Datasets:

Languages:
English
ArXiv:
Libraries:
Datasets
License:
parquet-converter commited on
Commit
14c0afe
·
1 Parent(s): 9b3a92f

Update parquet files

Browse files
.gitattributes DELETED
@@ -1,27 +0,0 @@
1
- *.7z filter=lfs diff=lfs merge=lfs -text
2
- *.arrow filter=lfs diff=lfs merge=lfs -text
3
- *.bin filter=lfs diff=lfs merge=lfs -text
4
- *.bin.* filter=lfs diff=lfs merge=lfs -text
5
- *.bz2 filter=lfs diff=lfs merge=lfs -text
6
- *.ftz filter=lfs diff=lfs merge=lfs -text
7
- *.gz filter=lfs diff=lfs merge=lfs -text
8
- *.h5 filter=lfs diff=lfs merge=lfs -text
9
- *.joblib filter=lfs diff=lfs merge=lfs -text
10
- *.lfs.* filter=lfs diff=lfs merge=lfs -text
11
- *.model filter=lfs diff=lfs merge=lfs -text
12
- *.msgpack filter=lfs diff=lfs merge=lfs -text
13
- *.onnx filter=lfs diff=lfs merge=lfs -text
14
- *.ot filter=lfs diff=lfs merge=lfs -text
15
- *.parquet filter=lfs diff=lfs merge=lfs -text
16
- *.pb filter=lfs diff=lfs merge=lfs -text
17
- *.pt filter=lfs diff=lfs merge=lfs -text
18
- *.pth filter=lfs diff=lfs merge=lfs -text
19
- *.rar filter=lfs diff=lfs merge=lfs -text
20
- saved_model/**/* filter=lfs diff=lfs merge=lfs -text
21
- *.tar.* filter=lfs diff=lfs merge=lfs -text
22
- *.tflite filter=lfs diff=lfs merge=lfs -text
23
- *.tgz filter=lfs diff=lfs merge=lfs -text
24
- *.xz filter=lfs diff=lfs merge=lfs -text
25
- *.zip filter=lfs diff=lfs merge=lfs -text
26
- *.zstandard filter=lfs diff=lfs merge=lfs -text
27
- *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
AIC/scitldr-test.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fe2ea3bf6e17da6835f801e38abd4d66f6ce55f617e5efdcbde3cf25f8691163
3
+ size 2420420
AIC/scitldr-train.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bed1d1dcde67bb575f8951b78f1b3b375ab82cef4ede2b9ba6b832741f45f8e4
3
+ size 7225242
AIC/scitldr-validation.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:06260d11e4edd5c736e939e747ea52079250accea0e95cf9a34e9489257ca164
3
+ size 2246859
Abstract/scitldr-test.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b1ecde965d8a5a0fa21872e43ffc402169669c2485edcfa6e2f401b09fcfe735
3
+ size 569390
Abstract/scitldr-train.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5dd84670937fc827083fa628172c0a0533ee2682e2de26dda74d57c99e6bf1bd
3
+ size 1462133
Abstract/scitldr-validation.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:830c859b310797d32d6e45ff3f9bd57d3badb4c937ad75623159ecc23d5dd4da
3
+ size 529207
FullText/scitldr-test.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:01943acff58815c010834dd5a14c1b33cda1f8d64350d089db108ef1b72fc1c8
3
+ size 10345754
FullText/scitldr-train.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8397c92d09f158f1d979dcbeae4d7fbc6d5ee45eaf6a4a04bb93adce626c70ab
3
+ size 32345466
FullText/scitldr-validation.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:23ba62cea503284c8e8cd16615d94f948facac26037d82b685e68cba46bfb1c0
3
+ size 9642576
README.md DELETED
@@ -1,305 +0,0 @@
1
- ---
2
- annotations_creators:
3
- - no-annotation
4
- language_creators:
5
- - found
6
- language:
7
- - en
8
- license:
9
- - unknown
10
- multilinguality:
11
- - monolingual
12
- size_categories:
13
- - 1K<n<10K
14
- source_datasets:
15
- - original
16
- task_categories:
17
- - summarization
18
- task_ids: []
19
- paperswithcode_id: scitldr
20
- pretty_name: SciTLDR
21
- tags:
22
- - scientific-documents-summarization
23
- dataset_info:
24
- - config_name: Abstract
25
- features:
26
- - name: source
27
- sequence: string
28
- - name: source_labels
29
- sequence:
30
- class_label:
31
- names:
32
- 0: non-oracle
33
- 1: oracle
34
- - name: rouge_scores
35
- sequence: float32
36
- - name: paper_id
37
- dtype: string
38
- - name: target
39
- sequence: string
