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

why predict values are strange #14

Closed
liangdashuang opened this issue Jan 7, 2018 · 3 comments
Closed

why predict values are strange #14

liangdashuang opened this issue Jan 7, 2018 · 3 comments

Comments

@liangdashuang
Copy link

liangdashuang commented Jan 7, 2018

I trained a model with my own data, the test output are as follows with "pred.py":
image

it seems there is something wrong, what's the reason?

I trained with my own data, my mask's pixel value is 0,1,2,3,4

@liangdashuang liangdashuang changed the title why predict value are not right why predict value are strange Jan 8, 2018
@liangdashuang liangdashuang changed the title why predict value are strange why predict values are strange Jan 8, 2018
@akirasosa
Copy link
Owner

akirasosa commented Jan 11, 2018

@liangdashuang I'm not sure it's wrong or not. Have you tried to draw it?

@liangdashuang
Copy link
Author

liangdashuang commented Jan 15, 2018

@akirasosa I only have 5 class(include the backgroud), the mask value is 0,1,2,3,4, so the predict output value should be 0,1,2,3,4 as well, am i right?

@akirasosa
Copy link
Owner

@liangdashuang
In this case, your output shape will be (h, w, 5). It means you have 5 channels.

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

2 participants