search

abess-team / abess

Fast Best-Subset Selection Library
https://abess.readthedocs.io/
Other
474 stars 41 forks source link

Add score function for OrdinalRegression #427

Closed chenpnn closed 2 years ago

chenpnn commented 2 years ago

add score function for ordinal regression

Mamba413 commented 2 years ago

Hi @chenpnn , many thanks for your contributions. Yet, the installation fails. The error is IndentationError: unexpected indent (see https://github.com/abess-team/abess/runs/6548472300?check_suite_focus=true). Please fixed this.

Before a new pull request, we strongly recommend you insure the package can be properly installed on your device.

codecov[bot] commented 2 years ago

Codecov Report

Merging #427 (30edf83) into master (dceabe9) will increase coverage by 0.00%. The diff coverage is 100.00%.

@@           Coverage Diff           @@
##           master     #427   +/-   ##
=======================================
  Coverage   95.85%   95.85%           
=======================================
  Files          29       29           
  Lines        3110     3115    +5     
=======================================
+ Hits         2981     2986    +5     
  Misses        129      129
Flag Coverage Δ
Python 96.23% <100.00%> (+0.01%) :arrow_up:
rpackage 95.65% <ø> (ø)

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files Coverage Δ
python/abess/linear.py 100.00% <100.00%> (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 778f015...30edf83. Read the comment docs.

oooo26 commented 2 years ago

Thank you for the contribution! There is only a small suggestion: could you please add a test [here](https://github.com/abess-team/abess/blob/master/python/pytest/test_alg.py#:~:text=%40staticmethod-,def%20test_ordinal()%3A,-np.random.seed) in python/pytest/test_alg.py to ensure it work?

It can be quite simple, just like:

model = abess.OrdinalRegression()
model.fit(....)
score = model.score(....)

And the desired result (if exists), like:

assert score > 0.5
Mamba413 commented 2 years ago

Hi @chenpnn , thanks for your contributions. Why not add the following command:

assert score > 0.5

If OrdinalRegression properly runs, we expect it gives a score that is larger than 0.5.

Mamba413 commented 2 years ago

@chenpnn , thanks! We will merge now.