toriato

[Tensorflow maps all VRAMs at startup by default.](https://www.tensorflow.org/guide/gpu#limiting_gpu_memory_growth) This could be the cause of the problem. https://github.com/toriato/stable-diffusion-webui-wd14-tagger/blob/aa51c991b9bdf4d1a956f195aa642194222dc424/tagger/interrogator.py#L17-L21

```python import os from modules import scripts from modules.sd_samplers import KDiffusionSampler, sample_to_image orig_callback_state = KDiffusionSampler.callback_state def callback_state(self: KDiffusionSampler, d: dict) -> None: ''' callback from k-diffusion sampling methods https://github.com/crowsonkb/k-diffusion/blob/60e5042ca0da89c14d1dd59d73883280f8fce991/k_diffusion/sampling.py#L67 '''...

There may be a problem caused by the architecture of the C toolchain you are using being different from the `GOARCH` environment variable. If you are using vscode, you can...

That's a great idea, I'll make it when I have time!

I unable to replicate the issue, It seems like a bug related to pydantic. Just in case, did you do PIP upgrade after reinstalling? `pip install --upgrade -r requirements.txt`