I'm drawing the bloxplot shown below using python and matplotlib. Is there any way I can reduce the distance between the two boxplots on the X axis?
This is the code that I'm using to get the figure above:
import matplotlib.pyplot as plt
from matplotlib import rcParams
rcParams['ytick.direction'] = 'out'
rcParams['xtick.direction'] = 'out'
fig = plt.figure()
xlabels = ["CG", "EG"]
ax = fig.add_subplot(111)
ax.boxplot([values_cg, values_eg])
ax.set_xticks(np.arange(len(xlabels))+1)
ax.set_xticklabels(xlabels, rotation=45, ha='right')
fig.subplots_adjust(bottom=0.3)
ylabels = yticks = np.linspace(0, 20, 5)
ax.set_yticks(yticks)
ax.set_yticklabels(ylabels)
ax.tick_params(axis='x', pad=10)
ax.tick_params(axis='y', pad=10)
plt.savefig(os.path.join(output_dir, "output.pdf"))
And this is an example closer to what I'd like to get visually (although I wouldn't mind if the boxplots were even a bit closer to each other):
You can either change the aspect ratio of plot or use the widths kwarg (doc) as such:
ax.boxplot([values_cg, values_eg], widths=1)
to make the boxes wider.
Try changing the aspect ratio using
ax.set_aspect(1.5) # or some other float
The larger then number, the narrower (and taller) the plot should be:
a circle will be stretched such that the height is num times the width. aspect=1 is the same as aspect=’equal’.
http://matplotlib.org/api/axes_api.html#matplotlib.axes.Axes.set_aspect
When your code writes:
ax.set_xticks(np.arange(len(xlabels))+1)
You're putting the first box plot on 0 and the second one on 1 (event though you change the tick labels afterwards), just like in the second, "wanted" example you gave they are set on 1,2,3.
So i think an alternative solution would be to play with the xticks position and the xlim of the plot.
for example using
ax.set_xlim(-1.5,2.5)
would place them closer.
positions : array-like, optional
Sets the positions of the boxes. The ticks and limits are automatically set to match the positions. Defaults to range(1, N+1) where N is the number of boxes to be drawn.
https://matplotlib.org/3.1.1/api/_as_gen/matplotlib.pyplot.boxplot.html
This should do the job!
As #Stevie mentioned, you can use the positions kwarg (doc) to manually set the x-coordinates of the boxes:
ax.boxplot([values_cg, values_eg], positions=[1, 1.3])
Related
I want to create a bar-plot in python. I want this plot to be beautiful though and I don't like the looks of python's axes.bar() function. Therefore, I have decided to use plt.vlines(). The challenge here is that my x-data is a list that contains strings and not numerical data. When I plot my graph, the spacing between the two columns (in my example column 2 = 0) is pretty big:
Furthermore, I want a grid. However, I would like to have minor grid lines as well. I know how to get all of this if my data was numerical. But since my x-data contains strings, I don't know how to set x_max. Any suggestions?
Internally, the positions of the labels are numbered 0,1,... So setting the x-limits a bit before 0 and after the last, shows them more centered.
Usually, bars are drawn with their 'feet' on the ground, which can be set via plt.ylim(0, ...). Minor ticks can be positioned for example at multiples of 0.2. Setting the length of the ticks to zero lets the position count for the grid, but suppresses the tick mark.
from matplotlib import pyplot as plt
from matplotlib.ticker import MultipleLocator
import numpy as np
labels = ['Test 1', 'Test 2']
values = [1, 0.7]
fig, ax = plt.subplots()
plt.vlines(labels, 0, values, colors='dodgerblue', alpha=.4, lw=7)
plt.xlim(-0.5, len(labels) - 0.5) # add some padding left and right of the bars
plt.ylim(0, 1.1) # bars usually have their 0 at the bottom
ax.xaxis.set_minor_locator(MultipleLocator(.2))
plt.tick_params(axis='x', which='both', length=0) # ticks not shown, but position serves for gridlines
plt.grid(axis='both', which='both', ls=':') # optionally set the linestyle of the grid
plt.show()
I can't figure out how to rotate the text on the X Axis. Its a time stamp, so as the number of samples increase, they get closer and closer until they overlap. I'd like to rotate the text 90 degrees so as the samples get closer together, they aren't overlapping.
