Datasets:
Tasks:
Text Generation
Formats:
parquet
Sub-tasks:
language-modeling
Languages:
Danish
Size:
1M - 10M
License:
KennethEnevoldsen
commited on
Added overview table to the main readme
Browse files- .gitignore +5 -1
- .vscode/launch.json +0 -16
- README.md +57 -46
- data/adl/adl.md +7 -39
- data/botxt/botxt.md +4 -1
- data/dannet/dannet.md +3 -0
- data/depbank/depbank.md +4 -1
- data/ep/ep.md +4 -1
- data/ft/ft.md +4 -1
- data/gutenberg/gutenberg.md +4 -1
- data/hest/hest.md +7 -39
- data/jvj/jvj.md +6 -31
- data/naat/naat.md +4 -1
- data/nordjyllandnews/nordjyllandnews.md +4 -2
- data/relig/relig.md +4 -1
- data/retsinformationdk/retsinformationdk.md +4 -1
- data/retspraksis/retspraksis.md +4 -2
- data/skat/skat.md +3 -1
- data/spont/spont.md +4 -2
- data/synne/synne.md +3 -1
- data/tv2r/tv2r.md +7 -5
- data/wiki/wiki.md +3 -11
- data/wikibooks/wikibooks.md +5 -3
- data/wikisource/wikisource.md +3 -1
- makefile +2 -2
- pyproject.toml +1 -0
- src/tests/readme_parsing.py +20 -1
- src/update_descriptive_statistics.py +76 -14
- uv.lock +11 -0
.gitignore
CHANGED
@@ -5,5 +5,9 @@ __pycache__/*
|
|
5 |
# cSpell
|
6 |
cspell.json
|
7 |
|
|
|
|
|
|
|
8 |
# tmp files
|
9 |
-
tmp.py
|
|
|
|
5 |
# cSpell
|
6 |
cspell.json
|
7 |
|
8 |
+
# debugfile
|
9 |
+
.vscode/launch.json
|
10 |
+
|
11 |
# tmp files
|
12 |
+
tmp.py
|
13 |
+
|
.vscode/launch.json
DELETED
@@ -1,16 +0,0 @@
|
|
1 |
-
{
|
2 |
-
// Use IntelliSense to learn about possible attributes.
|
3 |
-
// Hover to view descriptions of existing attributes.
|
4 |
-
// For more information, visit: https://go.microsoft.com/fwlink/?linkid=830387
|
5 |
-
"version": "0.2.0",
|
6 |
-
"configurations": [
|
7 |
-
{
|
8 |
-
"name": "Python Debugger: Current File with Arguments",
|
9 |
-
"type": "debugpy",
|
10 |
-
"request": "launch",
|
11 |
-
"program": "${file}",
|
12 |
-
"console": "integratedTerminal",
|
13 |
-
"args": "--force"
|
14 |
-
}
|
15 |
-
]
|
16 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
README.md
CHANGED
@@ -206,6 +206,11 @@ The dataset contains text from different sources which are thoroughly defined in
|
|
206 |
|
207 |
Each entry in the dataset consists of a single text with associated metadata
|
208 |
|
|
|
|
|
|
|
|
|
|
|
209 |
```py
|
210 |
{
|
211 |
"text": "SAMLEDE VÆRKER\n\nJEPPE AAKJÆR GYLDENDALSKE BOGHANDEL...",
|
@@ -252,61 +257,67 @@ This data generally contains no annotation besides the metadata attached to each
|
|
252 |
|
253 |
Below follows a brief overview of the sources in the corpus along with their individual license.
|
254 |
|
255 |
-
<!-- START-MAIN TABLE -->
|
256 |
-
<!-- END-MAIN TABLE -->
|
257 |
|
258 |
-
| Source | License |
|
259 |
-
| ------------------- | -------------------- |
|
260 |
-
| [adl] | [CC-0] |
|
261 |
-
| [botxt] | [CC-0] |
|
262 |
-
| [dannet] | [dannet license] |
|
263 |
-
| [depbank] | [CC-BY-SA 4.0] |
|
264 |
-
| [ep] | [CC-0] |
|
265 |
-
| [ft] | [CC-0] |
|
266 |
-
| [gutenberg] | [gutenberg license] |
|
267 |
-
| [hest] | [CC-0] |
|
268 |
-
| [jvj] | [CC-BY-SA 4.0] |
|
269 |
-
| [naat] | [CC-0] |
|
270 |
-
| [nordjyllandnews] | [CC-0] |
|
271 |
-
| [relig] | [CC-0] |
|
272 |
-
| [retsinformationdk] | [Other (Danish Law)] |
|
273 |
-
| [retspraksis] | [CC-0] |
|
274 |
-
| [skat] | [CC-0] |
|
275 |
-
| [spont] | [CC-0] |
|
276 |
-
| [synne] | [CC-0] |
|
277 |
-
| [tv2r] | [Custom, CC-BY 4.0] |
|
278 |
-
| [wiki] | [CC-0] |
|
279 |
-
| [wikibooks] | [CC-0] |
|
280 |
-
| [wikisource] | [CC-0] |
|
281 |
|
282 |
-
|
283 |
-
|
284 |
-
|
285 |
-
|
286 |
-
|
287 |
-
|
288 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
289 |
[hest]: data/hest/hest.md
|
290 |
-
[jvj]: data/jvj/jvj.md
|
291 |
-
[naat]: data/naat/naat.md
|
292 |
-
[nordjyllandnews]: data/nordjyllandnews/nordjyllandnews.md
|
293 |
-
[relig]: data/relig/relig.md
|
294 |
-
[retsinformationdk]: data/retsinformationdk/retsinformationdk.md
|
295 |
-
[retspraksis]: data/retspraksis/retspraksis.md
|
296 |
-
[skat]: data/skat/skat.md
|
297 |
[spont]: data/spont/spont.md
|
298 |
-
[
|
299 |
-
[
|
|
|
|
|
|
|
|
|
300 |
[wiki]: data/wiki/wiki.md
|
301 |
[wikibooks]: data/wikibooks/wikibooks.md
|
302 |
-
[
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
303 |
|
304 |
[CC-0]: https://creativecommons.org/publicdomain/zero/1.0/legalcode.en
|
305 |
[CC-BY-SA 4.0]: https://creativecommons.org/licenses/by-sa/4.0/deed.en
|
306 |
-
[
|
307 |
-
[
|
308 |
-
[
|
309 |
-
|
310 |
|
311 |
|
312 |
|
|
|
206 |
|
207 |
Each entry in the dataset consists of a single text with associated metadata
|
208 |
|
209 |
+
|
210 |
+
<!-- START-SAMPLE -->
|
211 |
+
<!-- END-SAMPLE -->
|
212 |
+
|
213 |
+
|
214 |
```py
|
215 |
{
|
216 |
"text": "SAMLEDE VÆRKER\n\nJEPPE AAKJÆR GYLDENDALSKE BOGHANDEL...",
|
|
|
257 |
|
258 |
Below follows a brief overview of the sources in the corpus along with their individual license.
