albertvillanova HF staff commited on
Commit
5754856
1 Parent(s): f31a317

Add 2.1.2 version with cased raw strings (#3)

Browse files

- Add 2.1.2 data files (04f83833dba762753c1c4d12c5cff6e6e4926b96)
- Move 1.0.0 data files (074aeed8e1a0d039c13b171a77a67950f71c5496)
- Update dataset loading script (0f27e656ac5751add50114c74468dfb433fdec30)
- Update instructions about version in dataset card (a815ad4f6cfd2bc49b15b3c734a8e29f8fe336c6)
- Update metadata in dataset card (30ea432618e70552803977936d6854c9b03c1433)
- Update metadata in legacy JSON (ef871d08aeb1fe2293f40380298f7f2071f485a4)

README.md CHANGED
@@ -42,16 +42,16 @@ dataset_info:
42
  dtype: string
43
  splits:
44
  - name: train
45
- num_bytes: 23363518650
46
  num_examples: 1207222
47
  - name: validation
48
- num_bytes: 1290154487
49
  num_examples: 67068
50
  - name: test
51
- num_bytes: 1296234391
52
  num_examples: 67072
53
- download_size: 6447221554
54
- dataset_size: 25949907528
55
  - config_name: a
56
  features:
57
  - name: description
@@ -60,16 +60,16 @@ dataset_info:
60
  dtype: string
61
  splits:
62
  - name: train
63
- num_bytes: 3329778447
64
  num_examples: 174134
65
  - name: validation
66
- num_bytes: 184116486
67
  num_examples: 9674
68
  - name: test
69
- num_bytes: 185987552
70
  num_examples: 9675
71
- download_size: 6447221554
72
- dataset_size: 3699882485
73
  - config_name: b
74
  features:
75
  - name: description
@@ -78,16 +78,16 @@ dataset_info:
78
  dtype: string
79
  splits:
80
  - name: train
81
- num_bytes: 2574594655
82
  num_examples: 161520
83
  - name: validation
84
- num_bytes: 143029380
85
  num_examples: 8973
86
  - name: test
87
- num_bytes: 140741033
88
  num_examples: 8974
89
- download_size: 6447221554
90
- dataset_size: 2858365068
91
  - config_name: c
92
  features:
93
  - name: description
@@ -96,16 +96,16 @@ dataset_info:
96
  dtype: string
97
  splits:
98
  - name: train
99
- num_bytes: 2641973267
100
  num_examples: 101042
101
  - name: validation
102
- num_bytes: 145441704
103
  num_examples: 5613
104
  - name: test
105
- num_bytes: 149052258
106
  num_examples: 5614
107
- download_size: 6447221554
108
- dataset_size: 2936467229
109
  - config_name: d
110
  features:
111
  - name: description
@@ -114,16 +114,16 @@ dataset_info:
114
  dtype: string
115
  splits:
116
  - name: train
117
- num_bytes: 160467163
118
  num_examples: 10164
119
  - name: validation
120
- num_bytes: 8667961
121
  num_examples: 565
122
  - name: test
123
- num_bytes: 8713720
124
  num_examples: 565
125
- download_size: 6447221554
126
- dataset_size: 177848844
127
  - config_name: e
128
  features:
129
  - name: description
@@ -132,16 +132,16 @@ dataset_info:
132
  dtype: string
133
  splits:
134
  - name: train
135
- num_bytes: 535567259
136
  num_examples: 34443
137
  - name: validation
138
- num_bytes: 28549964
139
  num_examples: 1914
140
  - name: test
141
- num_bytes: 29843613
142
  num_examples: 1914
143
- download_size: 6447221554
144
- dataset_size: 593960836
145
  - config_name: f
146
  features:
147
  - name: description
@@ -150,16 +150,16 @@ dataset_info:
150
  dtype: string
151
  splits:
152
  - name: train
153
- num_bytes: 1297707404
154
  num_examples: 85568
155
  - name: validation
156
- num_bytes: 72367466
157
  num_examples: 4754
158
  - name: test
159
- num_bytes: 71676041
160
  num_examples: 4754
161
- download_size: 6447221554
162
- dataset_size: 1441750911
163
  - config_name: g
164
  features:
165
  - name: description
@@ -168,16 +168,16 @@ dataset_info:
168
  dtype: string
169
  splits:
170
  - name: train
171
- num_bytes: 5571186559
172
  num_examples: 258935
173
  - name: validation
174
- num_bytes: 309182447
175
  num_examples: 14385
176
  - name: test
177
- num_bytes: 310624265
178
  num_examples: 14386
179
- download_size: 6447221554
180
- dataset_size: 6190993271
181
  - config_name: h
182
  features:
183
  - name: description
@@ -186,16 +186,16 @@ dataset_info:
186
  dtype: string
187
  splits:
188
  - name: train
189
- num_bytes: 4988365946
190
  num_examples: 257019
191
  - name: validation
192
- num_bytes: 275293153
193
  num_examples: 14279
194
  - name: test
195
- num_bytes: 274505113
196
  num_examples: 14279
197
- download_size: 6447221554
198
- dataset_size: 5538164212
199
  - config_name: y
200
  features:
201
  - name: description
@@ -204,16 +204,16 @@ dataset_info:
204
  dtype: string
205
  splits:
206
  - name: train
207
- num_bytes: 2263877990
208
  num_examples: 124397
209
  - name: validation
210
- num_bytes: 123505958
211
  num_examples: 6911
212
  - name: test
213
- num_bytes: 125090828
214
  num_examples: 6911
215
- download_size: 6447221554
216
- dataset_size: 2512474776
217
  ---
218
 
219
  # Dataset Card for Big Patent
@@ -252,17 +252,36 @@ dataset_info:
252
 
253
  ### Dataset Summary
254
 
255
- BIGPATENT, consisting of 1.3 million records of U.S. patent documents along with human written abstractive summaries. Each US patent application is filed under a Cooperative Patent Classification (CPC) code. There are nine such classification categories:
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
256
 
257
- - A (Human Necessities)
258
- - B (Performing Operations; Transporting)
259
- - C (Chemistry; Metallurgy)
260
- - D (Textiles; Paper)
261
- - E (Fixed Constructions)
262
- - F (Mechanical Engineering; Lightning; Heating; Weapons; Blasting)
263
- - G (Physics)
264
- - H (Electricity)
265
- - Y (General tagging of new or cross-sectional technology)
266
 
267
  ### Supported Tasks and Leaderboards
268
 
@@ -277,6 +296,12 @@ English
277
  ### Data Instances
278
 
279
  Each instance contains a pair of `description` and `abstract`. `description` is extracted from the Description section of the Patent while `abstract` is extracted from the Abstract section.
 
 
 
 
 
 
280
 
281
  ### Data Fields
282
 
42
  dtype: string
43
  splits:
44
  - name: train
45
+ num_bytes: 38367048389
46
  num_examples: 1207222
47
  - name: validation
48
+ num_bytes: 2115827002
49
  num_examples: 67068
50
  - name: test
51
+ num_bytes: 2129505280
52
  num_examples: 67072
53
+ download_size: 10142923776
54
+ dataset_size: 42612380671
55
  - config_name: a
56
  features:
57
  - name: description
60
  dtype: string
61
  splits:
62
  - name: train
63
+ num_bytes: 5683460620
64
  num_examples: 174134
65
  - name: validation
66
+ num_bytes: 313324505
67
  num_examples: 9674
68
  - name: test
69
+ num_bytes: 316633277
70
  num_examples: 9675
71
+ download_size: 10142923776
72
+ dataset_size: 6313418402
73
  - config_name: b
74
  features:
75
  - name: description
78
  dtype: string
79
  splits:
80
  - name: train
81
+ num_bytes: 4236070976
82
  num_examples: 161520
83
  - name: validation
84
+ num_bytes: 234425138
85
  num_examples: 8973
86
  - name: test
87
+ num_bytes: 231538734
88
  num_examples: 8974
89
+ download_size: 10142923776
90
+ dataset_size: 4702034848
91
  - config_name: c
92
  features:
93
  - name: description
96
  dtype: string
97
  splits:
98
  - name: train
99
+ num_bytes: 4506249306
100
  num_examples: 101042
101
  - name: validation
102
+ num_bytes: 244684775
103
  num_examples: 5613
104
  - name: test
105
+ num_bytes: 252566793
106
  num_examples: 5614
107
+ download_size: 10142923776
108
+ dataset_size: 5003500874
109
  - config_name: d
110
  features:
111
  - name: description
114
  dtype: string
115
  splits:
116
  - name: train
117
+ num_bytes: 264717412
118
  num_examples: 10164
119
  - name: validation
120
+ num_bytes: 14560482
121
  num_examples: 565
122
  - name: test
123
+ num_bytes: 14403430
124
  num_examples: 565
125
+ download_size: 10142923776
126
+ dataset_size: 293681324
127
  - config_name: e
128
  features:
129
  - name: description
132
  dtype: string
133
  splits:
134
  - name: train
135
+ num_bytes: 881101433
136
  num_examples: 34443
137
  - name: validation
138
+ num_bytes: 48646158
139
  num_examples: 1914
140
  - name: test
141
+ num_bytes: 48586429
142
  num_examples: 1914
143
+ download_size: 10142923776
144
+ dataset_size: 978334020
145
  - config_name: f
146
  features:
147
  - name: description
150
  dtype: string
151
  splits:
152
  - name: train
153
+ num_bytes: 2146383473
154
  num_examples: 85568
155
  - name: validation
156
+ num_bytes: 119632631
157
  num_examples: 4754
158
  - name: test
159
+ num_bytes: 119596303
160
  num_examples: 4754
161
+ download_size: 10142923776
162
+ dataset_size: 2385612407
163
  - config_name: g
164
  features:
165
  - name: description
168
  dtype: string
169
  splits:
170
  - name: train
171
+ num_bytes: 8877854206
172
  num_examples: 258935
173
  - name: validation
174
+ num_bytes: 492581177
175
  num_examples: 14385
176
  - name: test
177
+ num_bytes: 496324853
178
  num_examples: 14386
179
+ download_size: 10142923776
180
+ dataset_size: 9866760236
181
  - config_name: h
182
  features:
183
  - name: description
186
  dtype: string
187
  splits:
188
  - name: train
189
+ num_bytes: 8075621958
190
  num_examples: 257019
191
  - name: validation
192
+ num_bytes: 447602356
193
  num_examples: 14279
194
  - name: test
195
+ num_bytes: 445460513
196
  num_examples: 14279
197
+ download_size: 10142923776
198
+ dataset_size: 8968684827
199
  - config_name: y
200
  features:
201
  - name: description
204
  dtype: string
205
  splits:
206
  - name: train
207
+ num_bytes: 3695589005
208
  num_examples: 124397
209
  - name: validation
210
+ num_bytes: 200369780
211
  num_examples: 6911
212
  - name: test
213
+ num_bytes: 204394948
214
  num_examples: 6911
215
+ download_size: 10142923776
216
+ dataset_size: 4100353733
217
  ---
218
 
