Python Numpy 2D plot set total number of y-tics with autoscaling -

i limit total number of tics on y-axis in 2d plot. in example image provided can see f_x plot has 6 total tics (0,0.2,0.4,0.6,0.8,1) while f_y plot has 8. there way still have axis scale automatically fix number of total tics plots greater 6 tics in example busy?

ax.locator_params(axis='y', nbins=num) easiest way this.

nbins set maximum number of bins (i.e. spaces between ticks) each axis. "even" numbers still chosen, value controls density of ticks on axis. default value new axes 9 (in other words, maximum of 10 ticks/ticklabels).

for example, let's set rather busy default:

import matplotlib.pyplot plt import numpy np np.random.seed(1977)  # generate data different ranges x = np.linspace(0, 8.8, 1000) ydata = np.random.normal(0, 1, (4, x.size)).cumsum(axis=1) ydata *= np.array([1e-3, 1e3, 10, 1e-2])[:,none]  fig, axes = plt.subplots(nrows=4) y, ax in zip(ydata, axes):     ax.plot(x, y, color='salmon') plt.show() 

oy!! not good! let's see if can better:

fig, axes = plt.subplots(nrows=4) y, ax in zip(ydata, axes):     ax.plot(x, y, color='salmon')     ax.locator_params(axis='y', nbins=5) plt.show() 

getting there, still bit busy. reduce nbins further, we'll wind no ticks. instead, 1 trick use "prune" first , last ticks axes. can controlled locator_params:

fig, axes = plt.subplots(nrows=4) y, ax in zip(ydata, axes):     ax.plot(x, y, color='salmon')     ax.locator_params(axis='y', nbins=5, prune='both') plt.show() 

this "pruning" more effective when combined shared x-axes, in case of type of plot you're making. main effect post turn off of x-ticklabels. however, link interactive zooming , panning of axes x-range shared:

fig, axes = plt.subplots(nrows=4, sharex=true) y, ax in zip(ydata, axes):     ax.plot(x, y, color='salmon')     ax.locator_params(axis='y', nbins=5, prune='both') plt.show() 

now can move things bit closer together:

fig, axes = plt.subplots(nrows=4, sharex=true) y, ax in zip(ydata, axes):     ax.plot(x, y, color='salmon')     ax.locator_params(axis='y', nbins=5, prune='both')  fig.subplots_adjust(hspace=0) plt.show() 

finally, in particular case, might consider using ax.margins(...) add padding in y-direction , force "tight" scaling of data range in x-direction.

fig, axes = plt.subplots(nrows=4, sharex=true) y, ax in zip(ydata, axes):     ax.plot(x, y, color='salmon')     ax.margins(x=0, y=0.05)     ax.locator_params(axis='y', nbins=5, prune='both')  fig.subplots_adjust(hspace=0) plt.show()