Python LIBSVM usage Summary

Python LIBSVM usage Summary

1) download the installation package python-2.7.3.msi under windows from the python official website and install

2) Open IDLE (python GUI) and enter
>>> Import sys
>>> Sys. version
If your python is 32-bit, the following characters will appear:
'2. 7.3 (default, Apr 10 2012, 23:31:26) [MSC v.1500 32 bit (Intel)]'
In this case, the python interface setting of LIBSVM is very simple. Find the dynamic link library libsvm. dll in the windows folder under the libsvm-3.16 folder and add it to the system directory, such as 'C: windowssystem32', you can use libsvm in python

3) If your python is 64-bit, that is, open IDLE (python GUI) and enter
>>> Import sys
>>> Sys. version
The following characters are displayed:
'2. 7.3 (default, Apr 10 2012, 23:24:47) [MSC v.1500 64 bit (AMD64)]'
At this time, you need to compile the 64-bit dynamic link library libsvm. dll by yourself. The method is as follows:
Find Visual Studio x64 Win64 Command Prompt (2010) in Microsoft Visual Studio 2010/Visual Studio Tools in the program list. Be sure to use the 64-bit command prompt.
Libsvm-3.16 from cd to LIBSVM folder
Enter nmake-f Makefile. win clean all
A 64-bit dynamic link library is generated under the windows directory of the libsvm-3.16. Copy the newly generated libsvm. dll to the system directory (for example, 'c: windowssystem32.

 

========================================================== ============================================
====================================================== Awesome split line ==== ======================================

========================================================== ============================================

4) Simple SVM operations are directly implemented according to this step.
Enable IDLE
>>> Import OS
>>> OS. chdir ('C: Program Fileslibsvm-3.16python ')
>>> From svmutil import *
>>> Y, x = svm_read_problem ('../heart_scale ')
>>> M = svm_train (y [: 200], x [: 200], '-c 4 ')
>>> P_label, p_acc, p_val = svm_predict (y [200:], x [200:], m)

5) python Interface Description
The python folder in the libsvm-3.16 mainly contains two files: svm. py and svmutil. py.
The svmutil. py interface mainly includes high-level functions. The usage of these functions is similar to the MATLAB interface of LIBSVM.
Svmutil mainly includes the following functions:
Svm_train (): train an SVM model
Svm_predict (): predict testing data
Svm_read_problem (): read the data from a LIBSVM-format file.
Svm_load_model (): load a LIBSVM model.
Svm_save_model (): save model to a file.
Evaluations (): evaluate prediction results.
The svm. py interface mainly includes some low-level applications. The built-in ctypes library of python is used in svm. py, so that python can directly access the C structure and interface functions defined in svm. h. Svm. py mainly uses four data structures: svm_node, svm_problem, svm_parameter, and svm_model.