test

example/ner/standard/data/README.md

@@ -0,0 +1,28 @@
+## People's Daily(人民日报) dataset
+### Task
+Named Entity Recognition
+### Description
+**Tags**: LOC(地名), ORG(机构名), PER(人名)   
+**Tag Strategy**：BIO  
+**Split**: '*space*' (北 B-LOC)  
+**Data Size**:  
+Train data set ( [example.train](example.train) ):  
+
+|句数|字符数|LOC数|ORG数|PER数|
+|:-:|:-:|:-:|:-:|:-:|
+|20864|979180|16571|9277|8144|
+
+Dev data set ( [example.dev](example.dev) ):  
+
+|句数|字符数|LOC数|ORG数|PER数|
+|:-:|:-:|:-:|:-:|:-:|
+|2318|109870|1951|984|884|
+
+Test data set ( [example.test](example.test) )
+
+|句数|字符数|LOC数|ORG数|PER数|
+|:-:|:-:|:-:|:-:|:-:|
+|4636|219197|3658|2185|1864|
+
+**Reference**:   
+<https://github.com/zjy-ucas/ChineseNER>

example/ner/standard/predict.py

@@ -133,13 +133,13 @@ class Ner:
                     tmp.append(word)
                 else:
                     wordstype = result[i-1][1][2:]
-                    tag[wordstype].append(' '.join(tmp))
+                    tag[wordstype].append(''.join(tmp))
                     tmp.clear()
                     tmp.append(word)
             elif i==len(result)-1:
                 tmp.append(word)
                 wordstype = result[i][1][2:]
-                tag[wordstype].append(' '.join(tmp))
+                tag[wordstype].append(''.join(tmp))
             else:
                 tmp.append(word)