Assignment 1 - All About torch.Tensor

Deep Learning with PyTorch: Zero to GANs

PyTorch is an open source machine learning library written in Python and C++. PyTorch is a favorite among researchers because it allows for quick prototyping and has a simple API.

Loss function is a mathematical optimization function. Loss functions are located in torch.nn module in the Pytorch library. Some of the loss functions are as follows:

torch.nn.L1Loss
torch.nn.MSELoss
torch.nn.NLLLoss
torch.nn.CrossEntropyLoss
torch.nn.KLDivLoss

# Import torch and other required modules
import torch
import torch.nn as nn

Function 1 - torch.nn.L1Loss

the mean absolute error is measured

# Example 1 - 
loss = nn.L1Loss()
pred = torch.tensor(4.)
target = torch.tensor(5.)
output = loss(pred, target)
output

tensor(1.)

Explanation about example: loss(4.,5.)=|4.-5.|=|-1.|=1.