[scikit-learn] Possible Feature Suggestions.

[Lekan Wahab]

Hello, new contributor here.
I've been meaning to contribute to the library for a while now but I
haven't found anything easy or clearenough for me to.
While going through the source today, I noticed some possible features I
could implement and would love to run it by the team here to see which is
feasible and which is not.

   1.

   More dataset benchmark. I noticed the benchmarksfolder only has
   benchmark on one of the datasets. MNIST. My plan is to add benchmarks
   for more datasets like iris, 'wine' and boston datasets.
   2.

   Implement a Batch Gradient Descent Regressor and a Mini Batch Gradient
   Regressor just like the Stochastic Gradient Regressor available in the
   linear_model module.

This is really my first attempt at contributing to the package so if
there's anything i'm missing about either feature suggestions, please, do
let me know.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20180531/34547ead/attachment.html>