Suggestion on: http://stackoverflow.com/questions/17949475/intelligibly-displaying-multiple-dense-scatter-plots-on-one-axis-in-python-mat

scatter_vs_multiple_plot_calls.py

import numpy as np
import matplotlib.pyplot as plt
from matplotlib.colors import ListedColormap

num = 100000
x, y = np.random.random((2, num))
category = np.random.randint(1, 5, num)

# Using multiple "plot" calls
fig, ax = plt.subplots()
ax.set_color_cycle(['black', 'red', 'orange', 'yellow'])
for i in range(1, 5):
    mask = category == i
    ax.plot(x[mask], y[mask], marker='+', linestyle='none')
ax.set(title="The way you're probably plotting things now")

# Using a single "scatter" call
fig, ax = plt.subplots()
cmap = ListedColormap(['black', 'red', 'orange', 'yellow'])
ax.scatter(x, y, c=category, cmap=cmap, marker='+')
ax.set(title="Using scatter to plot instead")
ax.axis([0, 1, 0, 1])

plt.show()