Attend the Women in ML Symposium on December 7 Register now

tfl.kronecker_factored_lattice_lib.custom_reduce_prod

Stay organized with collections Save and categorize content based on your preferences.

tf.reduce_prod(t, axis) with faster custom gradient.

Shows comparable speed on CPU, up to 2x speed up on GPU, and 7x on TPU.

t The tensor to reduce.
axis The dimension to reduce.

prod(t) and grad(prod(t))