1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
| import os
import json
import sys
import unittest
from rougescore import rouge_n, rouge_l
from pythonrouge.pythonrouge import Pythonrouge
from sumeval.metrics.rouge import RougeCalculator
class TestRouge(unittest.TestCase):
DATA_DIR = os.path.join(os.path.dirname(__file__), "data/rouge")
def load_test_data(self):
test_file = os.path.join(self.DATA_DIR, "ROUGE-test.json")
with open(test_file, encoding="utf-8") as f:
data = json.load(f)
return data
def test_rouge(self):
data = self.load_test_data()
rouge = RougeCalculator(stopwords=False)
for eval_id in data:
summaries = data[eval_id]["summaries"]
references = data[eval_id]["references"]
for n in [1, 2]:
for s in summaries:
baseline = Pythonrouge(
summary_file_exist=False,
summary=[[s]],
reference=[[[r] for r in references]],
n_gram=n, recall_only=False,
length_limit=False,
stemming=False, stopwords=False)
b1_v = baseline.calc_score()
b2_v = rouge_n(rouge.tokenize(s),
[rouge.tokenize(r) for r in references],
n, 0.5)
v = rouge.rouge_n(s, references, n)
self.assertAlmostEqual(b1_v["ROUGE-{}".format(n)], b2_v, places=4)
self.assertAlmostEqual(b1_v["ROUGE-{}".format(n)], v, places=4)
def test_rouge_with_stop_word(self):
data = self.load_test_data()
rouge = RougeCalculator(stopwords=True)
for eval_id in data:
summaries = data[eval_id]["summaries"]
references = data[eval_id]["references"]
for n in [1, 2]:
for s in summaries:
baseline = Pythonrouge(
summary_file_exist=False,
summary=[[s]],
reference=[[[r] for r in references]],
n_gram=n, recall_only=False,
length_limit=False,
stemming=False, stopwords=True)
b1_v = baseline.calc_score()
b2_v = rouge_n(rouge.tokenize(s),
[rouge.tokenize(r) for r in references],
n, 0.5)
v = rouge.rouge_n(s, references, n)
self.assertAlmostEqual(b1_v["ROUGE-{}".format(n)], b2_v, places=4)
self.assertAlmostEqual(b1_v["ROUGE-{}".format(n)], v, places=4)
def test_rouge_with_length_limit(self):
data = self.load_test_data()
rouge = RougeCalculator(stopwords=True, length_limit=50)
for eval_id in data:
summaries = data[eval_id]["summaries"]
references = data[eval_id]["references"]
for n in [1, 2]:
for s in summaries:
baseline = Pythonrouge(
summary_file_exist=False,
summary=[[s]],
reference=[[[r] for r in references]],
n_gram=n, recall_only=False,
length_limit=True, length=50,
word_level=False,
stemming=False, stopwords=True)
b1_v = baseline.calc_score()
b2_v = rouge_n(rouge.tokenize(s),
[rouge.tokenize(r) for r in references],
n, 0.5)
v = rouge.rouge_n(s, references, n)
self.assertAlmostEqual(b1_v["ROUGE-{}".format(n)], b2_v, places=4)
self.assertAlmostEqual(b1_v["ROUGE-{}".format(n)], v, places=4)
def test_rouge_with_word_limit(self):
data = self.load_test_data()
rouge = RougeCalculator(stopwords=True, word_limit=5)
for eval_id in data:
summaries = data[eval_id]["summaries"]
references = data[eval_id]["references"]
for n in [1, 2]:
for s in summaries:
baseline = Pythonrouge(
summary_file_exist=False,
summary=[[s]],
reference=[[[r] for r in references]],
n_gram=n, recall_only=False,
length_limit=True, length=5,
word_level=True,
stemming=False, stopwords=True)
b1_v = baseline.calc_score()
b2_v = rouge_n(rouge.tokenize(s),
[rouge.tokenize(r) for r in references],
n, 0.5)
v = rouge.rouge_n(s, references, n)
self.assertAlmostEqual(b1_v["ROUGE-{}".format(n)], b2_v, places=4)
self.assertAlmostEqual(b1_v["ROUGE-{}".format(n)], v, places=4)
def test_rouge_l(self):
data = self.load_test_data()
rouge = RougeCalculator(stopwords=True)
for eval_id in data:
summaries = data[eval_id]["summaries"]
references = data[eval_id]["references"]
for s in summaries:
baseline = Pythonrouge(
summary_file_exist=False,
summary=[[s]],
reference=[[[r] for r in references]],
n_gram=1, recall_only=False, ROUGE_L=True,
length_limit=True, length=50,
stemming=False, stopwords=True)
b1_v = baseline.calc_score()
b2_v = rouge_l(rouge.tokenize(s),
[rouge.tokenize(r) for r in references],
0.5)
v = rouge.rouge_l(s, references)
self.assertAlmostEqual(b1_v["ROUGE-L"], b2_v, places=4)
self.assertAlmostEqual(b1_v["ROUGE-L"], v, places=4)
if __name__ == "__main__":
unittest.main()
|
