Skip to content

Instantly share code, notes, and snippets.

@mikaelbr
Created March 13, 2013 09:56
Show Gist options
  • Save mikaelbr/5150684 to your computer and use it in GitHub Desktop.
Save mikaelbr/5150684 to your computer and use it in GitHub Desktop.
Score output for predictions
TASK B
Comparing predictions demo.pred to gold standard twitter-dev-gold-B.tsv.
File: demo.pred -- no errors found
Prediction file: 1654 lines
1654 unique
Breakdown by class:
positive 580
neutral 891
negative 183
File: twitter-dev-gold-B.tsv -- no errors found
Gold standard: 1654 lines
1654 unique
Breakdown by class:
positive 575
neutral 739
negative 340
Confusion table:
gs \ pred| positive| negative| neutral
---------------------------------------
positive| 385| 29| 161
negative| 77| 118| 145
neutral| 118| 36| 585
Scores:
class prec recall fscore
positive (385/580) 0.6638 (385/575) 0.6696 0.6667
negative (118/183) 0.6448 (118/340) 0.3471 0.4512
neutral (585/891) 0.6566 (585/739) 0.7916 0.7178
-----------------------------------------------------------------------
average(pos and neg) 0.5590
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment