search

glm-tools / pyglmnet

Python implementation of elastic-net regularized generalized linear models
http://glm-tools.github.io/pyglmnet/
MIT License
279 stars 84 forks source link

[MRG] gamma link function #213

Closed jasmainak closed 7 years ago

jasmainak commented 7 years ago

closes https://github.com/glm-tools/pyglmnet/issues/189

jasmainak commented 7 years ago

Success @pavanramkumar ! The only problem is that right now the exponential explodes, but the way I wrote the gradients, it should be easy to replace it with any other link function. :beers:

codecov-io commented 7 years ago

Codecov Report

Merging #213 into master will increase coverage by 0.6%. The diff coverage is 75%.

Impacted file tree graph

@@            Coverage Diff            @@
##           master     #213     +/-   ##
=========================================
+ Coverage   77.26%   77.86%   +0.6%     
=========================================
  Files           4        4             
  Lines         475      497     +22     
  Branches       86       92      +6     
=========================================
+ Hits          367      387     +20     
- Misses         76       78      +2     
  Partials       32       32
Impacted Files Coverage Δ
pyglmnet/utils.py 44.61% <66.66%> (+2.67%) :arrow_up:
pyglmnet/pyglmnet.py 81.81% <76.19%> (+0.37%) :arrow_up:
doc/_themes/sphinx_rtd_theme/__init__.py
pyglmnet/__init__.py 100% <0%> (ø)

Continue to review full report at Codecov.

Legend - Click here to learn more Δ = absolute <relative> (impact), ø = not affected, ? = missing data Powered by Codecov. Last update ed545eb...d01ebd1. Read the comment docs.

jasmainak commented 7 years ago

@pavanramkumar I am going to merge this PR with the hessian loss unimplemented. If you don't mind, I want to add proper tests before implementing it and do a couple of refactoring which will make my life much easier