I have too many ticks on my graph and they are running into each other.

How can I reduce the number of ticks?

For example, I have ticks:

1E-6, 1E-5, 1E-4, ... 1E6, 1E7

And I only want:

1E-5, 1E-3, ... 1E5, 1E7

I’ve tried playing with the LogLocator, but I haven’t been able to figure this out.

Alternatively, if you want to simply set the number of ticks while allowing matplotlib to position them (currently only with MaxNLocator), there is pyplot.locator_params,

pyplot.locator_params(nbins=4)

You can specify specific axis in this method as mentioned below, default is both:

# To specify the number of ticks on both or any single axes
pyplot.locator_params(axis='y', nbins=6)
pyplot.locator_params(axis='x', nbins=10)

If somebody still gets this page in search results:

fig, ax = plt.subplots()

plt.plot(...)

every_nth = 4
for n, label in enumerate(ax.xaxis.get_ticklabels()):
    if n % every_nth != 0:
        label.set_visible(False)

To solve the issue of customisation and appearance of the ticks, see the Tick Locators guide on the matplotlib website

ax.xaxis.set_major_locator(plt.MaxNLocator(3))

Would set the total number of ticks in the x-axis to 3, and evenly distribute it across the axis.

There is also a nice tutorial about this

There’s a set_ticks() function for axis objects.

in case somebody still needs it, and since nothing here really worked for me, i came up with a very simple way that keeps the appearance of the generated plot “as is” while fixing the number of ticks to exactly N:

import numpy as np
import matplotlib.pyplot as plt

f, ax = plt.subplots()
ax.plot(range(100))

ymin, ymax = ax.get_ylim()
ax.set_yticks(np.round(np.linspace(ymin, ymax, N), 2))

The solution @raphael gave is straightforward and quite helpful.

Still, the displayed tick labels will not be values sampled from the original distribution but from the indexes of the array returned by np.linspace(ymin, ymax, N).

To display N values evenly spaced from your original tick labels, use the set_yticklabels() method. Here is a snippet for the y axis, with integer labels:

import numpy as np
import matplotlib.pyplot as plt

ax = plt.gca()

ymin, ymax = ax.get_ylim()
custom_ticks = np.linspace(ymin, ymax, N, dtype=int)
ax.set_yticks(custom_ticks)
ax.set_yticklabels(custom_ticks)

When a log scale is used the number of major ticks can be fixed with the following command

import matplotlib.pyplot as plt

....

plt.locator_params(numticks=12)
plt.show()

The value set to numticks determines the number of axis ticks to be displayed.

Credits to @bgamari’s post for introducing the locator_params() function, but the nticks parameter throws an error when a log scale is used.

To remove frame in figure, I write

frameon=False

works perfect with pyplot.figure, but with matplotlib.Figure it only removes the gray background, the frame stays . Also, I only want the lines to show, and all the rest of figure be transparent.

with pyplot I can do what I want, I want to do it with matplotlib for some long reason I ‘d rather not mention to extend my question.

First off, if you’re using savefig, be aware that it will override the figure’s background color when saving unless you specify otherwise (e.g. fig.savefig('blah.png', transparent=True)).

However, to remove the axes’ and figure’s background on-screen, you’ll need to set both ax.patch and fig.patch to be invisible.

E.g.

import matplotlib.pyplot as plt

fig, ax = plt.subplots()
ax.plot(range(10))

for item in [fig, ax]:
    item.patch.set_visible(False)

with open('test.png', 'w') as outfile:
    fig.canvas.print_png(outfile)

(Of course, you can’t tell the difference on SO’s white background, but everything is transparent…)

If you don’t want to show anything other than the line, turn the axis off as well using ax.axis('off'):

import matplotlib.pyplot as plt

fig, ax = plt.subplots()
ax.plot(range(10))

fig.patch.set_visible(False)
ax.axis('off')

with open('test.png', 'w') as outfile:
    fig.canvas.print_png(outfile)

In that case, though, you may want to make the axes take up the full figure. If you manually specify the location of the axes, you can tell it to take up the full figure (alternately, you can use subplots_adjust, but this is simpler for the case of a single axes).

import matplotlib.pyplot as plt

fig = plt.figure(frameon=False)
ax = fig.add_axes([0, 0, 1, 1])
ax.axis('off')

ax.plot(range(10))

with open('test.png', 'w') as outfile:
    fig.canvas.print_png(outfile)

ax.axis('off'), will as Joe Kington pointed out, remove everything except the plotted line.

