bqplot documentation in Python

Question

I'm trying to develop some graphs using bqplot in Python but I'm having some questions regarding the documentation and where I can find the appropriate options to personalize a graph. I'm trying to make a boxplot graph and chenge the axes_options, but I don't find anywhere what's the possible options of 'axes_options' argument.

Here's a toy example of the boxplot that I want to change:

import bqplot.pyplot as plt
x = [1]
y = [[7,9,7,6,4,3,2,5,6,7,1,2,3,1,4.4,7.7]]
fig = plt.figure()
# add x and y labels using axes_options
axes_options = {'x': {'label': 'X'}, 'y': {'label': 'Variable Value'}}
boxes = plt.boxplot(x, y, box_fill_color='gray', outlier_fill_color='black', axes_options=axes_options)
fig

How can I, for example, change axes limits inside the axes_options dictionary? And how can I find the available options?

bqplot has two API: one inspired in matplotlib (which is from my example) and another one called verbose API. They explain this here in this notebook: https://hub.mybinder.org/user/bloomberg-bqplot-g3kywf1d/notebooks/examples/Introduction.ipynb — Renan Xavier Cortes, Aug 17 '18 at 17:35
The link isn't working, but I suppose you mean [this one](https://github.com/bloomberg/bqplot/blob/4722f21209cd8a1b6baa818e556b4cea4d5373e0/examples/Marks/Pyplot/Boxplot.ipynb)? Is there a reason you don't want to use `plt.ylim()`? — ImportanceOfBeingErnest, Aug 17 '18 at 18:09
I guess the available options are listed in [bqplot.axes.Axis](https://bqplot.readthedocs.io/en/latest/_generate/bqplot.axes.Axis.html). It doesn't look like it's possible to change the axis limits with this though. — ImportanceOfBeingErnest, Aug 17 '18 at 21:13
Hi! Yes, I actually changed my code to use plt.ylim and it worked :) However, I'm still having another issue which is a 'panning' problem of bqplot. Whenever I pan the boxplot, the axes update but the boxplot doesn't move. But I think this is another issue that I should open. Thanks! — Renan Xavier Cortes, Aug 20 '18 at 12:25

score 0 · Answer 1 · answered Aug 22 '18 at 08:06

One way I achieve this is to access the Scale that the Axis is attached to once the figure has been instantiated, e.g. fig.axes[0].scale.min = 0, or fig.axes[0].scale.max = max(data).

I have come across some issues with boxplot generally, and as you highlighted on Github. But this method seems to work well for other chart types e.g. Scatter.

bqplot documentation in Python

1 Answers1