Bubble-plot weighted 2d scatter plot data onto mesh/grid

Question

Background

I have a 2D set of "raw" data points, in the form of an array, i.e.:

[[0, 0, 1, 0, 0],
 [0, 1, 2, 1, 0],
 [1, 2, 4, 2, 1],
 [0, 1, 2, 1, 0],
 [0, 0, 1, 0, 0]]

The data represents a circle-like "mass distribution" plot in 2D space. The x-y coordinates represent the x-y distribution of data points, and the value at each x-y coordinate is the mass/intensity measured at that data point.

I would like to plot this data, using only Python, (only at the integral x-y intersections), like in the plot below, but using my own x-y scatterplot data, rather than plotting a 2D line/function.

Additionally, I would like to combine the "specify dot size" logic from another SO question, which allows me to specify the dot size on a "per-sample/value" basis, i.e.:

---

Question

How do I combine the logic above to render a data set like so:

[[0, 0, 1, 0, 0],
 [0, 1, 2, 1, 0],
 [1, 2, 4, 2, 1],
 [0, 1, 2, 1, 0],
 [0, 0, 1, 0, 0]]

As something like this, via matplotlib/pyplot (discrete-domain, discrete-range, continuous-valued):

---

Extra

How can I re-use the above data set to generate a heatmap of the same data (i.e. continuous-domain, continuous-range, continuous valued)?

Or, more like this:

Answer 1

You can do it by first creating a meshgrid defining your x and y coordinates and then using the data array to define the size of your points. The entries which are 0 will not be shown because of 0 size. I am using a rescaling factor of 100 to just enlarge the points.

Complete working code:

---

import numpy as np
import matplotlib.pyplot as plt

data = np.array([[0, 0, 1, 0, 0],
 [0, 1, 2, 1, 0],
 [1, 2, 4, 2, 1],
 [0, 1, 2, 1, 0],
 [0, 0, 1, 0, 0]])

mesh = np.arange(len(data))
x, y = np.meshgrid(mesh, mesh)
plt.scatter(x, y, s=data*100)
plt.xticks(range(len(data))) # To put ticks at integer values
plt.yticks(range(len(data))) # To put ticks at integer values
plt.show()

Generating a heat map

import numpy as np
import matplotlib.cm as cm
import matplotlib.pyplot as plt

# data here

x, y = np.meshgrid(np.arange(len(data)), np.arange(len(data)))
plt.scatter(x, y, s=data*100, c=data, cmap=cm.Oranges)
plt.xticks(range(len(data)))
plt.yticks(range(len(data)))