『nvidia 2080ti』吾支持半精度類型, 早知買『nvidia 3090ti』
| modules.devices.NansException: A tensor with NaNs was produced in Unet. This could be either because there’s not enough precision to represent the picture, or because your video card does not support half type. Try setting the “Upcast cross attention layer to float32” option in Settings > Stable Diffusion or using the –no-half commandline argument to fix this. Use –disable-nan-check commandline argument to disable this check. |
| A tensor with all NaNs was produced in VAE.
Web UI will now convert VAE into 32-bit float and retry. To disable this behavior, disable the ‘Automatically revert VAE to 32-bit floats’ setting. To always start with 32-bit VAE, use –no-half-vae commandline flag. |
- 𢴇行『webui-user.bat』著『Stable Diffusion webui』
- 撳『Settings』->『Stable Diffusion』
- 勾『Upcast cross attention layer to float32』
- 撳『Apply settings』應用設定
- 撳『Reload UI』重啟界面
另壹條橋係加入命令行參數
- 編緝『C:\stable-diffusion-webui\webui-user.bat』
- 係命令行參數『COMMANDLINE_ARGS』加入『–no-half』冇半精度類型, -『-no-half-vae』『–disable-nan-check』禁nan檢驗.
| webui-user.bat |
| set COMMANDLINE_ARGS= –no-half –no-half-vae –disable-nan-check |
| –no-half-vae | 吾將VAE模型切換成16bit浮點數 |
| –no-half | 吾將模型切換成16bit浮點數 |
| –disable-nan-check | 吾檢查生成圖像潛在空間是否包含nan值,係持續集成中運行時無需檢查點 |