Skip to content
This repository has been archived by the owner on Jul 27, 2018. It is now read-only.

Normalize option to diversity metrics from 0 to 1 #5

Open
dvro opened this issue Mar 24, 2015 · 0 comments
Open

Normalize option to diversity metrics from 0 to 1 #5

dvro opened this issue Mar 24, 2015 · 0 comments

Comments

@dvro
Copy link
Member

dvro commented Mar 24, 2015

Diversity class (https://github.com/viisar/brew/blob/master/brew/metrics/diversity/base.py) accepts several diversity metrics, but different diversity metrics have different behavior (some is from 0 to 1, being 1 the most diverse; others 1 to -1, being -1 the most diverse ...).

In this feature, we want the "normalized" option, in which all methods would return a float in the range 0 to 1, being 1 the most diverse. This will be useful for generation methods using Diversity metrics (example).

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

No branches or pull requests

1 participant