正负值的热图

Question

我试图为某些数据生成热图。我需要积极的蓝色渐变和负红色渐变。

import matplotlib.pyplot as plt
import numpy as np
arr = array([[  2.155,   3.093,  -5.377,   7.973,   6.564,   6.348,  10.279,
      1.536],
   [  0.355,  -2.848,   0.65 ,   6.877,   1.08 ,   4.001,  11.427,   0.   ],
   [ -4.23 ,  -2.855,  -2.507,   2.44 ,   2.78 ,   1.163,  27.868,   0.   ],
   [ -5.373,   1.429,  -4.247,  -3.534,  -7.044,   0.688,   3.255,   0.   ],
   [ -5.364,  -3.792,  -3.203,  -7.212,  -5.303,   0.928,  -3.028,   0.   ],
   [  1.257,  -5.088,  -4.209,  -8.229,  -3.4  ,  -7.856, -29.007,   0.   ],
   [ -2.956,  -4.787,  -0.855,  -7.551,  -5.735,  -3.882,  -4.183,   0.   ],
   [ -4.307, -19.643,  -0.422,  -5.797, -12.939,  -9.516,   4.866,   0.   ],
   [-22.003, -20.167, -17.983, -46.882, -44.439,   2.678,   3.308,   0.   ],
   [ -4.706,  -8.41 ,   0.885, -12.78 , -13.647,  10.902,   6.622,   0.   ],
   [ -6.822,  -2.396,   2.941,   3.965,  -5.588,  14.765,  11.879,   0.   ],
   [ -2.674,   0.173,   8.768,   9.349,   4.73 ,  10.9  ,  12.983,   0.   ]])
x_header = [2007, 2008, 2009, 2010, 2011, 2012, 2013, 2014]
y_header = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12]
fig, ax = plt.subplots()
heatmap = plt.pcolor(arr, cmap = 'RdBu')
ax.set_xticks(np.arange(arr.shape[1]), minor=False)
ax.set_yticks(np.arange(arr.shape[0])+0.5, minor=False)
ax.set_xticklabels(x_header)
ax.set_yticklabels(y_header)
ax.xaxis.tick_top()
ax.invert_yaxis()
plt.show()

但是，我没有得到我选择的渐变。我得到下图。如果您注意到大多数负值都具有蓝色渐变，并且只有少数值具有红色渐变。 是否有办法强制将特定颜色渐变用于正值，将另一种颜色用于负值？

感谢。