Datasets:
Modalities:
Text
Size:
10K - 100K
Text revision + New sources
Browse files- README.md +18 -46
- index.html +33 -20
README.md
CHANGED
@@ -1,58 +1,26 @@
|
|
1 |
-
---
|
2 |
-
annotations_creators:
|
3 |
-
- expert-generated
|
4 |
-
language:
|
5 |
-
- bam
|
6 |
-
- fr
|
7 |
-
language_creators:
|
8 |
-
- expert-generated
|
9 |
-
- found
|
10 |
-
- crowdsourced
|
11 |
-
license:
|
12 |
-
- cc-by-sa-4.0
|
13 |
-
multilinguality:
|
14 |
-
- translation
|
15 |
-
pretty_name: "BAY\u0190L\u0190MABAGA: Parallel French - Bambara Dataset for Machine\
|
16 |
-
\ Learning"
|
17 |
-
size_categories:
|
18 |
-
- 10K<n<100K
|
19 |
-
source_datasets: []
|
20 |
-
tags:
|
21 |
-
- bambara
|
22 |
-
- french
|
23 |
-
- periodicals
|
24 |
-
task_categories:
|
25 |
-
- translation
|
26 |
-
- text-generation
|
27 |
-
task_ids:
|
28 |
-
- language-modeling
|
29 |
-
|
30 |
-
viewer: true
|
31 |
-
---
|
32 |
-
|
33 |
# BAYƐLƐMABAGA: Parallel French - Bambara Dataset for Machine Learning
|
34 |
|
35 |
## Overview
|
36 |
-
The Bayelemabaga dataset is a collection of
|
37 |
|
38 |
-
## Snapshot:
|
39 |
| | |
|
40 |
|:---|---:|
|
41 |
-
| **Lines** | **
|
42 |
-
| French Tokens |
|
43 |
-
| Bambara Tokens |
|
44 |
-
| French Types |
|
45 |
-
| Bambara Types |
|
46 |
-
| Avg. Fr line length |
|
47 |
-
| Avg. Bam line length |
|
48 |
-
| Number of text sources |
|
49 |
|
50 |
## Data Splits
|
51 |
| | | |
|
52 |
|:-----:|:---:|------:|
|
53 |
-
| Train | 80% |
|
54 |
-
| Valid | 10% |
|
55 |
-
| Test | 10% |
|
56 |
||
|
57 |
|
58 |
## Remarks
|
@@ -76,9 +44,10 @@ The Bayelemabaga dataset is a collection of 44160 aligned machine translation re
|
|
76 |
|
77 |
## Version
|
78 |
|
79 |
-
- `1.0.
|
80 |
|
81 |
## Citation
|
|
|
82 |
```
|
83 |
@misc{bayelemabagamldataset2022
|
84 |
title={Machine Learning Dataset Development for Manding Languages},
|
@@ -97,3 +66,6 @@ The Bayelemabaga dataset is a collection of 44160 aligned machine translation re
|
|
97 |
year={2022}
|
98 |
}
|
99 |
```
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
# BAYƐLƐMABAGA: Parallel French - Bambara Dataset for Machine Learning
|
2 |
|
3 |
## Overview
|
4 |
+
The Bayelemabaga dataset is a collection of 44562 aligned machine translation ready Bambara-French lines, originating from [Corpus Bambara de Reference](http://cormande.huma-num.fr/corbama/run.cgi/first_form). The dataset is constitued of text extracted from **231** text files, varing from periodicals, books, short stories, blog posts, part of the Bible and the Quran.
