To implement my Binary CNN in keras, I had a dataset of ~~35000 images but only 700 is from one class and all the others are from the other class, so what I did: I get the 700 unique images from class 1 and get 700 images randomly from class 0 totalizing 1400 samples (700 each class). After I perform this augmentation in my set:
datagen = ImageDataGenerator( rotation_range=30, zoom_range=0.15, featurewise_std_normalization=True, width_shift_range=0.2, height_shift_range=0.2, shear_range=0.15, horizontal_flip=True, fill_mode="nearest")
I did it right? or there's a better way to do?