跳转至

课时四 XTuner微调多模态Agent

alt text

飞书地址

算力平台

1. 提交的作业结果

作业要求

1.1 训练自己的小助手认知（记录复现过程并截图）

  • 训练完小助手对话截图：

alt text - 复现笔记： 复现步骤

1.2 将自我认知的模型上传到 OpenXLab，并将应用部署到 OpenXLab

1.3 复现多模态微调

  • 复现结果截图

alt text

  • 复现过程文档

多模态微调复现

2. 文档复现笔记

文档地址

2.1 前置知识

2.2 复现步骤

2.2.1 环境配置

  • 创建开发机

alt text

  • 进入开发机创建conda环境
cd /root
/share/install_conda_env_internlm_base.sh xtuner0.1.9

alt text

  • 激活xtuner0.1.9环境&创建文件夹
conda activate xtuner0.1.9
cd /root
mkdir xtuner019 && cd xtuner019

alt text

  • 获取0.1.9版本源码

git clone -b v0.1.9  https://github.com/InternLM/xtuner
alt text

  • 从源码安装Xtuner
cd xtuner
pip install -e '.[all]'

alt text

  • 创建数据集文件夹地址
# 创建一个微调 oasst1 数据集的工作路径，进入
mkdir ~/ft-oasst1 && cd ~/ft-oasst1

alt text

2.2.2 微调

2.2.2.1 准备配置文件
# 列出所有内置配置
xtuner list-cfg

假如显示bash: xtuner: command not found的话可以考虑在终端输入 export PATH=$PATH:'/root/.local/bin'

我是又执行了一遍源码安装命令

cd xtuner
pip install -e '.[all]'
  • 配置文件 
    xtuner list-cfg

alt text

展开内容 
==========================CONFIGS===========================
baichuan2_13b_base_qlora_alpaca_e3
baichuan2_13b_base_qlora_alpaca_enzh_e3
baichuan2_13b_base_qlora_alpaca_enzh_oasst1_e3
baichuan2_13b_base_qlora_alpaca_zh_e3
baichuan2_13b_base_qlora_arxiv_gentitle_e3
baichuan2_13b_base_qlora_code_alpaca_e3
baichuan2_13b_base_qlora_colorist_e5
baichuan2_13b_base_qlora_lawyer_e3
baichuan2_13b_base_qlora_oasst1_512_e3
baichuan2_13b_base_qlora_oasst1_e3
baichuan2_13b_base_qlora_open_platypus_e3
baichuan2_13b_base_qlora_sql_e3
baichuan2_13b_chat_qlora_alpaca_e3
baichuan2_13b_chat_qlora_alpaca_enzh_e3
baichuan2_13b_chat_qlora_alpaca_enzh_oasst1_e3
baichuan2_13b_chat_qlora_alpaca_zh_e3
baichuan2_13b_chat_qlora_code_alpaca_e3
baichuan2_13b_chat_qlora_lawyer_e3
baichuan2_13b_chat_qlora_oasst1_512_e3
baichuan2_13b_chat_qlora_oasst1_e3
baichuan2_13b_chat_qlora_open_platypus_e3
baichuan2_7b_base_qlora_alpaca_e3
baichuan2_7b_base_qlora_alpaca_enzh_e3
baichuan2_7b_base_qlora_alpaca_enzh_oasst1_e3
baichuan2_7b_base_qlora_alpaca_zh_e3
baichuan2_7b_base_qlora_arxiv_gentitle_e3
baichuan2_7b_base_qlora_code_alpaca_e3
baichuan2_7b_base_qlora_colorist_e5
baichuan2_7b_base_qlora_lawyer_e3
baichuan2_7b_base_qlora_oasst1_512_e3
baichuan2_7b_base_qlora_oasst1_e3
baichuan2_7b_base_qlora_open_platypus_e3
baichuan2_7b_base_qlora_sql_e3
baichuan2_7b_chat_qlora_alpaca_e3
baichuan2_7b_chat_qlora_alpaca_enzh_e3
baichuan2_7b_chat_qlora_alpaca_enzh_oasst1_e3
baichuan2_7b_chat_qlora_alpaca_zh_e3
baichuan2_7b_chat_qlora_code_alpaca_e3
baichuan2_7b_chat_qlora_lawyer_e3
baichuan2_7b_chat_qlora_oasst1_512_e3
baichuan2_7b_chat_qlora_oasst1_e3
baichuan2_7b_chat_qlora_open_platypus_e3
baichuan_13b_base_qlora_alpaca_e3
baichuan_13b_base_qlora_alpaca_enzh_e3
baichuan_13b_base_qlora_alpaca_enzh_oasst1_e3
baichuan_13b_base_qlora_alpaca_zh_e3
baichuan_13b_base_qlora_arxiv_gentitle_e3
baichuan_13b_base_qlora_code_alpaca_e3
baichuan_13b_base_qlora_colorist_e5
baichuan_13b_base_qlora_lawyer_e3
baichuan_13b_base_qlora_medical_e1
baichuan_13b_base_qlora_moss_sft_all_e1
baichuan_13b_base_qlora_moss_sft_all_e2_gpu8
baichuan_13b_base_qlora_moss_sft_plugins_e1
baichuan_13b_base_qlora_oasst1_512_e3
baichuan_13b_base_qlora_oasst1_e3
baichuan_13b_base_qlora_open_platypus_e3
baichuan_13b_base_qlora_openorca_e1
baichuan_13b_base_qlora_sql_e3
baichuan_13b_base_qlora_tiny_codes_e1
baichuan_13b_chat_qlora_alpaca_e3
baichuan_13b_chat_qlora_alpaca_enzh_e3
baichuan_13b_chat_qlora_alpaca_enzh_oasst1_e3
baichuan_13b_chat_qlora_alpaca_zh_e3
baichuan_13b_chat_qlora_arxiv_gentitle_e3
baichuan_13b_chat_qlora_code_alpaca_e3
baichuan_13b_chat_qlora_colorist_e5
baichuan_13b_chat_qlora_lawyer_e3
baichuan_13b_chat_qlora_medical_e1
baichuan_13b_chat_qlora_oasst1_512_e3
baichuan_13b_chat_qlora_oasst1_e3
baichuan_13b_chat_qlora_open_platypus_e3
baichuan_13b_chat_qlora_openorca_e1
baichuan_13b_chat_qlora_sql_e3
baichuan_13b_chat_qlora_tiny_codes_e1
baichuan_7b_qlora_alpaca_e3
baichuan_7b_qlora_alpaca_enzh_e3
baichuan_7b_qlora_alpaca_enzh_oasst1_e3
baichuan_7b_qlora_alpaca_zh_e3
baichuan_7b_qlora_arxiv_gentitle_e3
baichuan_7b_qlora_code_alpaca_e3
baichuan_7b_qlora_colorist_e5
baichuan_7b_qlora_lawyer_e3
baichuan_7b_qlora_medical_e1
baichuan_7b_qlora_moss_sft_all_e1
baichuan_7b_qlora_moss_sft_all_e2_gpu8
baichuan_7b_qlora_moss_sft_plugins_e1
baichuan_7b_qlora_oasst1_512_e3
baichuan_7b_qlora_oasst1_e3
baichuan_7b_qlora_open_platypus_e3
baichuan_7b_qlora_openorca_e1
baichuan_7b_qlora_sql_e3
baichuan_7b_qlora_tiny_codes_e1
chatglm2_6b_qlora_alpaca_e3
chatglm2_6b_qlora_alpaca_enzh_e3
chatglm2_6b_qlora_alpaca_enzh_oasst1_e3
chatglm2_6b_qlora_alpaca_zh_e3
chatglm2_6b_qlora_arxiv_gentitle_e3
chatglm2_6b_qlora_code_alpaca_e3
chatglm2_6b_qlora_colorist_e5
chatglm2_6b_qlora_lawyer_e3
chatglm2_6b_qlora_medical_e1
chatglm2_6b_qlora_oasst1_512_e3
chatglm2_6b_qlora_oasst1_e3
chatglm2_6b_qlora_open_platypus_e3
chatglm2_6b_qlora_openorca_e1
chatglm2_6b_qlora_sql_e3
chatglm2_6b_qlora_tiny_codes_e1
chatglm3_6b_base_qlora_alpaca_e3
chatglm3_6b_base_qlora_alpaca_enzh_e3
chatglm3_6b_base_qlora_alpaca_enzh_oasst1_e3
chatglm3_6b_base_qlora_alpaca_zh_e3
chatglm3_6b_base_qlora_arxiv_gentitle_e3
chatglm3_6b_base_qlora_code_alpaca_e3
chatglm3_6b_base_qlora_colorist_e5
chatglm3_6b_base_qlora_lawyer_e3
chatglm3_6b_base_qlora_medical_e1
chatglm3_6b_base_qlora_oasst1_512_e3
chatglm3_6b_base_qlora_oasst1_e3
chatglm3_6b_base_qlora_open_platypus_e3
chatglm3_6b_base_qlora_openorca_e1
chatglm3_6b_base_qlora_sql_e3
chatglm3_6b_base_qlora_tiny_codes_e1
chatglm3_6b_qlora_alpaca_e3
chatglm3_6b_qlora_alpaca_enzh_e3
chatglm3_6b_qlora_alpaca_enzh_oasst1_e3
chatglm3_6b_qlora_alpaca_zh_e3
chatglm3_6b_qlora_arxiv_gentitle_e3
chatglm3_6b_qlora_code_alpaca_e3
chatglm3_6b_qlora_colorist_e5
chatglm3_6b_qlora_lawyer_e3
chatglm3_6b_qlora_medical_e1
chatglm3_6b_qlora_oasst1_512_e3
chatglm3_6b_qlora_oasst1_e3
chatglm3_6b_qlora_open_platypus_e3
chatglm3_6b_qlora_openorca_e1
chatglm3_6b_qlora_sql_e3
chatglm3_6b_qlora_tiny_codes_e1
deepspeed_zero1
deepspeed_zero2
deepspeed_zero2_offload
deepspeed_zero3
deepspeed_zero3_offload
internlm_20b_qlora_alpaca_e3
internlm_20b_qlora_alpaca_enzh_e3
