Update README.md - fix class name
#3
by
matospiso
- opened
I'm getting ImportError using FluxControlPipeline, but FluxControlNetPipeline seems to work.
Reference: https://huggingface.co/docs/diffusers/main/en/api/pipelines/controlnet_flux#diffusers.FluxControlNetPipeline
matospiso
changed pull request title from
Update README.md
to Update README.md - fix class name
however, it then also fails with the following message:
The config attributes {'invert_sigmas': False} were passed to FlowMatchEulerDiscreteScheduler, but are not expected and will be ignored. Please verify your scheduler_config.json configuration file.
The config attributes {'out_channels': 64} were passed to FluxTransformer2DModel, but are not expected and will be ignored. Please verify your config.json configuration file.
---------------------------------------------------------------------------
ValueError Traceback (most recent call last)
<ipython-input-3-a3e95a9cf1af> in <cell line: 1>()
----> 1 pipe = FluxControlNetPipeline.from_pretrained("black-forest-labs/FLUX.1-Canny-dev", torch_dtype=torch.bfloat16).to("cuda")
7 frames
/usr/local/lib/python3.10/dist-packages/accelerate/utils/modeling.py in set_module_tensor_to_device(module, tensor_name, device, value, dtype, fp16_statistics, tied_params_map)
284 # In other cases, we want to make sure we're not loading checkpoints that do not match the config.
285 if old_value.shape != value.shape and param_cls.__name__ != "Params4bit":
--> 286 raise ValueError(
287 f'Trying to set a tensor of shape {value.shape} in "{tensor_name}" (which has shape {old_value.shape}), this looks incorrect.'
288 )
ValueError: Trying to set a tensor of shape torch.Size([64]) in "bias" (which has shape torch.Size([128])), this looks incorrect.