Copy gates in LSTM, GRU ops before applying activations #193
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
When the batch size is > 1, slices of the
gates
scratch space to extract specific gates are non-contiguous, which makessigmoid_in_place
andtanh_in_place
much slower. See #192. The GRU op had a workaround by using the non-in-placetanh
instead. This applies the same solution to the sigmoid activation in the GRU op and all activations in the LSTM op.