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

There is a problem about convergence of the networks #6687

Open
guokainew opened this Issue Feb 1, 2019 · 0 comments

Comments

Projects
None yet
1 participant
@guokainew
Copy link

guokainew commented Feb 1, 2019

When i train my image use this Code,there is a problem about convergence: Training results shows that the 'total loss' is fluctuate around 2 ,and did not decrease anymore; Second, the 'loss_box' always decay to zero and it will rise a little only; Third, the training is always cut off, no reason; Fourth, the candidate boxes are generally small than the fact; Fifth, when i run the demo there are no object has been detected. So, what happen to my project, please reply to me as soon as possible,thank U!

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment