Add test to compare encoder inference on input with and without padding #1770
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
erichan1
commented
Jun 7, 2022
| ) | ||
| model = RobertaModel(encoder_conf) | ||
| model = model.eval() | ||
| # TODO: make return_all_layers a property of RobertaEncoderConf so it can be passed as arg |
There was a problem hiding this comment.
Currently, return_all_layers cannot be set from RobertaEncoderConf because it's not a property of the config. Thus we need to manually set it below. Not a big deal right now, but something to consider changing.
parmeet
approved these changes
Jun 7, 2022
Nayef211
pushed a commit
to Nayef211/text
that referenced
this pull request
Jun 7, 2022
…ng (pytorch#1770) * add test to compare encoder inference on input with and without padding * remove bottom whitespace * remove unnecessary import * lint
Nayef211
added a commit
that referenced
this pull request
Jun 8, 2022
…ng (#1770) (#1772) * add test to compare encoder inference on input with and without padding * remove bottom whitespace * remove unnecessary import * lint Co-authored-by: erichan1 <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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.
Encoder inference should output identical embeddings if given a padded vs. non-padded input (assuming the non-padded data is the same). Check that this is the case.
We make this check for four paths by setting the below two flags.