CalvinBackup/neuralchen-SimSwap
1
0
Fork 0
Code Issues 288 Pull Requests 16 Actions Packages Projects Releases 2 Wiki Activity

Training on CelebAMask-HQ #329

New Issue
Open
opened 2022-10-14 14:47:06 +02:00 by lizian1611 · 4 comments
lizian1611 commented 2022-10-14 14:47:06 +02:00 (Migrated from github.com)
Copy Link
Copy Source

Hello, I used your model to train on CelebAMask-HQ, and the identity information of the generated image was confused, which was completely different from the identity of the source. May I ask what is the cause of this problem?

Hello, I used your model to train on CelebAMask-HQ, and the identity information of the generated image was confused, which was completely different from the identity of the source. May I ask what is the cause of this problem?
Fibonacci134 commented 2022-10-14 19:36:18 +02:00 (Migrated from github.com)
Copy Link
Copy Source

How many iterations did you train, and what was the result? Was the face kind of white washed? Or did have missing layers?

How many iterations did you train, and what was the result? Was the face kind of white washed? Or did have missing layers?
lizian1611 commented 2022-10-15 08:39:08 +02:00 (Migrated from github.com)
Copy Link
Copy Source

你训练了多少次迭代，结果如何？脸是不是洗白了？还是缺少图层？

Due to the device, I set batchsize to 6. Due to the confusion of identity information, I increased the weight of identity. I have not changed other Settings. The following is the picture of training 15W step.
微信图片_20221015143841

> 你训练了多少次迭代，结果如何？脸是不是洗白了？还是缺少图层？ Due to the device, I set batchsize to 6. Due to the confusion of identity information, I increased the weight of identity. I have not changed other Settings. The following is the picture of training 15W step. ![微信图片_20221015143841](https://user-images.githubusercontent.com/112329687/195973127-e5324098-795d-48cc-ac7a-d6b163f6adad.jpg)
Fibonacci134 commented 2022-10-15 18:59:04 +02:00 (Migrated from github.com)
Copy Link
Copy Source

How much did you increase the weight of the identity by? Your results seem to be following the right path but you might need more iterations.
( Enhanced using gfgpan)

1665852936649.jpg

How much did you increase the weight of the identity by? Your results seem to be following the right path but you might need more iterations. ( Enhanced using gfgpan) ![1665852936649.jpg](https://user-images.githubusercontent.com/97559161/195998774-906e2db9-f3be-459a-8730-ea9241d5fc83.jpg)
Fibonacci134 commented 2022-10-17 13:02:44 +02:00 (Migrated from github.com)
Copy Link
Copy Source

Can you post the model so the community can figure out how to make it work?

Can you post the model so the community can figure out how to make it work?
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
GIF
GUI
Increase inference time
A solution to increase inference time
 bug
Something isn't working
 documentation
Improvements or additions to documentation
 duplicate
This issue or pull request already exists
 enhancement
New feature or request
 good first issue
Good for newcomers
 help wanted
Extra attention is needed
 invalid
This doesn't seem right
 question
Further information is requested
 wontfix
This will not be worked on
No labels
Milestone
No items
No Milestone
Projects
Clear projects
No projects
Assignees
Clear assignees
CalvinBackup (Calvin Backup)
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: CalvinBackup/neuralchen-SimSwap#329