Skip to content

Documenting how to implement a DataLoader for image augmentation#141

Open
rkube wants to merge 1 commit intoJuliaML:mainfrom
rkube:main
Open

Documenting how to implement a DataLoader for image augmentation#141
rkube wants to merge 1 commit intoJuliaML:mainfrom
rkube:main

Conversation

@rkube
Copy link

@rkube rkube commented Feb 2, 2023

Hi,
I've added an example on how to integrate MLUtils.DataLoader with an image augmentation pipeline for a custom dataset. That's a common task for CNN-based image learning. Documentation for the individual pieces (dataloading, augmentation) is good, but I couldn't find a good example of how to integrate them into one piece.

@darsnack
Copy link
Member

darsnack commented Mar 6, 2023

This is great, thanks for the contribution! It might make more sense to add a new top-level section for "Tutorials" instead of adding more to the first page?

Also, MLUtils.jl (especially DataLoader) is the result of work directed towards building FastAI.jl. There, DataAugmentation.jl is used over Augmentor.jl. Maybe it makes sense to favor DataAugmentation.jl in the example, since future improvements would be to that package over Augmentor.jl. cc @lorenzoh for thoughts on this too.

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.

2 participants