numpy_torch_basic

import numpy as np

Create

arr1 = np.array([1.,2.,3.])
arr1

array([1., 2., 3.])

arr2 = arr1.astype(int)
arr2

array([1, 2, 3])

arr3 = np.ones((3,2,1))
arr3

array([[[1.],
[1.]],

[[1.],
[1.]],

[[1.],
[1.]]])

a = np.arange(1,21,2)
a

array([ 1, 3, 5, 7, 9, 11, 13, 15, 17, 19])

a = np.random.random((3,2))
a

array([[0.30421515, 0.53292776],
[0.28661949, 0.55374416],
[0.54982564, 0.451857 ]])

a = np.random.rand(3,2)
a

array([[0.77216069, 0.40891772],
[0.92578388, 0.08103115],
[0.51501061, 0.55870138]])

a = np.random.randint(1,10,(2,3))
a

array([[1, 4, 2],
[3, 2, 9]])

a = np.random.normal(0,1,(2,3))
a

array([[-1.52450694, 0.81997198, 0.22341405],
[-1.23589707, 1.75922823, -0.77042462]])

a = np.random.randn(2,3)
a

array([[-1.9286721 , -0.49341625, -0.72808546],
[ 0.56516771, -0.04973357, -0.47164904]])

Fancy Index

a = np.arange(1,17).reshape(4,4)
a

array([[ 1, 2, 3, 4],
[ 5, 6, 7, 8],
[ 9, 10, 11, 12],
[13, 14, 15, 16]])

print(a[[0,1],[0,2]])
print(a[[0,1,2,3],[3,2,1,0]])
print(a[[1],[1]])
print(a[1][1])

[ 1 100]
[100 100 100 100]
[6]
6

a[[0,1,2,3],[3,2,1,0]] = 100
a

array([[ 1, 2, 3, 100],
[ 5, 6, 100, 8],
[ 9, 100, 11, 12],
[100, 14, 15, 16]])

Slice

a = np.arange(1,21).reshape(4,5)
b = a.copy()
print(a)
print(b)
a

[[ 1 2 3 4 5]
[ 6 7 8 9 10]
[11 12 13 14 15]
[16 17 18 19 20]]
[[ 1 2 3 4 5]
[ 6 7 8 9 10]
[11 12 13 14 15]
[16 17 18 19 20]]

array([[ 1, 2, 3, 4, 5],
[ 6, 7, 8, 9, 10],
[11, 12, 13, 14, 15],
[16, 17, 18, 19, 20]])

a[1::2,2::]

array([[ 8, 9, 10],
[18, 19, 20]])

print(a[2])
print(a[2,:])

[11 12 13 14 15]
[11 12 13 14 15]

print(a[:,2])
print(a[:,1:3])

[ 3 8 13 18]
[[ 2 3]
[ 7 8]
[12 13]
[17 18]]

a[:2] = 100
a

array([[100, 100, 100, 100, 100],
[100, 100, 100, 100, 100],
[ 11, 12, 13, 14, 15],
[ 16, 17, 18, 19, 20]])

Deform

a = np.arange(1,7).reshape(2,3)
a,a.T

(array([[1, 2, 3],
[4, 5, 6]]),
array([[1, 4],
[2, 5],
[3, 6]]))

a = np.arange(10)
a_ud = np.flipud(a)
a,a_ud

array([0, 1, 2, 3, 4, 5, 6, 7, 8, 9])

a = np.arange(20).reshape(4,5)
a_ud = np.flipud(a)
a_lr = np.fliplr(a)
print(a)
print(a_ud)
print(a_lr)

[[ 0 1 2 3 4]
[ 5 6 7 8 9]
[10 11 12 13 14]
[15 16 17 18 19]]
[[15 16 17 18 19]
[10 11 12 13 14]
[ 5 6 7 8 9]
[ 0 1 2 3 4]]
[[ 4 3 2 1 0]
[ 9 8 7 6 5]
[14 13 12 11 10]
[19 18 17 16 15]]

a = np.array([
    [1,2,3],
    [4,5,6]
])
b = np.array([
    [7,8,9],
    [10,11,12]
])
c = np.concatenate([a, b])
d = np.concatenate([a, b], axis=1)
print(c)
print(d)

[[ 1 2 3]
[ 4 5 6]
[ 7 8 9]
[10 11 12]]
[[ 1 2 3 7 8 9]
[ 4 5 6 10 11 12]]

Functions

a = np.arange(5)
b = np.arange(5)
c = np.dot(a,b)
a,b,c

(array([0, 1, 2, 3, 4]), array([0, 1, 2, 3, 4]), 30)

a = np.arange(12).reshape(3,4)
b = np.arange(20).reshape(4,5)
c = np.dot(a,b)
a,b,c

(array([[ 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, 12, 13, 14],
[15, 16, 17, 18, 19]]),
array([[ 70, 76, 82, 88, 94],
[190, 212, 234, 256, 278],
[310, 348, 386, 424, 462]]))

x = np.arange(-2,3)
x1 = np.abs(x)
print(x)
print(x1)

[-2 -1 0 1 2]
[2 1 0 1 2]

theta = np.arange(3)*np.pi/2.
print(theta)
print(np.sin(theta))
print(np.cos(theta))
print(np.tan(np.pi/4))

[0. 1.57079633 3.14159265]
[0.0000000e+00 1.0000000e+00 1.2246468e-16]
[ 1.000000e+00 6.123234e-17 -1.000000e+00]
0.9999999999999999

a = np.random.randn(10000000)
num = np.sum(a<0)
num

5000092

a = np.arange(11)
b = np.flipud(a)
print(a)
print(b)
print(np.any(a==b))
print(np.all(a==b))

[ 0 1 2 3 4 5 6 7 8 9 10]
[10 9 8 7 6 5 4 3 2 1 0]
True
False

print(np.where(a>5))
print(np.where(a==np.max(a)))

(array([ 6, 7, 8, 9, 10], dtype=int64),)
(array([10], dtype=int64),)

torch

import torch

x = np.random.randn(2,8,3,4)
y = torch.randn(2,8,3,4)
z = torch.randn(2,8,4,5)
res = torch.matmul(y,z)
res2 = y@z
print(res.shape)
print(torch.all(res==res2))

torch.Size([2, 8, 3, 5])
tensor(True)

a = torch.arange(5)
b = a.type(torch.float32)
b

tensor([0., 1., 2., 3., 4.])

a = np.arange(5)
b = a.astype(float)
b

array([0., 1., 2., 3., 4.])