1 | /* |
---|
2 | * This program is free software; you can redistribute it and/or modify |
---|
3 | * it under the terms of the GNU General Public License as published by |
---|
4 | * the Free Software Foundation; either version 2 of the License, or |
---|
5 | * (at your option) any later version. |
---|
6 | * |
---|
7 | * This program is distributed in the hope that it will be useful, |
---|
8 | * but WITHOUT ANY WARRANTY; without even the implied warranty of |
---|
9 | * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
---|
10 | * GNU General Public License for more details. |
---|
11 | * |
---|
12 | * You should have received a copy of the GNU General Public License |
---|
13 | * along with this program; if not, write to the Free Software |
---|
14 | * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. |
---|
15 | */ |
---|
16 | |
---|
17 | /* |
---|
18 | * HTML.java |
---|
19 | * Copyright (C) 2009 University of Waikato, Hamilton, New Zealand |
---|
20 | */ |
---|
21 | |
---|
22 | package weka.classifiers.evaluation.output.prediction; |
---|
23 | |
---|
24 | import weka.classifiers.Classifier; |
---|
25 | import weka.core.Instance; |
---|
26 | import weka.core.Utils; |
---|
27 | |
---|
28 | /** |
---|
29 | <!-- globalinfo-start --> |
---|
30 | * Outputs the predictions in HTML. |
---|
31 | * <p/> |
---|
32 | <!-- globalinfo-end --> |
---|
33 | * |
---|
34 | <!-- options-start --> |
---|
35 | * Valid options are: <p/> |
---|
36 | * |
---|
37 | * <pre> -p <range> |
---|
38 | * The range of attributes to print in addition to the classification. |
---|
39 | * (default: none)</pre> |
---|
40 | * |
---|
41 | * <pre> -distribution |
---|
42 | * Whether to turn on the output of the class distribution. |
---|
43 | * Only for nominal class attributes. |
---|
44 | * (default: off)</pre> |
---|
45 | * |
---|
46 | * <pre> -decimals <num> |
---|
47 | * The number of digits after the decimal point. |
---|
48 | * (default: 3)</pre> |
---|
49 | * |
---|
50 | * <pre> -file <path> |
---|
51 | * The file to store the output in, instead of outputting it on stdout. |
---|
52 | * Gets ignored if the supplied path is a directory. |
---|
53 | * (default: .)</pre> |
---|
54 | * |
---|
55 | * <pre> -suppress |
---|
56 | * In case the data gets stored in a file, then this flag can be used |
---|
57 | * to suppress the regular output. |
---|
58 | * (default: not suppressed)</pre> |
---|
59 | * |
---|
60 | <!-- options-end --> |
---|
61 | * |
---|
62 | * @author fracpete (fracpete at waikato dot ac dot nz) |
---|
63 | * @version $Revision: 5987 $ |
---|
64 | */ |
---|
65 | public class HTML |
---|
66 | extends AbstractOutput { |
---|
67 | |
---|
68 | /** for serialization. */ |
---|
69 | private static final long serialVersionUID = 7241252244954353300L; |
---|
70 | |
---|
71 | /** |
---|
72 | * Returns a string describing the output generator. |
---|
73 | * |
---|
74 | * @return a description suitable for |
---|
75 | * displaying in the GUI |
---|
76 | */ |
---|
77 | public String globalInfo() { |
---|
78 | return "Outputs the predictions in HTML."; |
---|
79 | } |
---|
80 | |
---|
81 | /** |
---|
82 | * Returns a short display text, to be used in comboboxes. |
---|
83 | * |
---|
84 | * @return a short display text |
---|
85 | */ |
---|
86 | public String getDisplay() { |
---|
87 | return "HTML"; |
---|
88 | } |
---|
89 | |
---|
90 | /** |
---|
91 | * Replaces certain characters with their HTML entities. |
---|
92 | * |
---|
93 | * @param s the string to process |
---|
94 | * @return the processed string |
---|
95 | */ |
---|
