losca commited on
Commit
9144e1b
1 Parent(s): 8f1941f

Update README.md

Browse files

Added example usage. It SHOULD work

Files changed (1) hide show
  1. README.md +17 -0
README.md CHANGED
@@ -13,6 +13,23 @@ metrics:
13
 
14
  This is a finetuning of a MarianMT pretrained on Chinese-English. The target language pair is Vietnamese-English.
15
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
16
 
17
  ### Training results
18
 
 
13
 
14
  This is a finetuning of a MarianMT pretrained on Chinese-English. The target language pair is Vietnamese-English.
15
 
16
+ ### Example
17
+ ```
18
+ %%capture
19
+ !pip install transformers transformers[sentencepiece]
20
+
21
+ from transformers import AutoModelForSeq2SeqLM, AutoTokenizer
22
+ # Download the pretrained model for English-Vietnamese available on the hub
23
+ model = AutoModelForSeq2SeqLM.from_pretrained("CLAck/vi-en")
24
+
25
+ tokenizer = AutoTokenizer.from_pretrained("CLAck/vi-en")
26
+
27
+ sentence = your_vietnamese_sentence
28
+ # This token is needed to identify the source language
29
+ input_sentence = "<2vi> " + sentence
30
+ translated = model.generate(**tokenizer(input_sentence, return_tensors="pt", padding=True))
31
+ output_sentence = [tokenizer.decode(t, skip_special_tokens=True) for t in translated]
32
+ ```
33
 
34
  ### Training results
35