betterftr

inference code for super-resolution?

2

Is it on the todo list?

AttributeError: 'HifiGAN' object has no attribute 'device'

6

on windows, anaconda

need help: better quality on old man's voice

3

Anyone can help, what parameters to change in the config file to improve resemblance of trained voice closer to ground truth? Like n_layers? After 80k steps the male voice I'm...

finetune

6

Finetuning code will be released as well? Awesome project btw! Cant wait to train a custom model

### Question How do I train the new 1.6 model? just put --model_name_or_path liuhaotian/llava-v1.6-vicuna-13b in finetune_lora.sh?

Any chance for an implementation of the new Siglip vision processor? https://huggingface.co/google/siglip-large-patch16-256 for minigpt-4 and v2

suggestion

3

any way to implement guidance? (humming for example)

8bit CAME optimizer

1

### Feature request I would really love a 8bit version of CAME optimizer being supported. ### Motivation If the implementation is possible. ### Your contribution Pixart Sigma model was trained...

SD 3.0 Incoming!

1

Just got announced! HYPE https://stability.ai/news/stable-diffusion-3

[Bug]: Prodigy lr stays at Initial D value and not dynamic

13

### What happened? trying to train cascade tenc only with prodigy, but no matter what I tried to change the lr stays at the initial D value of the optimizer...