96 | protected String sanitize(String s) { |
---|
97 | String result; |
---|
98 | |
---|
99 | result = s; |
---|
100 | result = result.replaceAll("&", "&"); |
---|
101 | result = result.replaceAll("<", "<"); |
---|
102 | result = result.replaceAll(">", ">"); |
---|
103 | result = result.replaceAll("\"", """); |
---|
104 | |
---|
105 | return result; |
---|
106 | } |
---|
107 | |
---|
108 | /** |
---|
109 | * Performs the actual printing of the header. |
---|
110 | */ |
---|
111 | protected void doPrintHeader() { |
---|
112 | append("<html>\n"); |
---|
113 | append("<head>\n"); |
---|
114 | append("<title>Predictions for dataset " + sanitize(m_Header.relationName()) + "</title>\n"); |
---|
115 | append("</head>\n"); |
---|
116 | append("<body>\n"); |
---|
117 | append("<div align=\"center\">\n"); |
---|
118 | append("<h3>Predictions for dataset " + sanitize(m_Header.relationName()) + "</h3>\n"); |
---|
119 | append("<table border=\"1\">\n"); |
---|
120 | append("<tr>\n"); |
---|
121 | if (m_Header.classAttribute().isNominal()) |
---|
122 | if (m_OutputDistribution) |
---|
123 | append("<td>inst#</td><td>actual</td><td>predicted</td><td>error</td><td colspan=\"" + m_Header.classAttribute().numValues() + "\">distribution</td>"); |
---|
124 | else |
---|
125 | append("<td>inst#</td><td>actual</td><td>predicted</td><td>error</td><td>prediction</td>"); |
---|
126 | else |
---|
127 | append("<td>inst#</td><td>actual</td><td>predicted</td><td>error</td>"); |
---|
128 | |
---|
129 | if (m_Attributes != null) { |
---|
130 | append("<td>"); |
---|
131 | boolean first = true; |
---|
132 | for (int i = 0; i < m_Header.numAttributes(); i++) { |
---|
133 | if (i == m_Header.classIndex()) |
---|
134 | continue; |
---|
135 | |
---|
136 | if (m_Attributes.isInRange(i)) { |
---|
137 | if (!first) |
---|
138 | append("</td><td>"); |
---|
139 | append(sanitize(m_Header.attribute(i).name())); |
---|
140 | first = false; |
---|
141 | } |
---|
142 | } |
---|
143 | append("</td>"); |
---|
144 | } |
---|
145 | |
---|
146 | append("</tr>\n"); |
---|
147 | } |
---|
148 | |
---|
149 | /** |
---|
150 | * Builds a string listing the attribute values in a specified range of indices, |
---|
151 | * separated by commas and enclosed in brackets. |
---|
152 | * |
---|
153 | * @param instance the instance to print the values from |
---|
154 | * @return a string listing values of the attributes in the range |
---|
155 | */ |
---|
156 | protected String attributeValuesString(Instance instance) { |
---|
157 | StringBuffer text = new StringBuffer(); |
---|
158 | if (m_Attributes != null) { |
---|
159 | boolean firstOutput = true; |
---|
160 | m_Attributes.setUpper(instance.numAttributes() - 1); |
---|
161 | for (int i=0; i<instance.numAttributes(); i++) |
---|
162 | if (m_Attributes.isInRange(i) && i != instance.classIndex()) { |
---|
163 | if (!firstOutput) |
---|
164 | text.append("</td>"); |
---|
165 | if (m_Header.attribute(i).isNumeric()) |
---|
166 | text.append("<td align=\"right\">"); |
---|
167 | else |
---|
168 | text.append("<td>"); |
---|
169 | text.append(sanitize(instance.toString(i))); |
---|
170 | firstOutput = false; |
---|
171 | } |
---|
172 | if (!firstOutput) |
---|
173 | text.append("</td>"); |
---|
174 | } |
---|
175 | return text.toString(); |
---|
176 | } |
---|
177 | |
---|
178 | /** |
---|
179 | * Store the prediction made by the classifier as a string. |
---|
180 | * |
---|
181 | * @param classifier the classifier to use |
---|
182 | * @param inst the instance to generate text from |
---|
183 | * @param index the index in the dataset |