219
  # Dataset Card for Big Patent
252
 
253
  ### Dataset Summary
254
 
255
+ BIGPATENT, consisting of 1.3 million records of U.S. patent documents along with human written abstractive summaries.
256
+ Each US patent application is filed under a Cooperative Patent Classification (CPC) code.
257
+ There are nine such classification categories:
258
+ - a: Human Necessities
259
+ - b: Performing Operations; Transporting
260
+ - c: Chemistry; Metallurgy
261
+ - d: Textiles; Paper
262
+ - e: Fixed Constructions
263
+ - f: Mechanical Engineering; Lightning; Heating; Weapons; Blasting
264
+ - g: Physics
265
+ - h: Electricity
266
+ - y: General tagging of new or cross-sectional technology
267
+
268
+ Current defaults are 2.1.2 version (fix update to cased raw strings) and 'all' CPC codes:
269
+ ```python
270
+ from datasets import load_dataset
271
+
272
+ ds = load_dataset("big_patent") # default is 'all' CPC codes
273
+ ds = load_dataset("big_patent", "all") # the same as above
274
+ ds = load_dataset("big_patent", "a") # only 'a' CPC codes
275
+ ds = load_dataset("big_patent", codes=["a", "b"])
276
+ ```
277
+
278
+ To use 1.0.0 version (lower cased tokenized words), pass both parameters `codes` and `version`:
279
+ ```python
280
+ ds = load_dataset("big_patent", codes="all", version="1.0.0")
281
+ ds = load_dataset("big_patent", codes="a", version="1.0.0")
282
+ ds = load_dataset("big_patent", codes=["a", "b"], version="1.0.0")
283
+ ```
284
 
 
 
 
 
 
 
 
 
 
285
 
286
  ### Supported Tasks and Leaderboards
287
 
296
  ### Data Instances
297
 
298
  Each instance contains a pair of `description` and `abstract`. `description` is extracted from the Description section of the Patent while `abstract` is extracted from the Abstract section.
299
+ ```
300
+ {
301
+ 'description': 'FIELD OF THE INVENTION \n [0001] This invention relates to novel calcium phosphate-coated implantable medical devices and processes of making same. The unique calcium-phosphate coated implantable medical devices minimize...',
302
+ 'abstract': 'This invention relates to novel calcium phosphate-coated implantable medical devices...'
303
+ }
304
+ ```
305
 
306
  ### Data Fields
307
 
big_patent.py CHANGED
@@ -15,7 +15,6 @@
15
 
16
  # Lint as: python3
17
  """BigPatent Dataset."""
18
- import glob
19
  import gzip
20
  import json
21
  import os
@@ -53,12 +52,8 @@ There are two features:
53
 
54
  _LICENSE = "Creative Commons Attribution 4.0 International"
55
 
56
- _REPO = "https://huggingface.co/datasets/big_patent/resolve/main/data"
57
- _URLS = {
58
- "train": f"{_REPO}/train.zip",
59
- "validation": f"{_REPO}/val.zip",
60
- "test": f"{_REPO}/test.zip",
61
- }
62
 
63
  _DOCUMENT = "description"
64
  _SUMMARY = "abstract"
@@ -78,38 +73,44 @@ _CPC_DESCRIPTION = {
78
  # Available versions:
79
  # 1.0.0 lower cased tokenized words.
80
  # 2.0.0 cased raw strings.
81
- # 2.1.0 cased raw strings (fixed).
82
- # TODO Add raw string versions
83
 
84
- _VERSION = "1.0.0"
85
 
86
 
87
  class BigPatentConfig(datasets.BuilderConfig):
88
  """BuilderConfig for BigPatent."""
89
 
90
- def __init__(self, *args, cpc_codes=None, **kwargs):
91
  """BuilderConfig for BigPatent.
92
  Args:
93
- cpc_codes: str, cpc_codes
 
94
  **kwargs: keyword arguments forwarded to super.
95
  """
96
- super().__init__(*args, version=_VERSION, **kwargs)
97
- self.cpc_codes = cpc_codes
 
 
 
 
 
 
 
98
 
99
 
100
  class BigPatent(datasets.GeneratorBasedBuilder):
101
  """BigPatent datasets."""
102
 
 
103
  BUILDER_CONFIGS = [
104
  BigPatentConfig(
105
- cpc_codes=list(_CPC_DESCRIPTION),
106
- name="all",
107
  description="Patents under all categories.",
108
  ),
109
  ] + [
110
- BigPatentConfig( # pylint:disable=g-complex-comprehension
111
- cpc_codes=[k],
112
- name=k,
113
  description=f"Patents under Cooperative Patent Classification (CPC) {k}: {v}",
114
  )
115
  for k, v in sorted(_CPC_DESCRIPTION.items())
@@ -129,22 +130,30 @@ class BigPatent(datasets.GeneratorBasedBuilder):
129
 
130
  def _split_generators(self, dl_manager):
131
  """Returns SplitGenerators."""
132
- dl_paths = dl_manager.download_and_extract(_URLS)
133
- split_dirs = {datasets.Split.TRAIN: "train", datasets.Split.VALIDATION: "val", datasets.Split.TEST: "test"}
 
 
 
 
 
 
 
 
 
134
  return [
135
  datasets.SplitGenerator(
136
  name=split,
137
- gen_kwargs={"path": dl_paths[split], "split_dir": split_dirs[split]},
138
  )
139
- for split in split_dirs
140
  ]
141
 
142
- def _generate_examples(self, path=None, split_dir=None):
143
  """Yields examples."""
144
- for cpc_code in self.config.cpc_codes:
145
- filenames = glob.glob(os.path.join(path, split_dir, cpc_code, "*"))
146
- for filename in sorted(filenames):
147
- with open(filename, "rb") as fin:
148
  fin = gzip.GzipFile(fileobj=fin)
149
  for row in fin:
150
  json_obj = json.loads(row)
15
 
16
  # Lint as: python3
17
  """BigPatent Dataset."""
 
18
  import gzip
19
  import json
20
  import os
52
 
53
  _LICENSE = "Creative Commons Attribution 4.0 International"
54
 
55
+ _SPLIT_NAMES = {datasets.Split.TRAIN: "train", datasets.Split.VALIDATION: "val", datasets.Split.TEST: "test"}
56
+ _URL = "data/{version}/{split_name}.zip"
 
 
 
 
57
 
58
  _DOCUMENT = "description"
59
  _SUMMARY = "abstract"
73
  # Available versions:
74
  # 1.0.0 lower cased tokenized words.
75
  # 2.0.0 cased raw strings.
76
+ # 2.1.2 cased raw strings (fixed).
 
