diff --git a/finetune/configs/ds_config_zero2.json b/finetune/configs/ds_config_zero2.json new file mode 100644 index 0000000..b6fb07f --- /dev/null +++ b/finetune/configs/ds_config_zero2.json @@ -0,0 +1,32 @@ +{ + "train_batch_size": "auto", + "train_micro_batch_size_per_gpu": "auto", + "gradient_accumulation_steps": "auto", + "gradient_clipping": 1.0, + "fp16": { + "enabled": "auto", + "loss_scale": 0, + "loss_scale_window": 1000, + "initial_scale_power": 16, + "hysteresis": 2, + "min_loss_scale": 1 + }, + "bf16": { + "enabled": "auto" + }, + "zero_optimization": { + "stage": 2, + "overlap_comm": true, + "reduce_scatter": true, + "contiguous_gradients": true + }, + "wall_clock_breakdown": false, + "flops_profiler": { + "enabled": false, + "profile_step": 1, + "module_depth": -1, + "top_modules": 1, + "detailed": true, + "output_file": null + } +} \ No newline at end of file diff --git a/finetune/configs/ds_config_zero2_offload.json b/finetune/configs/ds_config_zero2_offload.json new file mode 100644 index 0000000..8dd6fc5 --- /dev/null +++ b/finetune/configs/ds_config_zero2_offload.json @@ -0,0 +1,38 @@ +{ + "train_batch_size": "auto", + "train_micro_batch_size_per_gpu": "auto", + "gradient_accumulation_steps": "auto", + "gradient_clipping": 1.0, + "fp16": { + "enabled": "auto", + "loss_scale": 0, + "loss_scale_window": 1000, + "initial_scale_power": 16, + "hysteresis": 2, + "min_loss_scale": 1 + }, + "bf16": { + "enabled": "auto" + }, + "zero_optimization": { + "stage": 2, + "overlap_comm": true, + "reduce_scatter": true, + "contiguous_gradients": true, + "offload_optimizer": { + "device": "cpu" + }, + "offload_param": { + "device": "cpu" + } + }, + "wall_clock_breakdown": false, + "flops_profiler": { + "enabled": false, + "profile_step": 1, + "module_depth": -1, + "top_modules": 1, + "detailed": true, + "output_file": null + } +} \ No newline at end of file diff --git a/finetune/configs/ds_config_zero3.json b/finetune/configs/ds_config_zero3.json new file mode 100644 index 0000000..1d091df --- /dev/null +++ b/finetune/configs/ds_config_zero3.json @@ -0,0 +1,22 @@ +{ + "train_batch_size": "auto", + "train_micro_batch_size_per_gpu": "auto", + "gradient_accumulation_steps": "auto", + "gradient_clipping": 1.0, + "bf16": { + "enabled": "auto" + }, + "zero_optimization": { + "stage": 3, + "overlap_comm": true, + "stage3_gather_16bit_weights_on_model_save": true + }, + "flops_profiler": { + "enabled": false, + "profile_step": 1, + "module_depth": -1, + "top_modules": 1, + "detailed": true, + "output_file": null + } +} \ No newline at end of file diff --git a/finetune/finetune.py b/finetune/finetune.py new file mode 100644 index 0000000..01eeb92 --- /dev/null +++ b/finetune/finetune.py @@ -0,0 +1,198 @@ +# -*- coding: utf-8 -*- +import json +import time +from typing import Dict, Optional +from datetime import datetime +from dataclasses import dataclass, field + +import torch +from torch.utils.data import Dataset + +import transformers +from transformers import AutoTokenizer, AutoModelForCausalLM, TrainingArguments, Trainer + + +@dataclass +class ModelArguments: + model_name_or_path: Optional[str] = field(default="baichuan-inc/Baichuan2-7B-Base") + + +@dataclass +class DataArguments: + train_data_path: str = field( + default="data/AdvertiseGenChatML/train.json", + metadata={"help": "Path to the training data."