From 353ce5e8dfa10489bf034041edc754c85f57ba1e Mon Sep 17 00:00:00 2001
From: iqiukp <44639227+iqiukp@users.noreply.github.com>
Date: Wed, 25 Mar 2020 17:05:13 +0800
Subject: [PATCH] Update README.md
---
README.md | 169 +++++++++++++++++++++++++++++++++++++++++++++++++++++-
1 file changed, 167 insertions(+), 2 deletions(-)
diff --git a/README.md b/README.md
index 77bf4f2..102c178 100644
--- a/README.md
+++ b/README.md
@@ -1,2 +1,167 @@
-# SVDD
- Python code for abnormal detection or fault detection using Support Vector Data Description (SVDD)
+# Support Vector Data Description (SVDD)
+
+Python Code for abnormal detection or fault detection using SVDD.
+
+Version 1.0, 1-DEC-2019
+
+Email: iqiukp@outlook.com
+
+-------------------------------------------------------------------
+
+## Main features
+
+* SVDD model for training dataset containing only positive training data. (SVDD)
+* SVDD model for training dataset containing both positive training data and negative training data. (nSVDD)
+* Multiple kinds of kernel functions.
+* Visualization module including ROC curve plotting, test result plotting, and decision boundary.
+
+-------------------------------------------------------------------
+
+## Requirements
+
+* matplotlib
+* cvxopt
+* scipy
+* numpy
+* scikit_learn
+
+-------------------------------------------------------------------
+
+## About SVDD model
+
+Two types of SVDD models are built according to the following references:
+
+[1] Tax D M J, Duin R P W. Support vector data description[J]. Machine learning, 2004, 54(1): 45-66.
+
+-------------------------------------------------------------------
+
+## A simple application for decision boundary (using differnent kernel functions)
+
+```
+
+# -*- coding: utf-8 -*-
+
+import sys
+sys.path.append("..")
+from src.svdd import SVDD
+from src.visualize import Visualization as draw
+from data import PrepareData as load
+
+# load banana-shape data
+trainData, testData, trainLabel, testLabel = load.banana()
+
+
+# kernel list
+kernelList = {"1": {"type": 'gauss', "width": 1/24},
+ "2": {"type": 'linear', "offset": 0},
+ "3": {"type": 'ploy', "degree": 2, "offset": 0},
+ "4": {"type": 'tanh', "gamma": 1e-4, "offset": 0},
+ "5": {"type": 'lapl', "width": 1/12}
+ }
+
+
+for i in range(len(kernelList)):
+
+ # set SVDD parameters
+ parameters = {"positive penalty": 0.9,
+ "negative penalty": 0.8,
+ "kernel": kernelList.get(str(i+1)),
+ "option": {"display": 'on'}}
+
+ # construct an SVDD model
+ svdd = SVDD(parameters)
+
+ # train SVDD model
+ svdd.train(trainData, trainLabel)
+
+ # test SVDD model
+ distance, accuracy = svdd.test(testData, testLabel)
+
+ # visualize the results
+ # draw.testResult(svdd, distance)
+ # draw.testROC(testLabel, distance)
+ draw.boundary(svdd, trainData, trainLabel)
+
+```
+
+* gaussian kernel function
+
+
+ 
+
+
+* linear kernel function
+
+
+ 
+
+
+* polynomial kernel function
+
+
+ 
+
+
+* sigmoid kernel function
+
+
+ 
+
+
+* laplacian kernel function
+
+
+ 
+
+
+
+## A simple application for abnormal detection or fault detection
+
+```
+
+# -*- coding: utf-8 -*-
+
+import sys
+sys.path.append("..")
+from src.svdd import SVDD
+from src.visualize import Visualization as draw
+from data import PrepareData as load
+
+# load banana-shape data
+trainData, testData, trainLabel, testLabel = load.iris()
+
+# set SVDD parameters
+parameters = {"positive penalty": 0.9,
+ "negative penalty": 0.8,
+ "kernel": {"type": 'gauss', "width": 1/24},
+ "option": {"display": 'on'}}
+
+
+# construct an SVDD model
+svdd = SVDD(parameters)
+
+# train SVDD model
+svdd.train(trainData, trainLabel)
+
+
+# test SVDD model
+distance, accuracy = svdd.test(testData, testLabel)
+
+# visualize the results
+draw.testResult(svdd, distance)
+draw.testROC(testLabel, distance)
+
+```
+
+* test result
+
+
+ 
+
+
+* ROC curve
+
+
+ 
+
+