corels / pycorels

Public home of pycorels, the python binding to CORELS

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

predict_proba : is there way to calculate probability of correct decision

Sandy4321 opened this issue · comments

is there way to calculate probability of correct decision
usually it classier has attribute predict_proba

like
https://scikit-learn.org/stable/modules/generated/sklearn.linear_model.LogisticRegression.html

predict_proba(self, X) Probability estimates.

Doesn't look like it, but it could be implemented like for a decision tree, rather than for a logistic regression. Remember this is not the "probability of correct decision"; only the fraction of the training data that falls into each class for each leaf.

See https://scikit-learn.org/stable/modules/generated/sklearn.tree.DecisionTreeClassifier.html#sklearn.tree.DecisionTreeClassifier.predict_proba.

Great observation thanks
But what is leave in Corel case
I get only one rule
For example
If (*********)
Then yes
Else
No