从源代码开始安装,因为docker镜像库不能用了。。。
参考官网:
https://docs.dbgpt.site/docs/latest/quickstart
安装文档 - source
https://docs.dbgpt.site/docs/latest/installation/sourcecode
下载源代码
Download DB-GPT
git clone https://github.com/eosphoros-ai/DB-GPT.git
报错
remote: Counting objects: 100% (208/208), done.
remote: Compressing objects: 100% (175/175), done.
error: RPC failed; curl 56 GnuTLS recv error (-9): Error decoding the received TLS packet.
error: 57342 bytes of body are still expected
fetch-pack: unexpected disconnect while reading sideband packet
fatal: early EOF
fatal: fetch-pack: invalid index-pack output
奇怪 ,貌似我的这个Ubuntu有问题,不深究,从别的Ubuntu系统下载,然后拷贝过去
tar -zcvf DB-GPT.tgz DB-GPT/
scp DB-GPT.tgz root@192.168.254.194:/data/
下载minoconda和python
wget https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh
chmod +x Miniconda3-latest-Linux-x86_64.sh
./Miniconda3-latest-Linux-x86_64.sh
You can undo this by running `conda init --reverse $SHELL`? [yes|no]
[no] >>>
这里要选yes
source ~/.bashrc
注意:
Do you wish to update your shell profile to automatically initialize conda?
This will activate conda on startup and change the command prompt when activated.
If you'd prefer that conda's base environment not be activated on startup,
run the following command when conda is activated:
conda config --set auto_activate_base false
You can undo this by running `conda init --reverse $SHELL`
conda create -n dbgpt_env python=3.10
conda activate dbgpt_env
cd /data/DB-GPT/
conda install pytorch
conda install fastapi
conda install python-dotenv
conda install cachetools
cd /data/DB-GPT/
pip install -e ".[default]"
提示:
Running setup.py develop for dbgpt
很快完成
sudo yum install qemu-kvm libvirt libvirt-devel libguestfs-tools virt-install bridge-utils
sudo systemctl start libvirtd
sudo systemctl enable libvirtd
sudo yum install virt-manager
conda install pytorch
报错:
Retrying (Retry(total=0, connect=None, read=None, redirect=None, status=None)) after connection broken
by 'ReadTimeoutError("HTTPSConnectionPool(host='repo.anaconda.com', port=443): Read timed out. (read timeout=9.15)")': /pkgs/main/linux-64/gmpy2-2.1.2-py310heeb90bb_0.conda
解决办法:
conda config --add channels https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/free/
conda config --add channels https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/main/
conda config --set show_channel_urls yes
pip install -e ".[default]"
报错:
Downloading nvidia_cudnn_cu12-8.9.2.26-py3-none-manylinux1_x86_64.whl (731.7 MB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━ 622.3/731.7 MB 489.4 kB/s eta 0:03:44
pip._vendor.urllib3.exceptions.ReadTimeoutError: HTTPSConnectionPool(host='files.pythonhosted.org', port=443): Read timed out.
网络问题,后面加个参数proxy配置即可
pip install -e ".[default]" --proxy http://myproxy:8086
export PYTHONPATH=/data/DB-GPT
export PYTHONPATH=/data/DB-GPT
models
需要下载
cd DB-GPT
mkdir models and cd models
Add the Git LFS package repository
curl -s https://packagecloud.io/install/repositories/github/git-lfs/script.deb.sh | sudo bash
Update your package lists
sudo apt-get update
Install Git LFS
sudo apt install git-lfs
embedding model
git clone https://huggingface.co/GanymedeNil/text2vec-large-chinese
also you can use m3e-large model, you can choose one of them according to your needs
git clone https://huggingface.co/moka-ai/m3e-large
LLM model, if you use openai or Azure or tongyi llm api service, you don't need to download llm model
git clone https://huggingface.co/THUDM/glm-4-9b-chat
apt install rar
rar a -v500m models.rar models/*
scp models.part00*.rar root@192.168.254.194:/data/DB-GPT/
scp root@172.21.254.215:/mnt/bigstore/nfs_zstack/upload_zstack/temp/dbgpt.models/*.rar
环境说明:
model_name: glm-4-9b-chat
model_path: /data/DB-GPT/models/glm-4-9b-chat
========================================
启动DB-GPT的脚本
#!/bin/sh
conda activate dbgpt_env
cd /data/DB-GPT
/data/miniconda3/envs/dbgpt_env/bin/python dbgpt/app/dbgpt_server.py
========================================
db-gpt.service
[Unit]
Description=My Conda Script Service DB-GPT
After=network.target
[Service]
Type=onshot
#simple
ExecStart=/data/startDB-GPT.sh
Restart=on-failure
User=root
Environment="PATH=/usr/bin:/data/miniconda3/condabin:$PATH"
WorkingDirectory=/data/DB-GPT
Environment="CONDA_DEFAULT_ENV=dbgpt_env"
StandardOutput=journal
StandardError=file://data/DB-GPT/error.log
Environment="CONDA_EXE=/data/miniconda3/bin/conda"
Environment="CONDA_PREFIX=/data/miniconda3/envs/dbgpt_env"
Environment="CONDA_PREFIX_1=/data/miniconda3"
Environment="CONDA_PROMPT_MODIFIER='(dbgpt_env) '"
Environment="CONDA_PYTHON_EXE=/data/miniconda3/bin/python"
[Install]
WantedBy=multi-user.target