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

Regarding last layer FC output #111

Open
iitbombombay opened this issue Sep 16, 2020 · 2 comments
Open

Regarding last layer FC output #111

iitbombombay opened this issue Sep 16, 2020 · 2 comments

Comments

@iitbombombay
Copy link

iitbombombay commented Sep 16, 2020

hi @jiecaoyu , I have fused Bnorm and linear of Final FC layer of Imagenet Alexnet model. after fusing final FC output values (all) negative? I have checked the prediction I am getting 42% accuracy by taking fused weights and bias.

My concern is can all values of an output can be negative?

@jiecaoyu
Copy link
Owner

@iitbombombay Algorithmically it is possible. Is it the same case for all images or only for some images?

@iitbombombay
Copy link
Author

Thanks @jiecaoyu , I took 1000 classes from Imagenet as a dataset where each class having 5 images soo in total I have evaluated 5000 images where i am getting negative values for all images

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