Visualization problem

[ZH3nGsQ]

When I use your pretrained model and run the infer_mdt.py to generate images. The outcome--"sample.jpg" was very noisy. Can you tell me what is happened?

[ZH3nGsQ]

And I also encountered with another problem in training stage is that--"RuntimeError: a leaf Variable that requires grad is being used in an in-place operation." I don't know how to solve that.

[xv994]

hey bro, I met some problems when I ran the "infer_mdt.py". Can you tell me how can I run this .py file? Should I change or add some parameters?

[gasvn]

Have you do the pip install -e .? I tested on my side, and it works fine.

[gasvn]

When I use your pretrained model and run the infer_mdt.py to generate images. The outcome--"sample.jpg" was very noisy. Can you tell me what is happened?

did you download the pretrained model from hugglingface?

[ZH3nGsQ]

hey bro, I met some problems when I ran the "infer_mdt.py". Can you tell me how can I run this .py file? Should I change or add some parameters?

I didn't change or add any parameters.

[ZH3nGsQ]

When I use your pretrained model and run the infer_mdt.py to generate images. The outcome--"sample.jpg" was very noisy. Can you tell me what is happened?

did you download the pretrained model from hugglingface?

I already download pretrained model "mdt_xl2_v2_ckpt.pt" from hugging face. And I use the pip install -e . . The outcome is still worse.

[gasvn]

Did you find any warning or so? If looks like the pretrained model is not correctly loaded.