I'm trying to change a colorbar attached to a scatter plot so that the minimum and maximum of the colorbar are the minimum and maximum of the data, but I want the data to be centred at zero as I'm using a colormap with white at zero. Here is my example
import numpy as np
import matplotlib.pyplot as plt
x = np.linspace(0, 1, 61)
y = np.linspace(0, 1, 61)
C = np.linspace(-10, 50, 61)
M = np.abs(C).max() # used for vmin and vmax
fig, ax = plt.subplots(1, 1, figsize=(5,3), dpi=150)
sc=ax.scatter(x, y, c=C, marker='o', edgecolor='k', vmin=-M, vmax=M, cmap=plt.cm.RdBu_r)
cbar=fig.colorbar(sc, ax=ax, label='$R - R_0$ (mm)')
ax.set_xlabel('x')
ax.set_ylabel('y')
As you can see from the attached figure, the colorbar goes down to -M, where as I want the bar to just go down to -10, but if I let vmin=-10 then the colorbar won't be zerod at white. Normally, setting vmin to +/- M when using contourf the colorbar automatically sorts to how I want. This sort of behaviour is what I expect when contourf uses levels=np.linspace(-M,M,61) rather than setting it with vmin and vmax with levels=62. An example showing the default contourf colorbar behaviour I want in my scatter example is shown below
plt.figure(figsize=(6,5), dpi=150)
plt.contourf(x, x, np.reshape(np.linspace(-10, 50, 61*61), (61,61)),
levels=62, vmin=-M, vmax=M, cmap=plt.cm.RdBu_r)
plt.colorbar(label='$R - R_0$ (mm)')
Does anyone have any thoughts? I found this link which I thought might solve the problem, but when executing the cbar.outline.set_ydata line I get this error AttributeError: 'Polygon' object has no attribute 'set_ydata' .
EDIT a little annoyed that someone has closed this question without allowing me to clarify any questions they might have, as none of the proposed solutions are what I'm asking for.
As for Normalize.TwoSlopeNorm, I do not want to rescale the smaller negative side to use the entire colormap range, I just want the colorbar attached to the side of my graph to stop at -10.
This link also does not solve my issue, as it's the TwoSlopeNorm solution again.
After changing the ylim of the colorbar, the rectangle formed by the surrounding spines is too large. You can make this outline invisible. And then add a new rectangular border:
import numpy as np
import matplotlib.pyplot as plt
x = np.linspace(0, 1, 61)
y = np.linspace(0, 1, 61)
C = np.linspace(-10, 50, 61)
M = np.abs(C).max() # used for vmin and vmax
fig, ax = plt.subplots(1, 1, figsize=(5, 3), dpi=150)
sc = ax.scatter(x, y, c=C, marker='o', edgecolor='k', vmin=-M, vmax=M, cmap=plt.cm.RdBu_r)
cbar = fig.colorbar(sc, ax=ax, label='$R - R_0$ (mm)')
cb_ymin = C.min()
cb_ymax = C.max()
cb_xmin, cb_xmax = cbar.ax.get_xlim()
cbar.ax.set_ylim(cb_ymin, cb_ymax)
cbar.outline.set_visible(False) # hide the surrounding spines, which are too large after set_ylim
cbar.ax.add_patch(plt.Rectangle((cb_xmin, cb_ymin), cb_xmax - cb_xmin, cb_ymax - cb_ymin,
fc='none', ec='black', clip_on=False))
plt.show()
Another approach until v3.5 is released is to make a custom colormap that does what you want (see also https://matplotlib.org/stable/tutorials/colors/colormap-manipulation.html#sphx-glr-tutorials-colors-colormap-manipulation-py)
import matplotlib.pyplot as plt
import numpy as np
import matplotlib.cm as cm
from matplotlib.colors import ListedColormap
fig, axs = plt.subplots(2, 1)
X = np.random.randn(32, 32) + 2
pc = axs[0].pcolormesh(X, vmin=-6, vmax=6, cmap='RdBu_r')
fig.colorbar(pc, ax=axs[0])
import matplotlib.pyplot as plt
import numpy as np
import matplotlib.cm as cm
from matplotlib.colors import ListedColormap
fig, axs = plt.subplots(2, 1)
X = np.random.randn(32, 32) + 2
pc = axs[0].pcolormesh(X, vmin=-6, vmax=6, cmap='RdBu_r')
fig.colorbar(pc, ax=axs[0])
def keep_center_colormap(vmin, vmax, center=0):
vmin = vmin - center
vmax = vmax - center
dv = max(-vmin, vmax) * 2
N = int(256 * dv / (vmax-vmin))
RdBu_r = cm.get_cmap('RdBu_r', N)
newcolors = RdBu_r(np.linspace(0, 1, N))
beg = int((dv / 2 + vmin)*N / dv)
end = N - int((dv / 2 - vmax)*N / dv)
newmap = ListedColormap(newcolors[beg:end])
return newmap
newmap = keep_center_colormap(-2, 6, center=0)
pc = axs[1].pcolormesh(X, vmin=-2, vmax=6, cmap=newmap)
fig.colorbar(pc, ax=axs[1])
plt.show()
Related
I've been creating uneven subplots in matplotlib based on this question. The gridspec solution (third answer) worked a little better for me as it gives a bit more flexibility for the exact sizes of the subplots.
