LLaMA-Efficient-Tuning参数

毛小聪

已于 2023-12-04 00:14:02 修改

阅读量568

点赞数 9

文章标签： llama 人工智能深度学习

于 2023-12-03 23:58:38 首次发布

本文链接：https://blog.csdn.net/weixin_45448034/article/details/134773144

版权

背景

这个文章单纯记录参数，不分析

参数

llmtuner.tuner.core.parser.parse_train_args

ModelArguments

@dataclass
class ModelArguments:
    r"""
    Arguments pertaining to which model/config/tokenizer we are going to fine-tune.
    """
    model_name_or_path: str = field(
        metadata={"help": "Path to pretrained model or model identifier from huggingface.co/models."}
    )
    cache_dir: Optional[str] = field(
        default=None,
        metadata={"help": "Where to store the pretrained models downloaded from huggingface.co."}
    )
    use_fast_tokenizer: Optional[bool] = field(
        default=True,
        metadata={"help": "Whether to use one of the fast tokenizer (backed by the tokenizers library) or not."}
    )
    use_auth_token: Optional[bool] = field(
        default=False,
        metadata={"help": "Will use the token generated when running `huggingface-cli login`."}
    )
    model_revision: Optional[str] = field(
        default="main",
        metadata={"help": "The specific model version to use (can be a branch name, tag name or commit id)."}
    )
    quantization_bit: Optional[int] = field(
        default=None,
        metadata={"help": "The number of bits to quantize the model."}
    )
    quantization_type: Optional[Literal["fp4", "nf4"]] = field(
        default="nf4",
        metadata={"help": "Quantization data type to use in int4 training."}
    )
    double_quantization: Optional[bool] = field(
        default=True,
        metadata={"help": "Whether to use double quantization in int4 training or not."}
    )
    rope_scaling: Optional[Literal["linear", "dynamic"]] = field(
        default=None,
        metadata={"help": "Adopt scaled rotary positional embeddings."}
    )
    flash_attn: Optional[bool] = field(
        default=False,
        metadata={"help": "Enable flash attention for faster training."}
    )
    checkpoint_dir: Optional[str] = field(
        default=None,
        metadata={"help": "Path to the directory(s) containing the delta model checkpoints as well as the configurations."}
    )
    reward_model: Optional[str] = field(
        default=None,
        metadata={"help": "Path to the directory containing the checkpoints of the reward model."}
    )
    plot_loss: Optional[bool] = field(
        default=False,
        metadata={"help": "Whether to plot the training loss after fine-tuning or not."}
    )
    hf_auth_token: Optional[str] = field(
        default=None,
        metadata={"help": "Auth token to log in with Hugging Face Hub."}
    )
    compute_dtype: Optional[torch.dtype] = field(
        default=None,
        metadata={"help": "Used in quantization configs. Do not specify this argument manually."}
    )
    model_max_length: Optional[int] = field(
        default=None,
        metadata={"help": "Used in rope scaling. Do not specify this argument manually."}
    )

