Caught exception in thread: '_training_0'

If training is failing to start, and you are not receiving an error message telling you what to do, tell us about it here


Forum rules

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

This forum is for reporting errors with the Training process. If you want to get tips, or better understand the Training process, then you should look in the Training Discussion forum.

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

Locked
User avatar
DutchMex
Posts: 1
Joined: Sat Jul 11, 2020 4:15 pm

Caught exception in thread: '_training_0'

Post by DutchMex »

Hi Faceswap Community,
I am relatively new here (only been experimenting with this software for a few months). I have been training a model for a while, then gave it a break. When I came back to it, even after updating, it always crashes. I have tried to add photos to give the model more to work with, and made sure that they are all the same size when extracted. I have tried different models as well and all of them seem to crash the same way. Ideally I want to use Villain or realface. At best I can only get the system to run one cycle before crashing.

PC Hardware:
GPU: RTX 2070 Super
CPU: Intel i7 9700
RAM: 16 GB

(sorry I don't know how to make the log look nice so I linked to it on gitlab)
Log:
https://gitlab.com/CmdQue/faceswap-file ... .gitignore

User avatar
torzdf
Posts: 2651
Joined: Fri Jul 12, 2019 12:53 am
Answers: 159
Has thanked: 129 times
Been thanked: 622 times

Re: Caught exception in thread: '_training_0'

Post by torzdf »

Unfortunately the Memory Saving Gradients has pushed the useful information off the log.

However, this error usually occurs when the training images are not all the same size, so I would start with checking that.

My word is final

Locked