Skip to content

Conversation

@Laurent2916
Copy link
Contributor

@Laurent2916 Laurent2916 commented Feb 25, 2023

The following adds instructions on how to create a DataLoader from the guide on how to use object detection with augmentations (#4710). I am open to hearing any suggestions for improvement !

@Laurent2916 Laurent2916 marked this pull request as ready for review February 27, 2023 10:48
@HuggingFaceDocBuilderDev

The docs for this PR live here. All of your documentation changes will be reflected on that endpoint.

@mariosasko
Copy link
Collaborator

I'm not sure we need this part as we provide a link to the notebook that shows how to train an object detection model, and this notebook instantiates a DataLoader before training the model. I'd like to hear what @stevhliu thinks.

PS: Your collate_fn calls torch.stack on the bbox tensors, which don't have the same shape, so this will fail.

@stevhliu
Copy link
Member

stevhliu commented Mar 7, 2023

I agree with @mariosasko; we also have a Use with PyTorch guide that shows how you can create a DataLoader.

@Laurent2916 Laurent2916 deleted the object-detection-dataloader branch March 23, 2023 19:24
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

4 participants