How to do inference?

#1
by Actis92 - opened

Trying the README i get this error loading pytorch_lora_weights.safetensors:
wrapper, _ = create_lycoris_from_weights(lora_scale, adapter_id, pipe.tran
sformer)
Traceback (most recent call last):
File "", line 1, in
File "/home/wizard/mambaforge/lib/python3.10/site-packages/lycoris/wrapper.py", line 172, in create_lycoris_from_weights
module = make_module(lyco_type, params, lora_name, orig_modules)
File "/home/wizard/mambaforge/lib/python3.10/site-packages/torch/utils/_contextlib.py", line 116, in decorate_context
return func(*args, **kwargs)
File "/home/wizard/mambaforge/lib/python3.10/site-packages/lycoris/modules/init.py", line 46, in make_module
w1, w1a, w1b, w2, w2a, w2b, _, t2, alpha = params
ValueError: too many values to unpack (expected 9)

Actis92 changed discussion status to closed

You need to use the pipeline from SimpleTuner with attention masking on.

https://github.com/bghira/SimpleTuner/tree/main/helpers/models/flux

Sign up or log in to comment