Add test case for train(save_loc=)

PyThaiNLP · Aug 23, 2020 · 83eee2e · 83eee2e
1 parent ee0f5e5
commit 83eee2e
Show file tree

Hide file tree

Showing 2 changed files with 33 additions and 5 deletions.
diff --git a/pythainlp/tag/_tag_perceptron.py b/pythainlp/tag/_tag_perceptron.py
@@ -71,7 +71,7 @@ def train(
         """
         self._make_tagdict(sentences)
         self.model.classes = self.classes
-        for iter_ in range(nr_iter):
+        for _ in range(nr_iter):
             c = 0
             n = 0
             for sentence in sentences:
@@ -192,5 +192,6 @@ def _make_tagdict(self, sentences: List[List[Tuple[str, str]]]) -> None:
                 self.tagdict[word] = tag
 
 
-def _pc(n, d) -> float:
-    return (float(n) / d) * 100
+# for logging
+# def _pc(n, d) -> float:
+#     return (float(n) / d) * 100
diff --git a/tests/test_tag.py b/tests/test_tag.py
@@ -1,6 +1,7 @@
 # -*- coding: utf-8 -*-
 
 import unittest
+from os import path
 
 from pythainlp.tag import (
     PerceptronTagger,
@@ -99,11 +100,12 @@ def test_pos_tag(self):
 
     def test_perceptron_tagger(self):
         tagger = PerceptronTagger()
+        # train data, with "กิน" > 20 instances to trigger conditions in _make_tagdict()
         data = [
             [("คน", "N"), ("เดิน", "V")],
+            [("ฉัน", "N"), ("เดิน", "V")],
             [("แมว", "N"), ("เดิน", "V")],
             [("คน", "N"), ("วิ่ง", "V")],
-            [("แมว", "N"), ("วิ่ง", "V")],
             [("ปลา", "N"), ("ว่าย", "V")],
             [("นก", "N"), ("บิน", "V")],
             [("คน", "N"), ("พูด", "V")],
@@ -118,8 +120,33 @@ def test_perceptron_tagger(self):
             [("นก", "N"), ("กิน", "V"), ("ปลา", "N")],
             [("คน", "N"), ("กิน", "V"), ("กาแฟ", "N")],
             [("คน", "N"), ("คน", "V"), ("กาแฟ", "N")],
+            [("พระ", "N"), ("ฉัน", "V"), ("กาแฟ", "N")],
+            [("พระ", "N"), ("คน", "V"), ("กาแฟ", "N")],
+            [("พระ", "N"), ("ฉัน", "V"), ("ข้าว", "N")],
+            [("ฉัน", "N"), ("กิน", "V"), ("ข้าว", "N")],
+            [("เธอ", "N"), ("กิน", "V"), ("ปลา", "N")],
+            [("ปลา", "N"), ("กิน", "V"), ("แมลง", "N")],
+            [("แมวน้ำ", "N"), ("กิน", "V"), ("ปลา", "N")],
+            [("หนู", "N"), ("กิน", "V")],
+            [("เสือ", "N"), ("กิน", "V")],
+            [("ยีราฟ", "N"), ("กิน", "V")],
+            [("แรด", "N"), ("กิน", "V")],
+            [("หมู", "N"), ("กิน", "V")],
+            [("แมลง", "N"), ("กิน", "V")],
+            [("สิงโต", "N"), ("กิน", "V")],
+            [("เห็บ", "N"), ("กิน", "V")],
+            [("เหา", "N"), ("กิน", "V")],
+            [("เต่า", "N"), ("กิน", "V")],
+            [("กระต่าย", "N"), ("กิน", "V")],
+            [("จิ้งจก", "N"), ("กิน", "V")],
+            [("หมี", "N"), ("กิน", "V")],
+            [("หมา", "N"), ("กิน", "V")],
+            [("ตะพาบ", "N"), ("กิน", "V")],
+            [("เม่น", "N"), ("กิน", "V")],
+            [("หนอน", "N"), ("กิน", "V")],
         ]
-        tagger.train(data)
+        tagger.train(data, save_loc="temp.pkl")
+        self.assertTrue(path.exists("temp.pkl"))
         self.assertEqual(len(tagger.tag(["นก", "เดิน"])), 2)
 
     # ### pythainlp.tag.locations