书生浦语大模型实战营第三期 L2 InternVL 多模态模型部署微调实践

1. InternVL 模型总览

2. 环境部署

conda create --name xtuner python=3.10 -y

# 激活虚拟环境（注意：后续的所有操作都需要在这个虚拟环境中进行）
conda activate xtuner

# 安装一些必要的库
conda install pytorch==2.1.2 torchvision==0.16.2 torchaudio==2.1.2 pytorch-cuda=12.1 -c pytorch -c nvidia -y
# 安装其他依赖
apt install libaio-dev
pip install transformers==4.39.3
pip install streamlit==1.36.0
pip install flash_attn

安装xtuner v0.1.23

# 创建一个目录，用来存放源代码
mkdir -p /root/InternLM/code
cd /root/InternLM/code
git clone -b v0.1.23  https://github.com/InternLM/XTuner
cd /root/InternLM/code/XTuner
pip install -e '.[deepspeed]'

安装LMDeploy

pip install lmdeploy==0.5.3

验证安装情况

xtuner version

3. 准备微调数据集

使用huggingface上的zhongshsh/CLoT-Oogiri-GO数据集，已经在开发机的共享空间中了：

## 首先安装一下需要的包
pip install datasets matplotlib Pillow timm

## 把数据集挪出来
cp -r /root/share/new_models/datasets/CLoT_cn_2000 /root/InternLM/datasets/

4. 开始微调

修改配置文件

XTuner下 InternVL的config，文件在：/root/InternLM/code/XTuner/xtuner/configs/internvl/v2/internvl_v2_internlm2_2b_qlora_finetune.py，修改以后如下：

# Copyright (c) OpenMMLab. All rights reserved.
from mmengine.hooks import (CheckpointHook, DistSamplerSeedHook, IterTimerHook,
                            LoggerHook, ParamSchedulerHook)
from mmengine.optim import AmpOptimWrapper, CosineAnnealingLR, LinearLR
from peft import LoraConfig
from torch.optim import AdamW
from transformers import AutoTokenizer

from xtuner.dataset import InternVL_V1_5_Dataset
from xtuner.dataset.collate_fns import default_collate_fn
from xtuner.dataset.samplers import LengthGroupedSampler
from xtuner.engine.hooks import DatasetInfoHook
from xtuner.engine.runner import TrainLoop
from xtuner.model import InternVL_V1_5
from xtuner.utils import PROMPT_TEMPLATE

#######################################################################
#                          PART 1  Settings                           #
#######################################################################
# Model
path = '/root/model/InternVL2-2B'

# Data
data_root = '/root/InternLM/datasets/CLoT_cn_2000/'
data_path = data_root + 'ex_cn.json'
image_folder = data_root
prompt_template = PROMPT_TEMPLATE.internlm2_chat
max_length = 6656

# Scheduler & Optimizer
batch_size = 4  # per_device
accumulative_counts = 4
dataloader_num_workers = 4
max_epochs = 1
optim_type = AdamW
# official 1024 -> 4e-5
lr = 2e-5
betas = (0.9, 0.999)
weight_decay = 0.05
max_norm = 1  # grad clip
warmup_ratio = 0.03

# Save
save_steps = 1000
save_total_limit = 1  # Maximum checkpoints to keep (-1 means unlimited)

#######################################################################
#            PART 2  Model & Tokenizer & Image Processor              #
#######################################################################
model = dict(
    type=InternVL_V1_5,
    model_path=path,
    freeze_llm=True,
    freeze_visual_encoder=True,
    quantization_llm=True,  # or False
    quantization_vit=False,  # or True and uncomment visual_encoder_lora
    # comment the following lines if you don't want to use Lora in llm
    llm_lora=dict(
        type=LoraConfig,
        r=128,
        lora_alpha=256,
        lora_dropout=0.05,
        target_modules=None,
        task_type='CAUSAL_LM'),
    # uncomment the following lines if you don't want to use Lora in visual encoder # noqa
    # visual_encoder_lora=dict(
    #     type=LoraConfig, r=64, lora_alpha=16, lora_dropout=0.05,
    #     target_modules=['attn.qkv', 'attn.proj', 'mlp.fc1', 'mlp.fc2'])
)

#######################################################################
#                      PART 3  Dataset & Dataloader                   #
#######################################################################
llava_dataset = dict(
    type=InternVL_V1_5_Dataset,
    model_path=path,
    data_paths=data_path,
    image_folders=image_folder,
    template=prompt_template,
    max_length=max_length)

train_dataloader = dict(
    batch_size=batch_size,
    num_workers=dataloader_num_workers,
    dataset=llava_dataset,
    sampler=dict(
        type=LengthGroupedSampler,
        length_property='modality_length',
        per_device_batch_size=batch_size * accumulative_counts),
    collate_fn=dict(type=default_collate_fn))

