TorchEnsemble-Community / Ensemble-Pytorch

A unified ensemble framework for PyTorch to improve the performance and robustness of your deep learning model.
https://ensemble-pytorch.readthedocs.io
BSD 3-Clause "New" or "Revised" License
1.05k stars 95 forks source link

Logging performance on training dataset #145

Closed singhabhinav closed 1 year ago

singhabhinav commented 1 year ago

Hi,

I am using voting classifier. Is there anyway to log performance on training set also, so that I can plot training and validation loss and figure out if my model is underfitting or overfitting.

xuyxu commented 1 year ago

Maybe you could redirect the print result to a local text file, and parse it to get the training loss. Tensorboard is not used in VotingClassifier since it cannot support parallelization well.