Open
Conversation
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Work performed
After exporting a training dataset in YOLO format from Labelflow and send it as a Hub dataset using labelflow/labelflow#659, I created here a training script to use any Hub dataset for training, instead of a local Coco dataset as done originally. Additionally, I created a benchmark script "speed_test_hub.py" that compares the speed of data loading whether it's local or remote w/ hub. Results are encouraging.
Problems encountered
On macOS there is a PyTorch issue that hinders multiprocessing for data loading. I couldn't find a proper workaround (see https://stackoverflow.com/questions/64772335/pytorch-w-parallelnative-cpp206). Interestingly, that issue does not appear for regular training (with a local coco dataset), so there are some side effects due to the hub usage.