77
 
78
+ _VERSION = "2.1.2"
79
 
80
 
81
  class BigPatentConfig(datasets.BuilderConfig):
82
  """BuilderConfig for BigPatent."""
83
 
84
+ def __init__(self, codes="all", version=_VERSION, **kwargs):
85
  """BuilderConfig for BigPatent.
86
  Args:
87
+ codes (str or list, default 'all'): CPC codes. Either 'all' or a combination
88
+ of {'a', 'b', 'c', 'd', 'e', 'f', 'g', 'h', 'y'}.
89
  **kwargs: keyword arguments forwarded to super.
90
  """
91
+ if isinstance(codes, str):
92
+ codes = [codes]
93
+ name = "+".join(codes)
94
+ if name == "all":
95
+ codes = list(_CPC_DESCRIPTION)
96
+ if version != _VERSION:
97
+ name = f"{name}-{version}"
98
+ super().__init__(name=name, version=version, **kwargs)
99
+ self.codes = codes
100
 
101
 
102
  class BigPatent(datasets.GeneratorBasedBuilder):
103
  """BigPatent datasets."""
104
 
105
+ BUILDER_CONFIG_CLASS = BigPatentConfig
106
  BUILDER_CONFIGS = [
107
  BigPatentConfig(
108
+ codes="all",
 
109
  description="Patents under all categories.",
110
  ),
111
  ] + [
112
+ BigPatentConfig(
113
+ codes=k,
 
114
  description=f"Patents under Cooperative Patent Classification (CPC) {k}: {v}",
115
  )
116
  for k, v in sorted(_CPC_DESCRIPTION.items())
130
 
131
  def _split_generators(self, dl_manager):
132
  """Returns SplitGenerators."""
133
+ urls = {
134
+ split: _URL.format(version=self.config.version, split_name=split_name)
135
+ for split, split_name in _SPLIT_NAMES.items()
136
+ }
137
+ dl_paths = dl_manager.download_and_extract(urls)
138
+ paths = {
139
+ split: [
140
+ dl_manager.iter_files(os.path.join(dl_paths[split], split_name, code)) for code in self.config.codes
141
+ ]
142
+ for split, split_name in _SPLIT_NAMES.items()
143
+ }
144
  return [
145
  datasets.SplitGenerator(
146
  name=split,
147
+ gen_kwargs={"paths": paths[split]},
148
  )
149
+ for split in _SPLIT_NAMES
150
  ]
151
 
152
+ def _generate_examples(self, paths=None):
153
  """Yields examples."""
154
+ for paths_per_code in paths:
155
+ for path in paths_per_code:
156
+ with open(path, "rb") as fin:
 
