1 year ago

#342191

test-img

leskovecg

UserWarning: One or more of the test scores are non-finite

I'm a relatively new user to python and have question about warning. I have dataframe that has shape of (96350, 156). With that I am building my decison tree model and using grid search cross validation. Why I am getting that output and how can I solve it?

Input:

grid_param = {
    'criterion': ['squared_error', 'friedman_mse', 'absolute_error'],
    'splitter': ['best', 'random'],
    'min_samples_split': [2, 3],
    'max_features': ['auto', 'sqrt', 'log2']
}

gd_sr = GridSearchCV(estimator = DtReg,
                     param_grid = grid_param,
                     scoring = 'accuracy',
                     cv = 5,
                     n_jobs = -1)

DtReg = gd_sr.fit(X_train, y_train)

plt.figure(figsize = (20, 10))
plot_tree(DtReg.best_estimator_, filled=True, rounded=True)

Output:

UserWarning: One or more of the test scores are non-finite: [nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan]

python

pandas

dataframe

cross-validation

grid-search

0 Answers

Your Answer

Accepted video resources