 768) 0 mixed3[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_30 (Conv2D) (None, 12, 12, 192) 147456 mixed3[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_33 (Conv2D) (None, 12, 12, 192) 172032 activation_32[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_38 (Conv2D) (None, 12, 12, 192) 172032 activation_37[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_39 (Conv2D) (None, 12, 12, 192) 147456 average_pooling2d_3[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_30 (BatchNo (None, 12, 12, 192) 576 conv2d_30[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_33 (BatchNo (None, 12, 12, 192) 576 conv2d_33[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_38 (BatchNo (None, 12, 12, 192) 576 conv2d_38[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_39 (BatchNo (None, 12, 12, 192) 576 conv2d_39[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_30 (Activation) (None, 12, 12, 192) 0 batch_normalization_30[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_33 (Activation) (None, 12, 12, 192) 0 batch_normalization_33[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_38 (Activation) (None, 12, 12, 192) 0 batch_normalization_38[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_39 (Activation) (None, 12, 12, 192) 0 batch_normalization_39[0][0] \n",
"__________________________________________________________________________________________________\n",
"mixed4 (Concatenate) (None, 12, 12, 768) 0 activation_30[0][0] \n",
" activation_33[0][0] \n",
" activation_38[0][0] \n",
" activation_39[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_44 (Conv2D) (None, 12, 12, 160) 122880 mixed4[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_44 (BatchNo (None, 12, 12, 160) 480 conv2d_44[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_44 (Activation) (None, 12, 12, 160) 0 batch_normalization_44[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_45 (Conv2D) (None, 12, 12, 160) 179200 activation_44[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_45 (BatchNo (None, 12, 12, 160) 480 conv2d_45[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_45 (Activation) (None, 12, 12, 160) 0 batch_normalization_45[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_41 (Conv2D) (None, 12, 12, 160) 122880 mixed4[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_46 (Conv2D) (None, 12, 12, 160) 179200 activation_45[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_41 (BatchNo (None, 12, 12, 160) 480 conv2d_41[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_46 (BatchNo (None, 12, 12, 160) 480 conv2d_46[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_41 (Activation) (None, 12, 12, 160) 0 batch_normalization_41[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_46 (Activation) (None, 12, 12, 160) 0 batch_normalization_46[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_42 (Conv2D) (None, 12, 12, 160) 179200 activation_41[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_47 (Conv2D) (None, 12, 12, 160) 179200 activation_46[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_42 (BatchNo (None, 12, 12, 160) 480 conv2d_42[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_47 (BatchNo (None, 12, 12, 160) 480 conv2d_47[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_42 (Activation) (None, 12, 12, 160) 0 batch_normalization_42[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_47 (Activation) (None, 12, 12, 160) 0 batch_normalization_47[0][0] \n",
"__________________________________________________________________________________________________\n",
"average_pooling2d_4 (AveragePoo (None, 12, 12, 768) 0 mixed4[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_40 (Conv2D) (None, 12, 12, 192) 147456 mixed4[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_43 (Conv2D) (None, 12, 12, 192) 215040 activation_42[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_48 (Conv2D) (None, 12, 12, 192) 215040 activation_47[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_49 (Conv2D) (None, 12, 12, 192) 147456 average_pooling2d_4[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_40 (BatchNo (None, 12, 12, 192) 576 conv2d_40[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_43 (BatchNo (None, 12, 12, 192) 576 conv2d_43[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_48 (BatchNo (None, 12, 12, 192) 576 conv2d_48[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_49 (BatchNo (None, 12, 12, 192) 576 conv2d_49[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_40 (Activation) (None, 12, 12, 192) 0 batch_normalization_40[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_43 (Activation) (None, 12, 12, 192) 0 batch_normalization_43[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_48 (Activation) (None, 12, 12, 192) 0 batch_normalization_48[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_49 (Activation) (None, 12, 12, 192) 0 batch_normalization_49[0][0] \n",
