the-database

The first two issues should be fixed in v0.0.8. Please retry your last example on v0.0.8 to see if the issue still happens.

These issues should all be fixed in v1.0.0, please open a new issue if you see any similar errors in v1.0.0.

Follow these instructions: https://github.com/the-database/mpv-upscale-2x_animejanai/wiki/Setup-for-AMD-or-Intel-Arc-users Let me know if you run into any issues.

> https://github.com/chaiNNer-org/chaiNNer/blob/main/backend/src/packages/chaiNNer_pytorch/pytorch/io/load_model.py#L18 [chaiNNer-org/spandrel@c3aa562#diff-b335630551682c19a781afebcf4d07bf978fb1f8ac04c6bf87428ed5106870f5R63](https://github.com/chaiNNer-org/spandrel/commit/c3aa562d05388bc654cf6cc7b082eed3ea318ccf#diff-b335630551682c19a781afebcf4d07bf978fb1f8ac04c6bf87428ed5106870f5R63) As far as I remember, this change in the extra_arches connection interface breaks the load_model node for some people Do you know under what conditions `load_model` is...

Thanks for the feedback. onnx upscale models are typically only a single scale, and VideoJaNai runs the model at that scale. The "upscale2x" function name is just an old function...

Could you please retry with v1.1.1 (https://github.com/the-database/VideoJaNai/releases/tag/1.1.1) and see if you have the same issue? If so, could you post the error message?

This looks related to the use of `y4m` as a container, which is currently hard-coded here. Will look into it. https://github.com/the-database/VideoJaNai/blob/7e4609d635610aefdc2247a9e6e88a56a9ea5f77/VideoJaNai/ViewModels/MainWindowViewModel.cs#L669

Thanks for the info, I didn't know about `autorot()`. Unfortunately it doesn't seem to affect the result of my test. Also, I just tried the same test on a jpg...

Hi, what GPU do you have?

Yes, see this project https://github.com/the-database/mpv-upscale-2x_animejanai