157
  fin = gzip.GzipFile(fileobj=fin)
158
  for row in fin:
159
  json_obj = json.loads(row)
data/{test.zip → 1.0.0/test.zip} RENAMED
File without changes
data/{train.zip → 1.0.0/train.zip} RENAMED
File without changes
data/{val.zip → 1.0.0/val.zip} RENAMED
File without changes
data/2.1.2/test.zip ADDED
@@ -0,0 +1,3 @@
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:614821d8d8c92451ba495e0b381f157a0e0b36977a70f2d4cb3ae4b6b706539a
3
+ size 507619065
data/2.1.2/train.zip ADDED
@@ -0,0 +1,3 @@
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:59cae1439b01102faa5b9e1bb3061867ba292a6093cf66ee3d1bc33030d7a91f
3
+ size 9129045386
data/2.1.2/val.zip ADDED
@@ -0,0 +1,3 @@
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9206de008607095bdb29f405ef9b525565556860d015cbd4261154644b51ce9b
3
+ size 506259325
dataset_infos.json CHANGED
@@ -1 +1 @@
1
- {"all": {"description": "\nBIGPATENT, consisting of 1.3 million records of U.S. patent documents\nalong with human written abstractive summaries.\nEach US patent application is filed under a Cooperative Patent Classification\n(CPC) code. There are nine such classification categories:\nA (Human Necessities), B (Performing Operations; Transporting),\nC (Chemistry; Metallurgy), D (Textiles; Paper), E (Fixed Constructions),\nF (Mechanical Engineering; Lightning; Heating; Weapons; Blasting),\nG (Physics), H (Electricity), and\nY (General tagging of new or cross-sectional technology)\nThere are two features:\n - description: detailed description of patent.\n - abstract: Patent abastract.\n", "citation": "\n@misc{sharma2019bigpatent,\n title={BIGPATENT: A Large-Scale Dataset for Abstractive and Coherent Summarization},\n author={Eva Sharma and Chen Li and Lu Wang},\n year={2019},\n eprint={1906.03741},\n archivePrefix={arXiv},\n primaryClass={cs.CL}\n}\n", "homepage": "https://evasharma.github.io/bigpatent/", "license": "Creative Commons Attribution 4.0 International", "features": {"description": {"dtype": "string", "id": null, "_type": "Value"}, "abstract": {"dtype": "string", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": {"input": "description", "output": "abstract"}, "task_templates": null, "builder_name": "big_patent", "config_name": "all", "version": {"version_str": "1.0.0", "description": null, "major": 1, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 23363518650, "num_examples": 1207222, "dataset_name": "big_patent"}, "validation": {"name": "validation", "num_bytes": 1290154487, "num_examples": 67068, "dataset_name": "big_patent"}, "test": {"name": "test", "num_bytes": 1296234391, "num_examples": 67072, "dataset_name": "big_patent"}}, "download_checksums": {"https://huggingface.co/datasets/big_patent/resolve/main/data/train.zip": {"num_bytes": 5802341237, "checksum": "89831e047474822ff7be521707e9a58bd9acd4d359a194cc564e74012a44185d"}, "https://huggingface.co/datasets/big_patent/resolve/main/data/val.zip": {"num_bytes": 321731760, "checksum": "c8a3c745cac4216bd32a54149dbc044ea397e405a08b230041a554e3eb75080d"}, "https://huggingface.co/datasets/big_patent/resolve/main/data/test.zip": {"num_bytes": 323148557, "checksum": "ce605cafb69b1757276326c51d18a1c09275b1911252dd95e8e0dcbf386d1b77"}}, "download_size": 6447221554, "post_processing_size": null, "dataset_size": 25949907528, "size_in_bytes": 32397129082}, "a": {"description": "\nBIGPATENT, consisting of 1.3 million records of U.S. patent documents\nalong with human written abstractive summaries.\nEach US patent application is filed under a Cooperative Patent Classification\n(CPC) code. There are nine such classification categories:\nA (Human Necessities), B (Performing Operations; Transporting),\nC (Chemistry; Metallurgy), D (Textiles; Paper), E (Fixed Constructions),\nF (Mechanical Engineering; Lightning; Heating; Weapons; Blasting),\nG (Physics), H (Electricity), and\nY (General tagging of new or cross-sectional technology)\nThere are two features:\n - description: detailed description of patent.\n - abstract: Patent abastract.\n", "citation": "\n@misc{sharma2019bigpatent,\n title={BIGPATENT: A Large-Scale Dataset for Abstractive and Coherent Summarization},\n author={Eva Sharma and Chen Li and Lu Wang},\n year={2019},\n eprint={1906.03741},\n archivePrefix={arXiv},\n primaryClass={cs.CL}\n}\n", "homepage": "https://evasharma.github.io/bigpatent/", "license": "Creative Commons Attribution 4.0 International", "features": {"description": {"dtype": "string", "id": null, "_type": "Value"}, "abstract": {"dtype": "string", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": {"input": "description", "output": "abstract"}, "task_templates": null, "builder_name": "big_patent", "config_name": "a", "version": {"version_str": "1.0.0", "description": null, "major": 1, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 3329778447, "num_examples": 174134, "dataset_name": "big_patent"}, "validation": {"name": "validation", "num_bytes": 184116486, "num_examples": 9674, "dataset_name": "big_patent"}, "test": {"name": "test", "num_bytes": 185987552, "num_examples": 9675, "dataset_name": "big_patent"}}, "download_checksums": {"https://huggingface.co/datasets/big_patent/resolve/main/data/train.zip": {"num_bytes": 5802341237, "checksum": "89831e047474822ff7be521707e9a58bd9acd4d359a194cc564e74012a44185d"}, "https://huggingface.co/datasets/big_patent/resolve/main/data/val.zip": {"num_bytes": 321731760, "checksum": "c8a3c745cac4216bd32a54149dbc044ea397e405a08b230041a554e3eb75080d"}, "https://huggingface.co/datasets/big_patent/resolve/main/data/test.zip": {"num_bytes": 323148557, "checksum": "ce605cafb69b1757276326c51d18a1c09275b1911252dd95e8e0dcbf386d1b77"}}, "download_size": 6447221554, "post_processing_size": null, "dataset_size": 3699882485, "size_in_bytes": 10147104039}, "b": {"description": "\nBIGPATENT, consisting of 1.3 million records of U.S. patent documents\nalong with human written abstractive summaries.\nEach US patent application is filed under a Cooperative Patent Classification\n(CPC) code. There are nine such classification categories:\nA (Human Necessities), B (Performing Operations; Transporting),\nC (Chemistry; Metallurgy), D (Textiles; Paper), E (Fixed Constructions),\nF (Mechanical Engineering; Lightning; Heating; Weapons; Blasting),\nG (Physics), H (Electricity), and\nY (General tagging of new or cross-sectional technology)\nThere are two features:\n - description: detailed description of patent.\n - abstract: Patent abastract.\n", "citation": "\n@misc{sharma2019bigpatent,\n title={BIGPATENT: A Large-Scale Dataset for Abstractive and Coherent Summarization},\n author={Eva Sharma and Chen Li and Lu Wang},\n year={2019},\n eprint={1906.03741},\n archivePrefix={arXiv},\n primaryClass={cs.CL}\n}\n", "homepage": "https://evasharma.github.io/bigpatent/", "license": "Creative Commons Attribution 4.0 International", "features": {"description": {"dtype": "string", "id": null, "_type": "Value"}, "abstract": {"dtype": "string", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": {"input": "description", "output": "abstract"}, "task_templates": null, "builder_name": "big_patent", "config_name": "b", "version": {"version_str": "1.0.0", "description": null, "major": 1, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 2574594655, "num_examples": 161520, "dataset_name": "big_patent"}, "validation": {"name": "validation", "num_bytes": 143029380, "num_examples": 8973, "dataset_name": "big_patent"}, "test": {"name": "test", "num_bytes": 140741033, "num_examples": 8974, "dataset_name": "big_patent"}}, "download_checksums": {"https://huggingface.co/datasets/big_patent/resolve/main/data/train.zip": {"num_bytes": 5802341237, "checksum": "89831e047474822ff7be521707e9a58bd9acd4d359a194cc564e74012a44185d"}, "https://huggingface.co/datasets/big_patent/resolve/main/data/val.zip": {"num_bytes": 321731760, "checksum": "c8a3c745cac4216bd32a54149dbc044ea397e405a08b230041a554e3eb75080d"}, "https://huggingface.co/datasets/big_patent/resolve/main/data/test.zip": {"num_bytes": 323148557, "checksum": "ce605cafb69b1757276326c51d18a1c09275b1911252dd95e8e0dcbf386d1b77"}}, "download_size": 6447221554, "post_processing_size": null, "dataset_size": 2858365068, "size_in_bytes": 9305586622}, "c": {"description": "\nBIGPATENT, consisting of 1.3 million records of U.S. patent documents\nalong with human written abstractive summaries.\nEach US patent application is filed under a Cooperative Patent Classification\n(CPC) code. There are nine such classification categories:\nA (Human Necessities), B (Performing Operations; Transporting),\nC (Chemistry; Metallurgy), D (Textiles; Paper), E (Fixed Constructions),\nF (Mechanical Engineering; Lightning; Heating; Weapons; Blasting),\nG (Physics), H (Electricity), and\nY (General tagging of new or cross-sectional technology)\nThere are two features:\n - description: detailed description of patent.\n - abstract: Patent abastract.