4.500 iterations and 16 hours i lauched the training

Want to understand the training process better? Got tips for which model to use and when? This is the place for you


Forum rules

Read the FAQs and search the forum before posting a new topic.

This forum is for discussing tips and understanding the process involved with Training a Faceswap model.

If you have found a bug are having issues with the Training process not working, then you should post in the Training Support forum.

Please mark any answers that fixed your problems so others can find the solutions.

Locked
User avatar
danyfany
Posts: 1
Joined: Thu Mar 24, 2022 1:54 pm

4.500 iterations and 16 hours i lauched the training

Post by danyfany »

Hello guy,

I have a PC intel i3 with 8 giga of RAM and yersterday i started a training project.

Video of face A is about 1 minute and video of Face B about 2 minutes, both very good quality (1080p).

For the configuration, i let everything by default.

It has been 16 hours from now that i launched the training and the iteration is only 4.500!! Is it normal that it is so slow or i miss something

16 hours for only 4.500 iterations, i will need weeks to have at least the reasonable value of 100k iterations.

How can i fix it or it is just my pc? i use CPU for info

User avatar
torzdf
Posts: 2649
Joined: Fri Jul 12, 2019 12:53 am
Answers: 159
Has thanked: 128 times
Been thanked: 623 times

Re: 4.500 iterations and 16 hours i lauched the training

Post by torzdf »

There is no way to speed up CPU training. ML on CPU is extremely slow and not really a viable option.

You can train on Google Colab (viewtopic.php?f=23&t=1886) or you will need to purchase a (preferably Nvidia) GPU.

My word is final

Locked