|
5 |
|
6 |
+
## Snapshot: 46976
|
7 |
| | |
|
8 |
|:---|---:|
|
9 |
+
| **Lines** | **46976** |
|
10 |
+
| French Tokens | --- |
|
11 |
+
| Bambara Tokens | --- |
|
12 |
+
| French Types | --- |
|
13 |
+
| Bambara Types | --- |
|
14 |
+
| Avg. Fr line length | -- |
|
15 |
+
| Avg. Bam line length | -- |
|
16 |
+
| Number of text sources | 264 |
|
17 |
|
18 |
## Data Splits
|
19 |
| | | |
|
20 |
|:-----:|:---:|------:|
|
21 |
+
| Train | 80% | 37580 |
|
22 |
+
| Valid | 10% | 4698 |
|
23 |
+
| Test | 10% | 4698 |
|
24 |
||
|
25 |
|
26 |
## Remarks
|
|
|
44 |
|
45 |
## Version
|
46 |
|
47 |
+
- `1.0.1`
|
48 |
|
49 |
## Citation
|
50 |
+
|
51 |
```
|
52 |
@misc{bayelemabagamldataset2022
|
53 |
title={Machine Learning Dataset Development for Manding Languages},
|
|
|
66 |
year={2022}
|
67 |
}
|
68 |
```
|
69 |
+
|
70 |
+
## Contacts
|
71 |
+
- `sdiarra <at> robotsmali.org`
|
index.html
CHANGED
@@ -22,9 +22,10 @@
|
|
22 |
<h1>
|
23 |
BAYƐLƐMABAGA: Parallel French - Bambara Dataset for Machine Learning
|
24 |
</h1>
|
|
|
25 |
<h2>Overview</h2>
|
26 |
<p>
|
27 |
-
The Bayelemabaga dataset is a collection of
|
28 |
translation ready Bambara-French lines, originating from
|
29 |
<a href="http://cormande.huma-num.fr/corbama/run.cgi/first_form"
|
30 |
>Corpus Bambara de Reference</a
|
@@ -43,35 +44,35 @@
|
|
43 |
<tbody>
|
44 |
<tr>
|
45 |
<td align="left"><strong>Lines</strong></td>
|
46 |
-
<td align="right"><strong>
|
47 |
</tr>
|
48 |
<tr>
|
49 |
<td align="left">French Tokens</td>
|
50 |
-
<td align="right"
|
51 |
</tr>
|
52 |
<tr>
|
53 |
<td align="left">Bambara Tokens</td>
|
54 |
-
<td align="right"
|
55 |
</tr>
|
56 |
<tr>
|
57 |
<td align="left">French Types</td>
|
58 |
-
<td align="right"
|
59 |
</tr>
|
60 |
<tr>
|
61 |
<td align="left">Bambara Types</td>
|
62 |
-
<td align="right"
|
63 |
</tr>
|
64 |
<tr>
|
65 |
<td align="left">Avg. Fr line length</td>
|
66 |
-
<td align="right"
|
67 |
</tr>
|
68 |
<tr>
|
69 |
<td align="left">Avg. Bam line length</td>
|
70 |
-
<td align="right"
|
71 |
</tr>
|
72 |
<tr>
|
73 |
<td align="left">Number of text sources</td>
|
74 |
-
<td align="right"
|
75 |
</tr>
|
76 |
</tbody>
|
77 |
</table>
|
@@ -88,17 +89,17 @@
|
|
88 |
<tr>
|
89 |
<td align="center">Train</td>
|
90 |
<td align="center">80%</td>
|
91 |
-
<td align="right">
|
92 |
</tr>
|
93 |
<tr>
|
94 |
<td align="center">Valid</td>
|
95 |
<td align="center">10%</td>
|
96 |
-
<td align="right">
|
97 |
</tr>
|
98 |
<tr>
|
99 |
<td align="center">Test</td>
|
100 |
<td align="center">10%</td>
|
101 |
-
<td align="right">
|
102 |
</tr>
|
103 |
<tr>
|
104 |
<td align="center"></td>
|
@@ -135,21 +136,33 @@
|
|
135 |
<h2>Version</h2>
|
136 |
<ul>
|
137 |
<li>
|
138 |
-
<code>1.0.
