app.py

from datasets import load_dataset
from transformers import GPT2LMHeadModel, GPT2Tokenizer

tokenizer = GPT2Tokenizer.from_pretrained('gpt2')
model = GPT2LMHeadModel.from_pretrained('gpt2')

examples = load_dataset("Andyrasika/Ecommerce_FAQ")

def tokenize_function(examples):
    # Concatenate the question and answer into a single text field
    text_pairs = [f"Q: {q} A: {a}" for q, a in zip(examples['question'], examples['answer'])]
    return tokenizer(text_pairs, truncation=True, padding=True)


#def tokenize_function(examples):
 #   return tokenizer(examples['text'], truncation=True, padding=True)

tokenized_dataset = dataset.map(tokenize_function, batched=True)

from transformers import Trainer, TrainingArguments

training_args = TrainingArguments(
    output_dir='./results',          # Output directory
    num_train_epochs=3,              # Number of epochs
    per_device_train_batch_size=4,   # Batch size per device
    save_steps=10_000,               # Save checkpoint every 10,000 steps
    save_total_limit=2,              # Only last 2 checkpoints are kept
)


trainer = Trainer(
    model=model,                          # The pre-trained GPT-2 model
    args=training_args,                   # Training arguments
    train_dataset=tokenized_dataset,      # The dataset you prepared
)


trainer.train()


model.save_pretrained('./faqmodel')
tokenizer.save_pretrained('./faqmodel')