📄 instructions.htm
字号:
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN"><html><head><title>Instructions</title><meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1"><style type="text/css">body { margin: 50px; font-family: Arial, Helvetica, sans-serif; font-size: 10pt; line-height: 25px; }.titletext { font-family: Arial, Helvetica, sans-serif; font-size: 18pt;}.section { font-family: Arial, Helvetica, sans-serif; font-size: 12pt; font-weight: bold;}</style></head><body><div align="center"> <p><font class="titletext">Instructions for Execution</font> </p> <hr align="center" size="1" width="40%"> <p>Not as easy as you'd think. </p></div><p><font class="section">Known method</font> </p><p>This k-nearest-neighbor classifier program was written in Netbeans to save time on GUI design, however, I have yet to find a way to execute the program outside of Netbeans. Typically, one must load all the source files in a single project, compile, and execute using netbeans.</p><p>Whether or not it will function when compiled with javac is unknown, because I was unable to test it on a university computer due to disk space requirements of the Weka library that this program needs.</p><p>Mainly tested with the segment-challenge.arff and segment-test.arff datasets included with the Weka distribution. </p><p><font class="section">Relevant Libraries</font> </p><p>The program needs certain libraries to compile.</p><ul> <li>Java default (I used 1.4.2, but later versions may work also).</li> <li>Weka,weka.jar in distribution.</li> <li>Absolute Layout - AbsoluteLayout.jar org.netbeans.lib.awtextra, was necessary in order to avoid needless tweeking of the GUI.</li></ul><p> </p><div align="center"> <p><font class="titletext">Instructions for operation</font></p> <hr align="center" size="1" width="40%"></div><p>Once you have the program running, the rest is straight-forward. There are 2 tabs, one for loading data sets, the other for setting classifiers. </p><p><strong>Load training set</strong></p><p>Load the training set (ARFF format).</p><p><strong>Load test set</strong></p><p> Loas the test set (also ARFF format).</p><p> </p><p>In the second tab, one can choose the classifier, and which dataset to run the test on. The results are displayed in a box on the side.</p><p><strong>Classifier</strong></p><p>Simple k-nearest-neighbor - working, single threaded unfortunately.</p><p>knn partial distance - presently broken, I would like to say that it will be fixed in a future version, but that would be a flat-out lie. It's been a nightmare from hell to get this far with it.</p><p><strong>Settings</strong></p><p>Value of k - essential to the knn, specifies how many nearest neighbors to consider in classification.</p><p>Run test on test set or training set.</p><p>Save results to file - saves the results to a file in the same directory named results.txt</p><p><strong>Start Button</strong></p><p>Begins the classification of the dataset with the specified parameters.</p></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 + -