Interpretting machine learning regression scores

10 views
Skip to first unread message

Niveditha Kumar

unread,
Jul 25, 2019, 7:27:59 AM7/25/19
to scikit-fmm
Please find attached my results for the machine learning algorithm used. I tried to predict revenue with a number of features. Clearly, Random forest is a good predictor compared to Linear and Ridge. Is it safe to say that Random Forest performs better than Gradient boosting with respect to MAE, even though the R^2 score is higher in the case of Gradient boosting? This is because I have a lot of outliers in my data and hence MAE seems like a good comparison.
outliers.PNG
Reply all
Reply to author
Forward
0 new messages