fix serve_multiplexed_model_id and mm data process#120
Merged
Conversation
Loading
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
PR type
PR information
This pull request introduces a new client script for multimodal training with Tinker, updates existing client examples, and significantly refactors the Twinkle server's resource lifecycle management to be asynchronous. This refactoring involves converting numerous synchronous methods for session, model, adapter, and processor management to async/await patterns, consolidating common logic into a new SessionResourceMixin, and updating all corresponding calls throughout the server and client components. A patch for Ray Serve is also included to correctly handle HTTP header normalization for multiplexed model IDs. A review comment suggests refactoring duplicated logic for packing multimodal tensors into loss_fn_inputs within the input_feature_to_datum function for improved maintainability.
Experiment results
Paste your experiment result here(if needed).