This is a slightly tricky one to explain. Basically, I want to make an inset plot and then utilize the convenience of mpl_toolkits.axes_grid1.inset_locator.mark_inset, but I want the data in the inset plot to be completely independent of the data in the parent axes.
Example code with the functions I'd like to use:
import numpy as np
import matplotlib.pyplot as plt
from mpl_toolkits.axes_grid1.inset_locator import inset_axes
from mpl_toolkits.axes_grid1.inset_locator import mark_inset
from mpl_toolkits.axes_grid1.inset_locator import InsetPosition
data = np.random.normal(size=(2000,2000))
plt.imshow(data, origin='lower')
parent_axes = plt.gca()
ax2 = inset_axes(parent_axes, 1, 1)
ax2.plot([900,1100],[900,1100])
# I need more control over the position of the inset axes than is given by the inset_axes function
ip = InsetPosition(parent_axes,[0.7,0.7,0.3,0.3])
ax2.set_axes_locator(ip)
# I want to be able to control where the mark is connected to, independently of the data in the ax2.plot call
mark_inset(parent_axes, ax2, 2,4)
# plt.savefig('./inset_example.png')
plt.show()
The example code produces the following image:
So to sum up: The location of the blue box is entire controlled by the input data to ax2.plot(). I would like to manually place the blue box and enter whatever I want into ax2. Is this possible?
quick edit: to be clear, I understand why inset plots would have the data linked, as that's the most likely usage. So if there's a completely different way in matplotlib to accomplish this, do feel free to reply with that. However, I am trying to avoid manually placing boxes and lines to all of the axes I would place, as I need quite a few insets into a large image.
If I understand correctly, you want an arbitrarily scaled axis at a given position that looks like a zoomed inset, but has no connection to the inset marker's position.
Following your approach you can simply add another axes to the plot and position it at the same spot of the true inset, using the set_axes_locator(ip) function. Since this axis is drawn after the original inset, it will be on top of it and you'll only need to hide the tickmarks of the original plot to let it disappear completely (set_visible(False) does not work here, as it would hide the lines between the inset and the marker position).
import numpy as np
import matplotlib.pyplot as plt
from mpl_toolkits.axes_grid1.inset_locator import inset_axes, mark_inset, InsetPosition
data = np.random.normal(size=(200,200))
plt.imshow(data, origin='lower')
parent_axes = plt.gca()
ax2 = inset_axes(parent_axes, 1, 1)
ax2.plot([60,75],[90,110])
# hide the ticks of the linked axes
ax2.set_xticks([])
ax2.set_yticks([])
#add a new axes to the plot and plot whatever you like
ax3 = plt.gcf().add_axes([0,0,1,1])
ax3.plot([0,3,4], [2,3,1], marker=ur'$\u266B$' , markersize=30, linestyle="")
ax3.set_xlim([-1,5])
ax3.set_ylim([-1,5])
ip = InsetPosition(parent_axes,[0.7,0.7,0.3,0.3])
ax2.set_axes_locator(ip)
# set the new axes (ax3) to the position of the linked axes
ax3.set_axes_locator(ip)
# I want to be able to control where the mark is connected to, independently of the data in the ax2.plot call
mark_inset(parent_axes, ax2, 2,4)
plt.show()
FWIW, I came up with a hack that works.
In the source code for inset_locator, I added a version of mark_inset that takes another set of axes used to define the TransformedBbox:
def mark_inset_hack(parent_axes, inset_axes, hack_axes, loc1, loc2, **kwargs):
rect = TransformedBbox(hack_axes.viewLim, parent_axes.transData)
pp = BboxPatch(rect, **kwargs)
parent_axes.add_patch(pp)
p1 = BboxConnector(inset_axes.bbox, rect, loc1=loc1, **kwargs)
inset_axes.add_patch(p1)
p1.set_clip_on(False)
p2 = BboxConnector(inset_axes.bbox, rect, loc1=loc2, **kwargs)
inset_axes.add_patch(p2)
p2.set_clip_on(False)
return pp, p1, p2
Then in my original-post code I make an inset axis where I want the box to be, pass it to my hacked function, and make it invisible:
# location of desired axes
axdesire = inset_axes(parent_axes,1,1)
axdesire.plot([100,200],[100,200])
mark_inset_hack(parent_axes, ax2, axdesire, 2,4)
axdesire.set_visible(False)
Now I have a marked box at a different location in data units than the inset that I'm marking:
It is certainly a total hack, and at this point I'm not sure it's cleaner than simply drawing lines manually, but I think for a lot of insets this will keep things conceptually cleaner.
