Kazami Michiru

> After testing both pp and ic, I've found that pp works better in cmd, but ic is better in jupyter lab. >  > See that pp does not...

Same issue and you PR does fix the case, Thanks!

As issue [](url)https://github.com/AUTOMATIC1111/stable-diffusion-webui/issues/1862 suggest

As i tried this batched implementation, it seems large batch size (e.g. 8) generally prohibits learning textual inversion. either with original learning rate(0.005, working when batch size = 1) or...

> I was already working on this for both TI and hypernets; my version is not merged it. Also looking at code I think you're increasing step by batch count...

seems a machine translation. such as "restore face" as "还原面".

Need more experiments. I tried much lower LR (say, 2e-7) after some steps, and the problem continues on making noisy image output. But lack of evidence of solving the problem...

Same here

Hi, I also found this problem also in my experiments. It seems in generation some parameters are not used. Except the [PR](https://github.com/microsoft/DeepSpeed/pull/4966#issue-2085323920), a simple workaround can be passing a dummy...

looking forward to the training and finetuning code