|
259 |
|
|
|
|
|
260 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
261 |
|
262 |
+
|
263 |
+
|
264 |
+
|
265 |
+
|
266 |
+
<!-- START-MAIN TABLE -->
|
267 |
+
| Source | Description | N. Tokens | License |
|
268 |
+
| :------------------ | :--------------------------------------------------------------------------------------------------------------------------- | :-------- | :--------------------- |
|
269 |
+
| [retsinformationdk] | [retsinformation.dk](https://www.retsinformation.dk) (legal-information.dk) the official legal information system of Denmark | 516.54M | [Danish Copyright Law] |
|
270 |
+
| [hest] | Samples from the Danish debate forum www.heste-nettet.dk | 389.33M | [CC-0] |
|
271 |
+
| [spont] | Conversational samples collected as a part of research projects at Aarhus University | 1.56M | [CC-0] |
|
272 |
+
| [tv2r] | Contemporary Danish newswire articles published between 2010 and 2019 | 21.67M | [CC-BY-SA 4.0] |
|
273 |
+
| [ep] | The Danish subsection of [Europarl](https://aclanthology.org/2005.mtsummit-papers.11/) | 100.89M | [CC-0] |
|
274 |
+
| [gutenberg] | The Danish subsection from Project [Gutenberg](https://www.gutenberg.org) | 6.76M | [Gutenberg License] |
|
275 |
+
| [depbank] | The Danish subsection of the [Universal Dependencies Treebank](https://github.com/UniversalDependencies/UD_Danish-DDT) | 185.45K | [CC-BY-SA 4.0] |
|
276 |
+
| [jvj] | The works of the Danish author and poet, [Johannes V. Jensen](https://da.wikipedia.org/wiki/Johannes_V._Jensen) | 3.55M | [CC-BY-SA 4.0] |
|
277 |
+
| [wikisource] | The Danish subsection of [Wikisource](https://en.wikisource.org/wiki/Main_Page) | 5.34M | [CC-0] |
|
278 |
+
| [wiki] | The Danish subsection of [wikipeadia](https://en.wikipedia.org/wiki/Main_Page) | 122.00M | [CC-0] |
|
279 |
+
| [wikibooks] | The Danish Subsection of [Wikibooks](https://www.wikibooks.org) | 6.24M | [CC-0] |
|
280 |
+
| [nordjyllandnews] | Articles from the Danish Newspaper [TV2 Nord](https://www.tv2nord.dk) | 37.91M | [CC-0] |
|
281 |
+
| [adl] | Danish literature from 1700-2023 from the Archive for Danish Literature (ADL) | 58.49M | [CC-0] |
|
282 |
+
| [retspraksis] | Case law or judical practice in Denmark derived from [Retspraksis](https://da.wikipedia.org/wiki/Retspraksis) | 57.08M | [CC-0] |
|
283 |
+
| [relig] | Danish religious text from the 1700-2022 | 1.24M | [CC-0] |
|
284 |
+
| [dannet] | [DanNet](https://cst.ku.dk/projekter/dannet) is a Danish WordNet | 1.52M | [DanNet 1.0 License] |
|
285 |
+
| [synne] | Dataset collected from [synnejysk forening's website](https://www.synnejysk.dk), covering the Danish dialect sønderjysk | 52.51K | [CC-0] |
|
286 |
+
| [naat] | A dataset of Danish speeches from 1930-2022 | 286.68K | [CC-0] |
|
287 |
+
| [botxt] | The Bornholmsk Ordbog Dictionary Projec | 847.97K | [CC-0] |
|
288 |
+
| [ft] | This dataset consists of records from all meetings of The Danish parliament (Folketinget) in the parliament hall | 114.09M | [CC-0] |
|
289 |
+
| [skat] | Skat is the Danish tax authority. This dataset contains content from its website skat.dk | 122.12M | [CC-0] |
|
290 |
+
| **Total** | | 1.57B | |
|
291 |
+
|
292 |
+
[retsinformationdk]: data/retsinformationdk/retsinformationdk.md
|
293 |
[hest]: data/hest/hest.md
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
294 |
[spont]: data/spont/spont.md
|
295 |
+
[tv2r]: data/tv2r/tv2r.md
|
296 |
+
[ep]: data/ep/ep.md
|
297 |
+
[gutenberg]: data/gutenberg/gutenberg.md
|
298 |
+
[depbank]: data/depbank/depbank.md
|
299 |
+
[jvj]: data/jvj/jvj.md
|
300 |
+
[wikisource]: data/wikisource/wikisource.md
|
301 |
[wiki]: data/wiki/wiki.md
|
302 |
[wikibooks]: data/wikibooks/wikibooks.md
|
303 |
+
[nordjyllandnews]: data/nordjyllandnews/nordjyllandnews.md
|
304 |
+
[adl]: data/adl/adl.md
|
305 |
+
[retspraksis]: data/retspraksis/retspraksis.md
|
306 |
+
[relig]: data/relig/relig.md
|
307 |
+
[dannet]: data/dannet/dannet.md
|
308 |
+
[synne]: data/synne/synne.md
|
309 |
+
[naat]: data/naat/naat.md
|
310 |
+
[botxt]: data/botxt/botxt.md
|
311 |
+
[ft]: data/ft/ft.md
|
312 |
+
[skat]: data/skat/skat.md
|
313 |
+
|
314 |
|
315 |
[CC-0]: https://creativecommons.org/publicdomain/zero/1.0/legalcode.en
|
316 |
[CC-BY-SA 4.0]: https://creativecommons.org/licenses/by-sa/4.0/deed.en
|
317 |
+
[Danish Copyright Law]: ./data/retsinformationdk/retsinformationdk.md#license-information
|
318 |
+
[Gutenberg License]: ./data/gutenberg/gutenberg.md#license-information
|
319 |
+
[DanNet 1.0 License]: ./data/dannet/dannet.md#license-information
|
320 |
+
<!-- END-MAIN TABLE -->
|
321 |
|
322 |
|
323 |
|
data/adl/adl.md
CHANGED
@@ -3,7 +3,7 @@ pretty_name: Archive for Danish Literature
|
|
3 |
language:
|
4 |
- da
|
5 |
license: cc0-1.0
|
6 |
-
license_name:
|
7 |
size_categories:
|
8 |
- 1-10k
|
9 |
task_categories:
|
@@ -11,13 +11,16 @@ task_categories:
|
|
11 |
- fill-mask
|
12 |
task_ids:
|
13 |
- language-modeling
|
|
|
|
|
14 |
---
|
|
|
15 |
# Dataset Card for Archive for Danish Literature
|
16 |
|
17 |
## Dataset Description
|
18 |
|
19 |
<!-- START-SHORT DESCRIPTION -->
|
20 |
-
Danish literature from 1700-2023
|
21 |
<!-- END-SHORT DESCRIPTION -->
|
22 |
|
23 |
|
@@ -33,44 +36,9 @@ Danish literature from 1700-2023 stemming for the Archive for Danish Literature
|
|
33 |
|
34 |
## Dataset Sturcture
|
35 |
An example from the dataset looks as follows.
|
36 |
-
```yaml
|
37 |
-
{
|
38 |
-
'text': 'SAMLEDE VÆRKER
|
39 |
-
|
40 |
-
JEPPE AAKJÆR GYLDENDALSKE BOGHANDE',
|
41 |
-
'source': 'adl',
|
42 |
-
'id': 'adl_aakjaer06val',
|
43 |
-
'added': '2020-09-14',
|
44 |
-
'created': '1700-01-01, 2022-01-01',
|
45 |
-
'metadata': {
|
46 |
-
'domain': 'Wiki & Books',
|
47 |
-
'license': 'Creative Commons Legal Code
|
48 |
-
|
49 |
-
CC0 1.0 Universal',
|
50 |
-
'source-pretty': ' Archive for Danish Literature'
|
51 |
-
}
|
52 |
-
}
|
53 |
-
```
|
54 |
-
|
55 |
-
## Data Fields
|
56 |
-
|
57 |
-
- **id**: source-specific identifier.
|
58 |
-
- **text**: textual content of the document.
|
59 |
-
- **source**: source of the data.
|
60 |
-
- **added**: timestamp ai2 acquired this data.
|
61 |
-
- **created**": timestamp when original document was created (best-guess if not available)
|
62 |
-
- **metadata**: source-specific metadata.
|
63 |
-
|
64 |
-
## License Information
|
65 |
-
<details>
|
66 |
-
<summary>Creative Commons Zero v1.0 Universal</summary>
|
67 |
-
<p>
|
68 |
-
Creative Commons Legal Code
|
69 |
-
|
70 |
-
CC0 1.0 Universal
|
71 |
-
</p>
|
72 |
-
</details>
|
73 |
|
|
|
|
|
74 |
|
75 |
|
76 |
## Additional Information
|
|
|
3 |
language:
|
4 |
- da
|
5 |
license: cc0-1.0
|
6 |
+
license_name: CC-0
|
7 |
size_categories:
|
8 |
- 1-10k
|
9 |
task_categories:
|
|
|
11 |
- fill-mask
|
12 |
task_ids:
|
13 |
- language-modeling
|
14 |
+
source_datasets:
|
15 |
+
- danish-foundation-models/danish-gigaword
|
16 |
---
|
17 |
+
|
18 |
# Dataset Card for Archive for Danish Literature
|
19 |
|
20 |
## Dataset Description
|
21 |
|
22 |
<!-- START-SHORT DESCRIPTION -->
|
23 |
+
Danish literature from 1700-2023 from the Archive for Danish Literature (ADL).
