ValueError: Contour levels must be increasing - Contour plot in Python

Question

I am trying to plot density estimates using a contour plot and getting the following error.

A = np.random.uniform(size=(100, 2))
#mean = np.mean(x)
#cov = np.cov(x)
mean = np.array([0.5, 0.1])
cov = np.array([[0.1, 0.0], [0.0, 1.5,]])

B = multivariate_normal.pdf(A, mean=mean, cov=cov)


# visualize
contours = plt.contour(A, B, linewidths=2)
plt.clabel(contours, inline=True,  fontsize=12)
#plt.plot(x, y)
plt.colorbar();

Answer 1

For someone in future that has this problem in seaborn, I discovered that my data had some extreme outliers, meaning there was effectively no density to plot as 99% of the samples were around the origin. Using the 'clip' functionality in kdeplot worked to reduce the axis and thus plot the actual levels.