Adaptive Learning

[rodrigo-arenas]

This PR introduces the following changes

• New Adaptive Schedulers to enable adaptive mutation and crossover probabilities; currently, supported schedulers are:

  • ExponentialAdapter
  • InverseAdapter
  • PotentialAdapter

• Changed the default values of mutation_probability and crossover_probability to 0.8 and 0.2, respectively.

• The weighted_choice function used in :class:~sklearn_genetic.GAFeatureSelectionCV was re-written to give more probability to a number of features closer to the max_features parameter

[codecov[bot]]

Codecov Report

Merging #95 (7a817e1) into master (667e396) will decrease coverage by 0.42%. The diff coverage is 92.18%.

@@            Coverage Diff             @@
##           master      #95      +/-   ##
==========================================
- Coverage   98.98%   98.55%   -0.43%     
==========================================
  Files          20       23       +3     
  Lines         982     1036      +54     
==========================================
+ Hits          972     1021      +49     
- Misses         10       15       +5     

Impacted Files	Coverage Δ
sklearn_genetic/algorithms.py	93.04% <81.81%> (-1.60%)	:arrow_down:
sklearn_genetic/schedules/base.py	90.90% <90.90%> (ø)
sklearn_genetic/__init__.py	100.00% <100.00%> (ø)
sklearn_genetic/_version.py	100.00% <100.00%> (ø)
sklearn_genetic/genetic_search.py	99.73% <100.00%> (-0.01%)	:arrow_down:
sklearn_genetic/schedules/__init__.py	100.00% <100.00%> (ø)
sklearn_genetic/schedules/schedulers.py	100.00% <100.00%> (ø)
sklearn_genetic/utils/random.py	100.00% <100.00%> (ø)

:mega: Codecov can now indicate which changes are the most critical in Pull Requests. Learn more