0

I ran this code:

import pandas as pd
import sklearn.linear_model as lm
...
regression = lm.LogisticRegressionCV(random_state=0, solver='lbfgs', multi_class='ovr').fit(X, y)
print(round(regression.score(X,y), 4))

The regression score is 0.9765. I'd like to find the attributes that are most predictive of the classification. It is merely identifying the highest values in regression.coef_?

jsf80238
  • 1,577
  • 2
  • 11
  • 24
  • Sounds like linear discriminant analysis (LDA) is what you need. – Julien Dec 27 '19 at 01:59
  • This seems to have already been asked: https://stackoverflow.com/questions/34052115/how-to-find-the-importance-of-the-features-for-a-logistic-regression-model – jsf80238 Dec 27 '19 at 17:50

0 Answers0