Pierre-Paul De Breuck

https://github.com/ppdebreuck/modnet/blob/a5ab7c66e7c8a35f5bf0fe455945ed513f2e1f11/modnet/hyper_opt/fit_genetic.py#L94 introduced in https://github.com/ppdebreuck/modnet/pull/198 will be problematic in python 3.11. Solution: add sorted()

missing features for prediction

2

If some features are missing (from the optimal ones), when featurizing novel compounds (for predicting stage), `predict()` will not work. This happens for instance when some elements are not present...

See https://github.com/ppdebreuck/modnet/blob/1fbf7b2f45aee5970ee3f5c6fd88461faefdbc65/modnet/preprocessing.py#L319 Not used at this stage but may be integrated to get RR score. _Reported by @gbrunin_

The [feature selection](https://github.com/ppdebreuck/modnet/blob/1fbf7b2f45aee5970ee3f5c6fd88461faefdbc65/modnet/preprocessing.py#L770) procedure relies on `mutual_info_regression` & `mutual_info_classif`which is stochastic. Therefore, feature selection is nondeterministic (the ranked list of optimal descriptors can have slight changes from run to run)....

Sklearn interface

1

The sklearn API implementation of MODNet can be found *under modnet.sklearn*. It enables integration with scikit-learn methods such as pipelines, model selection functions (e.g. gridsearch), and integration with other sklearn...