Below is what I have, it works fine with the exception that I can't figure out how to rotate the X axis text.
import sys
import matplotlib
matplotlib.use('Agg')
import matplotlib.pyplot as plt
import datetime
font = {'family' : 'normal',
'weight' : 'bold',
'size' : 8}
matplotlib.rc('font', **font)
values = open('stats.csv', 'r').readlines()
time = [datetime.datetime.fromtimestamp(float(i.split(',')[0].strip())) for i in values[1:]]
delay = [float(i.split(',')[1].strip()) for i in values[1:]]
plt.plot(time, delay)
plt.grid(b='on')
plt.savefig('test.png')
This works for me:
plt.xticks(rotation=90)
Many "correct" answers here but I'll add one more since I think some details are left out of several. The OP asked for 90 degree rotation but I'll change to 45 degrees because when you use an angle that isn't zero or 90, you should change the horizontal alignment as well; otherwise your labels will be off-center and a bit misleading (and I'm guessing many people who come here want to rotate axes to something other than 90).
Easiest / Least Code
Option 1
plt.xticks(rotation=45, ha='right')
As mentioned previously, that may not be desirable if you'd rather take the Object Oriented approach.
Option 2
Another fast way (it's intended for date objects but seems to work on any label; doubt this is recommended though):
fig.autofmt_xdate(rotation=45)
fig you would usually get from:
fig = plt.gcf()
fig = plt.figure()
fig, ax = plt.subplots()
fig = ax.figure
Object-Oriented / Dealing directly with ax
Option 3a
If you have the list of labels:
labels = ['One', 'Two', 'Three']
ax.set_xticks([1, 2, 3])
ax.set_xticklabels(labels, rotation=45, ha='right')
In later versions of Matplotlib (3.5+), you can just use set_xticks alone:
ax.set_xticks([1, 2, 3], labels, rotation=45, ha='right')
Option 3b
If you want to get the list of labels from the current plot:
# Unfortunately you need to draw your figure first to assign the labels,
# otherwise get_xticklabels() will return empty strings.
plt.draw()
ax.set_xticks(ax.get_xticks())
ax.set_xticklabels(ax.get_xticklabels(), rotation=45, ha='right')
As above, in later versions of Matplotlib (3.5+), you can just use set_xticks alone:
ax.set_xticks(ax.get_xticks(), ax.get_xticklabels(), rotation=45, ha='right')
Option 4
Similar to above, but loop through manually instead.
for label in ax.get_xticklabels():
label.set_rotation(45)
label.set_ha('right')
Option 5
We still use pyplot (as plt) here but it's object-oriented because we're changing the property of a specific ax object.
plt.setp(ax.get_xticklabels(), rotation=45, ha='right')
Option 6
This option is simple, but AFAIK you can't set label horizontal align this way so another option might be better if your angle is not 90.
ax.tick_params(axis='x', labelrotation=45)
Edit:
There's discussion of this exact "bug" but a fix hasn't been released (as of 3.4.0):
https://github.com/matplotlib/matplotlib/issues/13774
Easy way
As described here, there is an existing method in the matplotlib.pyplot figure class that automatically rotates dates appropriately for you figure.
You can call it after you plot your data (i.e.ax.plot(dates,ydata) :
fig.autofmt_xdate()
If you need to format the labels further, checkout the above link.
Non-datetime objects
As per languitar's comment, the method I suggested for non-datetime xticks would not update correctly when zooming, etc. If it's not a datetime object used as your x-axis data, you should follow Tommy's answer:
for tick in ax.get_xticklabels():
tick.set_rotation(45)
Try pyplot.setp. I think you could do something like this:
x = range(len(time))
plt.xticks(x, time)
locs, labels = plt.xticks()
plt.setp(labels, rotation=90)
plt.plot(x, delay)
Appart from
plt.xticks(rotation=90)
this is also possible:
plt.xticks(rotation='vertical')
I came up with a similar example. Again, the rotation keyword is.. well, it's key.
from pylab import *
fig = figure()
ax = fig.add_subplot(111)
ax.bar( [0,1,2], [1,3,5] )
ax.set_xticks( [ 0.5, 1.5, 2.5 ] )
ax.set_xticklabels( ['tom','dick','harry'], rotation=45 ) ;
If you want to apply rotation on the axes object, the easiest way is using tick_params. For example.
ax.tick_params(axis='x', labelrotation=90)
Matplotlib documentation reference here.
