Construct a queued batch of images and labels. Args: image: 3-D Tensor of [height, width, 3] of type.float32. label: 1-D Tensor of type.int3

Add operations to minimize `loss` by updating `var_list` with SGD and compute the batch size for the next step according to the CABS rule.

Reads and parses examples from CIFAR10 data files. Recommendation: if you want N-way read parallelism, call this function N times. This will give

Computes gradient moment for a bias variable through an Add operation. Assumes ``Z = tf.add(Zz, b)``, where ``b`` is a bias parameter and ``Zz``

Computes gradient moment for the filter of a Conv2D operation. Assumes ``Z=tf.nn.conv2d(A, f)``, where ``f`` is a ``[h_f, w_f, c_in, c_out]`` c

Wrapper function for the operation type-specific GradMom functions below. Inputs: :op: A tensorflow operation of type in VALID_TYPES.

Computes gradient moment for a weight matrix through a MatMul operation. Assumes ``Z=tf.matmul(A, W)``, where ``W`` is a d1xd2 weight matrix, ``A

Sort a list of ops according to type into valid types for which we can compute the gradient moment) and regularizers. Raise an exception when enco

Construct a new gradient descent optimizer with coupled adaptive batch size (CABS). Args: :learning_rate: A Tensor or a floating po

Construct distorted input for CIFAR training using the Reader ops. Args: data_dir: Path to the CIFAR-10 data directory. batch_size: Number o

Compute the gradients and gradient moments of ``loss`` w.r.t. to the variables in ``var_list`` Inputs: :loss: The tensor containing the s

Construct input for CIFAR evaluation using the Reader ops. Args: eval_data: bool, indicating if one should use the train or eval data set. d