我刚刚在Windows 10中通过点子安装了Jupyter(Python版本为3.7.2),通过调用jupyter notebook
启动了jupyter服务器,在我的Web浏览器中创建了一个带有内核python3的新笔记本,但是与内核永远无法实现。
在命令行中没有明显的错误:
C:\Users\xxxx>jupyter notebook
[I 21:18:21.005 NotebookApp] Serving notebooks from local directory: C:\Users\xxxx
[I 21:18:21.005 NotebookApp] The Jupyter Notebook is running at:
[I 21:18:21.006 NotebookApp] http://localhost:8888/?token=5743fcbbc805efa4b36c983a7beb63b95cf922957378d64f
[I 21:18:21.007 NotebookApp] Use Control-C to stop this server and shut down all kernels (twice to skip confirmation).
[C 21:18:21.087 NotebookApp]
To access the notebook, open this file in a browser:
file:///C:/Users/xxxx/AppData/Roaming/jupyter/runtime/nbserver-18416-open.html
Or copy and paste one of these URLs:
http://localhost:8888/?token=5743fcbbc805efa4b36c983a7beb63b95cf922957378d64f
[I 21:18:33.777 NotebookApp] Creating new notebook in
[I 21:18:35.512 NotebookApp] Kernel started: ab30c1b3-55d2-45e9-9f01-b4bfc077bb01
[I 21:18:36.883 NotebookApp] Adapting to protocol v5.1 for kernel ab30c1b3-55d2-45e9-9f01-b4bfc077bb01
c:\users\xxxx\appdata\local\programs\python\python37-32\lib\site-packages\notebook\base\zmqhandlers.py:284: RuntimeWarning: coroutine 'WebSocketHandler.get' was never awaited
super(AuthenticatedZMQStreamHandler, self).get(*args, **kwargs)
RuntimeWarning: Enable tracemalloc to get the object allocation traceback
[W 21:18:37.912 NotebookApp] Replacing stale connection: ab30c1b3-55d2-45e9-9f01-b4bfc077bb01:397de5420f9e4bce82d95967bf5fa1eb
[W 21:19:14.026 NotebookApp] 404 GET /static/components/preact/preact.min.js.map (::1) 18.98ms referer=None
[W 21:19:14.040 NotebookApp] 404 GET /static/components/preact-compat/preact-compat.min.js.map (::1) 6.49ms referer=None
[W 21:19:14.050 NotebookApp] 404 GET /static/components/proptypes/index.js.map (::1) 15.99ms referer=None
[I 21:20:36.067 NotebookApp] Saving file at /Untitled1.ipynb
[W 21:22:40.052 NotebookApp] Replacing stale connection: ab30c1b3-55d2-45e9-9f01-b4bfc077bb01:397de5420f9e4bce82d95967bf5fa1eb
我尝试了所有Chrome,Firefox和Microsoft Edge,但均未成功。我什至尝试了PyCharm。
然后我查看了浏览器的js控制台,似乎问题是websocket连接失败:
menubar.js:277 actions jupyter-notebook:find-and-replace does not exist, still binding it in case it will be defined later...
MenuBar.bind_events @ menubar.js:277
utils.js:60 load_extensions Arguments(1)
bidi.js:19 Loaded moment locale en
utils.js:37 Loading extension: jupyter-js-widgets/extension
session.js:54 Session: kernel_created (003a523b-3d40-41ac-b65e-154db92d5528)
kernel.js:459 Starting WebSockets: ws://localhost:8888/api/kernels/f8809bf9-988f-4666-b183-e01bed63fa76
kernel.js:461 WebSocket connection to 'ws://localhost:8888/api/kernels/f8809bf9-988f-4666-b183-e01bed63fa76/channels?session_id=e49c21f065e64e7a89847a0859d689dd' failed: Error during WebSocket handshake: Unexpected response code: 200
Kernel.start_channels @ kernel.js:461
kernel.js:103 Kernel: kernel_disconnected (f8809bf9-988f-4666-b183-e01bed63fa76)
kernel.js:544 WebSocket connection failed: ws://localhost:8888/api/kernels/f8809bf9-988f-4666-b183-e01bed63fa76 true
kernel.js:562 Connection lost, reconnecting in 1 seconds.
kernel.js:103 Kernel: kernel_reconnecting (f8809bf9-988f-4666-b183-e01bed63fa76)
kernel.js:459 Starting WebSockets: ws://localhost:8888/api/kernels/f8809bf9-988f-4666-b183-e01bed63fa76
这可能是什么原因?
答案 0 :(得分:83)
我遇到了同样的问题,并且遵循此vvk2001github https://github.com/jupyter/notebook/issues/4399为我修复了该问题。
卸载龙卷风6并重新安装龙卷风5。
sudo pip3 uninstall tornado
sudo pip3 install tornado==5.1.1
@kaleo也对@Jason Nichols的类似问题给出了答案
答案 1 :(得分:3)
问题在于软件包tornado
6和notebook
5.7.4之间不兼容。 notebook
was updated将于2019年3月6日解决此问题。目前,解决此问题的最佳方法如下:
$ pip install -U jupyter
$ pipenv update jupyter
答案 2 :(得分:2)
Jupyter Notebook 5.7.5更新应解决龙卷风6的问题。请参见tweeted公告和Jupyter团队的github release tag。
从03-06-19开始,docs和conda distribution都处于待更改状态。但是,您可以update using pip
。
从文档中升级为:
> pip install notebook --upgrade
验证版本:
> pip --version
答案 3 :(得分:0)
通过使用以下命令重置Jupyter Notebook配置文件,以重置jupyter_notebook_config.py
的默认配置
jupyter notebook --generate-config
答案 4 :(得分:0)
不要降级龙卷风。相反,请强制重新安装jupyter。
pip3 install --force-reinstall upgrade jupyter
答案 5 :(得分:0)
在安装TensorFlow.js时会发生此问题。
在安装TensorFlow.js时,还会安装prompt-toolkit-1.0.14
,这会引起问题。
在安装TensorFlow.js之后,您只需安装prompt-toolkit-3.0.5
即可通过在命令提示符下运行以下代码来解决此问题:
pip install prompt-toolkit==3.0.5