This is useful when you have an array of axes as returned by plt.subplots, and it is more convenient than using set_xticks because in that case you need to also set the tick labels, and also more convenient that those that iterate over the ticks (for obvious reasons)
If using plt:
plt.xticks(rotation=90)
In case of using pandas or seaborn to plot, assuming ax as axes for the plot:
ax.set_xticklabels(ax.get_xticklabels(), rotation=90)
Another way of doing the above:
for tick in ax.get_xticklabels():
tick.set_rotation(45)
My answer is inspired by cjohnson318's answer, but I didn't want to supply a hardcoded list of labels; I wanted to rotate the existing labels:
for tick in ax.get_xticklabels():
tick.set_rotation(45)
The simplest solution is to use:
plt.xticks(rotation=XX)
but also
# Tweak spacing to prevent clipping of tick-labels
plt.subplots_adjust(bottom=X.XX)
e.g for dates I used rotation=45 and bottom=0.20 but you can do some test for your data
import pylab as pl
pl.xticks(rotation = 90)
To rotate the x-axis label to 90 degrees
for tick in ax.get_xticklabels():
tick.set_rotation(45)
It will depend on what are you plotting.
import matplotlib.pyplot as plt
x=['long_text_for_a_label_a',
'long_text_for_a_label_b',
'long_text_for_a_label_c']
y=[1,2,3]
myplot = plt.plot(x,y)
for item in myplot.axes.get_xticklabels():
item.set_rotation(90)
For pandas and seaborn that give you an Axes object:
df = pd.DataFrame(x,y)
#pandas
myplot = df.plot.bar()
#seaborn
myplotsns =sns.barplot(y='0', x=df.index, data=df)
# you can get xticklabels without .axes cause the object are already a
# isntance of it
for item in myplot.get_xticklabels():
item.set_rotation(90)
If you need to rotate labels you may need change the font size too, you can use font_scale=1.0 to do that.
I want to add a scale indicator to a plot like the one labelled '10kpc' in the (otherwise) empty plot below. So basically, the axis use one unit of measure and I want to indicate a length in the plot in a different unit. It has to have the same style as below, i.e. a |----| bar with text above.
Is there a convenient way in matplotlib to do that or do I have to draw three lines (two small vertical, one horizontal) and add the text? An ideal solution would not even require me to set coordinates in the data dimensions, i.e. I just say something along the line of horizontalalignment='left', verticalalignment='bottom', transform=ax.transAxes and specify only the width in data coordinates.
I fought with annotate() and arrow() and their documentations for quiet a bit until I concluded, they were not exactly useful, but I might be wrong.
Edit:
The code below is the closest, I have come so far. I still don't like having to specify the x-coordinates in the data coordinate system. The only thing I want to specify in data is the width of the bar. The rest should be placed in the plot system and ideally the bar should be placed relative to the text (a few pixels above).
import matplotlib.pyplot as plt
import matplotlib.transforms as tfrms
plt.imshow(somedata)
plt.colorbar()
ax = plt.gca()
trans = tfrms.blended_transform_factory( ax.transData, ax.transAxes )
plt.errorbar( 5, 0.06, xerr=10*arcsecperkpc/2, color='k', capsize=5, transform=trans )
plt.text( 5, 0.05, '10kpc', horizontalalignment='center', verticalalignment='top', transform=trans )
Here is a code that adds a horizontal scale bar (or scale indicator or scalebar) to a plot. The bar's width is given in data units, while the height of the edges is in fraction of axes units.
The solution is based on an AnchoredOffsetbox, which contains a VPacker. The VPacker has a label in its lower row, and an AuxTransformBox in its upper row.
