-
Notifications
You must be signed in to change notification settings - Fork 759
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Predict stop sequence matches during streaming #541
Open
Y4hL
wants to merge
5
commits into
ml-explore:main
Choose a base branch
from
Y4hL:sequence-predict
base: main
Could not load branches
Branch not found: {{ refName }}
Could not load tags
Nothing to show
Are you sure you want to change the base?
Some commits from the old base branch may be removed from the timeline,
and old review comments may become outdated.
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
Oh much better, thank you 😄 |
awni
reviewed
Mar 14, 2024
What about adding your test (modified for unittest) as a test case to a new test file |
awni
reviewed
Mar 14, 2024
Yeah, I'll have a look into writing a unittest |
Check for overlap of stop sequences and the tokens array for potential sequence matches after more tokens get generated. Generate tokens until we can confirm that the stop sequence is not met.
Added a test for the sequence_overlap method
Hi! Is this ready to be reviewed again? |
@awni yes |
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Outdated suggestions cannot be applied.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.
When streaming using mlx_lm/server.py we should predict potential stop sequence matches, and generate tokens until we know that there is no match. This prevents the server from sending parts of a stop sequence to the client before it finds the match.
Fixes #524
My implementation adds a new function called "sequence_overlap" which checks how much sequence 1 has overlap with sequence 2. It checks for larger overlaps first, and returns the overlap as an integer.
The server checks for overlaps, and generates more tokens before allowing the server to send them.
The sequence_overlap implementation can be tested with this example: