Implementation of depthwise separable 2D convolution operator as in MobileNet paper

Attach a lot of summaries to a Tensor (for TensorBoard visualization). :param var: variable to be summarized :return: None

Create a variable with L2 Regularization (Weight Decay) :param kernel_shape: the size of the convolving weight kernel. :param initializer

This block is responsible for a convolution 2D layer followed by optional (non-linearity, dropout, max-pooling). Note that: "is_training" sho

Add the summaries to tensorboard :param step: :param summaries_dict: :param summaries_merged: :return:

Generate batch from X_train/X_test and y_train/y_test using a python DataGenerator

Save Model Checkpoint :return:

Convolution 2D Wrapper :param name: (string) The name scope provided by the upper tf.name_scope('name') as scope. :param x: (tf.tensor) T

Fully connected layer :param name: (string) The name scope provided by the upper tf.name_scope('name') as scope. :param x: (tf.tensor) Th

Create a global epoch tensor to totally save the process of the training :return:

Create a global step variable to be a reference to the number of iterations :return:

Create the summary part of the graph :return:

Average pooling 2D Wrapper :param x: (tf.tensor) The input to the layer (N,H,W,C). :param size: (tuple) This specifies the si

Create Directories of a regular tensorflow experiment directory :param exp_dir: :return summary_dir, checkpoint_dir:

Implementation of depthwise 2D convolution wrapper

Dropout special layer

Flatten a (N,H,W,C) input into (N,D) output. Used for fully connected layers after conolution layers :param x: (tf.tensor) representing input

Max pooling 2D Wrapper :param x: (tf.tensor) The input to the layer (N,H,W,C). :param size: (tuple) This specifies the size of the filter

Parse the arguments of the program :return: (config_args) :rtype: tuple

This block is responsible for a fully connected followed by optional (non-linearity, dropout, max-pooling). Note that: "is_training" should b