Datasets:
Upload 2 files
Browse files- DiscoEval.py +13 -4
- DiscoEvalConstants.py +32 -7
DiscoEval.py
CHANGED
@@ -115,7 +115,7 @@ class DiscoEvalSentence(datasets.GeneratorBasedBuilder):
|
|
115 |
DiscoEvalConstants.TEXT_COLUMN_NAME[i]: datasets.Value('string')
|
116 |
for i in range(DiscoEvalConstants.BSO_TEXT_COLUMNS)
|
117 |
}
|
118 |
-
features_dict[DiscoEvalConstants.LABEL_NAME] = datasets.ClassLabel(names=DiscoEvalConstants.BSO_LABELS)
|
119 |
features = datasets.Features(features_dict)
|
120 |
|
121 |
elif self.config.name in [DiscoEvalConstants.DCCHAT, DiscoEvalConstants.DCWIKI]:
|
@@ -123,7 +123,7 @@ class DiscoEvalSentence(datasets.GeneratorBasedBuilder):
|
|
123 |
DiscoEvalConstants.TEXT_COLUMN_NAME[i]: datasets.Value('string')
|
124 |
for i in range(DiscoEvalConstants.DC_TEXT_COLUMNS)
|
125 |
}
|
126 |
-
features_dict[DiscoEvalConstants.LABEL_NAME] = datasets.ClassLabel(names=DiscoEvalConstants.DC_LABELS)
|
127 |
features = datasets.Features(features_dict)
|
128 |
|
129 |
elif self.config.name in [DiscoEvalConstants.RST]:
|
@@ -155,7 +155,7 @@ class DiscoEvalSentence(datasets.GeneratorBasedBuilder):
|
|
155 |
DiscoEvalConstants.TEXT_COLUMN_NAME[i]: datasets.Value('string')
|
156 |
for i in range(DiscoEvalConstants.SSPABS_TEXT_COLUMNS)
|
157 |
}
|
158 |
-
features_dict[DiscoEvalConstants.LABEL_NAME] = datasets.ClassLabel(names=DiscoEvalConstants.SSPABS_LABELS)
|
159 |
features = datasets.Features(features_dict)
|
160 |
|
161 |
return datasets.DatasetInfo(
|
@@ -253,5 +253,14 @@ class DiscoEvalSentence(datasets.GeneratorBasedBuilder):
|
|
253 |
for key, line in enumerate(f):
|
254 |
line = line.strip().split("\t")
|
255 |
example = {DiscoEvalConstants.TEXT_COLUMN_NAME[i]: sent for i, sent in enumerate(line[1:])}
|
256 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
257 |
yield key, example
|
|
|
115 |
DiscoEvalConstants.TEXT_COLUMN_NAME[i]: datasets.Value('string')
|
116 |
for i in range(DiscoEvalConstants.BSO_TEXT_COLUMNS)
|
117 |
}
|
118 |
+
features_dict[DiscoEvalConstants.LABEL_NAME] = datasets.ClassLabel(names=DiscoEvalConstants.BSO_LABELS.values())
|
119 |
features = datasets.Features(features_dict)
|
120 |
|
121 |
elif self.config.name in [DiscoEvalConstants.DCCHAT, DiscoEvalConstants.DCWIKI]:
|
|
|
123 |
DiscoEvalConstants.TEXT_COLUMN_NAME[i]: datasets.Value('string')
|
124 |
for i in range(DiscoEvalConstants.DC_TEXT_COLUMNS)
|
125 |
}
|
126 |
+
features_dict[DiscoEvalConstants.LABEL_NAME] = datasets.ClassLabel(names=DiscoEvalConstants.DC_LABELS.values())
|
127 |