For those wanting to only remove the frame (border), and keep labels, tickers etc, one can do that by accessing the spines object on the axis. Given an axis object ax, the following should remove borders on all four sides:

ax.spines['top'].set_visible(False)
ax.spines['right'].set_visible(False)
ax.spines['bottom'].set_visible(False)
ax.spines['left'].set_visible(False)

And, in case of removing x and y ticks from the plot:

 ax.get_xaxis().set_ticks([])
 ax.get_yaxis().set_ticks([])

The easiest way to get rid of the the ugly frame in newer versions of matplotlib:

import matplotlib.pyplot as plt
plt.box(False)

If you really must always use the object oriented approach, then do: ax.set_frame_on(False).

Building up on @peeol’s excellent answer, you can also remove the frame by doing

for spine in plt.gca().spines.values():
    spine.set_visible(False)

To give an example (the entire code sample can be found at the end of this post), let’s say you have a bar plot like this,

you can remove the frame with the commands above and then either keep the x- and ytick labels (plot not shown) or remove them as well doing

plt.tick_params(top='off', bottom='off', left='off', right='off', labelleft='off', labelbottom='on')

In this case, one can then label the bars directly; the final plot could look like this (code can be found below):

Here is the entire code that is necessary to generate the plots:

import matplotlib.pyplot as plt
import numpy as np

plt.figure()

xvals = list('ABCDE')
yvals = np.array(range(1, 6))

position = np.arange(len(xvals))

mybars = plt.bar(position, yvals, align='center', linewidth=0)
plt.xticks(position, xvals)

plt.title('My great data')
# plt.show()

# get rid of the frame
for spine in plt.gca().spines.values():
    spine.set_visible(False)

# plt.show()
# remove all the ticks and directly label each bar with respective value
plt.tick_params(top='off', bottom='off', left='off', right='off', labelleft='off', labelbottom='on')

# plt.show()

# direct label each bar with Y axis values
for bari in mybars:
    height = bari.get_height()
    plt.gca().text(bari.get_x() + bari.get_width()/2, bari.get_height()-0.2, str(int(height)),
                 ha='center', color='white', fontsize=15)
plt.show()

As I answered here, you can remove spines from all your plots through style settings (style sheet or rcParams):

import matplotlib as mpl

mpl.rcParams['axes.spines.left'] = False
mpl.rcParams['axes.spines.right'] = False
mpl.rcParams['axes.spines.top'] = False
mpl.rcParams['axes.spines.bottom'] = False

Problem

I had a similar problem using axes. The class parameter is frameon but the kwarg is frame_on. axes_api
>>> plt.gca().set(frameon=False)
AttributeError: Unknown property frameon

Solution

frame_on

Example

data = range(100)
import matplotlib.pyplot as plt
fig, ax = plt.subplots()
ax.plot(data)
#ax.set(frameon=False)  # Old
ax.set(frame_on=False)  # New
plt.show()

I use to do so:

from pylab import *
axes(frameon = 0)
...
show()

To remove the frame of the chart

for spine in plt.gca().spines.values():
  spine.set_visible(False)

I hope this could work

df = pd.DataFrame({
'client_scripting_ms' : client_scripting_ms,
 'apimlayer' : apimlayer, 'server' : server
}, index = index)

ax = df.plot(kind = 'barh', 
     stacked = True,
     title = "Chart",
     width = 0.20, 
     align='center', 
     figsize=(7,5))

plt.legend(loc='upper right', frameon=True)

ax.spines['right'].set_visible(False)
ax.spines['top'].set_visible(False)

ax.yaxis.set_ticks_position('left')
ax.xaxis.set_ticks_position('right')

plt.box(False)
plt.xticks([])
plt.yticks([])
plt.savefig('fig.png')

should do the trick.

plt.axis('off')
plt.savefig(file_path, bbox_inches="tight", pad_inches = 0)

plt.savefig has those options in itself, just need to set axes off before

surprisingly I didn’t find a straight-forward description on how to draw a circle with matplotlib.pyplot (please no pylab) taking as input center (x,y) and radius r. I tried some variants of this:

import matplotlib.pyplot as plt
circle=plt.Circle((0,0),2)
# here must be something like circle.plot() or not?
plt.show()

… but still didn’t get it working.

