-
Notifications
You must be signed in to change notification settings - Fork 10
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
How to test the trained model's auc. #9
Comments
Hello @wzk101, I've uploaded the results of our model detection to the result folder. Below is the code I've used to compute the AUC. If you wish to compute the AUC for your own detection, you can use the code I've provided below. Enjoy!
|
thanks a lot, I tested the model in the LQ FF++, and the auc was good while the acc performed bad. Is that normal? In celeb and HQ FF++ the model is awesome. |
Which FF++ sub class did you test? can you post the result you get? |
youtobe original, deepfakes, face2face, neuraltexture and face swap, I detected in video level and set 15 frames, and got ACC:0.4856 , AUC: 0.884. |
That's a very low accuracy rate. Maybe the calculation of the accuracy is incorrect? I'll test it on my system later today or tomorrow. |
No description provided.
The text was updated successfully, but these errors were encountered: