the Jupyter Extension for VS Code

原文：Introducing the Jupyter Extension for VS Code – Python (microsoft.com)

Jupyter Notebook enables creating and sharing documents that contain live code, equations, text, and visualizations, and is the de facto data science tool for its simplicity and interactivity. At Microsoft we’re all in to embrace its power. The Python extension for VS Code has shipped with Jupyter Notebook support for over a year with growing popularity. We understand that Jupyter Notebooks are also used with other programming languages beyond Python, such as R, Julia, and Scala. To enable the same rich Jupyter Notebook experience for other languages, we’ve refactored the Jupyter support out of the Python extension and into the Jupyter extension that we’re shipping today. This makes it much easier to build new Jupyter experiences for languages beyond Python by taking a dependency on the Jupyter extension, which itself has no dependency on the Python runtime or the Python extension. For our existing VS Code Python Notebooks users, thank you again for all your feedback that helps shape our investments, there’s no change in the experience for you as the Jupyter extension comes installed with the Python extension.

The Jupyter extension provides basic notebook support for any language kernel that is supported in Jupyter Notebooks today. Many language kernels will work with no modification. To enable advanced features such as full IntelliSense and debugging, there may be modifications needed in the VS Code language extensions. These will be documented as the project evolves. The screenshot below shows viewing and running a Julia Notebook in VS Code using the Jupyter extension.

The code for the new Jupyter extension lives in the GitHub repo under the MIT license. This work is rapidly evolving. We’ll continue to polish the extension in the coming weeks and months, and we look forward to collaborating with the Jupyter community to refine and evolve this extension to support a variety of languages and scenarios.

Try it out!

• To work with Python Jupyter Notebooks in VS Code, simply install or update the Python extension. There’s no need to install the Jupyter extension separately. For those who are using the VS Code Insiders build, you may notice that the new preview notebooks experience that was first introduced in July has now been turned on by default.
• To use the Jupyter extension for notebooks other than Python, install the VS Code Insiders build, the Jupyter extension, the Jupyter kernel of your language (e.g. kernel for the Julia language), and start playing around!

原文：Setting Up Run by Line and Debugging for Notebooks · microsoft/vscode-jupyter Wiki · GitHub

The Run by Line and Debugging features for Python notebooks requires ipykernel v6.0.0 or greater to be installed in the notebook’s kernel.

Anaconda

1. Use Anaconda Navigator or an Anaconda prompt to install ipykernel into your desired notebook environment.
2. Close and reopen VS Code and your desired notebook. The Developer: Reload Window command works well for this.

Pip

1. Open a Python terminal and activate your desired notebook environment.
2. Run pip install -U ipykernel
3. Close and reopen VS Code and your desired notebook. The Developer: Reload Window command works well for this.

conda、miniconda、anaconda、仓库的详解 | 霍小强博客 (huoxiaoqiang.com)

conda是 由Anaconda®公司用python语言开发维护的一个支持安装并管理python多版本、包管理、包依赖管理、虚拟环境管理四大功能于一身的开源免费工具，支持任何编程语言：Python, R, Ruby, Lua, Scala, Java, JavaScript, C/ C++, FORTRAN等。

miniconda 是 由Anaconda®公司开发维护的conda的免费最小安装程序。它是 Anaconda 的轻量级替代，仅包含 conda、python、它们所依赖的包以及少量其它有用的包，包括 pip、zlib 和其它一些包。

Anaconda 是 由Anaconda®公司开发维护的conda的安装程序，分为个人版、商业版、团队版、企业版，除了提供miniconda内容以外，各大版本还包含大量科学计算方面的包以及其他服务。个人版免费，商业版、团队版、企业版均为付费产品。

官方包仓库

repo.anaconda.com/pkgs/是 由Anaconda®公司自己构建的包的公开仓库，您仅仅具有包的使用权。此仓库仅可用于个人、教育机构学习用，如果用于商用需付费购买商业产品。

conda-forge

conda-forge是一个 GitHub 组织，包含 conda 包的存储库。构建的发行版上传到anaconda.org/conda-forge 并且可以使用 conda 安装。conda-forge、miniforge由社区驱动，开源免费。

miniforge致力于提供除了类似 miniconda 的安装程序，并配置了 conda-forge 是默认(且唯一)频道。

miniforge与miniconda的区别：

• miniforge是开源社区驱动，以防miniconda的开发者Anaconda®公司可能未来存在的法律风险。
• miniforge默认(且唯一)conda-forge仓库，miniconda除了提供不可商用的官方仓库外，也可以添加conda-forge仓库。
• miniforge支持Mamba(conda的C++实现版本)。
• miniforge支持各种 CPU 架构（x86_64、ppc64le 和 aarch64，包括 Apple M1）。

为什么要使用conda-forge？

来自Anaconda, Inc.的 conda 团队打包了大量软件包，并在其default频道中免费提供给所有用户。

但是如果您要查找的包不在默认频道中怎么办？过去，用户只能选择创建Anaconda Cloud帐户并创建自己的频道。

这带来了一系列缺点：

• 由于包裹分散在许多频道中，因此很难定位包。
• 由于二进制不兼容，跨频道组合包并不总是可行的。
• 包仅适用于开发人员感兴趣或可以访问的架构。
• 频道经常被放弃，更新需要定位新频道。

conda-forge 是解决这些问题的社区努力：

• 所有包都在名为 conda-forge 的单个频道中共享。
• 注意所有软件包都是最新的。
• 通用标准确保所有包都有兼容的版本。
• 默认情况下，我们为 macOS、Linux AMD64 和 Windows AMD64 构建包。
• 许多软件包由多个维护者更新，您可以轻松选择成为维护者。
• 一个活跃的核心开发团队也在努力维护废弃的包。

原文地址：http://www.cnblogs.com/panpanwelcome/p/16909542.html