Skip to content

Unable to load the variant=None of the model #1198

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
sk8boi opened this issue Apr 10, 2025 · 0 comments
Open

Unable to load the variant=None of the model #1198

sk8boi opened this issue Apr 10, 2025 · 0 comments
Assignees

Comments

@sk8boi
Copy link

sk8boi commented Apr 10, 2025

你好、飞桨开发人员和用户们、我在尝试使用
export FLAGS_conv_workspace_size_limit=4096
python -u train_txt2img_control_trainer.py
--do_train
--output_dir ./sd15_control
--per_device_train_batch_size 4
--gradient_accumulation_steps 1
--learning_rate 1e-5
--weight_decay 0.02
--lr_scheduler_type "constant"
--warmup_steps 0
--sd_locked True
--max_steps 10000000
--logging_steps 50
--image_logging_steps 400
--save_steps 2000
--save_total_limit 2
--seed 23
--dataloader_num_workers 4
--pretrained_model_name_or_path runwayml/stable-diffusion-v1-5
--max_grad_norm -1
--file_path ./fill50k
--recompute True
--overwrite_output_dir
命令训练controlnet的时候、因为需要下载stable-diffusion-v1-5模型、而官网上

Image
这个huggingface的模型地址已经失效、因此我直接使用git clone https://huggingface.co/stable-diffusion-v1-5/stable-diffusion-v1-5
在huggingface、新地址下载stable-diffusion-v1-5、将pretrained_model_name_or_path路径改为下载完后的地址、但是仍然出现无法找到vae模型的错误
Unable to load the variant=None of the model from /home/PaddleMIX/ppdiffusers/runwayml/stable-diffusion-v1-5/vae! Please make sure the specified variant exists and is correct.
Unable to load the variant=fp16 of the model from /home/PaddleMIX/ppdiffusers/runwayml/stable-diffusion-v1-5/vae! Please make sure the specified variant exists and is correct.
Unable to load the variant=fp32 of the model from /home/PaddleMIX/ppdiffusers/runwayml/stable-diffusion-v1-5/vae! Please make sure the specified variant exists and is correct.
Traceback (most recent call last):
File "/home/PaddleMIX/ppdiffusers/examples/controlnet/train_txt2img_control_trainer.py", line 101, in
main()
File "/home/PaddleMIX/ppdiffusers/examples/controlnet/train_txt2img_control_trainer.py", line 64, in main
model = ControlNet(model_args)
File "/home/PaddleMIX/ppdiffusers/examples/controlnet/control/model.py", line 66, in init
self.vae = AutoencoderKL.from_pretrained(vae_name_or_path)
File "/usr/local/lib/python3.10/dist-packages/ppdiffusers-0.29.1-py3.10.egg/ppdiffusers/models/modeling_utils.py", line 990, in from_pretrained
assert len(resolved_model_files) > 0, (
AssertionError: We are attempting to load the variant in [None, fp16, fp32]. But unfortunately, no model files were found in the path /home/PaddleMIX/ppdiffusers/runwayml/stable-diffusion-v1-5/vae. Please check if the provided path is correct and ensure that it contains the necessary model files. If the issue persists, consider redownloading the model files or contacting the model provider for assistance.

Image
这是我下载的vae模型文件、请问是目前框架不支持现有的huggingface的stable-diffusion-v1-5模型吗

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

3 participants
@westfish @sk8boi and others