I have an issue while running:
import torch
Below is my Pipfile
:
[[source]]
url = "https://pypi.org/simple"
verify_ssl = true
name = "pypi"
[packages]
streamlit-searchbox = "*"
pytesseract = "*"
pdf2image = "*"
pipenv = "*"
python-dotenv = "*"
openpyxl = "*"
sentence-transformers = "*"
torch = {version = "==2.3.1", index = "https://download.pytorch.org/whl/cpu"}
[dev-packages]
ruff = "*"
pytest = "*"
[requires]
python_version = "3.11"
This is the link to the app:
The error logs are below:
/home/adminuser/venv/lib/python3.11/site-packages/torch/__init__.py:238 in
<module>
235 │ #
236 │ # See Note [Global dependencies]
237 │ if USE_GLOBAL_DEPS:
❱ 238 │ │ _load_global_deps()
239 │ from torch._C import * # noqa: F403
240
241 # Appease the type checker; ordinarily this binding is inserted by th
/home/adminuser/venv/lib/python3.11/site-packages/torch/__init__.py:199 in
_load_global_deps
196 │ │ if not is_cuda_lib_err:
197 │ │ │ raise err
198 │ │ for lib_folder, lib_name in cuda_libs.items():
❱ 199 │ │ │ _preload_cuda_deps(lib_folder, lib_name)
200 │ │ ctypes.CDLL(lib_path, mode=ctypes.RTLD_GLOBAL)
201
202
/home/adminuser/venv/lib/python3.11/site-packages/torch/__init__.py:164 in
_preload_cuda_deps
161 │ │ if lib_path:
162 │ │ │ break
163 │ if not lib_path:
❱ 164 │ │ raise ValueError(f"{lib_name} not found in the system path {s
165 │ ctypes.CDLL(lib_path)
166
167
────────────────────────────────────────────────────────────────────────────────
ValueError: libcublas.so.*[0-9] not found in the system path
['src/sonar/app.py', 'src/sonar', '/home/adminuser/venv/bin',
'/usr/local/lib/python311.zip', '/usr/local/lib/python3.11',
'/usr/local/lib/python3.11/lib-dynload',
'/home/adminuser/venv/lib/python3.11/site-packages']
2024-07-01 17:37:02.682 503 GET /script-health-check (127.0.0.1) 108.03ms