features = datasets.Features(features_dict)
|
128 |
|
129 |
elif self.config.name in [DiscoEvalConstants.RST]:
|
|
|
155 |
DiscoEvalConstants.TEXT_COLUMN_NAME[i]: datasets.Value('string')
|
156 |
for i in range(DiscoEvalConstants.SSPABS_TEXT_COLUMNS)
|
157 |
}
|
158 |
+
features_dict[DiscoEvalConstants.LABEL_NAME] = datasets.ClassLabel(names=DiscoEvalConstants.SSPABS_LABELS.values())
|
159 |
features = datasets.Features(features_dict)
|
160 |
|
161 |
return datasets.DatasetInfo(
|
|
|
253 |
for key, line in enumerate(f):
|
254 |
line = line.strip().split("\t")
|
255 |
example = {DiscoEvalConstants.TEXT_COLUMN_NAME[i]: sent for i, sent in enumerate(line[1:])}
|
256 |
+
if self.config.name in (DiscoEvalConstants.PDTB_E, DiscoEvalConstants.PDTB_I):
|
257 |
+
example[DiscoEvalConstants.LABEL_NAME] = line[0]
|
258 |
+
elif self.config.name in (DiscoEvalConstants.DCCHAT, DiscoEvalConstants.DCWIKI):
|
259 |
+
example[DiscoEvalConstants.LABEL_NAME] = DiscoEvalConstants.DC_LABELS[line[0]]
|
260 |
+
elif self.config.name == DiscoEvalConstants.SSPABS:
|
261 |
+
example[DiscoEvalConstants.LABEL_NAME] = DiscoEvalConstants.SSPABS_LABELS[line[0]]
|
262 |
+
elif self.config.name in (DiscoEvalConstants.SPWIKI, DiscoEvalConstants.SPROCSTORY, DiscoEvalConstants.SPARXIV):
|
263 |
+
example[DiscoEvalConstants.LABEL_NAME] = DiscoEvalConstants.SP_LABELS[line[0]]
|
264 |
+
elif self.config.name in (DiscoEvalConstants.BSOARXIV, DiscoEvalConstants.BSOWIKI, DiscoEvalConstants.BSOROCSTORY):
|
265 |
+
example[DiscoEvalConstants.LABEL_NAME] = DiscoEvalConstants.BSO_LABELS[line[0]]
|
266 |
yield key, example
|
DiscoEvalConstants.py
CHANGED
@@ -1,6 +1,18 @@
|
|
1 |
# General Constants:
|
2 |
LABEL_NAME = 'label'
|
3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
4 |
|
5 |
# SSPabs Constants:
|
6 |
SSPABS = 'SSPabs'
|
@@ -8,7 +20,10 @@ SSPABS_TRAIN_NAME = 'train.txt'
|
|
8 |
SSPABS_VALID_NAME = 'valid.txt'
|
9 |
SSPABS_TEST_NAME = 'test.txt'
|
10 |
SSPABS_DATA_DIR = 'data/SSP/abs'
|
11 |
-
SSPABS_LABELS =
|
|
|
|
|
|
|
12 |
SSPABS_TEXT_COLUMNS = 1
|
13 |
|
14 |
# PDTB Constants:
|
@@ -49,7 +64,6 @@ PDTB_I_LABELS = [
|
|
49 |
PDTB_E_TEXT_COLUMNS = 2
|
50 |
PDTB_I_TEXT_COLUMNS = 2
|
51 |
|
52 |
-
|
53 |
# SP Constants:
|
54 |
SPARXIV = 'SParxiv'
|
55 |
SPROCSTORY = 'SProcstory'
|
@@ -59,7 +73,13 @@ SP_VALID_NAME = 'valid.txt'
|
|
59 |
SP_TEST_NAME = 'test.txt'
|
60 |
SP_DATA_DIR = 'data/SP'
|
61 |
SP_DIRS = {SPARXIV: 'arxiv', SPROCSTORY: 'rocstory', SPWIKI: 'wiki'}
|
62 |
-
SP_LABELS =
|
|
|
|
|
|
|
|
|
|
|
|
|
63 |
SP_TEXT_COLUMNS = 5
|
64 |
|
65 |
# BSO Constants:
|
