import numpy as np# initialize with values
np.array([[1, 2, 3],[4, 5, 6]])
>> array([[1, 2, 3],
>> [4, 5, 6]])
# 1d array
np.zeros(4)
>> array([ 0., 0., 0., 0.])
# 2d array
np.zeros((4,2))
>> array([[ 0., 0.],
>> [ 0., 0.],
>> [ 0., 0.],
>> [ 0., 0.]])
# multi-dimension array
np.zeros((4,2,3))
>> array([[[ 0., 0., 0.],
>> [ 0., 0., 0.]],
>>
>> [[ 0., 0., 0.],
>> [ 0., 0., 0.]],
>>
>> [[ 0., 0., 0.],
>> [ 0., 0., 0.]],
>>
>> [[ 0., 0., 0.],
>> [ 0., 0., 0.]]])
# np.arange(start, end, increment)
np.arange(1,10,0.5)
>> array([ 1. , 1.5, 2. , 2.5, 3. , 3.5, 4. , 4.5, 5. , 5.5, 6. ,
>> 6.5, 7. , 7.5, 8. , 8.5, 9. , 9.5])
# np.linspace(start, end, numsteps)
np.linspace(1,10,20)
>> array([ 1. , 1.47368421, 1.94736842, 2.42105263,
>> 2.89473684, 3.36842105, 3.84210526, 4.31578947,
>> 4.78947368, 5.26315789, 5.73684211, 6.21052632,
>> 6.68421053, 7.15789474, 7.63157895, 8.10526316,
>> 8.57894737, 9.05263158, 9.52631579, 10. ])# access element
A = np.zeros((3,4))
A[0][0] = 1
A[1,1] = 2
A[2,2] = 3
A
>> array([[ 1., 0., 0., 0.],
>> [ 0., 2., 0., 0.],
>> [ 0., 0., 3., 0.]])
# slicing [start:end:stride], start + n*stride < end
rows = np.size(A,0)
cols = np.size(A,1)
# column 0, rows 0..rows-1
print(A[0:rows:1, 0])
print(A[:,0])
>> [ 1. 0. 0.]
>> [ 1. 0. 0.]
# sub matrix
A[1:3,1:3]
>> array([[ 2., 0.],
>> [ 0., 3.]])
# array indexing
B = np.arange(10,21)
i = [2, 4, 5]
print(B)
print(B[i])
>> [10 11 12 13 14 15 16 17 18 19 20]
>>
>> [12 14 15]
# return intersections of rows 0,2 and columns 0,1
i = [[0,2],[0,1]]
print(A)
print(A[i])
>> [[ 1. 0. 0. 0.]
>> [ 0. 2. 0. 0.]
>> [ 0. 0. 3. 0.]]
>>
>> [ 1. 0.]# flatten returns a new independant array
print(A.flatten())
>> [ 1. 0. 0. 0. 0. 2. 0. 0. 0. 0. 3. 0.]
# ravel returns a new array but accesses the same data
print(A.ravel())
>> [ 1. 0. 0. 0. 0. 2. 0. 0. 0. 0. 3. 0.]
# reshape array, new array accesses the same data
A = np.arange(12)
print(A)
A.reshape((3,4))
>> [ 0 1 2 3 4 5 6 7 8 9 10 11]
>>
>> array([[ 0, 1, 2, 3],
>> [ 4, 5, 6, 7],
>> [ 8, 9, 10, 11]])a = np.arange(20)
b = np.arange(100,120)
c = np.vstack((a,b))
print(c)
>> [[ 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
>> 18 19]
>> [100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117
>> 118 119]]# sum
print(np.sum(c,0)) # applying along the first axis will collapse the first dimension, leaving the second
print(np.sum(c,1)) # applying along the second axis will collapse the second dimension, leaving the first
print(np.sum(c)) # apply on all elements
>> [100 102 104 106 108 110 112 114 116 118 120 122 124 126 128 130 132 134
>> 136 138]
>> [ 190 2190]
>> 2380