GVKun编程网logo

docker使用Dockerfile python:3.6-alpine配置tensorflow、numpy和Pillow运行环境(docker run python)

5

关于docker使用Dockerfilepython:3.6-alpine配置tensorflow、numpy和Pillow运行环境和dockerrunpython的问题就给大家分享到这里,感谢你花时

关于docker使用Dockerfile python:3.6-alpine配置tensorflow、numpy和Pillow运行环境docker run python的问题就给大家分享到这里,感谢你花时间阅读本站内容,更多关于CMake 不断从 cygwin python 中获取 Python,如何从 Windows 安装的 Python 中获取、Core Python | 2 - Core Python: Getting Started | 2.5 - Modularity | 2.5.5 - The Python Execution Mod、Error: Can‘t find Python executable “python“, you can set the PYTHON env variable、ERROR: Command "python setup.py egg_info" python-nss等相关知识的信息别忘了在本站进行查找喔。

本文目录一览:

docker使用Dockerfile python:3.6-alpine配置tensorflow、numpy和Pillow运行环境(docker run python)

docker使用Dockerfile python:3.6-alpine配置tensorflow、numpy和Pillow运行环境(docker run python)

大概在整个互联网上搜索了一天,浏览了不下100个搜索页,尝试了无数种方法,如
1、 FROM python:3.6-alpine改成FROM python:3.6-slim或者FROM ubuntu
这种方式最终都没法安装上tensorflow==1.15.0,始终显示找不到这个包的任何版本,最终也没解决

于是后来就想,怎么把tensorflow整合进去,于是又尝试在Dockerfile里使用两个FROM,不过由于自己对docker的了解实在有限,没办法,参考了这个https://www.cnblogs.com/leoyang63/articles/13733967.html,才知道自己的整合方式是错的,如下


FROM tensorflow/tensorflow

FROM python:3.6-alpine

LABEL maintainer="xxx <xxx@qq.com>"

# 这些都是后来加的,因为numpy和Pillow安装不上,才知道需要这些依赖包
#RUN echo -e http://dl-cdn.alpinelinux.org/alpine/v3.13/main > /etc/apk/repositories
#RUN cat /etc/apk/repositories
RUN apk --update add --no-cache libxml2-dev libxslt-dev libffi-dev gcc musl-dev libgcc openssl-dev curl jpeg-dev zlib-dev freetype-dev lcms2-dev openjpeg-dev tiff-dev tk-dev tcl-dev

copY . /detectweb

workdir /detectweb

RUN pip install -r requirements.txt -i https://mirror.baidu.com/pypi/simple && pip install gunicorn&&chmod 755 run_server.sh

EXPOSE 8080

ENTRYPOINT [ "./run_server.sh" ]

后来也尝试使用docker commit等来进行整合,最终还是失败了

2、又尝试直接放弃alpine,使用Ubuntu

FROM ubuntu

RUN apt update \
        && apt install -y python3-pip python3-dev vim \
        && pip3 install --upgrade pip \
        && pip3 install tensorflow \
        && ln -s /usr/bin/python3 /usr/bin/python \
        && ln -s /usr/bin/pip3 /usr/bin/pip

LABEL maintainer="xxx <xx@qq.com>"

copY . /detectweb

workdir /detectweb

RUN pip install -r requirements.txt -i https://mirror.baidu.com/pypi/simple && pip install gunicorn&&chmod 755 run_server.sh

EXPOSE 8080

ENTRYPOINT [ "./run_server.sh" ]

这样好不容易tensorflow终于安装成功了,docker 的image也构建好了,结果一运行,发现报错
TypeError: required field "type_ignores" missing from Module
原因是python是3.8的,和werkzeug包冲突了,真是吐了

没办法,又想办法降低版本,在Dockerfile里捣鼓了很久,如

# 这里还添加了各种依赖、工具
#RUN apt-get update
#RUN apt-get install -y software-properties-common
#RUN add-apt-repository ppa:jonathonf/python-3.6
RUN apt update \
        && apt install -y python3.6 python3.6-pip python3.6-dev vim \
        && pip3 install --upgrade pip \
        && pip3 install tensorflow \
        && ln -s /usr/bin/python3 /usr/bin/python \
        && ln -s /usr/bin/pip3 /usr/bin/pip

