MCPcopy Create free account
hub / github.com/MethodsOfMachineLearning/cabs / _MatMulGradMom

Function _MatMulGradMom

gradient_moment.py:134–156  ·  view source on GitHub ↗

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`` are the nxd1 activations of the previous layer (n being the batch size). ``out_grad`` is the gradient w.r.t. ``Z``, as computed by ``tf.gradien

(op, W, out_grad, batch_size, mom=2)

Source from the content-addressed store, hash-verified

source not stored for this graph (policy: none)

Callers 1

_GradMomFunction · 0.85

Calls

no outgoing calls

Tested by

no test coverage detected