EnVision-Research / LucidDreamer

Official implementation of "LucidDreamer: Towards High-Fidelity Text-to-3D Generation via Interval Score Matching"
MIT License
749 stars 32 forks source link

how to find Pretrained Diffusion Models config.json file #14

Closed jianan-kristine closed 10 months ago

jianan-kristine commented 10 months ago

Firstly, you may need to change model_key: in the configs\.yaml to link the local Pretrained Diffusion Models ( Stable Diffusion 2.1-base in default)

I wonder how to find config.json file. After downloading the file to "/home/xxx/LucidDreamer-main/stabilityai/stable-diffusion-2-1-base" ,I came across this error. image

This is the files in the StableDiffusion model. image

YixunLiang commented 10 months ago

Hi, sorry for the confusion. We will update detailed training instructions later.

Firstly, you need to download the pre-trained diffusion model on this page: stabilityai/stable-diffusion-2-1-base · Hugging Face. Then, change the model_key to link the location of pretrained diffusion models.

Here is an example:

the location of the models is: /home/yixunliang/pretrained/stable-diffusion-2-1-base image

Then, we need to modify the model_key to: image

After that, you can use it.

Feel free to contact me if you encounter other problems.

jianan-kristine commented 10 months ago

image here are the files in your link, I solved the problem of finding model_index.json. But I still couldn't find the file named config.json. image

YixunLiang commented 10 months ago

Since our model is built on the diffusers, you need to download the whole file in huggingface and then link them. You can do it as follow:

# HTTPS
git lfs install
git clone https://huggingface.co/stabilityai/stable-diffusion-2-1-base

Then, link it. Would not directly clone the repo of Stable Diffusion 2.1-base .

Or you can use the model_key that in .yaml. like model_key: 'stabilityai/stable-diffusion-2-1-base' . Your computer will download it automatically.