質問編集履歴
2
誤字
test
CHANGED
|
@@ -1 +1 @@
|
|
|
1
|
-
UnicodeDecodeErrorの直し方知りたい
|
|
1
|
+
UnicodeDecodeErrorの直し方を知りたい
|
test
CHANGED
|
File without changes
|
1
コード内容の変更
test
CHANGED
|
File without changes
|
test
CHANGED
|
@@ -1,36 +1,108 @@
|
|
|
1
1
|
#知りたいこと
|
|
2
2
|
|
|
3
|
-
PythonでKNPを使って構文解析をやりたいのですが、以下のコードでエラーが出ます。参考にした[サイト](https://
|
|
3
|
+
PythonでKNPを使って構文解析をやりたいのですが、以下のコードでエラーが出ます。参考にした[サイト](https://www.trifields.jp/how-to-select-dependency-structure-by-knp-in-ubuntu-1404-and-python-3-1393)通りなのですが、何が悪いのでしょうか。
|
|
4
4
|
|
|
5
5
|
```python
|
|
6
6
|
|
|
7
|
-
import
|
|
7
|
+
from pyknp import KNP
|
|
8
8
|
|
|
9
|
-
knp = pyknp.KNP()
|
|
10
9
|
|
|
11
|
-
knp = pyknp.KNP(jumanpp=True)
|
|
12
10
|
|
|
13
|
-
|
|
11
|
+
def select_normalization_representative_notation(fstring):
|
|
14
12
|
|
|
13
|
+
""" 正規化代表表記を抽出します
|
|
14
|
+
|
|
15
|
+
"""
|
|
16
|
+
|
|
17
|
+
begin = fstring.find('正規化代表表記:')
|
|
18
|
+
|
|
19
|
+
end = fstring.find('/', begin + 1)
|
|
20
|
+
|
|
21
|
+
return fstring[begin + len('正規化代表表記:') : end]
|
|
22
|
+
|
|
23
|
+
|
|
24
|
+
|
|
25
|
+
def select_dependency_structure(line):
|
|
26
|
+
|
|
27
|
+
"""係り受け構造を抽出します
|
|
28
|
+
|
|
29
|
+
"""
|
|
30
|
+
|
|
31
|
+
|
|
32
|
+
|
|
33
|
+
# KNP
|
|
34
|
+
|
|
35
|
+
knp = KNP(option = '-tab -anaphora')
|
|
36
|
+
|
|
37
|
+
|
|
38
|
+
|
|
39
|
+
# 解析
|
|
40
|
+
|
|
15
|
-
result.
|
|
41
|
+
result = knp.parse(line)
|
|
42
|
+
|
|
43
|
+
|
|
44
|
+
|
|
45
|
+
# 文節リスト
|
|
46
|
+
|
|
47
|
+
bnst_list = result.bnst_list()
|
|
48
|
+
|
|
49
|
+
|
|
50
|
+
|
|
51
|
+
# 文節リストをidによるディクショナリ化する
|
|
52
|
+
|
|
53
|
+
bnst_dic = dict((x.bnst_id, x) for x in bnst_list)
|
|
54
|
+
|
|
55
|
+
|
|
56
|
+
|
|
57
|
+
tuples = []
|
|
58
|
+
|
|
59
|
+
for bnst in bnst_list:
|
|
60
|
+
|
|
61
|
+
if bnst.parent_id != -1:
|
|
62
|
+
|
|
63
|
+
# (from, to)
|
|
64
|
+
|
|
65
|
+
tuples.append((select_normalization_representative_notation(bnst.fstring), select_normalization_representative_notation(bnst_dic[bnst.parent_id].fstring)))
|
|
66
|
+
|
|
67
|
+
|
|
68
|
+
|
|
69
|
+
return tuples
|
|
70
|
+
|
|
71
|
+
|
|
72
|
+
|
|
73
|
+
|
|
74
|
+
|
|
75
|
+
if __name__ == '__main__' :
|
|
76
|
+
|
|
77
|
+
line = '太郎は花子が読んでいる本を次郎に渡した'
|
|
78
|
+
|
|
79
|
+
tuples = select_dependency_structure(line)
|
|
80
|
+
|
|
81
|
+
for t in tuples:
|
|
82
|
+
|
|
83
|
+
print(t[0] + ' => ' + t[1])
|
|
16
84
|
|
|
17
85
|
|
|
18
86
|
|
|
19
87
|
エラー内容
|
|
20
88
|
|
|
21
|
-
File "C:/Users/proto/PycharmProjects/cutyasai_1/koubun.py", line
|
|
89
|
+
File "C:/Users/proto/PycharmProjects/cutyasai_1/koubun.py", line 37, in <module>
|
|
22
90
|
|
|
23
|
-
|
|
91
|
+
tuples = select_dependency_structure(line)
|
|
24
92
|
|
|
93
|
+
File "C:/Users/proto/PycharmProjects/cutyasai_1/koubun.py", line 18, in select_dependency_structure
|
|
94
|
+
|
|
95
|
+
result = knp.parse(line)
|
|
96
|
+
|
|
25
|
-
File "C:\Users\proto\venv\lib\site-packages\pyknp\knp\knp.py", line 63, in parse
|
|
97
|
+
File "C:\Users\proto\venv\lib\site-packages\pyknp\knp\knp.py", line 63, in parse
|
|
26
98
|
|
|
27
99
|
juman_lines = self.juman.juman_lines(sentence)
|
|
28
100
|
|
|
29
|
-
File "C:\Users\proto\venv\lib\site-packages\pyknp\juman\juman.py", line 70, in juman_lines
|
|
101
|
+
File "C:\Users\proto\venv\lib\site-packages\pyknp\juman\juman.py", line 70, in juman_lines
|
|
30
102
|
|
|
31
103
|
return self.subprocess.query(input_str, pattern=self.pattern)
|
|
32
104
|
|
|
33
|
-
File "C:\Users\proto\venv\lib\site-packages\pyknp\juman\process.py", line 71, in query
|
|
105
|
+
File "C:\Users\proto\venv\lib\site-packages\pyknp\juman\process.py", line 71, in query
|
|
34
106
|
|
|
35
107
|
line = self.stdouterr.readline()[:-1].decode('utf-8')
|
|
36
108
|
|