A Python Implementation of GLAD
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
data made multilabel inference Nov 29, 2014
.gitignore Initial commit Nov 27, 2014
README.md improved process speed Nov 29, 2014
glad.py made multilabel inference Nov 29, 2014
glad_bin.py made multilabel inference Nov 29, 2014
glad_naive.py improved process speed Nov 29, 2014
gladtest.py made multilabel inference Nov 29, 2014

README.md

python-glad

A Python Implementation of GLAD.

GLAD

Algorithm for aggregating labels given by labeler more acculately than majority vote.

  • Whitehill, Jacob, et al. "Whose vote should count more: Optimal integration of labels from labelers of unknown expertise." Advances in neural information processing systems. 2009.

Dependency

  • numpy
  • scipy

Usage

$ python glad.py data/data.txt

Input Format

  First line:
    <numGivenLabels> <numLabelers> <numTasks> <Prior p(Z=1)>
  Following <numGivenLabels> lines:
    <taskId> <labelerId> <label:0|1>
  • The task IDs must be integers and must be 0....
  • The labeler IDs must be integers and must be 0....