TensorFlow: how is dataset.train.next_batch defined?

Question

I am trying to learn TensorFlow and studying the example at: https://github.com/aymericdamien/TensorFlow-Examples/blob/master/notebooks/3_NeuralNetworks/autoencoder.ipynb

<

Answer 1

The mnist object is returned from the read_data_sets() function defined in the tf.contrib.learn module. The mnist.train.next_batch(batch_size) method is implemented here, and it returns a tuple of two arrays, where the first represents a batch of batch_size MNIST images, and the second represents a batch of batch-size labels corresponding to those images.

The images are returned as a 2-D NumPy array of size [batch_size, 784] (since there are 784 pixels in an MNIST image), and the labels are returned as either a 1-D NumPy array of size [batch_size] (if read_data_sets() was called with one_hot=False) or a 2-D NumPy array of size [batch_size, 10] (if read_data_sets() was called with one_hot=True).