[python] Refactor output formatter for rolling batch #916
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
xyang16
force-pushed
the
lmi
branch
2 times, most recently
from
f9abb7b to
c43dab1
Compare
lanking520
reviewed
Jul 7, 2023
| @@ -173,8 +173,6 @@ def initialize(self, properties: dict): | |||
| self.device = int(os.getenv("LOCAL_RANK", 0)) | |||
| _rolling_batch_cls = get_rolling_batch_class_from_str( | |||
| self.rolling_batch_type, is_mpi, self.model_config) | |||
|
|
|||
| # TODO: Allow user to set output formatter | |||
| self.rolling_batch = _rolling_batch_cls(model_id_or_path, | |||
There was a problem hiding this comment.
you can make output formatter as part of the kwargs, user can pass in a function if needed
lanking520
reviewed
Jul 7, 2023
frankfliu
reviewed
Jul 7, 2023
| @@ -143,7 +143,13 @@ def _prefill_and_decode(self, new_requests): | |||
| for request_id, generated_token, request in zip( | |||
| request_ids, generated_tokens, self.pending_requests): | |||
| is_last_token = (request_id in exit_req_ids) | |||
| request.set_next_token(generated_token, last_token=is_last_token) | |||
| if self.output_formatter is not None: | |||
There was a problem hiding this comment.
request.set_next_token(generated_token, self.output_formatter, last_token=is_last_token)
lanking520
reviewed
Jul 7, 2023
| """ | ||
|
|
||
| self.device = device | ||
| self.pending_requests = [] | ||
| self.req_id_counter = 0 | ||
| if 'rolling_batch_output_formatter' in kwargs: | ||
| # TODO: Allow user to set custom output formatter | ||
| pass |
There was a problem hiding this comment.
self.output_formatter = kwargs['rolling_batch_output_formatter']
lanking520
reviewed
Jul 7, 2023
| # TODO: find a way to return content-type for custom output formatter | ||
| if self.output_formatter == _default_output_formatter: | ||
| return "application/jsonlines" | ||
| return None |
There was a problem hiding this comment.
Maybe some other types
lanking520
approved these changes
Jul 7, 2023
KexinFeng
pushed a commit
to KexinFeng/djl-serving-forked
that referenced
this pull request
Aug 16, 2023
…#916) * [python] Refactor output formatter for rolling batch * Review changes * Review changes * Review changes * Review changes
KexinFeng
pushed a commit
to KexinFeng/djl-serving-forked
that referenced
this pull request
Aug 16, 2023
…#916) * [python] Refactor output formatter for rolling batch * Review changes * Review changes * Review changes * Review changes
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.
Description
Brief description of what this PR is about