vscode jupyter笔记本正在升级

介绍 (Introduction)

I’ve always wanted to use Visual Studio Code (VSCode) as my daily driver for everything, including iPython Notebooks. Don’t get me wrong: JupyterLab is awesome to use for data science purposes (Jupyter Notebooks as well), but not so much for regular code scripting, not to mention Jupyter only (realistically) supports Python, Julia, or a few other programming languages.

我一直想将Visual Studio Code(VSCode)用作所有驱动程序的日常驱动程序，包括iPython Notebooks 。 不要误会我的意思：JupyterLab非常适合用于数据科学目的(也适用于Jupyter Notebook)，但对于常规代码脚本而言却不是那么多，更不用说Jupyter了(实际上)仅支持Python，Julia或其他一些编程语言。

When VSCode — or specifically the VSCode Python extension team — first announced Notebooks support in 2019, I had to try it out. In the end, apart from covering all the keybindings and features of Jupyter Notebooks, it accomplished some notable feats:

VSCode时-或特别是VSCode Python扩展团队时 -在2019年首次宣布对笔记本电脑的支持，我不得不尝试一下。 最后，除了涵盖Jupyter笔记本的所有键绑定和功能之外，它还实现了一些显着的壮举：

• Powerful variable explorer and plot viewer

  强大的变量浏览 器和绘图查看器

• Integrating VSCode’s IntelliSense autocomplete into code cells

  将VSCode的IntelliSense自动完成功能集成到代码单元中

But, I found it a bit disappointing at other fronts:

但是，我在其他方面感到有些失望：

• It’s painfully slow to launch, especially through WSL or SSH

  启动非常缓慢 ，尤其是通过WSL或SSH时

• The interface was awkward and felt half-baked

  界面很尴尬 ，半生不熟

• The Notebook session is irrecoverable after closing it out in VSCode, unlike JupyterLab

  与JupyterLab不同，Notebook 会话在VSCode中关闭后无法恢复

Nonetheless, introducing Jupyter Notebook integration was a huge step forward for VSCode in the Python data science realm. And with the team recently announcing major updates for VSCode Jupyter Notebooks, it seems there will be another huge step in the near future!

尽管如此，引入Jupyter Notebook集成还是VSCode在Python数据科学领域迈出的一大步。 并且随着团队最近宣布VSCode Jupyter Notebook的重大更新，似乎在不久的将来将会迈出又一大步！

什么是新的？ (What’s New?)

Let’s get a sample Jupyter Notebook to view these new changes first. I’ll be using a Notebook from one of my earlier articles on Predicting League of Legends Matches with PyTorch since it has a good amount of code, a pandas DataFrame, and a few graphs — a nice example for a real-world scenario.

让我们获取一个示例Jupyter Notebook，以首先查看这些新更改。 我将使用之前的一篇有关使用PyTorch预测英雄联盟比赛的文章中的笔记本 因为它具有大量的代码，pandas DataFrame和一些图形 -这是真实场景的一个很好的示例。

You can take a look at the notebook here. The complimentary article is available to read below 👇.

您可以在