Thund3rPat
Thund3rPat
Hi kenan3507, sorry for the really late answer. You can of course use payloads developed in C++. Compile it to an executable and use pe_to_shellcode to convert it to shellcode...
In your first error message, you accidentally set the wrong model path: `pretrained_model_name_or_path=H:/kohya_ss/train/Config.json` And in the screenshot, it seems like you ran out of memory: `Cuda out of memory` Can...
I was able to resolve this issue by merging the dev branch into the master branch. This commit 641a168e55f429c79f9114bcdb123a13bc9b2167 resolved it for me and was probably forgotten.
@Maranpani With the last update it is now merged. No need to do it yourself anymore.
Hi, of course. To update to the newest release, open the terminal and go to the kohya_ss folder. Execute the upgrade script with: ``` .\upgrade.ps1 ``` I hope this helps.
@Maranpani Can you deactivate 8bit adam and try again?
@FlareWaterLily The error message says that you have set the resolution parameter wrong. Can you check again and set it like this: ``` 640,768 ```
You simply run out of Memory. You can try to use AdamW8bit optimizer, but with 4 GiB total capacity, it will be very difficult.
Do you have CUDA Toolkit installed? I had a similar problem and was able to fix it by installing it from the official NVIDIA page.
What else would come to my mind would be to disable 8bit adam. However, this would lead to a very high memory consumption...