Python基礎入門（Numpy）
阿新 • • 發佈：2018-12-18
#numpy 矩陣運算庫 import numpy as np #常用的建立函式 #array 從列表或者元祖建立陣列 ar1 = np.array([1,2,3,4,5,6]) print(ar1) #建立陣列0-10之間，步長為1 ar2 = np.arange(0,10,1) print(ar2) #將0-10分成5個元素的陣列 ar3 = np.linspace(0,10,5) print(ar3) #建立3行3列的陣列 ar4 = np.indices((3,3)) print(ar4) #建立3行3列的隨機陣列 ar5 = np.random.rand(3,3) print(ar5) #建立3行3列的全為1的陣列 ar6 = np.ones((3,3),float) print(ar6) #建立3行3列全為0的陣列 np.zeros((3,3)) ar8 = np.zeros((3,3)) print(ar8) #對角線為1，其他為0 ar7 = np.eye(3,3) print(ar7) [1 2 3 4 5 6] [0 1 2 3 4 5 6 7 8 9] [ 0. 2.5 5. 7.5 10. ] [[[0 0 0] [1 1 1] [2 2 2]] [[0 1 2] [0 1 2] [0 1 2]]] [[0.13282886 0.50496395 0.7787284 ] [0.80039057 0.40068522 0.1187407 ] [0.51833548 0.11563461 0.72263001]] [[1. 1. 1.] [1. 1. 1.] [1. 1. 1.]] [[0. 0. 0.] [0. 0. 0.] [0. 0. 0.]] [[1. 0. 0.] [0. 1. 0.] [0. 0. 1.]] In [32]: #陣列的基礎屬性 #陣列軸的個數，秩 print(ar6.ndim) #每個維度的大小 print(ar6.shape) #元素的總數 print(ar6.size) #元素資料的型別 print(ar6.dtype) #每個元素的位元組數 print(ar6.itemsize) #陣列元素緩衝區的地址 print(ar6.data) #元素迭代器 print(ar6.flat) 2 (3, 3) 9 float64 8 <memory at 0x00000000053A47E0> <numpy.flatiter object at 0x00000000065992D0> In [16]: #陣列的形態操作 ar7 = np.ones((3,3)) print(ar7) #np.resize 修改陣列 ar7.resize(2,2) print(ar7) #對陣列進行將維度 ar8= np.ones((3,4)) print(ar8.flatten()) ar8= np.ones((3,4)) print(ar8.ravel()) #reshape ar9= np.ones((3,4)) print(ar9) print(ar9.reshape(2,6)) #astype 型別轉換 ar10 = np.arange(10,dtype=float).reshape(2,5) print(ar10) print(ar10.astype(np.int32)) [[1. 1. 1.] [1. 1. 1.] [1. 1. 1.]] [[1. 1.] [1. 1.]] [1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1.] [1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1.] [[1. 1. 1. 1.] [1. 1. 1. 1.] [1. 1. 1. 1.]] [[1. 1. 1. 1. 1. 1.] [1. 1. 1. 1. 1. 1.]] [[0. 1. 2. 3. 4.] [5. 6. 7. 8. 9.]] [[0 1 2 3 4] [5 6 7 8 9]] In [34]: #橫向相加，垂直相加 a = np.arange(5) b = np.arange(5,10) print(a,b) #按行拼接 ar = np.vstack((a,b)) print(ar) #按列拼接 ar = np.hstack((a,b)) print(ar) ar = np.stack((a,b),1) print(ar) ar = np.stack((a,b),0) print(ar) [0 1 2 3 4] [5 6 7 8 9] [[0 1 2 3 4] [5 6 7 8 9]] [0 1 2 3 4 5 6 7 8 9] [[0 5] [1 6] [2 7] [3 8] [4 9]] [[0 1 2 3 4] [5 6 7 8 9]] In [28]: #陣列的拆分 ar = np.arange(16).reshape(4,4) print(ar) #按列拆分 ar1 = np.hsplit(ar,2) print(ar1) #按行拆分 ar1 = np.vsplit(ar,2) print(ar1) [[ 0 1 2 3] [ 4 5 6 7] [ 8 9 10 11] [12 13 14 15]] [array([[ 0, 1], [ 4, 5], [ 8, 9], [12, 13]]), array([[ 2, 3], [ 6, 7], [10, 11], [14, 15]])] [array([[0, 1, 2, 3], [4, 5, 6, 7]]), array([[ 8, 9, 10, 11], [12, 13, 14, 15]])] In [44]: #陣列的簡單運算 ar = np.arange(10) print(ar) print(ar*2) print(ar/2) print(ar+2) print(ar//2) print(ar**2) [0 1 2 3 4 5 6 7 8 9] [ 0 2 4 6 8 10 12 14 16 18] [0. 0.5 1. 1.5 2. 2.5 3. 3.5 4. 4.5] [ 2 3 4 5 6 7 8 9 10 11] [0 0 1 1 2 2 3 3 4 4] [ 0 1 4 9 16 25 36 49 64 81]