You need to add it to an axes. A Circle is a subclass of an Artist, and an axes has an add_artist method.

Here’s an example of doing this:

import matplotlib.pyplot as plt

circle1 = plt.Circle((0, 0), 0.2, color='r')
circle2 = plt.Circle((0.5, 0.5), 0.2, color='blue')
circle3 = plt.Circle((1, 1), 0.2, color='g', clip_on=False)

fig, ax = plt.subplots() # note we must use plt.subplots, not plt.subplot
# (or if you have an existing figure)
# fig = plt.gcf()
# ax = fig.gca()

ax.add_artist(circle1)
ax.add_artist(circle2)
ax.add_artist(circle3)

fig.savefig('plotcircles.png')

This results in the following figure:

The first circle is at the origin, but by default clip_on is True, so the circle is clipped when ever it extends beyond the axes. The third (green) circle shows what happens when you don’t clip the Artist. It extends beyond the axes (but not beyond the figure, ie the figure size is not automatically adjusted to plot all of your artists).

The units for x, y and radius correspond to data units by default. In this case, I didn’t plot anything on my axes (fig.gca() returns the current axes), and since the limits have never been set, they defaults to an x and y range from 0 to 1.

Here’s a continuation of the example, showing how units matter:

circle1 = plt.Circle((0, 0), 2, color='r')
# now make a circle with no fill, which is good for hi-lighting key results
circle2 = plt.Circle((5, 5), 0.5, color='b', fill=False)
circle3 = plt.Circle((10, 10), 2, color='g', clip_on=False)

ax = plt.gca()
ax.cla() # clear things for fresh plot

# change default range so that new circles will work
ax.set_xlim((0, 10))
ax.set_ylim((0, 10))
# some data
ax.plot(range(11), 'o', color='black')
# key data point that we are encircling
ax.plot((5), (5), 'o', color='y')

ax.add_artist(circle1)
ax.add_artist(circle2)
ax.add_artist(circle3)
fig.savefig('plotcircles2.png')

which results in:

You can see how I set the fill of the 2nd circle to False, which is useful for encircling key results (like my yellow data point).

import matplotlib.pyplot as plt
circle1=plt.Circle((0,0),.2,color='r')
plt.gcf().gca().add_artist(circle1)

A quick condensed version of the accepted answer, to quickly plug a circle into an existing plot. Refer to the accepted answer and other answers to understand the details.

By the way:

• gcf() means Get Current Figure
• gca() means Get Current Axis

If you want to plot a set of circles, you might want to see this post or this gist(a bit newer). The post offered a function named circles.

The function circles works like scatter, but the sizes of plotted circles are in data unit.

Here’s an example:

from pylab import *
figure(figsize=(8,8))
ax=subplot(aspect='equal')

#plot one circle (the biggest one on bottom-right)
circles(1, 0, 0.5, 'r', alpha=0.2, lw=5, edgecolor='b', transform=ax.transAxes)

#plot a set of circles (circles in diagonal)
a=arange(11)
out = circles(a, a, a*0.2, c=a, alpha=0.5, edgecolor='none')
colorbar(out)

xlim(0,10)
ylim(0,10)

#!/usr/bin/python
import matplotlib.pyplot as plt
import numpy as np

def xy(r,phi):
  return r*np.cos(phi), r*np.sin(phi)

fig = plt.figure()
ax = fig.add_subplot(111,aspect='equal')  

phis=np.arange(0,6.28,0.01)
r =1.
ax.plot( *xy(r,phis), c='r',ls='-' )
plt.show()

