Could not consume arg: --local_rank

[Droliven]

After finetuning the model with A100 80G * 4 GPU with the following commands:

CUDA_VISIBLE_DEVICES=0,1,2,3 python3 -m torch.distributed.launch --nproc_per_node=4 --nnodes=$WORLD_SIZE --node_rank=$RANK --master_addr=$MASTER_ADDR --master_port=29005 \
    lora_finetune.py \
    --base_model model_zoos/lmsys_vicuna-13b-v1.3 \
    --data_path datas/gpt4tools_instructions/origin/gpt4tools_71k.json \
    --output_dir outputs/gpt4tools \
    --prompt_template_name gpt4tools \
    --num_epochs 6 \
    --batch_size 512 \
    --cutoff_len 2048 \
    --group_by_length \
    --lora_target_modules '[q_proj,k_proj,v_proj,o_proj]' \
    --lora_r 16 \
    --micro_batch_size=8 \
    2>&1 | tee outputs/gpt4tools/log_`date +%Y%m%d-%H%M%S`.out

it shows:

100%|██████████| 834/834 [16:13:11<00:00, 70.01s/it]

 If there's a warning about missing keys above, please disregard :)
ERROR: Could not consume arg: --local_rank=3
Usage: lora_finetune.py --local_rank=3 --base_model /input/danglingwei.dlw/model_zoos/lmsys_vicuna-13b-v1.3 --data_path /input/danglingwei.dlw/datas/gpt4tools_instructions/origin/gpt4tools_71k.json --output_dir /input/danglingwei.dlw/outputs/gpt4tools --prompt_template_name gpt4tools --num_epochs 6 --batch_size 512 --cutoff_len 2048 --group_by_length --lora_target_modules '[q_proj,k_proj,v_proj,o_proj]' --lora_r 16 -

For detailed information on this command, run:
  lora_finetune.py --local_rank=3 --base_model /input/danglingwei.dlw/model_zoos/lmsys_vicuna-13b-v1.3 --data_path /input/danglingwei.dlw/datas/gpt4tools_instructions/origin/gpt4tools_71k.json --output_dir /input/danglingwei.dlw/outputs/gpt4tools --prompt_template_name gpt4tools --num_epochs 6 --batch_size 512 --cutoff_len 2048 --group_by_length --lora_target_modules '[q_proj,k_proj,v_proj,o_proj]' --lora_r 16 - --help

 If there's a warning about missing keys above, please disregard :)
ERROR: Could not consume arg: --local_rank=2
Usage: lora_finetune.py --local_rank=2 --base_model /input/danglingwei.dlw/model_zoos/lmsys_vicuna-13b-v1.3 --data_path /input/danglingwei.dlw/datas/gpt4tools_instructions/origin/gpt4tools_71k.json --output_dir /input/danglingwei.dlw/outputs/gpt4tools --prompt_template_name gpt4tools --num_epochs 6 --batch_size 512 --cutoff_len 2048 --group_by_length --lora_target_modules '[q_proj,k_proj,v_proj,o_proj]' --lora_r 16 -

For detailed information on this command, run:
  lora_finetune.py --local_rank=2 --base_model /input/danglingwei.dlw/model_zoos/lmsys_vicuna-13b-v1.3 --data_path /input/danglingwei.dlw/datas/gpt4tools_instructions/origin/gpt4tools_71k.json --output_dir /input/danglingwei.dlw/outputs/gpt4tools --prompt_template_name gpt4tools --num_epochs 6 --batch_size 512 --cutoff_len 2048 --group_by_length --lora_target_modules '[q_proj,k_proj,v_proj,o_proj]' --lora_r 16 - --help

 If there's a warning about missing keys above, please disregard :)
ERROR: Could not consume arg: --local_rank=1
Usage: lora_finetune.py --local_rank=1 --base_model /input/danglingwei.dlw/model_zoos/lmsys_vicuna-13b-v1.3 --data_path /input/danglingwei.dlw/datas/gpt4tools_instructions/origin/gpt4tools_71k.json --output_dir /input/danglingwei.dlw/outputs/gpt4tools --prompt_template_name gpt4tools --num_epochs 6 --batch_size 512 --cutoff_len 2048 --group_by_length --lora_target_modules '[q_proj,k_proj,v_proj,o_proj]' --lora_r 16 -

