model.predict() gives constant values

Question

model.predict() gives constant values

marastadler opened this issue 2 years ago · comments

Hi,

This package is super helpful :-)!

When applying LassonetRegressor to my data I get a constant model with model.predict(X) (for test or validation set), i.e. a vector of predictions where all entries are equal. But the feature importances still make sense. The same observation I made in the diabetes.py example.

Do you have any idea what this is?

Thanks a lot,
Mara

Louis Abraham · Answer 1 · Thu Mar 24 2022 18:30:25 GMT+0800 (China Standard Time)

After computing the path, the model is sparse and uses zero features. You need to load some history item with the number of features you want. I can provide an example if needed!

MaraStadler · Answer 2 · Thu Mar 24 2022 19:30:26 GMT+0800 (China Standard Time)

Thanks. Got it!

Louis Abraham · Answer 3 · Thu Mar 24 2022 20:05:41 GMT+0800 (China Standard Time)

You can find an example here: https://github.com/lasso-net/lassonet/blob/e3a3754ad258e67470e15457cb3fa6b8289653a5/examples/friedman.py