bad contour lines plot

Question

I'm having a hard time with contour plotting. The contour lines are going crazy in my plot, and I'm not sure why. You can see some of the data points in the background.

print positive_train_data.shape
#returns (1131,2)

def GaMM():
  GaussMM = GMM(n_components=3)
  GaussMM.fit(positive_train_data)
  X, Y = np.meshgrid(positive_train_data[:, 0], positive_train_data[:, 1])
  XX = np.array([X.ravel(), Y.ravel()]).T
  Z = -GaussMM.score(XX)
  Z = Z.reshape(X.shape)
  CS = plt.contour(X, Y, Z)
  CB = plt.colorbar(CS, shrink=0.8, extend='both')
  plt.scatter(positive_train_data[:, 0], positive_train_data[:, 1])

GaMM()

score 2 · Accepted Answer · answered Aug 04 '17 at 08:46

The data seems to be completely unordered. This is similar to what happens in the left picture below.

This is taken from one answer to this question: Why does pyplot.contour() require Z to be a 2D array? and the solution would be to use tricontour instead of contour as shown in the right picture.

The other option is to interpolate the data on a 2D grid, e.g. using matplotlib.mlab.griddata

Further suggested reading are:

Plotting Isolines/contours in matplotlib from (x, y, z) data set for options on how to plot data from x,y,z dataset.
How do I interpolate a 2D gridded point cloud to a continuous area? for a more detailed answer concerning filled contours.

bad contour lines plot

1 Answers1