PLEASD: A Matlab Toolbox for Structured Learning
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
data
doc
references
README.txt
bmrm.m
bmrmGetGradient.m
bmrmGetLoss.m
bmrmUpdateModelL2.m
colsInCell.m
getField.m
perceptron.m
perceptronGetLoss.m
println.m
rowsInCell.m
slpaGetLinearUpperBound.m
slpa_bmrm.m
slpa_perceptron.m
trackingDemo.m
trackingGetFullLoss.m
trackingJointFeature.m
trackingLoss.m
trackingPredictor.m

README.txt

[About]
PLEASD: A Matlab Toolbox for Structured Learning
PLEASD stands for Prediction and LEArning for Structured Data. It is a Matlab toolbox of algorithmic frameworks for training structured prediction models. We provide this toolbox to ease the process of applying structured learning to new problems. We attempt to minimize users’ involvement in coding the structured learning framework such that they can focus on issues related to their specific problems. Currently, PLEASD has included the following structured learning frameworks:
1. Bundle method for risk minimization (type help bmrm);
2. Structured perceptron learning (type help perceptron);
3. Structured learning from partial annotations (type help slpa_bmrm);
4. Structured perceptron learning from partial annotations (type help slpa_perceptron).

[License]
(c) MIT License for worry-free use and distribution.

[Examples and User Guide]
Demo: see trackingDemo.m for examples.
User Guide: included in the package.

[Contact]
Please forward any suggestions, bug reports, questions to xinghua.lou@gmail.com. Your feedback is highly appreciated.