CLP Tutorial 02: COIL-100

[elvinhajizada]

Issue Number:

706

Objective of pull request: Create the first tutorial on the initial version of CLP (Continually Learning Prototypes) algorithm, explaining how to set up and use the CLP processes from #706

Pull request checklist

Your PR fulfills the following requirements:

• [x] Issue created that explains the change and why it's needed
• [ ] Tests are part of the PR (for bug fixes / features)
• [ ] Docs reviewed and added / updated if needed (for bug fixes / features)
• [x] PR conforms to Coding Conventions
• [x] PR applys BSD 3-clause or LGPL2.1+ Licenses to all code files
• [x] Lint (flakeheaven lint src/lava tests/) and (bandit -r src/lava/.) pass locally
• [x] Build tests (pytest) passes locally

Pull request type

Please check your PR type:

• [ ] Bugfix
• [x] Feature
• [ ] Code style update (formatting, renaming)
• [ ] Refactoring (no functional changes, no api changes)
• [ ] Build related changes
• [ ] Documentation changes
• [ ] Other (please describe):

What is the current behavior?

• No CLP tutorial

What is the new behavior?

• The first CLP tutorial, describing the basics on the synthetic data

Does this introduce a breaking change?

• [ ] Yes
• [x] No

Supplemental information

[elvinhajizada]

Please ignore everything except the new tutorial notebook

[weidel-p]

I will review that in more detail, but I already saw some minor things you could change.

1. images should not go in this repository directly but should go here: https://github.com/lava-nc/lava-nc.github.io/tree/main/_static/images . This is to avoid the lava repository growing bigger.
2. Your tutorials are called tutorial01 and tutorial03. Is there a tutorial02 still coming? Otherwise relabel.
3. Usually, we have this "do you want to learn more" and "how to stay in contact" footer in our tutorials. Could you add them?
4. Add your tutorials to the test pipeline here: tests/lava/tutorials/test_tutorials.py

[weidel-p]

I added comments to the tutorials and started a PR to your branch. https://github.com/elvinhajizada/lava/pull/1

[PhilippPlank]

Tutorials use sklearn module, which is currently not part of lava.

[elvinhajizada]

The significant update to this PR includes:

• New CLP experiments are devised to test different aspects of the algorithm
• The abstract CLP class is improved to facilitate these experiments
• The user is given two options: use pre-extracted features or extracts features on your own (if you have torch & torchvision)
• Torch and torchvision dependencies are optional and if not there, code will use pre-extracted features
• Scikit-learn dependency is added and will be added to Lava requirements for the next release

[elvinhajizada]

The tests for CLP tutorials are commented out temporarily, till sk-learn is added to CI. Once that happens, I will add these tests before merge