40
- splits:
41
- - name: train
42
- num_bytes: 2738065
43
- num_examples: 1992
44
- - name: test
45
- num_bytes: 1073656
46
- num_examples: 618
47
- - name: validation
48
- num_bytes: 994876
49
- num_examples: 619
50
- download_size: 5483987
51
- dataset_size: 4806597
52
- - config_name: AIC
53
- features:
54
- - name: source
55
- sequence: string
56
- - name: source_labels
57
- sequence:
58
- class_label:
59
- names:
60
- 0: 0
61
- 1: 1
62
- - name: rouge_scores
63
- sequence: float32
64
- - name: paper_id
65
- dtype: string
66
- - name: ic
67
- dtype: bool_
68
- - name: target
69
- sequence: string
70
- splits:
71
- - name: train
72
- num_bytes: 14473822
73
- num_examples: 1992
74
- - name: test
75
- num_bytes: 4822026
76
- num_examples: 618
77
- - name: validation
78
- num_bytes: 4476237
79
- num_examples: 619
80
- download_size: 25545108
81
- dataset_size: 23772085
82
- - config_name: FullText
83
- features:
84
- - name: source
85
- sequence: string
86
- - name: source_labels
87
- sequence:
88
- class_label:
89
- names:
90
- 0: non-oracle
91
- 1: oracle
92
- - name: rouge_scores
93
- sequence: float32
94
- - name: paper_id
95
- dtype: string
96
- - name: target
97
- sequence: string
98
- splits:
99
- - name: train
100
- num_bytes: 66917363
101
- num_examples: 1992
102
- - name: test
103
- num_bytes: 20182554
104
- num_examples: 618
105
- - name: validation
106
- num_bytes: 18790651
107
- num_examples: 619
108
- download_size: 110904552
109
- dataset_size: 105890568
110
- ---
111
-
112
- # Dataset Card for SciTLDR
113
-
114
- ## Table of Contents
115
- - [Dataset Description](#dataset-description)
116
- - [Dataset Summary](#dataset-summary)
117
- - [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
118
- - [Languages](#languages)
119
- - [Dataset Structure](#dataset-structure)
120
- - [Data Instances](#data-instances)
121
- - [Data Fields](#data-fields)
122
- - [Data Splits](#data-splits)
123
- - [Dataset Creation](#dataset-creation)
124
- - [Curation Rationale](#curation-rationale)
125
- - [Source Data](#source-data)
126
- - [Annotations](#annotations)
127
- - [Personal and Sensitive Information](#personal-and-sensitive-information)
128
- - [Considerations for Using the Data](#considerations-for-using-the-data)
129
- - [Social Impact of Dataset](#social-impact-of-dataset)
130
- - [Discussion of Biases](#discussion-of-biases)
131
- - [Other Known Limitations](#other-known-limitations)
132
- - [Additional Information](#additional-information)
133
- - [Dataset Curators](#dataset-curators)
134
- - [Licensing Information](#licensing-information)
135
- - [Citation Information](#citation-information)
136
- - [Contributions](#contributions)
137
-
138
- ## Dataset Description
139
-
140
- - **Homepage:** https://github.com/allenai/scitldr
141
- - **Repository:** https://github.com/allenai/scitldr
142
- - **Paper:** https://arxiv.org/abs/2004.15011
143
- - **Leaderboard:**
144
- - **Point of Contact:** {isabelc,kylel,armanc,danw}@allenai.org
145
-
146
- ### Dataset Summary
147
- `SciTLDR`: Extreme Summarization of Scientific Documents
148
-
149
- SciTLDR is a new multi-target dataset of 5.4K TLDRs over 3.2K papers. SciTLDR contains both author-written and expert-derived TLDRs, where the latter are collected using a novel annotation protocol that produces high-quality summaries while minimizing annotation burden.
150
-
151
- ### Supported Tasks and Leaderboards
152
-
153
- summarization
154
-
155
- ### Languages
156
-
157
- English
158
-
159
- ## Dataset Structure
160
-
161
- SciTLDR is split in to a 60/20/20 train/dev/test split. For each file, each line is a json, formatted as follows
162
- ```
163
- {
164
- "source":[
165
- "sent0",
166
- "sent1",
167
- "sent2",
168
- ...
169
- ],
170
- "source_labels":[binary list in which 1 is the oracle sentence],
171
- "rouge_scores":[precomputed rouge-1 scores],
172
- "paper_id":"PAPER-ID",
173
- "target":[
174
- "author-tldr",
175
- "pr-tldr0",
176
- "pr-tldr1",
177
- ...
178
- ],
179
- "title":"TITLE"
180
- }
181
- ```
182
- The keys `rouge_scores` and `source_labels` are not necessary for any code to run, precomputed Rouge scores are provided for future research.
183
-
184
- ### Data Instances
185
-
186
- {
187
- "source": [
188
- "Mixed precision training (MPT) is becoming a practical technique to improve the speed and energy efficiency of training deep neural networks by leveraging the fast hardware support for IEEE half-precision floating point that is available in existing GPUs.",
189
- "MPT is typically used in combination with a technique called loss scaling, that works by scaling up the loss value up before the start of backpropagation in order to minimize the impact of numerical underflow on training.",
190
- "Unfortunately, existing methods make this loss scale value a hyperparameter that needs to be tuned per-model, and a single scale cannot be adapted to different layers at different training stages.",
191
- "We introduce a loss scaling-based training method called adaptive loss scaling that makes MPT easier and more practical to use, by removing the need to tune a model-specific loss scale hyperparameter.",
192
- "We achieve this by introducing layer-wise loss scale values which are automatically computed during training to deal with underflow more effectively than existing methods.",
193
- "We present experimental results on a variety of networks and tasks that show our approach can shorten the time to convergence and improve accuracy, compared with using the existing state-of-the-art MPT and single-precision floating point."
194
- ],
195
- "source_labels": [
196
- 0,
197
- 0,
198
- 0,
199
- 1,
200
- 0,
201
- 0
202
- ],
203
- "rouge_scores": [
204
- 0.2399999958000001,
205
- 0.26086956082230633,
206
- 0.19999999531250012,
207
- 0.38095237636054424,
208
- 0.2051282003944774,
209
- 0.2978723360796741
210
- ],
211
- "paper_id": "rJlnfaNYvB",
212
- "target": [
213
- "We devise adaptive loss scaling to improve mixed precision training that surpass the state-of-the-art results.",
214
- "Proposal for an adaptive loss scaling method during backpropagation for mix precision training where scale rate is decided automatically to reduce the underflow.",
215
- "The authors propose a method to train models in FP16 precision that adopts a more elaborate way to minimize underflow in every layer simultaneously and automatically."
216
- ],
217
- "title": "Adaptive Loss Scaling for Mixed Precision Training"
218
- }
219
-
220
- ### Data Fields
221
-
222
- - `source`: The Abstract, Introduction and Conclusion (AIC) or Full text of the paper, with one sentence per line.
223
- - `source_labels`: Binary 0 or 1, 1 denotes the oracle sentence.
224
- - `rouge_scores`: Precomputed ROUGE baseline scores for each sentence.
225
- - `paper_id`: Arxiv Paper ID.
226
- - `target`: Multiple summaries for each sentence, one sentence per line.
227
- - `title`: Title of the paper.
228
- ### Data Splits
229
-
230
- | | train | valid | test |
231
- |-------------------|-------|--------|------|
232
- | SciTLDR-A | 1992 | 618 | 619 |
233
- | SciTLDR-AIC | 1992 | 618 | 619 |
234
- | SciTLDR-FullText | 1992 | 618 | 619 |
235
-
236
- ## Dataset Creation
237
-
238
- [More Information Needed]
239
-
240
- ### Curation Rationale
241
-
242
- [More Information Needed]
243
-
244
- ### Source Data
245
-
246
- #### Initial Data Collection and Normalization
247
-
248
- [More Information Needed]
249
-
250
- #### Who are the source language producers?
251
- https://allenai.org/
252
-
253
- ### Annotations
254
-
255
- #### Annotation process
256
-
257
- Given the title and first 128 words of a reviewer comment about a paper,
258
- re-write the summary (if it exists) into a single sentence or an incomplete
259
- phrase. Summaries must be no more than one sentence.
260
- Most summaries are between 15 and 25 words. The average rewritten summary is
261
- 20 words long.
