We provide Matlab code that implements the training and evaluation of hidden-unit CRFs, as well as code to reproduce the results of our experiments. The code implements four different training algorithms: (1) a batch learner that uses L-BFGS, (2) a stochastic gradient descent learner, (3) an online perceptron training algorithm, and (4) an online large-margin perceptron algorithm. The code can also be used to perform (conditional) herding in hidden-unit CRFs.

hucrf_code
..........\.DS_Store
__MACOSX
........\hucrf_code
........\..........\._.DS_Store
hucrf_code\cb513_experiment.m
__MACOSX\hucrf_code\._cb513_experiment.m
hucrf_code\crf_grad.cpp
__MACOSX\hucrf_code\._crf_grad.cpp
hucrf_code\crf_grad.m
__MACOSX\hucrf_code\._crf_grad.m
hucrf_code\crf_herding.cpp
..........\crf_herding.m
__MACOSX\hucrf_code\._crf_herding.m
hucrf_code\crf_herding_2nd_order.cpp
..........\crf_herding_2nd_order.m
__MACOSX\hucrf_code\._crf_herding_2nd_order.m
hucrf_code\data
..........\....\.DS_Store
__MACOSX\hucrf_code\data
........\..........\....\._.DS_Store
hucrf_code\faq_experiment.m
__MACOSX\hucrf_code\._faq_experiment.m
hucrf_code\forward_backward_crf.cpp
..........\forward_backward_crf.m
__MACOSX\hucrf_code\._forward_backward_crf.m
hucrf_code\forward_backward_crf_2nd_order.m
__MACOSX\hucrf_code\._forward_backward_crf_2nd_order.m
hucrf_code\hidden_crf_herding.cpp
..........\hidden_crf_herding.m
__MACOSX\hucrf_code\._hidden_crf_herding.m
hucrf_code\hidden_crf_herding_2nd_order.cpp
..........\hidden_crf_herding_2nd_order.m
__MACOSX\hucrf_code\._hidden_crf_herding_2nd_order.m
hucrf_code\mexall.m
__MACOSX\hucrf_code\._mexall.m
hucrf_code\minFunc
..........\.......\ArmijoBacktrack.m
..........\.......\autoGrad.m
..........\.......\autoHess.m
..........\.......\autoHv.m
..........\.......\autoTensor.m
..........\.......\callOutput.m
..........\.......\conjGrad.m
..........\.......\dampedUpdate.m
..........\.......\example_minFunc.m
..........\.......\example_minFunc_LR.m
..........\.......\isLegal.m
..........\.......\lbfgs.m
..........\.......\lbfgsC.c
..........\.......\lbfgsC.mexa64
..........\.......\lbfgsC.mexglx
..........\.......\lbfgsC.mexmac
..........\.......\lbfgsC.mexmaci
..........\.......\lbfgsC.mexmaci64
..........\.......\lbfgsC.mexw32
..........\.......\lbfgsC.mexw64
..........\.......\lbfgsUpdate.m
..........\.......\logistic
..........\.......\........\LogisticDiagPrecond.m
..........\.......\........\LogisticHv.m
..........\.......\........\LogisticLoss.m
..........\.......\........\mexutil.c
..........\.......\........\mexutil.h
..........\.......\........\mylogsumexp.m
..........\.......\........\repmatC.c
..........\.......\........\repmatC.dll
..........\.......\........\repmatC.mexglx
..........\.......\........\repmatC.mexmac
..........\.......\mchol.m
..........\.......\mcholC.c
..........\.......\mcholC.mexmaci64
..........\.......\mcholC.mexw32
..........\.......\mcholC.mexw64
..........\.......\mcholinc.m
..........\.......\minFunc.m
..........\.......\minFunc_processInputOptions.m
..........\.......\polyinterp.m
..........\.......\precondDiag.m
..........\.......\precondTriu.m
..........\.......\precondTriuDiag.m
..........\.......\rosenbrock.m
..........\.......\taylorModel.m
..........\.......\WolfeLineSearch.m
..........\minimize.m
__MACOSX\hucrf_code\._minimize.m
hucrf_code\ocr_experiment.m
__MACOSX\hucrf_code\._ocr_experiment.m
hucrf_code\train_crf.m
__MACOSX\hucrf_code\._train_crf.m
hucrf_code\train_crf_sgd.m
__MACOSX\hucrf_code\._train_crf_sgd.m
hucrf_code\treebank_experiment.m
__MACOSX\hucrf_code\._treebank_experiment.m
hucrf_code\util.cpp
..........\util.h
..........\viterbi_crf.cpp
__MACOSX\hucrf_code\._viterbi_crf.cpp
hucrf_code\viterbi_crf.m
__MACOSX\hucrf_code\._viterbi_crf.m

• We are an exchange download platform that only provides communication channels. The downloaded content comes from the internet. Except for download issues, please Google on your own.
• The downloaded content is provided for members to upload. If it unintentionally infringes on your copyright, please contact us.
• Please use Winrar for decompression tools
• If download fail, Try it againg or Feedback to us.
• If downloaded content did not match the introduction, Feedback to us，Confirm and will be refund.
• Before downloading, you can inquire through the uploaded person information

Nothing．