2D Plotting¶
There are several ways to visualize 2-Dimensional gridded data using matplotlib.pyplot:
- How to generate 2-D gridded data from a known math function (meshgrid)
- Contour
- Contourf (filled contour) and color bar
- Pcolormesh (filled grid)
- Surface (3-D effects)
np.meshgrid¶
In the following example, we will plot the function $z=exp(-sin(2x)^{2}-cos(2y)^2)-0.5$.
Given each pair of x and y values, z (height) can be computed. Therefore the function z can be plotted on a "map" with axes along x and y using contour or surface.
First we need to generate the data of z using many combinations x and y values covering the ranges of x and y values of interest.
The most efficient way to do this is to specific 1-D arrays of x(0:n) and y(0:m) (for example, using np.linspace), then use np.meshgrid to "expand" the 1-D arrays into 2-D arrays of size (0:m,0:n):
import numpy as np
# let's first use a smaller array to demonstrate what meshgrid does --
x=np.linspace(-1,1,3)
y=np.linspace(-1,1,5)
xx,yy=np.meshgrid(x,y)
print(x)
print(xx.shape)
print(xx)
print(y)
print(yy.shape)
print(yy)
# Now we use finer grid (larger arrays) to continue our example:
x=np.linspace(-1,1,25)
y=np.linspace(-1,1,25)
xx,yy=np.meshgrid(x,y)
print(xx.shape)
# compute function z using the grids constructed by xx and yy
z=np.exp(-np.sin(2*xx)**2-np.cos(2*yy)**2)-0.5
print(z.shape)
plt.contour 畫等值線¶
CS=plt.contour(x,y,z,cmap=...,levels=...,colors=...)
- x: array (can be 1D or 2D) of x to evaluate z
- y: array (can be 1D or 2D) of y to evaluate z
- z: 2-D array of data to plot contour
- cmap: color map (remember to import matplotlib.cm)
- levels: contour levels
- colors: line color
reference: http://matplotlib.org/api/_as_gen/matplotlib.pyplot.contour.html#matplotlib.pyplot.contour
import matplotlib.pyplot as plt
CS = plt.contour(x, y, z)
plt.title(r'z=exp(-sin(2x)$^2$-cos(2y)$^2$)-0.5')
plt.xlabel('x')
plt.ylabel('y')
plt.show()
plt.contour: change color map 換等值線色階¶
CS=plt.contour(x,y,z,cmap=...,levels=...,colors=...)
- cmap: color map (remember to import matplotlib.cm)
Matplotlib provides many pre-defined color map options. Here are some examples:
Use cmap=cm.colormapname in the contour (and contourf) command to change color map. (Default is cm.jet)
More color map choices here: https://matplotlib.org/examples/color/colormaps_reference.html
import matplotlib.cm as cm
CS = plt.contour(x, y, z,cmap=cm.ocean)
plt.show()
plot.label:add contour label 加等值線標籤¶
plt.clabel(CS, inline=1, fontsize=10)
- CS: plotting object
- inline: location of label
- 1=lines do not cross over label text (default)
- 0=lines cross over label text
- fontsize: size of label text
CS = plt.contour(x, y, z,cmap=cm.ocean)
plt.clabel(CS, inline=1, fontsize=10)
plt.show()
plt.contour: change contour levels, single color lines 設定等值線數值、畫單色等值線¶
CS=plt.contour(x,y,z,cmap=...,levels=...,colors=...)
- levels: 1D-array to specify contour levels
- colors: line color (applied to all contour lines. Solid line for positive values, dashed for negative)
- If None, the colormap specified by cmap will be used.
- If a string, like ‘r’ or ‘red’, all levels will be plotted in this color.
- If a tuple of matplotlib color args (string, float, rgb, etc), different levels will be plotted in different colors in the order specified.
CS = plt.contour(x, y, z, levels=np.linspace(-0.5,0.5,11),colors='r')
plt.clabel(CS,fontsize=7)
plt.show()
plt.contourf: filled contour 色塊等值線（等值線內填入相同顏色）¶
Similar to contour but fill the space between contours with solid colors.
CS=plt.contourf(x,y,z,cmap=...,levels=...)
- x: array (can be 1D or 2D) of x to evaluate z
- y: array (can be 1D or 2D) of y to evaluate z
- z: 2-D array of data to plot contour
- cmap: color map (remember to import matplotlib.cm)
- levels: contour levels
CS = plt.contourf(x, y, z, levels= np.linspace(-0.5,0.5,11),cmap=cm.rainbow)
plt.show()
plt.colorbar: add color bar to the figure¶
plt.colorbar(CS, orientation=...)
- CS: plotting object
- orientation: orientation of color bar on the figure 'horizontal' 'vertical'
CS = plt.contourf(x, y, z, levels= np.linspace(-0.5,0.5,11),cmap=cm.jet)
plt.colorbar(CS,orientation='vertical')
plt.show()
You can also overlay filled contour with contour lines:
CS = plt.contourf(x, y, z, levels=np.linspace(-0.5,0.5,11),cmap=cm.jet)
CS2 = plt.contour(x, y, z, levels=np.linspace(-0.5,0.5,11),colors='k')
plt.show()
plt.pcolormesh: fill grid with solid color (mosaic)¶
plt.pcolormesh(xx, yy, z, cmap=...,shading=...,edgecolors=...)
- xx: 2D array of x mesh
- yy: 2D array of y mesh
- z: 2D array of data to be plotted
- shading: blending 渲染
- 'flat': no blending (default)
- 'gouraud: blending
- edgecolors: grid edge color (default is 'None')
CS = plt.pcolormesh(xx, yy, z, cmap=cm.jet,shading='flat')
plt.show()
CS = plt.pcolormesh(xx, yy, z, cmap=cm.jet,shading='flat',edgecolors='grey')
plt.show()
CS = plt.pcolormesh(xx, yy, z, cmap=cm.jet,shading='gouraud')
plt.show()
ax.plot_surface: plot the function on a 3D surface¶
Several steps are involved to show the function z(x,y) with 3D effects
- from mpl_toolkits.mplot3d import Axes3D
- generate a figure window object by plt.figure()
- set the projection to '3d by defining plotting object ax = fig.gca()
- plot the surface using ax.plot_surface (xx and yy has to be 2D mesh)
from mpl_toolkits.mplot3d import Axes3D
fig = plt.figure()
ax = fig.gca(projection='3d')
surf = ax.plot_surface(xx, yy, z, cmap=cm.jet)
ax.set_xlabel('x')
ax.set_ylabel('y')
ax.set_zlabel('z')
ax.set_title(r'z=exp(-sin(2x)$^2$-cos(2y)$^2$)-0.5')
plt.show()
Other 2D plot from Matplotlib library¶
- Vectors on 2D coordinate: plt.quiver(x,y,u,v,color=...)
- Streamlines: plt.streamplot(x,y,u,v,density=...,linewidth=...,color=...,cmap=...)
- 2D histogram: plt.hist2d(x,y,bins=...,range=...)
Sample plots in Matplotlib¶
Here you’ll find a host of example plots with the code that generated them: http://matplotlib.org/tutorials/introductory/sample_plots.html#sphx-glr-tutorials-introductory-sample-plots-py