Nick Becker

@divyegala , I was thinking of the explicit cross-validator utilities like `cross_val_score`. `KFold` and similar functions should work, as you suggested. ```python from cuml.datasets import make_regression from sklearn.model_selection import KFold...

As a note, the memory pressure is not observed during fit, regardless of whether the PredictionData is being generated. ```python import cuml import numpy as np import pandas as pd...

Gave this a test locally. It's really smooth! Quick comment: - When dispatching linear regression to CPU, the weights/bias attributes are not propagated back up to the cuML estimator --...

This would also be useful for tools like Boruta, a popular feature selection library that's part of scikit-learn-contrib. There is [a Boruta issue](https://github.com/scikit-learn-contrib/boruta_py/issues/99) asking for support for cuML estimators

Cross linking an issue that asks for this feature and OOB support https://github.com/rapidsai/cuml/issues/3361

It looks like this was resolved. @cjnolet is this still relevant?

With the merge of https://github.com/rapidsai/cuml/pull/4800, soft clustering the original dataset with `all_points_membership_vectors` is now available. Please give it a try and file issues if you run into any issues or...

Good question. Yes. In the canonical CPU library, this function is accessible from the module level namespace rather than as a method of the class. We've matched this user experience....

Thanks for commenting about this namespace issue. This will be resolved in #4895

The following example illustrates the general issue. Non-standard striding isn't supported by our input validation utilities which makes some estimators (including LinearRegression and Ridge Regression) throw errors. Where this works,...