1
2
3
4
| import torch
import d2lzh_pytorch
import os
import pandas as pd
|
1
2
3
4
5
6
| x=torch.arange(12)
print(x)
x.shape
x.numel
x.reshape(3,4)
y=torch.zeros(2,3,4)
|
tensor([ 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11])
1
2
3
4
5
6
7
8
9
10
11
12
13
| x = torch.tensor([[1, 2, 3], [3, 4, 5], [4, 3, 2]])
print(x)
x = torch.tensor([2.0, 2, 2, 2])
print(x)
x = torch.arange(12, dtype=torch.float32).reshape((3, 4))
print(x)
y = torch.tensor([[2.0, 1, 5, 3], [3, 2, 2, 3], [3, 2, 3, 3]])
print(y)
torch.cat((x, y), dim=0), torch.cat((x, y), dim=1)
x==y
x.sum()
|
tensor([[1, 2, 3],
[3, 4, 5],
[4, 3, 2]])
tensor([2., 2., 2., 2.])
tensor([[ 0., 1., 2., 3.],
[ 4., 5., 6., 7.],
[ 8., 9., 10., 11.]])
tensor([[2., 1., 5., 3.],
[3., 2., 2., 3.],
[3., 2., 3., 3.]])
tensor(66.)
1
2
3
4
5
|
a = torch.arange(3).reshape((3, 1))
b = torch.arange(2).reshape((1, 2))
a, b
a+b
|
tensor([[0, 1],
[1, 2],
[2, 3]])
1
2
3
4
5
| print(x)
x[-1], x[1:3]
|
tensor([[ 0., 1., 2., 3.],
[ 4., 5., 6., 7.],
[ 8., 9., 10., 11.]])
(tensor([ 8., 9., 10., 11.]),
tensor([[ 4., 5., 6., 7.],
[ 8., 9., 10., 11.]]))
1
2
3
4
5
6
7
8
9
| x=torch.zeros(3,4)
print(x)
x[1,2]=9
print(x)
x[0:2,:]=12
print(x)
before = id(y)
y+=x
id(y)==before
|
tensor([[0., 0., 0., 0.],
[0., 0., 0., 0.],
[0., 0., 0., 0.]])
tensor([[0., 0., 0., 0.],
[0., 0., 9., 0.],
[0., 0., 0., 0.]])
tensor([[12., 12., 12., 12.],
[12., 12., 12., 12.],
[ 0., 0., 0., 0.]])
True
1
2
3
| before=id(x)
x+=y
id(x)==before
|
True
1
2
3
4
5
6
7
8
9
10
11
| os.makedirs(os.path.join('./', "data"), exist_ok=True)
data_file = os.path.join('./', 'data', 'house_tiny.csv')
with open(data_file, 'w') as f:
f.write('NumRoom,Alley,Price\n')
f.write('Na,Pave,127500\n')
f.write('2,Na,199999\n')
f.write('5,Jager,199999\n')
f.write('3,Bruse,199999\n')
data = pd.read_csv(data_file)
data
|
|
NumRoom |
Alley |
Price |
| 0 |
Na |
Pave |
127500 |
| 1 |
2 |
Na |
199999 |
| 2 |
5 |
Jager |
199999 |
| 3 |
3 |
Bruse |
199999 |
1
2
3
4
5
6
| a = torch.arange(12)
b = a.reshape((3, 4))
b[:] = 2
print(a)
print(b)
|
tensor([2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2])
tensor([[2, 2, 2, 2],
[2, 2, 2, 2],
[2, 2, 2, 2]])
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
| x = torch.tensor([3.0])
y = torch.tensor([3.9])
x+y
x = torch.arange(4)
print(len(x))
print(x.shape)
a = torch.arange(20).reshape(5, 4)
print(a)
print(a.T)
a = torch.arange(20, dtype=torch.float32).reshape(5, 4)
b = a.clone()
print(b)
print(a*b)
a = 2
|
4
torch.Size([4])
tensor([[ 0, 1, 2, 3],
[ 4, 5, 6, 7],
[ 8, 9, 10, 11],
[12, 13, 14, 15],
[16, 17, 18, 19]])
tensor([[ 0, 4, 8, 12, 16],
[ 1, 5, 9, 13, 17],
[ 2, 6, 10, 14, 18],
[ 3, 7, 11, 15, 19]])
tensor([[ 0., 1., 2., 3.],
[ 4., 5., 6., 7.],
[ 8., 9., 10., 11.],
[12., 13., 14., 15.],
[16., 17., 18., 19.]])
tensor([[ 0., 1., 4., 9.],
[ 16., 25., 36., 49.],
[ 64., 81., 100., 121.],
[144., 169., 196., 225.],
[256., 289., 324., 361.]])
