Goal¶

This post aims to introduce basic PyTorch operations e.g., addition, multiplication,

Libraries¶

In [2]:

import numpy as np
import pandas as pd
import torch

Create a Tensor¶

In [5]:

t_x1 = torch.Tensor([[1, 2, 3],
                     [4, 5, 6],
                     [7, 8, 9]])

t_x2 = torch.Tensor([[9, 8, 7],
                     [6, 5, 4],
                     [3, 2, 1]])
print(t_x1)
print(t_x2)

tensor([[1., 2., 3.],
        [4., 5., 6.],
        [7., 8., 9.]])
tensor([[9., 8., 7.],
        [6., 5., 4.],
        [3., 2., 1.]])

Addition¶

+ operator¶

In [6]:

t_x1 + t_x2

Out[6]:

tensor([[10., 10., 10.],
        [10., 10., 10.],
        [10., 10., 10.]])

`torch.add` method¶

In [7]:

torch.add(t_x1, t_x2)

Out[7]:

tensor([[10., 10., 10.],
        [10., 10., 10.],
        [10., 10., 10.]])

In [8]:

torch.add(t_x1, -t_x2)

Out[8]:

tensor([[-8., -6., -4.],
        [-2.,  0.,  2.],
        [ 4.,  6.,  8.]])

Multiplication¶

Operator¶

In [9]:

t_x1 * t_x2

Out[9]:

tensor([[ 9., 16., 21.],
        [24., 25., 24.],
        [21., 16.,  9.]])

`torch.mm` method¶

In [11]:

torch.mm(t_x1, t_x2)

Out[11]:

tensor([[ 30.,  24.,  18.],
        [ 84.,  69.,  54.],
        [138., 114.,  90.]])

Inner product `torch.dot`¶

In [13]:

print(t_x1[:, 1], t_x2[:, 1])
torch.dot(t_x1[:, 1], t_x2[:, 1])

tensor([2., 5., 8.]) tensor([8., 5., 2.])

Out[13]:

tensor(57.)

Trace¶

In [15]:

torch.trace(t_x1)

Out[15]:

tensor(15.)

Diag¶

In [16]:

torch.diag(t_x1)

Out[16]:

tensor([1., 5., 9.])

Determinant¶

In [17]:

torch.det(t_x1)

Out[17]:

tensor(0.)