在运行cli_demo.py进行部署时报错如下:
[2023-07-24 06:19:37,012] [INFO] [real_accelerator.py:133:get_accelerator] Setting ds_accelerator to cuda (auto detect)
/remote-home/cwli/anaconda3/envs/minigpt4/lib/python3.9/site-packages/torchvision/io/image.py:13: UserWarning: Failed to load image Python extension: libtorch_cuda_cu.so: cannot open shared object file: No such file or directory
warn(f"Failed to load image Python extension: {e}")
===================================BUG REPORT===================================
Welcome to bitsandbytes. For bug reports, please submit your error trace to: https://github.com/TimDettmers/bitsandbytes/issues
[2023-07-24 06:19:39,268] [WARNING] Failed to load bitsandbytes:cannot import name 'LinearNF4' from 'bitsandbytes.nn' (/remote-home/cwli/anaconda3/envs/minigpt4/lib/python3.9/site-packages/bitsandbytes/nn/init.py)
Exception ignored in: <function BaseFileLock.del at 0x7f0ad473d820>
Traceback (most recent call last):
File "/remote-home/cwli/anaconda3/envs/minigpt4/lib/python3.9/site-packages/filelock/_api.py", line 240, in del
self.release(force=True)
File "/remote-home/cwli/anaconda3/envs/minigpt4/lib/python3.9/site-packages/filelock/_api.py", line 201, in release
with self._thread_lock:
AttributeError: 'UnixFileLock' object has no attribute '_thread_lock'
╭─────────────────────────────── Traceback (most recent call last) ────────────────────────────────╮
│ /remote-home/cwli/VisualGLM-6B/cli_demo.py:103 in │
│ │
│ 100 │
│ 101 │
│ 102 if name == "main": │
│ ❱ 103 │ main() │
│ 104 │
│ │
│ /remote-home/cwli/VisualGLM-6B/cli_demo.py:30 in main │
│ │
│ 27 │ args = parser.parse_args() │
│ 28 │ │
│ 29 │ # load model │
│ ❱ 30 │ model, model_args = AutoModel.from_pretrained( │
│ 31 │ │ args.from_pretrained, │
│ 32 │ │ args=argparse.Namespace( │
│ 33 │ │ fp16=True, │
│ │
│ /remote-home/cwli/anaconda3/envs/minigpt4/lib/python3.9/site-packages/sat/model/base_model.py:31 │
│ 0 in from_pretrained │
│ │
│ 307 │ @classmethod │
│ 308 │ def from_pretrained(cls, name, args=None, *, home_path=None, url=None, prefix='', bu │
│ 309 │ │ if build_only or 'model_parallel_size' not in overwrite_args: │
│ ❱ 310 │ │ │ return cls.from_pretrained_base(name, args=args, home_path=home_path, url=ur │
│ 311 │ │ else: │
│ 312 │ │ │ new_model_parallel_size = overwrite_args['model_parallel_size'] │
│ 313 │ │ │ model, model_args = cls.from_pretrained_base(name, args=args, home_path=home │
│ │
│ /remote-home/cwli/anaconda3/envs/minigpt4/lib/python3.9/site-packages/sat/model/base_model.py:28 │
│ 5 in from_pretrained_base │
│ │
│ 282 │ │ if os.path.exists(name) and os.path.isdir(name): │
│ 283 │ │ │ model_path = name │
│ 284 │ │ else: │
│ ❱ 285 │ │ │ model_path = auto_create(name, path=home_path, url=url) │
│ 286 │ │ if args is None: │
│ 287 │ │ │ args = argparse.Namespace() # null, fill later │
│ 288 │ │ │ null_args = True │
│ │
│ /remote-home/cwli/anaconda3/envs/minigpt4/lib/python3.9/site-packages/sat/resources/download.py: │
│ 50 in auto_create │
│ │
│ 47 │ if url == 'local': │
│ 48 │ │ return model_path │
│ 49 │ os.makedirs(os.path.dirname(model_path), exist_ok=True) │
│ ❱ 50 │ lock = FileLock(model_path + '.lock', mode=0o777) │
│ 51 │ with lock: │
│ 52 │ │ if url is None: │
│ 53 │ │ │ url = MODEL_URLS[name] │
╰──────────────────────────────────────────────────────────────────────────────────────────────────╯
TypeError: init() got an unexpected keyword argument 'mode'
请问这是什么问题呢
