Is there a way to make two subplots that share their x axis scale while allowing them to have different limits.
Here is a diagram of what I'm trying to achieve:
(the faint vertical lines are to indicate that the ticks are in the same horizontal position and will not be in the final graph)
The extra space in the bottom right will be filled with another axis. I'm guessing there is no easy way, but if there is I'd greatly appreciate it!
Ok, so I discovered the transforms in the matplotlib docs, and this works. Not exactly trivial, but that's ok. The graph produced here is reversed from the example: the top has the short x-axis. The concept ok though.
import matplotlib.pyplot as plt
short_ax_lim = 4.
f, (top_ax, bottom_ax) = plt.subplots(2,1,figsize=[9,5])
bottom_ax.set_xlim(0,5)
top_ax.set_xlim(0,short_ax_lim)
# first find the pixel where the bottom x-axis equals the desired limit.
short_ax_stop_pix = bottom_ax.transData.transform((short_ax_lim,0)) # (x,y) in pixels at data point (0,4)
short_ax_stop_fig = pix_to_fig.transform(short_ax_stop_pix) # (x,y) in figure space (0,1)
top_orig_position_px = top_ax.bbox.corners() # (ll, ul, lr, ur)
top_orig_anchor_px = top_orig_position_px[0] # this is lower left corner.
top_orig_anchor_fig = pix_to_fig.transform(top_orig_anchor_px) #convert to figure space
top_x_anchor, top_y_anchor = top_orig_anchor_fig
top_width = short_ax_stop_fig[0] - top_x_anchor
new_pos = (top_x_anchor, top_y_anchor, top_width, top_ax.get_position().height)
top_ax.set_position(new_pos)
f.show()
Using sharex will couple the axes. It's not pretty but you could use subplot2grid like this,
import matplotlib.pyplot as plt
ax1 = plt.subplot2grid((2,5), (0,0), colspan=5)
ax2 = plt.subplot2grid((2,5), (1,0), colspan=4)
ax1.set_xlim(0,5)
ax2.set_xlim(0,4)
plt.show()
