Python Interactive window

The goal

Python Interactive window allows you to interactively develop code with VS Code.

Questions to David Rotermund

Please use # %% with .py files instead of .ipynb Jupyter notebook files for interactive cells!!!

If you don’t know what I am talking about, please read: Python Interactive window

Executing a cell

In the cell mode (# %%) you can use

• SHIFT + ENTER to execute a cell or
• CTRL + ENTER.

The former moves the cursor into the next cell, while the later keeps the cursor that the original position.

Interactive plotting in # %% cells​

pip install ipympl

We can activate the interactive plotting mode via ​

%matplotlib widget ​

The first time we use this command, vs code will need to download a plugin).

Here an example:

# %%
%matplotlib widget

import matplotlib.pyplot as plt
import numpy as np

fig = plt.figure()

plt.plot(np.sin(np.linspace(0, 20, 100)))

Yes, mypy will give you an error and yes, you can not suppress it… but non you can zoom and pan the figure… You can not have everything at once!

Modifing a plot

First we plot something but keep the handle:

# %%
import matplotlib.pyplot as plt
import numpy as np

fig = plt.figure()
line = plt.plot(np.sin(np.linspace(0, 20, 100)))

Then we can use to change the plot using the handle:

# %%
line[0].remove()
line = plt.plot(np.sin(5.0*np.linspace(0, 20, 100)))

If you don’t know the handle, you can retrieve it like this:

# %%
handles = fig.gca().get_children()
print(handles)

[<matplotlib.lines.Line2D object at 0x7fe130f75310>, <matplotlib.spines.Spine object at 0x7fe130f1d710>, <matplotlib.spines.Spine object at 0x7fe130f1dcd0>, <matplotlib.spines.Spine object at 0x7fe130f1e210>, <matplotlib.spines.Spine object at 0x7fe130f1e750>, <matplotlib.axis.XAxis object at 0x7fe130f1ecd0>, <matplotlib.axis.YAxis object at 0x7fe1310b7350>, Text(0.5, 1.0, ''), Text(0.0, 1.0, ''), Text(1.0, 1.0, ''), <matplotlib.patches.Rectangle object at 0x7fe130f3cf10>]

This allows us to do this:

# %%
fig.gca().get_children()[0].remove()

Re-Plotting

You can use

# %%
display(fig)

for replotting the plot. However you will get a message from VS Code that display is not defined. It is defined but it doesn’t know it.

The source code is Open Source and can be found on GitHub.