I would like to create three figures showing how a quantity varies with z in three different simulations. The different simulations have different domains, the first extends from z=0 to 1.4, the next 1.1, and the next 1.0. I would like to have three side by side graphs displaying the quantity on the x-axis and z on the y-axis, all scaled in the same way, but with the axes stopping at the maximum z-value, so that the graphs get progressively shorter. The below is how I'm currently plotting them, and I'd like to lose the blank space in the second and third axis. fig, axes = plt.subplots(1, 3, sharey=True) z1 = np.linspace(0, 1.4, 20) z2 = np.linspace(0, 1.1, 20) z3 = np.linspace(0, 1.0, 20) axes[0].plot(np.exp(-z1), z1, c='k') axes[1].plot(np.exp(-z2), z2, c='k') axes[2].plot(np.exp(-z3), z3, c='k') I've attempted using gridspec and subplots_mosaic, but both seem to insist on keeping the axes the same relative size. I suppose there's a solution using subplots_mosaic where the mosaic is something like """ A.. A.. A.. AB. ABC ABC ABC ABC ABC ABC ABC ABC ABC """ to get the right relative heights, but this seems very clunky, and I am hoping there is an easier solution. Continue reading...
