[Feature]: Abstraction of GenericDataset
and DfDataset
#101
Labels
🌀 cleanup
Related to code or repository cleanup
🔧 enhancement
Requests for improvements or enhancements to existing features
✨ feature
Requests or discussions about new features
🚨 high priority
Needs immediate attention or urgent resolution
💡 idea
Ideas or suggestions for future enhancements
📋 needs review
Requires code or design review
♻️ refactoring
Related to refactoring or restructuring code
✏️ Problem Description
At the moment there is a
KerasGenericDataset
and aKerasDfDataset
that implementsGenericDataset
orDfDataset
.This is because there are some customizations for Keras. If we use PyTorch in a future release, there are some PyTorch specific functions within the abstract
GenericDataset
orDfDataset
. This should be fixed.PyTorch-specific💡 Feature Request
Make
GenericDataset
andDfDataset
more abstract and move PyTorch-specific functions to a corresponding PyTorch dataset.I would prefer to move every framework-specific code inside a separate library.
🌍 Context
This makes it easier to make frame-specific changes in the right place.
🔍 Additional Information
No response
👍 Code of Conduct
The text was updated successfully, but these errors were encountered: