Llama-Factory 简介二，脚本详解 LLaMA-Factory/src/train_bash.py LLaMA-Factory/src/llmtuner/train/pt/workfl

examples/
├── lora_single_gpu/
│ ├── pretrain.sh: 基于 LoRA 进行增量预训练
│ ├── sft.sh: 基于 LoRA 进行指令监督微调
│ ├── reward.sh: 基于 LoRA 进行奖励模型训练
│ ├── ppo.sh: 基于 LoRA 进行 PPO 训练
│ ├── dpo.sh: 基于 LoRA 进行 DPO 训练
│ ├── orpo.sh: 基于 LoRA 进行 ORPO 训练
│ ├── prepare.sh: 保存预处理后的数据集
│ └── predict.sh: 基于 LoRA 进行批量预测并计算 BLEU 和 ROUGE 分数
├── qlora_single_gpu/
│ ├── bitsandbytes.sh: 基于 QLoRA 微调 4/8 比特 BNB 模型
│ ├── gptq.sh: 基于 QLoRA 微调 4/8 比特 GPTQ 模型
│ ├── awq.sh: 基于 QLoRA 微调 4 比特 AWQ 模型
│ └── aqlm.sh: 基于 QLoRA 微调 2 比特 AQLM 模型
├── lora_multi_gpu/
│ ├── single_node.sh: 使用 Accelerate 进行单节点 LoRA 训练
│ └── multi_node.sh: 使用 Accelerate 进行多节点 LoRA 训练
├── full_multi_gpu/
│ ├── single_node.sh: 使用 DeepSpeed 进行单节点全量训练
│ ├── multi_node.sh: 使用 DeepSpeed 进行多节点全量训练
│ └── predict.sh: 基于全量训练进行批量预测并计算 BLEU 和 ROUGE 分数
├── merge_lora/
│ ├── merge.sh: 将 LoRA 权重合并到预训练模型中
│ └── quantize.sh: 使用 AutoGPTQ 量化微调后的模型
├── inference/
│ ├── cli_demo.sh: 启动 LoRA 模型的命令行推理接口
│ ├── api_demo.sh: 启动 LoRA 模型的 OpenAI 风格 API
│ ├── web_demo.sh: 启动 LoRA 模型的浏览器推理接口
│ └── evaluate.sh: 在 MMLU/CMMLU/C-Eval 数据集上评测 LoRA 模型
└── extras/
├── galore/
│ └── sft.sh: 使用 GaLore 训练模型
├── badam/
│ └── sft.sh: 使用 BAdam 训练模型
├── loraplus/
│ └── sft.sh: 使用 LoRA+ 训练模型
├── llama_pro/
│ ├── expand.sh: 扩展模型中的层
│ └── sft.sh: 训练扩展后的模型
└── fsdp_qlora/
└── sft.sh: 使用 FSDP+QLoRA 微调量化模型

├── lora_single_gpu/
│ ├── pretrain.sh: 基于 LoRA 进行增量预训练

#!/bin/bash CUDA_VISIBLE_DEVICES=0 python ../../src/train_bash.py \ --stage pt \ --do_train \ --model_name_or_path meta-llama/Llama-2-7b-hf \ --dataset c4_demo \ --dataset_dir ../../data \ --finetuning_type lora \ --lora_target q_proj,v_proj \ --output_dir ../../saves/LLaMA2-7B/lora/pretrain \ --overwrite_cache \ --overwrite_output_dir \ --cutoff_len 1024 \ --preprocessing_num_workers 16 \ --per_device_train_batch_size 1 \ --per_device_eval_batch_size 1 \ --gradient_accumulation_steps 8 \ --lr_scheduler_type cosine \ --logging_steps 10 \ --warmup_steps 20 \ --save_steps 100 \ --eval_steps 100 \ --evaluation_strategy steps \ --load_best_model_at_end \ --learning_rate 5e-5 \ --num_train_epochs 3.0 \ --max_samples 10000 \ --val_size 0.1 \ --plot_loss \ --fp16

LLaMA-Factory/src/train_bash.py

from llmtuner import run_exp def main(): run_exp() def _mp_fn(index): # For xla_spawn (TPUs) main() if __name__ == "__main__": main()

