Skip to content

Navigation Menu

Sign in
Appearance settings

Search code, repositories, users, issues, pull requests...

Provide feedback

We read every piece of feedback, and take your input very seriously.

Saved searches

Use saved searches to filter your results more quickly
Appearance settings

Commit aa7f3a8

Browse filesBrowse files
puneithJon Wayne Parrott
puneith
authored and
Jon Wayne Parrott
committed
Movie sample changes
1 parent a236a5a commit aa7f3a8
Copy full SHA for aa7f3a8

File tree

Expand file treeCollapse file tree

5 files changed

+48
-68
lines changed
Filter options
Expand file treeCollapse file tree

5 files changed

+48
-68
lines changed

‎language/movie_nl/main.py

Copy file name to clipboardExpand all lines: language/movie_nl/main.py
+18-52Lines changed: 18 additions & 52 deletions
Original file line numberDiff line numberDiff line change
@@ -21,6 +21,7 @@
2121

2222
from googleapiclient import discovery
2323
from googleapiclient.errors import HttpError
24+
import httplib2
2425
from oauth2client.client import GoogleCredentials
2526
import requests
2627

@@ -30,10 +31,7 @@ def analyze_document(service, document):
3031
the movie name."""
3132
logging.info('Analyzing {}'.format(document.doc_id))
3233

33-
sentences, entities = document.extract_all_sentences(service)
34-
35-
sentiments = [get_sentiment(service, sentence) for sentence in sentences]
36-
34+
sentiments, entities = document.extract_sentiment_entities(service)
3735
return sentiments, entities
3836

3937

@@ -56,62 +54,35 @@ def get_request_body(text, syntax=True, entities=True, sentiment=True):
5654
return body
5755

5856

59-
def get_sentiment(service, sentence):
60-
"""Get the sentence-level sentiment."""
61-
body = get_request_body(
62-
sentence, syntax=False, entities=True, sentiment=True)
63-
64-
docs = service.documents()
65-
request = docs.annotateText(body=body)
66-
67-
response = request.execute(num_retries=3)
68-
69-
sentiment = response.get('documentSentiment')
70-
71-
if sentiment is None:
72-
return (None, None)
73-
else:
74-
pol = sentiment.get('polarity')
75-
mag = sentiment.get('magnitude')
76-
77-
if pol is None and mag is not None:
78-
pol = 0
79-
return (pol, mag)
80-
81-
8257
class Document(object):
8358
"""Document class captures a single document of movie reviews."""
8459

8560
def __init__(self, text, doc_id, doc_path):
8661
self.text = text
8762
self.doc_id = doc_id
8863
self.doc_path = doc_path
89-
self.sentence_entity_pair = None
64+
self.sentiment_entity_pair = None
9065
self.label = None
9166

92-
def extract_all_sentences(self, service):
67+
def extract_sentiment_entities(self, service):
9368
"""Extract the sentences in a document."""
9469

95-
if self.sentence_entity_pair is not None:
70+
if self.sentiment_entity_pair is not None:
9671
return self.sentence_entity_pair
9772

9873
docs = service.documents()
9974
request_body = get_request_body(
10075
self.text,
101-
syntax=True,
76+
syntax=False,
10277
entities=True,
103-
sentiment=False)
78+
sentiment=True)
10479
request = docs.annotateText(body=request_body)
10580

10681
ent_list = []
10782

10883
response = request.execute()
10984
entities = response.get('entities', [])
110-
sentences = response.get('sentences', [])
111-
112-
sent_list = [
113-
sentence.get('text', {}).get('content') for sentence in sentences
114-
]
85+
documentSentiment = response.get('documentSentiment', {})
11586

11687
for entity in entities:
11788
ent_type = entity.get('type')
@@ -120,9 +91,9 @@ def extract_all_sentences(self, service):
12091
if ent_type == 'PERSON' and wiki_url is not None:
12192
ent_list.append(wiki_url)
12293

123-
self.sentence_entity_pair = (sent_list, ent_list)
94+
self.sentiment_entity_pair = (documentSentiment, ent_list)
12495

125-
return self.sentence_entity_pair
96+
return self.sentiment_entity_pair
12697

12798

12899
def to_sentiment_json(doc_id, sent, label):
@@ -200,18 +171,9 @@ def get_sentiment_entities(service, document):
200171
"""
201172

202173
sentiments, entities = analyze_document(service, document)
174+
score = sentiments.get('score')
203175

204-
sentiments = [sent for sent in sentiments if sent[0] is not None]
205-
negative_sentiments = [
206-
polarity for polarity, magnitude in sentiments if polarity < 0.0]
207-
positive_sentiments = [
208-
polarity for polarity, magnitude in sentiments if polarity > 0.0]
209-
210-
negative = sum(negative_sentiments)
211-
positive = sum(positive_sentiments)
212-
total = positive + negative
213-
214-
return (total, entities)
176+
return (score, entities)
215177

216178

217179
def get_sentiment_label(sentiment):
@@ -318,8 +280,12 @@ def get_service():
318280
"""Build a client to the Google Cloud Natural Language API."""
319281

320282
credentials = GoogleCredentials.get_application_default()
321-
322-
return discovery.build('language', 'v1beta1',
283+
scoped_credentials = credentials.create_scoped(
284+
['https://www.googleapis.com/auth/cloud-platform'])
285+
http = httplib2.Http()
286+
scoped_credentials.authorize(http)
287+
return discovery.build('language', 'v1',
288+
http=http,
323289
credentials=credentials)
324290