\n", "citation": "\n@misc{sharma2019bigpatent,\n title={BIGPATENT: A Large-Scale Dataset for Abstractive and Coherent Summarization},\n author={Eva Sharma and Chen Li and Lu Wang},\n year={2019},\n eprint={1906.03741},\n archivePrefix={arXiv},\n primaryClass={cs.CL}\n}\n", "homepage": "https://evasharma.github.io/bigpatent/", "license": "Creative Commons Attribution 4.0 International", "features": {"description": {"dtype": "string", "id": null, "_type": "Value"}, "abstract": {"dtype": "string", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": {"input": "description", "output": "abstract"}, "task_templates": null, "builder_name": "big_patent", "config_name": "c", "version": {"version_str": "1.0.0", "description": null, "major": 1, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 2641973267, "num_examples": 101042, "dataset_name": "big_patent"}, "validation": {"name": "validation", "num_bytes": 145441704, "num_examples": 5613, "dataset_name": "big_patent"}, "test": {"name": "test", "num_bytes": 149052258, "num_examples": 5614, "dataset_name": "big_patent"}}, "download_checksums": {"https://huggingface.co/datasets/big_patent/resolve/main/data/train.zip": {"num_bytes": 5802341237, "checksum": "89831e047474822ff7be521707e9a58bd9acd4d359a194cc564e74012a44185d"}, "https://huggingface.co/datasets/big_patent/resolve/main/data/val.zip": {"num_bytes": 321731760, "checksum": "c8a3c745cac4216bd32a54149dbc044ea397e405a08b230041a554e3eb75080d"}, "https://huggingface.co/datasets/big_patent/resolve/main/data/test.zip": {"num_bytes": 323148557, "checksum": "ce605cafb69b1757276326c51d18a1c09275b1911252dd95e8e0dcbf386d1b77"}}, "download_size": 6447221554, "post_processing_size": null, "dataset_size": 2936467229, "size_in_bytes": 9383688783}, "d": {"description": "\nBIGPATENT, consisting of 1.3 million records of U.S. patent documents\nalong with human written abstractive summaries.\nEach US patent application is filed under a Cooperative Patent Classification\n(CPC) code. There are nine such classification categories:\nA (Human Necessities), B (Performing Operations; Transporting),\nC (Chemistry; Metallurgy), D (Textiles; Paper), E (Fixed Constructions),\nF (Mechanical Engineering; Lightning; Heating; Weapons; Blasting),\nG (Physics), H (Electricity), and\nY (General tagging of new or cross-sectional technology)\nThere are two features:\n - description: detailed description of patent.\n - abstract: Patent abastract.\n", "citation": "\n@misc{sharma2019bigpatent,\n title={BIGPATENT: A Large-Scale Dataset for Abstractive and Coherent Summarization},\n author={Eva Sharma and Chen Li and Lu Wang},\n year={2019},\n eprint={1906.03741},\n archivePrefix={arXiv},\n primaryClass={cs.CL}\n}\n", "homepage": "https://evasharma.github.io/bigpatent/", "license": "Creative Commons Attribution 4.0 International", "features": {"description": {"dtype": "string", "id": null, "_type": "Value"}, "abstract": {"dtype": "string", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": {"input": "description", "output": "abstract"}, "task_templates": null, "builder_name": "big_patent", "config_name": "d", "version": {"version_str": "1.0.0", "description": null, "major": 1, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 160467163, "num_examples": 10164, "dataset_name": "big_patent"}, "validation": {"name": "validation", "num_bytes": 8667961, "num_examples": 565, "dataset_name": "big_patent"}, "test": {"name": "test", "num_bytes": 8713720, "num_examples": 565, "dataset_name": "big_patent"}}, "download_checksums": {"https://huggingface.co/datasets/big_patent/resolve/main/data/train.zip": {"num_bytes": 5802341237, "checksum": "89831e047474822ff7be521707e9a58bd9acd4d359a194cc564e74012a44185d"}, "https://huggingface.co/datasets/big_patent/resolve/main/data/val.zip": {"num_bytes": 321731760, "checksum": "c8a3c745cac4216bd32a54149dbc044ea397e405a08b230041a554e3eb75080d"}, "https://huggingface.co/datasets/big_patent/resolve/main/data/test.zip": {"num_bytes": 323148557, "checksum": "ce605cafb69b1757276326c51d18a1c09275b1911252dd95e8e0dcbf386d1b77"}}, "download_size": 6447221554, "post_processing_size": null, "dataset_size": 177848844, "size_in_bytes": 6625070398}, "e": {"description": "\nBIGPATENT, consisting of 1.3 million records of U.S. patent documents\nalong with human written abstractive summaries.\nEach US patent application is filed under a Cooperative Patent Classification\n(CPC) code. There are nine such classification categories:\nA (Human Necessities), B (Performing Operations; Transporting),\nC (Chemistry; Metallurgy), D (Textiles; Paper), E (Fixed Constructions),\nF (Mechanical Engineering; Lightning; Heating; Weapons; Blasting),\nG (Physics), H (Electricity), and\nY (General tagging of new or cross-sectional technology)\nThere are two features:\n - description: detailed description of patent.\n - abstract: Patent abastract.\n", "citation": "\n@misc{sharma2019bigpatent,\n title={BIGPATENT: A Large-Scale Dataset for Abstractive and Coherent Summarization},\n author={Eva Sharma and Chen Li and Lu Wang},\n year={2019},\n eprint={1906.03741},\n archivePrefix={arXiv},\n primaryClass={cs.CL}\n}\n", "homepage": "https://evasharma.github.io/bigpatent/", "license": "Creative Commons Attribution 4.0 International", "features": {"description": {"dtype": "string", "id": null, "_type": "Value"}, "abstract": {"dtype": "string", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": {"input": "description", "output": "abstract"}, "task_templates": null, "builder_name": "big_patent", "config_name": "e", "version": {"version_str": "1.0.0", "description": null, "major": 1, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 535567259, "num_examples": 34443, "dataset_name": "big_patent"}, "validation": {"name": "validation", "num_bytes": 28549964, "num_examples": 1914, "dataset_name": "big_patent"}, "test": {"name": "test", "num_bytes": 29843613, "num_examples": 1914, "dataset_name": "big_patent"}}, "download_checksums": {"https://huggingface.co/datasets/big_patent/resolve/main/data/train.zip": {"num_bytes": 5802341237, "checksum": "89831e047474822ff7be521707e9a58bd9acd4d359a194cc564e74012a44185d"}, "https://huggingface.co/datasets/big_patent/resolve/main/data/val.zip": {"num_bytes": 321731760, "checksum": "c8a3c745cac4216bd32a54149dbc044ea397e405a08b230041a554e3eb75080d"}, "https://huggingface.co/datasets/big_patent/resolve/main/data/test.zip": {"num_bytes": 323148557, "checksum": "ce605cafb69b1757276326c51d18a1c09275b1911252dd95e8e0dcbf386d1b77"}}, "download_size": 6447221554, "post_processing_size": null, "dataset_size": 593960836, "size_in_bytes": 7041182390}, "f": {"description": "\nBIGPATENT, consisting of 1.3 million records of U.S. patent documents\nalong with human written abstractive summaries.\nEach US patent application is filed under a Cooperative Patent Classification\n(CPC) code. There are nine such classification categories:\nA (Human Necessities), B (Performing Operations; Transporting),\nC (Chemistry; Metallurgy), D (Textiles; Paper), E (Fixed Constructions),\nF (Mechanical Engineering; Lightning; Heating; Weapons; Blasting),\nG (Physics), H (Electricity), and\nY (General tagging of new or cross-sectional technology)\nThere are two features:\n - description: detailed description of patent.\n - abstract: Patent abastract.\n", "citation": "\n@misc{sharma2019bigpatent,\n title={BIGPATENT: A Large-Scale Dataset for Abstractive and Coherent Summarization},\n author={Eva Sharma and Chen Li and Lu Wang},\n year={2019},\n eprint={1906.03741},\n archivePrefix={arXiv},\n primaryClass={cs.CL}\n}\n", "homepage": "https://evasharma.github.io/bigpatent/", "license": "Creative Commons Attribution 4.0 International", "features": {"description": {"dtype": "string", "id": null, "_type": "Value"}, "abstract": {"dtype": "string", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": {"input": "description", "output": "abstract"}, "task_templates": null, "builder_name": "big_patent", "config_name": "f", "version": {"version_str": "1.0.0", "description": null, "major": 1, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 1297707404, "num_examples": 85568, "dataset_name": "big_patent"}, "validation": {"name": "validation", "num_bytes": 72367466, "num_examples": 4754, "dataset_name": "big_patent"}, "test": {"name": "test", "num_bytes": 71676041, "num_examples": 4754, "dataset_name": "big_patent"}}, "download_checksums": {"https://huggingface.co/datasets/big_patent/resolve/main/data/train.zip": {"num_bytes": 5802341237, "checksum": "89831e047474822ff7be521707e9a58bd9acd4d359a194cc564e74012a44185d"}, "https://huggingface.co/datasets/big_patent/resolve/main/data/val.zip": {"num_bytes": 321731760, "checksum": "c8a3c745cac4216bd32a54149dbc044ea397e405a08b230041a554e3eb75080d"}, "https://huggingface.co/datasets/big_patent/resolve/main/data/test.zip": {"num_bytes": 323148557, "checksum": "ce605cafb69b1757276326c51d18a1c09275b1911252dd95e8e0dcbf386d1b77"}}, "download_size": 6447221554, "post_processing_size": null, "dataset_size": 1441750911, "size_in_bytes": 7888972465}, "g": {"description": "\nBIGPATENT, consisting of 1.3 million records of U.S. patent documents\nalong with human written abstractive summaries.\nEach US patent application is filed under a Cooperative Patent Classification\n(CPC) code. There are nine such classification categories:\nA (Human Necessities), B (Performing Operations; Transporting),\nC (Chemistry; Metallurgy), D (Textiles; Paper), E (Fixed Constructions),\nF (Mechanical Engineering; Lightning; Heating; Weapons; Blasting),\nG (Physics), H (Electricity), and\nY (General tagging of new or cross-sectional technology)\nThere are two features:\n - description: detailed description of patent.\n - abstract: Patent abastract.