1
2
3
4
5
6
7
| a = torch.arange(12).reshape(3, 4)
print(a)
sum_1 = a.sum(axis=0)
print(sum_1)
sum_2 = a.sum(axis=1)
print(sum_2)
|
tensor([[ 0, 1, 2, 3],
[ 4, 5, 6, 7],
[ 8, 9, 10, 11]])
tensor([12, 15, 18, 21])
tensor([ 6, 22, 38])
1
2
3
| a = torch.arange(12, dtype=float).reshape(3, 4)
print(a.mean())
print(a.sum()/a.numel())
|
tensor(5.5000, dtype=torch.float64)
tensor(5.5000, dtype=torch.float64)
1
2
3
4
5
6
7
|
a = torch.arange(12, dtype=float).reshape(3, 4)
sum_a = a.sum(axis=1, keepdim=True)
print(sum_a)
sum1=a.cumsum(axis=0)
print(sum1)
|
tensor([[ 6.],
[22.],
[38.]], dtype=torch.float64)
tensor([[ 0., 1., 2., 3.],
[ 4., 6., 8., 10.],
[12., 15., 18., 21.]], dtype=torch.float64)
1
2
3
4
5
6
7
8
| a = torch.arange(12, dtype=float)
b = torch.ones(12, dtype=float)
print(torch.dot(a, b))
a = torch.arange(12, dtype=float).reshape(3, 4)
b = torch.ones((3, 4))
torch.sum(a*b)
|
tensor(66., dtype=torch.float64)
tensor(66., dtype=torch.float64)
1
2
| u = torch.tensor([3.0, 4, 5])
torch.norm(u)
|
tensor(7.0711)
1
2
3
4
5
6
| a = torch.ones((2, 5, 4))
print(a)
print(a.sum(axis=1).shape)
print(a.sum(axis=0).shape)
print(a.sum(axis=[0, 2]).shape)
|
tensor([[[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.]]])
torch.Size([2, 4])
torch.Size([5, 4])
torch.Size([5])
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
| a = torch.arange(4, dtype=float, requires_grad=True)
print(a.grad)
b = 2*torch.dot(a, a)
b.backward()
print(a.grad)
a.grad.zero_()
y=a.sum()
y.backward()
print(a.grad)
a.grad.zero_()
y=a*a
y.sum().backward()
a.grad
a.grad.zero_()
y=a*a
u=y.detach()
z=u*a
z.sum().backward()
a.grad==u
|
None
tensor([ 0., 4., 8., 12.], dtype=torch.float64)
tensor([1., 1., 1., 1.], dtype=torch.float64)
tensor([True, True, True, True])
1
2
3
4
5
6
7
8
9
10
| a = torch.arange(4, dtype=float).reshape(2, 2)
b = torch.arange(4, dtype=float).reshape(2, 2)
result = torch.empty(2, 2)
print(result)
torch.add(a, b, out=result)
print(result)
print(a)
print(b)
a.add_(b)
print(a)
|
tensor([[0., 2.],
[0., 0.]])
tensor([[0., 2.],
[4., 6.]])
tensor([[0., 1.],
[2., 3.]], dtype=torch.float64)
tensor([[0., 1.],
[2., 3.]], dtype=torch.float64)
tensor([[0., 2.],
[4., 6.]], dtype=torch.float64)
1
2
3
4
| x = torch.arange(12, dtype=float).reshape(3, 4)
print(x)
print(x[:, 2])
print(x[1:3, :])
|
tensor([[ 0., 1., 2., 3.],
[ 4., 5., 6., 7.],
[ 8., 9., 10., 11.]], dtype=torch.float64)
tensor([ 2., 6., 10.], dtype=torch.float64)
tensor([[ 4., 5., 6., 7.],
[ 8., 9., 10., 11.]], dtype=torch.float64)
