python产生随机样本数据

一、产生X样本

　x_train = np.random.random((5, 3)) 随机产生一个5行3列的样本矩阵，也就是5个维度为3的训练样本。

array([[ 0.56644011,  0.75185718,  0.98654195],
       [ 0.46676905,  0.2452094 ,  0.28035157],
       [ 0.69687126,  0.85162556,  0.23118269],
       [ 0.69127369,  0.32235362,  0.90172209],
       [ 0.64421882,  0.65666665,  0.37091807]])

二、产生Y样本

y_train = np.random.randint(10, size=(20, 1)) 产生一个20行1列的Y样本，值分布为10个，也就是0~9。也就是20个多类别样本标签。

　array([[8],

       [8],
       [0],
       [4],
       [9],
       [9],
       [7],
       [3],
       [0],
       [9],
       [0],
       [2],
       [1],
       [0],
       [3],
       [4],
       [6],
       [8],
       [9],
       [7]])

三、产生2D卷积X样本
x_train = np.random.random((4,2,5, 3))产生4个卷积样本，每个样本两层，没层矩阵是5X3结构。

array([[[[ 0.81108075,  0.75130404,  0.32276459],
         [ 0.84803225,  0.95347097,  0.98392204],
         [ 0.82862565,  0.60562112,  0.12725719],
         [ 0.66517274,  0.80061288,  0.56373024],
         [ 0.33360791,  0.15615631,  0.01854572]],

        [[ 0.95840439,  0.62069117,  0.98154442],
         [ 0.22812983,  0.83663549,  0.79360161],
         [ 0.40764592,  0.1903219 ,  0.75269041],
         [ 0.89337384,  0.48268712,  0.98336301],
         [ 0.00515764,  0.41898271,  0.17870325]]],

       [[[ 0.16303286,  0.30437622,  0.80772764],
         [ 0.99838344,  0.78417382,  0.52251551],
         [ 0.81561737,  0.20268081,  0.15342787],
         [ 0.77666367,  0.26014027,  0.01359609],
         [ 0.76491115,  0.23499911,  0.75797289]],

        [[ 0.0221104 ,  0.92696779,  0.16339887],
         [ 0.93589062,  0.64230156,  0.54570248],
         [ 0.01895301,  0.23444549,  0.03577822],
         [ 0.06956943,  0.05085453,  0.58532944],
         [ 0.01029333,  0.99890575,  0.22400419]]],

       [[[ 0.33587317,  0.38829797,  0.76169893],
         [ 0.8067067 ,  0.29012318,  0.01406736],
         [ 0.99158238,  0.60665312,  0.52777604],
         [ 0.06333543,  0.9294594 ,  0.0571626 ],
         [ 0.02463482,  0.9234842 ,  0.68864325]],

        [[ 0.23725655,  0.8793853 ,  0.49002114],
         [ 0.86578146,  0.93386534,  0.48375739],
         [ 0.5304713 ,  0.44797753,  0.79250569],
         [ 0.92835088,  0.17855765,  0.27783737],
         [ 0.17801198,  0.2095321 ,  0.64932004]]],

       [[[ 0.35564935,  0.98168517,  0.75135149],
         [ 0.79403744,  0.06994751,  0.95484361],
         [ 0.14493514,  0.11813182,  0.61482502],
         [ 0.5031048 ,  0.91276372,  0.2315978 ],
         [ 0.57193754,  0.20402079,  0.75060145]],

        [[ 0.0099759 ,  0.37148569,  0.89472595],
         [ 0.91443219,  0.17405477,  0.78021433],
         [ 0.84789989,  0.34975548,  0.85220165],
         [ 0.85179668,  0.04264071,  0.36531178],
         [ 0.72911524,  0.85494955,  0.60118721]]]])

四、产生2D卷积Y样本

y_train = np.random.randint(10, size=(20, 1)) y样本不变，同上