internlm_20b_qlora_alpaca_enzh_oasst1_e3
internlm_20b_qlora_alpaca_zh_e3
internlm_20b_qlora_arxiv_gentitle_e3
internlm_20b_qlora_code_alpaca_e3
internlm_20b_qlora_colorist_e5
internlm_20b_qlora_lawyer_e3
internlm_20b_qlora_msagent_react_e3_gpu8
internlm_20b_qlora_oasst1_512_e3
internlm_20b_qlora_oasst1_e3
internlm_20b_qlora_open_platypus_e3
internlm_20b_qlora_sql_e3
internlm_7b_full_alpaca_e3
internlm_7b_full_alpaca_enzh_e3
internlm_7b_full_alpaca_enzh_oasst1_e3
internlm_7b_full_alpaca_zh_e3
internlm_7b_full_oasst1_e3
internlm_7b_qlora_alpaca_e3
internlm_7b_qlora_alpaca_enzh_e3
internlm_7b_qlora_alpaca_enzh_oasst1_e3
internlm_7b_qlora_alpaca_zh_e3
internlm_7b_qlora_arxiv_gentitle_e3
internlm_7b_qlora_code_alpaca_e3
internlm_7b_qlora_colorist_e5
internlm_7b_qlora_lawyer_e3
internlm_7b_qlora_medical_e1
internlm_7b_qlora_moss_sft_all_e1
internlm_7b_qlora_moss_sft_all_e2_gpu8
internlm_7b_qlora_moss_sft_plugins_e1
internlm_7b_qlora_msagent_react_e3_gpu8
internlm_7b_qlora_oasst1_512_e3
internlm_7b_qlora_oasst1_e3
internlm_7b_qlora_oasst1_e3_hf
internlm_7b_qlora_oasst1_mmlu_e3
internlm_7b_qlora_open_platypus_e3
internlm_7b_qlora_openorca_e1
internlm_7b_qlora_sql_e3
internlm_7b_qlora_tiny_codes_e1
internlm_chat_20b_qlora_alpaca_e3
internlm_chat_20b_qlora_alpaca_enzh_e3
internlm_chat_20b_qlora_alpaca_enzh_oasst1_e3
internlm_chat_20b_qlora_alpaca_zh_e3
internlm_chat_20b_qlora_code_alpaca_e3
internlm_chat_20b_qlora_lawyer_e3
internlm_chat_20b_qlora_oasst1_512_e3
internlm_chat_20b_qlora_oasst1_e3
internlm_chat_20b_qlora_open_platypus_e3
internlm_chat_7b_qlora_alpaca_e3
internlm_chat_7b_qlora_alpaca_enzh_e3
internlm_chat_7b_qlora_alpaca_enzh_oasst1_e3
internlm_chat_7b_qlora_alpaca_zh_e3
internlm_chat_7b_qlora_arxiv_gentitle_e3
internlm_chat_7b_qlora_code_alpaca_e3
internlm_chat_7b_qlora_colorist_e5
internlm_chat_7b_qlora_lawyer_e3
internlm_chat_7b_qlora_medical_e1
internlm_chat_7b_qlora_oasst1_512_e3
internlm_chat_7b_qlora_oasst1_e3
internlm_chat_7b_qlora_open_platypus_e3
internlm_chat_7b_qlora_openorca_e1
internlm_chat_7b_qlora_sql_e3
internlm_chat_7b_qlora_tiny_codes_e1
llama2_70b_int8_lora_open_platypus_e1
llama2_70b_int8_lora_open_platypus_e1_hf
llama2_70b_qlora_open_platypus_e1
llama2_70b_qlora_open_platypus_e1_hf
llama2_7b_chat_qlora_alpaca_e3
llama2_7b_chat_qlora_alpaca_enzh_e3
llama2_7b_chat_qlora_alpaca_enzh_oasst1_e3
llama2_7b_chat_qlora_alpaca_zh_e3
llama2_7b_chat_qlora_arxiv_gentitle_e3
llama2_7b_chat_qlora_code_alpaca_e3
llama2_7b_chat_qlora_colorist_e5
llama2_7b_chat_qlora_lawyer_e3
llama2_7b_chat_qlora_medical_e1
llama2_7b_chat_qlora_oasst1_512_e3
llama2_7b_chat_qlora_oasst1_e3