@@ -71,7 +91,10 @@ BSO_VALID_NAME = 'valid.txt'
|
|
71 |
BSO_TEST_NAME = 'test.txt'
|
72 |
BSO_DATA_DIR = 'data/BSO'
|
73 |
BSO_DIRS = {BSOARXIV: 'arxiv', BSOROCSTORY: 'rocstory', BSOWIKI: 'wiki'}
|
74 |
-
BSO_LABELS =
|
|
|
|
|
|
|
75 |
BSO_TEXT_COLUMNS = 2
|
76 |
|
77 |
# DC Constants:
|
@@ -82,10 +105,12 @@ DC_VALID_NAME = 'valid.txt'
|
|
82 |
DC_TEST_NAME = 'test.txt'
|
83 |
DC_DATA_DIR = 'data/DC'
|
84 |
DC_DIRS = {DCCHAT: 'chat', DCWIKI: 'wiki'}
|
85 |
-
DC_LABELS =
|
|
|
|
|
|
|
86 |
DC_TEXT_COLUMNS = 6
|
87 |
|
88 |
-
|
89 |
# RST Constants:
|
90 |
RST = 'RST'
|
91 |
RST_TRAIN_NAME = 'RST_TRAIN.pkl'
|
|
|
1 |
# General Constants:
|
2 |
LABEL_NAME = 'label'
|
3 |
+
numbers = {
|
4 |
+
1: "first",
|
5 |
+
2: "second",
|
6 |
+
3: "third",
|
7 |
+
4: "fourth",
|
8 |
+
5: "fifth",
|
9 |
+
6: "sixth",
|
10 |
+
7: "seventh",
|
11 |
+
8: "eighth",
|
12 |
+
9: "ninth"
|
13 |
+
|
14 |
+
}
|
15 |
+
TEXT_COLUMN_NAME = [f"{numbers[i]}_sentence" for i in range(1, 10)]
|
16 |
|
17 |
# SSPabs Constants:
|
18 |
SSPABS = 'SSPabs'
|
|
|
20 |
SSPABS_VALID_NAME = 'valid.txt'
|
21 |
SSPABS_TEST_NAME = 'test.txt'
|
22 |
SSPABS_DATA_DIR = 'data/SSP/abs'
|
23 |
+
SSPABS_LABELS = {
|
24 |
+
"0": "Does not belong to abstract",
|
25 |
+
"1": "Belongs to abstract",
|
26 |
+
}
|
27 |
SSPABS_TEXT_COLUMNS = 1
|
28 |
|
29 |
# PDTB Constants:
|
|
|
64 |
PDTB_E_TEXT_COLUMNS = 2
|
65 |
PDTB_I_TEXT_COLUMNS = 2
|
66 |
|
|
|
67 |
# SP Constants:
|
68 |
SPARXIV = 'SParxiv'
|
69 |
SPROCSTORY = 'SProcstory'
|
|
|
73 |
SP_TEST_NAME = 'test.txt'
|
74 |
SP_DATA_DIR = 'data/SP'
|
75 |
SP_DIRS = {SPARXIV: 'arxiv', SPROCSTORY: 'rocstory', SPWIKI: 'wiki'}
|
76 |
+
SP_LABELS = {
|
77 |
+
"0": 'First sentence',
|
78 |
+
"1": 'Second sentence',
|
79 |
+
"2": 'Third sentence',
|
80 |
+
"3": "Fourth sentence",
|
81 |
+
"4": "Fifth sentence",
|
82 |
+
}
|
83 |
SP_TEXT_COLUMNS = 5
|
84 |
|
85 |
# BSO Constants:
|
|
|
91 |
BSO_TEST_NAME = 'test.txt'
|
92 |
BSO_DATA_DIR = 'data/BSO'
|
93 |
BSO_DIRS = {BSOARXIV: 'arxiv', BSOROCSTORY: 'rocstory', BSOWIKI: 'wiki'}
|
94 |
+
BSO_LABELS = {
|
95 |
+
"0": 'Incorrect order',
|
96 |
+
"1": 'Correct order',
|
97 |
+
}
|
98 |
BSO_TEXT_COLUMNS = 2
|
99 |
|
100 |
# DC Constants:
|
|
|
105 |
DC_TEST_NAME = 'test.txt'
|
106 |
DC_DATA_DIR = 'data/DC'
|
107 |
DC_DIRS = {DCCHAT: 'chat', DCWIKI: 'wiki'}
|
108 |
+
DC_LABELS = {
|
109 |
+
"0": "Incoherent",
|
110 |
+
"1": "Coherent",
|
111 |
+
}
|
112 |
DC_TEXT_COLUMNS = 6
|
113 |
|
|
|
114 |
# RST Constants:
|
115 |
RST = 'RST'
|
116 |
RST_TRAIN_NAME = 'RST_TRAIN.pkl'
|