在运行cli_demo.py进行部署时报错如下: [2023-07-24 06:19:37,012] [INFO] [real_accelerator.py:133:get_accelerator] Setting ds_accelerator to cuda (auto detect) /remote-home/cwli/anaconda3/envs/minigpt4/lib/python3.9/site-packages/torchvision/io/image.py:13: UserWarning: Failed to load image Python extension: libtorch_cuda_cu.so: cannot open shared object file: No such file or directory warn(f"Failed to load image Python extension: {e}")
===================================BUG REPORT=================================== Welcome to bitsandbytes. For bug reports, please submit your error trace to: https://github.com/TimDettmers/bitsandbytes/issues
[2023-07-24 06:19:39,268] [WARNING] Failed to load bitsandbytes:cannot import name 'LinearNF4' from 'bitsandbytes.nn' (/remote-home/cwli/anaconda3/envs/minigpt4/lib/python3.9/site-packages/bitsandbytes/nn/init.py) Exception ignored in: <function BaseFileLock.del at 0x7f0ad473d820> Traceback (most recent call last): File "/remote-home/cwli/anaconda3/envs/minigpt4/lib/python3.9/site-packages/filelock/_api.py", line 240, in del self.release(force=True) File "/remote-home/cwli/anaconda3/envs/minigpt4/lib/python3.9/site-packages/filelock/_api.py", line 201, in release with self._thread_lock: AttributeError: 'UnixFileLock' object has no attribute '_thread_lock' ╭─────────────────────────────── Traceback (most recent call last) ────────────────────────────────╮ │ /remote-home/cwli/VisualGLM-6B/cli_demo.py:103 in │
│ │
│ 100 │
│ 101 │
│ 102 if name == "main": │
│ ❱ 103 │ main() │
│ 104 │
│ │
│ /remote-home/cwli/VisualGLM-6B/cli_demo.py:30 in main │
│ │
│ 27 │ args = parser.parse_args() │
│ 28 │ │
│ 29 │ # load model │
│ ❱ 30 │ model, model_args = AutoModel.from_pretrained( │
│ 31 │ │ args.from_pretrained, │
│ 32 │ │ args=argparse.Namespace( │
│ 33 │ │ fp16=True, │
│ │
│ /remote-home/cwli/anaconda3/envs/minigpt4/lib/python3.9/site-packages/sat/model/base_model.py:31 │
│ 0 in from_pretrained │
│ │
│ 307 │ @classmethod │
│ 308 │ def from_pretrained(cls, name, args=None, *, home_path=None, url=None, prefix='', bu │
│ 309 │ │ if build_only or 'model_parallel_size' not in overwrite_args: │
│ ❱ 310 │ │ │ return cls.from_pretrained_base(name, args=args, home_path=home_path, url=ur │
│ 311 │ │ else: │
│ 312 │ │ │ new_model_parallel_size = overwrite_args['model_parallel_size'] │
│ 313 │ │ │ model, model_args = cls.from_pretrained_base(name, args=args, home_path=home │
│ │
│ /remote-home/cwli/anaconda3/envs/minigpt4/lib/python3.9/site-packages/sat/model/base_model.py:28 │
│ 5 in from_pretrained_base │
│ │
│ 282 │ │ if os.path.exists(name) and os.path.isdir(name): │
│ 283 │ │ │ model_path = name │
│ 284 │ │ else: │
│ ❱ 285 │ │ │ model_path = auto_create(name, path=home_path, url=url) │
│ 286 │ │ if args is None: │
│ 287 │ │ │ args = argparse.Namespace() # null, fill later │
│ 288 │ │ │ null_args = True │
│ │
│ /remote-home/cwli/anaconda3/envs/minigpt4/lib/python3.9/site-packages/sat/resources/download.py: │
│ 50 in auto_create │
│ │
│ 47 │ if url == 'local': │
│ 48 │ │ return model_path │
│ 49 │ os.makedirs(os.path.dirname(model_path), exist_ok=True) │
│ ❱ 50 │ lock = FileLock(model_path + '.lock', mode=0o777) │
│ 51 │ with lock: │
│ 52 │ │ if url is None: │
│ 53 │ │ │ url = MODEL_URLS[name] │
╰──────────────────────────────────────────────────────────────────────────────────────────────────╯
TypeError: init() got an unexpected keyword argument 'mode'
请问这是什么问题呢