}, + ) + eval_data_path: str = field( + default="data/AdvertiseGenChatML/dev.json", + metadata={"help": "Path to the test data."}, + ) + + +@dataclass +class TrainingArguments(transformers.TrainingArguments): + cache_dir: Optional[str] = field(default=None) + optim: str = field(default="adamw_torch") + model_max_length: int = field( + default=512, + metadata={ + "help": "Maximum sequence length. Sequences will be right padded (and possibly truncated)." + }, + ) + use_lora: bool = field(default=False) + + +class SupervisedDataset(Dataset): + """Dataset for supervised fine-tuning.""" + + def __init__( + self, + data_path, + tokenizer, + model_max_length=4096, + user_tokens=[1786, 4194, 95388], + assistant_tokens=[1786, 10850, 95388], + ): + super(SupervisedDataset, self).__init__() + self.data = json.load(open(data_path)) + self.tokenizer = tokenizer + self.model_max_length = model_max_length + self.user_tokens = user_tokens + self.assistant_tokens = assistant_tokens + self.ignore_index = -100 + item = self.preprocessing(self.data[0]) + print("input:", self.tokenizer.decode(item["input_ids"])) + labels = [] + for id_ in item["label_ids"]: + if id_ == -100: + continue + + labels.append(id_) + print("label:", self.tokenizer.decode(labels)) + + def __len__(self): + return len(self.data) + + def preprocessing(self, example): + input_ids = [self.tokenizer.bos_token_id] + label_ids = [self.tokenizer.bos_token_id] + + for message in example["messages"]: + role = message["role"] + content = message["content"] + content_ids = self.tokenizer.encode(content, add_special_tokens=False) + + if role == "user": + input_ids += self.user_tokens + content_ids + label_ids += ( + [self.ignore_index] * len(self.user_tokens) + + [self.tokenizer.eos_token_id] + + [self.ignore_index] * len(content_ids) + ) + else: + input_ids += self.assistant_tokens + content_ids + label_ids += [self.ignore_index] * len( + self.assistant_tokens + ) + content_ids + input_ids.append(self.tokenizer.eos_token_id) + label_ids.append(self.tokenizer.eos_token_id) + input_ids = input_ids[: self.model_max_length] + label_ids = label_ids[: self.model_max_length] + # input_ids += [self.tokenizer.eos_token_id] * (len(label_ids) - len(input_ids)) + input_ids += [self.tokenizer.eos_token_id] * ( + self.model_max_length - len(input_ids) + ) + label_ids += [self.ignore_index] * (self.model_max_length - len(label_ids)) + input_ids = torch.LongTensor(input_ids) + label_ids = torch.LongTensor(label_ids) + # print(f"len input_ids: {len(input_ids)}, len label_ids: {len(label_ids)}") + attention_mask = input_ids.ne(self.tokenizer.eos_token_id) + return { + "input_ids": input_ids, + "label_ids": label_ids, + "attention_mask": attention_mask, + } + + def __getitem__(self, idx) -> Dict[str, torch.Tensor]: + return self.preprocessing(self.data[idx]) + + +def load_model_and_tokenizer( + model_path: str, + max_length: int = 4096, + use_lora: bool = True, + bf16: bool = False, + fp16: bool = False, +): + """load model and tokenizer""" + tokenizer = AutoTokenizer.from_pretrained(model_path, trust_remote_code=True) + tokenizer.pad_token = tokenizer.eos_token + + assert not (bf16 and fp16), "bf16 or fp16, not both" + if bf16: + dtype = torch.bfloat16 + elif fp16: + dtype = torch.float16 + else: + dtype = torch.float32 + model = AutoModelForCausalLM.from_pretrained( + model_path, + torch_dtype=dtype, + trust_remote_code=True, + ) + if use_lora: + from peft import LoraConfig, TaskType, get_peft_model + + lora_config = LoraConfig( + init_lora_weights="gaussian", + task_type=TaskType.CAUSAL_LM, + target_modules=["q_proj", "v_proj"], + r=8, + lora_alpha=32, + lora_dropout=0.1, + inference_mode=False, + ) + model = get_peft_model(model, lora_config) + # trainable params: 2,949,120 || all params: 3,010,652,928 || trainable%: 0.09795616002669305 + model.print_trainable_parameters() + model.enable_input_require_grads() + + return model, tokenizer + + +if __name__ == "__main__": + model_path = "/mnt/data/user/tc_agi/yh/models/MiniCPM" + max_length = 512 + parser = transformers.HfArgumentParser( + (ModelArguments, DataArguments, TrainingArguments) + ) + model_args, data_args, training_args = parser.parse_args_into_dataclasses() + model, tokenizer = load_model_and_tokenizer( + model_path=model_args.model_name_or_path, + max_length=training_args.model_max_length, + use_lora=training_args.use_lora, + ) + + train_dataset = SupervisedDataset( + data_path=data_args.train_data_path, + tokenizer=tokenizer, + model_max_length=training_args.model_max_length, + ) + eval_dataset = SupervisedDataset( + data_path=data_args.eval_data_path, + tokenizer=tokenizer, + model_max_length=training_args.model_max_length, + ) + + formatted_time = datetime.now().strftime("%Y%m%d%H%M%S") + + trainer = Trainer( + model=model, + args=training_args, + train_dataset=train_dataset, + eval_dataset=eval_dataset, + tokenizer=tokenizer, + # compute_metrics=compute_metrics, + ) + + trainer.train() diff --git a/finetune/lora_finetune.ipynb b/finetune/lora_finetune.ipynb new file mode 100644 index 0000000..dc6e91c --- /dev/null +++ b/finetune/lora_finetune.ipynb @@ -0,0 +1,129 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# MiniCPM-2B 参数高效微调（LoRA）消费级单卡示例\n", + "\n", + "本 notebook 是一个使用 `AdvertiseGen` 数据集对 MiniCPM-2B 进行 LoRA 微调，使其具备专业的广告生成能力的代码示例。\n", + "\n", + "## 硬件需求\n", + "- 显存：24GB\n", + "- 显卡架构：安培架构（推荐）\n", + "- 内存：16GB" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## 1. 准备数据集\n", + "\n", + "下载 AdvertiseGen 数据集\n", + "- [Google Drive](https://drive.google.com/file/d/13_vf0xRTQsyneRKdD1bZIr93vBGOczrk/view?usp=sharing)\n", + "- [Tsinghua Cloud](https://cloud.tsinghua.edu.cn/f/b3f119a008264b1cabd1/?dl=1)\n", + "\n", + "下载后的数据集格式为 `.tar.gz` 的压缩格式，接下来的操作中，假设该压缩包被置于 `finetune/data/`。\n" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# 校验文件完整性\n", + "!md5sum data/AdvertiseGen.tar.gz " + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# 解压数据集\n", + "!tar xvf data/AdvertiseGen.tar.gz " + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# 转换为 ChatML 格式\n", + "import os\n", + "import shutil\n", + "import json\n", + "\n", + "input_dir = \"data/AdvertiseGen\"\n", + "output_dir = \"data/AdvertiseGenChatML\"\n", + "if os.path.exists(output_dir):\n", + " shutil.rmtree(output_dir)\n", + "os.makedirs(output_dir, exist_ok=True)\n", + "\n", + "for fn in [\"train.json\", \"dev.json\"]:\n", + " data_out_list = []\n", + " with open(os.path.join(input_dir, fn), \"r\") as f, open(os.path.join(output_dir, fn), \"w\") as fo:\n", + " for line in f:\n", + " if len(line.strip()) > 0:\n", + " data = json.loads(line)\n", + " data_out = {\n", + " \"messages\": [\n", + " {\n", + " \"role\": \"user\",\n", + " \"content\": data[\"content\"],\n", + " },\n", + " {\n", + " \"role\": \"assistant\",\n", + " \"content\": data[\"summary\"],\n", + " },\n", + " ]\n", + " }\n", + " data_out_list.append(data_out)\n", + " json.dump(data_out_list, fo, ensure_ascii=False, indent=4)\n" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## 2. 