CODEWITHSUNDEEP
pythonmatplotlibdata-visualization
NeutronStar
NeutronStar

Reputation: 2157

Python: how to plot a series of regions in a stacked/layered way?

I have several different regions that measure the 2-d spatial event of something given a condition. The condition is encoded with the color of the region. The regions in general are polygons (even though I've plotted circles) and can have holes in them, as in:

enter image description here

To increase clarity of visualization, I have been requested to turn this into what has been called a "wedding cake plot", where each region has a vertical offset from each other, with vertical sides around each region. Something kind of like the following Paint image:

enter image description here

How to go about doing this with Matplotlib, or some other Python plotting package?

Upvotes: 1

Views: 132

Answers (2)

r-beginners
r-beginners

Reputation: 35155

The plotly library has a function to draw a funnel, so you can use that instead. The following is a graph from the official reference, with some modifications to make it look the way you want.

from plotly import graph_objects as go

x_sorted = sorted([39, 27.4, 20.6, 11, 2])
y_sorted = ["Website visit", "Downloads", "Potential customers", "Requested price", "invoice sent"]
fig = go.Figure(go.Funnel(
    y = y_sorted[::-1],
    x = x_sorted,
    marker={'color':['black','yellow','green','blue','red']})
               )
fig.show()

enter image description here

Upvotes: 0

Quang Hoang
Quang Hoang

Reputation: 150745

You can do something with plot_surface like this:

# sample data
data = np.zeros((20,24))
data[2:18, 2:18] = 1
data[4:16, 4:16] = 2
data[6:14,6:14] = 3
data[8:12,8:12] = 2

# 2D view
plt.imshow(data, cmap='Reds')

# mesh
X,Y=np.mgrid[0:data.shape[0], 0:data.shape[1]]


# plot
fig, ax = plt.subplots(subplot_kw={"projection": "3d"})

surf = ax.plot_surface(X, Y, data, cmap='hot',
                       linewidth=0, antialiased=False)

Output:

enter image description here

Note You would need to hierarchically label your data correctly.

Upvotes: 2

Related Questions