Closed aolney closed 4 years ago
Email from TS:
I was looking at your final version and the solutions, I think it would be easier to teach them to us LogisticRegressionCV for this problem instead. Then, the tuning parameter selection is natural. Maybe and update for the future?
Closing with https://github.com/memphis-iis/datawhys-content-notebooks/commit/2a5efe95876e9a946bbdab4c4e79269854cdb2fc to clear the board.
But we will reopen at some point for revisions :smile_cat:
Need random seed to ensure train/test split gives the same answer across notebooks
See the spreadsheet for details
Ideas/prereqs: regularization, data splits/crossvalidation, overfit, bias/variance tradeoff,
Direct link https://jupyter.olney.ai/hub/user-redirect/git-pull?repo=https%3A%2F%2Fgithub.com%2Fmemphis-iis%2Fdatawhys-content-notebooks&subPath=Lasso.ipynb&app=lab