Other ideas are still welcome.
Related
I'd like to plot two scatter plots into the same Axes and turn the upper one's data points transparent such that the other plot shines through. However, I want the whole upper plot to have a homogeneous transparency level, such that superimposed markers of the upper plot do not add up their opacity as they would do if I simply set alpha=0.5.
In other words, I'd like both scatter plots to be rendered first and being set to one constant transparency level. Technically this should be possible for both raster and vector graphics (as SVG supports layer transparency, afaik), but either would be fine for me.
Here is some example code that displays what I do not want to achieve. ;)
import numpy as np
import matplotlib.pyplot as plt
fig = plt.figure(1, figsize=(6,4), dpi=160)
ax = fig.gca()
s1 = ax.scatter(np.random.randn(1000), np.random.randn(1000), color="b", edgecolors="none")
s2 = ax.scatter(np.random.randn(1000), np.random.randn(1000), color="g", edgecolors="none")
s2.set_alpha(0.5) # sadly the same as setting `alpha=0.5`
fig.show() # or display(fig)
I'd like the green markers around (2,2) to not be darker where they superimpose, for example. Is this possible with matplotlib?
Thanks for your time! :)
After searching some more, I found related questions and two solutions, of which at least one kind of works for me:
As I hoped one can render one layer and then afterwards blend them together like this:
import numpy as np
import matplotlib.pyplot as plt
fig = plt.figure(1, figsize=(6,4), dpi=160)
ax1 = fig.gca()
s1 = ax1.scatter(np.random.randn(1000), np.random.randn(1000), color="#3355ff", edgecolors="none")
ax1.set_xlim(-3.5,3.5)
ax1.set_ylim(-3.5,3.5)
ax1.patch.set_facecolor("none")
ax1.patch.set_edgecolor("none")
fig.canvas.draw()
w, h = fig.canvas.get_width_height()
img1 = np.frombuffer(fig.canvas.buffer_rgba(), np.uint8).reshape(h, w, -1).copy()
ax1.clear()
s2 = ax1.scatter(np.random.randn(1000), np.random.randn(1000), color="#11aa44", edgecolors="none")
ax1.set_xlim(-3.5,3.5)
ax1.set_ylim(-3.5,3.5)
ax1.patch.set_facecolor("none")
ax1.patch.set_edgecolor("none")
fig.canvas.draw()
img2 = np.frombuffer(fig.canvas.buffer_rgba(), np.uint8).reshape(h, w, -1).copy()
fig.clf()
plt.imshow(np.minimum(img1,img2))
plt.subplots_adjust(0, 0, 1, 1)
plt.axis("off")
plt.show()
I may have to come up with better methods than just taking the np.minimum of both layers to keep more color options (and probably save the axes and labels to a third layer).
I did not try mplcairo as suggested in the other linked answer as it has too many dependencies for my use case (my solution should be portable).
I am still happy for additional input. :)
The Problem
How can I access the colorbar instance created when plotting a GeoDataFrame? In this example, I've plotted troop movements and sizes during the French invasion of Russia, with army sizes less than 10000 plotted in red. How do I get the colorbar to show that red means under 10000?
MCVE:
import pandas as pd
import matplotlib as mpl
import matplotlib.pyplot as plt
from shapely.geometry import Point, LineString
import geopandas as gpd
# ingest troop movement data
PATH = ("https://vincentarelbundock.github.io/Rdatasets/csv/HistData/"
"Minard.troops.csv")
troops = pd.read_csv(PATH)
troops['geometry'] = troops.apply(lambda row: Point(row.long, row.lat), axis=1)
troops = gpd.GeoDataFrame(troops)
# get army group paths
grouped = troops.groupby('group')
groups = [[LineString(group[['long', 'lat']].values), name]
for name, group in grouped]
groups = pd.DataFrame(groups, columns=['geometry', 'army group'])
groups = gpd.GeoDataFrame(groups)
groups.plot(color=['red', 'green', 'blue'], lw=0.5)
# plot troop sizes
cmap = mpl.cm.get_cmap('cool')
cmap.set_under('red')
troops.plot(column='survivors', ax=plt.gca(),
cmap=cmap, vmin=10000, legend=True, markersize=50)
Output:
Here, legend=True in the last line adds the colorbar.
Attempts at a solution
I know that if I made the colorbar myself, I could just pass the argument extend='min' to add a red triangle to the bottom of the colorbar.
