聚合操作

import numpy as npL=np.random.random(100) L array([6.40912934e-01, 6.68707312e-01, 3.34817109e-01, 6.30394721e-01,1.33956105e-01, 5.77271515e-01, 6.48859014e-01, 5.07756100e-01,9.32359865e-03, 7.92954838e-01, 6.27917762e-01, 8.50336965e-01,2.40358614e-01, 3.41050480e-01, 6.65260877e-01, 3.18950655e-02,3.00673031e-01, 9.27826583e-01, 2.05861208e-01, 7.33610830e-01,3.61208111e-02, 9.32883308e-01, 2.38428809e-01, 5.89047404e-01,3.47473823e-01, 3.67105653e-01, 1.30058476e-03, 1.59895273e-01,4.58360452e-01, 1.95779715e-04, 8.93418797e-01, 1.85457181e-02,3.71156107e-01, 5.20282142e-01, 1.36484920e-01, 4.38857565e-01,1.03275714e-02, 4.03485916e-01, 6.75369123e-01, 4.89373844e-01,7.04879473e-01, 9.49022821e-01, 8.34890126e-01, 3.76801980e-01,4.28500622e-01, 2.62809258e-01, 5.23769311e-01, 9.28819885e-01,7.63581830e-01, 3.25866298e-02, 5.51025883e-01, 5.35687967e-01,3.39079879e-01, 2.59346840e-01, 7.68451174e-01, 3.96283705e-01,2.30806644e-01, 9.95081769e-01, 4.43257664e-01, 2.26709949e-01,5.91877109e-01, 2.20734191e-01, 1.78324415e-01, 7.09976504e-01,1.50176596e-01, 8.61794981e-01, 7.05957604e-01, 2.35726863e-01,1.91863521e-01, 8.03723376e-01, 2.94690054e-01, 5.84773442e-01,7.28967637e-01, 9.10776044e-01, 1.96923346e-01, 4.83698087e-01,8.06441846e-01, 9.54663309e-01, 3.57682219e-01, 6.97190492e-01,6.82302385e-01, 3.20093123e-01, 3.42721050e-01, 6.85886395e-01,8.67474233e-01, 4.06543995e-01, 7.42878119e-01, 6.25685640e-01,4.27487322e-01, 4.60038125e-01, 9.28615336e-01, 3.93480542e-01,4.54296195e-01, 1.17550075e-01, 2.22505188e-01, 9.87210073e-01,2.53840068e-01, 3.08736344e-01, 4.47973833e-01, 8.46861898e-01]) sum(L) 48.69739325141593 np.sum(L) 48.69739325141593 big_array=np.random.rand(1000000) %timeit sum(big_array) %timeit np.sum(big_array) 302 ms ± 32.4 ms per loop (mean ± std. dev. of 7 runs, 1 loop each) 1.12 ms ± 22.4 μs per loop (mean ± std. dev. of 7 runs, 1000 loops each) np.min(big_array) 6.792467711624894e-07 np.max(big_array) 0.9999979467731644 big_array.max() 0.9999979467731644 big_array.sum() 500255.9883044571 X=np.arange(16).reshape(4,-1) X array([[ 0, 1, 2, 3],[ 4, 5, 6, 7],[ 8, 9, 10, 11],[12, 13, 14, 15]]) np.sum(X) 120 np.sum(X, axis=0) #每一列的和 array([24, 28, 32, 36]) np.sum(X, axis=1) #每一行的和 array([ 6, 22, 38, 54]) np.prod(X) 0 np.prod(X+1) %python不用擔心整形溢出 2004189184 np.mean(X) 7.5 np.median(X) 7.5 v=np.array([1,1,2,2,10]) np.mean(v) 3.2 np.median(v) 2.0 np.percentile(big_array,q=50) 0.5002877203077256 np.median(big_array) 0.5002877203077256 np.percentile(big_array,q=100) 0.9999979467731644 np.max(big_array) 0.9999979467731644 for percent in [0,25,50,75,100]: ##統計中很多繪圖用這幾個數據print(np.percentile(big_array,q=percent)) 6.792467711624894e-07 0.25028904658519036 0.5002877203077256 0.7502613491753023 0.9999979467731644 np.var(big_array) #方差 0.08327892606438625 np.std(big_array) #標準差 0.2885808830542769 x=np.random.normal(0,1,size=1000000) np.mean(x) 0.0007654919189879631 np.std(x) 0.9986988097009861

總結

以上是生活随笔為你收集整理的[云炬python3玩转机器学习笔记] 3-8Numpy中的聚合运算的全部內容，希望文章能夠幫你解決所遇到的問題。

如果覺得生活随笔網站內容還不錯，歡迎將生活随笔推薦給好友。