|
24 |
<!-- END-SHORT DESCRIPTION -->
|
25 |
|
26 |
|
|
|
36 |
|
37 |
## Dataset Sturcture
|
38 |
An example from the dataset looks as follows.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
39 |
|
40 |
+
<!-- START-SAMPLE -->
|
41 |
+
<!-- END-SAMPLE -->
|
42 |
|
43 |
|
44 |
## Additional Information
|
data/botxt/botxt.md
CHANGED
@@ -3,7 +3,7 @@ pretty_name: Bornholmsk
|
|
3 |
language:
|
4 |
- da
|
5 |
license: cc0-1.0
|
6 |
-
license_name:
|
7 |
size_categories:
|
8 |
- 1-10k
|
9 |
task_categories:
|
@@ -11,7 +11,10 @@ task_categories:
|
|
11 |
- fill-mask
|
12 |
task_ids:
|
13 |
- language-modeling
|
|
|
|
|
14 |
---
|
|
|
15 |
# Dataset Card for Bornholmsk
|
16 |
|
17 |
## Dataset Description
|
|
|
3 |
language:
|
4 |
- da
|
5 |
license: cc0-1.0
|
6 |
+
license_name: CC-0
|
7 |
size_categories:
|
8 |
- 1-10k
|
9 |
task_categories:
|
|
|
11 |
- fill-mask
|
12 |
task_ids:
|
13 |
- language-modeling
|
14 |
+
source_datasets:
|
15 |
+
- danish-foundation-models/danish-gigaword
|
16 |
---
|
17 |
+
|
18 |
# Dataset Card for Bornholmsk
|
19 |
|
20 |
## Dataset Description
|
data/dannet/dannet.md
CHANGED
@@ -11,7 +11,10 @@ task_categories:
|
|
11 |
- fill-mask
|
12 |
task_ids:
|
13 |
- language-modeling
|
|
|
|
|
14 |
---
|
|
|
15 |
# Dataset Card for DanNet
|
16 |
|
17 |
<!-- START-SHORT DESCRIPTION -->
|
|
|
11 |
- fill-mask
|
12 |
task_ids:
|
13 |
- language-modeling
|
14 |
+
source_datasets:
|
15 |
+
- danish-foundation-models/danish-gigaword
|
16 |
---
|
17 |
+
|
18 |
# Dataset Card for DanNet
|
19 |
|
20 |
<!-- START-SHORT DESCRIPTION -->
|
data/depbank/depbank.md
CHANGED
@@ -3,7 +3,7 @@ pretty_name: Danish Dependency Treebank
|
|
3 |
language:
|
4 |
- da
|
5 |
license: cc-by-sa-4.0
|
6 |
-
license_name:
|
7 |
size_categories:
|
8 |
- 1-10k
|
9 |
task_categories:
|
@@ -11,7 +11,10 @@ task_categories:
|
|
11 |
- fill-mask
|
12 |
task_ids:
|
13 |
- language-modeling
|
|
|
|
|
14 |
---
|
|
|
15 |
# Dataset Card for Danish Dependency Treebank
|
16 |
|
17 |
<!-- START-SHORT DESCRIPTION -->
|
|
|
3 |
language:
|
4 |
- da
|
5 |
license: cc-by-sa-4.0
|
6 |
+
license_name: CC-BY-SA 4.0
|
7 |
size_categories:
|
8 |
- 1-10k
|
9 |
task_categories:
|
|
|
11 |
- fill-mask
|
12 |
task_ids:
|
13 |
- language-modeling
|
14 |
+
source_datasets:
|
15 |
+
- danish-foundation-models/danish-gigaword
|
16 |
---
|
17 |
+
|
18 |
# Dataset Card for Danish Dependency Treebank
|
19 |
|
20 |
<!-- START-SHORT DESCRIPTION -->
|
data/ep/ep.md
CHANGED
@@ -3,7 +3,7 @@ pretty_name: European Parliament
|
|
3 |
language:
|
4 |
- da
|
5 |
license: cc0-1.0
|
6 |
-
license_name:
|
7 |
size_categories:
|
8 |
- 1-10k
|
9 |
task_categories:
|
@@ -11,7 +11,10 @@ task_categories:
|
|
11 |
- fill-mask
|
12 |
task_ids:
|
13 |
- language-modeling
|
|
|
|
|
14 |
---
|
|
|
15 |
# Dataset Card for European Parliament
|
16 |
|
17 |
<!-- START-SHORT DESCRIPTION -->
|
|
|
3 |
language:
|
4 |
- da
|
5 |
license: cc0-1.0
|
6 |
+
license_name: CC-0
|
7 |
size_categories:
|
8 |
- 1-10k
|
9 |
task_categories:
|
|
|
11 |
- fill-mask
|
12 |
task_ids:
|
13 |
- language-modeling
|
14 |
+
source_datasets:
|
15 |
+
- danish-foundation-models/danish-gigaword
|
16 |
---
|
17 |
+
|
18 |
# Dataset Card for European Parliament
|
19 |
|
20 |
<!-- START-SHORT DESCRIPTION -->
|
data/ft/ft.md
CHANGED
@@ -3,7 +3,7 @@ pretty_name: Folketinget
|
|
3 |
language:
|
4 |
- da
|
5 |
license: cc0-1.0
|
6 |
-
license_name:
|
7 |
size_categories:
|
8 |
- 1-10k
|
9 |
task_categories:
|
@@ -11,7 +11,10 @@ task_categories:
|
|
11 |
- fill-mask
|
12 |
task_ids:
|
13 |
- language-modeling
|
|
|
|
|
14 |
---
|
|
|
15 |
# Dataset Card for Folketinget
|
16 |
|
17 |
## Dataset Description
|
|
|
3 |
language:
|
4 |
- da
|
5 |
license: cc0-1.0
|
6 |
+
license_name: CC-0
|
7 |
size_categories:
|
8 |
- 1-10k
|
9 |
task_categories:
|
|
|
11 |
- fill-mask
|
12 |
task_ids:
|
13 |
- language-modeling
|
14 |
+
source_datasets:
|
15 |
+
- danish-foundation-models/danish-gigaword
|
16 |
---
|
17 |
+
|
18 |
# Dataset Card for Folketinget
|
19 |
|
20 |
## Dataset Description
|
data/gutenberg/gutenberg.md
CHANGED
@@ -11,13 +11,16 @@ task_categories:
|
|
11 |
- fill-mask
|
12 |
task_ids:
|
13 |
- language-modeling
|
|
|
|
|
14 |
---
|
|
|
15 |
# Dataset Card for Gutenberg
|
16 |
|
17 |
## Dataset Description
|
18 |
|
19 |
<!-- START-SHORT DESCRIPTION -->
|
20 |
-
|
21 |
<!-- END-SHORT DESCRIPTION -->
|
22 |
|
23 |
|
|
|
11 |
- fill-mask
|
12 |
task_ids:
|
13 |
- language-modeling
|
14 |
+
source_datasets:
|
15 |
+
- danish-foundation-models/danish-gigaword
|
16 |
---
|
17 |
+
|
18 |
# Dataset Card for Gutenberg
|
19 |
|
20 |
## Dataset Description
|
21 |
|
22 |
<!-- START-SHORT DESCRIPTION -->
|
23 |
+
The Danish subsection from Project [Gutenberg](https://www.gutenberg.org).
|
24 |
<!-- END-SHORT DESCRIPTION -->
|
25 |
|
26 |
|
data/hest/hest.md
CHANGED
@@ -3,7 +3,7 @@ pretty_name: Hestenettet (Danish debate forum)
|
|
3 |
language:
|
4 |
- da
|
5 |
license: cc0-1.0
|
6 |
-
license_name:
|
7 |
size_categories:
|
8 |
- 10k-100k
|
9 |
task_categories:
|
@@ -11,11 +11,14 @@ task_categories:
|
|
11 |
- fill-mask
|
12 |
task_ids:
|
13 |
- language-modeling
|
|
|
|
|
14 |
---
|
|
|
15 |
# Dataset Card for Hestenettet
|
16 |
|
17 |
<!-- START-SHORT DESCRIPTION -->
|
18 |
-
|
19 |
<!-- END-SHORT DESCRIPTION -->
|
20 |
|
21 |
|
@@ -37,44 +40,9 @@ Its inclusion as training data for large language models have multiple times rea
|
|
37 |
|
38 |
## Dataset Sturcture
|
39 |
An example from the dataset looks as follows.