使用 LoRA 进行微调\n", + "\n", + "命令行一键运行" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "!bash lora_finetune.sh" + ] + } + ], + "metadata": { + "kernelspec": { + "display_name": "base", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.10.13" + } + }, + "nbformat": 4, + "nbformat_minor": 2 +} diff --git a/finetune/lora_finetune.sh b/finetune/lora_finetune.sh new file mode 100644 index 0000000..86738f7 --- /dev/null +++ b/finetune/lora_finetune.sh @@ -0,0 +1,16 @@ +formatted_time=$(date +"%Y%m%d%H%M%S") +echo $formatted_time + +CUDA_VISIBLE_DEVICES=0 python finetune.py \ + --model_name_or_path \ + --output_dir output/AdvertiseGenLoRA/$formatted_time/ \ + --train_data_path data/AdvertiseGenChatML/train.json \ + --eval_data_path data/AdvertiseGenChatML/dev.json \ + --learning_rate 1e-3 --per_device_train_batch_size 1 \ + --per_device_eval_batch_size 32 --fp16\ + --gradient_accumulation_steps 8 --warmup_steps 100 \ + --max_steps 3000 --weight_decay 0.01 \ + --evaluation_strategy steps --eval_steps 500 \ + --save_strategy steps --save_steps 500 \ + --use_lora true --seed 42 \ + --log_level info --logging_strategy steps --logging_steps 10 diff --git a/finetune/lora_finetune_ds.sh b/finetune/lora_finetune_ds.sh new file mode 100644 index 0000000..5ced717 --- /dev/null +++ b/finetune/lora_finetune_ds.sh @@ -0,0 +1,17 @@ +formatted_time=$(date +"%Y%m%d%H%M%S") +echo $formatted_time + + +deepspeed --include localhost:0,1 finetune.py \ + --model_name_or_path \ + --output_dir output/AdvertiseGenLoRA/$formatted_time/ \ + --train_data_path data/AdvertiseGenChatML/train.json \ + --eval_data_path data/AdvertiseGenChatML/dev.json \ + --learning_rate 1e-3 --per_device_train_batch_size 1 \ + --per_device_eval_batch_size 1 --fp16 --use_lora \ + --gradient_accumulation_steps 8 --warmup_steps 100 \ + --max_steps 3000 --weight_decay 0.01 \ + --evaluation_strategy steps --eval_steps 500 \ + --save_strategy steps --save_steps 500 --seed 42 \ + --log_level info --logging_strategy steps --logging_steps 10 \ + --deepspeed configs/ds_config_zero2_offload.json diff --git a/finetune/requirements.txt b/finetune/requirements.txt new file mode 100644 index 0000000..6e51c33 --- /dev/null +++ b/finetune/requirements.txt @@ -0,0 +1,9 @@ +# for finetune +jieba>=0.42.1 +ruamel_yaml>=0.18.5 +rouge_chinese>=1.0.3 +jupyter>=1.0.0 +datasets>=2.16.1 +peft>=0.7.1 +deepspeed>=0.13.1 +flash_attn>=2.5.1 diff --git a/finetune/sft_finetune.sh b/finetune/sft_finetune.sh new file mode 100644 index 0000000..6338da8 --- /dev/null +++ b/finetune/sft_finetune.sh @@ -0,0 +1,17 @@ +formatted_time=$(date +"%Y%m%d%H%M%S") +echo $formatted_time + + +deepspeed --include localhost:1,2 finetune.py \ + --model_name_or_path \ + --output_dir output/AdvertiseGenLoRA/$formatted_time/ \ + --train_data_path data/AdvertiseGenChatML/train.json \ + --eval_data_path data/AdvertiseGenChatML/dev.json \ + --learning_rate 1e-3 --per_device_train_batch_size 1 \ + --per_device_eval_batch_size 32 --fp16 \ + --gradient_accumulation_steps 8 --warmup_steps 100 \ + --max_steps 3000 --weight_decay 0.01 \ + --evaluation_strategy steps --eval_steps 500 \ + --save_strategy steps --save_steps 500 --seed 42 \ + --log_level info --logging_strategy steps --logging_steps 10 \ + --deepspeed configs/ds_config_zero2.json