Load check point with Diffusers library.

#27
by dkackman - opened

Hi,

I'm trying to load one of the checkpoints with 0.14.0 version of the Diffusers library (safetensors also installed) and am not able to get it working. The code looks like this:

    pipeline = pipeline_type.from_pretrained(
        "Lykon/DreamShaper",
        revision="main",
        torch_dtype=torch.float16,
    )

    pipeline.unet.load_attn_procs("Lykon/DreamShaper", revision="main", weight_name="DreamShaper_4BakedVae_fp16.safetensors", use_safetensors=True)

It downloads the file but then errors with invalid load key, '\x9f'. which kind of implies that diffusers thinks that it is a pickle file.

I am not having much luck fining working examples of load_attn_procs that work with anything other than a pytorch.bin file so maybe it just doesn't work.

If you've gotten this to work would love to know it!

Don

have you solved it? I'd like to help you, but I don't use that. The model works on auto1111, so it should be fine

I figured out how to convert the various checkpoints into individual models. Thanks for following up!

dkackman changed discussion status to closed

no problem. Sorry it took me some time. Feel free to ask any time you need help

Hey @dkackman Can you share how to use it with diffusers?

Sign up or log in to comment