"__________________________________________________________________________________________________\n",
"mixed5 (Concatenate) (None, 12, 12, 768) 0 activation_40[0][0] \n",
" activation_43[0][0] \n",
" activation_48[0][0] \n",
" activation_49[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_54 (Conv2D) (None, 12, 12, 160) 122880 mixed5[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_54 (BatchNo (None, 12, 12, 160) 480 conv2d_54[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_54 (Activation) (None, 12, 12, 160) 0 batch_normalization_54[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_55 (Conv2D) (None, 12, 12, 160) 179200 activation_54[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_55 (BatchNo (None, 12, 12, 160) 480 conv2d_55[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_55 (Activation) (None, 12, 12, 160) 0 batch_normalization_55[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_51 (Conv2D) (None, 12, 12, 160) 122880 mixed5[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_56 (Conv2D) (None, 12, 12, 160) 179200 activation_55[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_51 (BatchNo (None, 12, 12, 160) 480 conv2d_51[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_56 (BatchNo (None, 12, 12, 160) 480 conv2d_56[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_51 (Activation) (None, 12, 12, 160) 0 batch_normalization_51[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_56 (Activation) (None, 12, 12, 160) 0 batch_normalization_56[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_52 (Conv2D) (None, 12, 12, 160) 179200 activation_51[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_57 (Conv2D) (None, 12, 12, 160) 179200 activation_56[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_52 (BatchNo (None, 12, 12, 160) 480 conv2d_52[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_57 (BatchNo (None, 12, 12, 160) 480 conv2d_57[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_52 (Activation) (None, 12, 12, 160) 0 batch_normalization_52[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_57 (Activation) (None, 12, 12, 160) 0 batch_normalization_57[0][0] \n",
"__________________________________________________________________________________________________\n",
"average_pooling2d_5 (AveragePoo (None, 12, 12, 768) 0 mixed5[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_50 (Conv2D) (None, 12, 12, 192) 147456 mixed5[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_53 (Conv2D) (None, 12, 12, 192) 215040 activation_52[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_58 (Conv2D) (None, 12, 12, 192) 215040 activation_57[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_59 (Conv2D) (None, 12, 12, 192) 147456 average_pooling2d_5[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_50 (BatchNo (None, 12, 12, 192) 576 conv2d_50[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_53 (BatchNo (None, 12, 12, 192) 576 conv2d_53[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_58 (BatchNo (None, 12, 12, 192) 576 conv2d_58[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_59 (BatchNo (None, 12, 12, 192) 576 conv2d_59[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_50 (Activation) (None, 12, 12, 192) 0 batch_normalization_50[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_53 (Activation) (None, 12, 12, 192) 0 batch_normalization_53[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_58 (Activation) (None, 12, 12, 192) 0 batch_normalization_58[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_59 (Activation) (None, 12, 12, 192) 0 batch_normalization_59[0][0] \n",
"__________________________________________________________________________________________________\n",
"mixed6 (Concatenate) (None, 12, 12, 768) 0 activation_50[0][0] \n",
" activation_53[0][0] \n",
" activation_58[0][0] \n",
" activation_59[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_64 (Conv2D) (None, 12, 12, 192) 147456 mixed6[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_64 (BatchNo (None, 12, 12, 192) 576 conv2d_64[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_64 (Activation) (None, 12, 12, 192) 0 batch_normalization_64[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_65 (Conv2D) (None, 12, 12, 192) 258048 activation_64[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_65 (BatchNo (None, 12, 12, 192) 576 conv2d_65[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_65 (Activation) (None, 12, 12, 192) 0 batch_normalization_65[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_61 (Conv2D) (None, 12, 12, 192) 147456 mixed6[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_66 (Conv2D) (None, 12, 12, 192) 258048 activation_65[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_61 (BatchNo (None, 12, 12, 192) 576 conv2d_61[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_66 (BatchNo (None, 12, 12, 192) 576 conv2d_66[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_61 (Activation) (None, 12, 12, 192) 0 batch_normalization_61[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_66 (Activation) (None, 12, 12, 192) 0 batch_normalization_66[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_62 (Conv2D) (None, 12, 12, 192) 258048 activation_61[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_67 (Conv2D) (None, 12, 12, 192) 258048 activation_66[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_62 (BatchNo (None, 12, 12, 192) 576 conv2d_62[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_67 (BatchNo (None, 12, 12, 192) 576 conv2d_67[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_62 (Activation) (None, 12, 12, 192) 0 batch_normalization_62[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_67 (Activation) (None, 12, 12, 192) 0 batch_normalization_67[0][0] \n",
"__________________________________________________________________________________________________\n",
"average_pooling2d_6 (AveragePoo (None, 12, 12, 768) 0 mixed6[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_60 (Conv2D) (None, 12, 12, 192) 147456 mixed6[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_63 (Conv2D) (None, 12, 12, 192) 258048 activation_62[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_68 (Conv2D) (None, 12, 12, 192) 258048 activation_67[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_69 (Conv2D) (None, 12, 12, 192) 147456 average_pooling2d_6[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_60 (BatchNo (None, 12, 12, 192) 576 conv2d_60[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_63 (BatchNo (None, 12, 12, 192) 576 conv2d_63[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_68 (BatchNo (None, 12, 12, 192) 576 conv2d_68[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_69 (BatchNo (None, 12, 12, 192) 576 conv2d_69[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_60 (Activation) (None, 12, 12, 192) 0 batch_normalization_60[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_63 (Activation) (None, 12, 12, 192) 0 batch_normalization_63[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_68 (Activation) (None, 12, 12, 192) 0 batch_normalization_68[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_69 (Activation) (None, 12, 12, 192) 0 batch_normalization_69[0][0] \n",
"__________________________________________________________________________________________________\n",
"mixed7 (Concatenate) (None, 12, 12, 768) 0 activation_60[0][0] \n",
" activation_63[0][0] \n",
" activation_68[0][0] \n",
" activation_69[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_72 (Conv2D) (None, 12, 12, 192) 147456 mixed7[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_72 (BatchNo (None, 12, 12, 192) 576 conv2d_72[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_72 (Activation) (None, 12, 12, 192) 0 batch_normalization_72[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_73 (Conv2D) (None, 12, 12, 192) 258048 activation_72[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_73 (BatchNo (None, 12, 12, 192) 576 conv2d_73[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_73 (Activation) (None, 12, 12, 192) 0 batch_normalization_73[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_70 (Conv2D) (None, 12, 12, 192) 147456 mixed7[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_74 (Conv2D) (None, 12, 12, 192) 258048 activation_73[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_70 (BatchNo (None, 12, 12, 192) 576 conv2d_70[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_74 (BatchNo (None, 12, 12, 192) 576 conv2d_74[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_70 (Activation) (None, 12, 12, 192) 0 batch_normalization_70[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_74 (Activation) (None, 12, 12, 192) 0 batch_normalization_74[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_71 (Conv2D) (None, 5, 5, 320) 552960 activation_70[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_75 (Conv2D) (None, 5, 5, 192) 331776 activation_74[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_71 (BatchNo (None, 5, 5, 320) 960 conv2d_71[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_75 (BatchNo (None, 5, 5, 192) 576 conv2d_75[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_71 (Activation) (None, 5, 5, 320) 0 batch_normalization_71[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_75 (Activation) (None, 5, 5, 192) 0 batch_normalization_75[0][0] \n",
"__________________________________________________________________________________________________\n",
"max_pooling2d_3 (MaxPooling2D) (None, 5, 5, 768) 0 mixed7[0][0] \n",
"__________________________________________________________________________________________________\n",
"mixed8 (Concatenate) (None, 5, 5, 1280) 0 activation_71[0][0] \n",
" activation_75[0][0] \n",
" max_pooling2d_3[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_80 (Conv2D) (None, 5, 5, 448) 573440 mixed8[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_80 (BatchNo (None, 5, 5, 448) 1344 conv2d_80[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_80 (Activation) (None, 5, 5, 448) 0 batch_normalization_80[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_77 (Conv2D) (None, 5, 5, 384) 491520 mixed8[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_81 (Conv2D) (None, 5, 5, 384) 1548288 activation_80[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_77 (BatchNo (None, 5, 5, 384) 1152 conv2d_77[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_81 (BatchNo (None, 5, 5, 384) 1152 conv2d_81[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_77 (Activation) (None, 5, 5, 384) 0 batch_normalization_77[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_81 (Activation) (None, 5, 5, 384) 0 batch_normalization_81[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_78 (Conv2D) (None, 5, 5, 384) 442368 activation_77[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_79 (Conv2D) (None, 5, 5, 384) 442368 activation_77[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_82 (Conv2D) (None, 5, 5, 384) 442368 activation_81[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_83 (Conv2D) (None, 5, 5, 384) 442368 activation_81[0][0] \n",
"__________________________________________________________________________________________________\n",
"average_pooling2d_7 (AveragePoo (None, 5, 5, 1280) 0 mixed8[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_76 (Conv2D) (None, 5, 5, 320) 409600 mixed8[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_78 (BatchNo (None, 5, 5, 384) 1152 conv2d_78[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_79 (BatchNo (None, 5, 5, 384) 1152 conv2d_79[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_82 (BatchNo (None, 5, 5, 384) 1152 conv2d_82[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_83 (BatchNo (None, 5, 5, 384) 1152 conv2d_83[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_84 (Conv2D) (None, 5, 5, 192) 245760 average_pooling2d_7[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_76 (BatchNo (None, 5, 5, 320) 960 conv2d_76[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_78 (Activation) (None, 5, 5, 384) 0 batch_normalization_78[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_79 (Activation) (None, 5, 5, 384) 0 batch_normalization_79[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_82 (Activation) (None, 5, 5, 384) 0 batch_normalization_82[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_83 (Activation) (None, 5, 5, 384) 0 batch_normalization_83[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_84 (BatchNo (None, 5, 5, 192) 576 conv2d_84[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_76 (Activation) (None, 5, 5, 320) 0 batch_normalization_76[0][0] \n",
"__________________________________________________________________________________________________\n",
"mixed9_0 (Concatenate) (None, 5, 5, 768) 0 activation_78[0][0] \n",
" activation_79[0][0] \n",
"__________________________________________________________________________________________________\n",
"concatenate (Concatenate) (None, 5, 5, 768) 0 activation_82[0][0] \n",
" activation_83[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_84 (Activation) (None, 5, 5, 192) 0 batch_normalization_84[0][0] \n",
"__________________________________________________________________________________________________\n",
"mixed9 (Concatenate) (None, 5, 5, 2048) 0 activation_76[0][0] \n",
" mixed9_0[0][0] \n",
" concatenate[0][0] \n",
" activation_84[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_89 (Conv2D) (None, 5, 5, 448) 917504 mixed9[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_89 (BatchNo (None, 5, 5, 448) 1344 conv2d_89[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_89 (Activation) (None, 5, 5, 448) 0 batch_normalization_89[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_86 (Conv2D) (None, 5, 5, 384) 786432 mixed9[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_90 (Conv2D) (None, 5, 5, 384) 1548288 activation_89[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_86 (BatchNo (None, 5, 5, 384) 1152 conv2d_86[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_90 (BatchNo (None, 5, 5, 384) 1152 conv2d_90[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_86 (Activation) (None, 5, 5, 384) 0 batch_normalization_86[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_90 (Activation) (None, 5, 5, 384) 0 batch_normalization_90[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_87 (Conv2D) (None, 5, 5, 384) 442368 activation_86[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_88 (Conv2D) (None, 5, 5, 384) 442368 activation_86[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_91 (Conv2D) (None, 5, 5, 384) 442368 activation_90[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_92 (Conv2D) (None, 5, 5, 384) 442368 activation_90[0][0] \n",
