luxiaolili comments

Results 6 comments of


                                            luxiaolili

How to prepare data, mask, etc,...

> Since you release training code, I really want to try finetune on my dataset but I cant understand how to prepare data you can prepare data follow https://github.com/luxiaolili/IDM-VTON-train/

how to generate tagged_json for jewelery?

> _No description provided._ blip2, llava GPT-4o，or other multimodel

How do you get an agnostic-mask

> Thank you for your code. How do you get the agnostic-mask, I think it is an important factor for the result. you can see for data prepare https://github.com/luxiaolili/IDM-VTON-train/

While training, how can you get the final 13 channels model, since I have obtained 4 channels model which seems like a trained sdxl base model not a sdxl inpaint model?

I realize it using sdxl inpaint model need a conv layer to change the in_channel, the train code for it see https://github.com/luxiaolili/IDM-VTON-train/

While training, how can you get the final 13 channels model, since I have obtained 4 channels model which seems like a trained sdxl base model not a sdxl inpaint model?

> hello, in the inference code, the input channel need to be changed in 13? I donot see any code operation in the inferenc.py in the training script, unet.config['in_channels'] =...

What's the loss function when fine tuning?

the loss both l1 loss for noise, only change the data and descripe of the data, and cloth mask , I realize the code for train see https://github.com/luxiaolili/IDM-VTON-train/