Skip to content

Initial Release#2

Open
michaelbeale-IL wants to merge 7 commits into
mainfrom
to_push
Open

Initial Release#2
michaelbeale-IL wants to merge 7 commits into
mainfrom
to_push

Conversation

@michaelbeale-IL
Copy link
Copy Markdown
Contributor

No description provided.

- Implement Food101 dataset class for few-shot learning support.
- Create ImageNet dataset class with preprocessed data handling.
- Add ImageNetA, ImageNetR, ImageNetSketch, and ImageNetV2 classes for various ImageNet subsets.
- Introduce OxfordFlowers dataset class with data splitting functionality.
- Develop OxfordPets dataset class with comprehensive data handling and subsampling.
- Implement StanfordCars dataset class with annotation loading.
- Create SUN397 dataset class with class name mapping.
- Add UCF101 dataset class for action recognition tasks.
- Introduce a new training script to manage training configurations and execution.
- Extend configuration management to support new dataset and training parameters.
- Introduced ImageNet templates for image classification tasks in `imagenet_templates.py`.
- Implemented LoRA (Low-Rank Adaptation) layers in `loralib` including `LoRALayer`, `LinearLoRA`, `Embedding`, and various convolutional layers.
- Added utility functions for managing LoRA parameters and saving/loading weights in `utils.py`.
- Enhanced multi-head attention with LoRA capabilities in `easymultiheadattention.py`.
- Updated `__init__.py` to include new modules from `loralib`.
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