"__________________________________________________________________________________________________\n",
"average_pooling2d_8 (AveragePoo (None, 5, 5, 2048) 0 mixed9[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_85 (Conv2D) (None, 5, 5, 320) 655360 mixed9[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_87 (BatchNo (None, 5, 5, 384) 1152 conv2d_87[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_88 (BatchNo (None, 5, 5, 384) 1152 conv2d_88[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_91 (BatchNo (None, 5, 5, 384) 1152 conv2d_91[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_92 (BatchNo (None, 5, 5, 384) 1152 conv2d_92[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_93 (Conv2D) (None, 5, 5, 192) 393216 average_pooling2d_8[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_85 (BatchNo (None, 5, 5, 320) 960 conv2d_85[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_87 (Activation) (None, 5, 5, 384) 0 batch_normalization_87[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_88 (Activation) (None, 5, 5, 384) 0 batch_normalization_88[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_91 (Activation) (None, 5, 5, 384) 0 batch_normalization_91[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_92 (Activation) (None, 5, 5, 384) 0 batch_normalization_92[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_93 (BatchNo (None, 5, 5, 192) 576 conv2d_93[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_85 (Activation) (None, 5, 5, 320) 0 batch_normalization_85[0][0] \n",
"__________________________________________________________________________________________________\n",
"mixed9_1 (Concatenate) (None, 5, 5, 768) 0 activation_87[0][0] \n",
" activation_88[0][0] \n",
"__________________________________________________________________________________________________\n",
"concatenate_1 (Concatenate) (None, 5, 5, 768) 0 activation_91[0][0] \n",
" activation_92[0][0] \n",
"__________________________________________________________________________________________________\n",
"activation_93 (Activation) (None, 5, 5, 192) 0 batch_normalization_93[0][0] \n",
"__________________________________________________________________________________________________\n",
"mixed10 (Concatenate) (None, 5, 5, 2048) 0 activation_85[0][0] \n",
" mixed9_1[0][0] \n",
" concatenate_1[0][0] \n",
" activation_93[0][0] \n",
"__________________________________________________________________________________________________\n",
"flatten_2 (Flatten) (None, 51200) 0 mixed10[0][0] \n",
"__________________________________________________________________________________________________\n",
"dense_2 (Dense) (None, 2) 102402 flatten_2[0][0] \n",
"==================================================================================================\n",
"Total params: 21,905,186\n",
"Trainable params: 102,402\n",
"Non-trainable params: 21,802,784\n",
"__________________________________________________________________________________________________\n"
],
"name": "stdout"
}
]
},
{
"cell_type": "code",
"metadata": {
"id": "E9ymGmQuGOA8"
},
"source": [
"# Use the Image Data Generator to import the images from the dataset\r\n",
"from keras.preprocessing.image import ImageDataGenerator\r\n",
"\r\n",
"train_datagen = ImageDataGenerator(rescale = 1./255,\r\n",
" shear_range = 0.2,\r\n",
" zoom_range = 0.2,\r\n",
" horizontal_flip = True)\r\n",
"\r\n",
"test_datagen = ImageDataGenerator(rescale = 1./255)"
],
"execution_count": 51,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/",
"height": 0
},
"id": "DhAIOMghGbTG",
"outputId": "ee3bee0e-3680-4a65-cbbe-4d68150de739"
},
"source": [
"# Make sure you provide the same target size as initialied for the image size\r\n",
"training_set = train_datagen.flow_from_directory('/content/chest_xray/train',\r\n",
" target_size = (224, 224),\r\n",
" batch_size = 32,\r\n",
" class_mode = 'categorical')"
],
"execution_count": 52,
"outputs": [
{
"output_type": "stream",
"text": [
"Found 5216 images belonging to 2 classes.\n"
],
"name": "stdout"
}
]
},
{
"cell_type": "code",
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/",
"height": 0
},
"id": "Zq5jU7stGgj-",
"outputId": "d83b8e75-adf0-483e-ec0d-075ee99cf9c3"
},
"source": [
"test_set = test_datagen.flow_from_directory('/content/chest_xray/test',\r\n",
" target_size = (224, 224),\r\n",
" batch_size = 32,\r\n",
" class_mode = 'categorical')"
],
"execution_count": 53,
"outputs": [
{
"output_type": "stream",
"text": [
"Found 624 images belonging to 2 classes.\n"
],
"name": "stdout"
}
]
},
{
"cell_type": "code",
"metadata": {
"id": "WGYuqy2lGl5E"
},
"source": [
"# tell the model what cost and optimization method to use\r\n",
"model.compile(\r\n",
" loss='categorical_crossentropy',\r\n",
" optimizer='adam',\r\n",
" metrics=['accuracy']\r\n",
")"
],
"execution_count": 54,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/",
"height": 0
},
"id": "xu4buVENGqt0",
"outputId": "9c48c951-379b-4ac8-ca7f-fc20b97a6b8d"
},