When I add a plot of a 2D array with imshow() the affected subplot is resized to the shape of the array. Is there any way to avoid that and keep the subplot-sizes (or rather aspect-ratio) fixed?
Here's the example code and the resulting image with the subplot-sizes I'm happy with:
import numpy as np
import matplotlib.pyplot as plt
from matplotlib import gridspec
# generate data
x = np.arange(0, 10, 0.2)
y = np.sin(x)
# plot
fig = plt.figure(figsize=(12, 9))
gs = gridspec.GridSpec(20, 20)
ax1 = fig.add_subplot(gs[0:5,0:11])
ax1.plot(x, y)
ax2 = fig.add_subplot(gs[6:11,0:11])
ax2.plot(y, x)
ax3 = fig.add_subplot(gs[12:20,0:11])
ax3.plot(y, x)
ax4 = fig.add_subplot(gs[0:9,13:20])
ax4.plot(x, y)
ax5 = fig.add_subplot(gs[11:20,13:20])
ax5.plot(y, x)
plt.show()
This is what happens if I additionally plot data from a 2D array with the following lines (insert before plt.show):
2Ddata = np.arange(0, 10, 0.1).reshape(10, 10)
im = ax3.imshow(2Ddata, cmap='rainbow')
How can I restore the original size of the subplot from ax3 (lower left corner)?
Including the line ax3.set_aspect('auto') seems to have solved the issue.
I am trying to color the errorbar points based on the color from an array. But getting an error. My code is shown below:
import numpy as np
import matplotlib.pyplot as plt
from matplotlib.cm import ScalarMappable, coolwarm as cmap
from matplotlib.colors import Normalize
fig = plt.figure(1)
sp = fig.add_subplot(1, 1, 1)
sp.set_xlabel(r'$x$')
sp.set_ylabel(r'$y$')
x = np.random.rand(10)
y = np.random.rand(10)
M = np.logspace(9, 10, 10)
norm = Normalize(vmin=8, vmax=11,clip=False) # controls the min and max of the colorbar
smap = ScalarMappable(cmap=cmap, norm=norm)
for xi, yi, Mi in zip(x, y, M):
c = cmap(norm(np.log10(Mi))) # make sure to color by log of mass, not mass
sp.errorbar(
xi,
yi,
yerr=[[.1], [.1]],
xerr=[[.1], [.1]],
ecolor=c,
marker='o',
mec=c,
mfc=c
)
cb = plt.colorbar(smap)
cb.set_label(r'$\log_{10}M$')
I am getting the following error:
TypeError: You must first set_array for mappable
For matplotlib < 3.1, you need to set an array - which can be empty
sm = ScalarMappable(cmap=cmap, norm=norm)
sm.set_array([])
fig.colorbar(sm)
For matplotlib >= 3.1, this is not necessary any more.
sm = ScalarMappable(cmap=cmap, norm=norm)
fig.colorbar(sm)
I can generate an error-bar plot using the code below. The graph produced by the code shows vertical lines that represent the errors in y. I would like to have horizontal lines at the tips of these errors ("error bars") and am not sure how to do so.
import numpy as np
import matplotlib.pyplot as plt
x = np.linspace(1, 10, 10, dtype=int)
y = 2**x
yerr = np.sqrt(y)*10
fig, ax = plt.subplots()
ax.errorbar(x, y, yerr, solid_capstyle='projecting')
ax.grid(alpha=0.5, linestyle=':')
plt.show()
plt.close(fig)
The code generates the figure below. I've played with the solid_capstyle kwarg. Is there a specific kwarg that does what I am trying to do?
