Update README.md
Browse files
README.md
CHANGED
|
@@ -14,6 +14,7 @@ it also contains duplicates, false translations and huge chunks of unrelated dat
|
|
| 14 |
making the dataset very noisy and almost unsuitable for any usage.
|
| 15 |
We have deduplicated, denoised and cleaned the dataset, reviewed the translations,
|
| 16 |
retranslated and cleared the meaning across all the entries to reach a standard version suitable for deep learning tasks.
|
|
|
|
| 17 |
|
| 18 |
## Dataset Details
|
| 19 |
|
|
|
|
| 14 |
making the dataset very noisy and almost unsuitable for any usage.
|
| 15 |
We have deduplicated, denoised and cleaned the dataset, reviewed the translations,
|
| 16 |
retranslated and cleared the meaning across all the entries to reach a standard version suitable for deep learning tasks.
|
| 17 |
+
This is ~3500 first entries of the entire OPUS-100 corpus.
|
| 18 |
|
| 19 |
## Dataset Details
|
| 20 |
|