util.py

import torch
import torch.nn.functional as F
from torch import nn, Tensor

class CircleLoss(nn.Module):
    def __init__(self, m: float, gamma: float) -> None:
        super(CircleLoss, self).__init__()
        self.m = m
        self.gamma = gamma
        self.soft_plus = nn.Softplus()

    def forward(self, sp: Tensor, sn: Tensor) -> Tensor:
        ap = torch.clamp_min(- sp.detach() + 1 + self.m, min=0.)
        an = torch.clamp_min(sn.detach() + self.m, min=0.)

        delta_p = 1 - self.m
        delta_n = self.m

        logit_p = - ap * (sp - delta_p) * self.gamma
        logit_n = an * (sn - delta_n) * self.gamma

        loss = self.soft_plus(torch.logsumexp(logit_n, dim=0) + torch.logsumexp(logit_p, dim=0))

        return loss

def cal_loss(pred, gold, smoothing=True):
    ''' Calculate cross entropy loss, apply label smoothing if needed. '''

    gold = gold.contiguous().view(-1)
    if smoothing:
        eps = 0.2
        n_class = pred.size(1)

        one_hot = torch.zeros_like(pred).scatter(1, gold.view(-1, 1), 1)
        one_hot = one_hot * (1 - eps) + (1 - one_hot) * eps / (n_class - 1)
        log_prb = F.log_softmax(pred, dim=1)

        loss = -(one_hot * log_prb).sum(dim=1).mean()
    else:
        loss = F.cross_entropy(pred, gold, reduction='mean')

    return loss


class IOStream():
    def __init__(self, path):
        self.f = open(path, 'a')

    def cprint(self, text):
        print(text)
        self.f.write(text+'\n')
        self.f.flush()

    def close(self):
        self.f.close()

def huber_loss(error, delta):
    abs_error = torch.abs(error)
    quadratic = torch.min(abs_error, torch.full_like(abs_error, fill_value=delta))
    losses = 0.5 * (quadratic ** 2) + delta * (abs_error - quadratic)
    return torch.mean(losses)