For detailed information on this command, run:
  lora_finetune.py --local_rank=1 --base_model /input/danglingwei.dlw/model_zoos/lmsys_vicuna-13b-v1.3 --data_path /input/danglingwei.dlw/datas/gpt4tools_instructions/origin/gpt4tools_71k.json --output_dir /input/danglingwei.dlw/outputs/gpt4tools --prompt_template_name gpt4tools --num_epochs 6 --batch_size 512 --cutoff_len 2048 --group_by_length --lora_target_modules '[q_proj,k_proj,v_proj,o_proj]' --lora_r 16 - --help

 If there's a warning about missing keys above, please disregard :)
ERROR: Could not consume arg: --local_rank=0
Usage: lora_finetune.py --local_rank=0 --base_model /input/danglingwei.dlw/model_zoos/lmsys_vicuna-13b-v1.3 --data_path /input/danglingwei.dlw/datas/gpt4tools_instructions/origin/gpt4tools_71k.json --output_dir /input/danglingwei.dlw/outputs/gpt4tools --prompt_template_name gpt4tools --num_epochs 6 --batch_size 512 --cutoff_len 2048 --group_by_length --lora_target_modules '[q_proj,k_proj,v_proj,o_proj]' --lora_r 16 -

For detailed information on this command, run:
  lora_finetune.py --local_rank=0 --base_model /input/danglingwei.dlw/model_zoos/lmsys_vicuna-13b-v1.3 --data_path /input/danglingwei.dlw/datas/gpt4tools_instructions/origin/gpt4tools_71k.json --output_dir /input/danglingwei.dlw/outputs/gpt4tools --prompt_template_name gpt4tools --num_epochs 6 --batch_size 512 --cutoff_len 2048 --group_by_length --lora_target_modules '[q_proj,k_proj,v_proj,o_proj]' --lora_r 16 - --help
/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py:180: FutureWarning: The module torch.distributed.launch is deprecated
and will be removed in future. Use torchrun.
Note that --use_env is set by default in torchrun.
If your script expects `--local_rank` argument to be set, please
change it to read from `os.environ['LOCAL_RANK']` instead. See 
https://pytorch.org/docs/stable/distributed.html#launch-utility for 
further instructions

  warnings.warn(
WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 570 closing signal SIGTERM
WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 571 closing signal SIGTERM
WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 572 closing signal SIGTERM
ERROR:torch.distributed.elastic.multiprocessing.api:failed (exitcode: 2) local_rank: 3 (pid: 573) of binary: /opt/conda/bin/python3
Traceback (most recent call last):
  File "/opt/conda/lib/python3.8/runpy.py", line 194, in _run_module_as_main
    return _run_code(code, main_globals, None,
  File "/opt/conda/lib/python3.8/runpy.py", line 87, in _run_code
    exec(code, run_globals)
  File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 195, in <module>
    main()
  File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 191, in main
    launch(args)
  File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 176, in launch
    run(args)
  File "/opt/conda/lib/python3.8/site-packages/torch/distributed/run.py", line 753, in run
    elastic_launch(
  File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launcher/api.py", line 132, in __call__
    return launch_agent(self._config, self._entrypoint, list(args))
  File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launcher/api.py", line 246, in launch_agent
    raise ChildFailedError(
torch.distributed.elastic.multiprocessing.errors.ChildFailedError: 
============================================================
lora_finetune.py FAILED
------------------------------------------------------------
Failures:
  <NO_OTHER_FAILURES>
------------------------------------------------------------
Root Cause (first observed failure):
[0]:
  time      : 2023-09-18_05:24:03
  host      : zcoregputrain-55-011068195071
  rank      : 3 (local_rank: 3)
  exitcode  : 2 (pid: 573)
  error_file: <N/A>
  traceback : To enable traceback see: https://pytorch.org/docs/stable/elastic/errors.html
============================================================

[Yangr116]

We use torchrun to launch the training script, you can check here to see its difference with torch.distributed.launch on the local_rank.
Moreover, you can change the fire into argparse.