Scale selection

[d-schindler]

I realised that the current implementation of the scale selection criterion does not recover all local minima of the criterion and that the criterion looks different to my original implementation. Therefore, I replaced the code with my original implementation. In particular, the moving mean is computed with pandas using a triangular window and this seems to be different to the numpy method used previously. However, we should get rid off pandas later.

[codecov-commenter]

Codecov Report

Merging #52 (2a61121) into master (f00fd8e) will increase coverage by 8.14%. The diff coverage is 60.30%.

@@            Coverage Diff             @@
##           master      #52      +/-   ##
==========================================
+ Coverage   41.85%   50.00%   +8.14%     
==========================================
  Files           8        8              
  Lines         571      586      +15     
==========================================
+ Hits          239      293      +54     
+ Misses        332      293      -39     

Flag	Coverage Δ
pytest	50.00% <60.30%> (+8.14%)	:arrow_up:

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

Impacted Files	Coverage Δ
src/pygenstability/plotting.py	0.00% <0.00%> (ø)
src/pygenstability/optimal_scales.py	50.72% <50.72%> (ø)
src/pygenstability/constructors.py	100.00% <100.00%> (+9.40%)	:arrow_up:
src/pygenstability/pygenstability.py	100.00% <100.00%> (+9.30%)	:arrow_up:

:mega: We’re building smart automated test selection to slash your CI/CD build times. Learn more

[arnaudon]

@d-schindler , I'm not sure this works, on the multiscale example it does not as compared to version on master branch. I'll let you fix it before we can merge

[d-schindler]

I made a small modification and it should work on our multiscale example. But it would be good to test it on other graphs and compare to the version of @peach-lucien

[d-schindler]

as scale selection has become a more central aspect of the whole package, I moved the file from contrib to pygenstability

[arnaudon]

@d-schindler , if this goes into the main package, could you add tests to it? thank you!

[arnaudon]

ah sorry, as it's called in the main function, it is tested, except the plotting, which we'll add later once the package it more stable, we can merge then!