262
-
263
- #### Who are the annotators?
264
-
265
- [More Information Needed]
266
-
267
- ### Personal and Sensitive Information
268
-
269
- [More Information Needed]
270
-
271
- ## Considerations for Using the Data
272
-
273
- ### Social Impact of Dataset
274
-
275
- To encourage further research in the area of extreme summarization of scientific documents.
276
-
277
- ### Discussion of Biases
278
-
279
- [More Information Needed]
280
-
281
- ### Other Known Limitations
282
-
283
- [More Information Needed]
284
-
285
- ## Additional Information
286
-
287
- ### Dataset Curators
288
-
289
- [More Information Needed]
290
-
291
- ### Licensing Information
292
-
293
- Apache License 2.0
294
-
295
- ### Citation Information
296
- @article{cachola2020tldr,
297
- title={{TLDR}: Extreme Summarization of Scientific Documents},
298
- author={Isabel Cachola and Kyle Lo and Arman Cohan and Daniel S. Weld},
299
- journal={arXiv:2004.15011},
300
- year={2020},
301
- }
302
-
303
- ### Contributions
304
-
305
- Thanks to [@Bharat123rox](https://github.com/Bharat123rox) for adding this dataset.
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
dataset_infos.json DELETED
@@ -1 +0,0 @@
1
- {"Abstract": {"description": "A new multi-target dataset of 5.4K TLDRs over 3.2K papers.\nSCITLDR contains both author-written and expert-derived TLDRs,\nwhere the latter are collected using a novel annotation protocol\nthat produces high-quality summaries while minimizing annotation burden.\n", "citation": "@article{cachola2020tldr,\n title={{TLDR}: Extreme Summarization of Scientific Documents},\n author={Isabel Cachola and Kyle Lo and Arman Cohan and Daniel S. Weld},\n journal={arXiv:2004.15011},\n year={2020},\n}\n", "homepage": "https://github.com/allenai/scitldr", "license": "Apache License 2.0", "features": {"source": {"feature": {"dtype": "string", "id": null, "_type": "Value"}, "length": -1, "id": null, "_type": "Sequence"}, "source_labels": {"feature": {"num_classes": 2, "names": ["non-oracle", "oracle"], "names_file": null, "id": null, "_type": "ClassLabel"}, "length": -1, "id": null, "_type": "Sequence"}, "rouge_scores": {"feature": {"dtype": "float32", "id": null, "_type": "Value"}, "length": -1, "id": null, "_type": "Sequence"}, "paper_id": {"dtype": "string", "id": null, "_type": "Value"}, "target": {"feature": {"dtype": "string", "id": null, "_type": "Value"}, "length": -1, "id": null, "_type": "Sequence"}}, "post_processed": null, "supervised_keys": {"input": "source", "output": "target"}, "builder_name": "scitldr", "config_name": "Abstract", "version": {"version_str": "0.0.0", "description": null, "major": 0, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 2738065, "num_examples": 1992, "dataset_name": "scitldr"}, "test": {"name": "test", "num_bytes": 1073656, "num_examples": 618, "dataset_name": "scitldr"}, "validation": {"name": "validation", "num_bytes": 994876, "num_examples": 619, "dataset_name": "scitldr"}}, "download_checksums": {"https://raw.githubusercontent.com/allenai/scitldr/master/SciTLDR-Data/SciTLDR-A/train.jsonl": {"num_bytes": 3155015, "checksum": "b222771d387be585cfdf5ae957b36757138415a352e0a3e3b23f73f87c3b1119"}, "https://raw.githubusercontent.com/allenai/scitldr/master/SciTLDR-Data/SciTLDR-A/dev.jsonl": {"num_bytes": 1124865, "checksum": "3191fa98ccc09521332b7a1cd63b1930be4e8df125a235ccd31e40329709525e"}, "https://raw.githubusercontent.com/allenai/scitldr/master/SciTLDR-Data/SciTLDR-A/test.jsonl": {"num_bytes": 1204107, "checksum": "fb42dd6cd4f4a1928ae8a01a189456fbfe994a07e938bd49f68653933f6503c9"}}, "download_size": 5483987, "post_processing_size": null, "dataset_size": 4806597, "size_in_bytes": 10290584}, "AIC": {"description": "A new multi-target dataset of 5.4K TLDRs over 3.2K papers.