fengsim
commented
5 months ago you should use linux, try ubuntu, it is very easy
Open stuffren opened 8 months ago
fengsim
commented
5 months ago you should use linux, try ubuntu, it is very easy
📜 Description
[zh：] 在执行到第10步时，这一步已经成功下载了很多依赖项，但最终还是有报错，这真的令人沮丧！！！ 为了成功安装这一步，尝试了很多次，我花了12个小时左右的时间，我甚至通过V-P-N连接国际互联网，网速高达50M/s，但还是有报错，这是为什么？ 下面是在执行 ./run-with-docker-compose.sh命令后，对应项成功构建的提示，以及对应项报错的提示：
[en:] On executing to step 10, this step has successfully downloaded many dependencies, but still ends up reporting errors, which is really frustrating!!!! It took many attempts to successfully install this step, I spent about 12 hours on it, I even connected to the international internet via V-P-N with an internet speed of up to 50M/s, but there are still errors reported, why is that? Here's what happens when executing . /run-with-docker-compose.sh command, the prompt that corresponds to the successful build of the item, and the prompt that corresponds to the error reported:
renhao@XTZJ-20231226DP:/var/www/docsGPT$ ./run-with-docker-compose.sh Running Plain Configuration [+] Building 1972.4s (17/27) docker:default => [worker internal] load build definition from Dockerfile 0.0s => => transferring dockerfile: 997B 0.0s => [backend internal] load build definition from Dockerfile 0.0s => => transferring dockerfile: 997B 0.0s => [worker internal] load metadata for docker.io/library/python:3.11-slim-bullseye 4.9s => [worker internal] load .dockerignore 0.1s => => transferring context: 2B 0.0s => [backend internal] load .dockerignore 0.1s => => transferring context: 2B 0.0s => CACHED [worker stage-1 1/5] FROM docker.io/library/python:3.11-slim-bullseye@sha256:6988e45203d29 0.0s => [backend internal] load build context 0.0s => => transferring context: 3.17kB 0.0s => [worker internal] load build context 0.1s => => transferring context: 3.17kB 0.0s => CACHED [backend builder 2/10] RUN apt-get update && apt-get install -y gcc curl 0.0s => CACHED [backend builder 3/10] RUN apt-get install -y wget unzip 0.0s => CACHED [backend builder 4/10] RUN wget https://d3dg1063dc54p9.cloudfront.net/models/embeddings/m 0.0s => CACHED [backend builder 5/10] RUN unzip mpnet-base-v2.zip -d model 0.0s => CACHED [backend builder 6/10] RUN rm mpnet-base-v2.zip 0.0s => CACHED [backend builder 7/10] RUN curl https://sh.rustup.rs -sSf | sh -s -- -y && apt-get instal 0.0s => CACHED [backend builder 8/10] RUN pip install --upgrade pip && pip install tiktoken==0.5.2 0.0s => CACHED [backend builder 9/10] COPY requirements.txt . 0.0s => ERROR [backend builder 10/10] RUN pip install -r requirements.txt 1967.1s
My system environment:
---wsl2 ---. WSL version: 2.1.5.0 Kernel version: 5.15.146.1-2 WSLg version: 1.0.60 MSRDC Version: 1.2.5105 Direct3D version: 1.611.1-81528511 DXCore version: 10.0.25131.1002-220531-1700.rs-onecore-base2-hyp Windows version: 10.0.19044.3803
----python verson---- Python 3.10.12 -----docker version----- Docker version 25.0.3, build 4debf41
DocsGPT is too complicated to install! Not as good as ollama! Why don't you guys make its installation steps as easy as using ollama?
👟 Reproduction steps
[en:] 1.Download the DocsGPT project file. Since it is already downloaded in Win10, copy it from Win10 to the project directory of wsl2 and execute the following command wslpath 'J:\Common packages\AI software\docsGPT\DocsGPT-0.8.0.zip' //Convert Windows path: Use the wslpath command to convert the Windows path to a Linux path that WSL 2 can understand. cp $(wslpath 'J:\Common Packages\AI Software\docsGPT\DocsGPT-0.8.0.zip') /var/www/ //copy files: Now you can use the cp command to copy files from the Windows filesystem to the Linux filesystem of WSL 2.
