ProGamerGov

So it it currently possible to train a LoRA model for SD 2.1 and then convert it to a safetensor or ckpt/pt? I see this PR added a safetensor to...

@dfaker I would try testing with a batch size of 50, as it fails on the second or third batch . I am unable to reproduce the bug at lower...

@dfaker Nothing to indicate why it happens on the terminal window or the UI page. It also just happened again while using the DPM++ 2S a Karras sampler after reaching...

@dfaker The generate button no longer works and I have to restart the UI to fix it. And I am not switching to another WebUI tab at all during the...

Just had it happen again when DPM++ 2S a Karras sampler after reaching 2100 images using a batch size of 50 and 50 steps. A simple F5 refresh fixes it,...

Just froze again while using DPM++ 2S a Karras sampler after reaching 400 images using a batch size of 50 and 50 steps. This time it stopped at 86% completion...

@dfaker I'm running it on a Google Cloud Compute VM.

Just ran into this issue: ``` [Epoch 0: 11/142]loss: 0.4385366: 0%| | 10/100000 [00:31 0, "No inf checks were recorded for this optimizer." AssertionError: No inf checks were recorded for...

Ran into this issue just now, but I resolved the issue by downgrading gradio. ``` Setting up a public link... we have recently upgraded the way public links are generated....

> To use it with Stable Diffusion, You can take the generated latent from Stable Diffusion and pass it into the upscaler before decoding with your standard VAE. Or you...