Placeholders for inputs and labels

The values of different layers are the objects, and are also called placeholders for inputs and labels. These placeholders are used for feeding the data into the network. The following lines of code are used for showing placeholders for the inputs and labels:

X = tf.placeholder(tf.float32, shape=[None, n_inputs])
y = tf.placeholder(tf.int64)

So we have a placeholder X for the features, which is the input layer, and we have a placeholder y for the target value. So this object will contain the actual true labels of the digits.