CODEWITHSUNDEEP
pythonnumpymatplotlibplotsubplot
Py-ser
Py-ser

Reputation: 2098

Multiple subplots to personalize

I make some subplots by using:

fig, ax = plt.subplots(nrows=2, ncols=3, sharex=True, sharey=True)

Then, I would like to personalize the subplots, e.g., with:

plt.subplots_adjust(hspace = 0., wspace= 0.)

However, I would like to personalize the ticks as well, like removing the ticks and the labels for some of those subplots. How could I do that?

The problem is that, after the definition, ax is a numpy array: I don't know that much of it, what I know is that it is impossible to use attributes (like, ax[0].set_yticks([])).

Upvotes: 0

Views: 852

Answers (2)

ali_m
ali_m

Reputation: 74154

If you create a 2D array of plots, e.g. with:

>>> fig, axarray = plt.subplots(3, 4)

then axarray is a 2D array of objects, with each element containing a matplotlib.axes.AxesSubplot:

>>> axarray.shape
(3, 4)

The problem is that when you index axarray[0], you're actually indexing a whole row of that array, containing several axes:

>>> axarray[0].shape
(4,)

>>> type(axarray[0])
numpy.ndarray

However, if you address a single element in the array then you can set its attributes in the normal way:

>>> type(axarray[0,0])
matplotlib.axes.AxesSubplot

>>> axarray[0,0].set_title('Top left')

A quick way of setting the attributes of all of the axes in the array is to loop over a flat iterator on the axis array:

for ii,ax in enumerate(axarray.flat):
    ax.set_title('Axis %i' %ii)

Another thing you can do is 'unpack' the axes in the array into a nested tuple of individual axis objects, although this gets a bit awkward when you're dealing with large numbers of rows/columns:

fig, ((ax1, ax2, ax3, ax4), (ax5, ax6, ax7, ax8), (ax9, ax10, ax11, ax12)) \
    = plt.subplots(3,4)

Upvotes: 3

Greg
Greg

Reputation: 12234

When using this method:

fig, ax = plt.subplots(nrows=2, ncols=3, sharex=True, sharey=True)

you have two choices, either call the elements of the array ax like you have suggested (but you will need to use two indices or flatten it):

ax[0][0].plot(...
ax.flat[0].plot(...

this second line is useful if you loop over the plots. Or you can modify in the following way:

fig, ((ax1, ax2, ax3), (ax4, ax5, ax6)) = plt.subplots(nrows=2, ncols=3, sharex=True, sharey=True)

It will depend on your use case which is better, I typically call the array ax if there is a chance I will change the number of subplot.

Upvotes: 1

Related Questions