lukemelas / realfusion

Hi @lukemelas, thanks for releasing your great work!

Could you please release the version of the packages you are using as well (e.g. direct export of your python environment)? I am trying out your code but there are some random issues. For example, stable_diffusion_model.text_encoder now gives a tuple of strings instead of the clip text model (looks like a version issue).

Having same issue. Also I had to install more packages not listed in requirements.txt like trimesh, PyMCubes.

I encountered the same issue. When the code runs to sd/utils.py line 15, it raises AttributeError: 'tuple' object has no attribute 'get_input_embeddings'. When viewing the content of text_encoder, I observe a tuple of strings: ('transformers', 'CLIPTextModel'). @lukemelas

I encountered the same issue. When the code runs to sd/utils.py line 15, it raises AttributeError: 'tuple' object has no attribute 'get_input_embeddings'. When viewing the content of text_encoder, I observe a tuple of strings: ('transformers', 'CLIPTextModel'). @lukemelas

I also got the same issue.

Also have the same issue

I encountered the same issue. When the code runs to sd/utils.py line 15, it raises AttributeError: 'tuple' object has no attribute 'get_input_embeddings'. When viewing the content of text_encoder, I observe a tuple of strings: ('transformers', 'CLIPTextModel'). @lukemelas

I have the same issue. @lukemelas

Could anyone have working version that they could share the requirements.txt?

I also had the same issues with having to manually install a couple of packages (trimesh and PyMCubes) but the latest versions for all packages worked for me...except for one, which leads to the attribute error @WenjiaR was talking about.

The library diffusers seems to deprecate some functionality if you install its latest version (0.15.1). Version 0.12.0 made running main.py fine for me. Run the below command and then run main.py again:

pip install diffusers==0.12.0

I'm not sure what's causing the issue between the different versions, and I'm also not sure if this version of diffusers is the one @lukemelas used.

diffusers==0.15.0 seems working. You cannot use diffusers==0.15.1 because of some deprecation problem. Check below PR from diffusers repo.

huggingface/diffusers#3129

As @ManuelSerna suggested, I tried diffusers==0.12.0 and it worked for main.py. But I couldn't run textual-inversion/textual_inversion.py.

realfusion/textual-inversion/textual_inversion.py

Lines 83 to 89 in 45c55eb

    
           # Will error if the minimal version of diffusers is not installed. Remove at your own risks. 
        
           check_min_version("0.15.0.dev0") 
        
           logger = get_logger(__name__)


	# Will error if the minimal version of diffusers is not installed. Remove at your own risks.
	check_min_version("0.15.0.dev0")

	logger = get_logger(__name__)

Package version