llama2_7b_chat_qlora_open_platypus_e3
llama2_7b_chat_qlora_openorca_e1
llama2_7b_chat_qlora_sql_e3
llama2_7b_chat_qlora_tiny_codes_e1
llama2_7b_full_wizardlm_e1
llama2_7b_qlora_alpaca_e3
llama2_7b_qlora_alpaca_enzh_e3
llama2_7b_qlora_alpaca_enzh_oasst1_e3
llama2_7b_qlora_alpaca_zh_e3
llama2_7b_qlora_arxiv_gentitle_e3
llama2_7b_qlora_code_alpaca_e3
llama2_7b_qlora_colorist_e5
llama2_7b_qlora_lawyer_e3
llama2_7b_qlora_medical_e1
llama2_7b_qlora_moss_sft_all_e1
llama2_7b_qlora_moss_sft_all_e2_gpu8
llama2_7b_qlora_moss_sft_plugins_e1
llama2_7b_qlora_msagent_react_e3_gpu8
llama2_7b_qlora_oasst1_512_e3
llama2_7b_qlora_oasst1_e3
llama2_7b_qlora_open_platypus_e3
llama2_7b_qlora_openorca_e1
llama2_7b_qlora_sql_e3
llama2_7b_qlora_tiny_codes_e1
llama_7b_qlora_alpaca_e3
llama_7b_qlora_alpaca_enzh_e3
llama_7b_qlora_alpaca_enzh_oasst1_e3
llama_7b_qlora_alpaca_zh_e3
llama_7b_qlora_arxiv_gentitle_e3
llama_7b_qlora_code_alpaca_e3
llama_7b_qlora_colorist_e5
llama_7b_qlora_lawyer_e3
llama_7b_qlora_medical_e1
llama_7b_qlora_moss_sft_all_e1
llama_7b_qlora_moss_sft_all_e2_gpu8
llama_7b_qlora_moss_sft_plugins_e1
llama_7b_qlora_oasst1_512_e3
llama_7b_qlora_oasst1_e3
llama_7b_qlora_open_platypus_e3
llama_7b_qlora_openorca_e1
llama_7b_qlora_sql_e3
llama_7b_qlora_tiny_codes_e1
mistral_7b_qlora_skypile_pretrain_e1
qwen_7b_chat_qlora_alpaca_e3
qwen_7b_chat_qlora_alpaca_enzh_e3
qwen_7b_chat_qlora_alpaca_enzh_oasst1_e3
qwen_7b_chat_qlora_alpaca_zh_e3
qwen_7b_chat_qlora_arxiv_gentitle_e3
qwen_7b_chat_qlora_code_alpaca_e3
qwen_7b_chat_qlora_colorist_e5
qwen_7b_chat_qlora_lawyer_e3
qwen_7b_chat_qlora_medical_e1
qwen_7b_chat_qlora_oasst1_512_e3
qwen_7b_chat_qlora_oasst1_e3
qwen_7b_chat_qlora_open_platypus_e3
qwen_7b_chat_qlora_openorca_e1
qwen_7b_chat_qlora_sql_e3
qwen_7b_chat_qlora_tiny_codes_e1
qwen_7b_qlora_alpaca_e3
qwen_7b_qlora_alpaca_enzh_e3
qwen_7b_qlora_alpaca_enzh_oasst1_e3
qwen_7b_qlora_alpaca_zh_e3
qwen_7b_qlora_arxiv_gentitle_e3
qwen_7b_qlora_code_alpaca_e3
qwen_7b_qlora_colorist_e5
qwen_7b_qlora_lawyer_e3
qwen_7b_qlora_medical_e1
qwen_7b_qlora_moss_sft_all_e1
qwen_7b_qlora_moss_sft_all_e2_gpu8
qwen_7b_qlora_moss_sft_plugins_e1
qwen_7b_qlora_oasst1_512_e3
qwen_7b_qlora_oasst1_e3
qwen_7b_qlora_open_platypus_e3
qwen_7b_qlora_openorca_e1
qwen_7b_qlora_sql_e3
qwen_7b_qlora_tiny_codes_e1
starcoder_qlora_stack_exchange_example
yi_34b_qlora_alpaca_enzh_e3
yi_6b_qlora_alpaca_enzh_e3
zephyr_7b_beta_qlora_alpaca_e3
=============================================================
  • 拷贝一个配置文件到当前目录： 
    cd ~/ft-oasst1
xtuner copy-cfg internlm_chat_7b_qlora_oasst1_e3 .
    alt text
2.2.2.2 模型设置
  • 设置软链接

ln -s /share/temp/model_repos/internlm-chat-7b ~/ft-oasst1/
alt text

2.2.2.3 数据集下载

数据集Huggingface地址

  • 从开发机share地址拷贝到ft-oasst1文件夹下 
    cd ~/ft-oasst1
# ...-guanaco 后面有个空格和英文句号啊
cp -r /root/share/temp/datasets/openassistant-guanaco .

alt text

2.2.2.4 修改配置文件
  • 修改internlm_chat_7b_qlora_oasst1_e3_copy.py配置文件模型及数据为本地模型及数据
# pretrained_model_name_or_path = 'internlm/internlm-chat-7b'
pretrained_model_name_or_path = 'internlm/internlm-chat-7b'

# Data
# data_path = 'timdettmers/openassistant-guanaco'
data_path = './openassistant-guanaco'

alt text

修改后的配置文件internlm_chat_7b_qlora_oasst1_e3_copy.py

internlm_chat_7b_qlora_oasst1_e3_copy.py 
  1
  2
  3
  4
  5
  6
  7
  8
  9
 10
 11
 12
 13
 14
 15
 16
 17
 18
 19
 20
 21
 22
 23
 24
 25
 26
 27
 28
 29
 30
 31
 32
 33
 34
 35
 36
 37
 38
 39
 40
 41
 42
 43
 44
 45
 46
 47
 48
 49
 50
 51
 52
 53
 54
 55
 56
 57
 58
 59
 60
 61
 62
 63
 64
 65
 66
 67
 68
 69
 70
 71
 72
 73
 74
 75
 76
 77
 78
 79
 80
 81
 82
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
# Copyright (c) OpenMMLab. All rights reserved.
import torch
from bitsandbytes.optim import PagedAdamW32bit
from datasets import load_dataset
from mmengine.dataset import DefaultSampler
from mmengine.hooks import (CheckpointHook, DistSamplerSeedHook, IterTimerHook,
                            LoggerHook, ParamSchedulerHook)
from mmengine.optim import AmpOptimWrapper, CosineAnnealingLR
from peft import LoraConfig
from transformers import (AutoModelForCausalLM, AutoTokenizer,
                          BitsAndBytesConfig)

from xtuner.dataset import process_hf_dataset
from xtuner.dataset.collate_fns import default_collate_fn
from xtuner.dataset.map_fns import oasst1_map_fn, template_map_fn_factory
from xtuner.engine import DatasetInfoHook, EvaluateChatHook
from xtuner.model import SupervisedFinetune
from xtuner.utils import PROMPT_TEMPLATE

#######################################################################
#                          PART 1  Settings                           #
#######################################################################
# Model
# pretrained_model_name_or_path = 'internlm/internlm-chat-7b'
pretrained_model_name_or_path = './internlm-chat-7b'

# Data
# data_path = 'timdettmers/openassistant-guanaco'
data_path = './openassistant-guanaco'
prompt_template = PROMPT_TEMPLATE.internlm_chat
max_length = 2048
pack_to_max_length = True

# Scheduler & Optimizer
batch_size = 1  # per_device
accumulative_counts = 16
dataloader_num_workers = 0
max_epochs = 3
optim_type = PagedAdamW32bit
lr = 2e-4
betas = (0.9, 0.999)
weight_decay = 0
max_norm = 1  # grad clip

# Evaluate the generation performance during the training
evaluation_freq = 500
SYSTEM = ''
evaluation_inputs = [
    '请给我介绍五个上海的景点', 'Please tell me five scenic spots in Shanghai'
]

#######################################################################
#                      PART 2  Model & Tokenizer                      #
#######################################################################
tokenizer = dict(
    type=AutoTokenizer.from_pretrained,
    pretrained_model_name_or_path=pretrained_model_name_or_path,
    trust_remote_code=True,
    padding_side='right')

model = dict(
    type=SupervisedFinetune,
    llm=dict(
        type=AutoModelForCausalLM.from_pretrained,
        pretrained_model_name_or_path=pretrained_model_name_or_path,
        trust_remote_code=True,
        torch_dtype=torch.float16,
        quantization_config=dict(
            type=BitsAndBytesConfig,
            load_in_4bit=True,
            load_in_8bit=False,
            llm_int8_threshold=6.0,
            llm_int8_has_fp16_weight=False,
            bnb_4bit_compute_dtype=torch.float16,
            bnb_4bit_use_double_quant=True,
            bnb_4bit_quant_type='nf4')),
    lora=dict(
        type=LoraConfig,
        r=64,
        lora_alpha=16,
        lora_dropout=0.1,
        bias='none',
        task_type='CAUSAL_LM'))