And I know I can get the colorbar axes through (as suggested in this answer):
cax = plt.gcf().axes[1]
but I don't know how that helps me edit the colorbar. I can't even add a label with cax.set_label('troop size'). (i.e. I can't see that label anywhere although cax.get_label() does return 'troop size')
These axes appear to consist of two polygons:
In[315]: cax.artists
Out[315]:
[<matplotlib.patches.Polygon at 0x1e0e73c8>,
<matplotlib.patches.Polygon at 0x190c8f28>]
No idea what to make of that. And even if I could find the actual colorbar instance, I wouldn't know how to extend it as the docs for the Colorbar class don't mention anything like that.
Alternatives
Is there a way to pass the extend keyword through the GeoDataFrame.plot function?
Can I somehow access the colorbar instance by either saving it when plotting or finding it in the figure?
How would I go about constructing the colorbar directly with Matplotlib? And how do I avoid that it deviates from the plot if I change parameters?
There is an issue (and PR) about making it possible to pass keywords to the colorbar construction: https://github.com/geopandas/geopandas/issues/697.
But for now, the best work-around is to create the colorbar yourself I think:
Create the same figure but with legend=False:
cmap = mpl.cm.get_cmap('cool')
cmap.set_under('red')
ax = troops.plot(column='survivors', cmap=cmap, vmin=10000, legend=False, markersize=50)
Now, we get the collection created for the points (from a scatter plot) as the first element of the ax.collections, so we can instruct matplotlib to create a colorbar based on this mapping (and now we can pass additional keywords):
scatter = ax.collections[0]
plt.colorbar(scatter, ax=ax, extend='min')
this gives me
Using Matplotlib I'd like to remove the grid lines inside the plot, while keeping the frame (i.e. the axes lines). I've tried the code below and other options as well, but I can't get it to work. How do I simply keep the frame while removing the grid lines?
I'm doing this to reproduce a ggplot2 plot in matplotlib. I've created a MWE below. Be aware that you need a relatively new version of matplotlib to use the ggplot2 style.
import matplotlib
matplotlib.use('Agg')
import matplotlib.pyplot as plt
import pylab as P
import numpy as np
if __name__ == '__main__':
values = np.random.uniform(size=20)
plt.style.use('ggplot')
fig = plt.figure()
_, ax1 = P.subplots()
weights = np.ones_like(values)/len(values)
plt.hist(values, bins=20, weights=weights)
ax1.set_xlabel('Value')
ax1.set_ylabel('Probability')
ax1.grid(b=False)
#ax1.yaxis.grid(False)
#ax1.xaxis.grid(False)
ax1.set_axis_bgcolor('white')
ax1.set_xlim([0,1])
P.savefig('hist.pdf', bbox_inches='tight')
OK, I think this is what you are asking (but correct me if I misunderstood):
You need to change the colour of the spines. You need to do this for each spine individually, using the set_color method:
for spine in ['left','right','top','bottom']:
ax1.spines[spine].set_color('k')
You can see this example and this example for more about using spines.
However, if you have removed the grey background and the grid lines, and added the spines, this is not really in the ggplot style any more; is that really the style you want to use?
EDIT
To make the edge of the histogram bars touch the frame, you need to either:
Change your binning, so the bin edges go to 0 and 1
n,bins,patches = plt.hist(values, bins=np.linspace(0,1,21), weights=weights)
# Check, by printing bins:
print bins[0], bins[-1]
# 0.0, 1.0
If you really want to keep the bins to go between values.min() and values.max(), you would need to change your plot limits to no longer be 0 and 1:
n,bins,patches = plt.hist(values, bins=20, weights=weights)
ax.set_xlim(bins[0],bins[-1])
The following code gives me a plot with significant margins above and below the figure. I don't know how to eliminate the noticeable margins. subplots_adjust does not work as expected.
import matplotlib.pyplot as plt
import numpy as np
fig = plt.figure()
ax = fig.add_subplot(111)
ax.plot(range(10),range(10))
ax.set_aspect('equal')
plt.tight_layout()
tight_layout eliminates some of the margin, but not all of the margins.
What I wanted is actually setting the aspect ratio to any customized value and eliminating the white space at the same time.
Update: as Pierre H. puts it, the key is to change the size of the figure container. So my question is: Could you suggest a way to accommodate the size of the figure to the size of the axes with arbitrary aspect ratio?
In other words, first I create a figure and an axes on it, and then I change the size of the axes (by changing aspect ratio for example), which in general will leave a portion of the figure container empty. At this stage, we need to change the size of the figure accordingly to eliminate the blank space on the figure container.