当然、最终还是没有把python版本降下来,但到这里已经很接近了,至少tensorflow和Pillow包这些没问题了

3、最后一种方法,直接干脆把ubuntu换了,换成python:3.6,连alpine都不要了。。就成功了,项目也成功部署了,吐了,就这?干了一天md

FROM python:3.6

LABEL maintainer="xxx <xx@qq.com>"

copY . /detectweb

workdir /detectweb

RUN pip install -r requirements.txt -i https://mirror.baidu.com/pypi/simple && pip install gunicorn&&chmod 755 run_server.sh

EXPOSE 8080

ENTRYPOINT [ "./run_server.sh" ]

CMake 不断从 cygwin python 中获取 Python,如何从 Windows 安装的 Python 中获取

CMake 不断从 cygwin python 中获取 Python,如何从 Windows 安装的 Python 中获取

如何解决CMake 不断从 cygwin python 中获取 Python,如何从 Windows 安装的 Python 中获取

我有一个看起来像这样的 CMake 脚本:

  1. find_program(PYTHON_COMMAND NAMES python3 python)

问题是它检测到安装在 Cygwin 安装中的 python。 输出总是:

  1. -- PYTHON_PATH:C:/cygwin64/bin/python3

我希望它取自:

  1. c:\\python36-64\\python

在windows PATH变量中,Cygwin bin在路径的最后一个,windows安装在第一个 但它只检测到 Cygwin python,
怎么改?

Core Python | 2 - Core Python: Getting Started | 2.5 - Modularity | 2.5.5 - The Python Execution Mod

Core Python | 2 - Core Python: Getting Started | 2.5 - Modularity | 2.5.5 - The Python Execution Mod

It's important to understand the Python execution model and precisely when function deFinitions and other important events occur when a module is imported or executed. Here, we show execution of our Python module as it's imported in a graphical debugging environment. We step through the top‑level statements in the module. What's important to realize here is that the def used for the fetch_words function isn't merely a declaration. It's actually a statement, which when executed in sequence with the other top‑level model scope code, causes the code within the function to be bound to the name of the function. When modules are imported or run, all of the top‑level statements are run, and this is the means by which the function within the module namespace are defined. We are sometimes asked about the difference between Python modules, Python scripts, and Python programs. Any .py file constitutes a Python module. But as we've seen, modules can be written for convenient import, convenient execution, or using the if dunder name = dunder main idiom, both. We strongly recommend making even simple scripts importable since it eases development and testing so much if you can access your code from within the REPL. Likewise, even modules, which are only ever meant to be imported in production settings, benefit from having executable test code. For this reason, nearly all modules we create have this form of defining one or more importable functions with a postscript to facilitate execution. Whether you consider a module to be a Python script or Python program is a matter of context and usage. It's certainly wrong to consider Python to be merely a scripting tool, in the vein of Windows batch files or UNIX Shell scripts, as many large and complex applications are built exclusively with python.

 

  • def不仅仅是一个declaration声明,更是一条statement语句。它将其中的python代码于函数名绑定在一起
  • 一个py文件就是一个模块,这个模块包含类或函数。你写python,要尽量将代码包装成函数和类,方便各种import
  • 一个py文件也可看作是一个脚本,在系统命令行中运行
  • python不仅仅是脚本语言,很多大型程序都是用python构建的

Error: Can‘t find Python executable “python“, you can set the PYTHON env variable

Error: Can‘t find Python executable “python“, you can set the PYTHON env variable

在启动vue项目的时候,安装node.js组件node-sass过程中报错了,错误提示如下
Error: Can’t find Python executable “python”, you can set the PYTHON env variable

由错误提示可知:Node.js 在安装模块组件node-sass的时候,node.js缺少Visual Studio2015 Build Tools相关的组件和python的环境,如果安装了vs2015组件的小伙伴们就不用安装Visual Studio2015 Build Tools相应的组件,只用安装python2.7即可解决缺少的python组件的问题。

