StructED 1.0.1

Structured tasks are distinctive: each task has its own measure of performance, such as the word error rate in speech recognition, the BLEU score in machine translation, the NDCG score in information retrieval, or the intersection-over-union score in visual object segmentation. StructED is a software package for learning structured prediction models with training methods that are aimed at optimizing the task measure of performance. The package was written in Java and was released under the MIT license.

Keywords: structured prediction, structured SVM, CRF, direct loss minimization, structured ramp loss, structured probit Loss, structured passive aggressive

Usage: In order to use StructED in your own project all you need to do is open new project and add StructED jar to your project’s build path. Documentations and more extensive usage examples can be found at: http://adiyoss.github.io/StructED/.

If you find our work useful please cite: [StructED: risk minimization in structured prediction] (http://www.jmlr.org/papers/volume17/15-531/15-531.pdf)

@article{adi2016structed,
  title={StructED: risk minimization in structured prediction},
  author={Adi, Yossi and Keshet, Joseph},
  journal={Journal of Machine Learning Research},
  volume={17},
  number={64},
  pages={1--5},
  year={2016}
}