📄 libtinysvm.html
字号:
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN" "http://www.w3.org/TR/html4/strict.dtd"><html><head><link rev=made href="mailto:taku-ku@is.aist-nara.ac.jp"><title>libtinysvm Reference Manual</title><link rel=stylesheet href="tinysvm.css"></head><body bgcolor="#ffffff"><h1>libtinysvm Reference Manual</h1><p>Last update: $Date: 2001/09/04 16:10:56 $</p><hr><h2>List of Contents</h2><ul><li><a href="#class">Class Reference Manual</a> <ul> <li><a href="#param_class">TinySVM::Param class</a> <li><a href="#base_example_class">TinySVM::BaseExample class</a> <li><a href="#example_class">TinySVM::Example class</a> <li><a href="#model_class">TinySVM::Model class</a> </ul><li><a href="#sample">Sample Programming</a><li><a href="#compile">How to compile</a></ul><h2><a name="class">Class Reference Manual</a></h2><ul><pre><a name="param_class"></a>namespace TinySVM { <font width=bold>class Param</font> { public: int set (int, char **); int set (const char *); };};</pre><dl><dt><code><font color=#dc143c>TinySVM::Param</font>::<font color=DarkBlue>set</font> (int argc, char **argv)</code><dd>Set parameter with <i>argc, argv</i> style, --- arguments for<i>main()</i> function. <br>This function accepts same parameterused for <a href="svm_learn.html">svm_learn</a>.<dt><code><font color=#dc143c>TinySVM::Param</font>::<font color=DarkBlue>set</font> (const char *s)</code><dd>Set parameter with string <i>s</i>, e.g.<i>"-t 1 -d 3 -c1"</i>.<br> This function accepts same parameterused for <a href="svm_learn.html">svm_learn</a>.</dl><pre><a name="base_example_class"></a>namespace TinySVM { <font width=bold>class BaseExample</font> { public: int add (const double, feature_node *); int add (const double, const char *); int add (const char *); int set (int i, const double, feature_node *); int set (int i, const double, const char *); int set (int i, const char *); int remove (int i); int clear (); int size (); virtual int read (const char *, const char *mode = "r", const int offset = 0) = 0; virtual int write (const char *, const char *mode = "w", const int offset = 0) = 0; int readSVindex (const char *, const char *mode = "r", const int offset = 0); int writeSVindex (const char *, const char *mode = "w", const int offset = 0); };};</pre><p><i>TinySVM::BaseExample</i> is a base class for <i>TinySVM::Example</i> and <i>TinySVM::Model</i>. You need not use this class directly.</p><dl><dt><code><font color=#dc143c>TinySVM::BaseExample</font>::<font color=DarkBlue>add</font> (const double y, feature_node *x)</code><dd>Add new example <i>y</i> and <i>x</i> into current instance.<br>Return 1 if success, return 0 if fail.<dt><code><font color=#dc143c>TinySVM::BaseExample</font>::<font color=DarkBlue>add</font> (const double y, char *x)</code><dd>Add new example <i>y</i> and <i>x</i> into current instance.x is represented with a string , e.g. <i>"1:2 3:4 5:6"</i>. <br>Return 1 if success, return 0 if fail.<dt><code><font color=#dc143c>TinySVM::BaseExample</font>::<font color=DarkBlue>add</font> (char *s)</code><dd>Add new example represented with single string, e.g. <i>"+1 1:2 3:4 5:1"</i> into current instance.<br>Return 1 if success, return 0 if fail.<dt><code><font color=#dc143c>TinySVM::BaseExample</font>::<font color=DarkBlue>set</font>(int i const double y, feature_node *x)</code><dd>Replace current i-th example with new example <i>y</i> and <i>x</i>.<br>Return 1 if success, return 0 if fail.<dt><code><font color=#dc143c>TinySVM::BaseExample</font>::<font color=DarkBlue>set</font> (int i const double y, char *x)</code><dd>Replace current i-th example with new example <i>y</i> and <i>x</i> .<br>x is represented with a string, (e.g. <i>"1:2 3:4 5:6"</i>). <br>Return 1 if success, return 0 if fail.<dt><code><font color=#dc143c>TinySVM::BaseExample</font>::<font color=DarkBlue>set</font> (int i char *s)</code><dd>Replace current i-th example with new example represented with singlestring s, (e.g. <i>"+1 1:2 3:4 5:1"</i>).