Datasets:
Tasks:
Translation
Languages:
English
Multilinguality:
translation
Size Categories:
10K<n<100K
Source Datasets:
original
Tags:
License:
blank lines removed, new total is 13232 lines
Browse files
README.md
CHANGED
@@ -20,11 +20,12 @@ source_datasets:
|
|
20 |
# Dataset Summary:
|
21 |
Pig-latin machine and English parallel machine translation corpus.
|
22 |
|
23 |
-
|
24 |
Based on [The Project Gutenberg EBook of "De Bello Gallico" and Other Commentaries](https://www.gutenberg.org/ebooks/10657)
|
25 |
|
26 |
Converted to pig-latin with https://github.com/bpabel/piglatin
|
27 |
|
|
|
|
|
28 |
## Dataset Structure
|
29 |
|
30 |
|
@@ -58,5 +59,5 @@ DatasetDict({
|
|
58 |
- `translation`: a dictionary containing two strings paired with a key indicating the corresponding language.
|
59 |
|
60 |
### Data Splits
|
61 |
-
- `train`: most of the data,
|
62 |
- `dev`: 1k holdout samples, created with the datasets.train_test_split() function
|
20 |
# Dataset Summary:
|
21 |
Pig-latin machine and English parallel machine translation corpus.
|
22 |
|
|
|
23 |
Based on [The Project Gutenberg EBook of "De Bello Gallico" and Other Commentaries](https://www.gutenberg.org/ebooks/10657)
|
24 |
|
25 |
Converted to pig-latin with https://github.com/bpabel/piglatin
|
26 |
|
27 |
+
Blank lines removed.
|
28 |
+
|
29 |
## Dataset Structure
|
30 |
|
31 |
|
59 |
- `translation`: a dictionary containing two strings paired with a key indicating the corresponding language.
|
60 |
|
61 |
### Data Splits
|
62 |
+
- `train`: most of the data, 13,232 samples total.
|
63 |
- `dev`: 1k holdout samples, created with the datasets.train_test_split() function
|