nf-core / deepmodeloptim

Stochastic Testing and Input Manipulation for Unbiased Learning Systems
https://nf-co.re/deepmodeloptim
MIT License
23 stars 9 forks source link

[feat] save the models during tuning #94

Closed suzannejin closed 6 months ago

mathysgrapotte commented 7 months ago

You can use what was done here : https://github.com/mathysgrapotte/model-check/blob/master/bin/launch_training.py

Specifically, dfs is the output of tuner.fit() , also the output of tune in the tunewrapper. Calling "get_best_result" on it should do what you want (saving the best model), there should be other way to save checkpoints