<br>Return 1 if success, return 0 if fail.<dt><code><font color=#dc143c>TinySVM::BaseExample</font>::<font color=DarkBlue>remove</font> (int i)</code><dd>Remove i-th example (pair of <i>y</i> and <i>x</i>) from current instance.<dt><code><font color=#dc143c>TinySVM::BaseExample</font>::<font color=DarkBlue>clear</font> ()</code><dd>Clear all examples (pair of <i>y</i> and <i>x</i>) stored in current instance.<dt><code><font color=#dc143c>TinySVM::BaseExample</font>::<font color=DarkBlue>size</font> ()</code><dd>Get the size of examples (pair of <i>y</i> and <i>x</i>).<dt><code><font color=#dc143c>TinySVM::BaseExample</font>::<font color=DarkBlue>read</font> (const char *file_name, const char *mode ="r", const int offset = 0)</code><dd>Read all examples from <i>file_name</i>.<i>mode</i> is used for calling <i>fopen (3)</i>.<br>if <i>offset</i> is given, seek <i>offset</i> byte using <i>fseek (3)</i> before reading examples.<br>Return 1 if success, return 0 if fail.<dt><code><font color=#dc143c>TinySVM::BaseExample</font>::<font color=DarkBlue>write</font> (const char *file_name, const char *mode ="w", const int offset = 0)</code><dd>Save all examples into <i>file_name</i>. <i>mode</i> is used for calling fopen(3). <br>By setting <i>"a"</i> to <i>mode</i>, you can append examples into <i>file_name</i>.<br><i>offset</i> is reserved argument for the future. <br>Return 1 if success, return 0 if fail.<dt><code><font color=#dc143c>TinySVM::BaseExample</font>::<font color=DarkBlue>writeSVindex</font> (const char *file_name const char *mode = "w", const int offset = 0)</code><dd></dt>Read pairs of alpha and gradiant from <i>file_name</i>.<i>mode</i> is used for calling <i>fopen (3)</i>.<br>if <i>offset</i> is given, seek <i>offset</i> byte using <i>fseek (3)</i> before reading examples.<br>Return 1 if success, return 0 if fail.<dt><code><font color=#dc143c>TinySVM::BaseExample</font>::<font color=DarkBlue>readSVindex</font> (const char *file_name const char *mode = "w", const int offset = 0)</code><dd></dt>Save pairs of alpha and gradiant into <i>file_name</i>.<i>mode</i> is used for calling <i>fopen (3)</i>.<br>if <i>offset</i> is given, seek <i>offset</i> byte using <i>fseek (3)</i> before reading examples.<br>Return 1 if success, return 0 if fail.</dl><pre><a name="example_class"></a>namespace TinySVM { <font width=bold>class Example</font>: public BaseExample { public: Model *learn (const Param &); Model *learn (const char *); int rebuildSVindex (Model *); };};</pre><p><i>TinySVM::Example</i> class inherits it's interface from <i>TinySVM::BaseExample</i> class. You can use all member functions defined in <i>TinySVM::BaseExample</i> class.</p><dl><dt><code><font color=#dc143c>TinySVM::Example</font>::<font color=DarkBlue>learn </font> (const TinySVM::Param &p)</code><dd>Execute learning with <i>TinySVM::Param</i> object <i>p</i>. <br>Return new instance (pointer) of <i>TinySVM::Model</i> object.<dt><code><font color=#dc143c>TinySVM::Example</font>::<font color=DarkBlue>learn </font> (const char *s)</code><dd>Execute learning with Parameter represented with single string<i>s</i>, e.g. <i>"-t 1 -d 2 -c 1"</i>. <br>Return new instance (pointer) of <i>TinySVM::Model</i> object.<dt><code><font color=#dc143c>TinySVM::Example</font>::<font color=DarkBlue>rebuldSVindex </font> (const TinySVM::Model *m)</code><dd>Rebuild alpha and gradient for all examples using Model <i>m</i>.</dl><pre><a name="model_class"></a>namespace TinySVM {<font width=bold>class Model</font>: public BaseExample { public: Model (); Model (const Param &); double classify (const char *); double classify (const feature_node *); double estimateMargin (); double estimateVC (); double estimateXA (const double rho = 2.0); double estimateSphere (); double getLoss (); int compress (); int getSVnum (); int getBSVnum(); int getTrainingDataSize(); };};</pre><p><i>TinySVM::Model</i> class inherits it's interface from <i>TinySVM::BaseExample</i> class. You can use all member functions defined in <i>TinySVM::BaseExample</i> class.