hafsteinn commited on
Commit
0321fb7
1 Parent(s): 759c047

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +17 -0
README.md CHANGED
@@ -20,3 +20,20 @@ This model was trained to translate multiple sentences at once, compared to one
20
  It will occasionally combine sentences or add an extra sentence.
21
 
22
  This is the same model as are provided on CLARIN: https://repository.clarin.is/repository/xmlui/handle/20.500.12537/278
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
20
  It will occasionally combine sentences or add an extra sentence.
21
 
22
  This is the same model as are provided on CLARIN: https://repository.clarin.is/repository/xmlui/handle/20.500.12537/278
23
+
24
+ You can use the following example to get started (note that it is necessary to alter the `decoder_start_token_id` of the model):
25
+
26
+ from transformers import AutoTokenizer, AutoModelForSeq2SeqLM, pipeline
27
+ import torch
28
+
29
+ device = torch.cuda.current_device() if torch.cuda.is_available() else -1
30
+
31
+ tokenizer = AutoTokenizer.from_pretrained("mideind/nmt-doc-en-is-2022-10",src_lang="en_XX",tgt_lang="is_IS")
32
+
33
+ model = AutoModelForSeq2SeqLM.from_pretrained("mideind/nmt-doc-en-is-2022-10")
34
+ model.config.decoder_start_token_id = 2
35
+
36
+ translate = pipeline("translation_XX_to_YY",model=model,tokenizer=tokenizer,device=device,src_lang="en_XX",tgt_lang="is_IS")
37
+
38
+ target_seq = translate("I am using a translation model to translate text from English to Icelandic.",src_lang="en_XX",tgt_lang="is_IS",max_length=128)
39
+ print(target_seq[0]['translation_text'].strip('YY '))