Datasets:
patriziobellan
commited on
Commit
•
28456d8
1
Parent(s):
d0040a3
Upload PET.py
Browse files
PET.py
CHANGED
@@ -1,4 +1,13 @@
|
|
1 |
-
#
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2 |
#
|
3 |
# Copyright 2020 The HuggingFace Datasets Authors and the current dataset script contributor.
|
4 |
#
|
@@ -55,7 +64,35 @@ _URL = "https://pdi.fbk.eu/pet/PETHuggingFace/"
|
|
55 |
# _TRAINING_FILE = "train.json"
|
56 |
# _DEV_FILE = "dev.json"
|
57 |
_TEST_FILE = "test.json"
|
58 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
59 |
|
60 |
|
61 |
class PETConfig(datasets.BuilderConfig):
|
@@ -70,105 +107,114 @@ class PETConfig(datasets.BuilderConfig):
|
|
70 |
|
71 |
class PET(datasets.GeneratorBasedBuilder):
|
72 |
"""PET DATASET."""
|
73 |
-
|
74 |
-
|
75 |
-
name
|
76 |
-
|
77 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
78 |
|
79 |
def _info(self):
|
80 |
-
|
|
|
|
|
|
|
|
|
81 |
{
|
82 |
-
|
83 |
-
|
84 |
-
|
85 |
-
|
86 |
-
|
87 |
-
|
88 |
-
"O",
|
89 |
-
"B-Actor",
|
90 |
-
"I-Actor",
|
91 |
-
"B-Activity",
|
92 |
-
"I-Activity",
|
93 |
-
"B-Activity Data",
|
94 |
-
"I-Activity Data",
|
95 |
-
"B-Further Specification",
|
96 |
-
"I-Further Specification",
|
97 |
-
"B-XOR Gateway",
|
98 |
-
"I-XOR Gateway",
|
99 |
-
"B-Condition Specification",
|
100 |
-
"I-Condition Specification",
|
101 |
-
"B-AND Gateway",
|
102 |
-
"I-AND Gateway",
|
103 |
-
]
|
104 |
-
)
|
105 |
-
),
|
106 |
}
|
107 |
)
|
108 |
-
|
109 |
return datasets.DatasetInfo(
|
110 |
-
# This is the description that will appear on the datasets page.
|
111 |
description=_DESCRIPTION,
|
112 |
-
|
113 |
-
features=features, # Here we define them above because they are different between the two configurations
|
114 |
-
# If there's a common (input, target) tuple from the features, uncomment supervised_keys line below and
|
115 |
-
# specify them. They'll be used if as_supervised=True in builder.as_dataset.
|
116 |
-
# supervised_keys=("sentence", "label"),
|
117 |
-
# Homepage of the dataset for documentation
|
118 |
homepage=_HOMEPAGE,
|
119 |
-
# License for the dataset if available
|
120 |
license=_LICENSE,
|
121 |
-
# Citation for the dataset
|
122 |
citation=_CITATION,
|
123 |
)
|
124 |
|
125 |
def _split_generators(self, dl_manager):
|
126 |
-
|
127 |
-
|
128 |
-
|
129 |
-
|
130 |
-
|
131 |
-
|
132 |
-
|
133 |
-
|
134 |
-
|
135 |
-
|
136 |
-
|
137 |
-
|
138 |
-
|
139 |
-
|
140 |
-
|
141 |
-
|
142 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
143 |
name=datasets.Split.TEST,
|
144 |
# These kwargs will be passed to _generate_examples
|
145 |
gen_kwargs={
|
146 |
"filepath": downloaded_files["test"],
|
147 |
"split": "test"
|
148 |
},
|
149 |
-
)
|
150 |
-
|
151 |
-
# datasets.SplitGenerator(
|
152 |
-
# name=datasets.Split.VALIDATION,
|
153 |
-
# # These kwargs will be passed to _generate_examples
|
154 |
-
# gen_kwargs={
|
155 |
-
# "filepath": os.path.join(data_dir, "dev.jsonl"),
|
156 |
-
# "split": "dev",
|
157 |
-
# },
|
158 |
-
# ),
|
159 |
-
]
|
160 |
-
|
161 |
-
# method parameters are unpacked from `gen_kwargs` as given in `_split_generators`
|
162 |
def _generate_examples(self, filepath, split):