\nSCITLDR contains both author-written and expert-derived TLDRs,\nwhere the latter are collected using a novel annotation protocol\nthat produces high-quality summaries while minimizing annotation burden.\n", "citation": "@article{cachola2020tldr,\n title={{TLDR}: Extreme Summarization of Scientific Documents},\n author={Isabel Cachola and Kyle Lo and Arman Cohan and Daniel S. Weld},\n journal={arXiv:2004.15011},\n year={2020},\n}\n", "homepage": "https://github.com/allenai/scitldr", "license": "Apache License 2.0", "features": {"source": {"feature": {"dtype": "string", "id": null, "_type": "Value"}, "length": -1, "id": null, "_type": "Sequence"}, "source_labels": {"feature": {"num_classes": 2, "names": [0, 1], "names_file": null, "id": null, "_type": "ClassLabel"}, "length": -1, "id": null, "_type": "Sequence"}, "rouge_scores": {"feature": {"dtype": "float32", "id": null, "_type": "Value"}, "length": -1, "id": null, "_type": "Sequence"}, "paper_id": {"dtype": "string", "id": null, "_type": "Value"}, "ic": {"dtype": "bool_", "id": null, "_type": "Value"}, "target": {"feature": {"dtype": "string", "id": null, "_type": "Value"}, "length": -1, "id": null, "_type": "Sequence"}}, "post_processed": null, "supervised_keys": {"input": "source", "output": "target"}, "builder_name": "scitldr", "config_name": "AIC", "version": {"version_str": "0.0.0", "description": null, "major": 0, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 14473822, "num_examples": 1992, "dataset_name": "scitldr"}, "test": {"name": "test", "num_bytes": 4822026, "num_examples": 618, "dataset_name": "scitldr"}, "validation": {"name": "validation", "num_bytes": 4476237, "num_examples": 619, "dataset_name": "scitldr"}}, "download_checksums": {"https://raw.githubusercontent.com/allenai/scitldr/master/SciTLDR-Data/SciTLDR-AIC/train.jsonl": {"num_bytes": 15569568, "checksum": "64b08af6de479671a12afd04770f66bcbc1c2c5f3098a08392b0fd7c1070d621"}, "https://raw.githubusercontent.com/allenai/scitldr/master/SciTLDR-Data/SciTLDR-AIC/dev.jsonl": {"num_bytes": 4811551, "checksum": "ac5168c27d25181fc17bb6f1fb41d11dbe30c627bebee14457feb3bad2c839dd"}, "https://raw.githubusercontent.com/allenai/scitldr/master/SciTLDR-Data/SciTLDR-AIC/test.jsonl": {"num_bytes": 5163989, "checksum": "7cb9230d3eb4863884762154918360d1c063aa18fc76de928801a14f4bcf4d37"}}, "download_size": 25545108, "post_processing_size": null, "dataset_size": 23772085, "size_in_bytes": 49317193}, "FullText": {"description": "A new multi-target dataset of 5.4K TLDRs over 3.2K papers.\nSCITLDR contains both author-written and expert-derived TLDRs,\nwhere the latter are collected using a novel annotation protocol\nthat produces high-quality summaries while minimizing annotation burden.\n", "citation": "@article{cachola2020tldr,\n title={{TLDR}: Extreme Summarization of Scientific Documents},\n author={Isabel Cachola and Kyle Lo and Arman Cohan and Daniel S. Weld},\n journal={arXiv:2004.15011},\n year={2020},\n}\n", "homepage": "https://github.com/allenai/scitldr", "license": "Apache License 2.0", "features": {"source": {"feature": {"dtype": "string", "id": null, "_type": "Value"}, "length": -1, "id": null, "_type": "Sequence"}, "source_labels": {"feature": {"num_classes": 2, "names": ["non-oracle", "oracle"], "names_file": null, "id": null, "_type": "ClassLabel"}, "length": -1, "id": null, "_type": "Sequence"}, "rouge_scores": {"feature": {"dtype": "float32", "id": null, "_type": "Value"}, "length": -1, "id": null, "_type": "Sequence"}, "paper_id": {"dtype": "string", "id": null, "_type": "Value"}, "target": {"feature": {"dtype": "string", "id": null, "_type": "Value"}, "length": -1, "id": null, "_type": "Sequence"}}, "post_processed": null, "supervised_keys": {"input": "source", "output": "target"}, "builder_name": "scitldr", "config_name": "FullText", "version": "0.0.0", "splits": {"train": {"name": "train", "num_bytes": 