ctransformers: OSError No such file or directory issue
#4
by
lazyDataScientist
- opened
After running this code I am receiving this error. Note: i am running this in a CoLab environment.
from ctransformers import AutoModelForCausalLM
# Set gpu_layers to the number of layers to offload to GPU. Set to 0 if no GPU acceleration is available on your system.
llm = AutoModelForCausalLM.from_pretrained("TheBloke/Mistral-7B-Instruct-v0.1-GGUF", model_file="mistral-7b-instruct-v0.1.Q2_K.gguf", model_type="mistral", gpu_layers=50)
---------------------------------------------------------------------------
OSError Traceback (most recent call last)
<ipython-input-7-a38c39cda463> in <cell line: 4>()
2
3 # Set gpu_layers to the number of layers to offload to GPU. Set to 0 if no GPU acceleration is available on your system.
----> 4 llm = AutoModelForCausalLM.from_pretrained("TheBloke/Mistral-7B-Instruct-v0.1-GGUF", model_file="mistral-7b-instruct-v0.1.Q2_K.gguf", model_type="mistral", gpu_layers=50)
3 frames
/usr/local/lib/python3.10/dist-packages/ctransformers/hub.py in from_pretrained(cls, model_path_or_repo_id, model_type, model_file, config, lib, local_files_only, revision, hf, **kwargs)
173 )
174
--> 175 llm = LLM(
176 model_path=model_path,
177 model_type=model_type,
/usr/local/lib/python3.10/dist-packages/ctransformers/llm.py in __init__(self, model_path, model_type, config, lib)
244 model_type = "gguf"
245
--> 246 self._lib = load_library(lib, gpu=config.gpu_layers > 0)
247 self._llm = self._lib.ctransformers_llm_create(
248 model_path.encode(),
/usr/local/lib/python3.10/dist-packages/ctransformers/llm.py in load_library(path, gpu)
124 if "cuda" in path:
125 load_cuda()
--> 126 lib = CDLL(path)
127
128 lib.ctransformers_llm_create.argtypes = [
/usr/lib/python3.10/ctypes/__init__.py in __init__(self, name, mode, handle, use_errno, use_last_error, winmode)
372
373 if handle is None:
--> 374 self._handle = _dlopen(self._name, mode)
375 else:
376 self._handle = handle
OSError: libcudart.so.12: cannot open shared object file: No such file or directory
I'm getting a different error while running ctransformers in my laptop: "Model type 'mistral' is not supported."
I've used falcon 7b and wizardlm 7b in the past with the exact same setup. Not sure whether ctransformers is actually compatible with this model or not :s
I have a similar issue.
RuntimeError: Failed to create LLM 'mistral' from 'Models\mistral-7b-v0.1.Q4_K_M.gguf'
After updating CTransformers, the issue was solved on my side (apologies for delay on reply).