from datasets import load_dataset
from transformers import AutoTokenizer, AutoModelForCausalLM, Trainer, TrainingArguments

# Load dataset
dataset = load_dataset('json', data_files='flirty_dataset.json')

# Tokenizer and model
model_name = "gpt2"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(model_name)

# Tokenize dataset
def tokenize_function(examples):
    return tokenizer(examples['prompt'], truncation=True, padding="max_length", max_length=128)

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

# Training arguments
training_args = TrainingArguments(
    output_dir="./fine_tuned_gpt2",
    evaluation_strategy="epoch",
    save_strategy="epoch",
    learning_rate=5e-5,
    num_train_epochs=3,
    per_device_train_batch_size=8,
    save_total_limit=2,
    logging_dir="./logs",
    logging_steps=10,
    fp16=True
)

# Trainer
trainer = Trainer(
    model=model,
    args=training_args,
    train_dataset=tokenized_dataset["train"],
    eval_dataset=tokenized_dataset["validation"],
    tokenizer=tokenizer
)

# Train the model
trainer.train()

# Save model
trainer.save_model("./fine_tuned_gpt2")
tokenizer.save_pretrained("./fine_tuned_gpt2")