lllyasviel / style2paints

sketch + style = paints :art: (TOG2018/SIGGRAPH2018ASIA)

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Unable to run style2paints v4

My12123 opened this issue · comments

Unable to run style2paints v4
ERROR: Could not find a version that satisfies the requirement llvmlite==0.36.0rc1 (from versions: 0.2.0, 0.2.1, 0.2.2, 0.4.0, 0.5.0, 0.6.0, 0.7.0, 0.8.0, 0.9.0, 0.10.0, 0.11.0, 0.12.0.1, 0.12.1, 0.13.0, 0.14.0, 0.15.0, 0.16.0, 0.17.0, 0.17.1, 0.18.0, 0.19.0, 0.20.0, 0.21.0, 0.22.0, 0.23.0, 0.23.2, 0.24.0, 0.25.0, 0.26.0, 0.27.0, 0.27.1, 0.28.0, 0.29.0, 0.30.0, 0.31.0, 0.32.0, 0.32.1, 0.33.0, 0.34.0, 0.35.0, 0.36.0)
ERROR: No matching distribution found for llvmlite==0.36.0rc
2022-07-06_12-55-04
How do I launch style2paints v4?

https://github.com/78Alpha/style2paints-Backup/tree/master/V4

Try this out (the instructions/requirements) or the EXE version, see if that works

The EXE file fails, closes. I have the NVIDIA GeForce GTX 1050 video card
CUDA_ERROR_OUT_OF_MEMORY

Yeah, V4 fails on anything that has less than 6 GB VRAM. It would be an XX60 and above. V4.5 seemed to be the one that works on everything, V4 is a resource hog.

Can we fix this?
I have six GB of memory.
2 GB allocated and 4 GB total.
Can you make a version for Google Colab?

Can we fix this?
I have six GB of memory.
2 GB allocated and 4 GB total.
Can you make a version for Google Colab?

Well, it could be possible, since the 4.5 has done it. It would just be figuring it out. I myself am unfamiliar with machine learning tools, so I'm of no help with correcting the issue.

@78Alpha thanks for your help.

@lllyasviel with the Google Colab version please help.
Can I have a link to style2paints v4 for Google Colab?

The versioning of tensorflow is very confusing. keras version control has gotten even worse since moving out of tensorflow last year (https://twitter.com/fchollet/status/1404967230048149506), and code that works this week may not work again in ten days.

I will look at it next when I'm free.

My group has fully moved to pytorch. The next version of s2p will be a complete departure from tensorflow and provide general operator-based inference model, perhaps in onnx or torchscript, or python code of pytorch.

I believe that needs h5py 2.10.0