Update README.md

README.md

@@ -14,25 +14,41 @@ pipeline_tag: translation
 library_name: transformers
 ---
 # Lao - English Transliteration
+Welcome to the forefront of linguistic innovation with our groundbreaking T5 model designed specifically for Lao to English translation. In a rapidly globalizing world where effective communication is paramount, our T5 model stands as a beacon of excellence, offering unparalleled accuracy, fluency, and efficiency in bridging the language gap between Lao and English.
 
+Built on state-of-the-art deep learning architecture and trained on vast datasets of Lao and English texts, our T5 model harnesses the power of transformer-based technology to deliver seamless and precise translations. Whether you're a business expanding into Laotian markets, a researcher seeking to access Lao-language resources, or an individual connecting with Lao-speaking communities, our T5 model is your ultimate solution for unlocking linguistic barriers and fostering meaningful cross-cultural exchanges.
 
-## How to use
-
-
+With a commitment to quality and innovation, our T5 model not only translates words but also preserves context, tone, and cultural nuances, ensuring that the essence of the original message remains intact in every translated sentence. Whether it's documents, websites, or multimedia content, our T5 model offers unmatched versatility and reliability, empowering users to communicate effortlessly across languages and borders.
 
+## How to use
+### On GPU
 ```python
 from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
 tokenizer = AutoTokenizer.from_pretrained("minhtoan/t5-translation-lao-english")  
 model = AutoModelForSeq2SeqLM.from_pretrained("minhtoan/t5-translation-lao-english")
 model.cuda()
-src = "ຂ້ອຍຢາກຊື້ປຶ້ມ"
+src = "ຂ້ອຍ​ຮັກ​ເຈົ້າ"
 tokenized_text = tokenizer.encode(src, return_tensors="pt").cuda()
 model.eval()
 translate_ids = model.generate(tokenized_text, max_length=140)
 output = tokenizer.decode(translate_ids[0], skip_special_tokens=True)
 output
 ```
-'I want to buy book'
+'I love you'
+
+### On CPU
+```python
+from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
+tokenizer = AutoTokenizer.from_pretrained("minhtoan/t5-translation-lao-english")  
+model = AutoModelForSeq2SeqLM.from_pretrained("minhtoan/t5-translation-lao-english")
+src = "ຂ້ອຍ​ຮັກ​ເຈົ້າ"
+input_ids = tokenizer(src, max_length=200, return_tensors="pt", padding="max_length", truncation=True).input_ids
+outputs = model.generate(input_ids=input_ids, max_new_tokens=140)
+output = tokenizer.batch_decode(outputs, skip_special_tokens=True)[0]
+output
+```
+'I love you'
+
 
 
 ## Author