Update NER

PyThaiNLP · Sep 17, 2022 · 4c69dd4 · 4c69dd4
1 parent e2a3404
commit 4c69dd4
Show file tree

Hide file tree

Showing 4 changed files with 20 additions and 11 deletions.
diff --git a/pythainlp/corpus/default_db.json b/pythainlp/corpus/default_db.json
@@ -1,21 +1,27 @@
 {
     "thainer": {
         "name": "thainer",
-        "latest_version": "1.5.1",
+        "latest_version": "1.5.2",
         "description": "Thai Named Entity Recognition",
         "long_description": "Thai Named Entity Recognition",
         "url": "https://github.com/wannaphong/thai-ner/",
         "authors": [
             "Wannaphong Phatthiyaphaibun"
         ],
-        "author_email": "wannaphong@kkumail.com",
+        "author_email": "wannaphong@yahoo.com",
         "license": "cc-by-4.0",
         "versions": {
             "1.5.1": {
-                "filename": "thainer_crf_1_5_1.model",
+                "filename": "thainer_crf_1_5_2.model",
                 "download_url": "https://github.com/wannaphong/thai-ner/releases/download/1.5/thai-ner-1-5-newmm-lst20.crfsuite",
                 "md5": "-",
                 "pythainlp_version": ">=2.2.7"
+            },
+            "1.5.2": {
+                "filename": "thainer_crf_1_5_2.model",
+                "download_url": "https://github.com/wannaphong/thai-ner/releases/download/1.5.2/thainer_crf_1_5_2.model",
+                "md5": "-",
+                "pythainlp_version": ">=3.1.0"
             }
         }
     }

diff --git a/pythainlp/corpus/thainer_crf_1_5_2.model b/pythainlp/corpus/thainer_crf_1_5_2.model
diff --git a/pythainlp/tag/named_entity.py b/pythainlp/tag/named_entity.py
@@ -20,7 +20,7 @@ class NER:
         * *tltk* - wrapper for `TLTK <https://pypi.org/project/tltk/>`_.
 
     **Options for corpus**
-        * *thaimer* - Thai NER corpus
+        * *thainer* - Thai NER corpus
         * *lst20* - lst20 corpus (wangchanberta only). \
             `LST20 <https://aiforthai.in.th/corpus.php>`_ corpus \
             by National Electronics and Computer Technology Center, Thailand \

diff --git a/pythainlp/tag/thainer.py b/pythainlp/tag/thainer.py
@@ -76,27 +76,27 @@ class ThaiNameTagger:
     """
     Thai named-entity recognizer.
     :param str version: Thai NER version.
-        It's support Thai NER 1.4 & 1.5.
-        The defualt value is `1.5`
+        It's support Thai NER 1.4, 1.5, 1.5.1 and 1.5.2.
+        The defualt value is `1.5.2`
 
     :Example:
     ::
 
         from pythainlp.tag.named_entity import ThaiNameTagger
 
-        thainer15 = ThaiNameTagger(version="1.5")
+        thainer15 = ThaiNameTagger(version="1.5.2")
         thainer15.get_ner("วันที่ 15 ก.ย. 61 ทดสอบระบบเวลา 14:49 น.")
 
         thainer14 = ThaiNameTagger(version="1.4")
         thainer14.get_ner("วันที่ 15 ก.ย. 61 ทดสอบระบบเวลา 14:49 น.")
     """
-    def __init__(self, version: str = "1.5") -> None:
+    def __init__(self, version: str = "1.5.2") -> None:
         """
         Thai named-entity recognizer.
 
         :param str version: Thai NER version.
-                            It's support Thai NER 1.4 & 1.5.
-                            The defualt value is `1.5`
+                            It's support Thai NER 1.4, 1.5, 1.5.1 and 1.5.2.
+                            The defualt value is `1.5.2`
         """
         self.crf = CRFTagger()
 
@@ -105,7 +105,10 @@ def __init__(self, version: str = "1.5") -> None:
             self.pos_tag_name = "orchid_ud"
         else:
             self.crf.open(get_corpus_path(_CORPUS_NAME, version="1.5"))
-            self.pos_tag_name = "lst20"
+            if version=="1.5" or version=="1.5.1":
+                self.pos_tag_name = "lst20"
+            else:
+                self.pos_tag_name = "orchid"
 
     def get_ner(
         self, text: str, pos: bool = True, tag: bool = False