[fix] Sampling Parameters related improvements #80
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
|
LGTM! |
kthui
previously approved these changes
Jan 9, 2025
| @@ -197,6 +317,22 @@ else | |||
| RET=1 | |||
| fi | |||
| fi | |||
|
|
|||
| # Test generate endpoint + LoRA enabled (boolean flag) | |||
There was a problem hiding this comment.
# Test generate endpoint + LoRA enabled (boolean flag)
LoRA enabled -> disabled ?
| @@ -243,6 +379,22 @@ else | |||
| RET=1 | |||
| fi | |||
| fi | |||
|
|
|||
| # Test generate endpoint + LoRA enabled (str flag) | |||
There was a problem hiding this comment.
# Test generate endpoint + LoRA enabled (str flag)
LoRA enabled -> disabled ?
oandreeva-nv
commented
Jan 9, 2025
oandreeva-nv
commented
Jan 9, 2025
kthui
approved these changes
Jan 9, 2025
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Outdated suggestions cannot be applied.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.
What does the PR do?
This PR removes dependencies on
vllm.SamplingParamsnames, present in_get_sampling_params_dictin favor of dynamically checking sampling parameters passed with the request withvllm.SamplingParams.__annotations__Previously, we needed to keep track about SamplingParams members and their types: e.g. float keys as defined here.
I suggest to infer the type from
vllm.SamplingParams.__annotations__and for simple ones (int, float, bool, str, Optional[int]) perform a conversion from string to the expected type. Suggested logic is implemented here.Also added proper handling of "guided_generation" parameter for constrained decoding test. Limited test added in
accuracy_testsMisc:
Fixed some inconsistencies with error response for LoRA-based inference via generate endpoint with parameters passed through "parameters" vs through client script with "sampling_parameters" input. Enhanced tests.
Checklist
<commit_type>: <Title>Commit Type:
Check the conventional commit type
box here and add the label to the github PR.
Related PRs:
Where should the reviewer start?
vllm_backend_utils.py -> main.py
Test plan:
22089744
Caveats:
Background
Related Issues: (use one of the action keywords Closes / Fixes / Resolves / Relates to)