semra06

from transformers import Seq2SeqTrainer, Seq2SeqTrainingArguments, AutoModelForSeq2SeqLM, AutoTokenizer
from datasets import load_dataset

model_name = "t5-small"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForSeq2SeqLM.from_pretrained(model_name)

dataset = load_dataset("xsum")  # Örnek: haber özeti
def preprocess(example):
    inputs = tokenizer("summarize: " + example["document"], truncation=True, padding="max_length", max_length=512)