hilenti
commented
5 months ago --- Logging error ---
Traceback (most recent call last):
File "C:\Users\hei/.cache\huggingface\modules\transformers_modules\THUDM\chatglm-6b-int4\6c5205c47d0d2f7ea2e44715d279e537cae0911f\quantization.py", line 34, in <module>
kernels = CPUKernel(
^^^^^^^^^
NameError: name 'CPUKernel' is not defined
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "D:\anaconda3\envs\chatglm\Lib\logging\__init__.py", line 1110, in emit
msg = self.format(record)
^^^^^^^^^^^^^^^^^^^
File "D:\anaconda3\envs\chatglm\Lib\logging\__init__.py", line 953, in format
return fmt.format(record)
^^^^^^^^^^^^^^^^^^
File "D:\anaconda3\envs\chatglm\Lib\logging\__init__.py", line 687, in format
record.message = record.getMessage()
^^^^^^^^^^^^^^^^^^^
File "D:\anaconda3\envs\chatglm\Lib\logging\__init__.py", line 377, in getMessage
msg = msg % self.args
~~~~^~~~~~~~~~~
TypeError: not all arguments converted during string formatting
Call stack:
File "E:\ChatGLM-6B\cli_demo.py", line 8, in <module>
model = AutoModel.from_pretrained("THUDM/chatglm-6b-int4", trust_remote_code=True).float()
File "D:\anaconda3\envs\chatglm\Lib\site-packages\transformers\models\auto\auto_factory.py", line 466, in from_pretrained
return model_class.from_pretrained(
File "D:\anaconda3\envs\chatglm\Lib\site-packages\transformers\modeling_utils.py", line 2498, in from_pretrained
model = cls(config, *model_args, **model_kwargs)
File "C:\Users\hei/.cache\huggingface\modules\transformers_modules\THUDM\chatglm-6b-int4\6c5205c47d0d2f7ea2e44715d279e537cae0911f\modeling_chatglm.py", line 1061, in __init__
self.quantize(self.config.quantization_bit, self.config.quantization_embeddings, use_quantization_cache=True, empty_init=True)
File "C:\Users\hei/.cache\huggingface\modules\transformers_modules\THUDM\chatglm-6b-int4\6c5205c47d0d2f7ea2e44715d279e537cae0911f\modeling_chatglm.py", line 1424, in quantize
from .quantization import quantize, QuantizedEmbedding, QuantizedLinear, load_cpu_kernel
File "<frozen importlib._bootstrap>", line 1176, in _find_and_load
File "<frozen importlib._bootstrap>", line 1147, in _find_and_load_unlocked
File "<frozen importlib._bootstrap>", line 690, in _load_unlocked
File "<frozen importlib._bootstrap_external>", line 940, in exec_module
File "<frozen importlib._bootstrap>", line 241, in _call_with_frames_removed
File "C:\Users\hei/.cache\huggingface\modules\transformers_modules\THUDM\chatglm-6b-int4\6c5205c47d0d2f7ea2e44715d279e537cae0911f\quantization.py", line 46, in <module>
logger.warning("Failed to load cpm_kernels:", exception)
Message: 'Failed to load cpm_kernels:'
Arguments: (NameError("name 'CPUKernel' is not defined"),)
No compiled kernel found.
Compiling kernels : C:\Users\hei\.cache\huggingface\modules\transformers_modules\THUDM\chatglm-6b-int4\6c5205c47d0d2f7ea2e44715d279e537cae0911f\quantization_kernels_parallel.c
Compiling gcc -O3 -fPIC -pthread -fopenmp -std=c99 C:\Users\hei\.cache\huggingface\modules\transformers_modules\THUDM\chatglm-6b-int4\6c5205c47d0d2f7ea2e44715d279e537cae0911f\quantization_kernels_parallel.c -shared -o C:\Users\hei\.cache\huggingface\modules\transformers_modules\THUDM\chatglm-6b-int4\6c5205c47d0d2f7ea2e44715d279e537cae0911f\quantization_kernels_parallel.so
Compile default cpu kernel failed, using default cpu kernel code.