The key here is that the AnchoredOffsetbox is positioned relative to the axes, using the loc argument similar to the legend positioning (e.g. loc=4 denotes the lower right corner). However, the AuxTransformBox contains a set of elements, which are positioned inside the box using a transformation. As transformation we can choose a blended transform which transforms x coordinates according to the data transform of the axes and y coordinates according to the axes transform. A tranformation which does this is actually the xaxis_transform of the axes itself. Supplying this transform to the AuxTransformBox allows us to specify the artists within (which are Line2Ds in this case) in a useful way, e.g. the line of the bar will be Line2D([0,size],[0,0]).
All of this can be packed into a class, subclassing the AnchoredOffsetbox, such that it is easy to be used in an existing code.
import matplotlib.pyplot as plt
import matplotlib.offsetbox
from matplotlib.lines import Line2D
import numpy as np; np.random.seed(42)
x = np.linspace(-6,6, num=100)
y = np.linspace(-10,10, num=100)
X,Y = np.meshgrid(x,y)
Z = np.sin(X)/X+np.sin(Y)/Y
fig, ax = plt.subplots()
ax.contourf(X,Y,Z, alpha=.1)
ax.contour(X,Y,Z, alpha=.4)
class AnchoredHScaleBar(matplotlib.offsetbox.AnchoredOffsetbox):
""" size: length of bar in data units
extent : height of bar ends in axes units """
def __init__(self, size=1, extent = 0.03, label="", loc=2, ax=None,
pad=0.4, borderpad=0.5, ppad = 0, sep=2, prop=None,
frameon=True, linekw={}, **kwargs):
if not ax:
ax = plt.gca()
trans = ax.get_xaxis_transform()
size_bar = matplotlib.offsetbox.AuxTransformBox(trans)
line = Line2D([0,size],[0,0], **linekw)
vline1 = Line2D([0,0],[-extent/2.,extent/2.], **linekw)
vline2 = Line2D([size,size],[-extent/2.,extent/2.], **linekw)
size_bar.add_artist(line)
size_bar.add_artist(vline1)
size_bar.add_artist(vline2)
txt = matplotlib.offsetbox.TextArea(label, minimumdescent=False)
self.vpac = matplotlib.offsetbox.VPacker(children=[size_bar,txt],
align="center", pad=ppad, sep=sep)
matplotlib.offsetbox.AnchoredOffsetbox.__init__(self, loc, pad=pad,
borderpad=borderpad, child=self.vpac, prop=prop, frameon=frameon,
**kwargs)
ob = AnchoredHScaleBar(size=3, label="3 units", loc=4, frameon=True,
pad=0.6,sep=4, linekw=dict(color="crimson"),)
ax.add_artist(ob)
plt.show()
In order to achieve a result as desired in the question, you can set the frame off and adjust the linewidth. Of course the transformation from the units you want to show (kpc) into data units (km?) needs to be done by yourself.
ikpc = lambda x: x*3.085e16 #x in kpc, return in km
ob = AnchoredHScaleBar(size=ikpc(10), label="10kpc", loc=4, frameon=False,
pad=0.6,sep=4, linekw=dict(color="k", linewidth=0.8))
I have a vexing issue with a colorbar and even after vigorous research I cannot find the question even being asked. I have a plot where I overlay a contour and a pcolormesh and I would like a colorbar to indicate values. That works fine except for one thing:
The colorbar frame and color are offset
The colorbar frame and the actual bar are offset such that below you have a white bit in the frame and on top the color is poking out. While the frame is aligned with the axis as desired, the colorbar is offset.
Here is a working example that emulates the situation I was in, i.e. multiple plots with insets.
