PyStruct implements learning for structured prediction. In particular it contains implementations of Structural SVMs (1-slack and n-slack formulations), Subgradient optimization for structural SVMs, Structured Preceptron, Latent Variable Structural SVMs
It also contains popular models to which these algorithms can be applied, for example: Multi-class SVMs (Crammer-Singer SVMs), Pairwise CRFs with shared pairwise weights, Pairwise CRFs with edge-features, Several latent variable models,
Interfaces for several inference algorithms and libraries are available, in particular OpenGM, LibDAI, AD^3, QPBO, GCO
PyStruct builds uppon cvxopt to solve the QPs in the cutting plane formulations of Structural Support Vector Machines.
- Changes to previous version:
Initial Announcement on mloss.org.
No one has posted any comments yet. Perhaps you'd like to be the first?
Leave a comment
You must be logged in to post comments.