【问题标题】:Jupyter Nbextensions work in some of my notebooks but not othersJupyter Nbextensions 在我的一些笔记本中有效,但在其他笔记本中无效
【发布时间】:2017-09-27 03:27:14
【问题描述】:

我越来越喜欢 Jupyter Nbextensions。它们似乎可以在我创建的任何新笔记本中使用,但它们从未在我的一些旧笔记本中使用。回顾 Jupyter 启动时的喷吐,我看不到任何可能暗示原因的东西,但也许其他人可以帮助我找到问题的根源。这是我打开一个没有出现 nbextensions 的笔记本时出现的问题:

G:\DATA\EDUCATION\Computer Science\Machine Learning\Projects - Machine Learning\iPython (Machine Learning Tutorial)>jupyter notebook
[I 13:39:52.478 NotebookApp] [nb_conda_kernels] enabled, 6 kernels found
[I 13:39:53.312 NotebookApp] [jupyter_nbextensions_configurator] enabled 0.2.5
[I 13:39:53.486 NotebookApp] [nb_anacondacloud] enabled
[I 13:39:53.534 NotebookApp] [nb_conda] enabled
[I 13:39:54.180 NotebookApp] \u2713 nbpresent HTML export ENABLED
[W 13:39:54.180 NotebookApp] \u2717 nbpresent PDF export DISABLED: No module named 'nbbrowserpdf'
[I 13:39:54.396 NotebookApp] Serving notebooks from local directory: G:\DATA\EDUCATION\Computer Science\Machine Learning\Projects - Machine Learning\iPython (Machine Learning Tutorial)
[I 13:39:54.396 NotebookApp] 0 active kernels
[I 13:39:54.396 NotebookApp] The Jupyter Notebook is running at: http://localhost:8888/?token=a32e730ddd80639a9d1ba52af82606856b2f028a0f2ea12a
[I 13:39:54.396 NotebookApp] Use Control-C to stop this server and shut down all kernels (twice to skip confirmation).
[C 13:39:54.411 NotebookApp]

    Copy/paste this URL into your browser when you connect for the first time,
    to login with a token:
        http://localhost:8888/?token=a32e730ddd80639a9d1ba52af82606856b2f028a0f2ea12a
[I 13:39:54.662 NotebookApp] Accepting one-time-token-authenticated connection from ::1
[W 13:40:06.038 NotebookApp] Notebook NumPy (ver. 2.0 2017-09-26).ipynb is not trusted
[I 13:40:47.014 NotebookApp] Kernel started: eb66a861-e1e0-426e-81ba-eaacdfd4c47e
[W 13:40:47.158 NotebookApp] 404 GET /notebooks/images/Frobenius%20Norm.png (::1): No such file or directory: images/Frobenius Norm.png
[W 13:40:47.174 NotebookApp] 404 GET /notebooks/images/Frobenius%20Norm.png (::1) 27.02ms referer=http://localhost:8888/notebooks/NumPy%20(ver.%202.0%202017-09-26).ipynb
[W 13:40:52.083 NotebookApp] 404 GET /nbextensions/widgets/notebook/js/extension.js?v=20170926133952 (::1) 2.00ms referer=http://localhost:8888/notebooks/NumPy%20(ver.%202.0%202017-09-26).ipynb
[I 13:40:53.787 NotebookApp] Adapting to protocol v5.1 for kernel eb66a861-e1e0-426e-81ba-eaacdfd4c47e
[I 13:41:50.017 NotebookApp] Kernel shutdown: eb66a861-e1e0-426e-81ba-eaacdfd4c47e
[I 13:41:51.648 NotebookApp] Kernel started: f32da744-5615-4579-b636-2183b4712eb8
[I 13:42:47.175 NotebookApp] Saving file at /NumPy (ver. 2.0 2017-09-26).ipynb
[W 13:42:47.189 NotebookApp] Saving untrusted notebook NumPy (ver. 2.0 2017-09-26).ipynb
[I 13:45:04.715 NotebookApp] Saving file at /NumPy (ver. 2.0 2017-09-26).ipynb
[W 13:45:48.451 NotebookApp] 404 GET /notebooks/images/Frobenius%20Norm.png (::1): No such file or directory: images/Frobenius Norm.png
[W 13:45:48.453 NotebookApp] 404 GET /notebooks/images/Frobenius%20Norm.png (::1) 3.00ms referer=http://localhost:8888/notebooks/NumPy%20(ver.%202.0%202017-09-26).ipynb
[W 13:45:50.924 NotebookApp] 404 GET /nbextensions/widgets/notebook/js/extension.js?v=20170926133952 (::1) 2.00ms referer=http://localhost:8888/notebooks/NumPy%20(ver.%202.0%202017-09-26).ipynb

当我打开一个使用 nbextensions 的笔记本时,我看到了额外的几行代码:

[W 13:57:27.521 NotebookApp] Notebook NLP/LSTM/Sentiment Analysis with LSTM/Experiments with Sentiment Analysis.ipynb is not trusted
[I 13:57:39.619 NotebookApp] Kernel started: 73225ddd-b5cd-46bb-80ec-b9841765357b

【问题讨论】:

标签: jupyter-notebook jupyter


【解决方案1】:

正如我所怀疑的,我在上面的评论中引用的the Github article 提供了答案。我的旧笔记本中有很多 LaTex 降价,而且 MathJax 在加载时与 Nbextensions 交战。显然,由于所有 Latex 降价导致加载时间延长,Nbextensions 正在超时。本文指出的唯一解决方案是增加./jupyter/custom/custom.js 中的超时间隔。就我而言,我不得不将它增加 4 到 120 秒:

window.requirejs.config({
    waitseconds: 120,  // default is 30s
});

【讨论】:

  • 跟进评论:我在上面列出的解决方案有帮助,但并不能完全解决问题。随着我的笔记本尺寸增大,它们总是会达到无法加载 nbextensions 的点,无论我在 window.requirejs.config 中将 waitseconds 设置为多大。 [叹气]
猜你喜欢
  • 2018-06-20
  • 1970-01-01
  • 2020-05-17
  • 2018-09-23
  • 2019-08-07
  • 2018-10-13
  • 1970-01-01
  • 2017-10-10
相关资源
最近更新 更多