import matplotlib.gridspec as gridspec
import numpy as np
import matplotlib
import matplotlib.pyplot as plt
figheight = 4.2 - (2.1 - 49.519 / 25.4)
matplotlib.rcParams['figure.figsize'] = (5.25, figheight)
matplotlib.rcParams['axes.linewidth'] = 0.5
fig = plt.figure()
grid = gridspec.GridSpec(2, 1, height_ratios=[49.519 / 25.4 / figheight, 2.1 / figheight])
ax0 = plt.subplot(grid[0, 0])
ax1 = plt.subplot(grid[1, 0])
plt.tight_layout()
###############################################################################################
#
# Define position of inset
#
###############################################################################################
ax1.axis('off')
pos1 = ax1.get_position()
pos2 = matplotlib.transforms.Bbox([[pos1.x0, pos1.y0],
[.8*pos1.x1,
0.8*pos1.height + pos1.y0]])
left, bottom, width, height = [pos2.x0, pos2.y0, pos2.width, pos2.height]
ax2 = fig.add_axes([left, bottom, width, height])
###############################################################################################
#
# ax2 (inset) plot
#
###############################################################################################
pos2 = ax2.get_position()
ax2.axis('on')
x = np.linspace(0,5)
z = (np.outer(np.sin(x), np.cos(x))+1)*0.5
im = ax2.pcolormesh(z)
c = ax2.contour(z, linewidths=7)
ax2pos = ax2.get_position()
cbar_axis = fig.add_axes([ax2pos.x1+0.05,ax2pos.y0, .02, ax2pos.height])
colorbar = fig.colorbar(im, ax = ax2,
cax = cbar_axis, ticks = [0.1, .5, .9])
colorbar.outline.set_visible(True)
plot = 'Minimal.pdf'
fig.savefig(plot)
plt.close()
The problem persists in both the inline display and the saved .pdf if 'Inline' graphics backend is chosen. Using tight layout or not changes how badly the offset is depending on the size of the bar - same with using PyQT5 rather than inline graphics backend. I thought it was gone when I was changing between the various combinations, but I just realized it's still there.
I would appreciate any input.
As suggested by ImportanceOfBeingErnest I have tried using np.round on the figsize and that didn't change things. While you can fiddle around with sizes to make it look okay, it always stands over on one or the other side by some amount. When I change the graphics backend on Spyder 3 from 'Inline' to 'QT5' the problem becomes less severe with or without rounding. A summary of this is in this picture Colorbar overlap cases. Note that with not rounded and PyQT5 the problem still occurs, but is not as severe.
On inspection, it is clear that the colorbar is not only bleeding out over the top of its axes, but it's also positioned slightly to the left.
So, the problem here appears to be a conflict between the position of the colorbar axis and the colorbar itself when rasterization occurs. You can find more details on this issue in matplotlib's github repository, but I'll summarize what's going on here.
Colorbars are rasterized when the output is produced, so as to avoid artifacting issues during rendering. The position of the colorbar is snapped to the nearest integer pixels during the rasterization process, while the axis is kept where it is supposed to be. Then, when the output is produced, the colorbar falls within borders of fixed pixels of the image, despite the fact that the image is, itself, vectorized. Thus, there are two strategies that can be employed to avoid this mishap.
Use a finer DPI
The conversion from vectorized coordinates to rasterized coordinates takes place assuming a given DPI on the image. By default, this is set to be 72. However, by using more DPI, the overall shift induced by the rasterization process will be smaller, as the closest pixel the colorbar will snap to will be much nearer. Here, we change the output to have fig.savefig(plot,dpi=4000), and the problem goes away:
Note, however, that on my machine, the output size changed from 62 KB to 78 KB due to this change (although the DPI adjustment was also, admittedly, extreme). If you are worried about file sizes, you should pick a lower DPI that fixes the problem.
Use a different colormap
This rasterization happens when more than 50 colors are in the colorbar. Thus, we can do a quick test, setting our colormap to Pastel1 via
im = ax2.pcolormesh(z,cmap='Pastel1'). Here, the colorbar / axis mismatch is mitigated.
As a fallback, adopting a colorbar with fewer than 50 colors should mitigate this problem.
Rasterize the Axis
For completeness, there is also a third option. If you rasterize the colorbar axis, both the axis boundaries and the colormap will be rasterized, and you'll lose the offset. This will also rasterize your labels, and the axis will shift as one, breaking alignment with the nearby axis. For this, you just need to include cbar_axis.set_rasterized(True).
