search

smthemex / ComfyUI_StoryDiffusion

You can using StoryDiffusion in ComfyUI
Apache License 2.0
156 stars 190 forks source link

在“laion/CLIP-ViT-bigG-14-laion2B-39B-b160k” 一栏里填写你的本地clip模型的绝对路径，使用“/” #19

Closed dy-gdut closed 1 week ago

dy-gdut commented 2 months ago

在“laion/CLIP-ViT-bigG-14-laion2B-39B-b160k” 一栏里填写你的本地clip模型的绝对路径，使用“/” 这句话怎么理解？

smthemex commented 2 months ago

如果你使用提前下载好的模型（直接下载那种，而不是hub自动下载的），你的模型文件会存在电脑的某个文件夹里，这里是提示你要填写F:/xxx/xxx/你的文件夹，不能填写windows的F:\xxx\xxx\你的文件夹 这种格式。

dy-gdut commented 2 months ago

这个路径填写在哪里呢 写在ComfyUI_StoryDiffusion/config/models.yaml下吗？

dy-gdut commented 2 months ago

Use_Single_XL_Model: path: "https://huggingface.co/RunDiffusion/Juggernaut-XL-v9/blob/main/Juggernaut-XL_v9_RunDiffusionPhoto_v2.safetensors" single_files: true ### if true, is a civitai model use_safetensors: true

RealVision: path: "SG161222/RealVisXL_V4.0" single_files: false use_safetensors: true

SDXL: path: "stabilityai/stable-diffusion-xl-base-1.0" single_files: false use_safetensors: true

Unstable: path: "stablediffusionapi/sdxl-unstable-diffusers-y" single_files: false use_safetensors: false

SDXL_Flash: path: "sd-community/sdxl-flash" single_files: false use_safetensors: true

Playground_v2p5: path: "playgroundai/playground-v2.5-1024px-aesthetic" single_files: false use_safetensors: true

这个配置文件里面没有看到这个clip的关键字

smthemex commented 2 months ago

这个是独立的，只是在使用双角色才会使用，如果不使用，是可以不填的。填写方式就是在节点栏aion/CLIP-ViT-bigG-14-laion2B-39B-b160k里填写，不需要去yaml文件

dy-gdut commented 2 months ago

有个问题想请教一下 模型读取还是读取不起来 能否微信截图指导一下 15377318053

dy-gdut commented 2 months ago

ERROR:root:!!! Exception during processing !!! ERROR:root:Traceback (most recent call last): File "C:\sd\ComfyUI-aki-v1.3\ComfyUI-aki-v1.3\execution.py", line 152, in recursive_execute output_data, output_ui = get_output_data(obj, input_data_all) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "C:\sd\ComfyUI-aki-v1.3\ComfyUI-aki-v1.3\execution.py", line 82, in get_output_data return_values = map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "C:\sd\ComfyUI-aki-v1.3\ComfyUI-aki-v1.3\execution.py", line 75, in map_node_over_list results.append(getattr(obj, func)(slice_dict(input_data_all, i))) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "C:\sd\ComfyUI-aki-v1.3\ComfyUI-aki-v1.3\custom_nodes\ComfyUI_StoryDiffusion\Storydiffusion_node.py", line 1069, in story_model_loader pipe = StableDiffusionXLPipeline.from_single_file( ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "C:\sd\ComfyUI-aki-v1.3\ComfyUI-aki-v1.3\venv\Lib\site-packages\huggingface_hub\utils_validators.py", line 114, in _inner_fn return fn(args, kwargs) ^^^^^^^^^^^^^^^^^^^ File "C:\sd\ComfyUI-aki-v1.3\ComfyUI-aki-v1.3\venv\Lib\site-packages\diffusers\loaders\single_file.py", line 491, in from_single_file loaded_sub_model = load_single_file_sub_model( ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "C:\sd\ComfyUI-aki-v1.3\ComfyUI-aki-v1.3\venv\Lib\site-packages\diffusers\loaders\single_file.py", line 127, in load_single_file_sub_model loaded_sub_model = _legacy_load_scheduler( ^^^^^^^^^^^^^^^^^^^^^^^ File "C:\sd\ComfyUI-aki-v1.3\ComfyUI-aki-v1.3\venv\Lib\site-packages\diffusers\loaders\single_file_utils.py", line 1472, in _legacy_load_scheduler return cls.from_config(scheduler_config) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "C:\sd\ComfyUI-aki-v1.3\ComfyUI-aki-v1.3\venv\Lib\site-packages\diffusers\configuration_utils.py", line 260, in from_config model = cls(init_dict) ^^^^^^^^^^^^^^^^ File "C:\sd\ComfyUI-aki-v1.3\ComfyUI-aki-v1.3\venv\Lib\site-packages\diffusers\configuration_utils.py", line 658, in inner_init init(self, args, init_kwargs) File "C:\sd\ComfyUI-aki-v1.3\ComfyUI-aki-v1.3\venv\Lib\site-packages\diffusers\schedulers\scheduling_ddim.py", line 209, in init self.betas = torch.linspace(beta_start0.5, beta_end0.5, num_train_timesteps, dtype=torch.float32) ** 2


TypeError: unsupported operand type(s) for ** or pow(): 'NoneType' and 'float'
smthemex commented 2 months ago

第一次使用单体模型，需要能外网下载config文件，而且最好不要用魔改的单体模型，用一个大众点的。你这个没法联网的报错

dy-gdut commented 2 months ago

关键是我开了梯子 可以从抱脸去下载模型 用的模型是Juggernaut-XL_v9_RunDiffusionPhoto_v2.safetensors

dy-gdut commented 2 months ago

有单体模型推荐吗 我试了很多个 都不行

smthemex commented 2 months ago

报错文件显示你是diffuser版本高于0.28，但是奇怪config文件读取失败。你试过选择非单体模型，会不会自动下载？如果不会，那是你没有开秋叶的镜像下载

dy-gdut commented 2 months ago

是不是开启了代理去下模型就可以了

dy-gdut commented 2 months ago

Error occurred when executing Storydiffusion_Model_Loader:

The provided pretrained_model_name_or_path "C:/sd/ComfyUI-aki-v1.3/ComfyUI-aki-v1.3/models/checkpoints/HoloAnime-XL——动漫风格SDXL模型_v2.7.safetensors" is neither a valid local path nor a valid repo id. Please check the parameter.

dy-gdut commented 2 months ago

错误变成这样子了

smthemex commented 2 months ago

单体模型首次运行是要开代理下config文件的，就是几M的文件，按理下载好了 以后就不用了。你去找我B站的微信群号吧，这里聊不是办法

dy-gdut commented 2 months ago

'Please add the triger word " img " behind the class word you want to customize, such as: man img or woman img' 现在是这个报错

smthemex commented 2 months ago

如果用图生图，必须有一个关键词 “img ”在角色的prompt里，你应该是不小心删掉了

dy-gdut commented 2 months ago

我采用的是text2ing那个节点

smthemex commented 2 months ago

是这样的，text2ing要生效，必须选择origin那个菜单的origin，迟点我改一下逻辑吧，有点啰嗦了。