Open G10Rg10C opened 1 year ago
Got the same error when running the example file, not sure what the ‘RANK’ variable meaning except know it should be a int. Much appreciated if any suggestion.
Got the same error when running the example file, not sure what the ‘RANK’ variable meaning except know it should be a int. Much appreciated if any suggestion.
The real problem are not the problems but the fact that everytime I fix one error two more errors come out... That variable is about your GPU I think, it's not settled by default so you have to set it by your own, set I with zero up to me.
File "C:\Users\giorgio\OneDrive\Desktop\LLAMA2 MAIN\Conda Environment\Lib\site-packages\fire\core.py", line 141, in Fire component_trace = _Fire(component, args, parsed_flag_args, context, name) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "C:\Users\giorgio\OneDrive\Desktop\LLAMA2 MAIN\Conda Environment\Lib\site-packages\fire\core.py", line 475, in _Fire component, remaining_args = _CallAndUpdateTrace( ^^^^^^^^^^^^^^^^^^^^ File "C:\Users\giorgio\OneDrive\Desktop\LLAMA2 MAIN\Conda Environment\Lib\site-packages\fire\core.py", line 691, in _CallAndUpdateTrace component = fn(*varargs, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^ File "C:\Users\giorgio\OneDrive\Desktop\LLAMA2 MAIN\example_chat_completion.py", line 20, in main generator = Llama.build( ^^^^^^^^^^^^ File "C:\Users\giorgio\OneDrive\Desktop\LLAMA2 MAIN\llama\generation.py", line 62, in build torch.distributed.init_process_group("nccl") File "C:\Users\giorgio\OneDrive\Desktop\LLAMA2 MAIN\Conda Environment\Lib\site-packages\torch\distributed\distributed_c10d.py", line 900, in init_process_group store, rank, world_size = next(rendezvous_iterator) ^^^^^^^^^^^^^^^^^^^^^^^^^ File "C:\Users\giorgio\OneDrive\Desktop\LLAMA2 MAIN\Conda Environment\Lib\site-packages\torch\distributed\rendezvous.py", line 235, in _env_rendezvous_handler rank = int(_get_env_or_raise("RANK")) ^^^^^^^^^^^^^^^^^^^^^^^^^ File "C:\Users\giorgio\OneDrive\Desktop\LLAMA2 MAIN\Conda Environment\Lib\site-packages\torch\distributed\rendezvous.py", line 220, in _get_env_or_raise raise _env_error(env_var) ValueError: Error initializing torch.distributed using env:// rendezvous: environment variable RANK expected, but not set