scikit-learn-contrib / boruta_py

Python implementations of the Boruta all-relevant feature selection method.
BSD 3-Clause "New" or "Revised" License
1.5k stars 256 forks source link

Fix count for rejected features in final iteration #71

Closed Naehil closed 4 years ago

Naehil commented 4 years ago

In the final iteration, the number of rejected features are now printed correctly, so that confirmed/tentative/rejected add up to the total number of features.

In the current implementation, the following output is possible:

Iteration:  99 / 100
Confirmed:  0
Tentative:  8
Rejected:   251

BorutaPy finished running.

Iteration:  100 / 100
Confirmed:  0
Tentative:  4
Rejected:   251

which should be


Iteration:  99 / 100
Confirmed:  0
Tentative:  8
Rejected:   251

BorutaPy finished running.

Iteration:  100 / 100
Confirmed:  0
Tentative:  4
Rejected:   255
danielhomola commented 4 years ago

thanks @Naehil