Upside_down_detector

-- coding: utf-8 --

''' Original file is located at https://colab.research.google.com/drive/1HrNm5UMZr2Zjmze_HKW799p6LAHM8BTa '''

from google.colab import files files.upload()

!pip install kaggle

!cp kaggle.json ~/.kaggle/

!chmod 600 ~/.kaggle/kaggle.json

!kaggle datasets download 'shaunthesheep/microsoft-catsvsdogs-dataset'

!unzip microsoft-catsvsdogs-dataset

import tensorflow as tf from tensorflow.keras.preprocessing.image import ImageDataGenerator

image_dir='/content/PetImages/Cat'

!mkdir train_folder

!mkdir test_folder

import os path='/content/train_folder/' dir='upside_down' dir2='normal' training_normal= os.path.join(path, dir2) training_upside= os.path.join(path, dir) os.mkdir(training_normal) os.mkdir(training_upside)

#creating classes directories path='/content/test_folder/' dir='upside_down' dir2='normal' training_normal= os.path.join(path, dir2) training_upside= os.path.join(path, dir) os.mkdir(training_normal) os.mkdir(training_upside)

#copying only the cat images to my train folder fnames = ['{}.jpg'.format(i) for i in range(2000)] for fname in fnames: src = os.path.join('/content/PetImages/Cat', fname) dst = os.path.join('/content/train_folder/normal', fname) shutil.copyfile(src, dst)

import os import shutil fnames = ['{}.jpg'.format(i) for i in range(2000, 4000)] for fname in fnames: src = os.path.join('/content/PetImages/Cat', fname) dst = os.path.join('/content/test_folder/normal', fname) shutil.copyfile(src, dst)

from scipy import ndimage, misc from PIL import Image import numpy as np import matplotlib.pyplot as plt import imageio import os import cv2

#inverting Training Images outPath = '/content/train_folder/upside_down' path ='/content/train_folder/normal'

# iterate through the names of contents of the folder

for image_path in os.listdir(path):

    # create the full input path and read the file
    input_path = os.path.join(path, image_path)
    image_to_rotate =plt.imread(input_path)

    # rotate the image
    rotated =  np.flipud(image_to_rotate)

    # create full output path, 'example.jpg' 
    # becomes 'rotate_example.jpg', save the file to disk
    fullpath = os.path.join(outPath, 'rotated_'+image_path)
    imageio.imwrite(fullpath, rotated)

#nverting images for Validation outPath = '/content/test_folder/upside_down' path ='/content/test_folder/normal'

# iterate through the names of contents of the folder

for image_path in os.listdir(path):

    # create the full input path and read the file
    input_path = os.path.join(path, image_path)
    image_to_rotate =plt.imread(input_path)

    # rotate the image
    rotated =  np.flipud(image_to_rotate)

    # create full output path, 'example.jpg' 
    # becomes 'rotate_example.jpg', save the file to disk
    fullpath = os.path.join(outPath, 'rotated_'+image_path)
    imageio.imwrite(fullpath, rotated)

ima='/content/train_folder/inverted/rotated_1001.jpg' image=plt.imread(ima) plt.imshow(image)

visualize the the figure

plt.show()

train_dir='/content/train_folder' train_gen=ImageDataGenerator(rescale=1./255) train_images= train_gen.flow_from_directory( train_dir, target_size=(250,250), batch_size=50, class_mode='binary' )

validation_dir='/content/test_folder' test_gen=ImageDataGenerator(rescale=1./255) test_images= test_gen.flow_from_directory( validation_dir, target_size=(250,250), batch_size=50, class_mode='binary' )

model=tf.keras.Sequential([ tf.keras.layers.Conv2D(16, (3,3), activation='relu', input_shape=(250,250,3)), tf.keras.layers.MaxPooling2D(2,2), tf.keras.layers.Conv2D(32, (3,3), activation='relu'), tf.keras.layers.MaxPooling2D(2,2), tf.keras.layers.Conv2D(64, (3,3), activation='relu'), tf.keras.layers.MaxPooling2D(2,2), tf.keras.layers.Conv2D(128, (3,3), activation='relu'), tf.keras.layers.MaxPooling2D(2,2), tf.keras.layers.Flatten(), tf.keras.layers.Dense(512, activation='relu'), tf.keras.layers.Dense(1, activation='sigmoid')

])

from tensorflow.keras.optimizers import RMSprop model.compile(optimizer=RMSprop(learning_rate=0.001), loss=tf.keras.losses.BinaryCrossentropy(), metrics=['acc'])

history=model.fit(train_images, validation_data=test_images, epochs=5, steps_per_epoch=40 )