beki commited on
Commit
c1c5fe8
1 Parent(s): 8aa33f7

Create privy.py

Browse files
Files changed (1) hide show
  1. privy.py +274 -0
privy.py ADDED
@@ -0,0 +1,274 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # Copyright 2020 The HuggingFace Datasets Authors and the current dataset script contributor.
2
+ #
3
+ # Licensed under the Apache License, Version 2.0 (the "License");
4
+ # you may not use this file except in compliance with the License.
5
+ # You may obtain a copy of the License at
6
+ #
7
+ # http://www.apache.org/licenses/LICENSE-2.0
8
+ #
9
+ # Unless required by applicable law or agreed to in writing, software
10
+ # distributed under the License is distributed on an "AS IS" BASIS,
11
+ # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
12
+ # See the License for the specific language governing permissions and
13
+ # limitations under the License.
14
+ import csv
15
+ import json
16
+ import os
17
+
18
+ import datasets
19
+
20
+ _DESCRIPTION="This labelled PII dataset consists of protocol traces (JSON, SQL (PostgreSQL, MySQL), HTML, and XML) generated from OpenAPI specifications and includes 60+ PII types."
21
+ _CITATION="""
22
+ @online{WinNT,
23
+ author = {Benjamin Kilimnik},
24
+ title = {{Privy} Synthetic PII Protocol Trace Dataset},
25
+ year = 2022,
26
+ url = {https://huggingface.co/datasets/beki/privy},
27
+ }
28
+ """
29
+
30
+ _HOMEPAGE = "https://github.com/pixie-io/pixie/tree/main/src/datagen/pii/privy/privy"
31
+
32
+ _LICENSE = "MIT"
33
+
34
+ class NewDataset(datasets.GeneratorBasedBuilder):
35
+ VERSION = datasets.Version("1.0.0")
36
+
37
+ # You will be able to load one or the other configurations in the following list with
38
+ # data = datasets.load_dataset('my_dataset', 'first_domain')
39
+ # data = datasets.load_dataset('my_dataset', 'second_domain')
40
+ BUILDER_CONFIGS = [
41
+ datasets.BuilderConfig(name="small", version=VERSION, description="Privy small"),
42
+ datasets.BuilderConfig(name="large", version=VERSION, description="Privy large"),
43
+ ]
44
+
45
+ DEFAULT_CONFIG_NAME = "small"
46
+
47
+ def _info(self):
48
+ features = datasets.Features(
49
+ {
50
+ "full_text": datasets.Value("string"),
51
+ "masked": datasets.Value("string"),
52
+ "spans": datasets.Sequence(datasets.Value("string")),
53
+ "tags": datasets.Sequence(
54
+ datasets.features.ClassLabel(
55
+ names=[
56
+ "O",
57
+ "B-O",
58
+ "I-O",
59
+ "L-O",
60
+ "U-O",
61
+
62
+ "B-PER",
63
+ "I-PER",
64
+ "L-PER",
65
+ "U-PER",
66
+
67
+ "B-LOC",
68
+ "I-LOC",
69
+ "L-LOC",
70
+ "U-LOC",
71
+
72
+ "B-ORG",
73
+ "I-ORG",
74
+ "L-ORG",
75
+ "U-ORG",
76
+
77
+ "B-NRP",
78
+ "I-NRP",
79
+ "L-NRP",
80
+ "U-NRP",
81
+
82
+ "B-DATE_TIME",
83
+ "I-DATE_TIME",
84
+ "L-DATE_TIME",
85
+ "U-DATE_TIME",
86
+
87
+ "B-CREDIT_CARD",
88
+ "I-CREDIT_CARD",
89
+ "L-CREDIT_CARD",
90
+ "U-CREDIT_CARD",
91
+
92
+ "B-URL",
93
+ "I-URL",
94
+ "L-URL",
95
+ "U-URL",
96
+
97
+ "B-IBAN_CODE",
98
+ "I-IBAN_CODE",
99
+ "L-IBAN_CODE",
100
+ "U-IBAN_CODE",
101
+
102
+ "B-US_BANK_NUMBER",
103
+ "I-US_BANK_NUMBER",
104
+ "L-US_BANK_NUMBER",
105
+ "U-US_BANK_NUMBER",
106
+
107
+ "B-PHONE_NUMBER",
108
+ "I-PHONE_NUMBER",
109
+ "L-PHONE_NUMBER",
110
+ "U-PHONE_NUMBER",
111
+
112
+ "B-US_SSN",
113
+ "I-US_SSN",
114
+ "L-US_SSN",
115
+ "U-US_SSN",
116
+
117
+ "B-US_PASSPORT",
118
+ "I-US_PASSPORT",
119
+ "L-US_PASSPORT",
120
+ "U-US_PASSPORT",
121
+
122
+ "B-US_DRIVER_LICENSE",
123
+ "I-US_DRIVER_LICENSE",
124
+ "L-US_DRIVER_LICENSE",
125
+ "U-US_DRIVER_LICENSE",
126
+
127
+ "B-US_LICENSE_PLATE",
128
+ "I-US_LICENSE_PLATE",
129
+ "L-US_LICENSE_PLATE",
130
