Convert in training settings
Posted: Mon Nov 21, 2022 9:57 am
Can someone explain why there is the option to adjust the convert batchsize in the global training settings?
Has this to do with the convert that is showing in the preview?
The place to discuss Faceswap and Deepfakes
https://forum.faceswap.dev/
Can someone explain why there is the option to adjust the convert batchsize in the global training settings?
Has this to do with the convert that is showing in the preview?
Honestly, it was added a long time ago, but makes very little difference to the conversion speed (the bottleneck is the CPU not the GPU during convert). The setting just dictates how many faces are fed through the model at once to get the swap at each call.