Source code for neodroidvision.segmentation.evaluation.focal_loss

#!/usr/bin/env python3
# -*- coding: utf-8 -*-

__author__ = "Christian Heider Nielsen"

import torch
from torch import nn
from torch.autograd import Variable

__all__ = ["FocalLoss"]


class FocalLoss(nn.Module):
    r"""
    This criterion is a implementation of Focal Loss, which is proposed in
    Focal Loss for Dense Object Detection.

    Loss(x, class) = - \alpha (1-softmax(x)[class])^gamma \log(softmax(x)[class])

    The loss_functions are averaged across observations for each mini batch.
    Args:
    alpha(1D Tensor, Variable) : the scalar factor for this criterion
    gamma(float, double) : gamma > 0; reduces the relative loss for well-classified examples (p > .5),
                       putting more focus on hard, misclassified examples
    size_average(bool): size_average(bool): By default, the loss_functions are averaged over
    observations for
    each mini batch.
                    However, if the field size_average is set to False, the loss_functions are
                    instead summed for each mini batch."""

    def __init__(
        self, class_num, alpha=None, gamma: float = 2.0, size_average: bool = True
    ):
        super(FocalLoss, self).__init__()
        if alpha is None:
            self.alpha = Variable(torch.ones(class_num, 1))
        else:
            if isinstance(alpha, Variable):
                self.alpha = alpha
            else:
                self.alpha = Variable(alpha)
        self.gamma = gamma
        self.class_num = class_num
        self.size_average = size_average

    def forward(self, inputs: torch.Tensor, targets: torch.Tensor) -> torch.Tensor:
        """

        Args:
          inputs:
          targets:

        Returns:

        """
        N = inputs.size(0)
        C = inputs.size(1)
        P = torch.softmax(
            inputs, 0
        )  # TODO: use log_softmax? Check dim maybe it should be 1

        class_mask = inputs.data.new(N, C).fill_(0)
        class_mask = Variable(class_mask)
        ids = targets.reshape(-1, 1)
        class_mask.scatter_(1, ids.data, 1.0)

        if inputs.is_cuda and not self.alpha.is_cuda:
            self.alpha = self.alpha.cuda()
        alpha = self.alpha[ids.data.view(-1)]

        probs = (P * class_mask).sum(1).reshape(-1, 1)

        log_p = probs.log()

        batch_loss = -alpha * (torch.pow((1 - probs), self.gamma)) * log_p

        if self.size_average:
            loss = batch_loss.mean()
        else:
            loss = batch_loss.sum()
        return loss


if __name__ == "__main__":
    alpha = torch.rand(21, 1)
    focal_loss_func = FocalLoss(class_num=5, gamma=0)
    cross_entropy_func = nn.CrossEntropyLoss()
    N = 4
    C = 5
    inputs = torch.rand(N, C)
    targets = torch.LongTensor(N).random_(C)
    inputs_fl = Variable(inputs.clone(), requires_grad=True)
    targets_fl = Variable(targets.clone())

    inputs_ce = Variable(inputs.clone(), requires_grad=True)
    targets_ce = Variable(targets.clone())
    print("----inputs----")
    print(inputs)
    print("---target-----")
    print(targets)

    fl_loss = focal_loss_func(inputs_fl, targets_fl)
    ce_loss = cross_entropy_func(inputs_ce, targets_ce)
    print(f"ce = {ce_loss.item()}, fl ={fl_loss.item()}")
    fl_loss.backward()
    ce_loss.backward()
    print(inputs_ce.grad.data)