★ dotplotlib

I hijacked matplotlib‘s scatterplot function to give the people what they want; a dot chart (also know as strip plot). You can install it with pip install dotplotlib, or see the source code here.

1
Current stars: 6

Great things come in tiny packages. A bare minimum extension library for creating tree dot plots, strip plots or dot charts w/ matplotlib or seaborn in Python

  • Designed to work with matplotlib and seaborn in Python
  • Fully customizable

installation

1
pip install dotplotlib

usage

dotplotlib can be used to generate dot charts with minimal code. Here are some examples:


Example 1: Simple Dot Chart

.dotchart returns x and y lists that can be inputted straight into matplotlib or seaborn scatterplots. 

1
2
3
4
5
6
7
8
9
10
11
from dotplotlib import dotchart
import matplotlib.pyplot as plt

data = {'size': [1, 2, 2, 3, 3, 3, 4]}

# Generate dot chart data
x, y = dotchart(data['size'])

# Plot
plt.scatter(x, y)
plt.show()

Example 2: Dot Chart with Color Mapping

Pass the data you would like to color by to the color_by= argument.

Returns an extra list c that should be passed into the c= parameter if using matplotlib or hue= if using seaborn. 

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
from dotplotlib import dotchart
import matplotlib.pyplot as plt

data = {'size': [1, 2, 2, 3, 3, 3, 4], 'rating': [3, 2, 5, 4, 3, 6, 4]}

# Generate dot chart data with color mapping
x, y, c = dotchart(data['size'], color_by=data['rating'])

# Plot with color mapping
plt.scatter(x, y, c=c, cmap='viridis')
plt.colorbar()
plt.xlabel('Size')
plt.ylabel('Number')
plt.title('Mushroom Size Count Colored by Rating')
plt.show()

Example 3: Using make_dotchart to plot in one step

Instead of just giving you x, y lists to make the plot yourself, make_dotplot() actually generates the plot. 

1
2
3
4
5
6
7
8
9
10
11
12
13
14
from dotplotlib import make_dotchart

df = {'size': [1, 2, 2, 3, 3, 3, 4], 'rating': [3, 2, 5, 4, 3, 6, 7]}

# Create a dot chart with optional arguments (only the first one is mandatory)
make_dotchart(df['size'], 
                  color_by=df['rating'], # list to color by
                  reverse=False, # inverts the color mapping
                  theme='gnuplot2', # scroll down to see all themes
                  colorbar=True, 
                  xlabel='Sizes', 
                  ylabel='Size Count', 
                  title='Mushroom Sizes Colored by Rating', 
                  dot_size=40):

Example 4: Plotting in a Jupyter Notebook

If plotting inline, use the default .dotchart() to obtain x and y lists, and then adjust as necessary with one of the following: 

1
2
3
plt.figure(figsize=(12,6))  # or
plt.figure().set_figwidth(12) # or
plt.figure().set_figheight(12)


preset themes

custom:lavender

cmap

Any cmap value supported by matplotlib (see here) will work when passed into theme='viridis'.

viridis:

gnuplot:

gallery:

features

  • generate strip plots/dot charts by exploiting matplotlib/seaborn scatterplots
  • supports any cmap color profile
  • the data can be automatically sorted for better visualization, especially when using color mapping.
  • accepts both list and pandas.Series as input data.
  • set custom labels, titles, and dot sizes for your charts.
  • works with Jupyter Notebook

attribution