learnables / torchml Goto Github PK

Should have:

• - a short description of project + logo
• - installation
• - link to docs
• - link to contribution.md and changelog.md
• - link to slack/zulip?
• - How to cite this repo.
• - Acknowledgement section.

Design a logo for torchml.

Provide guidelines for contributing to the library, in a CONTRIBUTING.md file. Should include:

• - Open an issues for discussion.
• - Start a PR with docs and tests.
• - Add your contribution / fix to CHANGELOG.md

Implementing ridge and lasso regression.

Please add the Nearest Neighbors Algos from scikit-learn.

• Color scheme
• Nice looking readme #15
• Full documentation (expand on what's already there)
• Tutorials sections with write-ups from our discussions
  • Ridge and Lasso
  • Gaussian Naive Bayes and Nearest Centroid
  • K Nearest Neighbors

• Linear Regression, Ridge, and Lasso
• Gaussian Naive Bayes, LDA, QDA, and Nearest Centroid
• K-Nearest Neighbors
• PCA and LDA for Dimensionality Reduction

Todos:

• debug so that the model produces the same output as scikit-learn
• vectorize the inner & outer loop to make the implementation more efficient
• rerun tests
• check all arguments are supported
• add docs
• add types

Please add KNeighborsClassifier

When running the LDA and QDA tests, they sometimes pass and sometimes fail. Track down what is happening and why that's the case.