yudonglee

Here is the log from : DeepSpeedExamples/applications/DeepSpeed-Chat/output/actor-models/1.3b/training.log [2023-04-13 22:52:19,362] [WARNING] [runner.py:190:fetch_hostfile] Unable to find hostfile, will proceed with training with local resources only. [2023-04-13 22:52:19,416] [INFO] [runner.py:540:main] cmd = /home/ps/anaconda3/envs/pt2/bin/python...

> @yudonglee This looks to be a problem with `transformers` and `huggingface_hub`. Can you try installing the latest master of `transformers` and update `huggingface_hub` to the latest release? `pip install...

> Nvidia recommends that the version match exactly, however we have found that matching major version is usually sufficient (e.g., 11.1 and 11.3 should be compatible). We have this assertion...