欲安装python2.7,请至python官网:www.python.org 下载,然后配置好python的环境变量即可。

不过博主我并不推荐上述的解决方案,因为对于程序员来说,效率第一,上述的问题一个命令就可以轻松解决你所遇到的麻烦,前面说了那么多,无非就是想告诉在看本篇博客的同仁们放下浮躁的心,遇到问题首先不是急着去解决问题,而是分析为什么会这样,然后才能水到聚成的去找到解决问题的方法。

运行下面这个命令即可解决你们遇到的Error问题

npm install --global --production windows-build-tools

:上面讲述了一堆就是为了讲述此命令是干嘛的,上面已经描述很详细了,就不再赘述了,该操作与上述的一堆操作无异,效果却是一样的。

然后运气不好的小伙伴可能接着会遇到一个坑,那就是执行了:npm install --global --production windows-build-tools这个命令的人细心点会发现执行到一半就卡住了,这个卡住了没有红字重点提示,而且下方还有英文在等待中,粗心的小伙伴可能以为是命令执行完了,组件安装好了,其实不然,我这边已经解决了,就无法复现了,具体点就是中文的提示,提示我们由于有类似组件在运行或者下载导致无法继续下载安装组件了。稳妥点的解决办法是,将电脑重启,将底层正在运行的模块干掉,待电脑重启后再执行npm install --global --production windows-build-tools这条命令即可,博主我就是这样解决的,稳稳的幸福就会浮现在你面前如下图所示,你的可能和我不一样,因为我已经跑成功过一次了,没有你的那么多细节的log打印。
在这里插入图片描述

然后就是在你的项目下shift+鼠标右击你的项目运行npm run dev即可启动vue项目了。

ERROR: Command

ERROR: Command "python setup.py egg_info" python-nss

[root@localhost ~]# pip install python-nss

DEPRECATION: Python 2.7 will reach the end of its life on January 1st, 2020. Please upgrade your Python as Python 2.7 won''t be maintained after that date. A future version of pip will drop support for Python 2.7.
Looking in indexes: http://pypi.douban.com/simple
Collecting python-nss
  Downloading http://pypi.doubanio.com/packages/6b/29/629098e34951c358b1f04f13a70b3590eb0cf2df817d945bd05c4169d71b/python-nss-1.0.1.tar.bz2 (222kB)
     |████████████████████████████████| 225kB 31kB/s 
    ERROR: Complete output from command python setup.py egg_info:
    ERROR: Traceback (most recent call last):
      File "<string>", line 1, in <module>
      File "/tmp/pip-install-JGnrT5/python-nss/setup.py", line 409, in <module>
        sys.exit(main(sys.argv))
      File "/tmp/pip-install-JGnrT5/python-nss/setup.py", line 333, in main
        nss_include_dir  = find_include_dir([''nss3'', ''nss''],   [''nss.h'',  ''pk11pub.h''], include_roots=include_roots)
      File "/tmp/pip-install-JGnrT5/python-nss/setup.py", line 94, in find_include_dir
        raise ValueError("unable to locate include directory containing header files %s" % include_files)

    ValueError: unable to locate include directory containing header files [''nss.h'', ''pk11pub.h'']

ERROR: Command "python setup.py egg_info" failed with error code 1 in /tmp/pip-install-JGnrT5/python-nss/


查看错误日志缺少头文件

进入python-nss官网,写着To build python-nss you the C language header files and libraries for both NSPR and NSS will need to be installed. This is system and distribution specific, as such we cannot give you explicit instructions. On Linux typically these packages are called:

  • nss-devel
  • nspr-devel

yum install nss-devel -y

yum install nspr-devel -y

关于docker使用Dockerfile python:3.6-alpine配置tensorflow、numpy和Pillow运行环境docker run python的问题我们已经讲解完毕,感谢您的阅读,如果还想了解更多关于CMake 不断从 cygwin python 中获取 Python,如何从 Windows 安装的 Python 中获取、Core Python | 2 - Core Python: Getting Started | 2.5 - Modularity | 2.5.5 - The Python Execution Mod、Error: Can‘t find Python executable “python“, you can set the PYTHON env variable、ERROR: Command "python setup.py egg_info" python-nss等相关内容,可以在本站寻找。

本文标签: