Commit 61b52167 authored by A. Unique TensorFlower's avatar A. Unique TensorFlower Committed by TensorFlower Gardener
Browse files

Collapse backprop implementation of LSTMBlockCell into one CUDA kernel. 

Add benchmarks for backprop. Speed difference is minor - will need to move
everything out of the graph for large speedups, I think.

Also template the fprop kernel on use_peephole.

Original change by @duckworthd

PiperOrigin-RevId: 172131001
parent 45b11515
Show whitespace changes
Inline Side-by-side
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Please to comment