Skip to content
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

Evaluation on own datasets #716

Open
Mayankjha1535 opened this issue Jul 23, 2019 · 4 comments
Open

Evaluation on own datasets #716

Mayankjha1535 opened this issue Jul 23, 2019 · 4 comments

Comments

@Mayankjha1535
Copy link

How to evaluate on our own datasets ?

@junyanz
Copy link
Owner

junyanz commented Jul 23, 2019

The evaluation depends on your specific task and application. There is no single metric that works for all the datasets and tasks.

@Mayankjha1535
Copy link
Author

Actually, I have trained the model on my own datasets and Day to night dataset. I want to evaluate it on Mean IOU, Mean pixel accuracy and Mean class accuracy. So, how to evaluate these on my datasets ?

@junyanz
Copy link
Owner

junyanz commented Jul 24, 2019

MeanIOU is only useful for the cityscape dataset and other segmentation datasets. It cannot be used in Day2night dataset. You need to design a new metric for your own dataset.

@diaosiji
Copy link

MeanIOU is only useful for the cityscape dataset and other segmentation datasets. It cannot be used in Day2night dataset. You need to design a new metric for your own dataset.

Yes, it depends on your specific task. You can find some paper in the area which your own data maybe related to, then you may find some metric,and could compare with Other one.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

3 participants