CombinedTimestepGuidanceTextProjEmbeddings.forward() missing 1 required positional argument: 'pooled_projection' #9370
Description
Describe the bug
Hello, I am using the latest version of transformer to train Flux's Lora slider. I used 'pooled_projection' in the middle, but the following error will be reported during runtime:
File "/opt/conda/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1511, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
File "/opt/conda/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1520, in _call_impl
return forward_call(*args, **kwargs)
File "/opt/conda/lib/python3.10/site-packages/diffusers/models/transformers/transformer_flux.py", line 450, in forward
self.time_text_embed(timestep, pooled_projections)
File "/opt/conda/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1511, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
File "/opt/conda/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1520, in _call_impl
return forward_call(*args, **kwargs)
TypeError: CombinedTimestepGuidanceTextProjEmbeddings.forward() missing 1 required positional argument: 'pooled_projection'
Strangely, I put the 450th line of code
self.time_text_embed(timestep, After changing "self.time_text_embed(timestep, timestep, pooled_projections)" to "self.time_text_embed(timestep, timestep, pooled_projections)", the training code can run normally. However, the trained Lora has no effect, and the generated image is exactly the same as the original image. I have tried many methods but failed to solve this problem. I hope to get an answer! Thank you!
Reproduction
python /tmp/code/sliders/conceptmod/textsliders/train_lora_flux.py --name 'light' --rank 16 --alpha 1 --config_file '/tmp/code/sliders/conceptmod/textsliders/data/config.yaml' --peft_type lora
Logs
python /tmp/code/sliders/conceptmod/textsliders/train_lora_flux.py --name 'light' --rank 16 --alpha 1 --config_file '/tmp/code/sliders/conceptmod/textsliders/data/config.yaml' --peft_type lora
File "/opt/conda/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1511, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
File "/opt/conda/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1520, in _call_impl
return forward_call(*args, **kwargs)
File "/opt/conda/lib/python3.10/site-packages/diffusers/models/transformers/transformer_flux.py", line 450, in forward
self.time_text_embed(timestep, pooled_projections)
File "/opt/conda/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1511, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
File "/opt/conda/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1520, in _call_impl
return forward_call(*args, **kwargs)
TypeError: CombinedTimestepGuidanceTextProjEmbeddings.forward() missing 1 required positional argument: 'pooled_projection'System Info
- 🤗 Diffusers version: 0.31.0.dev0
- Platform: Linux-4.15.0-189-generic-x86_64-with-glibc2.35
- Running on Google Colab?: No
- Python version: 3.10.13
- PyTorch version (GPU?): 2.2.1+cu121 (True)
- Flax version (CPU?/GPU?/TPU?): not installed (NA)
- Jax version: not installed
- JaxLib version: not installed
- Huggingface_hub version: 0.24.6
- Transformers version: 4.44.1
- Accelerate version: 0.26.1
- PEFT version: 0.7.1
- Bitsandbytes version: 0.42.0
- Safetensors version: 0.4.1
- xFormers version: 0.0.25
- Accelerator: NVIDIA A100-PCIE-40GB, 40960 MiB
- Using GPU in script?:
- Using distributed or parallel set-up in script?:
Who can help?
No response