from datasets import load_dataset
from transformers import pipeline

# 1. Mee dataset ni load cheyyi (mee dataset path ekkada create chesavo ade pettu)
dataset = load_dataset("file:///C:/Users/LENOVO/Downloads/my-data-txt.pdf")

# 2. Example: dataset lo oka sample chupinchu
print(dataset['train'][0])  # train split lo first record chupistundi

# 3. Hugging Face model load cheyyi (example: text-generation / Q&A)
generator = pipeline("text-generation", model="gpt2")

# 4. Mee dataset nundi oka data ni teesukoni prompt laga vadatam
sample_text = dataset['train'][0]["text"]   # column name "text" unte
response = generator(f"Mee data: {sample_text}\nQuestion: explain this")

print(response[0]["generated_text"])