tfl.kronecker_factored_lattice_lib.custom_reduce_prod

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

tfl.kronecker_factored_lattice_lib.custom_reduce_prod(
    t, axis
)

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

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

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

Except as otherwise noted, the content of this page is licensed under the Creative Commons Attribution 4.0 License, and code samples are licensed under the Apache 2.0 License. For details, see the Google Developers Site Policies. Java is a registered trademark of Oracle and/or its affiliates.

Last updated 2024-08-02 UTC.

Args