|
40 |
-
```yaml
|
41 |
-
{
|
42 |
-
'text': 'Er den ikke kær?
|
43 |
-
Jeg kan ikke forstå at der altid',
|
44 |
-
'source': 'hest',
|
45 |
-
'id': 'hest_forum112802271280227_0',
|
46 |
-
'added': '2020-10-05',
|
47 |
-
'created': '2000-01-01, 2022-01-01',
|
48 |
-
'metadata': {
|
49 |
-
'domain': 'Social Media',
|
50 |
-
'license': 'Creative Commons Legal Code
|
51 |
-
|
52 |
-
CC0 1.0 Universal',
|
53 |
-
'source-pretty': 'Hestenettet (Danish debate forum)'
|
54 |
-
}
|
55 |
-
}
|
56 |
-
```
|
57 |
-
|
58 |
-
## Data Fields
|
59 |
-
|
60 |
-
- **id**: source-specific identifier.
|
61 |
-
- **text**: textual content of the document.
|
62 |
-
- **source**: source of the data.
|
63 |
-
- **added**: timestamp ai2 acquired this data.
|
64 |
-
- **created**": timestamp when original document was created (best-guess if not available)
|
65 |
-
- **metadata**: source-specific metadata.
|
66 |
-
|
67 |
-
## License Information
|
68 |
-
<details>
|
69 |
-
<summary>Creative Commons Zero v1.0 Universal</summary>
|
70 |
-
<p>
|
71 |
-
Creative Commons Legal Code
|
72 |
-
|
73 |
-
CC0 1.0 Universal
|
74 |
-
</p>
|
75 |
-
</details>
|
76 |
-
|
77 |
|
|
|
|
|
78 |
|
79 |
## Additional Information
|
80 |
|
|
|
3 |
language:
|
4 |
- da
|
5 |
license: cc0-1.0
|
6 |
+
license_name: CC-0
|
7 |
size_categories:
|
8 |
- 10k-100k
|
9 |
task_categories:
|
|
|
11 |
- fill-mask
|
12 |
task_ids:
|
13 |
- language-modeling
|
14 |
+
source_datasets:
|
15 |
+
- danish-foundation-models/danish-gigaword
|
16 |
---
|
17 |
+
|
18 |
# Dataset Card for Hestenettet
|
19 |
|
20 |
<!-- START-SHORT DESCRIPTION -->
|
21 |
+
Samples from the Danish debate forum www.heste-nettet.dk.
|
22 |
<!-- END-SHORT DESCRIPTION -->
|
23 |
|
24 |
|
|
|
40 |
|
41 |
## Dataset Sturcture
|
42 |
An example from the dataset looks as follows.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
43 |
|
44 |
+
<!-- START-SAMPLE -->
|
45 |
+
<!-- END-SAMPLE -->
|
46 |
|
47 |
## Additional Information
|
48 |
|
data/jvj/jvj.md
CHANGED
@@ -3,7 +3,7 @@ pretty_name: Johannes V. Jensen
|
|
3 |
language:
|
4 |
- da
|
5 |
license: cc-by-sa-4.0
|
6 |
-
license_name:
|
7 |
size_categories:
|
8 |
- 1-10k
|
9 |
task_categories:
|
@@ -11,7 +11,10 @@ task_categories:
|
|
11 |
- fill-mask
|
12 |
task_ids:
|
13 |
- language-modeling
|
|
|
|
|
14 |
---
|
|
|
15 |
# Dataset Card for Johannes V. Jensen
|
16 |
|
17 |
<!-- START-SHORT DESCRIPTION -->
|
@@ -37,37 +40,9 @@ The works of the Danish author and poet, [Johannes V. Jensen](https://da.wikiped
|
|
37 |
|
38 |
## Dataset Sturcture
|
39 |
An example from the dataset looks as follows.
|
40 |
-
```yaml
|
41 |
-
{
|
42 |
-
'text': 'JØRGINE JØRGINE KØBENHAVN HAGE & CLAUSENS FORLAG (',
|
43 |
-
'source': 'jvj',
|
44 |
-
'id': 'jvj_Jørgine',
|
45 |
-
'added': '2020-06-26',
|
46 |
-
'created': '1873-01-01, 1951-01-01',
|
47 |
-
'metadata': {
|
48 |
-
'domain': 'Wiki & Books',
|
49 |
-
'license': 'Attribution-ShareAlike 4.0 International',
|
50 |
-
'source-pretty': 'Johannes V. Jensen (Danish poet)'
|
51 |
-
}
|
52 |
-
}
|
53 |
-
```
|
54 |
|
55 |
-
|
56 |
-
|
57 |
-
- **id**: source-specific identifier.
|
58 |
-
- **text**: textual content of the document.
|
59 |
-
- **source**: source of the data.
|
60 |
-
- **added**: timestamp ai2 acquired this data.
|
61 |
-
- **created**": timestamp when original document was created (best-guess if not available)
|
62 |
-
- **metadata**: source-specific metadata.
|
63 |
-
|
64 |
-
## License Information
|
65 |
-
<details>
|
66 |
-
<summary>Creative Commons Attribution Share Alike 4.0</summary>
|
67 |
-
<p>
|
68 |
-
Attribution-ShareAlike 4.0 International
|
69 |
-
</p>
|
70 |
-
</details>
|
71 |
|
72 |
|
73 |
## Additional Information
|
|
|
3 |
language:
|
4 |
- da
|
5 |
license: cc-by-sa-4.0
|
6 |
+
license_name: CC-BY-SA 4.0
|
7 |
size_categories:
|
8 |
- 1-10k
|
9 |
task_categories:
|
|
|
11 |
- fill-mask
|
12 |
task_ids:
|
13 |
- language-modeling
|
14 |
+
source_datasets:
|
15 |
+
- danish-foundation-models/danish-gigaword
|
16 |
---
|
17 |
+
|
18 |
# Dataset Card for Johannes V. Jensen
|
19 |
|
20 |
<!-- START-SHORT DESCRIPTION -->
|
|
|
40 |
|
41 |
## Dataset Sturcture
|
42 |
An example from the dataset looks as follows.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
43 |
|
44 |
+
<!-- START-SAMPLE -->
|
45 |
+
<!-- END-SAMPLE -->
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
46 |
|
47 |
|
48 |
## Additional Information
|
data/naat/naat.md
CHANGED
@@ -3,7 +3,7 @@ pretty_name: NAAT
|
|
3 |
language:
|
4 |
- da
|
5 |
license: cc0-1.0
|
6 |
-
license_name:
|
7 |
size_categories:
|
8 |
- 1-10k
|
9 |
task_categories:
|
@@ -11,7 +11,10 @@ task_categories:
|
|
11 |
- fill-mask
|
12 |
task_ids:
|
13 |
- language-modeling
|
|
|
|
|
14 |
---
|
|
|
15 |
# Dataset Card for NAAT
|
16 |
|
17 |
<!-- START-SHORT DESCRIPTION -->
|
|
|
3 |
language:
|
4 |
- da
|
5 |
license: cc0-1.0
|
6 |
+
license_name: CC-0
|
7 |
size_categories:
|
8 |
- 1-10k
|
9 |
task_categories:
|
|
|
11 |
- fill-mask
|
12 |
task_ids:
|
13 |
- language-modeling
|
14 |
+
source_datasets:
|
15 |
+
- danish-foundation-models/danish-gigaword
|
16 |
---
|
17 |
+
|
18 |
# Dataset Card for NAAT
|
19 |
|
20 |
<!-- START-SHORT DESCRIPTION -->
|
data/nordjyllandnews/nordjyllandnews.md
CHANGED
@@ -3,7 +3,7 @@ pretty_name: Nordjylland News
|
|
3 |
language:
|
4 |
- da
|
5 |
license: cc0-1.0
|
6 |
-
license_name:
|
7 |
size_categories:
|
8 |
- 10-100k
|
9 |
task_categories:
|
@@ -11,12 +11,14 @@ task_categories:
|
|
11 |
- fill-mask
|
12 |
task_ids:
|
13 |
- language-modeling
|
|
|
|
|
14 |
---
|
15 |
|
16 |
# Dataset Card for Nordjylland News
|
17 |
|
18 |
<!-- START-SHORT DESCRIPTION -->
|
19 |
-
Articles from Danish Newspaper [TV2 Nord](https://www.tv2nord.dk).
