Adding supervised

[ctr26]

PR Description: Supervised Branch Integration

Summary

This pull request introduces a new supervised learning framework to the bioimage_embed project. It integrates several key features and updates that enhance the model's capabilities and performance in supervised settings.

Key Changes

1. Supervised Learning Implementation:

   • Added modules and scripts to support training and evaluation of supervised models.
   • Updated pyproject.toml to include new dependencies required for supervised learning.

2. Data Handling Enhancements:

   • Modified data loading pipelines to accommodate labelled datasets.
   • Implemented new data augmentation techniques to improve model generalization.

3. Model Architecture:

   • Introduced new layers and architecture changes to optimize for supervised tasks.
   • Enhanced the model's ability to learn from labelled data by adding supervised loss functions.

4. Training and Evaluation Scripts:

   • Added comprehensive training scripts that support various supervised learning scenarios.
   • Implemented evaluation scripts to assess model performance on validation and test sets.

5. Documentation and Examples:

   • Updated documentation to reflect the new supervised learning capabilities.
   • Included example notebooks demonstrating how to train and evaluate models using the new supervised framework.

Improvements

• Enhanced model accuracy and robustness by leveraging supervised learning techniques.
• Improved data pipeline efficiency and flexibility.
• Expanded documentation and examples to facilitate user adoption of the new features.

Dependencies

• Added new dependencies such as torchvision and scikit-learn for handling supervised learning tasks.
• Updated existing dependencies to ensure compatibility with the new features.

Testing

• Thoroughly tested the new supervised learning features on multiple datasets.
• Ensured backward compatibility with existing unsupervised learning functionalities.

This update significantly boosts the bioimage_embed project by adding powerful supervised learning capabilities, improving model performance, and enhancing the overall usability of the codebase.

For more detailed information, please refer to the updated documentation and example notebooks provided in this PR.