{{ message }}
Repository navigation
Hide GPUs before spawning TensorFlow data workers - #8847
Open
jinchuntan wants to merge 1 commit into
Open
jinchuntan wants to merge 1 commit into
jinchuntan wants to merge 1 commit into
Conversation
Set worker GPU visibility before spawn imports user modules, then restore the parent environment. Cover eager TensorFlow initialization in imported collators, normal data conversion, and failed process startup. Fixes huggingface#6948 Assisted-by: Codex
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
to_tf_dataset(num_workers=...)can import a user's collator and initialize TensorFlow beforeworker_loophides its GPUs. Hide GPUs in the spawn environment so initialization already uses the CPU, then restore the parent environment. Fixes #6948.Six new cases cover imported collators with eager TensorFlow initialization, one/two workers, unset/existing
CUDA_VISIBLE_DEVICESand failed startup. All 10 selected worker/conversion tests pass, along with lint and formatting. Tested with TensorFlow 2.21 on CPU. GPU hardware was unavailable.