|
20 |
<!-- END-SHORT DESCRIPTION -->
|
21 |
|
22 |
|
|
|
3 |
language:
|
4 |
- da
|
5 |
license: cc0-1.0
|
6 |
+
license_name: CC-0
|
7 |
size_categories:
|
8 |
- 10-100k
|
9 |
task_categories:
|
|
|
11 |
- fill-mask
|
12 |
task_ids:
|
13 |
- language-modeling
|
14 |
+
source_datasets:
|
15 |
+
- alexandrainst/nordjylland-news-summarization
|
16 |
---
|
17 |
|
18 |
# Dataset Card for Nordjylland News
|
19 |
|
20 |
<!-- START-SHORT DESCRIPTION -->
|
21 |
+
Articles from the Danish Newspaper [TV2 Nord](https://www.tv2nord.dk).
|
22 |
<!-- END-SHORT DESCRIPTION -->
|
23 |
|
24 |
|
data/relig/relig.md
CHANGED
@@ -3,7 +3,7 @@ pretty_name: Religious texts
|
|
3 |
language:
|
4 |
- da
|
5 |
license: cc0-1.0
|
6 |
-
license_name:
|
7 |
size_categories:
|
8 |
- 1-10k
|
9 |
task_categories:
|
@@ -11,7 +11,10 @@ task_categories:
|
|
11 |
- fill-mask
|
12 |
task_ids:
|
13 |
- language-modeling
|
|
|
|
|
14 |
---
|
|
|
15 |
# Dataset Card for Religious texts
|
16 |
|
17 |
<!-- START-SHORT DESCRIPTION -->
|
|
|
3 |
language:
|
4 |
- da
|
5 |
license: cc0-1.0
|
6 |
+
license_name: CC-0
|
7 |
size_categories:
|
8 |
- 1-10k
|
9 |
task_categories:
|
|
|
11 |
- fill-mask
|
12 |
task_ids:
|
13 |
- language-modeling
|
14 |
+
source_datasets:
|
15 |
+
- danish-foundation-models/danish-gigaword
|
16 |
---
|
17 |
+
|
18 |
# Dataset Card for Religious texts
|
19 |
|
20 |
<!-- START-SHORT DESCRIPTION -->
|
data/retsinformationdk/retsinformationdk.md
CHANGED
@@ -11,11 +11,14 @@ task_categories:
|
|
11 |
- fill-mask
|
12 |
task_ids:
|
13 |
- language-modeling
|
|
|
|
|
14 |
---
|
|
|
15 |
# Dataset Card for retsinformation.dk (Danish legal information)
|
16 |
|
17 |
<!-- START-SHORT DESCRIPTION -->
|
18 |
-
[retsinformation.dk](https://www.retsinformation.dk) (legal-information.dk)
|
19 |
<!-- END-SHORT DESCRIPTION -->
|
20 |
|
21 |
|
|
|
11 |
- fill-mask
|
12 |
task_ids:
|
13 |
- language-modeling
|
14 |
+
source_datasets:
|
15 |
+
- danish-foundation-models/danish-gigaword
|
16 |
---
|
17 |
+
|
18 |
# Dataset Card for retsinformation.dk (Danish legal information)
|
19 |
|
20 |
<!-- START-SHORT DESCRIPTION -->
|
21 |
+
[retsinformation.dk](https://www.retsinformation.dk) (legal-information.dk) the official legal information system of Denmark.
|
22 |
<!-- END-SHORT DESCRIPTION -->
|
23 |
|
24 |
|
data/retspraksis/retspraksis.md
CHANGED
@@ -3,7 +3,7 @@ pretty_name: retspraksis (Danish legal information)
|
|
3 |
language:
|
4 |
- da
|
5 |
license: cc0-1.0
|
6 |
-
license_name:
|
7 |
size_categories:
|
8 |
- 1-10k
|
9 |
task_categories:
|
@@ -11,11 +11,13 @@ task_categories:
|
|
11 |
- fill-mask
|
12 |
task_ids:
|
13 |
- language-modeling
|
|
|
|
|
14 |
---
|
15 |
# Dataset Card for retspraksis
|
16 |
|
17 |
<!-- START-SHORT DESCRIPTION -->
|
18 |
-
[Retspraksis](https://da.wikipedia.org/wiki/Retspraksis)
|
19 |
<!-- END-SHORT DESCRIPTION -->
|
20 |
|
21 |
|
|
|
3 |
language:
|
4 |
- da
|
5 |
license: cc0-1.0
|
6 |
+
license_name: CC-0
|
7 |
size_categories:
|
8 |
- 1-10k
|
9 |
task_categories:
|
|
|
11 |
- fill-mask
|
12 |
task_ids:
|
13 |
- language-modeling
|
14 |
+
source_datasets:
|
15 |
+
- danish-foundation-models/danish-gigaword
|
16 |
---
|
17 |
# Dataset Card for retspraksis
|
18 |
|
19 |
<!-- START-SHORT DESCRIPTION -->
|
20 |
+
Case law or judical practice in Denmark derived from [Retspraksis](https://da.wikipedia.org/wiki/Retspraksis).
|
21 |
<!-- END-SHORT DESCRIPTION -->
|
22 |
|
23 |
|
data/skat/skat.md
CHANGED
@@ -3,7 +3,7 @@ pretty_name: skat.dk
|
|
3 |
language:
|
4 |
- da
|
5 |
license: cc0-1.0
|
6 |
-
license_name:
|
7 |
size_categories:
|
8 |
- 10k-100k
|
9 |
task_categories:
|
@@ -11,6 +11,8 @@ task_categories:
|
|
11 |
- fill-mask
|
12 |
task_ids:
|
13 |
- language-modeling
|
|
|
|
|
14 |
---
|
15 |
# Dataset Card for skat.dk
|
16 |
|
|
|
3 |
language:
|
4 |
- da
|
5 |
license: cc0-1.0
|
6 |
+
license_name: CC-0
|
7 |
size_categories:
|
8 |
- 10k-100k
|
9 |
task_categories:
|
|
|
11 |
- fill-mask
|
12 |
task_ids:
|
13 |
- language-modeling
|
14 |
+
source_datasets:
|
15 |
+
- danish-foundation-models/danish-gigaword
|
16 |
---
|
17 |
# Dataset Card for skat.dk
|
18 |
|
data/spont/spont.md
CHANGED
@@ -3,7 +3,7 @@ pretty_name: Spontaneous speech
|
|
3 |
language:
|
4 |
- da
|
5 |
license: cc0-1.0
|
6 |
-
license_name:
|
7 |
size_categories:
|
8 |
- 1-10k
|
9 |
task_categories:
|
@@ -11,11 +11,13 @@ task_categories:
|
|
11 |
- fill-mask
|
12 |
task_ids:
|
13 |
- language-modeling
|
|
|
|
|
14 |
---
|
15 |
# Dataset Card for Spontaneous speech
|
16 |
|
17 |
<!-- START-SHORT DESCRIPTION -->
|
18 |
-
|
19 |
<!-- END-SHORT DESCRIPTION -->
|
20 |
|
21 |
|
|
|
3 |
language:
|
4 |
- da
|
5 |
license: cc0-1.0
|
6 |
+
license_name: CC-0
|
7 |
size_categories:
|
8 |
- 1-10k
|
9 |
task_categories:
|
|
|
11 |
- fill-mask
|
12 |
task_ids:
|
13 |
- language-modeling
|
14 |
+
source_datasets:
|
15 |
+
- danish-foundation-models/danish-gigaword
|
16 |
---
|
17 |
# Dataset Card for Spontaneous speech
|
18 |
|
19 |
<!-- START-SHORT DESCRIPTION -->
|
20 |
+
Conversational samples collected as a part of research projects at Aarhus University.
