ImportError: cannot import name 'Tensor' from 'torch' (unknown location) about postgresml

[daimenhutchison]

I got pass this error by;

pip3 uninstall torch
pip3 install torch==2.0.1

Somehow I ended up with the wrong version. Looks like I'm in the same boat with CUDA;

org.jkiss.dbeaver.model.sql.DBSQLException: SQL Error [XX000]: ERROR: Traceback (most recent call last):
  File "transformers.py", line 10, in <module>
  File "/usr/local/lib/python3.9/site-packages/InstructorEmbedding/__init__.py", line 1, in <module>
    from .instructor import *
  File "/usr/local/lib/python3.9/site-packages/InstructorEmbedding/instructor.py", line 2, in <module>
    import torch
  File "/usr/local/lib64/python3.9/site-packages/torch/__init__.py", line 228, in <module>
    _load_global_deps()
  File "/usr/local/lib64/python3.9/site-packages/torch/__init__.py", line 190, in _load_global_deps
    ctypes.CDLL(lib_path, mode=ctypes.RTLD_GLOBAL)
  File "/usr/lib64/python3.9/ctypes/__init__.py", line 374, in __init__
    self._handle = _dlopen(self._name, mode)
 OSError: libcufft.so.10: cannot open shared object file: No such file or directory

I was careful to follow the dependencies in the various requirements files, guess I've messed that up somewhere.

[montanalow]

Did you install the platform specific dependencies, or the generic requirements.txt. Complex ML dependency graphs frequently break, so we provide a complete/working frozen set per platform.

[daimenhutchison]

do you mean the requirements.linux.txt? If so, yes.