Search found 15 matches

by sp13
Thu Mar 24, 2022 10:31 pm
Forum: Convert Discussion
Topic: Can the same model convert A to B and B to A?
Replies: 1
Views: 4211

Re: Can the same model convert A to B and B to A?

Yes click the "Swap Model" checkbox in the settings or use the "-s" command line option. While the guide mentions this is useful if you've gotten A & B confused, you can do it intentionally if you wish, changing the setting based on the video you are swapping into. Swap Model...
by sp13
Thu Jun 24, 2021 2:41 am
Forum: Training Discussion
Topic: Training from multiple sources - how to combine alignments?
Replies: 4
Views: 11013

Re: Training from multiple sources - how to combine alignments?

Put all the extracted face .png files (for one particular person) into one directory and use it as the training source Alignments are stored in header of the .png files The alignment files are not used for training at all, so there is no way to read multiple files (or even one) and no need to merge...
by sp13
Sat Jun 05, 2021 5:13 am
Forum: Convert Support
Topic: Crash Report
Replies: 3
Views: 11022

Re: Crash Report

Don't use spaces in your filename. I'm not sure if the Korean characters are ok or not.

by sp13
Thu Jun 03, 2021 3:06 pm
Forum: Cloud Discussion
Topic: [Outdated] Training Using Google Colab
Replies: 66
Views: 138329

Re: [Resource] Training Using Google Colab

foundmyway89 wrote: Thu Jun 03, 2021 12:32 am

Please do! I've been trying to get mine working for a week now...

The notebook from this post should work fine viewtopic.php?p=5136#p5136

by sp13
Sun May 09, 2021 8:08 am
Forum: Extract Discussion
Topic: Manual Alignment Bug?
Replies: 4
Views: 11786

Re: Manual Alignment Bug?

Yeah, when extracting with N > 1 like that, the first time I open the manual tool I let it load, then immediately close it without doing anything else. Then I open it again. It is always better the second time.

by sp13
Mon May 03, 2021 8:04 pm
Forum: Cloud Discussion
Topic: Some benchmarking numbers
Replies: 6
Views: 17243

Re: Some benchmarking numbers

I did have some spare credits so I've been playing with a V100 some more. Using something similar to Villain I get about 2.5 times the performance of the T4, which is exactly what my Realface experiments showed. The V100 is a hungry beast. On an n1-standard-4, faceswap averaged about 333% CPU so not...
by sp13
Sat May 01, 2021 9:20 pm
Forum: Cloud Discussion
Topic: Some benchmarking numbers
Replies: 6
Views: 17243

Re: Some benchmarking numbers

Yeah, I always use preempt when I can. The particular day that I ran the V100 I couldn't get a T4 even at non-preempt price. I haven't had a problem since. I stopped my realface experiments because they didn't play nice with my little 6 GB GTX 1060. But villain with mixed precision is very nice on a...
by sp13
Mon Apr 19, 2021 7:48 pm
Forum: Cloud Discussion
Topic: Some benchmarking numbers
Replies: 6
Views: 17243

Re: Some benchmarking numbers

I'm curious about the poor V100 numbers you were seeing. I haven't ran any specific benchmarks, but yesterday I was training a realface model on a n1-standard-2 + T4. I was getting around 78 EG/s (batchsize = 32, mixed precision enabled) with CPU utilization about 69%. Today I wanted to redo the mod...
by sp13
Sat Apr 17, 2021 5:09 pm
Forum: Cloud Discussion
Topic: Some benchmarking numbers
Replies: 6
Views: 17243

Re: Some benchmarking numbers

Interesting. I was wondering how the P100 would perform compared to the T4.
So far my experiments have been with a T4 on n1-standard-2. It doesn't seem to be having CPU or system RAM limitations and could save like 10 cents an hour ;)

by sp13
Thu Apr 15, 2021 5:31 pm
Forum: Cloud Discussion
Topic: [Outdated] Training Using Google Colab
Replies: 66
Views: 138329

Re: [Resource] Training Using Google Colab

Another random tip that I don't think is in this thread: If you put the following line somewhere after the git clone that installs faceswap !echo "{\"backend\": \"nvidia\"}" > faceswap/config/.faceswap then faceswap will know that you want to use nvidia and won't ask yo...
by sp13
Thu Apr 15, 2021 4:59 pm
Forum: Cloud Discussion
Topic: [Outdated] Training Using Google Colab
Replies: 66
Views: 138329

Re: [Resource] Training Using Google Colab

What are you using to bypass the colab interactive timeout? I use some Javascript snippet that I'm not sure actually works because I've gotten timed out before and sometimes get a "Are you still here?" captcha. I ran across this stackoverflow the other day with lots of things to try, thou...
by sp13
Wed Apr 14, 2021 4:54 pm
Forum: Hardware
Topic: Hardware Performance Data Request.
Replies: 8
Views: 19048

Re: Hardware Performance Data Request.

Google Colab There are different GPUs available, but you have no control over which one you get. No GUI, so I calculated EGs/sec manually as (iterations * batchsize * 2) / seconds . (The 2 because both the A and B sides get a batchsize of faces. This is the way the GUI calculates it now but I don't...
by sp13
Wed Apr 14, 2021 2:13 am
Forum: Cloud Discussion
Topic: [Outdated] Training Using Google Colab
Replies: 66
Views: 138329

Re: [Resource] Training Using Google Colab

You should be able to unzip them without even copying them. I imagine something like !unzip "/content/drive/MyDrive/colab_files/faceswap/faces/face_a.zip" -d /content/ should work, though I use tar instead of zip. Also for GPU check, I like to throw in !nvidia-smi import tensorflow as tf p...
by sp13
Tue Apr 13, 2021 11:14 pm
Forum: Cloud Discussion
Topic: [Outdated] Training Using Google Colab
Replies: 66
Views: 138329

Re: [Resource] Training Using Google Colab

I get the same error message about albumentations but it doesn't seem to hurt anything so I just ignore it. You do have to press the button to restart the runtime though (or at least I always do it.) One thing to note is that Colab comes with Tensorflow already installed and Google advises you not t...
by sp13
Sat Apr 10, 2021 12:25 am
Forum: Hardware
Topic: Hardware Performance Data Request.
Replies: 8
Views: 19048

Re: Hardware Performance Data Request.

GPU: GTX 1060 6 GB Mainboard: AMD FX8350 + 32 GB RAM OS: Debian 10 NVIDIA 418.181.07 driver + Tensorflow 2.2 Original BS = 2: EGs/sec = 20.5 BS = 140: EGs/sec = 132.3 BS = 256: EGs/ sec = 137.5 looks CPU limited at large batchsize Villain BS = 2: EGs/sec = 7.9 BS = 6: EGs/sec = 11.3 DFL H128 BS = 2:...