Open Reannvin opened 3 months ago
@Reannvin I'm also testing the code out and got that error but I've managed to fix it by using
pip install transformers==4.29.2
There is no ControlNet project under the model, and the transformers version (4.36.2) in your environment conflicts with the 4.24.0 version in ControlNet.
After I manually added the ControlNet project under the model, the following error occurred.
Traceback (most recent call last): File "/data/user/code/Ultraman/scripts/generate_texture.py", line 40, in <module> from lib.diffusion_helper import ( File "/data/user/code/Ultraman/./lib/diffusion_helper.py", line 26, in <module> from models.ControlNet.gradio_depth2image import init_model, process File "/data/user/code/Ultraman/./models/ControlNet/gradio_depth2image.py", line 29, in <module> model.load_state_dict(load_state_dict('./models/ControlNet/models/control_sd15_depth.pth', location='cuda')) File "/data/user/.conda/envs/ultraman/lib/python3.9/site-packages/torch/nn/modules/module.py", line 1604, in load_state_dict raise RuntimeError('Error(s) in loading state_dict for {}:\n\t{}'.format( RuntimeError: Error(s) in loading state_dict for ControlLDM: Unexpected key(s) in state_dict: "cond_stage_model.transformer.text_model.embeddings.position_ids".
I found the solution under ControlNet's issue page but does't work for me.
Thank you for supporting our work and asking valuable questions. We are recently optimizing our algorithm to achieve better results. We will update a version of the code when it is done. We will pay attention to the issues you raise.
There is no ControlNet project under the model, and the transformers version (4.36.2) in your environment conflicts with the 4.24.0 version in ControlNet.
After I manually added the ControlNet project under the model, the following error occurred.
Traceback (most recent call last): File "/data/user/code/Ultraman/scripts/generate_texture.py", line 40, in <module> from lib.diffusion_helper import ( File "/data/user/code/Ultraman/./lib/diffusion_helper.py", line 26, in <module> from models.ControlNet.gradio_depth2image import init_model, process File "/data/user/code/Ultraman/./models/ControlNet/gradio_depth2image.py", line 29, in <module> model.load_state_dict(load_state_dict('./models/ControlNet/models/control_sd15_depth.pth', location='cuda')) File "/data/user/.conda/envs/ultraman/lib/python3.9/site-packages/torch/nn/modules/module.py", line 1604, in load_state_dict raise RuntimeError('Error(s) in loading state_dict for {}:\n\t{}'.format( RuntimeError: Error(s) in loading state_dict for ControlLDM: Unexpected key(s) in state_dict: "cond_stage_model.transformer.text_model.embeddings.position_ids".
I found the solution under ControlNet's issue page but does't work for me.https://github.com/lllyasviel/ControlNet/issues/675