---
license: mit
language:
- my
pipeline_tag: text-generation
---

The Simbolo's Myanmar SAR GPT symbol is trained on a dataset of 1 million Burmese data and pre-trained using the GPT-2 architecture. Its purpose is to serve as a foundational pre-trained model for the Burmese language, facilitating fine-tuning for specific applications of different tasks such as creative writing, chatbot, machine translation etc.



### How to use

```python
!pip install transformers

from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Simbolo-Servicio/myanmar-sar-gpt")
model = AutoModelForCausalLM.from_pretrained("Simbolo-Servicio/myanmar-sar-gpt")

input_text = ""
input_ids = tokenizer.encode(input_text, return_tensors='pt')
output = model.generate(input_ids, max_length=100)
print(tokenizer.decode(output[0], skip_special_tokens=True))

```
### Data
The data utilized comprises 1 million sentences sourced from Wikipedia.

### Limitations and bias
We have yet to thoroughly investigate the potential bias inherent in this model. Regarding transparency, it's important to note that the model is primarily trained on data from the Unicode Burmese(Myanmar) language.

### References and Citations
1. Jiang, Shengyi & Huang, Xiuwen & Cai, Xiaonan & Lin, Nankai. (2021). Pre-trained Models and Evaluation Data for the Myanmar Language. 10.1007/978-3-030-92310-5_52. 
2. Lin, N., Fu, Y., Chen, C., Yang, Z., & Jiang, S. (2021). LaoPLM: Pre-trained Language Models for Lao. ArXiv. /abs/2110.05896