Transformers
Safetensors
michaelginn commited on
Commit
3b35b58
·
verified ·
1 Parent(s): ee91bc5

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +33 -2
README.md CHANGED
@@ -11,6 +11,37 @@ base_model:
11
  library_name: transformers
12
  ---
13
 
14
- Repo: https://github.com/foltaProject/glosslm
 
15
 
16
- Paper: https://arxiv.org/abs/2403.06399
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
11
  library_name: transformers
12
  ---
13
 
14
+ - Repo: https://github.com/foltaProject/glosslm
15
+ - Paper: https://arxiv.org/abs/2403.06399
16
 
17
+ Usage:
18
+ ```python
19
+ import transformers
20
+
21
+ # Your inputs
22
+ transcription = "o sey xtok rixoqiil"
23
+ translation = "O sea busca esposa."
24
+ lang = "Uspanteco"
25
+ metalang = "Spanish"
26
+ is_segmented = False
27
+
28
+ prompt = f"""Provide the glosses for the following transcription in {lang}.
29
+
30
+ Transcription in {lang}: {transcription}
31
+ Transcription segmented: {is_segmented}
32
+ Translation in {metalang}: {translation}\n
33
+ Glosses:
34
+ """
35
+
36
+ model = transformers.T5ForConditionalGeneration.from_pretrained("lecslab/glosslm")
37
+ tokenizer = transformers.ByT5Tokenizer.from_pretrained(
38
+ "google/byt5-base", use_fast=False
39
+ )
40
+
41
+ inputs = tokenizer(prompt, return_tensors="pt")
42
+ outputs = tokenizer.batch_decode(
43
+ model.generate(**inputs, max_length=1024), skip_special_tokens=True
44
+ )
45
+ print(outputs[0])
46
+ # o sea COM-buscar E3S-esposa
47
+ ```