import numpy as np In [12]: ar = np.arange(20) #一維陣列切片 ar = np.arange(20) print(ar[10]) print(ar[0:10]) print(ar[0:10:2]) 10 [0 1 2 3 4 5 6 7 8 9] [0 2 4 6 8] In [35]: #二維陣列的切片 ar = np.arange(16).reshape(4,4) print(ar) print(ar[0],ar[0,1]) print(ar[0:2]) print(ar[2:3]) print(ar[:4,2:]) [[ 0 1 2 3] [ 4 5 6 7] [ 8 9 10 11] [12 13 14 15]] [0 1 2 3] 1 [[0 1 2 3] [4 5 6 7]] [[ 8 9 10 11]] [[ 2 3] [ 6 7] [10 11] [14 15]] In [38]: #複製 #複製 br = ar.copy() cr = ar print(id(cr),id(br),id(ar)) 92916432 92916352 92916432 In [41]: _new = ar[ar > 5] print(ar_new) #小作業 ar = np.arange(10).reshape(2,5) ar_new = ar[ar > 5] print(ar_new) [6 7 8 9] In [48]: #隨機數 正態分佈 samples = np.random.normal(size=(100)) print(samples) import matplotlib.pyplot as plt % matplotlib inline plt.hist(samples) [-0.61227469 -0.36103937 0.28734323 0.63352977 1.13658948 1.15243507 0.70852784 0.40504143 0.2712728 -0.27399682 -0.68746028 0.15983535 -0.02442818 -1.71875477 -1.90132349 -1.52038551 2.04971197 -0.97968765 -0.7398707 2.58470461 -0.06499621 -0.52215928 1.11283527 1.77455139 -0.29510978 -0.3529555 0.25065179 -3.32293627 0.39312473 0.64401673 -0.24950078 1.29350219 -0.10785972 0.82839765 -0.14743582 1.28908868 1.38144382 1.1830453 1.55924739 1.00124522 -0.31871122 0.66595038 0.37629854 -0.20744411 -0.44468078 -0.57489172 0.16270152 1.96241238 1.69363656 -1.89301945 -0.80169326 -0.20132487 -0.4355692 -0.69024891 0.58913554 -1.62332398 -0.46826792 0.12339841 0.12177624 0.60421575 -0.00350087 1.28861972 0.10536274 1.36020219 -0.4496176 0.65117662 1.03193623 0.82908065 0.18313433 -0.02936144 -2.60726463 -1.58424282 -0.09221244 0.39805004 -0.46286613 -0.13994535 0.46165815 1.7918481 1.34881523 0.49391726 0.10413002 1.03006967 -1.00628197 0.556884 -0.86332625 -1.02390836 -0.46605733 -0.32249655 0.61331111 0.39275608 -0.70372193 0.37659617 -0.39106357 0.22215355 -1.40161871 1.04150908 -0.12478619 2.40395458 1.26045816 -0.45474683] Out[48]: (array([ 1., 1., 5., 5., 17., 25., 21., 17., 5., 3.]), array([-3.32293627, -2.73217218, -2.1414081 , -1.55064401, -0.95987992, -0.36911583, 0.22164826, 0.81241234, 1.40317643, 1.99394052, 2.58470461]), <a list of 10 Patch objects>)
#生成100個隨機浮點數 均勻分佈 a = np.random.rand(1000) % matplotlib inline plt.hist(a) (array([110., 93., 89., 106., 95., 105., 115., 90., 97., 100.]), array([5.47661376e-05, 1.00025836e-01, 1.99996906e-01, 2.99967976e-01, 3.99939045e-01, 4.99910115e-01, 5.99881185e-01, 6.99852255e-01, 7.99823325e-01, 8.99794394e-01, 9.99765464e-01]), <a list of 10 Patch objects>)
#生成一個浮點數或者浮點陣列 正態分佈
samples1 = np.random.randn(1000)
samples2 = np.random.randn(1000)
plt.scatter(samples1,samples2)
<matplotlib.collections.PathCollection at 0x112d2710>