[Continuous batching] In the event of OOM, return tokens generated so far for the request #661
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
iefode
approved these changes
Jul 23, 2024
Wovchena
approved these changes
Jul 23, 2024
github-merge-queue
bot
removed this pull request from the merge queue due to failed status checks
mzegla
added a commit
to mzegla/openvino.genai
that referenced
this pull request
Jul 24, 2024
… far for the request (openvinotoolkit#661)
| for (auto& sequence: m_sequences) { | ||
| GenerationOutput output; | ||
| output.generated_token_ids = sequence->get_generated_ids(); | ||
| output.score = sequence->get_beam_search_score(m_sampling_params); |
There was a problem hiding this comment.
beam search score can be used only for beam search, while this method push_outputs is used for greedy and multinomial as well. For non-beam search we need to use cumulative logprob.
There was a problem hiding this comment.
You're right, I'm already fixing that in my next PR
| for (auto& sequence: finished_sequences) { | ||
| GenerationOutput output; | ||
| output.generated_token_ids = sequence->get_generated_ids(); | ||
| output.score = sequence->get_cumulative_log_probs(); |
There was a problem hiding this comment.
currently logprobs are not used, but should be
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Outdated suggestions cannot be applied.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.
No description provided.