Ensemble-Pytorch

Ensemble-Pytorch copied to clipboard

split_data_target issue

4

Hi, I'm using a Dataset class (extending torch.utils.data.Dataset) which contains this method: ```python def __getitem__(self, idx) -> tuple[list[torch.Tensor], torch.Tensor] ``` Unfortunately, the `split_data_target` in `torchensemble.utils.io` can't handle that. ```python if...

jfulem

xuyxu

Bumps [black](https://github.com/psf/black) from 20.8b1 to 24.3.0. Release notes Sourced from black's releases. 24.3.0 Highlights This release is a milestone: it fixes Black's first CVE security vulnerability. If you run Black...

dependabot[bot]

can not fine-tuning after reload the model weights

6

``` n_bases=2 softGBM = SoftGradientBoostingRegressor( estimator=MLP, n_estimators=n_bases, shrinkage_rate=1.00, cuda=True ) io.load(softGBM, save_dir='./torch_ensemble_results/softGBM/') # reload criterion = StepwiseMSELoss() softGBM.set_criterion(criterion) softGBM.set_optimizer('Adam', lr=0.001, weight_decay=5e-4) softGBM.set_scheduler("ReduceLROnPlateau") # Re-training softGBM.fit(train_loader=new_train_loader, log_interval=128, epochs=20, test_loader=new_vali_loader, save_model=True, save_dir='./torch_ensemble_results/softGBM/')...

wubizhi

How can I save the ensemble which has best validation loss during training?

fatcatZF

question about expected speedup when using parallelization via joblib

3

Hi, I came across your repository while searching for ways to train multiple NN models simultaneously using 1 single GPU. My model is pretty small (just 1 layer MLP) and...

nguyentr17