Compiling gcc -O3 -fPIC -std=c99 C:\Users\hei\.cache\huggingface\modules\transformers_modules\THUDM\chatglm-6b-int4\6c5205c47d0d2f7ea2e44715d279e537cae0911f\quantization_kernels.c -shared -o C:\Users\hei\.cache\huggingface\modules\transformers_modules\THUDM\chatglm-6b-int4\6c5205c47d0d2f7ea2e44715d279e537cae0911f\quantization_kernels.so
'gcc' �����ڲ����ⲿ���Ҳ���ǿ����еij���
���������ļ���
'gcc' �����ڲ����ⲿ���Ҳ���ǿ����еij���
���������ļ���
Traceback (most recent call last):
File "E:\ChatGLM-6B\cli_demo.py", line 8, in <module>
model = AutoModel.from_pretrained("THUDM/chatglm-6b-int4", trust_remote_code=True).float()
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\anaconda3\envs\chatglm\Lib\site-packages\transformers\models\auto\auto_factory.py", line 466, in from_pretrained
return model_class.from_pretrained(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\anaconda3\envs\chatglm\Lib\site-packages\transformers\modeling_utils.py", line 2498, in from_pretrained
model = cls(config, *model_args, **model_kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\hei/.cache\huggingface\modules\transformers_modules\THUDM\chatglm-6b-int4\6c5205c47d0d2f7ea2e44715d279e537cae0911f\modeling_chatglm.py", line 1061, in __init__
self.quantize(self.config.quantization_bit, self.config.quantization_embeddings, use_quantization_cache=True, empty_init=True)
File "C:\Users\hei/.cache\huggingface\modules\transformers_modules\THUDM\chatglm-6b-int4\6c5205c47d0d2f7ea2e44715d279e537cae0911f\modeling_chatglm.py", line 1439, in quantize
self.transformer = quantize(self.transformer, bits, use_quantization_cache=use_quantization_cache, empty_init=empty_init, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\hei/.cache\huggingface\modules\transformers_modules\THUDM\chatglm-6b-int4\6c5205c47d0d2f7ea2e44715d279e537cae0911f\quantization.py", line 464, in quantize
assert kernels is not None
^^^^^^^^^^^^^^^^^^^
AssertionError
Compile default cpu kernel failed.
Failed to load kernel.
Cannot load cpu or cuda kernel, quantization failed:Environment
`- OS:win11
- Python:3.11.5
- Transformers:4.27.1
- PyTorch:pip3 install torch torchvision torchaudio
- cpu运行`
Is there an existing issue for this?
Current Behavior
D:\Anaconda\envs\langchain\python.exe E:/langchain-ChatGLM-master/cli_demo.py D:\Anaconda\envs\langchain\lib\site-packages\numpy_distributor_init.py:30: UserWarning: loaded more than 1 DLL from .libs: D:\Anaconda\envs\langchain\lib\site-packages\numpy.libs\libopenblas.FB5AE2TYXYH2IJRDKGDGQ3XBKLKTF43H.gfortran-win_amd64.dll D:\Anaconda\envs\langchain\lib\site-packages\numpy.libs\libopenblas64__v0.3.21-gcc_10_3_0.dll warnings.warn("loaded more than 1 DLL from .libs:" INFO 2023-07-04 20:10:16,665-1d: loading model config llm device: cuda embedding device: cuda dir: E:\langchain-ChatGLM-master flagging username: 3f1fea213a014dfeab250a87ae629172
INFO 2023-07-04 20:10:17,661-1d: Note: NumExpr detected 12 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 8. INFO 2023-07-04 20:10:17,661-1d: NumExpr defaulting to 8 threads. D:\Anaconda\envs\langchain\lib\site-packages\pkg_resources__init.py:121: DeprecationWarning: pkg_resources is deprecated as an API warnings.warn("pkg_resources is deprecated as an API", DeprecationWarning) D:\Anaconda\envs\langchain\lib\site-packages\pkg_resources__init__.py:2870: DeprecationWarning: Deprecated call to