325291

‎language/movie_nl/main_test.py

Copy file name to clipboardExpand all lines: language/movie_nl/main_test.py
+3-3Lines changed: 3 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -69,18 +69,18 @@ def test_process_movie_reviews():
6969
entities = [json.loads(entity) for entity in entities]
7070

7171
# assert sentiments
72-
assert sentiments[0].get('sentiment') == 1.0
72+
assert sentiments[0].get('sentiment') == 0.9
7373
assert sentiments[0].get('label') == 1
7474

75-
assert sentiments[1].get('sentiment') == 1.0
75+
assert sentiments[1].get('sentiment') == 0.9
7676
assert sentiments[1].get('label') == 1
7777

7878
# assert entities
7979
assert len(entities) == 1
8080
assert entities[0].get('name') == 'Tom Cruise'
8181
assert (entities[0].get('wiki_url') ==
8282
'http://en.wikipedia.org/wiki/Tom_Cruise')
83-
assert entities[0].get('sentiment') == 2.0
83+
assert entities[0].get('sentiment') == 1.8
8484

8585

8686
def test_rank_positive_entities(capsys):

‎language/ocr_nl/main.py

Copy file name to clipboardExpand all lines: language/ocr_nl/main.py
+4-2Lines changed: 4 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -115,10 +115,12 @@ class TextAnalyzer(object):
115115
def __init__(self, db_filename=None):
116116
credentials = GoogleCredentials.get_application_default()
117117
scoped_credentials = credentials.create_scoped(
118-
['https://www.googleapis.com/auth/cloud-platform'])
118+
['https://www.googleapis.com/auth/cloud-platform'])
119119
http = httplib2.Http()
120120
scoped_credentials.authorize(http)
121-
self.service = discovery.build('language', 'v1beta1', http=http)
121+
self.service = discovery.build('language', 'v1',
122+
http=http,
123+
credentials=credentials)
122124

123125
# This list will store the entity information gleaned from the
124126
# image files.

‎language/sentiment/sentiment_analysis.py

Copy file name to clipboardExpand all lines: language/sentiment/sentiment_analysis.py
+16-4Lines changed: 16 additions & 4 deletions
Original file line numberDiff line numberDiff line change
@@ -22,7 +22,7 @@ def main(movie_review_filename):
2222
'''Run a sentiment analysis request on text within a passed filename.'''
2323

2424
credentials = GoogleCredentials.get_application_default()
25-
service = discovery.build('language', 'v1beta1', credentials=credentials)
25+
service = discovery.build('language', 'v1', credentials=credentials)
2626

2727
with open(movie_review_filename, 'r') as review_file:
2828
service_request = service.documents().analyzeSentiment(
@@ -35,11 +35,23 @@ def main(movie_review_filename):
3535
)
3636
response = service_request.execute()
3737

38-
polarity = response['documentSentiment']['polarity']
38+
score = response['documentSentiment']['score']
3939
magnitude = response['documentSentiment']['magnitude']
4040

41-
print('Sentiment: polarity of {} with magnitude of {}'.format(
42-
polarity, magnitude))
41+
for i, sentence in enumerate(response['sentences']):
42+
sentence_sentiment = sentence['sentiment']['score']
43+
print('Sentence {} has a sentiment score of {}'.format(
44+
i,
45+
sentence_sentiment))
46+
47+
print('Overall Sentiment: score of {} with magnitude of {}'.format(
48+
score,
49+
magnitude)
50+
)
51+
return 0
52+
53+
print('Sentiment: score of {} with magnitude of {}'.format(
54+
score, magnitude))
4355
return 0
4456

4557

‎language/sentiment/sentiment_analysis_test.py

Copy file name to clipboardExpand all lines: language/sentiment/sentiment_analysis_test.py
+7-7Lines changed: 7 additions & 7 deletions
Original file line numberDiff line numberDiff line change
@@ -18,25 +18,25 @@
1818
def test_pos(resource, capsys):
1919
main(resource('pos.txt'))
2020
out, err = capsys.readouterr()
21-
polarity = float(re.search('polarity of (.+?) with', out).group(1))
21+
score = float(re.search('score of (.+?) with', out).group(1))
2222
magnitude = float(re.search('magnitude of (.+?)', out).group(1))
23-
assert polarity * magnitude > 0
23+
assert score * magnitude > 0
2424

2525

2626
def test_neg(resource, capsys):
2727
main(resource('neg.txt'))
2828
out, err = capsys.readouterr()
29-
polarity = float(re.search('polarity of (.+?) with', out).group(1))
29+
score = float(re.search('score of (.+?) with', out).group(1))
3030
magnitude = float(re.search('magnitude of (.+?)', out).group(1))
31-
assert polarity * magnitude < 0
31+
assert score * magnitude < 0
3232

3333

3434
def test_mixed(resource, capsys):
3535
main(resource('mixed.txt'))
3636
out, err = capsys.readouterr()
37-
polarity = float(re.search('polarity of (.+?) with', out).group(1))
38-
assert polarity <= 0.3
39-
assert polarity >= -0.3
37+
score = float(re.search('score of (.+?) with', out).group(1))
38+
assert score <= 0.3
39+
assert score >= -0.3
4040

4141

4242
def test_neutral(resource, capsys):

0 commit comments

Comments
0 (0)
Morty Proxy This is a proxified and sanitized view of the page, visit original site.