Skip to content
/ Support-Vector-machine-with-Truncated-Pinball-loss-for-Multi-class-probability-estimation Public

the code for the paper “ support vector machine with Truncated Pinball loss for Multi-class probability estimation”

3 stars 1 fork Branches Tags Activity
Star
Notifications You must be signed in to change notification settings

hyliush/Support-Vector-machine-with-Truncated-Pinball-loss-for-Multi-class-probability-estimation

BranchesTags

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

1 Commit
EGKL_cal.m
EGKL_cal.m
 
 
Mul2Sim.m
Mul2Sim.m
 
 
get_p_simp.m
get_p_simp.m
 
 
kernel_test.m
kernel_test.m
 
 
kernel_train.m
kernel_train.m
 
 
main_function.m
main_function.m
 
 
param_cal.m
param_cal.m
 
 
prediction.m
prediction.m
 
 
readme.txt
readme.txt
 
 
t_pin_svm.m
t_pin_svm.m
 
 

Repository files navigation

The working environment is MATLAB, to run this code, users also need to install Tomlab (https://tomopt.com/tomlab/).

The data saved in the training data should be a 2-D matrix with the first column as the classes and the next should be the variables, and the last few columns are the probability or pseudoprobability .

The data format in the test data should be a 2-D matrix which is the same as training data.

To run the code, simply run t_pin_svm.m file.

About

the code for the paper “ support vector machine with Truncated Pinball loss for Multi-class probability estimation”

Resources

Readme
Activity

Stars

3 stars

Watchers

1 watching

Forks

1 fork
Report repository

Releases

No releases published

Packages

No packages published

Languages