YuxinWenRick / tree-ring-watermark

MIT License
237 stars 30 forks source link

the model location #24

Closed jinganglang567 closed 6 months ago

jinganglang567 commented 6 months ago

sorry to bother you again,doctor . i run the script,However, when the program finishes running and I don't see the saved model file, am I missing something?

YuxinWenRick commented 6 months ago

Hi, do you mean the stable diffusion model weights? It should be saved in the default huggingface cache directory. You can check it through echo $HF_HOME with the command line or import os; print(os.environ['HF_HOME']) in python. The model weights should be under the directory with a name hub or something.

jinganglang567 commented 6 months ago

@YuxinWenRick , sorry to bother you ,doctor.It should be that I didn't express it clearly, I mean, shouldn't 256diffusion.pt be the trained model, why when I run the command, no similar model is saved to the local area.

jinganglang567 commented 6 months ago

@YuxinWenRick doctor, sorry to bother you again, I want to convert the Fourier transform to the wavelet transform and deal with the noise, what should I do?

YuxinWenRick commented 6 months ago

Oh, you meant the ImageNet model? You may download the model manually through this link.

I don't know what's the best way to do wavelet transform in python, but I found this library. Hope it's useful for you.

jinganglang567 commented 6 months ago

@YuxinWenRick thank you doctor, i got it.