Minhao Chou
Minhao Chou
Normally, you can hardly find any obvious performance gap between the two serving strategies. It is a good practice to deploy your service in docker, you can always expect to...
Yes，just recently I'm occupied with the work for the next few months, So the gpu support may come a little late.
嗯嗯，辛苦作者啦
I accidently figured out that you can use dape to debug specific test function in specific test file by simply runing this command: ``` debugpy-module :module "pytest" :args ["test_file_relative_path_to_base_project::test_method_name"] ```
Hi @inkcherry, Thanks for your quick reply, Did you saved checkpoint on shared filesystem? I guess this error only occurs when `use_node_local_storage` is enabled and model is divided to GPUs...
the bug disappears when I saved checkpoint to the shared filesystem.