#######################################################################
#                    PART 4  Scheduler & Optimizer                    #
#######################################################################
# optimizer
optim_wrapper = dict(
    type=AmpOptimWrapper,
    optimizer=dict(
        type=optim_type, lr=lr, betas=betas, weight_decay=weight_decay),
    clip_grad=dict(max_norm=max_norm, error_if_nonfinite=False),
    accumulative_counts=accumulative_counts,
    loss_scale='dynamic',
    dtype='float16')

# learning policy
# More information: https://github.com/open-mmlab/mmengine/blob/main/docs/en/tutorials/param_scheduler.md  # noqa: E501
param_scheduler = [
    dict(
        type=LinearLR,
        start_factor=1e-5,
        by_epoch=True,
        begin=0,
        end=warmup_ratio * max_epochs,
        convert_to_iter_based=True),
    dict(
        type=CosineAnnealingLR,
        eta_min=0.0,
        by_epoch=True,
        begin=warmup_ratio * max_epochs,
        end=max_epochs,
        convert_to_iter_based=True)
]

# train, val, test setting
train_cfg = dict(type=TrainLoop, max_epochs=max_epochs)

#######################################################################
#                           PART 5  Runtime                           #
#######################################################################
# Log the dialogue periodically during the training process, optional
tokenizer = dict(
    type=AutoTokenizer.from_pretrained,
    pretrained_model_name_or_path=path,
    trust_remote_code=True)

custom_hooks = [
    dict(type=DatasetInfoHook, tokenizer=tokenizer),
]

# configure default hooks
default_hooks = dict(
    # record the time of every iteration.
    timer=dict(type=IterTimerHook),
    # print log every 10 iterations.
    logger=dict(type=LoggerHook, log_metric_by_epoch=False, interval=10),
    # enable the parameter scheduler.
    param_scheduler=dict(type=ParamSchedulerHook),
    # save checkpoint per `save_steps`.
    checkpoint=dict(
        type=CheckpointHook,
        save_optimizer=False,
        by_epoch=False,
        interval=save_steps,
        max_keep_ckpts=save_total_limit),
    # set sampler seed in distributed evrionment.
    sampler_seed=dict(type=DistSamplerSeedHook),
)

# configure environment
env_cfg = dict(
    # whether to enable cudnn benchmark
    cudnn_benchmark=False,
    # set multi process parameters
    mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0),
    # set distributed parameters
    dist_cfg=dict(backend='nccl'),
)

# set visualizer
visualizer = None

# set log level
log_level = 'INFO'

# load from which checkpoint
load_from = None

# whether to resume training from the loaded checkpoint
resume = False

# Defaults to use random seed and disable `deterministic`
randomness = dict(seed=None, deterministic=False)

# set log processor
log_processor = dict(by_epoch=False)

开始训练

50%A100资源下大概每1000次迭代需要耗时2小时左右，一共3000次迭代微调共用时6个多小时：

NPROC_PER_NODE=1 xtuner train /root/InternLM/code/XTuner/xtuner/configs/internvl/v2/internvl_v2_internlm2_2b_qlora_finetune.py  --work-dir /root/InternLM/work_dir/internvl_ft_run_8_filter  --deepspeed deepspeed_zero1

结果如下：

合并权重&&模型转换

python3 /root/XTuner/code/XTuner/xtuner/configs/internvl/v1_5/convert_to_official.py /root/XTuner/code/XTuner/xtuner/configs/internvl/v2/internvl_v2_internlm2_2b_qlora_finetune.py /root/XTuner/gengtu/work_dir/internvl_ft_run_8_filter/iter_3000.pth /root/XTuner/gengtu/InternVL2-2B/

5. 微调后效果对比

首先试试微调前的效果，运行以下代码：

from lmdeploy import pipeline
from lmdeploy.vl import load_image

# 微调前
pipe = pipeline('/root/model/InternVL2-2B')
# 微调后
# pipe = pipeline('/root/InternLM/InternVL2-2B')

image = load_image('/root/share/new_models/datasets/CLoT_cn_2000/ex_images/007aPnLRgy1hb39z0im50j30ci0el0wm.jpg')
response = pipe(('请你根据这张图片，讲一个脑洞大开的梗', image))
print(response.text)

输入图片：

微调前：

微调后：