bugfix:修复上次更新引入bug.AlphabetFactory在linear类的继承问题

Author: xpqiu

fnlp-core/src/main/java/org/fnlp/ml/classifier/linear/Linear.java

@@ -51,7 +51,9 @@ public class Linear extends AbstractClassifier implements Serializable	{
 
 	private static final long serialVersionUID = -2626247109469506636L;
 
-	protected Inferencer inferencer;
+	protected Inferencer inferencer;
+	
+	protected AlphabetFactory factory;
 
 	protected Pipe pipe;
 
@@ -141,6 +143,10 @@ public void setPipe(Pipe pipe) {
 	}
 	public Pipe getPipe() {
 		return pipe;		
+	}
+	
+	public AlphabetFactory getAlphabetFactory() {
+		return factory;
 	}
 
 }

fnlp-core/src/main/java/org/fnlp/nlp/pipe/NGram.java

@@ -169,7 +169,8 @@ public static Set<String> ngramOnCharacter2Set(String data,int[] gramSizes) {
      * @param list
      */
     private static void ngramOnCharacter(String data, int[] gramSizes, Collection<String> list) {
-        for (int j = 0; j < gramSizes.length; j++) {
+    	data = data.replaceAll("\\s+", "");
+    	for (int j = 0; j < gramSizes.length; j++) {
             int len = gramSizes[j];
             if (len <= 0 || len > data.length())
                 continue;

fnlp-demo/src/main/java/org/fnlp/demo/nlp/tc/TextClassificationCustom1.java

@@ -59,7 +59,7 @@ public static void main(String[] args) throws Exception {
 		AlphabetFactory af = AlphabetFactory.buildFactory();
 
 		//使用n元特征
-		Pipe ngrampp = new NGram(new int[] {1});
+		Pipe ngrampp = new NGram(new int[] {1,2},true);
 		//将字符特征转换成字典索引
 		Pipe indexpp = new StringArray2IndexArray(af);
 		//将目标值对应的索引号作为类别
@@ -88,7 +88,7 @@ public static void main(String[] args) throws Exception {
 		 * 建立分类器
 		 */		
 		OnlineTrainer trainer = new OnlineTrainer(af,50);
-		Linear pclassifier = trainer.train(trainset,testset);
+		Linear pclassifier = trainer.train(trainset);
 		pp.removeTargetPipe();
 		pclassifier.setPipe(pp);
 		af.setStopIncrement(true);