Jupyter General
Help
To see parameters: SHIFT + TAB
For help use ?
before or after the command. For autocompletion: TAB + TAB
Merge cells, select cells then: Shift + M
`To see all available functions in numpy that have
sinin their name:
np.sin?`
Markdown
For raw text use:
```text
and close with
<```> but without the <> (that's the only one text that can't be displayed I guess!)
To add some empty vertical space one can use:
or add a new line use \_
where _
denotes space.
Magic commands
To autoreload edited python files (doesn’t include class signatures changes, these you have to re-import):
%load_ext autoreload
%autoreload 2
Links
Links are useful to jump between sections and files. Here are few examples:
To jump to upper section help use a hashtag: [help](#help)
To jump to some section in another file one needs to specify html link (not a jupyter file one), for example [SSH](/tips/030_Terminal/zsh.html#ssh)
for SSH.
To open a browser link in a new window use [text](link){target="_blank"}
, for example this SSH in new window.
Similar for HTML link <a href="https://example.com" target="_blank">Link Text</a>
.
nbimporter
To import methods from another jupyter notebook:
pip install nbimporter
For more see here.
nbconvert
nbconvert
is a command line tool that can convert notebooks to other formats. To convert to pdfs, use the following command:
jupyter nbconvert --to pdf notebook.ipynb
and it will create a pdf file with the same name as the notebook (in the case above notebook.pdf).
You will need to install MikTeX or TeX Live to convert to pdf (MikTex is smaller and easier to install).
One quirk is that it doesn’t render images properly, I found it works only if the image is referred to in markdown as:
![image](images/image.png)
.
The following WILL NOT work however:
- ![](images/image.png)
apparently image caption is necessary to reference it.
- ![image](images/image.png){width=500px}
this will fail to render properly, image will be squished weirdly
Another option is to use `notebook-as-pdf` package (this will install `nbconvert` as well, and bunch of other packages).
From [here](https://towardsdatascience.com/jupyter-notebook-to-pdf-in-a-few-lines-3c48d68a7a63):
```python
pip install -U notebook-as-pdf
wait to finish installation. Then:
pyppeteer-install
pip uninstall pypdf2
pip install "pypdf2<3.0"
Then run the following command line: jupyter-nbconvert --to PDFviaHTML <jupyter_notebook_file.ipynb>
this will create PDF file with the same name jupyter_notebook_file.pdf
(ignore the warning on IPython3 lexer unavailable). To rename a file: mv old_name new_name
Unfortunately only command line works, Download as
option in a browser doesn’t.
Extensions
conda install -c conda-forge jupyter_contrib_nbextensions
After installation and restart, you will see nbextensions
tab, most notable extensions are collapsable headings
, table of context 2 (toc2)
, spellcheck
.