Troubleshooting

From Upscale Wiki
This is the latest revision of this page; it has no approved revision.
Revision as of 23:10, 19 December 2020 by Alsa (talk | contribs) (Update PyTorch version in troubleshooting message // Edit via Wikitext Extension for VSCode)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to navigation Jump to search

NumPy

Windows 10 20H2

If you get this error:

 ** On entry to DGEBAL parameter number  3 had an illegal value
 ** On entry to DGEHRD  parameter number  2 had an illegal value
 ** On entry to DORGHR DORGQR parameter number  2 had an illegal value
 ** On entry to DHSEQR parameter number  4 had an illegal value
Traceback (most recent call last):
  File "C:\dev\lang\Python38\lib\site-packages\numpy\__init__.py", line 305, in <module>
    _win_os_check()
  File "C:\dev\lang\Python38\lib\site-packages\numpy\__init__.py", line 302, in _win_os_check
    raise RuntimeError(msg.format(__file__)) from None
RuntimeError: The current Numpy installation ('C:\\dev\\lang\\Python38\\lib\\site-packages\\numpy\\__init__.py') fails to pass a sanity check due to a bug in the windows runtime. See this issue for more information: https://tinyurl.com/y3dm3h86
Traceback (most recent call last):
  File "upscale.py", line 8, in <module>
    import cv2
  File "C:\dev\lang\Python38\lib\site-packages\cv2\__init__.py", line 5, in <module>
    from .cv2 import *
ImportError: numpy.core.multiarray failed to import

You can solve this by downgrading to NumPy 1.19.3

pip install numpy==1.19.3

This is an issue that Microsoft needs to fix, a fix is expected to come out in January of 2021. Until then, you should use the old version on Windows.

PyTorch

Wrong Version of Python, wrong CUDA version or wrong OS

If you get this error:

Looking in links: https://download.pytorch.org/whl/torch_stable.html
ERROR: Could not find a version that satisfies the requirement torch===1.7.1 (from versions: 0.1.2, 0.1.2.post1, 0.1.2.post2)
ERROR: No matching distribution found for torch===1.7.1

Chances are you have to downgrade your Python version. (Or upgrade it if your Python is still version 2). Check the installation guide for the currently tested version of Python, PyTorch and CUDA, uninstall your old versions and reinstall the ones we verified to work.

You can also try to get a newer configuration here: https://pytorch.org/get-started/locally/