pkg_resources.declare_namespace('mpl_toolkits'). Implementing implicit namespace packages (as specified in PEP 420) is preferred topkg_resources.declare_namespace. See https://setuptools.pypa.io/en/latest/references/keywords.html#keyword-namespace-packages declare_namespace(pkg) D:\Anaconda\envs\langchain\lib\site-packages\pkg_resources\init.py:2870: DeprecationWarning: Deprecated call topkg_resources.declare_namespace('google'). Implementing implicit namespace packages (as specified in PEP 420) is preferred topkg_resources.declare_namespace. See https://setuptools.pypa.io/en/latest/references/keywords.html#keyword-namespace-packages declare_namespace(pkg) D:\Anaconda\envs\langchain\lib\site-packages\pkg_resources\init__.py:2870: DeprecationWarning: Deprecated call topkg_resources.declare_namespace('zope'). Implementing implicit namespace packages (as specified in PEP 420) is preferred topkg_resources.declare_namespace. See https://setuptools.pypa.io/en/latest/references/keywords.html#keyword-namespace-packages declare_namespace(pkg)Expected Behavior
No response
Steps To Reproduce
Loading E:\langchain-ChatGLM-master\GLM-int4... num_gpus 1 No compiled kernel found. Compiling kernels : C:\Users\11025.cache\huggingface\modules\transformers_modules\GLM-int4\quantization_kernels_parallel.c Compiling gcc -O3 -fPIC -pthread -fopenmp -std=c99 C:\Users\11025.cache\huggingface\modules\transformers_modules\GLM-int4\quantization_kernels_parallel.c -shared -o C:\Users\11025.cache\huggingface\modules\transformers_modules\GLM-int4\quantization_kernels_parallel.so --- Logging error --- Traceback (most recent call last): File "C:\Users\11025/.cache\huggingface\modules\transformers_modules\GLM-int4\quantization.py", line 19, in
from cpm_kernels.kernels.base import LazyKernelCModule, KernelFunction, round_up
File "D:\Anaconda\envs\langchain\lib\site-packages\cpm_kernels__init.py", line 1, in
from . import library
File "D:\Anaconda\envs\langchain\lib\site-packages\cpm_kernels\library__init.py", line 1, in
from . import nvrtc
File "D:\Anaconda\envs\langchain\lib\site-packages\cpm_kernels\library\nvrtc.py", line 5, in
nvrtc = Lib("nvrtc")
File "D:\Anaconda\envs\langchain\lib\site-packages\cpm_kernels\library\base.py", line 45, in init__
lib_path = windows_find_lib(self. name)
File "D:\Anaconda\envs\langchain\lib\site-packages\cpm_kernels\library\base.py", line 39, in windows_find_lib
return lookup_dll(lib_name)
File "D:\Anaconda\envs\langchain\lib\site-packages\cpm_kernels\library\base.py", line 16, in lookup_dll
for name in os.listdir(path):
NotADirectoryError: [WinError 267] 目录名称无效。: 'D:\tomcat8\lib\servlet-api.jar'
During handling of the above exception, another exception occurred:
Traceback (most recent call last): File "D:\Anaconda\envs\langchain\lib\logging__init.py", line 1085, in emit msg = self.format(record) File "D:\Anaconda\envs\langchain\lib\logging__init.py", line 929, in format return fmt.format(record) File "D:\Anaconda\envs\langchain\lib\logging\init.py", line 668, in format record.message = record.getMessage() File "D:\Anaconda\envs\langchain\lib\logging\init.py", line 373, in getMessage msg = msg % self.args TypeError: not all arguments converted during string formatting Call stack: File "E:/langchain-ChatGLM-master/cli_demo.py", line 66, in
