|
--- |
|
license: mit |
|
language: |
|
- en |
|
library_name: transformers |
|
tags: |
|
- Question Answering |
|
datasets: |
|
- yahoo_answers_qa |
|
metrics: |
|
- rouge |
|
--- |
|
|
|
# Y5 Bot |
|
|
|
Y5 Bot is a text2text transfer transformer model working in English Language. |
|
|
|
It is lightweight, fast, and easy to use for text generation. |
|
|
|
Y5 Bot can answer and generate text short to the main points. |
|
|
|
# How to use |
|
|
|
``` |
|
pip install transformers |
|
``` |
|
|
|
## pipeline approach |
|
|
|
```python |
|
from transformers import pipeline |
|
|
|
generator = pipeline(model="jojo-ai-mst/Y5Bot") |
|
generator("What is Artificial Intelligence?") |
|
``` |
|
|
|
```python |
|
model = T5ForConditionalGeneration.from_pretrained("jojo-ai-mst/Y5Bot") |
|
tokenizer = T5Tokenizer.from_pretrained("jojo-ai-mst/Y5Bot") |
|
|
|
my_question = "What do you think about the benefit of Artificial Intelligence?" |
|
inputs = "Please answer to this question: " + my_question |
|
|
|
inputs = tokenizer(inputs, return_tensors="pt") |
|
outputs = model.generate(**inputs) |
|
answer = tokenizer.decode(outputs[0]) |
|
|
|
print(answer) |
|
``` |
|
|
|
# References |
|
|
|
- [https://arxiv.org/abs/1910.10683](https://arxiv.org/abs/1910.10683) |