|
21 |
<!-- END-SHORT DESCRIPTION -->
|
22 |
|
23 |
|
data/synne/synne.md
CHANGED
@@ -3,7 +3,7 @@ pretty_name: Synnejysk Forening
|
|
3 |
language:
|
4 |
- da
|
5 |
license: cc0-1.0
|
6 |
-
license_name:
|
7 |
size_categories:
|
8 |
- 1-10k
|
9 |
task_categories:
|
@@ -11,6 +11,8 @@ task_categories:
|
|
11 |
- fill-mask
|
12 |
task_ids:
|
13 |
- language-modeling
|
|
|
|
|
14 |
---
|
15 |
# Dataset Card for synnejysk Forening
|
16 |
|
|
|
3 |
language:
|
4 |
- da
|
5 |
license: cc0-1.0
|
6 |
+
license_name: CC-0
|
7 |
size_categories:
|
8 |
- 1-10k
|
9 |
task_categories:
|
|
|
11 |
- fill-mask
|
12 |
task_ids:
|
13 |
- language-modeling
|
14 |
+
source_datasets:
|
15 |
+
- danish-foundation-models/danish-gigaword
|
16 |
---
|
17 |
# Dataset Card for synnejysk Forening
|
18 |
|
data/tv2r/tv2r.md
CHANGED
@@ -1,9 +1,9 @@
|
|
1 |
---
|
2 |
-
pretty_name: TV 2 Radio
|
3 |
language:
|
4 |
- da
|
5 |
license: cc-by-sa-4.0
|
6 |
-
license_name:
|
7 |
size_categories:
|
8 |
- 10k-100k
|
9 |
task_categories:
|
@@ -11,13 +11,15 @@ task_categories:
|
|
11 |
- fill-mask
|
12 |
task_ids:
|
13 |
- language-modeling
|
|
|
|
|
14 |
---
|
15 |
-
# Dataset Card for TV 2 Radio
|
16 |
|
17 |
## Dataset Description
|
18 |
|
19 |
<!-- START-SHORT DESCRIPTION -->
|
20 |
-
|
21 |
<!-- END-SHORT DESCRIPTION -->
|
22 |
|
23 |
|
@@ -41,7 +43,7 @@ An example from the dataset looks as follows.
|
|
41 |
|
42 |
## License Information
|
43 |
<details>
|
44 |
-
<summary>
|
45 |
<p>
|
46 |
The owner of this content is TV2 Regionerne, Denmark.
|
47 |
Creative Commons Attribution 4.0 International
|
|
|
1 |
---
|
2 |
+
pretty_name: TV 2 Radio
|
3 |
language:
|
4 |
- da
|
5 |
license: cc-by-sa-4.0
|
6 |
+
license_name: CC-BY-SA 4.0
|
7 |
size_categories:
|
8 |
- 10k-100k
|
9 |
task_categories:
|
|
|
11 |
- fill-mask
|
12 |
task_ids:
|
13 |
- language-modeling
|
14 |
+
source_datasets:
|
15 |
+
- danish-foundation-models/danish-gigaword
|
16 |
---
|
17 |
+
# Dataset Card for TV 2 Radio
|
18 |
|
19 |
## Dataset Description
|
20 |
|
21 |
<!-- START-SHORT DESCRIPTION -->
|
22 |
+
Contemporary Danish newswire articles published between 2010 and 2019.
|
23 |
<!-- END-SHORT DESCRIPTION -->
|
24 |
|
25 |
|
|
|
43 |
|
44 |
## License Information
|
45 |
<details>
|
46 |
+
<summary>CC-BY-SA 4.0</summary>
|
47 |
<p>
|
48 |
The owner of this content is TV2 Regionerne, Denmark.
|
49 |
Creative Commons Attribution 4.0 International
|
data/wiki/wiki.md
CHANGED
@@ -3,7 +3,7 @@ pretty_name: Wikipedia
|
|
3 |
language:
|
4 |
- da
|
5 |
license: cc0-1.0
|
6 |
-
license_name:
|
7 |
size_categories:
|
8 |
- 100k-1M
|
9 |
task_categories:
|
@@ -11,6 +11,8 @@ task_categories:
|
|
11 |
- fill-mask
|
12 |
task_ids:
|
13 |
- language-modeling
|
|
|
|
|
14 |
---
|
15 |
# Dataset Card for Wikipedia
|
16 |
|
@@ -39,16 +41,6 @@ An example from the dataset looks as follows.
|
|
39 |
<!-- START-SAMPLE -->
|
40 |
<!-- END-SAMPLE -->
|
41 |
|
42 |
-
## License Information
|
43 |
-
<details>
|
44 |
-
<summary>Creative Commons Zero v1.0 Universal</summary>
|
45 |
-
<p>
|
46 |
-
Creative Commons Legal Code
|
47 |
-
|
48 |
-
CC0 1.0 Universal
|
49 |
-
</p>
|
50 |
-
</details>
|
51 |
-
|
52 |
|
53 |
## Additional Information
|
54 |
|
|
|
3 |
language:
|
4 |
- da
|
5 |
license: cc0-1.0
|
6 |
+
license_name: CC-0
|
7 |
size_categories:
|
8 |
- 100k-1M
|
9 |
task_categories:
|
|
|
11 |
- fill-mask
|
12 |
task_ids:
|
13 |
- language-modeling
|
14 |
+
source_datasets:
|
15 |
+
- danish-foundation-models/danish-gigaword
|
16 |
---
|
17 |
# Dataset Card for Wikipedia
|
18 |
|
|
|
41 |
<!-- START-SAMPLE -->
|
42 |
<!-- END-SAMPLE -->
|
43 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
44 |
|
45 |
## Additional Information
|
46 |
|
data/wikibooks/wikibooks.md
CHANGED
@@ -3,7 +3,7 @@ pretty_name: Wikibooks
|
|
3 |
language:
|
4 |
- da
|
5 |
license: cc0-1.0
|
6 |
-
license_name:
|
7 |
size_categories:
|
8 |
- 1-10k
|
9 |
task_categories:
|
@@ -11,12 +11,14 @@ task_categories:
|
|
11 |
- fill-mask
|
12 |
task_ids:
|
13 |
- language-modeling
|
|
|
|
|
14 |
---
|
15 |
|
16 |
# Dataset Card for Wikibooks
|
17 |
|
18 |
<!-- START-SHORT DESCRIPTION -->
|
19 |
-
The Danish Subsection of [
|
20 |
<!-- END-SHORT DESCRIPTION -->
|
21 |
|
22 |
|
@@ -40,7 +42,7 @@ An example from the dataset looks as follows.
|
|
40 |
|
41 |
## License Information
|
42 |
<details>
|
43 |
-
<summary>
|
44 |
<p>
|
45 |
Creative Commons Legal Code
|
46 |
|
|
|
3 |
language:
|
4 |
- da
|
5 |
license: cc0-1.0
|
6 |
+
license_name: CC-0
|
7 |
size_categories:
|
8 |
- 1-10k
|
9 |
task_categories:
|
|
|
11 |
- fill-mask
|
12 |
task_ids:
|
13 |
- language-modeling
|
14 |
+
source_datasets:
|
15 |
+
- danish-foundation-models/danish-gigaword
|
16 |
---
|
17 |
|
18 |
# Dataset Card for Wikibooks
|
19 |
|
20 |
<!-- START-SHORT DESCRIPTION -->
|
21 |
+
The Danish Subsection of [Wikibooks](https://www.wikibooks.org).