Or, if you prefer, look at the paths, http://matplotlib.sourceforge.net/users/path_tutorial.html

If you aim to have the “circle” maintain a visual aspect ratio of 1 no matter what the data coordinates are, you could use the scatter() method. http://matplotlib.org/1.3.1/api/pyplot_api.html#matplotlib.pyplot.scatter

import matplotlib.pyplot as plt
x = [1, 2, 3, 4, 5]
y = [10, 20, 30, 40, 50]
r = [100, 80, 60, 40, 20] # in points, not data units
fig, ax = plt.subplots(1, 1)
ax.scatter(x, y, s=r)
fig.show()

Extending the accepted answer for a common usecase. In particular:

1. View the circles at a natural aspect ratio.

2. Automatically extend the axes limits to include the newly plotted circles.

Self-contained example:

import matplotlib.pyplot as plt

fig, ax = plt.subplots()
ax.add_patch(plt.Circle((0, 0), 0.2, color='r', alpha=0.5))
ax.add_patch(plt.Circle((1, 1), 0.5, color='#00ffff', alpha=0.5))
ax.add_artist(plt.Circle((1, 0), 0.5, color='#000033', alpha=0.5))

#Use adjustable='box-forced' to make the plot area square-shaped as well.
ax.set_aspect('equal', adjustable='datalim')
ax.plot()   #Causes an autoscale update.
plt.show()

Note the difference between ax.add_patch(..) and ax.add_artist(..): of the two, only the former makes autoscaling machinery take the circle into account (reference: discussion), so after running the above code we get:

See also: set_aspect(..) documentation.

I see plots with the use of (.circle) but based on what you might want to do you can also try this out:

import matplotlib.pyplot as plt
import numpy as np

x = list(range(1,6))
y = list(range(10, 20, 2))

print(x, y)

for i, data in enumerate(zip(x,y)):
    j, k = data
    plt.scatter(j,k, marker = "o", s = ((i+1)**4)*50, alpha = 0.3)

centers = np.array([[5,18], [3,14], [7,6]])
m, n = make_blobs(n_samples=20, centers=[[5,18], [3,14], [7,6]], n_features=2, 
cluster_std = 0.4)
colors = ['g', 'b', 'r', 'm']

plt.figure(num=None, figsize=(7,6), facecolor='w', edgecolor='k')
plt.scatter(m[:,0], m[:,1])

for i in range(len(centers)):

    plt.scatter(centers[i,0], centers[i,1], color = colors[i], marker = 'o', s = 13000, alpha = 0.2)
    plt.scatter(centers[i,0], centers[i,1], color = 'k', marker = 'x', s = 50)

plt.savefig('plot.png')

Hello I have written a code for drawing a circle. It will help for drawing all kind of circles. The image shows the circle with radius 1 and center at 0,0 The center and radius can be edited of any choice.

## Draw a circle with center and radius defined
## Also enable the coordinate axes
import matplotlib.pyplot as plt
import numpy as np
# Define limits of coordinate system
x1 = -1.5
x2 = 1.5
y1 = -1.5
y2 = 1.5

circle1 = plt.Circle((0,0),1, color = 'k', fill = False, clip_on = False)
fig, ax = plt.subplots()
ax.add_artist(circle1)
plt.axis("equal")
ax.spines['left'].set_position('zero')
ax.spines['bottom'].set_position('zero')
ax.spines['right'].set_color('none')
ax.spines['top'].set_color('none')
ax.xaxis.set_ticks_position('bottom')
ax.yaxis.set_ticks_position('left')
plt.xlim(left=x1)
plt.xlim(right=x2)
plt.ylim(bottom=y1)
plt.ylim(top=y2)
plt.axhline(linewidth=2, color='k')
plt.axvline(linewidth=2, color='k')

##plt.grid(True)
plt.grid(color='k', linestyle='-.', linewidth=0.5)
plt.show()

Good luck

Similarly to scatter plot you can also use normal plot with circle line style. Using markersize parameter you can adjust radius of a circle:

import matplotlib.pyplot as plt

plt.plot(200, 2, 'o', markersize=7)