使用Python SDK启动Jupyter笔记本时遇到问题 [英] Problem starting Jupyter notebook using the Python SDK

问题描述

我一直按照https://docs.microsoft.com/en-us/azure/machine-learning/service/setup-create-workspace#portal上的说明设置Azure ML服务工作区。 进入门户网站并创建资源非常简单。 
我安装了Python SDK（使用Anaconda而不是miniConda - 都支持）。 我在"使用SDK创建工作区"中遇到了障碍。 在此过程中，我应该能够打开一个包含
Azure ML SDK的Jupyter笔记本。   但是，不是在浏览器中显示的普通Jupyter笔记本界面，我在URL localhost：8888 / tree上得到一个空白页面。 这就好像Azure特定的本地Jupyter服务器无法正常服务。

 

我会注意到，如果我关闭它并重新启动我的正常Jupyter设置开始菜单，工作正常。 不幸的是，它无法访问我刚为Azure设置的备用环境。

I have been following the instructions at https://docs.microsoft.com/en-us/azure/machine-learning/service/setup-create-workspace#portal to set up an Azure ML service workspace.  Going into the portal and creating a resource was straight forward.  I installed the Python SDK (using Anaconda instead of miniConda – both are supported).  I’m hitting a blockage at "Create a workspace with the SDK".  At that point in the process, I should be able to open a Jupyter notebook which incorporates the Azure ML SDK.    But instead of the normal Jupyter notebook interface showing up in the browser, I’m getting a blank page at URL localhost:8888/tree.  It’s as if the Azure specific local Jupyter server isn’t serving correctly.
 
I’ll note that if I shut this down and restart my normal Jupyter setup from the start menu, that works ok.  Unfortunately, it doesn’t have access to the alternate environment I’ve just set up for Azure.

推荐答案

Hello Mike，

Hello Mike,

我也能够重现你所看到的问题。在localhost上启动笔记本时显示空白页：8888 / tree

I was also able to reproduce the issue you have seen. A blank page shows up when the notebook is launched on localhost:8888/tree

在进一步检查时，我尝试以管理员身份启动anaconda提示，然后运行以下命令启动笔记本一个不同的端口：

On checking further I tried to launch the anaconda prompt as a administrator and then ran the following command to launch notebook on a different port:

jupyter notebook --port=4000 --log-level='DEBUG'

此后笔记本电脑可以访问，我能够成功运行单元格中的代码。

After this the notebook was accessible and I am able to run the code in cells successfully.

希望这会有所帮助!!

Hope this helps!!

这篇关于使用Python SDK启动Jupyter笔记本时遇到问题的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！