|
22 |
<!-- END-SHORT DESCRIPTION -->
|
23 |
|
24 |
|
|
|
42 |
|
43 |
## License Information
|
44 |
<details>
|
45 |
+
<summary>CC-0</summary>
|
46 |
<p>
|
47 |
Creative Commons Legal Code
|
48 |
|
data/wikisource/wikisource.md
CHANGED
@@ -3,7 +3,7 @@ pretty_name: Wikisource
|
|
3 |
language:
|
4 |
- da
|
5 |
license: cc0-1.0
|
6 |
-
license_name:
|
7 |
size_categories:
|
8 |
- 1-10k
|
9 |
task_categories:
|
@@ -11,6 +11,8 @@ task_categories:
|
|
11 |
- fill-mask
|
12 |
task_ids:
|
13 |
- language-modeling
|
|
|
|
|
14 |
---
|
15 |
# Dataset Card for Wikisource
|
16 |
|
|
|
3 |
language:
|
4 |
- da
|
5 |
license: cc0-1.0
|
6 |
+
license_name: CC-0
|
7 |
size_categories:
|
8 |
- 1-10k
|
9 |
task_categories:
|
|
|
11 |
- fill-mask
|
12 |
task_ids:
|
13 |
- language-modeling
|
14 |
+
source_datasets:
|
15 |
+
- danish-foundation-models/danish-gigaword
|
16 |
---
|
17 |
# Dataset Card for Wikisource
|
18 |
|
makefile
CHANGED
@@ -14,8 +14,8 @@ lint:
|
|
14 |
|
15 |
bump-version:
|
16 |
@echo "--- 🚀 Bumping patch version ---"
|
17 |
-
uv run
|
18 |
|
19 |
update-descriptive-statistics:
|
20 |
@echo "--- 🚀 Recomputing Descriptive statistics ---"
|
21 |
-
uv run
|
|
|
14 |
|
15 |
bump-version:
|
16 |
@echo "--- 🚀 Bumping patch version ---"
|
17 |
+
uv run src/bump_version.py
|
18 |
|
19 |
update-descriptive-statistics:
|
20 |
@echo "--- 🚀 Recomputing Descriptive statistics ---"
|
21 |
+
uv run src/update_descriptive_statistics.py
|
pyproject.toml
CHANGED
@@ -15,6 +15,7 @@ dependencies = [
|
|
15 |
"pytest>=8.3.4",
|
16 |
"ruff>=0.8.3",
|
17 |
"seaborn>=0.13.2",
|
|
|
18 |
"tomlkit>=0.13.2",
|
19 |
"transformers>=4.47.1",
|
20 |
]
|
|
|
15 |
"pytest>=8.3.4",
|
16 |
"ruff>=0.8.3",
|
17 |
"seaborn>=0.13.2",
|
18 |
+
"tabulate>=0.9.0",
|
19 |
"tomlkit>=0.13.2",
|
20 |
"transformers>=4.47.1",
|
21 |
]
|
src/tests/readme_parsing.py
CHANGED
@@ -15,7 +15,7 @@ def read_frontmatter_and_body(file_path: Path) -> tuple[dict[str, Any], str]:
|
|
15 |
raise ValueError(f"No frontmatter found in file: {file_path}")
|
16 |
|
17 |
|
18 |
-
def get_tag_idx(readme: str, tag: str):
|
19 |
tag_start = f"<!-- START-{tag} -->"
|
20 |
tag_end = f"<!-- END-{tag} -->"
|
21 |
start_idx = readme.find(tag_start)
|
@@ -23,3 +23,22 @@ def get_tag_idx(readme: str, tag: str):
|
|
23 |
if end_idx != -1 and start_idx != -1 and start_idx < end_idx:
|
24 |
return start_idx, end_idx
|
25 |
raise ValueError(f"tag ({tag}) not found in readme")
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
15 |
raise ValueError(f"No frontmatter found in file: {file_path}")
|
16 |
|
17 |
|
18 |
+
def get_tag_idx(readme: str, tag: str) -> tuple[int, int]:
|
19 |
tag_start = f"<!-- START-{tag} -->"
|
20 |
tag_end = f"<!-- END-{tag} -->"
|
21 |
start_idx = readme.find(tag_start)
|
|
|
23 |
if end_idx != -1 and start_idx != -1 and start_idx < end_idx:
|
24 |
return start_idx, end_idx
|
25 |
raise ValueError(f"tag ({tag}) not found in readme")
|
26 |
+
|
27 |
+
|
28 |
+
def get_tag_content(readme: str, tag: str) -> str:
|
29 |
+
s, e = get_tag_idx(readme, tag=tag)
|
30 |
+
tag_start = f"<!-- START-{tag} -->"
|
31 |
+
return readme[s + len(tag_start) : e].strip()
|
32 |
+
|
33 |
+
|
34 |
+
def replace_tag(markdown: str, package: str, tag: str) -> str:
|
35 |
+
tag_start = f"<!-- START-{tag} -->"
|
36 |
+
tag_end = f"<!-- END-{tag} -->"
|
37 |
+
|
38 |
+
if markdown.count(tag_start) != 1 or markdown.count(tag_end) != 1:
|
39 |
+
raise ValueError("Markers should appear exactly once in the markdown.")
|
40 |
+
|
41 |
+
start_md, _, remainder = markdown.partition(tag_start)
|
42 |
+
_, _, end_md = remainder.partition(tag_end)
|
43 |
+
|
44 |
+
return f"{start_md}\n{tag_start}\n{package.strip()}\n{tag_end}\n{end_md}"
|
src/update_descriptive_statistics.py
CHANGED
@@ -16,10 +16,13 @@ from pathlib import Path
|
|
16 |
from textwrap import dedent
|
17 |
from typing import Self, cast
|
18 |
|
|
|
19 |
from datasets import Dataset, load_dataset
|
20 |
-
from git_utilities import check_is_ancestor, get_current_revision, get_latest_revision
|
21 |
from transformers import AutoTokenizer
|
22 |
|
|
|
|
|
|
|
23 |
logger = logging.getLogger(__name__)
|
24 |
|
25 |
repo_path = Path(__file__).parent.parent
|
@@ -97,17 +100,9 @@ class DescriptiveStatsOverview:
|
|
97 |
return format
|
98 |
|
99 |
def add_to_markdown(self, markdown: str) -> str:
|
100 |
-
|
101 |
-
|
102 |
-
|
103 |
-
if markdown.count(start_identifier) != 1 or markdown.count(end_identifier) != 1:
|
104 |
-
raise ValueError("Markers should appear exactly once in the markdown.")
|
105 |
-
|
106 |
-
start_md, _, remainder = markdown.partition(start_identifier)
|
107 |
-
_, _, end_md = remainder.partition(end_identifier)
|
108 |
-
|
109 |
-
stats = self.to_markdown()
|
110 |
-
return f"{start_md}{start_identifier}{stats}{end_identifier}{end_md}"
|
111 |
|
112 |
def to_disk(self, path: Path):
|
113 |
data = self.__dict__
|
@@ -115,6 +110,15 @@ class DescriptiveStatsOverview:
|
|
115 |
with path.with_suffix(".json").open("w") as f:
|
116 |
json.dump(self.__dict__, f)
|
117 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
118 |
|
119 |
def update_statitics(
|
120 |
dataset_path: Path,
|
@@ -183,9 +187,66 @@ def create_parser():
|
|
183 |
)
|
184 |
return parser
|
185 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
186 |
def update_main_table(repo_path: Path = repo_path):
|
187 |
-
|
188 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
189 |
|
190 |
|
191 |
def main(
|
@@ -206,6 +267,7 @@ def main(
|
|
206 |
update_statitics(dataset_path, dataset_path.name, force=force)
|
207 |
|
208 |
update_statitics(repo_path, "default", "README.md", force=force)
|
|
|
209 |
|
210 |
|
211 |
if __name__ == "__main__":
|
|
|
16 |
from textwrap import dedent
|
17 |
from typing import Self, cast
|
18 |
|
19 |
+
import pandas as pd
|
20 |
from datasets import Dataset, load_dataset
|
|
|
21 |
from transformers import AutoTokenizer
|
22 |
|
23 |
+
from git_utilities import check_is_ancestor, get_current_revision, get_latest_revision
|
24 |
+
from tests.readme_parsing import get_tag_content, read_frontmatter_and_body, replace_tag
|
25 |
+
|
26 |
logger = logging.getLogger(__name__)
|
27 |
|
28 |
repo_path = Path(__file__).parent.parent
|
|
|
100 |
return format
|
101 |
|
102 |
def add_to_markdown(self, markdown: str) -> str:
|
103 |
+
return replace_tag(
|
104 |
+
markdown=markdown, package=self.to_markdown(), tag="DESC-STATS"
|
105 |
+
)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
106 |
|
107 |
def to_disk(self, path: Path):
|
108 |
data = self.__dict__
|
|
|
110 |
with path.with_suffix(".json").open("w") as f:
|
111 |
json.dump(self.__dict__, f)
|
112 |
|
113 |
+
@classmethod
|
114 |
+
def from_disk(cls, path: Path):
|
115 |
+
with path.open("r") as f:
|
116 |
+
data = json.load(f)
|
117 |
+
if "revision" in data:
|
118 |
+
data.pop("revision")
|
119 |
+
obj = cls(**data)
|
120 |
+
return obj
|
121 |
+
|
122 |
|
123 |
def update_statitics(
|
124 |
dataset_path: Path,
|
|
|
187 |
)
|
188 |
return parser
|
189 |
|
190 |
+
|
191 |
+
def create_main_table(repo_path: Path = repo_path) -> pd.DataFrame:
|
192 |
+
datasets = (repo_path / "data").glob("*")
|
193 |
+
|
194 |
+
table = {
|
195 |
+
"Source": [],
|
196 |
+
"Description": [],
|
197 |
+
# "Domain": [], # TODO Add domain
|
198 |
+
"N. Tokens": [],
|
199 |
+
"License": [],
|
200 |
+
}
|
201 |
+
readme_references = ""
|
202 |
+
license_references = (
|
203 |
+
"[CC-0]: https://creativecommons.org/publicdomain/zero/1.0/legalcode.en\n"
|
204 |
+
+ "[CC-BY-SA 4.0]: https://creativecommons.org/licenses/by-sa/4.0/deed.en\n"
|
205 |
+
)
|
206 |
+
|
207 |
+
for dataset in datasets:
|
208 |
+
readme_path = dataset / f"{dataset.name}.md"
|
209 |
+
frontmatter, body = read_frontmatter_and_body(readme_path)
|
210 |
+
desc_stats = DescriptiveStatsOverview.from_disk(
|
211 |
+
dataset / "descriptive_stats.json"
|
212 |
+
)
|
213 |
+
|
214 |
+
short_description = get_tag_content(body, tag="SHORT DESCRIPTION").strip()[:-1] # to exclude "."
