Skip to content

Instantly share code, notes, and snippets.

@anupamchugh

anupamchugh/CatDogCategoricalClassifcationKeras.ipynb

Last active September 30, 2019 17:52
Show Gist options
  • Star 0 You must be signed in to star a gist
  • Fork 0 You must be signed in to fork a gist
  • Save anupamchugh/9bdb12ee2daf28575c0667622ab1b420 to your computer and use it in GitHub Desktop.
Save anupamchugh/9bdb12ee2daf28575c0667622ab1b420 to your computer and use it in GitHub Desktop.
Download ZIP
Display the source blob
Display the rendered blob
Raw
CatDogCategoricalClassifcationKeras.ipynb
{
"nbformat": 4,
"nbformat_minor": 0,
"metadata": {
"colab": {
"name": "CategoricalClassification.ipynb",
"provenance": [],
"collapsed_sections": []
},
"kernelspec": {
"name": "python3",
"display_name": "Python 3"
}
},
"cells": [
{
"cell_type": "code",
"metadata": {
"id": "7Ihybkh9hivU",
"colab_type": "code",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 34
},
"outputId": "09e744aa-5e39-4fb2-9d34-301270480936"
},
"source": [
"from zipfile import ZipFile\n",
"file_name = \"predict.zip\"\n",
"\n",
"with ZipFile(file_name, 'r') as zip:\n",
" zip.extractall()\n",
" print('Done')"
],
"execution_count": 5,
"outputs": [
{
"output_type": "stream",
"text": [
"Done\n"
],
"name": "stdout"
}
]
},
{
"cell_type": "code",
"metadata": {
"id": "4CSzA0OmlExF",
"colab_type": "code",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 34
},
"outputId": "3a83f46d-3ebf-4e0b-ffa2-4b3cf0b15c9b"
},
"source": [
"from zipfile import ZipFile\n",
"file_name = \"data.zip\"\n",
"\n",
"with ZipFile(file_name, 'r') as zip:\n",
" zip.extractall()\n",
" print('Done')"
],
"execution_count": 6,
"outputs": [
{
"output_type": "stream",
"text": [
"Done\n"
],
"name": "stdout"
}
]
},
{
"cell_type": "code",
"metadata": {
"id": "oruMSnOci7PX",
"colab_type": "code",
"colab": {}
},
"source": [
"import pandas as pd\n",
"from keras.preprocessing.image import ImageDataGenerator\n",
"from keras.models import Model\n",
"from keras.layers import Input, Convolution2D, MaxPooling2D, Dropout, Flatten, Dense\n",
"from keras.callbacks import Callback, ModelCheckpoint\n",
"\n",
"IMG_WIDTH, IMG_HEIGHT = 150, 150\n",
"TRAIN_DATA_DIR = 'data/train'\n",
"VALIDATION_DATA_DIR = 'data/validation'\n",
"NB_TRAIN_SAMPLES = 2000\n",
"NB_VALIDATION_SAMPLES = 800\n",
"NB_EPOCH = 5\n",
"BATCH_SIZE = 32"
],
"execution_count": 0,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"id": "7pVtjjWPjHg-",
"colab_type": "code",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 377
},
"outputId": "681d9cc6-d45a-4918-c2c4-235647951412"
},
"source": [
"input = Input(shape=(IMG_WIDTH, IMG_HEIGHT, 3,))\n",
"x = Convolution2D(32, 3, 3, activation='relu')(input)\n",
"x = MaxPooling2D(pool_size=(2, 2))(x)\n",
"x = Convolution2D(32, 3, 3, activation='relu')(x)\n",
"x = MaxPooling2D(pool_size=(2, 2))(x)\n",
"x = Convolution2D(64, 3, 3, activation='relu')(x)\n",
"x = MaxPooling2D(pool_size=(2, 2))(x)\n",
"x = Flatten()(x)\n",
"x = Dense(64, activation='relu')(x)\n",
"x = Dropout(0.5)(x)\n",
"x = Dense(2, activation='softmax')(x)\n",
"model = Model(input=input, output=x)\n",
"\n",
"model.compile(loss='categorical_crossentropy',\n",
" optimizer='rmsprop',\n",
" metrics=['accuracy'])\n",
"\n",
"\n",
"train_datagen = ImageDataGenerator(\n",
" rescale=1./255,\n",
" shear_range=0.2,\n",
" zoom_range=0.2,\n",
" horizontal_flip=True)\n",
"\n",
"test_datagen = ImageDataGenerator(rescale=1./255)\n",
"\n",
"train_generator = train_datagen.flow_from_directory(\n",
" TRAIN_DATA_DIR,\n",
" target_size=(IMG_WIDTH, IMG_HEIGHT),\n",
" batch_size=BATCH_SIZE,\n",
" class_mode='categorical')\n",
"\n",
"validation_generator = test_datagen.flow_from_directory(\n",
" VALIDATION_DATA_DIR,\n",
" target_size=(IMG_WIDTH, IMG_HEIGHT),\n",
" batch_size=BATCH_SIZE,\n",
" class_mode='categorical')\n",
"\n",
"model.fit_generator(\n",
" train_generator,\n",
