[BFCL] Fix Hanging Inference for OSS Models on GPU Platforms #663
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
HuanzhiMao
changed the title
VishnuSuresh27
pushed a commit
to VishnuSuresh27/gorilla
that referenced
this pull request
Nov 11, 2024
…Patil#663) This PR addresses issues encountered when running locally-hosted models on GPU-renting platforms (e.g., Lambda Cloud). Specifically, there were problems with output display from `vllm` due to the use of subprocesses for launching these models. Additionally, some multi-turn functions (such as `xargs`) rely on subprocesses, which caused inference on certain test entries (such as `multi_turn_36 `) to hang indefinitely, resulting in an undesirable pipeline halt. To fix this, the terminal logging logic has been updated to utilize a separate thread for reading from the subprocess pipe and printing to the terminal. Alos, for readability, the `_format_prompt` function has been moved to the `Prompting methods` section; this would not change the leaderboard score.
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Outdated suggestions cannot be applied.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.
This PR addresses issues encountered when running locally-hosted models on GPU-renting platforms (e.g., Lambda Cloud). Specifically, there were problems with output display from
vllmdue to the use of subprocesses for launching these models. Additionally, some multi-turn functions (such asxargs) rely on subprocesses, which caused inference on certain test entries (such asmulti_turn_36) to hang indefinitely, resulting in an undesirable pipeline halt.To fix this, the terminal logging logic has been updated to utilize a separate thread for reading from the subprocess pipe and printing to the terminal.
Alos, for readability, the
_format_promptfunction has been moved to thePrompting methodssection; this would not change the leaderboard score.