Zoom the graph during 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
uequalsn
Posts: 4
Joined: Tue Dec 10, 2019 1:32 pm
Has thanked: 1 time
Been thanked: 1 time

Zoom the graph during the training

Post by uequalsn »

Hi All,

I was wondering if there's a possibility to zoom into the loss function graph during the training phase to make sure the model benefits from the training. Sometimes, when number of iterations is quite high, it is virtually impossible to tell if my mode has already converged just by looking at the graph at full scale.

Thanks in advance!

by torzdf » Wed Dec 11, 2019 11:29 am

Unfortunately that isn't possible at the moment, but will likely be added one day.

In the meantime you can zoom the current graph from the analysis tab (covered in this section of the training guide: viewtopic.php?f=6&t=146#monitor)

Go to full post
User avatar
torzdf
Posts: 2687
Joined: Fri Jul 12, 2019 12:53 am
Answers: 159
Has thanked: 135 times
Been thanked: 628 times

Re: Zoom the graph during the training

Post by torzdf »

Unfortunately that isn't possible at the moment, but will likely be added one day.

In the meantime you can zoom the current graph from the analysis tab (covered in this section of the training guide: viewtopic.php?f=6&t=146#monitor)

My word is final

User avatar
uequalsn
Posts: 4
Joined: Tue Dec 10, 2019 1:32 pm
Has thanked: 1 time
Been thanked: 1 time

Re: Zoom the graph during the training

Post by uequalsn »

I think this is exactly what I was looking for! Weird that I've missed this part of the tutorial.

Thanks for the prompt response!

Locked