Lightning-AI/pytorch-lightning
Ver no GitHubDocumentation: writing custom samplers compatible with multi GPU training
Open
#19.964 aberto em 10 de jun. de 2024
docshelp wanted
Métricas do repositório
- Stars
- (26.687 stars)
- Métricas de merge de PR
- (Mesclagem média 9d 15h) (3 fundiu PRs em 30d)
Description
📚 Documentation
Hi,
I'm trying to run distributed training with a custom sampler for the first time. The idea is rather simple (fixed budget for each class) and works fine in single GPU. When moving to multi GPU, unsurprisingly I get an error message, which tells me that I should subclass BatchSampler.
TypeError: Lightning can't inject a (distributed) sampler into your batch sampler, because it doesn't subclass PyTorch's `BatchSampler`. To mitigate this, either follow the API of `BatchSampler` or set `Trainer(use_distributed_sampler=False)`. If you choose the latter, you will be responsible for handling the distributed sampling within your batch sampler.
It is my understanding that torch's BatchSampler takes one (single-sample) Sampler and samples from that repeatedly to fill up the batch size. Are there any guidelines for how samplers should be built to be compatible with the sampler injection? I can't seem to find it in the docs.
cc @borda