This repository has been archived by the owner on Nov 3, 2023. It is now read-only.
[BlenderBot2] clean up reply during interactive mode #4379
Merged
Conversation
This file contains 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
| """ | ||
|
|
||
| if ( | ||
| self.opt('clean_reply', False) |
There was a problem hiding this comment.
So it returns here and doesn't go to cleaning the text if clean_reply is True?
There was a problem hiding this comment.
I'd love to ALWAYS clean the reply and I think it's the right thing to do?
There was a problem hiding this comment.
I'd love to ALWAYS clean the reply and I think it's the right thing to do?
added 2 commits
February 23, 2022 11:50
klshuster
approved these changes
Feb 24, 2022
|
could you maybe add some test cases to make sure that |
| if self_message is not None: | ||
| last_reply = self_message['text'] | ||
| clean_reply = self._clean_text(last_reply) | ||
| self.history.add_reply(clean_reply) |
There was a problem hiding this comment.
perhaps we should just override the History class and do this logic there?
added 2 commits
March 22, 2022 11:45
|
Go ahead and merge main into this branch |
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Outdated suggestions cannot be applied.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.
Patch description
Right now, during interactive mode, the
'_POTENTIALLY_UNSAFE__'from blenderbot2 generation would go into blenderbot 2 history and possibly affects its subsequent generation. However the model is never trained on observing such specially tokens, only generating them.After the change, we still keep the
'_POTENTIALLY_UNSAFE__'in the reply text, and only remove that from the blenderbot2 history.Testing steps
Other information