MyDataGenerator.py

import keras
from scipy.misc import imresize
import numpy as np
from PIL import Image
resize_factor = 0.6

class DataGenerator(keras.utils.Sequence):
    'Generates data for Keras'
    def __init__(self, path, list_IDs, labels, batch_size=32, dim=(32,32,32), n_channels=3,
                  shuffle=True, fully_convolutional = True):
        'Initialization'
        self.dim = dim
        self.batch_size = batch_size
        self.labels = labels
        self.list_IDs = list_IDs
        self.n_channels = n_channels
        self.shuffle = shuffle
        self.on_epoch_end()
        self.path = path

    def __len__(self):
        'Denotes the number of batches per epoch'
        return int(np.floor(len(self.list_IDs) / self.batch_size))

    def __getitem__(self, index):
        'Generate one batch of data'
        # Generate indexes of the batch
        indexes = self.indexes[index*self.batch_size:(index+1)*self.batch_size]

        # Find list of IDs
        list_IDs_temp = [self.list_IDs[k] for k in indexes]

        # Generate data
        X, y = self.__data_generation(list_IDs_temp)

        return X, y

    def on_epoch_end(self):
        'Updates indexes after each epoch'
        self.indexes = np.arange(len(self.list_IDs))
        if self.shuffle == True:
            np.random.shuffle(self.indexes)

    def __data_generation(self, list_IDs_temp):
        'Generates data containing batch_size samples' # X : (n_samples, *dim, n_channels)
        # Initialization
        X = np.empty((self.batch_size, *self.dim, self.n_channels))
        y = np.empty((self.batch_size, 2), dtype=float)

        # Generate data
        for i, ID in enumerate(list_IDs_temp):
            # Store sample
            X[i,] = imresize(np.array(Image.open(self.path + ID + '.png')), resize_factor)/255.0
            

            # Store class
            y[i,0], y[i,1] = self.labels[ID]
            

        return X, y