|
139 |
</li>
|
140 |
</ul>
|
141 |
<h2>Citation</h2>
|
142 |
-
<
|
143 |
-
|
144 |
-
|
145 |
-
|
146 |
-
|
147 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
148 |
|
149 |
<button onclick="download_triggered()">Download</button>
|
150 |
<div id="code_sent"></div>
|
151 |
<div>
|
152 |
-
For any questions, contact <code>
|
153 |
</div>
|
154 |
<br />
|
155 |
<div>This dataset is distributed under the CC-BY-SA-4.0 LICENSE</div>
|
|
|
22 |
<h1>
|
23 |
BAYƐLƐMABAGA: Parallel French - Bambara Dataset for Machine Learning
|
24 |
</h1>
|
25 |
+
<div><a href="https://huggingface.co/datasets/RobotsMaliAI/bayelemabaga">HuggingFace: Bayelemabaga</a></div>
|
26 |
<h2>Overview</h2>
|
27 |
<p>
|
28 |
+
The Bayelemabaga dataset is a collection of 46976 aligned machine
|
29 |
translation ready Bambara-French lines, originating from
|
30 |
<a href="http://cormande.huma-num.fr/corbama/run.cgi/first_form"
|
31 |
>Corpus Bambara de Reference</a
|
|
|
44 |
<tbody>
|
45 |
<tr>
|
46 |
<td align="left"><strong>Lines</strong></td>
|
47 |
+
<td align="right"><strong>44976</strong></td>
|
48 |
</tr>
|
49 |
<tr>
|
50 |
<td align="left">French Tokens</td>
|
51 |
+
<td align="right">--</td>
|
52 |
</tr>
|
53 |
<tr>
|
54 |
<td align="left">Bambara Tokens</td>
|
55 |
+
<td align="right">--</td>
|
56 |
</tr>
|
57 |
<tr>
|
58 |
<td align="left">French Types</td>
|
59 |
+
<td align="right">--</td>
|
60 |
</tr>
|
61 |
<tr>
|
62 |
<td align="left">Bambara Types</td>
|
63 |
+
<td align="right">---</td>
|
64 |
</tr>
|
65 |
<tr>
|
66 |
<td align="left">Avg. Fr line length</td>
|
67 |
+
<td align="right">---</td>
|
68 |
</tr>
|
69 |
<tr>
|
70 |
<td align="left">Avg. Bam line length</td>
|
71 |
+
<td align="right">---</td>
|
72 |
</tr>
|
73 |
<tr>
|
74 |
<td align="left">Number of text sources</td>
|
75 |
+
<td align="right">---</td>
|
76 |
</tr>
|
77 |
</tbody>
|
78 |
</table>
|
|
|
89 |
<tr>
|
90 |
<td align="center">Train</td>
|
91 |
<td align="center">80%</td>
|
92 |
+
<td align="right">37580</td>
|
93 |
</tr>
|
94 |
<tr>
|
95 |
<td align="center">Valid</td>
|
96 |
<td align="center">10%</td>
|
97 |
+
<td align="right">4698</td>
|
98 |
</tr>
|
99 |
<tr>
|
100 |
<td align="center">Test</td>
|
101 |
<td align="center">10%</td>
|
102 |
+
<td align="right">4698</td>
|
103 |
</tr>
|
104 |
<tr>
|
105 |
<td align="center"></td>
|
|
|
136 |
<h2>Version</h2>
|
137 |
<ul>
|
138 |
<li>
|
139 |
+
<code>1.0.1</code>
|
140 |
</li>
|
141 |
</ul>
|
142 |
<h2>Citation</h2>
|
143 |
+
<code>
|
144 |
+
@misc{bayelemabagamldataset2022
|
145 |
+
title={Machine Learning Dataset Development for Manding Languages},
|
146 |
+
author={
|
147 |
+
Valentin Vydrin and
|
148 |
+
Christopher Homan and
|
149 |
+
Michael Leventhal and
|
150 |
+
Allashera Auguste Tapo and
|
151 |
+
Marco Zampieri and
|
152 |
+
Jean-Jacques Meric and
|
153 |
+
Kirill Maslinsky and
|
154 |
+
Andrij Rovenchak and
|
155 |
+
Sebastien Diarra
|
156 |
+
},
|
157 |
+
howpublished = {url{https://github.com/robotsmali-ai/datasets}},
|
158 |
+
year={2022}
|
159 |
+
}
|
160 |
+
</code>
|
161 |
|
162 |
<button onclick="download_triggered()">Download</button>
|
163 |
<div id="code_sent"></div>
|
164 |
<div>
|
165 |
+
For any questions, contact <code>sdiarra@robotsmali.org</code>
|
166 |
</div>
|
167 |
<br />
|
168 |
<div>This dataset is distributed under the CC-BY-SA-4.0 LICENSE</div>
|