Kohaku-Blueleaf

Results 20 issues of Kohaku-Blueleaf

Reference to this issue: https://github.com/Stability-AI/stablediffusion/issues/104 The proj_out should convert inner_dim back to in_channels.

https://github.com/Stability-AI/stablediffusion/blob/d55bcd4d31d0316fcbdf552f2fd2628fdc812500/ldm/modules/attention.py#L318 Since the proj_in does in_ch->inner_dim Shouldn't the proj_out does inner_dim -> in_ch? The conv version of proj_out also does inner_dim -> in_ch https://github.com/Stability-AI/stablediffusion/blob/d55bcd4d31d0316fcbdf552f2fd2628fdc812500/ldm/modules/attention.py#L312-L316

![image](https://user-images.githubusercontent.com/59680068/222614157-39fc93df-f386-42f7-98fd-78e3b28b49ce.png) in both a1111 sd-webui and addnet, LoCon need extension to load correctly. Sd webui will load only part of locon. (BTW, you can use pip install locon now)

Add option for LoraConfig's modules_to_save, for training more modules/layers without LoRA applied. Like embedding or lm_head, which LoRA not good at.

After this commit: cfad895 I got an error like this: ``` RuntimeError: Windows not yet supported for torch.compile ``` It looks like torch.compile can only be used in MacOS and...

https://huggingface.co/JosephusCheung/Guanaco https://huggingface.co/datasets/JosephusCheung/GuanacoDataset

## Description Consistency Decoder: https://github.com/openai/consistencydecoder I added it Just like TAESD. It require lot of resource to "decode" the image (it is actually a latent guided consistency model on pixel...

## Description When disable built-in lora or enable some other ExtraNetworks extensions. The ui callback will not setup for reload/search. Change the check to >=6 which is Hypernet+TI+ckpt * 2....

In the restormer.py We can see attention is implemented as below: ```py qkv = self.qkv_dwconv(self.qkv(x)) q,k,v = qkv.chunk(3, dim=1) q = rearrange(q, 'b (head c) h w -> b head...

## Checklist: - [x] I have read [contributing wiki page](https://github.com/AUTOMATIC1111/stable-diffusion-webui/wiki/Contributing) - [x] I have performed a self-review of my own code - [x] My code follows the [style guidelines](https://github.com/AUTOMATIC1111/stable-diffusion-webui/wiki/Contributing#code-style) -...