Alexandre Ghelfi, PhD

Call `set_epoch` on `Dataloader.batch_sampler` if defined

1

# What does this PR do? Proposes a technical implementation to `set_epoch` on either `DataLoader.sampler` or `DataLoader.batch_sampler` if it is defined. # What issue(s) does this change relate to? -...

Set `torch.utils.data.DataLoader.batch_sampler` epoch if defined

1

## 🚀 Feature Request Current implementation of composer trainer call the `DistributedSampler.set_epoch` method only on the `Dataloader.sampler` attribute but not on the `Dataloader.batch_sampler`, even if it is defined. One example...

Adding Compare in torch.utils.benchmark documentation

3

`torch.utils.benchmark.Compare` is not directly exposed in torch.utils.benchmark documentation. I think this is a valuable resource to add since it can help people embracing the torch benchmark way of doing things,...

Error when using `Trainer.compile_config={}` in DDP mode

4

Training a toy example on DDP mode with the composer runtime while using both using `torch.compile` through `Trainer.compile_config={}` and `BlurPool` algorithm raises a dynamo error. ** To reproduce From develop...

Computing train metrics at a given frequency

1

## 🚀 Feature Request Add the possibility to compute metrics on the train dataset with a given frequency and avoid computing these every steps ## Motivation Having the possibility to...