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.

Sign up for GitHub

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

Jump to bottom

Reproduce paper results #7

Closed
aybassiouny opened this issue Mar 22, 2015 · 1 comment
Closed

Reproduce paper results #7

aybassiouny opened this issue Mar 22, 2015 · 1 comment
Assignees
@phanein

Comments

@aybassiouny
Copy link

Hi,

I am trying to reproduce paper results at the moment. As the paper indicated, I have tried using LIBLINEAR. I could not however find exact details into how each instance is evaluated. I speculated by using n one Vs. all binary classifiers (where n is number of classes) and then sort predictions by estimate value and compare the highest m to the groundtruth (where m is the number of true classes). Still, I got macro F1 score of 16%.
@phanein
Copy link
Owner

phanein commented Mar 23, 2015

Hello @aybassiouny,

We have a scoring script available that illustrates how we scored the results.

Note: Make sure to update your copy of the repository! (We fixed a small issue with it recently.)

Let me know if this resolves your issue.

@phanein phanein self-assigned this Mar 24, 2015
@phanein phanein closed this as completed Mar 24, 2015
This issue was closed.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees

@phanein phanein

Labels
None yet
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
2 participants
@aybassiouny @phanein