Stas Bekman
Stas Bekman
**Describe the bug** As the models get huge offloading is becoming more and more a thing, especially with inference. This Issue discusses inference, so it's quite simple moving blocks-wise -...
**Describe the bug** Currently HF Transformers integration users can finetune a model and save the checkpoint with given resources. However resuming from that same checkpoint requires much more CPU peak...
All examples I have seen use a docker image, which has a `user` parameter. But I'm not using docker. How do I then tell the action-runner to run as non-root...
To ensure that no more than one instance is running on several consequent pushes, the idea is to cancel the previous still running workflows on a new push, so we...
An attempt to fix the issue reported https://github.com/huggingface/transformers/issues/17336 Fixes: https://github.com/huggingface/transformers/issues/17336
### Feature request Add alibi cache for BLOOM ### Motivation Many training scenarios involve fixed input length. In which case re-creating the deterministic alibi tensor that is the same on...
As discussed here https://github.com/CFMTech/pytest-monitor/pull/38 `pytest` almost fully supports `unittest`, but currently `pytest-monitor` skips all `unittest`-based tests. I originally proposed to document this, but it was suggested that perhaps this can...
Splitting off from https://github.com/CFMTech/pytest-monitor/pull/38#issuecomment-831420330 to focus the discussion on nuances of memory measurement side of this extension. And also I will bring up some memory leak detection issue, which I...
Splitting of from https://github.com/benfred/py-spy/issues/395, some envs give no `sudo` access to its users (e.g. HPCs), and the need to be able to dump the trace on demand is big. The...
Came to this project for spelling in twitter text, but it doesn't quite work most of the time. 1. spell correction seems to only work when `annotate` is set as...