def run_exp() -> LLaMA-Factory/src/llmtuner/train/tuner.py

def run_exp(args: Optional[Dict[str, Any]] = None, callbacks: Optional[List["TrainerCallback"]] = None): model_args, data_args, training_args, finetuning_args, generating_args = get_train_args(args) callbacks = [LogCallback()] if callbacks is None else callbacks if finetuning_args.stage == "pt": run_pt(model_args, data_args, training_args, finetuning_args, callbacks) elif finetuning_args.stage == "sft": run_sft(model_args, data_args, training_args, finetuning_args, generating_args, callbacks) elif finetuning_args.stage == "rm": run_rm(model_args, data_args, training_args, finetuning_args, callbacks) elif finetuning_args.stage == "ppo": run_ppo(model_args, data_args, training_args, finetuning_args, generating_args, callbacks) elif finetuning_args.stage == "dpo": run_dpo(model_args, data_args, training_args, finetuning_args, callbacks) elif finetuning_args.stage == "orpo": run_orpo(model_args, data_args, training_args, finetuning_args, callbacks) else: raise ValueError("Unknown task.") def export_model(args: Optional[Dict[str, Any]] = None): model_args, data_args, finetuning_args, _ = get_infer_args(args) if model_args.export_dir is None: raise ValueError("Please specify `export_dir` to save model.") if model_args.adapter_name_or_path is not None and model_args.export_quantization_bit is not None: raise ValueError("Please merge adapters before quantizing the model.") tokenizer = load_tokenizer(model_args) get_template_and_fix_tokenizer(tokenizer, data_args.template) model = load_model(tokenizer, model_args, finetuning_args) # must after fixing tokenizer to resize vocab if getattr(model, "quantization_method", None) and model_args.adapter_name_or_path is not None: raise ValueError("Cannot merge adapters to a quantized model.") if not isinstance(model, PreTrainedModel): raise ValueError("The model is not a `PreTrainedModel`, export aborted.") if getattr(model, "quantization_method", None) is None: # cannot convert dtype of a quantized model output_dtype = getattr(model.config, "torch_dtype", torch.float16) setattr(model.config, "torch_dtype", output_dtype) model = model.to(output_dtype) model.save_pretrained( save_directory=model_args.export_dir, max_shard_size="{}GB".format(model_args.export_size), safe_serialization=(not model_args.export_legacy_format), ) if model_args.export_hub_model_id is not None: model.push_to_hub( model_args.export_hub_model_id, token=model_args.hf_hub_token, max_shard_size="{}GB".format(model_args.export_size), safe_serialization=(not model_args.export_legacy_format), ) try: tokenizer.padding_side = "left" # restore padding side tokenizer.init_kwargs["padding_side"] = "left" tokenizer.save_pretrained(model_args.export_dir) if model_args.export_hub_model_id is not None: tokenizer.push_to_hub(model_args.export_hub_model_id, token=model_args.hf_hub_token) except Exception: logger.warning("Cannot save tokenizer, please copy the files manually.") if __name__ == "__main__": run_exp()

from typing import TYPE_CHECKING, Any, Dict, List, Optional

import torch
from transformers import PreTrainedModel

from ..data import get_template_and_fix_tokenizer
from ..extras.callbacks import LogCallback
from ..extras.logging import get_logger
from ..hparams import get_infer_args, get_train_args
from ..model import load_model, load_tokenizer
from .dpo import run_dpo
from .orpo import run_orpo
from .ppo import run_ppo
from .pt import run_pt
from .rm import run_rm
from .sft import run_sft

if TYPE_CHECKING:
from transformers import TrainerCallback

logger = get_logger(__name__)

def run_pt() -> LLaMA-Factory/src/llmtuner/train/pt/workflow.py

def run_pt( model_args: "ModelArguments", data_args: "DataArguments", training_args: "Seq2SeqTrainingArguments", finetuning_args: "FinetuningArguments", callbacks: Optional[List["TrainerCallback"]] = None, ): tokenizer = load_tokenizer(model_args) dataset = get_dataset(tokenizer, model_args, data_args, training_args, stage="pt") model = load_model(tokenizer, model_args, finetuning_args, training_args.do_train) data_collator = DataCollatorForLanguageModeling(tokenizer=tokenizer, mlm=False) # Initialize our Trainer trainer = CustomTrainer( model=model, args=training_args, finetuning_args=finetuning_args, tokenizer=tokenizer, data_collator=data_collator, callbacks=callbacks, **split_dataset(dataset, data_args, training_args), ) # Training if training_args.do_train: train_result = trainer.train(resume_from_checkpoint=training_args.resume_from_checkpoint) trainer.save_model() trainer.log_metrics("train", train_result.metrics) trainer.save_metrics("train", train_result.metrics) trainer.save_state() if trainer.is_world_process_zero() and finetuning_args.plot_loss: plot_loss(training_args.output_dir, keys=["loss", "eval_loss"]) # Evaluation if training_args.do_eval: metrics = trainer.evaluate(metric_key_prefix="eval") try: perplexity = math.exp(metrics["eval_loss"]) except OverflowError: perplexity = float("inf") metrics["perplexity"] = perplexity trainer.log_metrics("eval", metrics) trainer.save_metrics("eval", metrics) # Create model card create_modelcard_and_push(trainer, model_args, data_args, training_args, finetuning_args)

# Inspired by: https://github.com/huggingface/transformers/blob/v4.34.1/examples/pytorch/language-modeling/run_clm.py import math from typing import TYPE_CHECKING, List, Optional from transformers import DataCollatorForLanguageModeling from ...data import get_dataset, split_dataset from ...extras.ploting import plot_loss from ...model import load_model, load_tokenizer from ..utils import create_modelcard_and_push from .trainer import CustomTrainer if TYPE_CHECKING: from transformers import Seq2SeqTrainingArguments, TrainerCallback from ...hparams import DataArguments, FinetuningArguments, ModelArguments