" steps_per_epoch=NB_TRAIN_SAMPLES // batch_size,\n",
" epochs=NB_EPOCH,\n",
" validation_data=validation_generator,\n",
" validation_steps=NB_VALIDATION_SAMPLES // batch_size)\n",
"\n",
"model.save('model.h5')\n"
],
"execution_count": 14,
"outputs": [
{
"output_type": "stream",
"text": [
"/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:2: UserWarning: Update your `Conv2D` call to the Keras 2 API: `Conv2D(32, (3, 3), activation=\"relu\")`\n",
" \n",
"/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:4: UserWarning: Update your `Conv2D` call to the Keras 2 API: `Conv2D(32, (3, 3), activation=\"relu\")`\n",
" after removing the cwd from sys.path.\n",
"/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:6: UserWarning: Update your `Conv2D` call to the Keras 2 API: `Conv2D(64, (3, 3), activation=\"relu\")`\n",
" \n",
"/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:12: UserWarning: Update your `Model` call to the Keras 2 API: `Model(inputs=Tensor(\"in..., outputs=Tensor(\"de...)`\n",
" if sys.path[0] == '':\n"
],
"name": "stderr"
},
{
"output_type": "stream",
"text": [
"Found 2001 images belonging to 2 classes.\n",
"Found 1008 images belonging to 2 classes.\n",
"Epoch 1/5\n",
"125/125 [==============================] - 152s 1s/step - loss: 0.6994 - acc: 0.5678 - val_loss: 0.6287 - val_acc: 0.6717\n",
"Epoch 2/5\n",
"125/125 [==============================] - 150s 1s/step - loss: 0.6277 - acc: 0.6614 - val_loss: 0.6046 - val_acc: 0.6888\n",
"Epoch 3/5\n",
"125/125 [==============================] - 151s 1s/step - loss: 0.5723 - acc: 0.7099 - val_loss: 0.6208 - val_acc: 0.6679\n",
"Epoch 4/5\n",
"125/125 [==============================] - 150s 1s/step - loss: 0.5546 - acc: 0.7254 - val_loss: 0.6821 - val_acc: 0.6467\n",
"Epoch 5/5\n",
"125/125 [==============================] - 151s 1s/step - loss: 0.5224 - acc: 0.7562 - val_loss: 0.5590 - val_acc: 0.7165\n"
],
"name": "stdout"
}
]
},
{
"cell_type": "code",
"metadata": {
"id": "djJED46Gmn3G",
"colab_type": "code",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 34
},
"outputId": "dd1ca4ac-d514-427c-9c08-cce3a9226234"
},
"source": [
"score = model.evaluate(validation_generator, verbose=0)\n",
"print(\"accuracy: \", score[1])"
],
"execution_count": 15,
"outputs": [
{
"output_type": "stream",
"text": [
"accuracy: 0.7212301587301587\n"
],
"name": "stdout"
}
]
},
{
"cell_type": "code",
"metadata": {
"id": "hq0Ju_CdmB5E",
"colab_type": "code",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 238
},
"outputId": "62687f9b-576a-4316-b5d7-c1c321955069"
},
"source": [
"from keras.models import load_model\n",
"from keras.preprocessing import image\n",
"import numpy as np\n",
"from os import listdir\n",
"from os.path import isfile, join\n",
"\n",
"# dimensions of our images\n",
"img_width, img_height = 150, 150\n",
"\n",
"# load the model we saved\n",
"model = load_model('model.h5')\n",
"model.compile(loss='categorical_crossentropy',\n",
" optimizer='rmsprop',\n",
" metrics=['accuracy'])\n",
"\n",
"mypath = \"predict/\"\n",
"\n",
"onlyfiles = [f for f in listdir(mypath) if isfile(join(mypath, f))]\n",
"print(onlyfiles)\n",
"# predicting images\n",
"for file in onlyfiles:\n",
" img = image.load_img(mypath+file, target_size=(img_width, img_height))\n",
" x = image.img_to_array(img)\n",
" x = np.expand_dims(x, axis=0)\n",
" \n",
" result = model.predict(x)\n",
" print(\"predict result..\")\n",
" print(result)"
],
"execution_count": 16,
"outputs": [
{
"output_type": "stream",
"text": [
"['1.jpg', '2.jpg', '6.jpg', '5.jpg', '4.jpg', '3.jpg']\n",
"predict result..\n",
"[[1. 0.]]\n",
"predict result..\n",
"[[1. 0.]]\n",
"predict result..\n",
"[[1. 0.]]\n",
"predict result..\n",
"[[1. 0.]]\n",
"predict result..\n",
"[[1. 0.]]\n",
"predict result..\n",
"[[0. 1.]]\n"
],
"name": "stdout"
}
]
}
]
}
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment