Question

我第一次使用matplotlib来可视化以下矩阵：

[[ 150.       0.       0.       0.       0.       0.       0.       0.       0.       0.       0.       0.       0.       0.       0.       0.       0.       0.       0.       0.       0.       0.       0.   ]
 [ 150.      69.388   35.36    18.211    7.851    0.       0.       0.       0.       0.       0.       0.       0.       0.       0.       0.       0.       0.      -0.03    -0.047   -0.044   -0.027    0.   ]
 [ 150.      92.192   53.842   29.633   13.192    0.       0.       0.       0.       0.       0.       0.       0.       0.       0.       0.       0.       0.      -0.075   -0.112   -0.104   -0.062    0.   ]
 [ 150.      95.538   58.184   33.287   15.285    0.       0.       0.       0.       0.       0.       0.       0.       0.       0.       0.       0.       0.      -0.156   -0.221   -0.199   -0.116    0.   ]
 [ 150.      81.776   50.068   30.045   14.659    0.       0.       0.       0.       0.       0.       0.       0.       0.       0.       0.       0.       0.      -0.331   -0.416   -0.357   -0.203    0.   ]
 [   0.      31.498   30.267   22.168   13.308    5.653    2.766    1.5      0.86     0.492    0.254    0.082   -0.058   -0.182   -0.298   -0.41    -0.515   -0.617   -0.751   -0.754   -0.609   -0.34     0.   ]
 [   0.      13.948   17.335   15.05    10.751    6.537    3.913    2.374    1.447    0.854    0.443    0.13    -0.132   -0.372   -0.602   -0.826   -1.034   -1.204   -1.301   -1.241   -0.985   -0.548    0.   ]
 [   0.       6.958   10.074    9.949    8.107    5.832    3.973    2.635    1.7      1.034    0.534    0.129   -0.229   -0.57    -0.913   -1.258   -1.59    -1.864   -2.007   -1.924   -1.543   -0.866    0.   ]
 [   0.       3.812    6.053    6.564    5.896    4.712    3.511    2.493    1.685    1.048    0.529    0.079   -0.342   -0.768   -1.22    -1.705   -2.203   -2.654   -2.941   -2.904   -2.395   -1.375    0.   ]
 [   0.       2.237    3.761    4.358    4.202    3.609    2.867    2.142    1.498    0.943    0.455    0.002   -0.452   -0.94    -1.495   -2.139   -2.864   -3.607   -4.2     -4.357   -3.76    -2.237    0.   ]
 [   0.       1.375    2.396    2.906    2.943    2.656    2.206    1.708    1.223    0.771    0.345   -0.076   -0.526   -1.045   -1.682   -2.491   -3.509   -4.71    -5.895   -6.563   -6.052   -3.812    0.   ]
 [   0.       0.867    1.543    1.925    2.009    1.865    1.592    1.261    0.915    0.573    0.232   -0.126   -0.531   -1.031   -1.698   -2.633   -3.971   -5.831   -8.105   -9.948  -10.073   -6.958    0.   ]
 [   0.       0.548    0.986    1.242    1.302    1.205    1.035    0.828    0.604    0.373    0.134   -0.128   -0.441   -0.852   -1.445   -2.372   -3.911   -6.536  -10.749  -15.049  -17.334  -13.947    0.   ]
 [   0.       0.34     0.609    0.755    0.751    0.618    0.516    0.411    0.299    0.183    0.059   -0.081   -0.253   -0.491   -0.859   -1.499   -2.766   -5.652  -13.307  -22.167  -30.267  -31.497    0.   ]
 [   0.       0.203    0.357    0.416    0.331    0.       0.       0.       0.       0.       0.       0.       0.       0.       0.       0.       0.       0.     -14.659  -30.045  -50.068  -81.776 -150.   ]
 [   0.       0.116    0.2      0.221    0.157    0.       0.       0.       0.       0.       0.       0.       0.       0.       0.       0.       0.       0.     -15.284  -33.286  -58.184  -95.538 -150.   ]
 [   0.       0.062    0.104    0.112    0.075    0.       0.       0.       0.       0.       0.       0.       0.       0.       0.       0.       0.       0.     -13.192  -29.633  -53.842  -92.192 -150.   ]
 [   0.       0.027    0.044    0.047    0.03     0.       0.       0.       0.       0.       0.       0.       0.       0.       0.       0.       0.       0.      -7.851  -18.211  -35.36   -69.388 -150.   ]
 [   0.       0.       0.       0.       0.       0.       0.       0.       0.       0.       0.       0.       0.       0.       0.       0.       0.       0.       0.       0.       0.       0.    -150.   ]]