#######################################################################
#                      PART 3  Dataset & Dataloader                   #
#######################################################################
train_dataset = dict(
    type=process_hf_dataset,
    dataset=dict(type=load_dataset, path=data_path),
    tokenizer=tokenizer,
    max_length=max_length,
    dataset_map_fn=oasst1_map_fn,
    template_map_fn=dict(
        type=template_map_fn_factory, template=prompt_template),
    remove_unused_columns=True,
    shuffle_before_pack=True,
    pack_to_max_length=pack_to_max_length)

train_dataloader = dict(
    batch_size=batch_size,
    num_workers=dataloader_num_workers,
    dataset=train_dataset,
    sampler=dict(type=DefaultSampler, shuffle=True),
    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=CosineAnnealingLR,
    eta_min=0.0,
    by_epoch=True,
    T_max=max_epochs,
    convert_to_iter_based=True)

# train, val, test setting
train_cfg = dict(by_epoch=True, max_epochs=max_epochs, val_interval=1)

#######################################################################
#                           PART 5  Runtime                           #
#######################################################################
# Log the dialogue periodically during the training process, optional
custom_hooks = [
    dict(type=DatasetInfoHook, tokenizer=tokenizer),
    dict(
        type=EvaluateChatHook,
        tokenizer=tokenizer,
        every_n_iters=evaluation_freq,
        evaluation_inputs=evaluation_inputs,
        system=SYSTEM,
        prompt_template=prompt_template)
]

# configure default hooks
default_hooks = dict(
    # record the time of every iteration.
    timer=dict(type=IterTimerHook),
    # print log every 100 iterations.
    logger=dict(type=LoggerHook, interval=10),
    # enable the parameter scheduler.
    param_scheduler=dict(type=ParamSchedulerHook),
    # save checkpoint per epoch.
    checkpoint=dict(type=CheckpointHook, interval=1),
    # 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)
2.2.2.5 开始微调
  • 训练
# 单卡
## 用刚才改好的config文件训练
xtuner train ./internlm_chat_7b_qlora_oasst1_e3_copy.py --deepspeed deepspeed_zero2

alt text

alt text

2.2.2.6 转换PTH模型为LoRA模型
mkdir hf
export MKL_SERVICE_FORCE_INTEL=1
export MKL_THREADING_LAYER=GNU
xtuner convert pth_to_hf ./internlm_chat_7b_qlora_oasst1_e3_copy.py ./work_dirs/internlm_chat_7b_qlora_oasst1_e3_copy/epoch_1.pth ./hf

alt text

2.2.3 部署&测试

2.2.3.1 合并hf的adapter到大模型
xtuner convert merge ./internlm-chat-7b ./hf ./merged --max-shard-size 2GB

alt text

2.2.3.2 与合并后的模型对话
xtuner chat ./merged --prompt-template internlm_chat
2.2.3.3 运行测试cli_demo.py

修改模型为合并后的模型

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
import torch
from transformers import AutoTokenizer, AutoModelForCausalLM


# model_name_or_path = "/root/model/Shanghai_AI_Laboratory/internlm-chat-7b"
model_name_or_path = "/root/ft-oasst1/merged"

tokenizer = AutoTokenizer.from_pretrained(model_name_or_path, trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained(model_name_or_path, trust_remote_code=True, torch_dtype=torch.bfloat16, device_map='auto')
model = model.eval()

system_prompt = """You are an AI assistant whose name is InternLM (书生·浦语).
- InternLM (书生·浦语) is a conversational language model that is developed by Shanghai AI Laboratory (上海人工智能实验室). It is designed to be helpful, honest, and harmless.
- InternLM (书生·浦语) can understand and communicate fluently in the language chosen by the user such as English and 中文.
"""

messages = [(system_prompt, '')]

print("=============Welcome to InternLM chatbot, type 'exit' to exit.=============")

while True:
    input_text = input("User  >>> ")
    input_text.replace(' ', '')
    if input_text == "exit":
        break
    response, history = model.chat(tokenizer, input_text, history=messages)
    messages.append((input_text, response))
    print(f"robot >>> {response}")