DataArguments

@dataclass
class DataArguments:
    r"""
    Arguments pertaining to what data we are going to input our model for training and evaluation.
    """
    template: Optional[str] = field(
        default=None,
        metadata={"help": "Which template to use for constructing prompts in training and inference."}
    )
    dataset: Optional[str] = field(
        default="alpaca_en",
        metadata={"help": "The name of provided dataset(s) to use. Use commas to separate multiple datasets."}
    )
    dataset_dir: Optional[str] = field(
        default="data",
        metadata={"help": "The name of the folder containing datasets."}
    )
    split: Optional[str] = field(
        default="train",
        metadata={"help": "Which dataset split to use for training and evaluation."}
    )
    streaming: Optional[bool] = field(
        default=False,
        metadata={"help": "Enable streaming mode."}
    )
    buffer_size: Optional[int] = field(
        default=1024,
        metadata={"help": "Size of the buffer to randomly sample examples from in streaming mode."}
    )
    mix_strategy: Optional[Literal["concat", "interleave_under", "interleave_over"]] = field(
        default="concat",
        metadata={"help": "Strategy to use in dataset mixing."}
    )
    interleave_probs: Optional[str] = field(
        default=None,
        metadata={"help": "Probabilities to sample data from datasets. Use commas to separate multiple datasets."}
    )
    overwrite_cache: Optional[bool] = field(
        default=False,
        metadata={"help": "Overwrite the cached training and evaluation sets."}
    )
    preprocessing_num_workers: Optional[int] = field(
        default=None,
        metadata={"help": "The number of processes to use for the preprocessing."}
    )
    max_source_length: Optional[int] = field(
        default=512,
        metadata={"help": "The maximum total input sequence length after tokenization."}
    )
    max_target_length: Optional[int] = field(
        default=512,
        metadata={"help": "The maximum total output sequence length after tokenization."}
    )
    max_samples: Optional[int] = field(
        default=None,
        metadata={"help": "For debugging purposes, truncate the number of examples for each dataset."}
    )
    eval_num_beams: Optional[int] = field(
        default=None,
        metadata={"help": "Number of beams to use for evaluation. This argument will be passed to `model.generate`"}
    )
    ignore_pad_token_for_loss: Optional[bool] = field(
        default=True,
        metadata={"help": "Whether to ignore the tokens corresponding to padded labels in the loss computation or not."}
    )
    system_prompt: Optional[str] = field(
        default=None,
        metadata={"help": "System prompt to add before the user query. Use `|` to separate multiple prompts in training."}
    )
    val_size: Optional[float] = field(
        default=0,
        metadata={"help": "Size of the development set, should be an integer or a float in range `[0,1)`."}
    )

Seq2SeqTrainingArguments

@dataclass
@add_start_docstrings(TrainingArguments.__doc__)
class Seq2SeqTrainingArguments(TrainingArguments):

TrainingArguments

class TrainingArguments:
    """
    TrainingArguments is the subset of the arguments we use in our example scripts **which relate to the training loop
    itself**.

    Using [`HfArgumentParser`] we can turn this class into
    [argparse](https://docs.python.org/3/library/argparse#module-argparse) arguments that can be specified on the
    command line.

    Parameters:
        output_dir (`str`):
            The output directory where the model predictions and checkpoints will be written.
        overwrite_output_dir (`bool`, *optional*, defaults to `False`):
            If `True`, overwrite the content of the output directory. Use this to continue training if `output_dir`
            points to a checkpoint directory.
        do_train (`bool`, *optional*, defaults to `False`):
            Whether to run training or not. This argument is not directly used by [`Trainer`], it's intended to be used
            by your training/evaluation scripts instead. See the [example
            scripts](https://github.com/huggingface/transformers/tree/main/examples) for more details.
        do_eval (`bool`, *optional*):
            Whether to run evaluation on the validation set or not. Will be set to `True` if `evaluation_strategy` is
            different from `"no"`. This argument is not directly used by [`Trainer`], it's intended to be used by your
            training/evaluation scripts instead. See the [example
            scripts](https://github.com/huggingface/transformers/tree/main/examples) for more details.
        do_predict (`bool`, *optional*, defaults to `False`):
            Whether to run predictions on the test set or not. This argument is not directly used by [`Trainer`], it's
            intended to be used by your training/evaluation scripts instead. See the [example
            scripts](https://github.com/huggingface/transformers/tree/main/examples) for more details.
        evaluation_strategy (`str` or [`~trainer_utils.IntervalStrategy`], *optional*, defaults to `"no"`):
            The evaluation strategy to adopt during training. Possible values are:

                - `"no"`: No evaluation is done during training.
                - `"steps"`: Evaluation is done (and logged) every `eval_steps`.
                - `"epoch"`: Evaluation is done at the end of each epoch.

        prediction_loss_only (`bool`, *optional*, defaults to `False`):
            When performing evaluation and generating predictions, only returns the loss.
        per_device_train_batch_size (`int`, *optional*, defaults to 8):
            The batch size per GPU/XPU/TPU/MPS/NPU core/CPU for training.
        per_device_eval_batch_size (`int`, *optional*, defaults to 8):
            The batch size per GPU/XPU/TPU/MPS/NPU core/CPU for evaluation.
        gradient_accumulation_steps (`int`, *optional*, defaults to 1):
            Number of updates steps to accumulate the gradients for, before performing a backward/update pass.

