2012-03-31 130 views
1

3D数据 - 这是一个更新的帖子,老撤职 -绘制的matplotlib

假设我有一个像下面我想在3D图形/表面绘制与matplotlib数据。我该怎么做?

enter image description here

继@JouniK.Seppänen的答案,我想我需要meshgrid(),但我觉得我的轴数据不是非常正确的。如果我有数据(JSON)像:

{ 
    "data": { 
     "Random": [ 
      [1834, 3353, 4602, 5471, 6522, 7873], 
      [2637, 8575, 8357, 10329, 9742, 10359], 
      [3648, 10602, 10667, 10751, 10666, 10552], 
      [4570, 10220, 9202, 10460, 10329, 10928], 
      [5879, 10788, 10832, 10923, 11075, 10989], 
      [6783, 11104, 10235, 10499, 11024, 10731], 
      [7074, 11097, 10222, 10613, 10508, 10767], 
      [7300, 11002, 10727, 11073, 10328, 10864]], 
     "LRU": [ 
      [123, 155, 201, 223, 296, 321], 
      [143, 590, 1046, 1566, 1924, 2434], 
      [163, 1167, 1774, 2578, 3363, 3980], 
      [182, 1172, 2259, 3038, 4200, 4907], 
      [219, 1718, 3044, 3658, 5236, 5680], 
      [709, 2263, 3588, 4551, 5262, 6197], 
      [2065, 3865, 4430, 5024, 5986, 6617], 
      [3048, 4249, 5029, 5790, 6157, 6826]], 
     "FIFO": [ 
      [180, 269, 350, 424, 580, 601], 
      [230, 906, 1446, 2009, 2408, 2902], 
      [316, 1590, 2261, 3042, 3848, 4457], 
      [473, 1664, 2781, 3542, 4764, 5398], 
      [665, 2290, 3609, 4194, 5781, 6207], 
      [1158, 2826, 4115, 5064, 5751, 6613], 
      [2457, 4375, 4974, 5471, 6464, 7077], 
      [3512, 4724, 5485, 6272, 6684, 7312]] 
    }, 
    "workingSetAxis": [2, 22, 42, 62, 82, 102, 122, 142], 
    "stabilityAxis": [0.0, 0.2, 0.4, 0.6, 0.8, 1.0] 
} 

workingSetAxis在哪里被认为是X轴,以及stabilityAxis y轴。我不喜欢的东西

plot(jsonObj["data"]["FIFO"], jsonObj["workingSetAxis"], jsonObj["stabilityAxis"]) 

def plot(data, workingSetAxis, stabilityAxis): 
    # make axis data 
    X, Y = numpy.meshgrid(workingSetAxis, stabilityAxis) 
    Z = data 

    fig = plt.figure() 
    ax = fig.add_subplot(111, projection="3d") 
    ax.plot_wireframe(X, Y, Z) 
    plt.show() 

和得到的东西,如:

enter image description here

从我从Excel中得到了情节非常不同(FIFO数据)

enter image description here

我相信我的坐标轴数据或者是错的,但是哪一个?

回答

5

尝试绘制你的数据的转置,例如,使用

def plot(data, workingSetAxis, stabilityAxis): 
    # make axis data 
    X, Y = numpy.meshgrid(workingSetAxis, stabilityAxis) 
    Z = numpy.transpose(data) #<--- This is the only line I have changed. 

    fig = plt.figure() 
    ax = fig.add_subplot(111, projection="3d") 
    ax.plot_wireframe(X, Y, Z) 
    plt.show() 

和您的示例数据给我的以下情节: plot demonstrating expected output

4

参见例如this example。把你的X和Y值成二维数组与meshgrid,然后用你的数据值作为Z.

+0

我也更新了我的问题,我只是想确认我得到的轴和订购权?将图像悬停在图像上有时候会产生像x值一样的奇怪坐标,也许我的猜测并不在正确的位置 – 2012-03-31 12:49:46