facebookresearch / esm

Evolutionary Scale Modeling (esm): Pretrained language models for proteins
MIT License
3.16k stars 627 forks source link

Warning bieng raised with 'cpu-offloading' option and generating structure takes forever #699

Open amani-acog opened 2 months ago

amani-acog commented 2 months ago

NOTE: if this is not a bug report, please use the GitHub Discussions for support questions (How do I do X?), feature requests, ideas, showcasing new applications, etc.

Bug description I am running esmfold in a docker container with GPUs enabled (not fully available gpu memory). I am getting the following warning: UserWarning: Module is put on CPU and will thus have flattening and sharding run on CPU, which is less efficient than on GPU. We recommend passing in device_id argument which will enable FSDP to put module on GPU device, module must also be on GPU device to work with sync_module_states=True flag which requires GPU communication. "Module is put on CPU and will thus have flattening and sharding"

However the there is no error and execution seems to be going on, but there is no trace of CPU or GPU usage. It did not generate predictions either.

Reproduction steps

Expected behavior Give a clear and concise description of what you expected to happen.

Logs Please paste the command line output:

Output goes here

Additional context Add any other context about the problem here. (like proxy settings, network setup, overall goals, etc.)