Datasets:
Tasks:
Text Generation
Modalities:
Text
Formats:
parquet
Languages:
Portuguese
Size:
1M - 10M
Tags:
portuguese
License:
Update README.md
Browse files
README.md
CHANGED
@@ -14,4 +14,81 @@ configs:
|
|
14 |
data_files:
|
15 |
- split: train
|
16 |
path: data/train-*
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
17 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
14 |
data_files:
|
15 |
- split: train
|
16 |
path: data/train-*
|
17 |
+
license: cc-by-sa-3.0
|
18 |
+
task_categories:
|
19 |
+
- text-generation
|
20 |
+
language:
|
21 |
+
- pt
|
22 |
+
pretty_name: Wikipedia-PT
|
23 |
+
size_categories:
|
24 |
+
- 1M<n<10M
|
25 |
+
tags:
|
26 |
+
- portuguese
|
27 |
---
|
28 |
+
|
29 |
+
# Portuguese-PT
|
30 |
+
|
31 |
+
- **Repository:** [nicholasKluge/wikipedia-PT](https://huggingface.co/datasets/nicholasKluge/wikipedia-PT)
|
32 |
+
-
|
33 |
+
## Dataset Summary
|
34 |
+
|
35 |
+
The Portuguese portion of the [Wikipedia dataset](https://huggingface.co/datasets/wikimedia/wikipedia).
|
36 |
+
|
37 |
+
### Supported Tasks and Leaderboards
|
38 |
+
|
39 |
+
The dataset is generally used for Language Modeling.
|
40 |
+
|
41 |
+
### Languages
|
42 |
+
|
43 |
+
Portuguese
|
44 |
+
|
45 |
+
## Dataset Structure
|
46 |
+
|
47 |
+
### Data Instances
|
48 |
+
|
49 |
+
An example looks as follows:
|
50 |
+
```
|
51 |
+
{
|
52 |
+
'text': 'Abril é o quarto mês...'
|
53 |
+
}
|
54 |
+
```
|
55 |
+
|
56 |
+
### Data Fields
|
57 |
+
|
58 |
+
- `text` (`str`): Text content of the article.
|
59 |
+
|
60 |
+
### Data Splits
|
61 |
+
|
62 |
+
All configurations contain a single `train` split.
|
63 |
+
|
64 |
+
## Dataset Creation
|
65 |
+
|
66 |
+
### Initial Data Collection and Normalization
|
67 |
+
|
68 |
+
The dataset is built from the Wikipedia dumps: https://dumps.wikimedia.org
|
69 |
+
|
70 |
+
You can find the full list of languages and dates here: https://dumps.wikimedia.org/backup-index.html
|
71 |
+
|
72 |
+
The articles have been parsed using the [`mwparserfromhell`](https://mwparserfromhell.readthedocs.io) tool.
|
73 |
+
|
74 |
+
When uploading the data files for the 20231101 dump, we noticed that the Wikimedia Dumps website does not contain this date dump
|
75 |
+
for the "bbc", "dga", nor "zgh" Wikipedias. We have reported the issue to the Wikimedia Phabricator: https://phabricator.wikimedia.org/T351761
|
76 |
+
|
77 |
+
### Licensing Information
|
78 |
+
|
79 |
+
Copyright licensing information: https://dumps.wikimedia.org/legal.html
|
80 |
+
|
81 |
+
All original textual content is licensed under the [GNU Free Documentation License](https://www.gnu.org/licenses/fdl-1.3.html) (GFDL)
|
82 |
+
and the [Creative Commons Attribution-Share-Alike 3.0 License](https://creativecommons.org/licenses/by-sa/3.0/).
|
83 |
+
Some text may be available only under the Creative Commons license; see their [Terms of Use](https://foundation.wikimedia.org/wiki/Policy:Terms_of_Use) for details.
|
84 |
+
Text written by some authors may be released under additional licenses or into the public domain.
|
85 |
+
|
86 |
+
## Citation Information
|
87 |
+
|
88 |
+
```
|
89 |
+
@ONLINE{wikidump,
|
90 |
+
author = "Wikimedia Foundation",
|
91 |
+
title = "Wikimedia Downloads",
|
92 |
+
url = "https://dumps.wikimedia.org"
|
93 |
+
}
|
94 |
+
```
|