cointegrated
commited on
Commit
•
b5de338
1
Parent(s):
0b6e564
Update README.md
Browse files
README.md
CHANGED
@@ -1,4 +1,13 @@
|
|
1 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2 |
|
3 |
Example inference code (with the correct NLLB preprocessing!):
|
4 |
|
@@ -84,3 +93,4 @@ def batched_translate(texts, batch_size=16, **kwargs):
|
|
84 |
return [p for i, p in sorted(zip(idxs, results))]
|
85 |
```
|
86 |
|
|
|
|
1 |
+
---
|
2 |
+
language:
|
3 |
+
- en
|
4 |
+
- az
|
5 |
+
pipeline_tag: translation
|
6 |
+
tags:
|
7 |
+
- translation
|
8 |
+
---
|
9 |
+
This is a version of NLLB fine-tuned to translate sentences between eng and azj languages,
|
10 |
+
using the corresponding subset of https://github.com/turkic-interlingua/til-mt/tree/master/til_corpus.
|
11 |
|
12 |
Example inference code (with the correct NLLB preprocessing!):
|
13 |
|
|
|
93 |
return [p for i, p in sorted(zip(idxs, results))]
|
94 |
```
|
95 |
|
96 |
+
Please beware that for translating a longer text, you need to split it into sentences and process them individually.
|