+ "U-US_LICENSE_PLATE",
131
+
132
+ "B-IP_ADDRESS",
133
+ "I-IP_ADDRESS",
134
+ "L-IP_ADDRESS",
135
+ "U-IP_ADDRESS",
136
+
137
+ "B-US_ITIN",
138
+ "I-US_ITIN",
139
+ "L-US_ITIN",
140
+ "U-US_ITIN",
141
+
142
+ "B-EMAIL_ADDRESS",
143
+ "I-EMAIL_ADDRESS",
144
+ "L-EMAIL_ADDRESS",
145
+ "U-EMAIL_ADDRESS",
146
+
147
+ "B-ORGANIZATION",
148
+ "I-ORGANIZATION",
149
+ "L-ORGANIZATION",
150
+ "U-ORGANIZATION",
151
+
152
+ "B-TITLE",
153
+ "I-TITLE",
154
+ "L-TITLE",
155
+ "U-TITLE",
156
+
157
+ "B-COORDINATE",
158
+ "I-COORDINATE",
159
+ "L-COORDINATE",
160
+ "U-COORDINATE",
161
+
162
+ "B-IMEI",
163
+ "I-IMEI",
164
+ "L-IMEI",
165
+ "U-IMEI",
166
+
167
+ "B-PASSWORD",
168
+ "I-PASSWORD",
169
+ "L-PASSWORD",
170
+ "U-PASSWORD",
171
+
172
+ "B-LICENSE_PLATE",
173
+ "I-LICENSE_PLATE",
174
+ "L-LICENSE_PLATE",
175
+ "U-LICENSE_PLATE",
176
+
177
+ "B-CURRENCY",
178
+ "I-CURRENCY",
179
+ "L-CURRENCY",
180
+ "U-CURRENCY",
181
+
182
+ "B-FINANCIAL",
183
+ "I-FINANCIAL",
184
+ "L-FINANCIAL",
185
+ "U-FINANCIAL",
186
+
187
+ "B-ROUTING_NUMBER",
188
+ "I-ROUTING_NUMBER",
189
+ "L-ROUTING_NUMBER",
190
+ "U-ROUTING_NUMBER",
191
+
192
+ "B-SWIFT_CODE",
193
+ "I-SWIFT_CODE",
194
+ "L-SWIFT_CODE",
195
+ "U-SWIFT_CODE",
196
+
197
+ "B-MAC_ADDRESS",
198
+ "I-MAC_ADDRESS",
199
+ "L-MAC_ADDRESS",
200
+ "U-MAC_ADDRESS",
201
+
202
+ "B-AGE",
203
+ "I-AGE",
204
+ "L-AGE",
205
+ "U-AGE",
206
+ ]
207
+ )
208
+ ),
209
+ "tokens": datasets.Sequence(datasets.Value("string")),
210
+ "template_id": datasets.Value("int32"),
211
+ "metadata": datasets.Value("int32"),
212
+ }
213
+ )
214
+
215
+ return datasets.DatasetInfo(
216
+ description=_DESCRIPTION,
217
+ features=features,
218
+ # supervised_keys=("sentence", "label"),
219
+ # Homepage of the dataset for documentation
220
+ homepage=_HOMEPAGE,
221
+ # License for the dataset if available
222
+ license=_LICENSE,
223
+ # Citation for the dataset
224
+ citation=_CITATION,
225
+ )
226
+
227
+ def _split_generators(self, dl_manager):
228
+ data_dir = "./"
229
+ size = "small"
230
+ if self.config.name == "large": # This is the name of the configuration selected in BUILDER_CONFIGS above
231
+ size = "large"
232
+ return [
233
+ datasets.SplitGenerator(
234
+ name=datasets.Split.TRAIN,
235
+ # These kwargs will be passed to _generate_examples
236
+ gen_kwargs={
237
+ "filepath": os.path.join(data_dir, f"train-{size}.json"),
238
+ "split": "train",
239
+ },
240
+ ),
241
+ datasets.SplitGenerator(
242
+ name=datasets.Split.VALIDATION,
243
+ # These kwargs will be passed to _generate_examples
244
+ gen_kwargs={
245
+ "filepath": os.path.join(data_dir, f"dev-{size}.json"),
246
+ "split": "dev",
247
+ },
248
+ ),
249
+ datasets.SplitGenerator(
250
+ name=datasets.Split.TEST,
251
+ # These kwargs will be passed to _generate_examples
252
+ gen_kwargs={
253
+ "filepath": os.path.join(data_dir, f"test-{size}.json"),
254
+ "split": "test"
255
+ },
256
+ ),
257
+ ]
258
+
259
+ # method parameters are unpacked from `gen_kwargs` as given in `_split_generators`
260
+ def _generate_examples(self, filepath, split):
261
+ # The `key` is for legacy reasons (tfds) and is not important in itself, but must be unique for each example.
262
+ with open(filepath, encoding="utf-8") as f:
263
+ dataset = json.load(f)
264
+ for key, row in enumerate(dataset):
265
+ # Yields examples as (key, example) tuples
266
+ yield key, {
267
+ "tokens": row["tokens"],
268
+ "tags": row["tags"],
269
+ "full_text": row["full_text"],
270
+ "spans": row["spans"],
271
+ "masked": row["masked"],
272
+ "template_id": row["template_id"],
273
+ "metadata": row["metadata"],
274
+ }