66917363, "num_examples": 1992, "dataset_name": "scitldr"}, "test": {"name": "test", "num_bytes": 20182554, "num_examples": 618, "dataset_name": "scitldr"}, "validation": {"name": "validation", "num_bytes": 18790651, "num_examples": 619, "dataset_name": "scitldr"}}, "download_checksums": {"https://raw.githubusercontent.com/allenai/scitldr/master/SciTLDR-Data/SciTLDR-FullText/train.jsonl": {"num_bytes": 71263949, "checksum": "e35461c1665cb4f7b46daba6dd5ac3cff03a61eb196e6ce9983edda44d867604"}, "https://raw.githubusercontent.com/allenai/scitldr/master/SciTLDR-Data/SciTLDR-FullText/dev.jsonl": {"num_bytes": 19111616, "checksum": "11c3fd77a7ec447adc44ca34c0fa41a7ab6bdacdf3b8e15748e6f8b8e4f698bf"}, "https://raw.githubusercontent.com/allenai/scitldr/master/SciTLDR-Data/SciTLDR-FullText/test.jsonl": {"num_bytes": 20528987, "checksum": "1584bd3f5fff5859cb8428cfbacc8d38c671f5fc6a24a8140ea5350cbd86a751"}}, "download_size": 110904552, "post_processing_size": null, "dataset_size": 105890568, "size_in_bytes": 216795120}}
 
 
scitldr.py DELETED
@@ -1,169 +0,0 @@
1
- # coding=utf-8
2
- # Copyright 2020 The HuggingFace Datasets Authors and the current dataset script contributor.
3
- #
4
- # Licensed under the Apache License, Version 2.0 (the "License");
5
- # you may not use this file except in compliance with the License.
6
- # You may obtain a copy of the License at
7
- #
8
- # http://www.apache.org/licenses/LICENSE-2.0
9
- #
10
- # Unless required by applicable law or agreed to in writing, software
11
- # distributed under the License is distributed on an "AS IS" BASIS,
12
- # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
13
- # See the License for the specific language governing permissions and
14
- # limitations under the License.
15
- """Dataset for TLDR: Extreme Summarization of Scientific Documents"""
16
-
17
-
18
- import json
19
- import os
20
-
21
- import datasets
22
-
23
-
24
- _SOURCE = "source"
25
- _TARGET = "target"
26
-
27
- _CITATION = """\
28
- @article{cachola2020tldr,
29
- title={{TLDR}: Extreme Summarization of Scientific Documents},
30
- author={Isabel Cachola and Kyle Lo and Arman Cohan and Daniel S. Weld},
31
- journal={arXiv:2004.15011},
32
- year={2020},
33
- }
34
- """
35
-
36
- _DESCRIPTION = """\
37
- A new multi-target dataset of 5.4K TLDRs over 3.2K papers.
38
- SCITLDR contains both author-written and expert-derived TLDRs,
39
- where the latter are collected using a novel annotation protocol
40
- that produces high-quality summaries while minimizing annotation burden.
41
- """
42
-
43
-
44
- _LICENSE = "Apache License 2.0"
45
-
46
- # TODO: Add link to the official dataset URLs here
47
- # The HuggingFace dataset library don't host the datasets but only point to the original files
48
- # This can be an arbitrary nested dict/list of URLs (see below in `_split_generators` method)
49
- _URLs = {
50
- "Abstract": "https://raw.githubusercontent.com/allenai/scitldr/master/SciTLDR-Data/SciTLDR-A/",
51
- "AIC": "https://raw.githubusercontent.com/allenai/scitldr/master/SciTLDR-Data/SciTLDR-AIC/",
52
- "FullText": "https://raw.githubusercontent.com/allenai/scitldr/master/SciTLDR-Data/SciTLDR-FullText/",
53
- }
54
-
55
- _TRAIN_DATA = "train.jsonl"
56
- _TEST_DATA = "test.jsonl"
57
- _VALID_DATA = "dev.jsonl"
58
-
59
-
60
- # There are several preprocessing scripts given in the original SciTLDR GitHub repository to preprocess this data.
61
- class Scitldr(datasets.GeneratorBasedBuilder):
62
- """Dataset for TLDR: Extreme Summarization of Scientific Documents."""
