Python Deep Learning Projects

Fitting ImageDataGenerator

Now, let's build a classifier using the same architecture as the deep convolution model with pooling and Dropout, but on augmented data.

First, define the features of the ImageDataGenerator, as follows:

train_datagen = ImageDataGenerator(        rescale = 1./255,        zoom_range = 0.2,        horizontal_flip = True)  

We have defined that the ImageDataGenerator can perform the following operations

• Rescaling
• Random zoom
• Random horizontal flip

The rescaling operation scales the pixel values to a range between 0 and 1.

The next step is to fit this generator on the train data:

train_datagen.fit(X_train)