Create README.md
Browse files
README.md
ADDED
@@ -0,0 +1,27 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
language:
|
3 |
+
- en
|
4 |
+
pipeline_tag: text2text-generation
|
5 |
+
metrics:
|
6 |
+
- f1
|
7 |
+
tags:
|
8 |
+
- grammatical error correction
|
9 |
+
- GEC
|
10 |
+
- russian
|
11 |
+
---
|
12 |
+
|
13 |
+
This is a fine-tuned version of Multilingual Bart trained on Russian in particular on the public dataset RULEC-GEC for Grammatical Error Correction.
|
14 |
+
|
15 |
+
To initialize the model:
|
16 |
+
|
17 |
+
|
18 |
+
from transformers import MBartForConditionalGeneration, MBart50TokenizerFast
|
19 |
+
model = MBartForConditionalGeneration.from_pretrained("MRNH/mbart-russian-grammar-corrector")
|
20 |
+
|
21 |
+
|
22 |
+
To generate text using the model:
|
23 |
+
|
24 |
+
|
25 |
+
tokenizer = MBart50TokenizerFast.from_pretrained("MRNH/mbart-russian-grammar-corrector", src_lang="ru_RU", tgt_lang="ru_RU")
|
26 |
+
input = tokenizer("I was here yesterday to studying",text_target="I was here yesterday to study", return_tensors='pt')
|
27 |
+
output = model.generate(input["input_ids"],attention_mask=input["attention_mask"],forced_bos_token_id=tokenizer_it.lang_code_to_id["ru_RU"])
|