Create README.md
Browse files
README.md
ADDED
@@ -0,0 +1,31 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
widget:
|
3 |
+
- text: "ความรัก"
|
4 |
+
---
|
5 |
+
|
6 |
+
# Generate Thai Lyrics
|
7 |
+
|
8 |
+
GPT2 for Thai lyrics generation. We use [GPT2 base Thai](https://huggingface.co/flax-community/gpt2-base-thai) as a pre-trained model
|
9 |
+
for [Siamzom lyrics](https://www.siamzone.com/music/thailyric/)
|
10 |
+
|
11 |
+
|
12 |
+
## Example use
|
13 |
+
|
14 |
+
``` py
|
15 |
+
from transformers import pipeline
|
16 |
+
from transformers import GPT2Model, GPT2TokenizerFast, AutoModelForCausalLM, AutoTokenizer
|
17 |
+
|
18 |
+
model_name = "tupleblog/generate-thai-lyrics"
|
19 |
+
model = AutoModelForCausalLM.from_pretrained(model_name)
|
20 |
+
tokenizer = AutoTokenizer.from_pretrained(model_name)
|
21 |
+
model.config.pad_token_id = model.config.eos_token_id
|
22 |
+
|
23 |
+
|
24 |
+
nlp = pipeline(
|
25 |
+
"text-generation",
|
26 |
+
model=model,
|
27 |
+
tokenizer=tokenizer
|
28 |
+
)
|
29 |
+
text = "ความรัก"
|
30 |
+
nlp(text, max_length=100, top_k=40, temperature=0.8) # varying the temperature and top-k produce different output
|
31 |
+
```
|