The example below generates a contour plot of sparse data.
import numpy as np
import matplotlib.pyplot as plt
from scipy.ndimage import gaussian_filter, zoom
# Get the data
data = np.loadtxt('data.txt')
# Create a contour plot of the original data
_, (ax1, ax2) = plt.subplots(ncols=2, tight_layout=True, figsize=[10, 4.8])
ax1.contour(data)
ax2.contourf(data)
The plot can be smoothed by using SciPy's zoom function or gaussian filter function. Input parameters for each function should be adjusted accordingly for the data.
# Resample the data with zoom then create contour plot
data2 = zoom(data, 3)
_, (ax1, ax2) = plt.subplots(ncols=2, tight_layout=True, figsize=[10, 4.8])
ax1.contour(data2)
ax2.contourf(data2)
# Resample data with gaussian filter then create contour plot
data3 = gaussian_filter(data, 0.7)
_, (ax1, ax2) = plt.subplots(ncols=2, tight_layout=True, figsize=[10, 4.8])
ax1.contour(data3)
ax2.contourf(data3)
The contour plot examples in this article are based on a Stack Overflow post which is where the original data was obtained from.
