-
Does ai-toolkit support training non-square images? I don't want to crop and scale the images.
I am training flux with 768*1024 images. On an A100-40G machine, 3000 steps take 2.5 hours. I set buck…
-
Thank you very much for your work. For the optical flow dataset, how can the finally generated mask_DPS and mask_PPF be converted into masks or valid images that can be used in optical flow model trai…
-
### Search before asking
- [X] I have searched the Ultralytics YOLO [issues](https://github.com/ultralytics/ultralytics/issues) and [discussions](https://github.com/ultralytics/ultralytics/discussion…
-
Thank you for the awesome dataset!
The description about the train/test set splits is pretty unclear in the paper and I couldn't find more information in this repository too.
> We split the datas…
-
**Info**
+ Hybrid Training "How to Publish in Springer Nature" How to Publish in Springer Nature
+ Hari /…
-
I would like to ask the following two questions:
1. When training the latent UNet (the additional U-Net for latent Refusion), how many images are in your training dataset? (Specifically, how many i…
-
Without the GAN loss, we find the reconstructed images are over-smooth, will you opensource the StyleGAN code when training Multi-scale VQ?
-
Hi, when trying to log the reconstructed images in the stage 1 training, I got this error
```
File "/home/rohol/projects/1d-tokenizer/utils/train_utils.py", line 762, in reconstruct_images
…
-
When my images in my dataset are higher than 1024, like 2048 or 2536 the likeness in my training is gone. is there any way to fix this or is it better to train on lower res images?
-
Hello,
I have a project that logs an image to wandb at each epoch. I'd like to download the sequence of images now to make a video.
I'm not seeing any way to download the entire sequence in one …