63
-
64
- VERSION = datasets.Version("1.1.0")
65
-
66
- # You will be able to load one or the other configurations in the following list with
67
- # data = datasets.load_dataset('scitldr', 'Abstract')
68
- # data = datasets.load_dataset('scitldr', 'AIC')
69
- BUILDER_CONFIGS = [
70
- datasets.BuilderConfig(name="Abstract", description="This part contains only abstracts of the paper"),
71
- datasets.BuilderConfig(
72
- name="AIC",
73
- description="This part contains Abstracts, Introduction and Conclusion (AIC) sections of the paper",
74
- ),
75
- datasets.BuilderConfig(name="FullText", description="This part contains the full text of the paper"),
76
- ]
77
-
78
- DEFAULT_CONFIG_NAME = (
79
- "Abstract" # It's not mandatory to have a default configuration. Just use one if it make sense.
80
- )
81
-
82
- def _info(self):
83
- if self.config.name == "AIC": # This is the name of the configuration selected in BUILDER_CONFIGS above
84
- features = datasets.Features(
85
- {
86
- "source": datasets.Sequence(datasets.Value("string")),
87
- "source_labels": datasets.Sequence(datasets.ClassLabel(num_classes=2, names=[0, 1])),
88
- "rouge_scores": datasets.Sequence(datasets.Value("float32")),
89
- "paper_id": datasets.Value("string"),
90
- "ic": datasets.Value("bool_"),
91
- "target": datasets.features.Sequence(datasets.Value("string"))
92
- # These are the features of your dataset like images, labels ...
93
- }
94
- )
95
- else:
96
- features = datasets.Features(
97
- {
98
- "source": datasets.Sequence(datasets.Value("string")),
99
- "source_labels": datasets.Sequence(
100
- datasets.ClassLabel(num_classes=2, names=["non-oracle", "oracle"])
101
- ),
102
- "rouge_scores": datasets.Sequence(datasets.Value("float32")),
103
- "paper_id": datasets.Value("string"),
104
- "target": datasets.Sequence(datasets.Value("string"))
105
- # These are the features of your dataset like images, labels ...
106
- }
107
- )
108
- return datasets.DatasetInfo(
109
- # This is the description that will appear on the datasets page.
110
- description=_DESCRIPTION,
111
- # This defines the different columns of the dataset and their types
112
- features=features, # Here we define them above because they are different between the two configurations
113
- # If there's a common (input, target) tuple from the features,
114
- # specify them here. They'll be used if as_supervised=True in
115
- # builder.as_dataset.
116
- supervised_keys=(_SOURCE, _TARGET),
117
- # Homepage of the dataset for documentation
118
- homepage="https://github.com/allenai/scitldr",
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
- """Returns SplitGenerators."""
127
- urls = {
128
- "train": _URLs[self.config.name] + _TRAIN_DATA,
129
- "valid": _URLs[self.config.name] + _VALID_DATA,
130
- "test": _URLs[self.config.name] + _TEST_DATA,
131
- }
132
- data_dir = dl_manager.download(urls)
133
- return [
134
- datasets.SplitGenerator(
135
- name=datasets.Split.TRAIN,
136
- gen_kwargs={"filepath": os.path.join(data_dir["train"])},
137
- ),
138
- datasets.SplitGenerator(
139
- name=datasets.Split.TEST,
140
- gen_kwargs={"filepath": os.path.join(data_dir["test"])},
141
- ),
142
- datasets.SplitGenerator(
143
- name=datasets.Split.VALIDATION,
144
- gen_kwargs={"filepath": os.path.join(data_dir["valid"])},
145
- ),
146
- ]
147
-
148
- def _generate_examples(self, filepath):
149
- """Yields examples."""
150
- with open(filepath, encoding="utf-8") as f:
151
- for id_, row in enumerate(f):
152
- data = json.loads(row)
153
- if self.config.name == "AIC":
154
- yield id_, {
155
- "source": data["source"],
156
- "source_labels": data["source_labels"],
157
- "rouge_scores": data["rouge_scores"],
158
- "paper_id": data["paper_id"],
159
- "ic": True if data["ic"] else False,
160
- "target": data["target"],
161
- }
162
- else:
163
- yield id_, {
164
- "source": data["source"],
165
- "source_labels": data["source_labels"],
166
- "rouge_scores": data["rouge_scores"],
167
- "paper_id": data["paper_id"],
168
- "target": data["target"],
169
- }