Several key updates, including:
- Addition of preliminary multiclass prediction capabilities (which is still under active development)
- Refinement of continuous module
- Bug fixes, both big and small
- Changes to munging and harmonization workflows to allow train and test data to be munged together upfront and ensure they are preprocessed under the same conditions
- File restructuring for increased modularity
- Addition of log file in outputs for reproducibility
thank you @spencermg for implementing these improvements and @nvk23 for feedback!
Contributors
nvk23 and spencermg