main()
File "E:/langchain-ChatGLM-master/cli_demo.py", line 16, in main
llm_model_ins = shared.loaderLLM()
File "E:\langchain-ChatGLM-master\models\shared.py", line 40, in loaderLLM
loaderCheckPoint.reload_model()
File "E:\langchain-ChatGLM-master\models\loader\loader.py", line 431, in reload_model
self.model, self.tokenizer = self._load_model(self.model_name)
File "E:\langchain-ChatGLM-master\models\loader\loader.py", line 123, in _load_model
LoaderClass.from_pretrained(checkpoint,
File "D:\Anaconda\envs\langchain\lib\site-packages\transformers\models\auto\auto_factory.py", line 462, in from_pretrained
return model_class.from_pretrained(
File "D:\Anaconda\envs\langchain\lib\site-packages\transformers\modeling_utils.py", line 2611, in from_pretrained
model = cls(config, *model_args, **model_kwargs)
File "C:\Users\11025/.cache\huggingface\modules\transformers_modules\GLM-int4\modeling_chatglm.py", line 1061, in init__
self.quantize(self.config.quantization_bit, self.config.quantization_embeddings, use_quantization_cache=True, empty_init=True)
File "C:\Users\11025/.cache\huggingface\modules\transformers_modules\GLM-int4\modeling_chatglm.py", line 1424, in quantize
from .quantization import quantize, QuantizedEmbedding, QuantizedLinear, load_cpu_kernel
File "", line 991, in _find_and_load
File "", line 975, in _find_and_load_unlocked
File "", line 671, in _load_unlocked
File "", line 843, in exec_module
File "", line 219, in _call_with_frames_removed
File "C:\Users\11025/.cache\huggingface\modules\transformers_modules\GLM-int4\quantization.py", line 46, in
logger.warning("Failed to load cpm_kernels:", exception)
Message: 'Failed to load cpm_kernels:'
Arguments: (NotADirectoryError(20, '目录名称无效。'),)
Compile default cpu kernel failed, using default cpu kernel code.
Compiling gcc -O3 -fPIC -std=c99 C:\Users\11025.cache\huggingface\modules\transformers_modules\GLM-int4\quantization_kernels.c -shared -o C:\Users\11025.cache\huggingface\modules\transformers_modules\GLM-int4\quantization_kernels.so
Compile default cpu kernel failed.
Failed to load kernel.
Cannot load cpu or cuda kernel, quantization failed:
'gcc' �����ڲ����ⲿ���Ҳ���ǿ����еij���
���������ļ���
'gcc' �����ڲ����ⲿ���Ҳ���ǿ����еij���
���������ļ���
┌───────────────────── Traceback (most recent call last) ─────────────────────┐
│ E:/langchain-ChatGLM-master/cli_demo.py:66 in │
│ │
│ 63 │ args = parser.parse_args() │
│ 64 │ args_dict = vars(args) │
│ 65 │ shared.loaderCheckPoint = LoaderCheckPoint(args_dict) │
│ > 66 │ main() │
│ 67 │
│ │
│ E:/langchain-ChatGLM-master/cli_demo.py:16 in main │
│ │
│ 13 │
│ 14 def main(): │
│ 15 │ │
│ > 16 │ llm_model_ins = shared.loaderLLM() │
│ 17 │ llm_model_ins.history_len = LLM_HISTORY_LEN │
│ 18 │ │
│ 19 │ local_doc_qa = LocalDocQA() │
│ │
│ E:\langchain-ChatGLM-master\models\shared.py:40 in loaderLLM │
│ │
│ 37 │ if 'FastChatOpenAILLM' in llm_model_info["provides"]: │
│ 38 │ │ loaderCheckPoint.unload_model() │
│ 39 │ else: │
│ > 40 │ │ loaderCheckPoint.reload_model() │
│ 41 │ │
