def main()

in trl/scripts/dpo.py [0:0]

• 56 lines of code
• 14 McCabe index (conditional complexity)

def main(script_args, training_args, model_args):
    ################
    # Model & Tokenizer
    ###################
    torch_dtype = (
        model_args.torch_dtype if model_args.torch_dtype in ["auto", None] else getattr(torch, model_args.torch_dtype)
    )
    quantization_config = get_quantization_config(model_args)
    model_kwargs = dict(
        revision=model_args.model_revision,
        attn_implementation=model_args.attn_implementation,
        torch_dtype=torch_dtype,
        use_cache=False if training_args.gradient_checkpointing else True,
        device_map=get_kbit_device_map() if quantization_config is not None else None,
        quantization_config=quantization_config,
    )
    model = AutoModelForCausalLM.from_pretrained(
        model_args.model_name_or_path, trust_remote_code=model_args.trust_remote_code, **model_kwargs
    )
    peft_config = get_peft_config(model_args)
    if peft_config is None:
        ref_model = AutoModelForCausalLM.from_pretrained(
            model_args.model_name_or_path, trust_remote_code=model_args.trust_remote_code, **model_kwargs
        )
    else:
        ref_model = None
    tokenizer = AutoTokenizer.from_pretrained(
        model_args.model_name_or_path, trust_remote_code=model_args.trust_remote_code
    )
    if tokenizer.pad_token is None:
        tokenizer.pad_token = tokenizer.eos_token
    if tokenizer.chat_template is None:
        tokenizer.chat_template = SIMPLE_CHAT_TEMPLATE
    if script_args.ignore_bias_buffers:
        # torch distributed hack
        model._ddp_params_and_buffers_to_ignore = [
            name for name, buffer in model.named_buffers() if buffer.dtype == torch.bool
        ]

    ################
    # Dataset
    ################
    dataset = load_dataset(
        script_args.dataset_name,
        name=script_args.dataset_config,
        streaming=script_args.dataset_streaming,
    )

    ##########
    # Training
    ################
    trainer = DPOTrainer(
        model,
        ref_model,
        args=training_args,
        train_dataset=dataset[script_args.dataset_train_split],
        eval_dataset=dataset[script_args.dataset_test_split] if training_args.eval_strategy != "no" else None,
        processing_class=tokenizer,
        peft_config=peft_config,
    )

    trainer.train()

    if training_args.eval_strategy != "no":
        metrics = trainer.evaluate()
        trainer.log_metrics("eval", metrics)
        trainer.save_metrics("eval", metrics)

    # Save and push to hub
    trainer.save_model(training_args.output_dir)
    if training_args.push_to_hub:
        trainer.push_to_hub(dataset_name=script_args.dataset_name)