极坐标图分块形状不匹配

Question

我想绘制一张从XYZ数据获得的方位角和天顶角的极坐标图。但是我传递给contourf函数的数组是畸形的，我不知道如何纠正这个问题？在

import numpy as np
import matplotlib.pyplot as plt

# Populated arrays with angles.
azimuths = np.random.random(200)*360
zeniths = np.random.random(200)*180

a_bins = np.linspace(0,360,13)
z_bins = np.linspace(0,180,7)

grid, ae, ze = np.histogram2d(azimuths, zeniths, bins=[a_bins,z_bins])

a_bins = np.radians(a_bins)
r, theta = np.meshgrid(z_bins, a_bins)

# Plot
fig, ax = plt.subplots(subplot_kw=dict(projection='polar'))
cax = ax.contourf(theta, r, grid, 30)
cb = fig.colorbar(cax)

plt.show()

代码运行，但抛出以下警告：x的形状与z的形状不匹配：找到（13，7）而不是（12，6）。

现在我想我理解了这个错误。方位角为13个（0-360），7个为天顶（0-180）。histogram2d函数返回的矩阵的形状为（12,6），因为这是边之间的槽数。我只是不知道怎么修理箱子。在

Answer 1

一种方法是将grid数组展开为与theta和{}相同的形状。这是必要的，这样极坐标图就可以一直延伸（并且在theta=0处匹配）。在

import numpy as np
import matplotlib.pyplot as plt

# Populated arrays with angles.
azimuths = np.random.random(200)*360
zeniths = np.random.random(200)*180

a_bins = np.linspace(0,360,13)
z_bins = np.linspace(0,180,7)

grid, ae, ze = np.histogram2d(azimuths, zeniths, bins=[a_bins,z_bins])

a_bins = np.radians(a_bins)
r, theta = np.meshgrid(z_bins, a_bins)

# Extend grid by one column row, using the 0th column and row
g = np.zeros(r.shape)
g[:-1,:-1] = grid 
g[-1] = g[0]      # copy the top row to the bottom
g[:,-1] = g[:,0]  # copy the left column to the right
print g.shape,r.shape,theta.shape
### (13, 7) (13, 7) (13, 7)

# Plot
fig, ax = plt.subplots(subplot_kw=dict(projection='polar'))
cax = ax.contourf(theta, r, g, 30)
cb = fig.colorbar(cax)

plt.show()