And as an example of what I'd like, the figure below:
In case it's relevant, I am using matplotlib 2.2.2
The argument you are looking for is capsize= in ax.errorbar(). The default is None so the length of the cap will default to the value of matplotlib.rcParams["errorbar.capsize"]. The number you give will be the length of the cap in points:
import numpy as np
import matplotlib.pyplot as plt
x = np.linspace(1, 10, 10, dtype=int)
y = 2**x
yerr = np.sqrt(y)*10
fig, ax = plt.subplots()
ax.errorbar(x, y, yerr, solid_capstyle='projecting', capsize=5)
ax.grid(alpha=0.5, linestyle=':')
plt.show()
I am making some density plots like so:
import matplotlib.pyplot as plt
import numpy as np
from sklearn.metrics import r2_score
import matplotlib
from scipy import stats
import matplotlib.gridspec as gridspec
from mpl_toolkits.axes_grid1.inset_locator import InsetPosition
from matplotlib.ticker import FormatStrFormatter
import matplotlib.cm as cm
from scipy.ndimage.filters import gaussian_filter
import random
matplotlib.rcParams.update({'font.size': 16})
matplotlib.rcParams['xtick.direction'] = 'in'
matplotlib.rcParams['ytick.direction'] = 'in'
x = random.sample(range(1, 10001), 1000)
y = random.sample(range(1, 10001), 1000)
def myplot(x, y, s, bins=1000):
heatmap, xedges, yedges = np.histogram2d(x, y, bins=bins)
heatmap = gaussian_filter(heatmap, sigma=s)
extent = [xedges[0], xedges[-1], yedges[0], yedges[-1]]
return heatmap.T, extent
cmap = cm.YlOrRd
fig, (ax, ax1, cax) = plt.subplots(ncols = 3, figsize = (15, 5),
gridspec_kw={"width_ratios":[1,1, 0.5]})
img, extent = myplot(x, y, 20)
im = ax.imshow(img, extent = extent, origin = 'lower', cmap = cmap)
ax.text(0.05, 0.92, '$R^2$ = {}'.format(np.round(r2_score(x, y), 2)), fontsize=14, color = 'k', transform = ax.transAxes)
ax.plot(ax.get_xlim(), ax.get_ylim(), ls="--", c=".3")
ax.set_xlabel("Black Sky")
ax.set_ylabel("Blue Sky")
img2, extent2 = myplot(x, y, 20)
ax1.imshow(img2, extent = extent2, origin = 'lower', cmap = cmap)
ax1.text(0.05, 0.92, '$R^2$ = {}'.format(np.round(r2_score(x, y), 2)), fontsize=14, color = 'k', transform = ax1.transAxes)
ax1.axes.get_yaxis().set_visible(False)
ax1.yaxis.set_ticks([])
ax1.plot(ax1.get_xlim(), ax1.get_ylim(), ls="--", c=".3")
ax1.set_xlabel("White Sky")
ip = InsetPosition(ax1, [1.05,0,0.05,1])
cax.set_axes_locator(ip)
fig.colorbar(im, cax=cax, ax=[ax,ax1], use_gridspec = True)
plt.subplots_adjust(wspace=0.1, hspace=0)
which gives me a plot like this:
No matter what I change wspace to the plot stays the same. I think this is because when I turn of the y-axis in ax1 I am just making the text blank instead of removing the y-axis all together. Is there a way to do this so that I can make the width spacing between the figures closer together?
As commented, wspace sets the minimal distance between plots. This distance may be larger in case of equal aspect axes. Then it will depend on the figure size, figure aspect and image aspect.
A. Use automatic aspect
You may set aspect = "auto" in your imshow plots,
ax.imshow(..., aspect = "auto")
B. Adjust the subplot parameters
You may set the left or right subplot parameter to something smaller. E.g.
plt.subplots_adjust(wspace=0.0, hspace=0, right=0.7)
C. Adjust the figure size
Using a smaller figure width, which is closer to the actual image aspect will also reduce whitespace around the figure.
E.g, making the figure only 11 inches wide and using 5% padding on the right,
plt.subplots(..., figsize = (11, 5))
plt.subplots_adjust(wspace=0.0, hspace=0, right=.95)
I'm currently trying to get an impression of continuous change in my contour plot. I have to use a logscale for the values, because some of them are some orders of magnitude bigger than the others.
import numpy as np
import matplotlib.pyplot as plt
from matplotlib import ticker
K = np.linspace(-0.99, 5, 100)
x = np.linspace(1, 5, 100)
K, x = np.meshgrid(K, x)
static_diff = 1 / (1 + K)
fig = plt.figure()
plot = plt.contourf(K, x, static_diff, locator=ticker.LogLocator(numticks=300))
plt.grid(True)
plt.xlabel('K')
plt.ylabel('x')
plt.xlim([-0.99, 5])
plt.ylim([1, 5])
fig.colorbar(plot)
plt.show()
Despite the number of ticks given to be 300 it returns a plot like:
Is there a way to get more of these lines? I also tried adding the number of parameters as the fourth parameter of the plt.contourf function.
To specify the levels of a contourf plot you may
use the levels argument and supply a list of values for the levels. E.g for 20 levels,
plot = plt.contourf(K, x, static_diff, levels=np.logspace(-2, 3, 20))
use the locator argument to which you would supply a matplotlib ticker
plt.contourf(K, x, static_diff, locator=ticker.LogLocator(subs=range(1,10)))
Note however that the LogLocator does not use a numticks argument but instead a base and a subs argument to determine the locations of the ticks. See documentation.
Complete example for the latter case, which also uses a LogNormto distribute the colors better in logspace:
import numpy as np
import matplotlib.pyplot as plt
from matplotlib import ticker
import matplotlib.colors
K = np.linspace(-0.99, 5, 100)
x = np.linspace(1, 5, 100)
K, x = np.meshgrid(K, x)
static_diff = 1 / (1 + K)
fig = plt.figure()
norm= matplotlib.colors.LogNorm(vmin=static_diff.min(), vmax=static_diff.max())
plot = plt.contourf(K, x, static_diff, locator=ticker.LogLocator(subs=range(1,10)), norm=norm)
#plot = plt.contourf(K, x, static_diff, levels=np.logspace(-2, 3, 20), norm=norm)
plt.grid(True)
plt.xlabel('K')
plt.ylabel('x')
plt.xlim([-0.99, 5])
plt.ylim([1, 5])
fig.colorbar(plot)
plt.show()