nom
nom
In the paper you mention you validate the hyperparameters for the input processing (the FGSM magnitude) and the feature ensemble using adversarial samples (right part of Table 2 in the...
Through experimentation I've observed that the identity transfer performance is quite poor for African ethnicity faces as opposed to other ethnicities. I've delved a bit deeper and I suspect it's...
## 🐛 Bug I've adapted https://github.com/crowsonkb/k-diffusion/ for torch_xla use. In the evaluation step (diffusion sampling), there is a very high (>7minute) compilation time after a number of sampling steps. It...
Hi, fantastic work with NATTEN. Is support for XLA on your roadmap? It would be great to enable neighborhood attention on TPUs and other non-Nvidia GPUs.
Hi, I've been looking at the face segmentation masking feature. IIUC it should work such that only the face is being learnt and nothing else. So the loss of non-face...
Hey, great work! Quick question on training. I was wondering how you're fitting two SDXL UNets (garment UNet and tryon UNet) on a single A800 with batch size 24/4=6 (assuming...
hey Nico, would it be helpful at all to add Polygon tx support? happy to do so as a little react practice for myself