Question

I have a subplot of 28 lines x 2 columns(it can change, actually). The yaxis scale of all the lines of the 1st column is supposed to be the same (that must work for that 2nd column as well)….

All the xaxis are supposed to be the same.

What I want to do is to make something inside the output figure that shows what is the yaxis for the 1st and 2nd columns and what is the xaxis for both columns… I also want to get a label to up top at the 1st and 2nd column (saying what are those data).

I also want to change the aspect of the plots so that it can be seen more clearly (probably increasing the yaxis aspect size and drecreasing the xaxis size a little).

The subplot I want, without the re-sizing, should be something like this:

It can be something different. I really don’t know whats possible to make out of my request.

The code i usd to generate the figure (without the paint-made labels):

def pltconc(conc,self):
t=self.t
idx1=0
conc=conc*1000000


c=len(find( self.ml[:,3]==1 ))

from scipy.stats import scoreatpercentile #To adjust the scales
ymin1 = max([median(scoreatpercentile(conc[:,i,:],0.05)) for i in range(28)])
ymax1 = max([median(scoreatpercentile(conc[:,i,:],99.95)) for i in range(28)])

for idx1 in range(c):
    a=subplot(c,2,2*idx1+1, adjustable='box-forced')
    plt.plot(t,conc[:,idx1,0],color='r')
    plt.plot(t,conc[:,idx1,1],color='b')
    plt.axis('tight')
    xlim(0,max(self.t))
    ylim(ymin1,ymax1)
    frame1 = plt.gca()
    a.set_yticklabels([])
    a.set_xticklabels([])


    ax=subplot(c,2,2*idx1+2, adjustable='box-forced')
    CBV = (conc[:,idx1,2]*100)/(90+conc[:,idx1,2])
    StO2 = (conc[:,idx1,0]*100)/(90+conc[:,idx1,2])
    ymin2 = max(median(scoreatpercentile(CBV,0.05)),median(scoreatpercentile(StO2,0.05)))
    ymax2 = max(median(scoreatpercentile(StO2,99.95)),median(scoreatpercentile(CBV,99.95)))
    plt.plot(t,CBV, color='m')
    plt.plot(t,StO2, color = 'b')
    plt.axis('tight')
    xlim(0,max(self.t))
    ylim(ymin2,ymax2)
    frame1 = plt.gca()
    ax.set_yticklabels([])
    ax.set_xticklabels([])

Thanks alot for the help.

I changed the code since I’ve realized they weren’t correctly scaled. The output figure should be a little bit differente, but it doesn’t really matters for this questions purpose.

Answer 1

I’m not entirely sure what you’re asking, but here’s how I’d go about plotting something along those lines…

The aspect ratio for your figure is controlled by the figsize kwarg to plt.figure (or plt.subplots, in this case).

The rest you can do with judicious application of annotate.

Here’s an example:

import matplotlib.pyplot as plt
import numpy as np

# Generate the data
data = (np.random.random((20, 2, 2, 1001)) - 0.5).cumsum(axis=-1)

# Set up the figure (the figsize is what's going to control your aspect ratio)
fig, axes = plt.subplots(nrows=20, ncols=2, sharex=True, figsize=(6, 10))
fig.subplots_adjust(wspace=0.1, hspace=0, bottom=0.05)

# Turn off tick labels everywhere
for ax in axes.flat:
    for axis in [ax.xaxis, ax.yaxis]:
        axis.set_ticklabels([])

# Plot the data
color = {(0,0):'red', (0,1):'green', (1,0):'blue', (1,1):'magenta'}
for (i,j), ax in np.ndenumerate(axes):
    for k in range(2):
        ax.plot(data[i,j,k,:], color=color[(j,k)])

# Add stacked titles (and text legends)
titles = [['TITLE:', 'Red: Data X', 'Green: Data Y'],
          ['TITLE:', 'Blue: Data W', 'Magenta: Data Z']]
for i, title in enumerate(titles):
    for text, ypos in zip(title, [35, 20, 5]):
        axes[0,i].annotate(text, xy=(0.05, 1.0), xytext=(0, ypos), va='bottom',
                           xycoords='axes fraction', textcoords='offset points')

# Add arrows on "super-Y" axes
xpos, length = -0.1, 5
axes[12,0].annotate('', xy=(xpos, 0), xytext=(xpos, length), 
        xycoords='axes fraction', textcoords='axes fraction',
        arrowprops=dict(arrowstyle='<|-'))
axes[12,0].annotate('{0} subplots'.format(length), xy=(xpos, length/2.0), 
        xycoords='axes fraction', rotation=90, va='center', ha='right')

# Add arrows on "super-X" axes
ypos, length = -0.7, 1000
axes[-1,0].annotate('', xy=(0, ypos), xytext=(length, ypos),
        xycoords=('data', 'axes fraction'), textcoords=('data', 'axes fraction'),
        arrowprops=dict(arrowstyle='<|-'))
axes[-1,0].annotate('{0} data units'.format(length), xy=(length/2.0, ypos), 
        xytext=(0, 5), xycoords=('data', 'axes fraction'), 
        textcoords='offset points', ha='center', va='bottom')

plt.show()