\n", "citation": "\n@misc{sharma2019bigpatent,\n title={BIGPATENT: A Large-Scale Dataset for Abstractive and Coherent Summarization},\n author={Eva Sharma and Chen Li and Lu Wang},\n year={2019},\n eprint={1906.03741},\n archivePrefix={arXiv},\n primaryClass={cs.CL}\n}\n", "homepage": "https://evasharma.github.io/bigpatent/", "license": "Creative Commons Attribution 4.0 International", "features": {"description": {"dtype": "string", "id": null, "_type": "Value"}, "abstract": {"dtype": "string", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": {"input": "description", "output": "abstract"}, "task_templates": null, "builder_name": "big_patent", "config_name": "g", "version": {"version_str": "1.0.0", "description": null, "major": 1, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 5571186559, "num_examples": 258935, "dataset_name": "big_patent"}, "validation": {"name": "validation", "num_bytes": 309182447, "num_examples": 14385, "dataset_name": "big_patent"}, "test": {"name": "test", "num_bytes": 310624265, "num_examples": 14386, "dataset_name": "big_patent"}}, "download_checksums": {"https://huggingface.co/datasets/big_patent/resolve/main/data/train.zip": {"num_bytes": 5802341237, "checksum": "89831e047474822ff7be521707e9a58bd9acd4d359a194cc564e74012a44185d"}, "https://huggingface.co/datasets/big_patent/resolve/main/data/val.zip": {"num_bytes": 321731760, "checksum": "c8a3c745cac4216bd32a54149dbc044ea397e405a08b230041a554e3eb75080d"}, "https://huggingface.co/datasets/big_patent/resolve/main/data/test.zip": {"num_bytes": 323148557, "checksum": "ce605cafb69b1757276326c51d18a1c09275b1911252dd95e8e0dcbf386d1b77"}}, "download_size": 6447221554, "post_processing_size": null, "dataset_size": 6190993271, "size_in_bytes": 12638214825}, "h": {"description": "\nBIGPATENT, consisting of 1.3 million records of U.S. patent documents\nalong with human written abstractive summaries.\nEach US patent application is filed under a Cooperative Patent Classification\n(CPC) code. There are nine such classification categories:\nA (Human Necessities), B (Performing Operations; Transporting),\nC (Chemistry; Metallurgy), D (Textiles; Paper), E (Fixed Constructions),\nF (Mechanical Engineering; Lightning; Heating; Weapons; Blasting),\nG (Physics), H (Electricity), and\nY (General tagging of new or cross-sectional technology)\nThere are two features:\n - description: detailed description of patent.\n - abstract: Patent abastract.\n", "citation": "\n@misc{sharma2019bigpatent,\n title={BIGPATENT: A Large-Scale Dataset for Abstractive and Coherent Summarization},\n author={Eva Sharma and Chen Li and Lu Wang},\n year={2019},\n eprint={1906.03741},\n archivePrefix={arXiv},\n primaryClass={cs.CL}\n}\n", "homepage": "https://evasharma.github.io/bigpatent/", "license": "Creative Commons Attribution 4.0 International", "features": {"description": {"dtype": "string", "id": null, "_type": "Value"}, "abstract": {"dtype": "string", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": {"input": "description", "output": "abstract"}, "task_templates": null, "builder_name": "big_patent", "config_name": "h", "version": {"version_str": "1.0.0", "description": null, "major": 1, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 4988365946, "num_examples": 257019, "dataset_name": "big_patent"}, "validation": {"name": "validation", "num_bytes": 275293153, "num_examples": 14279, "dataset_name": "big_patent"}, "test": {"name": "test", "num_bytes": 274505113, "num_examples": 14279, "dataset_name": "big_patent"}}, "download_checksums": {"https://huggingface.co/datasets/big_patent/resolve/main/data/train.zip": {"num_bytes": 5802341237, "checksum": "89831e047474822ff7be521707e9a58bd9acd4d359a194cc564e74012a44185d"}, "https://huggingface.co/datasets/big_patent/resolve/main/data/val.zip": {"num_bytes": 321731760, "checksum": "c8a3c745cac4216bd32a54149dbc044ea397e405a08b230041a554e3eb75080d"}, "https://huggingface.co/datasets/big_patent/resolve/main/data/test.zip": {"num_bytes": 323148557, "checksum": "ce605cafb69b1757276326c51d18a1c09275b1911252dd95e8e0dcbf386d1b77"}}, "download_size": 6447221554, "post_processing_size": null, "dataset_size": 5538164212, "size_in_bytes": 11985385766}, "y": {"description": "\nBIGPATENT, consisting of 1.3 million records of U.S. patent documents\nalong with human written abstractive summaries.\nEach US patent application is filed under a Cooperative Patent Classification\n(CPC) code. There are nine such classification categories:\nA (Human Necessities), B (Performing Operations; Transporting),\nC (Chemistry; Metallurgy), D (Textiles; Paper), E (Fixed Constructions),\nF (Mechanical Engineering; Lightning; Heating; Weapons; Blasting),\nG (Physics), H (Electricity), and\nY (General tagging of new or cross-sectional technology)\nThere are two features:\n - description: detailed description of patent.\n - abstract: Patent abastract.\n", "citation": "\n@misc{sharma2019bigpatent,\n title={BIGPATENT: A Large-Scale Dataset for Abstractive and Coherent Summarization},\n author={Eva Sharma and Chen Li and Lu Wang},\n year={2019},\n eprint={1906.03741},\n archivePrefix={arXiv},\n primaryClass={cs.CL}\n}\n", "homepage": "https://evasharma.github.io/bigpatent/", "license": "Creative Commons Attribution 4.0 International", "features": {"description": {"dtype": "string", "id": null, "_type": "Value"}, "abstract": {"dtype": "string", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": {"input": "description", "output": "abstract"}, "task_templates": null, "builder_name": "big_patent", "config_name": "y", "version": "1.0.0", "splits": {"train": {"name": "train", "num_bytes": 2263877990, "num_examples": 124397, "dataset_name": "big_patent"}, "validation": {"name": "validation", "num_bytes": 123505958, "num_examples": 6911, "dataset_name": "big_patent"}, "test": {"name": "test", "num_bytes": 125090828, "num_examples": 6911, "dataset_name": "big_patent"}}, "download_checksums": {"https://huggingface.co/datasets/big_patent/resolve/main/data/train.zip": {"num_bytes": 5802341237, "checksum": "89831e047474822ff7be521707e9a58bd9acd4d359a194cc564e74012a44185d"}, "https://huggingface.co/datasets/big_patent/resolve/main/data/val.zip": {"num_bytes": 321731760, "checksum": "c8a3c745cac4216bd32a54149dbc044ea397e405a08b230041a554e3eb75080d"}, "https://huggingface.co/datasets/big_patent/resolve/main/data/test.zip": {"num_bytes": 323148557, "checksum": "ce605cafb69b1757276326c51d18a1c09275b1911252dd95e8e0dcbf386d1b77"}}, "download_size": 6447221554, "post_processing_size": null, "dataset_size": 2512474776, "size_in_bytes": 8959696330}}
1
+ {"all": {"description": "\nBIGPATENT, consisting of 1.3 million records of U.S. patent documents\nalong with human written abstractive summaries.\nEach US patent application is filed under a Cooperative Patent Classification\n(CPC) code. There are nine such classification categories:\nA (Human Necessities), B (Performing Operations; Transporting),\nC (Chemistry; Metallurgy), D (Textiles; Paper), E (Fixed Constructions),\nF (Mechanical Engineering; Lightning; Heating; Weapons; Blasting),\nG (Physics), H (Electricity), and\nY (General tagging of new or cross-sectional technology)\nThere are two features:\n - description: detailed description of patent.\n - abstract: Patent abastract.\n", "citation": "\n@misc{sharma2019bigpatent,\n title={BIGPATENT: A Large-Scale Dataset for Abstractive and Coherent Summarization},\n author={Eva Sharma and Chen Li and Lu Wang},\n year={2019},\n eprint={1906.03741},\n archivePrefix={arXiv},\n primaryClass={cs.CL}\n}\n", "homepage": "https://evasharma.github.io/bigpatent/", "license": "Creative Commons Attribution 4.0 International", "features": {"description": {"dtype": "string", "id": null, "_type": "Value"}, "abstract": {"dtype": "string", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": {"input": "description", "output": "abstract"}, "task_templates": null, "builder_name": "big_patent", "config_name": "all", "version": "2.1.2", "splits": {"train": {"name": "train", "num_bytes": 38367048389, "num_examples": 1207222, "dataset_name": "big_patent"}, "validation": {"name": "validation", "num_bytes": 2115827002, "num_examples": 67068, "dataset_name": "big_patent"}, "test": {"name": "test", "num_bytes": 2129505280, "num_examples": 67072, "dataset_name": "big_patent"}}, "download_checksums": {"data/2.1.2/train.zip": {"num_bytes": 9129045386, "checksum": "59cae1439b01102faa5b9e1bb3061867ba292a6093cf66ee3d1bc33030d7a91f"}, "data/2.1.2/val.zip": {"num_bytes": 506259325, "checksum": "9206de008607095bdb29f405ef9b525565556860d015cbd4261154644b51ce9b"}, "data/2.1.2/test.zip": {"num_bytes": 507619065, "checksum": "614821d8d8c92451ba495e0b381f157a0e0b36977a70f2d4cb3ae4b6b706539a"}}, "download_size": 10142923776, "post_processing_size": null, "dataset_size": 42612380671, "size_in_bytes": 52755304447}, "a": {"description": "\nBIGPATENT, consisting of 1.3 million records of U.S. patent documents\nalong with human written abstractive summaries.\nEach US patent application is filed under a Cooperative Patent Classification\n(CPC) code. There are nine such classification categories:\nA (Human Necessities), B (Performing Operations; Transporting),\nC (Chemistry; Metallurgy), D (Textiles; Paper), E (Fixed Constructions),\nF (Mechanical Engineering; Lightning; Heating; Weapons; Blasting),\nG (Physics), H (Electricity), and\nY (General tagging of new or cross-sectional technology)\nThere are two features:\n - description: detailed description of patent.\n - abstract: Patent abastract.\n", "citation": "\n@misc{sharma2019bigpatent,\n title={BIGPATENT: A Large-Scale Dataset for Abstractive and Coherent Summarization},\n author={Eva Sharma and Chen Li and Lu Wang},\n year={2019},\n eprint={1906.03741},\n archivePrefix={arXiv},\n primaryClass={cs.CL}\n}\n", "homepage": "https://evasharma.github.io/bigpatent/", "license": "Creative Commons Attribution 4.0 International", "features": {"description": {"dtype": "string", "id": null, "_type": "Value"}, "abstract": {"dtype": "string", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": {"input": "description", "output": "abstract"}, "task_templates": null, "builder_name": "big_patent", "config_name": "a", "version": "2.1.2", "splits": {"train": {"name": "train", "num_bytes": 5683460620, "num_examples": 174134, "dataset_name": "big_patent"}, "validation": {"name": "validation", "num_bytes": 313324505, "num_examples": 9674, "dataset_name": "big_patent"}, "test": {"name": "test", "num_bytes": 316633277, "num_examples": 9675, "dataset_name": "big_patent"}}, "download_checksums": {"data/2.1.2/train.zip": {"num_bytes": 9129045386, "checksum": "59cae1439b01102faa5b9e1bb3061867ba292a6093cf66ee3d1bc33030d7a91f"}, "data/2.1.2/val.zip": {"num_bytes": 506259325, "checksum": "9206de008607095bdb29f405ef9b525565556860d015cbd4261154644b51ce9b"}, "data/2.1.2/test.zip": {"num_bytes": 507619065, "checksum": "614821d8d8c92451ba495e0b381f157a0e0b36977a70f2d4cb3ae4b6b706539a"}}, "download_size": 10142923776, "post_processing_size": null, "dataset_size": 6313418402, "size_in_bytes": 16456342178}, "b": {"description": "\nBIGPATENT, consisting of 1.3 million records of U.S. patent documents\nalong with human written abstractive summaries.\nEach US patent application is filed under a Cooperative Patent Classification\n(CPC) code. There are nine such classification categories:\nA (Human Necessities), B (Performing Operations; Transporting),\nC (Chemistry; Metallurgy), D (Textiles; Paper), E (Fixed Constructions),\nF (Mechanical Engineering; Lightning; Heating; Weapons; Blasting),\nG (Physics), H (Electricity), and\nY (General tagging of new or cross-sectional technology)\nThere are two features:\n - description: detailed description of patent.\n - abstract: Patent abastract.\n", "citation": "\n@misc{sharma2019bigpatent,\n title={BIGPATENT: A Large-Scale Dataset for Abstractive and Coherent Summarization},\n author={Eva Sharma and Chen Li and Lu Wang},\n year={2019},\n eprint={1906.03741},\n archivePrefix={arXiv},\n primaryClass={cs.CL}\n}\n", "homepage": "https://evasharma.github.io/bigpatent/", "license": "Creative Commons Attribution 4.0 International", "features": {"description": {"dtype": "string", "id": null, "_type": "Value"}, "abstract": {"dtype": "string", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": {"input": "description", "output": "abstract"}, "task_templates": null, "builder_name": "big_patent", "config_name": "b", "version": "2.1.2", "splits": {"train": {"name": "train", "num_bytes": 4236070976, "num_examples": 161520, "dataset_name": "big_patent"}, "validation": {"name": "validation", "num_bytes": 234425138, "num_examples": 8973, "dataset_name": "big_patent"}, "test": {"name": "test", "num_bytes": 231538734, "num_examples": 8974, "dataset_name": "big_patent"}}, "download_checksums": {"data/2.1.2/train.zip": {"num_bytes": 9129045386, "checksum": "59cae1439b01102faa5b9e1bb3061867ba292a6093cf66ee3d1bc33030d7a91f"}, "data/2.1.2/val.zip": {"num_bytes": 506259325, "checksum": "9206de008607095bdb29f405ef9b525565556860d015cbd4261154644b51ce9b"}, "data/2.1.2/test.zip": {"num_bytes": 507619065, "checksum": "614821d8d8c92451ba495e0b381f157a0e0b36977a70f2d4cb3ae4b6b706539a"}}, "download_size": 10142923776, "post_processing_size": null, "dataset_size": 4702034848, "size_in_bytes": 14844958624}, "c": {"description": "\nBIGPATENT, consisting of 1.3 million records of U.S. patent documents\nalong with human written abstractive summaries.\nEach US patent application is filed under a Cooperative Patent Classification\n(CPC) code. There are nine such classification categories:\nA (Human Necessities), B (Performing Operations; Transporting),\nC (Chemistry; Metallurgy), D (Textiles; Paper), E (Fixed Constructions),\nF (Mechanical Engineering; Lightning; Heating; Weapons; Blasting),\nG (Physics), H (Electricity), and\nY (General tagging of new or cross-sectional technology)\nThere are two features:\n - description: detailed description of patent.\n - abstract: Patent abastract.\n", "citation": "\n@misc{sharma2019bigpatent,\n title={BIGPATENT: A Large-Scale Dataset for Abstractive and Coherent Summarization},\n author={Eva Sharma and Chen Li and Lu Wang},\n year={2019},\n eprint={1906.03741},\n archivePrefix={arXiv},\n primaryClass={cs.CL}\n}\n", "homepage": "https://evasharma.github.io/bigpatent/", "license": "Creative Commons Attribution 4.0 International", "features": {"description": {"dtype": "string", "id": null, "_type": "Value"}, "abstract": {"dtype": "string", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": {"input": "description", "output": "abstract"}, "task_templates": null, "builder_name": "big_patent", "config_name": "c", "version": "2.1.2", "splits": {"train": {"name": "train", "num_bytes": 4506249306, "num_examples": 101042, "dataset_name": "big_patent"}, "validation": {"name": "validation", "num_bytes": 244684775, "num_examples": 5613, "dataset_name": "big_patent"}, "test": {"name": "test", "num_bytes": 252566793, "num_examples": 5614, "dataset_name": "big_patent"}}, "download_checksums": {"data/2.1.2/train.zip": {"num_bytes": 9129045386, "checksum": "59cae1439b01102faa5b9e1bb3061867ba292a6093cf66ee3d1bc33030d7a91f"}, "data/2.1.2/val.zip": {"num_bytes": 506259325, "checksum": "9206de008607095bdb29f405ef9b525565556860d015cbd4261154644b51ce9b"}, "data/2.1.2/test.zip": {"num_bytes": 507619065, "checksum": "614821d8d8c92451ba495e0b381f157a0e0b36977a70f2d4cb3ae4b6b706539a"}}, "download_size": 10142923776, "post_processing_size": null, "dataset_size": 5003500874, "size_in_bytes": 15146424650}, "d": {"description": "\nBIGPATENT, consisting of 1.3 million records of U.S. patent documents\nalong with human written abstractive summaries.\nEach US patent application is filed under a Cooperative Patent Classification\n(CPC) code. There are nine such classification categories:\nA (Human Necessities), B (Performing Operations; Transporting),\nC (Chemistry; Metallurgy), D (Textiles; Paper), E (Fixed Constructions),\nF (Mechanical Engineering; Lightning; Heating; Weapons; Blasting),\nG (Physics), H (Electricity), and\nY (General tagging of new or cross-sectional technology)\nThere are two features:\n - description: detailed description of patent.\n - abstract: Patent abastract.\n", "citation": "\n@misc{sharma2019bigpatent,\n title={BIGPATENT: A Large-Scale Dataset for Abstractive and Coherent Summarization},\n author={Eva Sharma and Chen Li and Lu Wang},\n year={2019},\n eprint={1906.03741},\n archivePrefix={arXiv},\n primaryClass={cs.CL}\n}\n", "homepage": "https://evasharma.github.io/bigpatent/", "license": "Creative Commons Attribution 4.0 International", "features": {"description": {"dtype": "string", "id": null, "_type": "Value"}, "abstract": {"dtype": "string", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": {"input": "description", "output": "abstract"}, "task_templates": null, "builder_name": "big_patent", "config_name": "d", "version": "2.1.2", "splits": {"train": {"name": "train", "num_bytes": 264717412, "num_examples": 10164, "dataset_name": "big_patent"}, "validation": {"name": "validation", "num_bytes": 14560482, "num_examples": 565, "dataset_name": "big_patent"}, "test": {"name": "test", "num_bytes": 14403430, "num_examples": 565, "dataset_name": "big_patent"}}, "download_checksums": {"data/2.1.2/train.zip": {"num_bytes": 9129045386, "checksum": "59cae1439b01102faa5b9e1bb3061867ba292a6093cf66ee3d1bc33030d7a91f"}, "data/2.1.2/val.zip": {"num_bytes": 506259325, "checksum": "9206de008607095bdb29f405ef9b525565556860d015cbd4261154644b51ce9b"}, "data/2.1.2/test.zip": {"num_bytes": 507619065, "checksum": "614821d8d8c92451ba495e0b381f157a0e0b36977a70f2d4cb3ae4b6b706539a"}}, "download_size": 10142923776, "post_processing_size": null, "dataset_size": 293681324, "size_in_bytes": 10436605100}, "e": {"description": "\nBIGPATENT, consisting of 1.3 million records of U.S. patent documents\nalong with human written abstractive summaries.\nEach US patent application is filed under a Cooperative Patent Classification\n(CPC) code. There are nine such classification categories:\nA (Human Necessities), B (Performing Operations; Transporting),\nC (Chemistry; Metallurgy), D (Textiles; Paper), E (Fixed Constructions),\nF (Mechanical Engineering; Lightning; Heating; Weapons; Blasting),\nG (Physics), H (Electricity), and\nY (General tagging of new or cross-sectional technology)\nThere are two features:\n - description: detailed description of patent.\n - abstract: Patent abastract.\n", "citation": "\n@misc{sharma2019bigpatent,\n title={BIGPATENT: A Large-Scale Dataset for Abstractive and Coherent Summarization},\n author={Eva Sharma and Chen Li and Lu Wang},\n year={2019},\n eprint={1906.03741},\n archivePrefix={arXiv},\n primaryClass={cs.CL}\n}\n", "homepage": "https://evasharma.github.io/bigpatent/", "license": "Creative Commons Attribution 4.0 International", "features": {"description": {"dtype": "string", "id": null, "_type": "Value"}, "abstract": {"dtype": "string", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": {"input": "description", "output": "abstract"}, "task_templates": null, "builder_name": "big_patent", "config_name": "e", "version": "2.1.2", "splits": {"train": {"name": "train", "num_bytes": 881101433, "num_examples": 34443, "dataset_name": "big_patent"}, "validation": {"name": "validation", "num_bytes": 48646158, "num_examples": 1914, "dataset_name": "big_patent"}, "test": {"name": "test", "num_bytes": 48586429, "num_examples": 1914, "dataset_name": "big_patent"}}, "download_checksums": {"data/2.1.2/train.zip": {"num_bytes": 9129045386, "checksum": "59cae1439b01102faa5b9e1bb3061867ba292a6093cf66ee3d1bc33030d7a91f"}, "data/2.1.2/val.zip": {"num_bytes": 506259325, "checksum": "9206de008607095bdb29f405ef9b525565556860d015cbd4261154644b51ce9b"}, "data/2.1.2/test.zip": {"num_bytes": 507619065, "checksum": "614821d8d8c92451ba495e0b381f157a0e0b36977a70f2d4cb3ae4b6b706539a"}}, "download_size": 10142923776, "post_processing_size": null, "dataset_size": 978334020, "size_in_bytes": 11121257796}, "f": {"description": "\nBIGPATENT, consisting of 1.3 million records of U.S. patent documents\nalong with human written abstractive summaries.\nEach US patent application is filed under a Cooperative Patent Classification\n(CPC) code. There are nine such classification categories:\nA (Human Necessities), B (Performing Operations; Transporting),\nC (Chemistry; Metallurgy), D (Textiles; Paper), E (Fixed Constructions),\nF (Mechanical Engineering; Lightning; Heating; Weapons; Blasting),\nG (Physics), H (Electricity), and\nY (General tagging of new or cross-sectional technology)\nThere are two features:\n - description: detailed description of patent.\n - abstract: Patent abastract.\n", "citation": "\n@misc{sharma2019bigpatent,\n title={BIGPATENT: A Large-Scale Dataset for Abstractive and Coherent Summarization},\n author={Eva Sharma and Chen Li and Lu Wang},\n year={2019},\n eprint={1906.03741},\n archivePrefix={arXiv},\n primaryClass={cs.CL}\n}\n", "homepage": "https://evasharma.github.io/bigpatent/", "license": "Creative Commons Attribution 4.0 International", "features": {"description": {"dtype": "string", "id": null, "_type": "Value"}, "abstract": {"dtype": "string", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": {"input": "description", "output": "abstract"}, "task_templates": null, "builder_name": "big_patent", "config_name": "f", "version": "2.1.2", "splits": {"train": {"name": "train", "num_bytes": 2146383473, "num_examples": 85568, "dataset_name": "big_patent"}, "validation": {"name": "validation", "num_bytes": 119632631, "num_examples": 4754, "dataset_name": "big_patent"}, "test": {"name": "test", "num_bytes": 119596303, "num_examples": 4754, "dataset_name": "big_patent"}}, "download_checksums": {"data/2.1.2/train.zip": {"num_bytes": 9129045386, "checksum": "59cae1439b01102faa5b9e1bb3061867ba292a6093cf66ee3d1bc33030d7a91f"}, "data/2.1.2/val.zip": {"num_bytes": 506259325, "checksum": "9206de008607095bdb29f405ef9b525565556860d015cbd4261154644b51ce9b"}, "data/2.1.2/test.zip": {"num_bytes": 507619065, "checksum": "614821d8d8c92451ba495e0b381f157a0e0b36977a70f2d4cb3ae4b6b706539a"}}, "download_size": 10142923776, "post_processing_size": null, "dataset_size": 2385612407, "size_in_bytes": 12528536183}, "g": {"description": "\nBIGPATENT, consisting of 1.3 million records of U.S. patent documents\nalong with human written abstractive summaries.\nEach US patent application is filed under a Cooperative Patent Classification\n(CPC) code. There are nine such classification categories:\nA (Human Necessities), B (Performing Operations; Transporting),\nC (Chemistry; Metallurgy), D (Textiles; Paper), E (Fixed Constructions),\nF (Mechanical Engineering; Lightning; Heating; Weapons; Blasting),\nG (Physics), H (Electricity), and\nY (General tagging of new or cross-sectional technology)\nThere are two features:\n - description: detailed description of patent.\n - abstract: Patent abastract.\n", "citation": "\n@misc{sharma2019bigpatent,\n title={BIGPATENT: A Large-Scale Dataset for Abstractive and Coherent Summarization},\n author={Eva Sharma and Chen Li and Lu Wang},\n year={2019},\n eprint={1906.03741},\n archivePrefix={arXiv},\n primaryClass={cs.CL}\n}\n", "homepage": "https://evasharma.github.io/bigpatent/", "license": "Creative Commons Attribution 4.0 International", "features": {"description": {"dtype": "string", "id": null, "_type": "Value"}, "abstract": {"dtype": "string", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": {"input": "description", "output": "abstract"}, "task_templates": null, "builder_name": "big_patent", "config_name": "g", "version": "2.1.2", "splits": {"train": {"name": "train", "num_bytes": 8877854206, "num_examples": 258935, "dataset_name": "big_patent"}, "validation": {"name": "validation", "num_bytes": 492581177, "num_examples": 14385, "dataset_name": "big_patent"}, "test": {"name": "test", "num_bytes": 496324853, "num_examples": 14386, "dataset_name": "big_patent"}}, "download_checksums": {"data/2.1.2/train.zip": {"num_bytes": 9129045386, "checksum": "59cae1439b01102faa5b9e1bb3061867ba292a6093cf66ee3d1bc33030d7a91f"}, "data/2.1.2/val.zip": {"num_bytes": 506259325, "checksum": "9206de008607095bdb29f405ef9b525565556860d015cbd4261154644b51ce9b"}, "data/2.1.2/test.zip": {"num_bytes": 507619065, "checksum": "614821d8d8c92451ba495e0b381f157a0e0b36977a70f2d4cb3ae4b6b706539a"}}, "download_size": 10142923776, "post_processing_size": null, "dataset_size": 9866760236, "size_in_bytes": 20009684012}, "h": {"description": "\nBIGPATENT, consisting of 1.3 million records of U.S. patent documents\nalong with human written abstractive summaries.\nEach US patent application is filed under a Cooperative Patent Classification\n(CPC) code. There are nine such classification categories:\nA (Human Necessities), B (Performing Operations; Transporting),\nC (Chemistry; Metallurgy), D (Textiles; Paper), E (Fixed Constructions),\nF (Mechanical Engineering; Lightning; Heating; Weapons; Blasting),\nG (Physics), H (Electricity), and\nY (General tagging of new or cross-sectional technology)\nThere are two features:\n - description: detailed description of patent.\n - abstract: Patent abastract.\n", "citation": "\n@misc{sharma2019bigpatent,\n title={BIGPATENT: A Large-Scale Dataset for Abstractive and Coherent Summarization},\n author={Eva Sharma and Chen Li and Lu Wang},\n year={2019},\n eprint={1906.03741},\n archivePrefix={arXiv},\n primaryClass={cs.CL}\n}\n", "homepage": "https://evasharma.github.io/bigpatent/", "license": "Creative Commons Attribution 4.0 International", "features": {"description": {"dtype": "string", "id": null, "_type": "Value"}, "abstract": {"dtype": "string", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": {"input": "description", "output": "abstract"}, "task_templates": null, "builder_name": "big_patent", "config_name": "h", "version": "2.1.2", "splits": {"train": {"name": "train", "num_bytes": 8075621958, "num_examples": 257019, "dataset_name": "big_patent"}, "validation": {"name": "validation", "num_bytes": 447602356, "num_examples": 14279, "dataset_name": "big_patent"}, "test": {"name": "test", "num_bytes": 445460513, "num_examples": 14279, "dataset_name": "big_patent"}}, "download_checksums": {"data/2.1.2/train.zip": {"num_bytes": 9129045386, "checksum": "59cae1439b01102faa5b9e1bb3061867ba292a6093cf66ee3d1bc33030d7a91f"}, "data/2.1.2/val.zip": {"num_bytes": 506259325, "checksum": "9206de008607095bdb29f405ef9b525565556860d015cbd4261154644b51ce9b"}, "data/2.1.2/test.zip": {"num_bytes": 507619065, "checksum": "614821d8d8c92451ba495e0b381f157a0e0b36977a70f2d4cb3ae4b6b706539a"}}, "download_size": 10142923776, "post_processing_size": null, "dataset_size": 8968684827, "size_in_bytes": 19111608603}, "y": {"description": "\nBIGPATENT, consisting of 1.3 million records of U.S. patent documents\nalong with human written abstractive summaries.\nEach US patent application is filed under a Cooperative Patent Classification\n(CPC) code. There are nine such classification categories:\nA (Human Necessities), B (Performing Operations; Transporting),\nC (Chemistry; Metallurgy), D (Textiles; Paper), E (Fixed Constructions),\nF (Mechanical Engineering; Lightning; Heating; Weapons; Blasting),\nG (Physics), H (Electricity), and\nY (General tagging of new or cross-sectional technology)\nThere are two features:\n - description: detailed description of patent.\n - abstract: Patent abastract.\n", "citation": "\n@misc{sharma2019bigpatent,\n title={BIGPATENT: A Large-Scale Dataset for Abstractive and Coherent Summarization},\n author={Eva Sharma and Chen Li and Lu Wang},\n year={2019},\n eprint={1906.03741},\n archivePrefix={arXiv},\n primaryClass={cs.CL}\n}\n", "homepage": "https://evasharma.github.io/bigpatent/", "license": "Creative Commons Attribution 4.0 International", "features": {"description": {"dtype": "string", "id": null, "_type": "Value"}, "abstract": {"dtype": "string", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": {"input": "description", "output": "abstract"}, "task_templates": null, "builder_name": "big_patent", "config_name": "y", "version": "2.1.2", "splits": {"train": {"name": "train", "num_bytes": 3695589005, "num_examples": 124397, "dataset_name": "big_patent"}, "validation": {"name": "validation", "num_bytes": 200369780, "num_examples": 6911, "dataset_name": "big_patent"}, "test": {"name": "test", "num_bytes": 204394948, "num_examples": 6911, "dataset_name": "big_patent"}}, "download_checksums": {"data/2.1.2/train.zip": {"num_bytes": 9129045386, "checksum": "59cae1439b01102faa5b9e1bb3061867ba292a6093cf66ee3d1bc33030d7a91f"}, "data/2.1.2/val.zip": {"num_bytes": 506259325, "checksum": "9206de008607095bdb29f405ef9b525565556860d015cbd4261154644b51ce9b"}, "data/2.1.2/test.zip": {"num_bytes": 507619065, "checksum": "614821d8d8c92451ba495e0b381f157a0e0b36977a70f2d4cb3ae4b6b706539a"}}, "download_size": 10142923776, "post_processing_size": null, "dataset_size": 4100353733, "size_in_bytes": 14243277509}}