First, a way to overlay a contour and a pcolormesh and create a colorbar would be the following
import matplotlib.pyplot as plt
from mpl_toolkits.axes_grid1 import make_axes_locatable
import numpy as np
x = np.linspace(0,5)
z = (np.outer(np.sin(x), np.cos(x))+1)*0.5
fig = plt.figure(figsize=(4, 4))
ax = fig.add_subplot(111)
im = ax.pcolormesh(z)
c = ax.contour(z, linewidths=7)
divider = make_axes_locatable(ax)
cax = divider.append_axes("right", "5%", pad="3%")
colorbar = fig.colorbar(im, cax=cax, ticks = [0.1, .5, .9])
plt.show()
Now to the problem from the question. It is of course possible to create the axes to put the colorbar in manually. Replacing the colorbar creation with the code from the question still produces a nice image.
import matplotlib.pyplot as plt
import numpy as np
x = np.linspace(0,5)
z = (np.outer(np.sin(x), np.cos(x))+1)*0.5
fig = plt.figure(figsize=(4, 4))
ax = fig.add_subplot(111)
plt.subplots_adjust(right=0.8)
im = ax.pcolormesh(z)
c = ax.contour(z, linewidths=7)
ax2pos = ax.get_position()
cbar_axis = fig.add_axes([ax2pos.x1+0.05,ax2pos.y0, .05, ax2pos.height])
colorbar = fig.colorbar(im, ax = ax,
cax = cbar_axis, ticks = [0.1, .5, .9])
colorbar.outline.set_visible(True)
plt.show()
Conclusion so far: The issue is not reproducible, at least not without a Minimal, Complete, and Verifiable example.
I'm uncertain about the reasons for the behaviour in the example from the question. However, it seems that it can be overcome by rounding the figure size to 3 significant digits
matplotlib.rcParams['figure.figsize'] = (5.25, np.round(figheight,3))
Is there a way to increase the padding/offset of the polar plot tick labels (theta)?
import matplotlib
import numpy as np
from matplotlib.pyplot import figure, show, grid
# make a square figure
fig = figure(figsize=(2, 2))
ax = fig.add_axes([0.1, 0.1, 0.8, 0.8], polar=True, axisbg='#d5de9c')
ax.set_yticklabels([])
r = np.arange(0, 3.0, 0.01)
theta = 2*np.pi*r
ax.plot(theta, r, color='#ee8d18', lw=3)
ax.set_rmax(2.0)
show()
I'd like to have theta tick labels further away from the polar plot so they don't overlap.
First of all; seeing as how you have specified the figsize to be (2,2) and having the ax occupy 80 % of both the width and height, you have very little space left over to pad the ticklabels. This could cause the ticklabels to be "cut off" at the figure's egdes. This can easily be "fixed" by either
Specifying bigger figsize
Make the ax occupy less space on the (2,2) sized figure
Use smaller fontsize for the ticklabels
or any combination of these. Another, in my opinion better, solution to this "problem" is to use a subplot rather than specifying the Axes's bounds;
ax = fig.add_subplot(111, polar=True, axisbg='#d5de9c')
as this makes it possible to use the method tight_layout() which automatically configures the figure layout to nicely include all elements.
Then over to the real problem at hand; the padding. On a PolarAxes you can set, among other things, the radial placement of the theta-ticks. This is done by specifying the fraction of the polar axes radius where you want the ticklabels to be placed as an argument to the frac parameter of the PolarAxes's set_thetagrids() method. The argument should be a fraction of the axes' radius where you want the ticklabels placed. I.e. for frac < 1 the ticklabels will be placed inside the axes, while for frac > 1 they will be placed outside the axes.
Your code could then be something like this:
import numpy as np
from matplotlib.pyplot import figure, show, grid, tight_layout
# make a square figure
fig = figure(figsize=(2, 2))
ax = fig.add_subplot(111, polar=True, axisbg='#d5de9c')
ax.set_yticklabels([])
r = np.arange(0, 3.0, 0.01)
theta = 2*np.pi*r
ax.plot(theta, r, color='#ee8d18', lw=3)
ax.set_rmax(2.0)
# tick locations
thetaticks = np.arange(0,360,45)
# set ticklabels location at 1.3 times the axes' radius
ax.set_thetagrids(thetaticks, frac=1.3)
tight_layout()
show()
You should try different values for frac to find a value that is best suited for your needs.
If you don't specify a value to the parameter frac as above, i.e. frac has default value None, the code outputs a plot as below. Notice how the radius of the plot is bigger, as the ticklabels don't "occupy as much space" as in the example above.
As of matplotlib 2.1.0, the functionality of the original answer is now deprecated - polar axes now obey to the parameters of ax.tick_params:
ax.tick_params(pad=123)
should do the trick.