What was the full command you ran?
You train until you are happy with the previews. You can view the previews with a pop up window by passing the -p
flag, or you can write the preview to the faceswap folder with the -w
flag and view it there
What was the full command you ran?
You train until you are happy with the previews. You can view the previews with a pop up window by passing the -p
flag, or you can write the preview to the faceswap folder with the -w
flag and view it there
I'm unlikely to post a video, sorry. I just don't have the time to make it at the moment.
Honestly?
I haven't started either of them yet. Your best resources for these currently are USAGE.md in the github repo or the pinned topics in the relevant channels on our Discord Server.
I'm not sure what it is you're asking?
You should be performing the extraction process separately for your "A" and "B" facesets. These shouldn't be merged with each other.
Also, training takes a long time and needs a lot if images. See https://faceswap.dev/forum/app.php/faqpage
We have identified a potential issues on certain batch-sizes.
It still doesn't explain your CPU usage (although, that may be numpy doing multiprocessing in the background).
When we have a bit of time, we are going to review the data augmentation code to see if we can optimize it a bit better.
I can't talk for cloud based options, as I don't use them, but as to your "preview" issue, if you use the -w
flag, it will write out a preview to your faceswap folder every save iteration, that you can pull down and check the progress.
Thanks for this.
I am currently having GPU issues, so I will have to wait for some parts to arrive before I can run my tests.
Off the top of my head, these figures seem reasonable though.
Ok. I'm away from my PC at the moment, and don't have the figures on the top of my head. I will double check though and get back to you tomorrow.
In the first instance, try a reboot. This normally means something has gone wrong in Cuda/cuDNN which needs a reboot to fix.
Can you open a new topic and post the crash_report (from your facewap folder) there please.
Also, please mark my post previous as "Answer" (the tick icon)
I'm going to lock this one off as it's solved
It will depend a lot on the model.
The most important thing is quality of data.
Which model are you using?
You should try using the preview tool to set your convert settings prior to convert.