|
215 |
+
license, license_name = frontmatter["license"], frontmatter["license_name"]
|
216 |
+
|
217 |
+
table["Source"] += [f"[{dataset.name}]"]
|
218 |
+
readme_references += (
|
219 |
+
f"[{dataset.name}]: data/{dataset.name}/{dataset.name}.md\n"
|
220 |
+
)
|
221 |
+
|
222 |
+
table["License"] += [f"[{license_name}]"]
|
223 |
+
if license == "other":
|
224 |
+
license_references += f"[{license_name}]: ./data/{dataset.name}/{dataset.name}.md#license-information\n"
|
225 |
+
table["Description"] += [short_description]
|
226 |
+
table["N. Tokens"] += [desc_stats.number_of_tokens]
|
227 |
+
|
228 |
+
# total
|
229 |
+
table["Source"] += ["**Total**"]
|
230 |
+
# table["Domain"] += [""]
|
231 |
+
table["License"] += [""]
|
232 |
+
table["Description"] += [""]
|
233 |
+
table["N. Tokens"] += [sum(table["N. Tokens"])]
|
234 |
+
|
235 |
+
df = pd.DataFrame.from_dict(table)
|
236 |
+
df["N. Tokens"] = df["N. Tokens"].apply(human_readable_large_int)
|
237 |
+
return df, readme_references, license_references
|
238 |
+
|
239 |
+
|
240 |
def update_main_table(repo_path: Path = repo_path):
|
241 |
+
logging.info("Updating MAIN TABLE")
|
242 |
+
main_table, readme_references, license_references = create_main_table(repo_path)
|
243 |
+
readme_path = repo_path / "README.md"
|
244 |
+
with readme_path.open("r") as f:
|
245 |
+
markdown = f.read()
|
246 |
+
package = f"{main_table.to_markdown(index=False)}\n\n{readme_references}\n\n{license_references}\n\n"
|
247 |
+
markdown = replace_tag(markdown, package=package, tag="MAIN TABLE")
|
248 |
+
with readme_path.open("w") as f:
|
249 |
+
f.write(markdown)
|
250 |
|
251 |
|
252 |
def main(
|
|
|
267 |
update_statitics(dataset_path, dataset_path.name, force=force)
|
268 |
|
269 |
update_statitics(repo_path, "default", "README.md", force=force)
|
270 |
+
update_main_table(repo_path)
|
271 |
|
272 |
|
273 |
if __name__ == "__main__":
|
uv.lock
CHANGED
@@ -268,6 +268,7 @@ dependencies = [
|
|
268 |
{ name = "pytest" },
|
269 |
{ name = "ruff" },
|
270 |
{ name = "seaborn" },
|
|
|
271 |
{ name = "tomlkit" },
|
272 |
{ name = "transformers" },
|
273 |
]
|
@@ -284,6 +285,7 @@ requires-dist = [
|
|
284 |
{ name = "pytest", specifier = ">=8.3.4" },
|
285 |
{ name = "ruff", specifier = ">=0.8.3" },
|
286 |
{ name = "seaborn", specifier = ">=0.13.2" },
|
|
|
287 |
{ name = "tomlkit", specifier = ">=0.13.2" },
|
288 |
{ name = "transformers", specifier = ">=4.47.1" },
|
289 |
]
|
@@ -1477,6 +1479,15 @@ wheels = [
|
|
1477 |
{ url = "https://files.pythonhosted.org/packages/1d/eb/cb8b01f5edf8f135eb3d0553d159db113a35b2948d0e51eeb735e7ae09ea/statsmodels-0.14.4-cp313-cp313-win_amd64.whl", hash = "sha256:81030108d27aecc7995cac05aa280cf8c6025f6a6119894eef648997936c2dd0", size = 9817574 },
|
1478 |
]
|
1479 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1480 |
[[package]]
|
1481 |
name = "tokenizers"
|
1482 |
version = "0.21.0"
|
|
|
268 |
{ name = "pytest" },
|
269 |
{ name = "ruff" },
|
270 |
{ name = "seaborn" },
|
271 |
+
{ name = "tabulate" },
|
272 |
{ name = "tomlkit" },
|
273 |
{ name = "transformers" },
|
274 |
]
|
|
|
285 |
{ name = "pytest", specifier = ">=8.3.4" },
|
286 |
{ name = "ruff", specifier = ">=0.8.3" },
|
287 |
{ name = "seaborn", specifier = ">=0.13.2" },
|
288 |
+
{ name = "tabulate", specifier = ">=0.9.0" },
|
289 |
{ name = "tomlkit", specifier = ">=0.13.2" },
|
290 |
{ name = "transformers", specifier = ">=4.47.1" },
|
291 |
]
|
|
|
1479 |
{ url = "https://files.pythonhosted.org/packages/1d/eb/cb8b01f5edf8f135eb3d0553d159db113a35b2948d0e51eeb735e7ae09ea/statsmodels-0.14.4-cp313-cp313-win_amd64.whl", hash = "sha256:81030108d27aecc7995cac05aa280cf8c6025f6a6119894eef648997936c2dd0", size = 9817574 },
|
1480 |
]
|
1481 |
|
1482 |
+
[[package]]
|
1483 |
+
name = "tabulate"
|
1484 |
+
version = "0.9.0"
|
1485 |
+
source = { registry = "https://pypi.org/simple" }
|
1486 |
+
sdist = { url = "https://files.pythonhosted.org/packages/ec/fe/802052aecb21e3797b8f7902564ab6ea0d60ff8ca23952079064155d1ae1/tabulate-0.9.0.tar.gz", hash = "sha256:0095b12bf5966de529c0feb1fa08671671b3368eec77d7ef7ab114be2c068b3c", size = 81090 }
|
1487 |
+
wheels = [
|
1488 |
+
{ url = "https://files.pythonhosted.org/packages/40/44/4a5f08c96eb108af5cb50b41f76142f0afa346dfa99d5296fe7202a11854/tabulate-0.9.0-py3-none-any.whl", hash = "sha256:024ca478df22e9340661486f85298cff5f6dcdba14f3813e8830015b9ed1948f", size = 35252 },
|
1489 |
+
]
|
1490 |
+
|
1491 |
[[package]]
|
1492 |
name = "tokenizers"
|
1493 |
version = "0.21.0"
|