我遇到的问题是，渲染的默认精度不会显示中心值之间的差异，这对我的应用程序而言最重要：

由于我的值介于150到-150之间，因此matplotlib为所有中间值（[-5,5]）赋予相同的颜色

我想控制缩放比例以生成某种热图，以显示整个图形的进度。

相关的python代码：

    plt.matshow(mat, aspect='auto', cmap=plt.get_cmap('magma'))
    plt.show()

达到此结果的最简单方法是什么？

Answer 1

您可以对此进行调整：

import matplotlib.pyplot as plt
import matplotlib
mat = [[ 150. ,  0.    ,   0.     ,  0.   , 0.   ,  0.   ,    0.     ,  0.     ,  0.    ,   0.    ,   0.    ,  0.  ,   0.  ,     0.    ,   0.   ,    0.    ,   0.     ,  0.   ,  0.     ,  0.     ,  0.      , 0.     ,  0.   ],
       [ 150. , 69.388 ,  35.36   , 18.211, 7.851,  0.   ,    0.     ,  0.     ,  0.    ,   0.    ,   0.    ,  0.  ,   0.  ,     0.    ,   0.    ,   0.   ,    0.     ,  0.   ,  -0.03  ,  -0.047 ,  -0.044  , -0.027 ,   0.   ],
       [ 150. , 92.192 ,  53.842  , 29.633,13.192,  0.   ,    0.     ,  0.     ,  0.    ,   0.    ,   0.    ,  0.  ,   0.  ,     0.    ,   0.    ,   0.   ,    0.     ,  0.   ,  -0.075 ,  -0.112 ,  -0.104  , -0.062 ,   0.   ],
       [ 150. , 95.538 ,  58.184  , 33.287,15.285,  0.   ,    0.     ,  0.     ,  0.    ,   0.    ,   0.    ,  0.  ,   0.  ,     0.    ,   0.    ,   0.   ,    0.     ,  0.   ,  -0.156 ,  -0.221 ,  -0.199  , -0.116 ,   0.   ],
       [ 150. , 81.776 ,  50.068  , 30.045,14.659,  0.   ,    0.     ,  0.     ,  0.    ,   0.    ,   0.    ,  0.  ,   0.  ,     0.    ,   0.    ,   0.   ,    0.     ,  0.   ,  -0.331 ,  -0.416 ,  -0.357  , -0.203 ,   0.   ],
       [   0. , 31.498 ,  30.267  , 22.168,13.308,  5.653,    2.766  ,  1.5    ,  0.86  ,   0.492  ,  0.254  , 0.082, -0.058,   -0.182  , -0.298 ,  -0.41 ,   -0.515  , -0.617,  -0.751 ,  -0.754 ,  -0.609  , -0.34  ,   0.   ],
       [   0. , 13.948 ,  17.335  , 15.05 ,10.751,  6.537,    3.913  ,  2.374  ,  1.447 ,   0.854  ,  0.443  , 0.13 , -0.132,   -0.372  , -0.602  , -0.826 ,  -1.034  , -1.204,  -1.301 ,  -1.241 ,  -0.985  , -0.548 ,   0.   ],
       [   0. ,  6.958 ,  10.074  ,  9.949, 8.107,  5.832,    3.973  ,  2.635  ,  1.7   ,   1.034  ,  0.534  , 0.129, -0.229,   -0.57   , -0.913  , -1.258 ,  -1.59   , -1.864,  -2.007 ,  -1.924 ,  -1.543  , -0.866 ,   0.   ],
       [   0. ,  3.812 ,   6.053  ,  6.564, 5.896,  4.712,    3.511  ,  2.493  ,  1.685 ,   1.048  ,  0.529  , 0.079, -0.342,   -0.768  , -1.22   , -1.705 ,  -2.203  , -2.654,  -2.941 ,  -2.904 ,  -2.395  , -1.375 ,   0.   ],
       [   0. ,  2.237 ,   3.761  ,  4.358, 4.202,  3.609,    2.867  ,  2.142  ,  1.498 ,   0.943  ,  0.455  , 0.002, -0.452,   -0.94   , -1.495  , -2.139 ,  -2.864  , -3.607,  -4.2   ,  -4.357 ,  -3.76   , -2.237 ,   0.   ],
       [   0. ,  1.375 ,   2.396  ,  2.906, 2.943,  2.656,    2.206  ,  1.708  ,  1.223 ,   0.771  ,  0.345  ,-0.076, -0.526,   -1.045  , -1.682  , -2.491 ,  -3.509  , -4.71 ,  -5.895 ,  -6.563 ,  -6.052  , -3.812 ,   0.   ],
       [   0. ,  0.867 ,   1.543  ,  1.925, 2.009,  1.865,    1.592  ,  1.261  ,  0.915 ,   0.573  ,  0.232  ,-0.126, -0.531,   -1.031  , -1.698  , -2.633 ,  -3.971  , -5.831,  -8.105 ,  -9.948 , -10.073  , -6.958 ,   0.   ],
       [   0. ,  0.548 ,   0.986  ,  1.242, 1.302,  1.205,    1.035  ,  0.828  ,  0.604 ,   0.373  ,  0.134  ,-0.128, -0.441,   -0.852  , -1.445  , -2.372 ,  -3.911  , -6.536, -10.749 , -15.049 , -17.334  ,-13.947 ,   0.   ],
       [   0. ,  0.34  ,   0.609  ,  0.755, 0.751,  0.618,    0.516  ,  0.411  ,  0.299 ,   0.183  ,  0.059  ,-0.081, -0.253,   -0.491  , -0.859  , -1.499 ,  -2.766  , -5.652, -13.307 , -22.167 , -30.267  ,-31.497 ,   0.   ],
       [   0. ,  0.203 ,   0.357  ,  0.416, 0.331,  0.   ,    0.     ,  0.     ,  0.    ,   0.    ,   0.    ,  0.  ,   0.  ,     0.    ,   0.     ,  0.    ,   0.     ,  0.   , -14.659 , -30.045 , -50.068  ,-81.776 ,-150.   ],
       [   0. ,  0.116 ,   0.2    ,  0.221, 0.157,  0.   ,    0.     ,  0.     ,  0.    ,   0.    ,   0.    ,  0.  ,   0.  ,     0.    ,   0.    ,   0.   ,    0.     ,  0.   , -15.284 , -33.286 , -58.184  ,-95.538 ,-150.   ],
       [   0. ,  0.062 ,   0.104  ,  0.112, 0.075,  0.   ,    0.     ,  0.     ,  0.    ,   0.    ,   0.    ,  0.  ,   0.  ,     0.    ,   0.    ,   0.   ,    0.     ,  0.   , -13.192 , -29.633 , -53.842  ,-92.192 ,-150.   ],
       [   0. ,  0.027 ,   0.044  ,  0.047, 0.03 ,  0.   ,    0.     ,  0.     ,  0.    ,   0.    ,   0.    ,  0.  ,   0.  ,     0.    ,   0.    ,   0.   ,    0.     ,  0.   ,  -7.851 , -18.211 , -35.36   ,-69.388 ,-150.   ],
       [   0. ,  0.    ,   0.     ,  0.   , 0.   ,  0.   ,    0.     ,  0.     ,  0.    ,   0.    ,   0.    ,  0.  ,   0.  ,     0.    ,   0.    ,   0.   ,    0.     ,  0.   ,   0.    ,   0.    ,   0.     ,  0.    ,-150.   ]]
plt.matshow(mat, aspect='auto', norm=matplotlib.colors.SymLogNorm(0.1))
plt.show()

将产生以下图像：

说明：
在matplotlib中，可以使用归一化的选项，当您处理相对较大的数字时，而较小的数字则更为重要，因此您需要记录（即对数）值以查看差异。因为您也有负数，只是记录它们不会减少它（log（）显然不会处理负数）。因此，SymLogNorm也以相同形式处理负数。 0.1是一个重要的参数，因为接近零的值趋于无穷大，因此需要在零附近具有线性范围。参数linthresh（{我在这里通过查找矩阵值在这里选择的0.1）允许用户指定此范围的大小（-linthresh，linthresh）。

有关更多详细信息：SymLogNorm documentation

matplotlib矩阵可视化：控制浮点值的精度

1 个答案: