Datasets:

Multilinguality:
translation
Size Categories:
10K<n<100K
Language Creators:
found
Annotations Creators:
expert-generated
found
Source Datasets:
original
ArXiv:
License:

added new details for the dev and train set, paper details and new github repo link

#3
by jmukiibi - opened
Files changed (1) hide show
  1. README.md +16 -10
README.md CHANGED
@@ -33,8 +33,14 @@ dataset_info:
33
  - name: train
34
  num_bytes: 2551273
35
  num_examples: 10070
36
- download_size: 2490852
37
- dataset_size: 2551273
 
 
 
 
 
 
38
  ---
39
 
40
  # Dataset Card for MENYO-20k
@@ -66,8 +72,8 @@ dataset_info:
66
  ## Dataset Description
67
 
68
  - **Homepage:** [Homepage for Menyo-20k](https://zenodo.org/record/4297448#.X81G7s0zZPY)
69
- - **Repository:**[Github Repo](https://github.com/dadelani/menyo-20k_MT)
70
- - **Paper:**
71
  - **Leaderboard:**
72
  - **Point of Contact:**
73
 
@@ -81,7 +87,7 @@ MENYO-20k is a multi-domain parallel dataset with texts obtained from news artic
81
 
82
  ### Languages
83
 
84
- Languages are English and YOruba
85
 
86
  ## Dataset Structure
87
 
@@ -90,7 +96,7 @@ Languages are English and YOruba
90
  The data consists of tab seperated entries
91
 
92
  ```
93
- {'translation':
94
  {'en': 'Unit 1: What is Creative Commons?',
95
  'yo': 'Ìdá 1: Kín ni Creative Commons?'
96
  }
@@ -106,7 +112,7 @@ The data consists of tab seperated entries
106
 
107
  ### Data Splits
108
 
109
- Only training dataset available
110
 
111
  ## Dataset Creation
112
 
@@ -160,7 +166,7 @@ Only training dataset available
160
 
161
  ### Licensing Information
162
 
163
- The dataset is open but for non-commercial use because some of the data sources like Ted talks and JW news requires permission for commercial use.
164
 
165
  ### Citation Information
166
  ```
@@ -171,7 +177,7 @@ The dataset is open but for non-commercial use because some of the data sources
171
  Adesina Ayeni and
172
  Mofe Adeyemi and
173
  Ayodele Awokoya},
174
- title = {{MENYO-20k: A Multi-domain English - Yorùbá Corpus
175
  for Machine Translation}},
176
  month = nov,
177
  year = 2020,
@@ -183,4 +189,4 @@ The dataset is open but for non-commercial use because some of the data sources
183
  ```
184
  ### Contributions
185
 
186
- Thanks to [@yvonnegitau](https://github.com/yvonnegitau) for adding this dataset.
 
33
  - name: train
34
  num_bytes: 2551273
35
  num_examples: 10070
36
+ download_size: 2490852
37
+ - name: dev
38
+ num_examples: 3397
39
+ download_size: 849789
40
+ - name: test
41
+ num_examples: 6633
42
+ download_size: 1865593
43
+
44
  ---
45
 
46
  # Dataset Card for MENYO-20k
 
72
  ## Dataset Description
73
 
74
  - **Homepage:** [Homepage for Menyo-20k](https://zenodo.org/record/4297448#.X81G7s0zZPY)
75
+ - **Repository:**[Github Repo](https://github.com/uds-lsv/menyo-20k_MT/)
76
+ - **Paper:** [African NLP 2021](https://www.lsv.uni-saarland.de/wp-content/uploads/2022/06/africanlp2021_EACL_menyo20k.pdf)
77
  - **Leaderboard:**
78
  - **Point of Contact:**
79
 
 
87
 
88
  ### Languages
89
 
90
+ Languages are English and Yoruba
91
 
92
  ## Dataset Structure
93
 
 
96
  The data consists of tab seperated entries
97
 
98
  ```
99
+ {'translation':
100
  {'en': 'Unit 1: What is Creative Commons?',
101
  'yo': 'Ìdá 1: Kín ni Creative Commons?'
102
  }
 
112
 
113
  ### Data Splits
114
 
115
+ Training, Validation and Test datasets are available.
116
 
117
  ## Dataset Creation
118
 
 
166
 
167
  ### Licensing Information
168
 
169
+ The dataset is open but for non-commercial use, because some data sources like Ted talks and JW news require permission for commercial use.
170
 
171
  ### Citation Information
172
  ```
 
177
  Adesina Ayeni and
178
  Mofe Adeyemi and
179
  Ayodele Awokoya},
180
+ title = {{MENYO-20k: A Multi-domain English - Yorùbá Corpus
181
  for Machine Translation}},
182
  month = nov,
183
  year = 2020,
 
189
  ```
190
  ### Contributions
191
 
192
+ Thanks to [@yvonnegitau](https://github.com/yvonnegitau) for adding this dataset.