jojo-ai-mst commited on
Commit
350ca93
·
1 Parent(s): 9919514

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +26 -0
README.md CHANGED
@@ -34,6 +34,32 @@ You can use the [MyanmarGPT model 125 M](https://huggingface.co/jojo-ai-mst/Myan
34
  - **License:** CreativeML-OpenRail-M
35
  - **Finetuned from model [optional]:** [mGPT]
36
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
37
  ## Applied Uses
38
 
39
  MyanmarGPT-Big can be used for the following use cases.
 
34
  - **License:** CreativeML-OpenRail-M
35
  - **Finetuned from model [optional]:** [mGPT]
36
 
37
+ ## How to use
38
+
39
+ ### Using pipeline
40
+
41
+ ```python
42
+ from transformers import pipeline
43
+
44
+ pipe = pipeline("text-generation", model="jojo-ai-mst/MyanmarGPT-Big")
45
+ outputs = pipe("အီတလီ",do_sample=False)
46
+ print(outputs)
47
+ ```
48
+
49
+ ### Using Model Generator
50
+ ```python
51
+ import torch
52
+ from transformers import AutoTokenizer, AutoModelForCausalLM
53
+
54
+ tokenizer = AutoTokenizer.from_pretrained("jojo-ai-mst/MyanmarGPT-Big")
55
+ model = AutoModelForCausalLM.from_pretrained("jojo-ai-mst/MyanmarGPT-Big")
56
+
57
+
58
+ input_ids = tokenizer.encode("ချစ်သား", return_tensors='pt')
59
+ output = model.generate(input_ids, max_length=50)
60
+ print(tokenizer.decode(output[0], skip_special_tokens=True))
61
+ ```
62
+
63
  ## Applied Uses
64
 
65
  MyanmarGPT-Big can be used for the following use cases.