---|
184 | * @throws Exception if something goes wrong |
---|
185 | */ |
---|
186 | protected void doPrintClassification(Classifier classifier, Instance inst, int index) throws Exception { |
---|
187 | int prec = m_NumDecimals; |
---|
188 | |
---|
189 | Instance withMissing = (Instance)inst.copy(); |
---|
190 | withMissing.setDataset(inst.dataset()); |
---|
191 | withMissing.setMissing(withMissing.classIndex()); |
---|
192 | double predValue = classifier.classifyInstance(withMissing); |
---|
193 | |
---|
194 | // index |
---|
195 | append("<tr>"); |
---|
196 | append("<td>" + (index+1) + "</td>"); |
---|
197 | |
---|
198 | if (inst.dataset().classAttribute().isNumeric()) { |
---|
199 | // actual |
---|
200 | if (inst.classIsMissing()) |
---|
201 | append("<td align=\"right\">" + "?" + "</td>"); |
---|
202 | else |
---|
203 | append("<td align=\"right\">" + Utils.doubleToString(inst.classValue(), prec) + "</td>"); |
---|
204 | // predicted |
---|
205 | if (Utils.isMissingValue(predValue)) |
---|
206 | append("<td align=\"right\">" + "?" + "</td>"); |
---|
207 | else |
---|
208 | append("<td align=\"right\">" + Utils.doubleToString(predValue, prec) + "</td>"); |
---|
209 | // error |
---|
210 | if (Utils.isMissingValue(predValue) || inst.classIsMissing()) |
---|
211 | append("<td align=\"right\">" + "?" + "</td>"); |
---|
212 | else |
---|
213 | append("<td align=\"right\">" + Utils.doubleToString(predValue - inst.classValue(), prec) + "</td>"); |
---|
214 | } else { |
---|
215 | // actual |
---|
216 | append("<td>" + ((int) inst.classValue()+1) + ":" + sanitize(inst.toString(inst.classIndex())) + "</td>"); |
---|
217 | // predicted |
---|
218 | if (Utils.isMissingValue(predValue)) |
---|
219 | append("<td>" + "?" + "</td>"); |
---|
220 | else |
---|
221 | append("<td>" + ((int) predValue+1) + ":" + sanitize(inst.dataset().classAttribute().value((int)predValue)) + "</td>"); |
---|
222 | // error? |
---|
223 | if (!Utils.isMissingValue(predValue) && !inst.classIsMissing() && ((int) predValue+1 != (int) inst.classValue()+1)) |
---|
224 | append("<td>" + "+" + "</td>"); |
---|
225 | else |
---|
226 | append("<td>" + " " + "</td>"); |
---|
227 | // prediction/distribution |
---|
228 | if (m_OutputDistribution) { |
---|
229 | if (Utils.isMissingValue(predValue)) { |
---|
230 | append("<td>" + "?" + "</td>"); |
---|
231 | } |
---|
232 | else { |
---|
233 | append("<td align=\"right\">"); |
---|
234 | double[] dist = classifier.distributionForInstance(withMissing); |
---|
235 | for (int n = 0; n < dist.length; n++) { |
---|
236 | if (n > 0) |
---|
237 | append("</td><td align=\"right\">"); |
---|
238 | if (n == (int) predValue) |
---|
239 | append("*"); |
---|
240 | append(Utils.doubleToString(dist[n], prec)); |
---|
241 | } |
---|
242 | append("</td>"); |
---|
243 | } |
---|
244 | } |
---|
245 | else { |
---|
246 | if (Utils.isMissingValue(predValue)) |
---|
247 | append("<td align=\"right\">" + "?" + "</td>"); |
---|
248 | else |
---|
249 | append("<td align=\"right\">" + Utils.doubleToString(classifier.distributionForInstance(withMissing) [(int)predValue], prec) + "</td>"); |
---|
250 | } |
---|
251 | } |
---|
252 | |
---|
253 | // attributes |
---|
254 | append(attributeValuesString(withMissing) + "</tr>\n"); |
---|
255 | } |
---|
256 | |
---|
257 | /** |
---|
258 | * Does nothing. |
---|
259 | */ |
---|
260 | protected void doPrintFooter() { |
---|
261 | append("</table>\n"); |
---|
262 | append("</div>\n"); |
---|
263 | append("</body>\n"); |
---|
264 | append("</html>\n"); |
---|
265 | } |
---|
266 | } |
---|