I just discovered how to eliminate all margins from my figures. I didn't use tight_layout(), instead I used:
import matplotlib.pyplot as plt
fig = plt.figure(figsize=(20,20))
ax = plt.subplot(111,aspect = 'equal')
plt.subplots_adjust(left=0, bottom=0, right=1, top=1, wspace=0, hspace=0)
Hope this helps.
After plotting your chart you can easily manipulate margins this way:
plot_margin = 0.25
x0, x1, y0, y1 = plt.axis()
plt.axis((x0 - plot_margin,
x1 + plot_margin,
y0 - plot_margin,
y1 + plot_margin))
This example could be changed to the aspect ratio you want or change the margins as you really want.
In other stacktoverflow posts many questions related to margins could make use of this simpler approach.
Best regards.
tight_layout(pad=0) will meet your need.
http://matplotlib.org/api/pyplot_api.html#matplotlib.pyplot.tight_layout
I think what you need is, and it works well for me.
plt.axis('tight')
This command will automatically scale the axis to fit tightly to the data. Also check the answer of Nuno Aniceto for a customized axis. The documents are in https://matplotlib.org/api/pyplot_api.html#matplotlib.pyplot.axis.
Be aware that
plt.savefig(filename, bbox_inches='tight')
will help remove white space of all the figure including labels, etc, but not the white space inside the axes.
You should use add_axes if you want exact control of the figure layout. eg.
left = 0.05
bottom = 0.05
width = 0.9
height = 0.9
ax = fig.add_axes([left, bottom, width, height])
I think the subplot_adjust call is irrelevant here since the adjustment is overridden by tight_layout. Anyway, this only change the size of the axes inside the figure.
As tcaswell pointed it, you need to change the size of the figure. Either at creation (my proposition below) or after, using fig.set_size_inches. I'm here creating a figure with a 1:1 aspect ratio using the figsize=(6,6) argument (of course 6 inches is an arbitrary choice):
import matplotlib.pyplot as plt
import numpy as np
fig = plt.figure(figsize=(6,6))
ax = fig.add_subplot(111)
ax.plot(range(10),range(10))
ax.set_aspect('equal')
plt.tight_layout()
You can use like:
plt.subplots_adjust(wspace=1,hspace=0.5,left=0.1,top=0.9,right=0.9,bottom=0.1)
And delete the item bbox_inches='tight' in plt.savefig().
I am working on using Matplotlib to produce plots of implicit equations (eg. y^x=x^y). With many thanks to the help I have already received I have got quite far with it. I have used a contour line to produce the plot. My remaining problem is with formatting the contour line eg width, color and especially zorder, where the contour appears behind my gridlines. These work fine when plotting a standard function of course.
import matplotlib.pyplot as plt
from matplotlib.ticker import MultipleLocator, FormatStrFormatter
import numpy as np
fig = plt.figure(1)
ax = fig.add_subplot(111)
# set up axis
ax.spines['left'].set_position('zero')
ax.spines['right'].set_color('none')
ax.spines['bottom'].set_position('zero')
ax.spines['top'].set_color('none')
ax.xaxis.set_ticks_position('bottom')
ax.yaxis.set_ticks_position('left')
# setup x and y ranges and precision
x = np.arange(-0.5,5.5,0.01)
y = np.arange(-0.5,5.5,0.01)
# draw a curve
line, = ax.plot(x, x**2,zorder=100,linewidth=3,color='red')
# draw a contour
X,Y=np.meshgrid(x,y)
F=X**Y
G=Y**X
ax.contour(X,Y,(F-G),[0],zorder=100,linewidth=3,color='green')
#set bounds
ax.set_xbound(-1,7)
ax.set_ybound(-1,7)
#add gridlines
ax.xaxis.set_minor_locator(MultipleLocator(0.2))
ax.yaxis.set_minor_locator(MultipleLocator(0.2))
ax.xaxis.grid(True,'minor',linestyle='-',color='0.8')
ax.yaxis.grid(True,'minor',linestyle='-',color='0.8')
plt.show()
This is rather hackish but...
Apparently in the current release Matplotlib does not support zorder on contours. This support, however, was recently added to the trunk.
So, the right way to do this is either to wait for the 1.0 release or just go ahead and re-install from trunk.
Now, here's the hackish part. I did a quick test and if I changed line 618 in
python/site-packages/matplotlib/contour.py
to add a zorder into the collections.LineCollection call, it fixes your specific problem.
col = collections.LineCollection(nlist,
linewidths = width,
linestyle = lstyle,
alpha=self.alpha,zorder=100)
Not the right way to do things, but might just work in a pinch.
Also off-topic, if you accept some responses to your previous questions, you probably get quicker help around here. People love those rep points :)