Ben

Results 1311 comments of Ben

TRaining of text encoder and changing learning rate....

did you experiment with learning_rate=1e-6 ? I'm testing the argument --train_text_encoder right now to make sure there is no memory error

TRaining of text encoder and changing learning rate....

Thanks for the feedback, great help.

TRaining of text encoder and changing learning rate....

@1blackbar you'll get the same results

TRaining of text encoder and changing learning rate....

by default the sample batch size is 4, and as for the new vae, it is already merged in the download model cell

TRaining of text encoder and changing learning rate....

1500 steps, 2e-6, 200 autogenerated class images, 28 instance images ![download (1](https://user-images.githubusercontent.com/110583491/197384923-6359d788-d72a-4975-a445-ad41bb1ef045.png) ![download (3)](https://user-images.githubusercontent.com/110583491/197384929-691264b7-2dc2-4c1c-9587-ed7f38c5df7a.png) ![download (4)](https://user-images.githubusercontent.com/110583491/197384932-9a8920eb-915d-4f3e-bb23-fe625e3483f0.png) ![download (5)](https://user-images.githubusercontent.com/110583491/197384939-114d6f40-54c4-4754-8893-0f3f24a0fabc.png)

apply_strength problem

remove the sd folder from your gdrive, AUT1111 repo was updated

runpod version

I'll look into it

Couldn't using META's new inference engine increase Fast Stable's speed even more?

I've been trying to get it to work, but I'm getting some errors about syntax, any feedback would be great

How to Enter Parameters

probably a different gradio version ?

How to Enter Parameters

thanks