│ 42 │ provides_class = getattr(sys.modules['models'], llm_model_info['pr │
│ 43 │ modelInsLLM = provides_class(checkPoint=loaderCheckPoint) │
│ │
│ E:\langchain-ChatGLM-master\models\loader\loader.py:431 in reload_model │
│ │
│ 428 │ │ │ except Exception as e: │
│ 429 │ │ │ │ print("加载PrefixEncoder config.json失败") │
│ 430 │ │ │
│ > 431 │ │ self.model, self.tokenizer = self._load_model(self.model_name │
│ 432 │ │ │
│ 433 │ │ if self.lora: │
│ 434 │ │ │ self._add_lora_to_model([self.lora]) │
│ │
│ E:\langchain-ChatGLM-master\models\loader\loader.py:123 in _load_model │
│ │
│ 120 │ │ │ │ print("num_gpus",num_gpus) │
│ 121 │ │ │ │ if num_gpus < 2 and self.device_map is None: │
│ 122 │ │ │ │ │ model = ( │
│ > 123 │ │ │ │ │ │ LoaderClass.from_pretrained(checkpoint, │
│ 124 │ │ │ │ │ │ │ │ │ │ │ │ │ config=self.model │
│ 125 │ │ │ │ │ │ │ │ │ │ │ │ │ torch_dtype=torch │
│ 126 │ │ │ │ │ │ │ │ │ │ │ │ │ trust_remotecode │
│ │
│ D:\Anaconda\envs\langchain\lib\site-packages\transformers\models\auto\auto │
│ factory.py:462 in from_pretrained │
│ │
│ 459 │ │ │ model_class = get_class_from_dynamic_module( │
│ 460 │ │ │ │ class_ref, pretrained_model_name_or_path, hub_kwarg │
│ 461 │ │ │ ) │
│ > 462 │ │ │ return model_class.from_pretrained( │
│ 463 │ │ │ │ pretrained_model_name_or_path, model_args, config=co │
│ 464 │ │ │ ) │
│ 465 │ │ elif type(config) in cls._model_mapping.keys(): │
│ │
│ D:\Anaconda\envs\langchain\lib\site-packages\transformers\modeling_utils.py │
│ :2611 in from_pretrained │
│ │
│ 2608 │ │ │ init_contexts.append(init_empty_weights()) │
│ 2609 │ │ │
│ 2610 │ │ with ContextManagers(init_contexts): │
│ > 2611 │ │ │ model = cls(config, model_args, model_kwargs) │
│ 2612 │ │ │
│ 2613 │ │ # Check first if we are init__ │
│ │
│ 1058 │ │ self.quantized = False │
│ 1059 │ │ │
│ 1060 │ │ if self.config.quantization_bit: │
│ > 1061 │ │ │ self.quantize(self.config.quantization_bit, self.config. │
│ 1062 │ │
│ 1063 │ def get_output_embeddings(self): │
│ 1064 │ │ return self.lm_head │
│ │
│ C:\Users\11025/.cache\huggingface\modules\transformers_modules\GLM-int4\mod │
│ eling_chatglm.py:1439 in quantize │
│ │
│ 1436 │ │ self.config.quantization_bit = bits │
│ 1437 │ │ self.config.quantization_embeddings = quantize_embeddings │
│ 1438 │ │ │
│ > 1439 │ │ self.transformer = quantize(self.transformer, bits, use_quan │
│ 1440 │ │ │
│ 1441 │ │ if self.device == torch.device("cpu"): │
│ 1442 │ │ │ dtype = torch.float32 │
│ │
│ C:\Users\11025/.cache\huggingface\modules\transformers_modules\GLM-int4\qua │
│ ntization.py:464 in quantize │
│ │
│ 461 │ if not cpu_kernels.load: │
│ 462 │ │ if kernels is None: # CUDA kernels failed │
│ 463 │ │ │ print("Cannot load cpu or cuda kernel, quantization faile │
│ > 464 │ │ │ assert kernels is not None │
│ 465 │ │ print("Cannot load cpu kernel, don't use quantized model on c │
│ 466 │ │
│ 467 │ current_device = model.device │
└─────────────────────────────────────────────────────────────────────────────┘
AssertionError
from_pt│ │ 2614 │ │ if use_keep_in_fp32_modules: │ │ │ │ C:\Users\11025/.cache\huggingface\modules\transformers_modules\GLM-int4\mod │ │ eling_chatglm.py:1061 inProcess finished with exit code 1
Environment
Anything else?
No response