|
|
--- |
|
|
license: apache-2.0 |
|
|
tags: |
|
|
- hindi |
|
|
- chatbot |
|
|
- question-answering |
|
|
- prakriti |
|
|
- flan-t5 |
|
|
--- |
|
|
|
|
|
# Prakriti AI — A Hindi Question-Answering Model |
|
|
|
|
|
**Prakriti-AI** is a custom-trained language model based on `flan-t5-small`, fine-tuned on over 10,000 Hindi question-answer pairs. |
|
|
It is designed to understand and respond to Hindi queries in a natural, human-like way. |
|
|
|
|
|
--- |
|
|
|
|
|
## ✨ Features |
|
|
|
|
|
- Base Model: `google/flan-t5-small` |
|
|
- Language: Hindi (and Hinglish supported) |
|
|
- Purpose: Q&A chatbot assistant |
|
|
- Trained by: **Asankhla01** |
|
|
- Use-case: Chatbots, education, AI assistants |
|
|
|
|
|
--- |
|
|
|
|
|
## 🔧 Example Usage |
|
|
|
|
|
```python |
|
|
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM |
|
|
|
|
|
model = AutoModelForSeq2SeqLM.from_pretrained("Asankhla01/prakriti-ai-model") |
|
|
tokenizer = AutoTokenizer.from_pretrained("Asankhla01/prakriti-ai-model") |
|
|
|
|
|
def ask_prakriti(question): |
|
|
input_text = "Q: " + question |
|
|
inputs = tokenizer(input_text, return_tensors="pt") |
|
|
outputs = model.generate(**inputs) |
|
|
answer = tokenizer.decode(outputs[0], skip_special_tokens=True) |
|
|
return answer |
|
|
|
|
|
ask_prakriti("AI kya hota hai?") |
|
|
|
|
|
--- |
|
|
library_name: transformers |
|
|
tags: [] |
|
|
--- |
|
|
|