Unzip to the current directory sudo unzip DocsGPT-0.8.0.zip
Delete the zip file sudo rm -rf DocsGPT-0.8.0.zip
Modify the name of the project file sudo mv DocsGPT-0.8.0 docsGPT
Enter docsGPT cd docsGPT
Enter the super administrator account and set the api key. sudo -i //enter super administrator account sudo echo "API_KEY=my-openai-key" > .env echo "VITE_API_STREAMING=true" >> .env
//Exit root exit
Check if the .env file is set up successfully ls -a To view all files and directories, including hidden files, you can use the ls -a command. The -a option lists all files and directories, including hidden files that begin with a dot.
Run the following command to start the installer with Docker Compose: . /run-with-docker-compose.sh This step, originally, also reported an error when installing the PIP, but I added a third-party docker image source that includes multiple Chinese and foreign sources, which solved the problem But the multiple errors reported in this 10th step, which I tried about 10 times or so, has made me lose confidence. Because I know in this case, it shouldn't be a user problem, but a problem caused by the software designers making the installation steps so complicated!
=> ERROR [backend builder 10/10] RUN pip install -r requirements.txt 1967.1s This is the step, strangely enough, it successfully installs all the dependencies contained in requirements.txt, but in addition to that it automatically installs a lot of other dependencies, including nvidia and such, and I can't figure out why this is so, it's just an error when installing these additional dependencies!
[cn:] 1.下载DocsGPT项目文件 由于已经在win10下载了，从win10复制到wsl2的项目目录中，执行以下命令 wslpath 'J:\常用软件包\AI软件\docsGPT\DocsGPT-0.8.0.zip' //转换Windows路径：使用wslpath命令将Windows路径转换为WSL 2可以理解的Linux路径。 cp $(wslpath 'J:\常用软件包\AI软件\docsGPT\DocsGPT-0.8.0.zip') /var/www/ //复制文件：现在你可以使用cp命令将文件从Windows文件系统复制到WSL 2的Linux文件系统中。
2.解压缩到当前目录 sudo unzip DocsGPT-0.8.0.zip
3.删除压缩包文件 sudo rm -rf DocsGPT-0.8.0.zip
修改项目文件名称 sudo mv DocsGPT-0.8.0 docsGPT
进入docsGPT cd docsGPT
进入超级管理员账户，设置api key sudo -i //进入超级管理员账户 sudo echo "API_KEY=my-openai-key" > .env echo "VITE_API_STREAMING=true" >> .env
//退出root exit
8.运行以下命令以使用 Docker Compose 启动安装程序： ./run-with-docker-compose.sh 这一步，本来在安装PIP时，也报错，但我添加了包含多个中国和国外的第三方docker镜像源，解决了 但这第10步的多次报错，我尝试了大概10次左右，已经让我失去了信心。 因为我知道这种情况，应该不是用户的问题，而是软件设计者把安装步骤搞的如此复杂而造成的问题！！
=> ERROR [backend builder 10/10] RUN pip install -r requirements.txt 1967.1s 就是这一步，很奇怪的是，它成功安装了requirements.txt所包含的全部依赖项，但除此之外它还额外在自动安装很多包括nvidia之类的其他依赖项，我搞不懂这是为什么，就是在安装这些额外的依赖项时，报出的错误！！
👍 Expected behavior
Seriously, I was pleasantly surprised when I first found DocsGPT, but it disappointed during the installation process! Hopefully it will be as quick to install as ollama! ollamag officially offers a lot of big models like gemini, ollama 2, most of them 2Gigabyte-8Gigabyte in size, but the download of these models is quite fast, it's done in about 10 minutes!!!!
👎 Actual Behavior with Screenshots
下面是屏幕截图：
The middle part, all downloaded successfully, so it was omitted!
💻 Operating system
Windows
What browsers are you seeing the problem on?
Chrome
🤖 What development environment are you experiencing this bug on?
Docker
🔒 Did you set the correct environment variables in the right path? List the environment variable names (not values please!)
My .evn file is in the project root directory and is set as follows: API_KEY=my-openai-key VITE_API_STREAMING=true LLM_NAME=openai
📃 Provide any additional context for the Bug.
no
📖 Relevant log output
👀 Have you spent some time to check if this bug has been raised before?
🔗 Are you willing to submit PR?
None
🧑⚖️ Code of Conduct