Table 1.
Model overview. Overview of the descriptors and parameters used for the development of the machine learning models.
|
Model |
Descriptor |
Split |
Training |
Hyper‐parameter search |
Balancing technique |
|---|---|---|---|---|---|
|
Random Forest |
10 RDKit |
80/20 |
CV+ External test |
no |
no |
|
Gradient Boosting |
Atom Pair FP |
80/20 |
Nested CV+ External test |
Across model algorithms and inputs |
no |
|
Deep learning |
RDKit all |
80/20 |
Nested CV+ External test |
Network architecture |
no |