質問編集履歴
2
markdownの変更
test
CHANGED
|
File without changes
|
test
CHANGED
|
@@ -10,7 +10,7 @@
|
|
|
10
10
|
|
|
11
11
|
|
|
12
12
|
|
|
13
|
-
|
|
13
|
+
```
|
|
14
14
|
|
|
15
15
|
from sklearn.pipeline import Pipeline
|
|
16
16
|
|
|
@@ -76,4 +76,4 @@
|
|
|
76
76
|
|
|
77
77
|
|
|
78
78
|
|
|
79
|
-
|
|
79
|
+
```
|
1
sdfs
test
CHANGED
|
File without changes
|
test
CHANGED
|
@@ -3,3 +3,77 @@
|
|
|
3
3
|
|
|
4
4
|
|
|
5
5
|
アウトオブコアの実装内容が理解できなかったのですが、なぜこのように高速に学習を終えるのかを教えてください。
|
|
6
|
+
|
|
7
|
+
|
|
8
|
+
|
|
9
|
+
|
|
10
|
+
|
|
11
|
+
|
|
12
|
+
|
|
13
|
+
'''
|
|
14
|
+
|
|
15
|
+
from sklearn.pipeline import Pipeline
|
|
16
|
+
|
|
17
|
+
from sklearn.linear_model import LogisticRegression
|
|
18
|
+
|
|
19
|
+
from sklearn.feature_extraction.text import TfidfVectorizer
|
|
20
|
+
|
|
21
|
+
from sklearn.model_selection import GridSearchCV
|
|
22
|
+
|
|
23
|
+
|
|
24
|
+
|
|
25
|
+
tfidf = TfidfVectorizer(strip_accents=None,
|
|
26
|
+
|
|
27
|
+
lowercase=False,
|
|
28
|
+
|
|
29
|
+
preprocessor=None)
|
|
30
|
+
|
|
31
|
+
|
|
32
|
+
|
|
33
|
+
param_grid = [{'vect__ngram_range': [(1, 1)],
|
|
34
|
+
|
|
35
|
+
'vect__stop_words': [stop, None],
|
|
36
|
+
|
|
37
|
+
'vect__tokenizer': [tokenizer, tokenizer_porter],
|
|
38
|
+
|
|
39
|
+
'clf__penalty': ['l1', 'l2'],
|
|
40
|
+
|
|
41
|
+
'clf__C': [1.0, 10.0, 100.0]},
|
|
42
|
+
|
|
43
|
+
{'vect__ngram_range': [(1, 1)],
|
|
44
|
+
|
|
45
|
+
'vect__stop_words': [stop, None],
|
|
46
|
+
|
|
47
|
+
'vect__tokenizer': [tokenizer, tokenizer_porter],
|
|
48
|
+
|
|
49
|
+
'vect__use_idf':[False],
|
|
50
|
+
|
|
51
|
+
'vect__norm':[None],
|
|
52
|
+
|
|
53
|
+
'clf__penalty': ['l1', 'l2'],
|
|
54
|
+
|
|
55
|
+
'clf__C': [1.0, 10.0, 100.0]},
|
|
56
|
+
|
|
57
|
+
]
|
|
58
|
+
|
|
59
|
+
|
|
60
|
+
|
|
61
|
+
lr_tfidf = Pipeline([('vect', tfidf),
|
|
62
|
+
|
|
63
|
+
('clf', LogisticRegression(random_state=0))])
|
|
64
|
+
|
|
65
|
+
|
|
66
|
+
|
|
67
|
+
gs_lr_tfidf = GridSearchCV(lr_tfidf, param_grid,
|
|
68
|
+
|
|
69
|
+
scoring='accuracy',
|
|
70
|
+
|
|
71
|
+
cv=5,
|
|
72
|
+
|
|
73
|
+
verbose=1,
|
|
74
|
+
|
|
75
|
+
n_jobs=-1)
|
|
76
|
+
|
|
77
|
+
|
|
78
|
+
|
|
79
|
+
'''
|