|
163 |
-
|
164 |
-
|
165 |
-
|
166 |
-
|
167 |
-
|
168 |
-
|
169 |
-
|
170 |
-
|
171 |
-
|
172 |
-
|
173 |
-
|
174 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
# https://huggingface.co/docs/datasets/v1.2.1/add_dataset.html
|
2 |
+
# TO login
|
3 |
+
# TO CREATE dataset_infos.json use:
|
4 |
+
# $ datasets-cli test PET --save_infos --all_configs
|
5 |
+
#
|
6 |
+
# DO
|
7 |
+
# $ huggingface-cli login
|
8 |
+
# then.
|
9 |
+
# in pytohn:
|
10 |
+
# dataset.push_to_hub(patriziobellan/PET) to set the preview on the web interface
|
11 |
#
|
12 |
# Copyright 2020 The HuggingFace Datasets Authors and the current dataset script contributor.
|
13 |
#
|
|
|
64 |
# _TRAINING_FILE = "train.json"
|
65 |
# _DEV_FILE = "dev.json"
|
66 |
_TEST_FILE = "test.json"
|
67 |
+
_TEST_FILE_RELATIONS = 'PETrelations.json'
|
68 |
+
|
69 |
+
_NER = 'token-classification'
|
70 |
+
_RELATIONS_EXTRACTION = 'relations-extraction'
|
71 |
+
_NER_TAGS = [ "O",
|
72 |
+
"B-Actor",
|
73 |
+
"I-Actor",
|
74 |
+
"B-Activity",
|
75 |
+
"I-Activity",
|
76 |
+
"B-Activity Data",
|
77 |
+
"I-Activity Data",
|
78 |
+
"B-Further Specification",
|
79 |
+
"I-Further Specification",
|
80 |
+
"B-XOR Gateway",
|
81 |
+
"I-XOR Gateway",
|
82 |
+
"B-Condition Specification",
|
83 |
+
"I-Condition Specification",
|
84 |
+
"B-AND Gateway",
|
85 |
+
"I-AND Gateway"]
|
86 |
+
|
87 |
+
_STR_PET = """\n
|
88 |
+
_______ _ _ _______ _____ _______ _______ ______ _______ _______ _______ _______ _______ _______
|
89 |
+
| |_____| |______ |_____] |______ | | \ |_____| | |_____| |______ |______ |
|
90 |
+
| | | |______ | |______ | |_____/ | | | | | ______| |______ |
|
91 |
+
|
92 |
+
\n\n\n
|
93 |
+
|
94 |
+
Discover more at: [https://pdi.fbk.eu/pet-dataset/]
|
95 |
+
"""
|
96 |
|
97 |
|
98 |
class PETConfig(datasets.BuilderConfig):
|
|
|
107 |
|
108 |
class PET(datasets.GeneratorBasedBuilder):
|
109 |
"""PET DATASET."""
|
110 |
+
|
111 |
+
features_ner = {
|
112 |
+
"document name": datasets.Value("string"),
|
113 |
+
"sentence-ID": datasets.Value("int8"),
|
114 |
+
"tokens": datasets.Sequence(datasets.Value("string")),
|
115 |
+
"ner-tags": datasets.Sequence(datasets.features.ClassLabel(names=_NER_TAGS)),
|
116 |
+
}
|
117 |
+
|
118 |
+
features_relations = datasets.Sequence(
|
119 |
+
datasets.Features(
|
120 |
+
|
121 |
+
{
|
122 |
+
'source-head-sentence-ID': datasets.Value("int8"),
|
123 |
+
'source-head-word-ID': datasets.Value("int8"),
|
124 |
+
'relation-type': datasets.Value("string"),
|
125 |
+
'target-head-sentence-ID': datasets.Value("int8"),
|
126 |
+
'target-head-word-ID' : datasets.Value("int8"),
|
127 |
+
}
|
128 |
+
))
|
129 |
+
BUILDER_CONFIGS = [ PETConfig(
|
130 |
+
name=_NER,
|
131 |
+
version=datasets.Version("1.0.1"),
|
132 |
+
description="The PET Dataset for Token Classification"
|
133 |
+
),
|
134 |
+
PETConfig(
|
135 |
+
name=_RELATIONS_EXTRACTION,
|
136 |
+
version=datasets.Version("1.0.1"),
|
137 |
+
description="The PET Dataset for Relation Extraction"
|
138 |
+
),
|
139 |
+
]
|
140 |
+
|
141 |
+
DEFAULT_CONFIG_NAME = _RELATIONS_EXTRACTION
|
142 |
|
143 |
def _info(self):
|
144 |
+
print(_STR_PET)
|
145 |
+
if self.config.name == _NER:
|
146 |
+
features = datasets.Features(self.features_ner)
|
147 |
+
else:
|
148 |
+
features = datasets.Features(
|
149 |
{
|
150 |
+
"document name": datasets.Value("string"),
|
151 |
+
'tokens':datasets.Sequence(datasets.Value("string")),
|
152 |
+
'tokens-IDs':datasets.Sequence(datasets.Value("int8")),
|
153 |
+
'ner_tags': datasets.Sequence(datasets.Value("string")),
|
154 |
+
'sentence-IDs':datasets.Sequence(datasets.Value("int8")),
|
155 |
+
"relations": self.features_relations
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
156 |
}
|
157 |
)
|
158 |
+
# print(features)
|
159 |
return datasets.DatasetInfo(
|
|
|
160 |
description=_DESCRIPTION,
|
161 |
+
features=datasets.Features(features),
|
|
|
|
|
|
|
|
|
|
|
162 |
homepage=_HOMEPAGE,
|
|
|
163 |
license=_LICENSE,
|
|
|
164 |
citation=_CITATION,
|
165 |
)
|
166 |
|
167 |
def _split_generators(self, dl_manager):
|
168 |
+
if self.config.name == _NER:
|
169 |
+
urls_to_download = {
|
170 |
+
# "train": f"{_URL}{_TRAINING_FILE}",
|
171 |
+
# "dev": f"{_URL}{_DEV_FILE}",
|
172 |
+
"test": f"{_URL}{_TEST_FILE}",
|
173 |
+
}
|
174 |
+
downloaded_files = dl_manager.download_and_extract(urls_to_download)
|
175 |
+
return [datasets.SplitGenerator(
|
176 |
+
name=datasets.Split.TEST,
|
177 |
+
# These kwargs will be passed to _generate_examples
|
178 |
+
gen_kwargs={
|
179 |
+
"filepath": downloaded_files["test"],
|
180 |
+
"split": "test"
|
181 |
+
},
|
182 |
+
)]
|
183 |
+
else:
|
184 |
+
urls_to_download = {
|
185 |
+
# "train": f"{_URL}{_TRAINING_FILE}",
|
186 |
+
# "dev": f"{_URL}{_DEV_FILE}",
|
187 |
+
"test": f"{_URL}{_TEST_FILE_RELATIONS}",
|
188 |
+
}
|
189 |
+
downloaded_files = dl_manager.download_and_extract(urls_to_download)
|
190 |
+
return [datasets.SplitGenerator(
|
191 |
name=datasets.Split.TEST,
|
192 |
# These kwargs will be passed to _generate_examples
|
193 |
gen_kwargs={
|
194 |
"filepath": downloaded_files["test"],
|
195 |
"split": "test"
|
196 |
},
|
197 |
+
)]
|
198 |
+
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
199 |
def _generate_examples(self, filepath, split):
|
200 |
+
if self.config.name == _NER:
|
201 |
+
with open(filepath, encoding="utf-8", mode='r') as f:
|
202 |
+
for key, row in enumerate(f):
|
203 |
+
row = json.loads(row)
|
204 |
+
yield key, {
|
205 |
+
"document name": row["document name"],
|
206 |
+
"sentence-ID": row["sentence-ID"],
|
207 |
+
"tokens": row["tokens"],
|
208 |
+
"ner-tags": row["ner-tags"]
|
209 |
+
}
|
210 |
+
else:
|
211 |
+
with open(filepath, encoding="utf-8", mode='r') as f:
|
212 |
+
for key, row in enumerate(json.load(f)):
|
213 |
+
yield key, {"document name": row["document name"], # datasets.Value("string"),
|
214 |
+
'tokens': row["tokens"], # sentences['tokens'],
|
215 |
+
'tokens-IDs': row["tokens-IDs"],
|
216 |
+
'ner_tags': row["ner_tags"],
|
217 |
+
'sentence-IDs': row["sentence-IDs"], # sentences['sentence-IDs'],
|
218 |
+
|
219 |
+
"relations": row["relations"]
|
220 |
+
}
|