This repository was archived by the owner on Jul 7, 2023. It is now read-only.
Avoid error in beam search when "f" is in cache #1302
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
added 4 commits
December 12, 2018 17:35
…duct in transformer model
… errors won't occur in beamsearch (nest.map function on cache)
lukaszkaiser
approved these changes
Jan 9, 2019
tensorflow-copybara
pushed a commit
that referenced
this pull request
Jan 9, 2019
PiperOrigin-RevId: 228581110
kpe
pushed a commit
to kpe/tensor2tensor
that referenced
this pull request
Mar 2, 2019
* add caching mechanism support for fast decoding with relative_dot_product in transformer model * fix typo * remove f in cache if use dense_relu_dense or conv_hidden_relu so that errors won't occur in beamsearch (nest.map function on cache)
kpe
pushed a commit
to kpe/tensor2tensor
that referenced
this pull request
Mar 2, 2019
PiperOrigin-RevId: 228581110
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
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Outdated suggestions cannot be applied.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.
If
ffn_layeris in["dense_relu_dense" or "conv_hidden_relu"], than the cache "f" won't be usedwhich means that the
shape of cache["f"]won't be changed to[beamsize*batch_size, decode_length, hparams.hidden_size]and may cause error when applyingnest.map reshape functionon it.This only happened in eager_execution mode because this part of graph may not be execute in non-eager_execution mode.