"source": [
"r = model.fit_generator(\r\n",
" training_set,\r\n",
" validation_data=test_set,\r\n",
" epochs=4,\r\n",
" steps_per_epoch=len(training_set),\r\n",
" validation_steps=len(test_set)\r\n",
")"
],
"execution_count": 55,
"outputs": [
{
"output_type": "stream",
"text": [
"/usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/engine/training.py:1844: UserWarning: `Model.fit_generator` is deprecated and will be removed in a future version. Please use `Model.fit`, which supports generators.\n",
" warnings.warn('`Model.fit_generator` is deprecated and '\n"
],
"name": "stderr"
},
{
"output_type": "stream",
"text": [
"Epoch 1/4\n",
"163/163 [==============================] - 106s 626ms/step - loss: 0.8582 - accuracy: 0.8780 - val_loss: 1.0335 - val_accuracy: 0.8670\n",
"Epoch 2/4\n",
"163/163 [==============================] - 100s 614ms/step - loss: 0.4948 - accuracy: 0.9343 - val_loss: 1.8882 - val_accuracy: 0.8413\n",
"Epoch 3/4\n",
"163/163 [==============================] - 100s 614ms/step - loss: 0.5985 - accuracy: 0.9379 - val_loss: 2.4259 - val_accuracy: 0.8173\n",
"Epoch 4/4\n",
"163/163 [==============================] - 100s 613ms/step - loss: 0.4942 - accuracy: 0.9456 - val_loss: 1.6890 - val_accuracy: 0.8590\n"
],
"name": "stdout"
}
]
},
{
"cell_type": "code",
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/",
"height": 545
},
"id": "keSE92hRG_Y4",
"outputId": "eb38243c-d2f3-4b7c-8218-7a028c1aa470"
},
"source": [
"# plot the loss\r\n",
"plt.plot(r.history['loss'], label='train loss')\r\n",
"plt.plot(r.history['val_loss'], label='val loss')\r\n",
"plt.legend()\r\n",
"plt.show()\r\n",
"plt.savefig('LossVal_loss')\r\n",
"\r\n",
"# plot the accuracy\r\n",
"acc_train = r.history['accuracy']\r\n",
"acc_val = r.history['val_accuracy']\r\n",
"epochs = range(1,5)\r\n",
"plt.plot(epochs, acc_train, 'g', label='Training accuracy')\r\n",
"plt.plot(epochs, acc_val, 'b', label='validation accuracy')\r\n",
"plt.title('Training and Validation accuracy')\r\n",
"plt.xlabel('Epochs')\r\n",
"plt.ylabel('Accuracy')\r\n",
"plt.legend()\r\n",
"plt.show()"
],
"execution_count": 56,
"outputs": [
{
"output_type": "display_data",
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 432x288 with 1 Axes>"
]
},
"metadata": {
"tags": [],
"needs_background": "light"
}
},
{
"output_type": "display_data",
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 432x288 with 1 Axes>"
]
},
"metadata": {
"tags": [],
"needs_background": "light"
}
}
]
},
{
"cell_type": "code",
"metadata": {
"id": "DPf_ix7aHAIO"
},
"source": [
"# save it as a h5 file\r\n",
"\r\n",
"import tensorflow as tf\r\n",
"\r\n",
"from keras.models import load_model\r\n",
"\r\n",
"model.save('model_inceptionv3.h5')"
],
"execution_count": 57,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"id": "iO03lcaRHDx1"
},
"source": [
"from keras.models import load_model\r\n",
"from keras.preprocessing import image\r\n",
"from keras.applications.inception_v3 import preprocess_input\r\n",
"import numpy as np"
],
"execution_count": 58,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"id": "NO2PVJ5aHH6v"
},
"source": [
"model=load_model('model_inceptionv3.h5')"
],
"execution_count": 59,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"id": "dyphMdnNHMEw"
},
"source": [
"img=image.load_img('/content/chest_xray/val/PNEUMONIA/person1947_bacteria_4876.jpeg',target_size=(224,224))"
],
"execution_count": 60,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/",
"height": 0
},
"id": "6ImqGTlrHRvV",
"outputId": "fe1eb630-9282-42ae-dfe2-fbfc45f1fcb8"
},
"source": [
"x=image.img_to_array(img)\r\n",
"x=np.expand_dims(x,axis=0)\r\n",
"img_data=preprocess_input(x)\r\n",
"classes=model.predict(img_data)\r\n",
"classes"
],
"execution_count": 61,
"outputs": [
{
"output_type": "execute_result",
"data": {
"text/plain": [
"array([[3.704202e-05, 9.999629e-01]], dtype=float32)"
]
},
"metadata": {
"tags": []
},
"execution_count": 61
}
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "75tBT-cUMspm"
},
"source": [
"Here for example the InceptionV3 model makes an error of prediction: the image is predicted to be normal whereas it is associated to pneumonia class."
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "hpnzax1RNEzl"
},
"source": [
"# Comparison between models"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "BWQViC6O0VSZ"
},
"source": [
""
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "scB6aGQGNLbs"
},
"source": [
" \r\n",
" \r\n",
"\r\n",
"* One one hand VGG19 and VGG16 outperform InceptionV3 in terms of accuracy. On the other hand the three algorithms perform closely in terms of computational speed.\r\n",
"* We compared our models only in terms of accuracy and computational speed. But this is not sufficient: we have to consider other important metrics such as f1_score, recall, precision.\r\n",
"* There are many other interesting models to test such as MobileNet, EfficientNet, ResNet...\r\n",
"\r\n"
]
}
]
}