XTuner 微调 LLM：1.8B、多模态、Agent

XTuner 一个大语言模型&多模态模型微调工具箱。由 MMRazor 和 MMDeploy 联合开发。

1.Xtuner安装

安装虚拟环境

然后git clone

成功后效果如下

我们将 自己构造 <question text><image>--<answer text> 数据对，基于InternLM2_Chat_1.8B这个文本单模态模型，使用LLaVA方案，训练一个给InternLM2_Chat_1.8B使用的Image Projector文件。

分为pretrain和fine-tuning

cd ~ && git clone https://github.com/InternLM/tutorial -b camp2 && conda activate xtuner0.1.17 && cd tutorial

python /root/tutorial/xtuner/llava/llava_data/repeat.py \
  -i /root/tutorial/xtuner/llava/llava_data/unique_data.json \
  -o /root/tutorial/xtuner/llava/llava_data/repeated_data.json \
  -n 200

cp /root/tutorial/xtuner/llava/llava_data/internlm2_chat_1_8b_llava_tutorial_fool_config.py /root/tutorial/xtuner/llava/llava_internlm2_chat_1_8b_qlora_clip_vit_large_p14_336_lora_e1_gpu8_finetune_copy.py

# 查询xtuner内置配置文件
xtuner list-cfg -p llava_internlm2_chat_1_8b

# 拷贝配置文件到当前目录
xtuner copy-cfg \
  llava_internlm2_chat_1_8b_qlora_clip_vit_large_p14_336_lora_e1_gpu8_finetune \
  /root/tutorial/xtuner/llava

修改llava_internlm2_chat_1_8b_qlora_clip_vit_large_p14_336_lora_e1_gpu8_finetune_copy.py文件中的：

• pretrained_pth
• llm_name_or_path
• visual_encoder_name_or_path
• data_root
• data_path
• image_folder

# Model
- llm_name_or_path = 'internlm/internlm2-chat-1_8b'
+ llm_name_or_path = '/root/share/new_models/Shanghai_AI_Laboratory/internlm2-chat-1_8b'
- visual_encoder_name_or_path = 'openai/clip-vit-large-patch14-336'
+ visual_encoder_name_or_path = '/root/share/new_models/openai/clip-vit-large-patch14-336'

# Specify the pretrained pth
- pretrained_pth = './work_dirs/llava_internlm2_chat_1_8b_clip_vit_large_p14_336_e1_gpu8_pretrain/iter_2181.pth'  # noqa: E501
+ pretrained_pth = '/root/share/new_models/xtuner/iter_2181.pth'

# Data
- data_root = './data/llava_data/'
+ data_root = '/root/tutorial/xtuner/llava/llava_data/'
- data_path = data_root + 'LLaVA-Instruct-150K/llava_v1_5_mix665k.json'
+ data_path = data_root + 'repeated_data.json'
- image_folder = data_root + 'llava_images'
+ image_folder = data_root

# Scheduler & Optimizer
- batch_size = 16  # per_device
+ batch_size = 1  # per_device

# evaluation_inputs
- evaluation_inputs = ['请描述一下这张图片','Please describe this picture']
+ evaluation_inputs = ['Please describe this picture','What is the equipment in the image?']

Finetune前

即：加载 1.8B 和 Pretrain阶段产物(iter_2181) 到显存。

# 解决小bug
export MKL_SERVICE_FORCE_INTEL=1
export MKL_THREADING_LAYER=GNU

# pth转huggingface
xtuner convert pth_to_hf \
  llava_internlm2_chat_1_8b_clip_vit_large_p14_336_e1_gpu8_pretrain \
  /root/share/new_models/xtuner/iter_2181.pth \
  /root/tutorial/xtuner/llava/llava_data/iter_2181_hf

# 启动！
xtuner chat /root/share/new_models/Shanghai_AI_Laboratory/internlm2-chat-1_8b \
  --visual-encoder /root/share/new_models/openai/clip-vit-large-patch14-336 \
  --llava /root/tutorial/xtuner/llava/llava_data/iter_2181_hf \
  --prompt-template internlm2_chat \
  --image /root/tutorial/xtuner/llava/llava_data/test_img/oph.jpg

Q1: Describe this image.
Q2: What is the equipment in the image?

Finetune后

即：加载 1.8B 和 Fintune阶段产物 到显存。

# 解决小bug
export MKL_SERVICE_FORCE_INTEL=1
export MKL_THREADING_LAYER=GNU

# pth转huggingface
xtuner convert pth_to_hf \
  /root/tutorial/xtuner/llava/llava_internlm2_chat_1_8b_qlora_clip_vit_large_p14_336_lora_e1_gpu8_finetune_copy.py \
  /root/tutorial/xtuner/llava/work_dirs/llava_internlm2_chat_1_8b_qlora_clip_vit_large_p14_336_lora_e1_gpu8_finetune_copy/iter_1200.pth \
  /root/tutorial/xtuner/llava/llava_data/iter_1200_hf

# 启动！
xtuner chat /root/share/new_models/Shanghai_AI_Laboratory/internlm2-chat-1_8b \
  --visual-encoder /root/share/new_models/openai/clip-vit-large-patch14-336 \
  --llava /root/tutorial/xtuner/llava/llava_data/iter_1200_hf \
  --prompt-template internlm2_chat \
  --image /root/tutorial/xtuner/llava/llava_data/test_img/oph.jpg

Q1: Describe this image.
Q2: What is the equipment in the image?

这是fintune前的，只会打标题，可以看出我们的fintune比较有效