</p><dl><dt><code><font color=#dc143c>TinySVM::Model</font>::<font color=DarkBlue>Model</font> (const TinySVM::Param &p)</code><dd>Create new <i>TinySVM::Model</i> instance with <i>TinySVM::Param</i> object <i>p</i>.<dt><code><font color=#dc143c>TinySVM::Model</font>::<font color=DarkBlue><font color=DarkBlue>classify</font> </font> (const char *s)</code><dd>Classify test example represented with single string <i>s</i>, e.g. <i>"1:1 2:1 3:1"</i>.<br>Return distance from the decision hyperplane. <dt><code><font color=#dc143c>TinySVM::Model</font>::<font color=DarkBlue>classify </font> (const feature_node *f)</code><dd>Classify test example represented with feature_node "f".<br>Return distance from the decision hyper plane. <dt><code><font color=#dc143c>TinySVM::Model</font>::<font color=DarkBlue>estimateMargin</font> ()</code><dd>Return the estimated margin value<dt><code><font color=#dc143c>TinySVM::Model</font>::<font color=DarkBlue>estimateVC</font> ()</code><dd>Return the estimated VC dimension. This value is roughly approximated.<dt><code><font color=#dc143c>TinySVM::Model</font>::<font color=DarkBlue>estimateXA</font>(const double rho = 2.0 )</code><dd>Return the Xi-Alpha estimation value, which is known as a tighter bound forLeave-One-Out bound, with parameter rho.<dt><code><font color=#dc143c>TinySVM::Model</font>::<font color=DarkBlue>estimateSphere</font> ()</code><dd>Return the minimum diameter of sphere which encloses the all examples stored in current model.This value is roughly approximated.<dt><code><font color=#dc143c>TinySVM::Model</font>::<font color=DarkBlue>getSVnum</font> ()</code><dd>Return the number of Support Vectors.<dt><code><font color=#dc143c>TinySVM::Model</font>::<font color=DarkBlue>getBSVnum</font> ()</code><dd>Return the number of bounded Support Vectors, -- the number of examples whose alpha is C.<dt><code><font color=#dc143c>TinySVM::Model</font>::<font color=DarkBlue>getTrainingDataSize</font> ()</code><dd>Return the size of training data.<dt><code><font color=#dc143c>TinySVM::Model</font>::<font color=DarkBlue>getLoss</font> ()</code><dd>Return the 1-norm loss evaluated with training data. This value is useally mentioned as "Empirical Risk".The case of C-SVR, this function returns the "1-norm epsilon-insenstiveloss" evaluated with given training data.<dt><code><font color=#dc143c>TinySVM::Model</font>::<font color=DarkBlue>compress</font> ()</code><dd>Replace the current representation of model instance (lists of alpha and trainingdata) with single vector w (w * x + b). </ul></dl><h2><a name="sample">Sample Programming</a></h2><ul><li>Learn with <i>file-name</i>, and store the result into <i>model-file</i>.<pre>#include <tinysvm.h> TinySVM::Example e; TinySVM::Param p; if (! e.read(file-name)) throw; if (! p.set("-t 1 -d 2 -c 1")) throw; TinySVM::Model *m = e.learn(p); m->write(model-file); delete m;</pre><li>Classify using stored <i>model-file</i><pre>#include <tinysvm.h> TinySVM::Model m; if (! m.read(model-file)) throw; cout << m.classify("1:1 2:1 3:1 4:1") << endl;</pre><li>svm_learn, svm_classify and svm_model, all of these included program arewritten with libtinysvm. <br>Tese sources would also help you to build your own programs.</ul><h2><a name="compile">How to compile</a></h2><ul><pre>c++ -I/usr/local/include -L/usr/local/lib sample.cc -o sample -ltinysvm</pre></ul><hr><p>$Id: libtinysvm.html,v 1.11 2001/09/04 16:10:56 taku-ku Exp $;</p><address>taku-ku@is.aist-nara.ac.jp</address></body></html>⌨️ 快捷键说明
复制代码
Ctrl + C
搜索代码
Ctrl + F
全屏模式
F11
切换主题
Ctrl + Shift + D
显示快捷键
?
增大字号
Ctrl + =
减小字号
Ctrl + -