1.install
ubuntu 14.0
1)sudo apt-get install liblinear-dev !!! liblinear1.8
2)https://github.com/cjlin1/liblinear/tree/v180 download directly liblinear 2.11
ubuntu->make get linear.o prediect.o train.o$(CXX) $${SHARED_LIB_FLAG} linear.o tron.o blas/blas.a -o liblinear.so.$(SHVER) --->to get .so
example:
program received signal SIGINT, Interrupt.
0x00007ffff6a303ac in l2r_lr_fun::Xv(double*, double*) () from /usr/lib/liblinear.so.1
www.cnblogs.com/dream-of-snail/p/3793236.html
windows ->
1). Open a dos command box and change to liblinear directory. If
environment variables of VC++ have not been set, type
""C:\Program Files (x86)\Microsoft Visual Studio 12.0\VC\bin\amd64\vcvars64.bat""
You may have to modify the above command according which version of
VC++ or where it is installed.
2). Type
nmake -f Makefile.win clean all
2)usage (1.8 1.9 2.11diffrent )
train
-s 0~7 11,12,13 ->solver
-C find best parameter c {only -s 0 2????}
-c 3 set parameter c
-v 5 nr_fold=5 {crossvaliadation->accuary on the valiation,,include train and predict; all solver ;without model }
-e
predict
-b 0 without output
-b 1 probability{only for logistic regression}
svm output
LIBSVM uses more advanced techniques for SVM probability outputs. The code is a bit complicated so we haven't decided if including it is suitable or no
maysimply deal as like
int check_probability_model(const struct model *model_)
{
return (model_->param.solver_type==L2R_LR ||
to
int check_probability_model(const struct model *model_)
{
return 1;