            <Tip warning={true}>

            When using gradient accumulation, one step is counted as one step with backward pass. Therefore, logging,
            evaluation, save will be conducted every `gradient_accumulation_steps * xxx_step` training examples.

            </Tip>

        eval_accumulation_steps (`int`, *optional*):
            Number of predictions steps to accumulate the output tensors for, before moving the results to the CPU. If
            left unset, the whole predictions are accumulated on GPU/NPU/TPU before being moved to the CPU (faster but
            requires more memory).
        eval_delay (`float`, *optional*):
            Number of epochs or steps to wait for before the first evaluation can be performed, depending on the
            evaluation_strategy.
        learning_rate (`float`, *optional*, defaults to 5e-5):
            The initial learning rate for [`AdamW`] optimizer.
        weight_decay (`float`, *optional*, defaults to 0):
            The weight decay to apply (if not zero) to all layers except all bias and LayerNorm weights in [`AdamW`]
            optimizer.
        adam_beta1 (`float`, *optional*, defaults to 0.9):
            The beta1 hyperparameter for the [`AdamW`] optimizer.
        adam_beta2 (`float`, *optional*, defaults to 0.999):
            The beta2 hyperparameter for the [`AdamW`] optimizer.
        adam_epsilon (`float`, *optional*, defaults to 1e-8):
            The epsilon hyperparameter for the [`AdamW`] optimizer.
        max_grad_norm (`float`, *optional*, defaults to 1.0):
            Maximum gradient norm (for gradient clipping).
        num_train_epochs(`float`, *optional*, defaults to 3.0):
            Total number of training epochs to perform (if not an integer, will perform the decimal part percents of
            the last epoch before stopping training).
        max_steps (`int`, *optional*, defaults to -1):
            If set to a positive number, the total number of training steps to perform. Overrides `num_train_epochs`.
            In case of using a finite iterable dataset the training may stop before reaching the set number of steps
            when all data is exhausted
        lr_scheduler_type (`str` or [`SchedulerType`], *optional*, defaults to `"linear"`):
            The scheduler type to use. See the documentation of [`SchedulerType`] for all possible values.

FinetuningArguments

@dataclass
class FinetuningArguments:
    r"""
    Arguments pertaining to which techniques we are going to fine-tuning with.
    """
    finetuning_type: Optional[Literal["lora", "freeze", "full", "none"]] = field(
        default="lora",
        metadata={"help": "Which fine-tuning method to use."}
    )
    num_hidden_layers: Optional[int] = field(
        default=32,
        metadata={"help": "Number of decoder blocks in the model for partial-parameter (freeze) fine-tuning. \
                  LLaMA choices: [\"32\", \"40\", \"60\", \"80\"], \
                  LLaMA-2 choices: [\"32\", \"40\", \"80\"], \
                  BLOOM choices: [\"24\", \"30\", \"70\"], \
                  Falcon choices: [\"32\", \"60\"], \
                  Baichuan choices: [\"32\", \"40\"] \
                  Qwen choices: [\"32\"], \
                  XVERSE choices: [\"40\"]"}
    )
    num_layer_trainable: Optional[int] = field(
        default=3,
        metadata={"help": "Number of trainable layers for partial-parameter (freeze) fine-tuning."}
    )
    name_module_trainable: Optional[Literal["mlp", "self_attn", "self_attention"]] = field(
        default="mlp",
        metadata={"help": "Name of trainable modules for partial-parameter (freeze) fine-tuning. \
                  LLaMA choices: [\"mlp\", \"self_attn\"], \
                  BLOOM & Falcon choices: [\"mlp\", \"self_attention\"], \
                  Baichuan choices: [\"mlp\", \"self_attn\"], \
                  Qwen choices: [\"mlp\", \"attn\"], \
                  LLaMA-2, InternLM, XVERSE choices: the same as LLaMA."}
    )
    lora_rank: Optional[int] = field(
        default=8,
        metadata={"help": "The intrinsic dimension for LoRA fine-tuning."}
    )
    lora_alpha: Optional[float] = field(
        default=32.0,
        metadata={"help": "The scale factor for LoRA fine-tuning (similar with the learning rate)."}
    )
    lora_dropout: Optional[float] = field(
        default=0.1,
        metadata={"help": "Dropout rate for the LoRA fine-tuning."}
    )
    lora_target: Optional[str] = field(
        default=None,
        metadata={"help": "Name(s) of target modules to apply LoRA. Use commas to separate multiple modules. \
                  LLaMA choices: [\"q_proj\", \"k_proj\", \"v_proj\", \"o_proj\", \"gate_proj\", \"up_proj\", \"down_proj\"], \
                  BLOOM & Falcon choices: [\"query_key_value\", \"self_attention.dense\", \"mlp.dense\"], \
                  Baichuan choices: [\"W_pack\", \"o_proj\", \"gate_proj\", \"up_proj\", \"down_proj\"], \
                  Qwen choices: [\"c_attn\", \"attn.c_proj\", \"w1\", \"w2\", \"mlp.c_proj\"], \
                  LLaMA-2, InternLM, XVERSE choices: the same as LLaMA."}
    )
    resume_lora_training: Optional[bool] = field(
        default=True,
        metadata={"help": "Whether to resume training from the last LoRA weights or create new weights after merging them."}
    )
    ppo_score_norm: Optional[bool] = field(
        default=False,
        metadata={"help": "Use score normalization in PPO Training."}
    )
    dpo_beta: Optional[float] = field(
        default=0.1,
        metadata={"help": "The beta parameter for the DPO loss."}
    )

GeneratingArguments

@dataclass
class GeneratingArguments:
    r"""
    Arguments pertaining to specify the decoding parameters.
    """
    do_sample: Optional[bool] = field(
        default=True,
# 
        metadata={"help": "Whether or not to use sampling, use greedy decoding otherwise."}
    )
    temperature: Optional[float] = field(
        default=0.95,
# 用于调整下一个令牌概率的值。
        metadata={"help": "The value used to modulate the next token probabilities."}
    )
    top_p: Optional[float] = field(
        default=0.7,
#保留概率加起来为top_p或更高的最可能令牌的最小集合。
        metadata={"help": "The smallest set of most probable tokens with probabilities that add up to top_p or higher are kept."}
    )
#为前k个过滤保留的最高概率词汇表标记的数量。
    top_k: Optional[int] = field(
        default=50,
        metadata={"help": "The number of highest probability vocabulary tokens to keep for top-k filtering."}
    )
    num_beams: Optional[int] = field(
        default=1,
        metadata={"help": "Number of beams for beam search. 1 means no beam search."}
    )
    max_length: Optional[int] = field(
        default=None,
        metadata={"help": "The maximum length the generated tokens can have. It can be overridden by max_new_tokens."}
    )
    max_new_tokens: Optional[int] = field(
        default=512,
        metadata={"help": "The maximum numbers of tokens to generate, ignoring the number of tokens in the prompt."}
    )
    repetition_penalty: Optional[float] = field(
        default=1.0,
#重复惩罚的参数。1.0表示不罚。
        metadata={"help": "The parameter for repetition penalty. 1.0 means no penalty."}
    )
#长度惩罚的参数。1.0表示不罚。
    length_penalty: Optional[float] = field(
        default=1.0,
        metadata={"help": "Exponential penalty to the length that is used with beam-based generation."}
    )

GeneralArguments

@dataclass
class GeneralArguments:
    r"""
    Arguments pertaining to which stage we are going to perform.
    """
    stage: Optional[Literal["pt", "sft", "rm", "ppo", "dpo"]] = field(
        default="sft",
        metadata={"help": "Which stage will be performed in training."}
    )