运行示例脚本 

python Tutorial/xtuner/cli_demo.py

alt text

2.2.4 模型&应用部署到OpenXLab

Streamlit部署参考文档

gradio部署参考文档

2.2.5 多模态微调

文档地址

2.2.5.1 环境设置
  • 创建开发机

填写 开发机名称 后，点击 选择镜像 使用 Cuda11.7-conda 镜像，然后在资源配置中，使用 50% A100 * 1 的选项，然后立即创建开发机器。 alt text

  • Xtuner安装
  • conda环境安装
  cd ~ && studio-conda xtuner0.1.17
  # 如果你是在其他平台：
  # conda create --name xtuner0.1.17 python=3.10 -y

alt text

  • 激活环境
# 激活环境
conda activate xtuner0.1.17
  • 获取源码&安装 
    # 进入家目录 （~的意思是 “当前用户的home路径”）
cd ~
# 创建版本文件夹并进入
mkdir -p /root/xtuner0117 && cd /root/xtuner0117

# 拉取 0.1.17 的版本源码
git clone -b v0.1.17  https://github.com/InternLM/xtuner
# 无法访问github的用户请从 gitee 拉取:
# git clone -b v0.1.15 https://gitee.com/Internlm/xtuner

# 进入源码目录
cd /root/xtuner0117/xtuner

# 从源码安装 XTuner
pip install -e '.[all]' && cd ~

alt text

alt text

2.2.5.2 预训练

提供了Pretrain阶段的产物——iter_2181.pth文件。它就是幼稚园阶段的Image Projector！大家带着iter_2181.pth文件继续进入下一阶段进行Finetune即可。

share/new_models/xtuner/iter_2181.pth

alt text

2.2.5.3 微调
  • 格式示意
[
  {
    "id": "000000033471",
    "image": "coco/train2017/000000033471.jpg",
    "conversations": [
      {
        "from": "human",
        "value": "<image>\nWhat are the colors of the bus in the image?"
      },
      {
        "from": "gpt",
        "value": "The bus in the image is white and red."
      },
      {
        "from": "human",
        "value": "What feature can be seen on the back of the bus?"
      },
      {
        "from": "gpt",
        "value": "The back of the bus features an advertisement."
      },
      {
        "from": "human",
        "value": "Is the bus driving down the street or pulled off to the side?"
      },
      {
        "from": "gpt",
        "value": "The bus is driving down the street, which is crowded with people and other vehicles."
      }
    ]
  },
  {
    "id": "000000052846",
    "image": "coco/train2017/000000052846.jpg",
    "conversations": [
      {
        "from": "human",
        "value": "<image>\nWhere is the cat positioned in the image?"
      },
      {
        "from": "gpt",
        "value": "The cat is positioned on top of the back of the couch in the living room."
      }
    ]
  }
 ]
 ```

 - 生成问答对数据(repeat_data.json)

 ```bash
 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
  ```

![alt text](image-213.png)

- 获取配置文件

```bash
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

alt text

  • 开始微调
cd /root/tutorial/xtuner/llava/
xtuner train /root/tutorial/xtuner/llava/llava_internlm2_chat_1_8b_qlora_clip_vit_large_p14_336_lora_e1_gpu8_finetune_copy.py --deepspeed deepspeed_zero2

alt text

alt text

alt text

2.2.5.4 微调模型对话
  • 多卡环境设置
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

alt text

  • chat
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

alt text

3. 视频总结