losses.py

import torch
import torch.nn as nn
from torch.nn import functional as F
import numpy as np
import contextlib



@contextlib.contextmanager
def _disable_tracking_bn_stats(model):
    def switch_attr(m):
        if hasattr(m, 'track_running_stats'):
            m.track_running_stats ^= True

    model.apply(switch_attr)
    yield
    model.apply(switch_attr)


def _l2_normalize(d):
    d_reshaped = d.view(d.shape[0], -1, *(1 for _ in range(d.dim() - 2)))
    d /= torch.norm(d_reshaped, dim=1, keepdim=True) + 1e-8
    return d


class VATLoss(nn.Module):

    def __init__(self, xi=10.0, eps=1.0, ip=1):
        """VAT loss
        :param xi: hyperparameter of VAT (default: 10.0)
        :param eps: hyperparameter of VAT (default: 1.0)
        :param ip: iteration times of computing adv noise (default: 1)
        """
        super(VATLoss, self).__init__()
        self.xi = xi
        self.eps = eps
        self.ip = ip

    def forward(self, model, x):
        with torch.no_grad():
            pred = F.softmax(model(x)[1], dim=1)

        # prepare random unit tensor
        d = torch.rand(x.shape).sub(0.5).to(x.device)
        d = _l2_normalize(d)

        with _disable_tracking_bn_stats(model):
            # calc adversarial direction
            for _ in range(self.ip):
                d.requires_grad_()
                _,pred_hat = model(x + self.xi * d)
                logp_hat = F.log_softmax(pred_hat, dim=1)
                adv_distance = F.kl_div(logp_hat, pred, reduction='batchmean')
                adv_distance.backward()
                d = _l2_normalize(d.grad)
                model.zero_grad()

            # calc LDS
            r_adv = d * self.eps
            _,pred_hat = model(x + r_adv)
            logp_hat = F.log_softmax(pred_hat, dim=1)
            lds = F.kl_div(logp_hat, pred, reduction='batchmean')

        return lds


class SupConLoss(nn.Module):
    """Supervised Contrastive Learning: https://arxiv.org/pdf/2004.11362.pdf.
    It also supports the unsupervised contrastive loss in SimCLR"""
    def __init__(self, temperature=0.07, contrast_mode='all',
                 base_temperature=0.07):
        super(SupConLoss, self).__init__()
        self.temperature = temperature
        self.contrast_mode = contrast_mode
        self.base_temperature = base_temperature

    def forward(self, features, labels=None, mask=None):
        """Compute loss for model. If both `labels` and `mask` are None,
        it degenerates to SimCLR unsupervised loss:
        https://arxiv.org/pdf/2002.05709.pdf
        Args:
            features: hidden vector of shape [bsz, n_views, ...].
            labels: ground truth of shape [bsz].
            mask: contrastive mask of shape [bsz, bsz], mask_{i,j}=1 if sample j
                has the same class as sample i. Can be asymmetric.
        Returns:
            A loss scalar.
        """
        device = (torch.device('cuda')
                  if features.is_cuda
                  else torch.device('cpu'))

        if len(features.shape) < 3:
            raise ValueError('`features` needs to be [bsz, n_views, ...],'
                             'at least 3 dimensions are required')
        if len(features.shape) > 3:
            features = features.view(features.shape[0], features.shape[1], -1)

        batch_size = features.shape[0]
        if labels is not None and mask is not None:
            raise ValueError('Cannot define both `labels` and `mask`')
        elif labels is None and mask is None:
            mask = torch.eye(batch_size, dtype=torch.float32).to(device)
        elif labels is not None:
            labels = labels.contiguous().view(-1, 1)
            if labels.shape[0] != batch_size:
                raise ValueError('Num of labels does not match num of features')
            mask = torch.eq(labels, labels.T).float().to(device)
        else:
            mask = mask.float().to(device)

        contrast_count = features.shape[1]
        contrast_feature = torch.cat(torch.unbind(features, dim=1), dim=0)
        if self.contrast_mode == 'one':
            anchor_feature = features[:, 0]
            anchor_count = 1
        elif self.contrast_mode == 'all':
            anchor_feature = contrast_feature
            anchor_count = contrast_count
        else:
            raise ValueError('Unknown mode: {}'.format(self.contrast_mode))

        # compute logits
        anchor_dot_contrast = torch.div(
            torch.matmul(anchor_feature, contrast_feature.T),
            self.temperature)
        # for numerical stability
        logits_max, _ = torch.max(anchor_dot_contrast, dim=1, keepdim=True)
        logits = anchor_dot_contrast - logits_max.detach()

        # tile mask
        mask = mask.repeat(anchor_count, contrast_count)
        # mask-out self-contrast cases
        logits_mask = torch.scatter(
            torch.ones_like(mask),
            1,
            torch.arange(batch_size * anchor_count).view(-1, 1).to(device),
            0
        )
        mask = mask * logits_mask

        # compute log_prob
        exp_logits = torch.exp(logits) * logits_mask
        log_prob = logits - torch.log(exp_logits.sum(1, keepdim=True))

        # compute mean of log-likelihood over positive
        mean_log_prob_pos = (mask * log_prob).sum(1) / mask.sum(1)

        # loss
        loss = - (self.temperature / self.base_temperature) * mean_log_prob_pos
        loss = loss.view(anchor_count, batch_size).mean()

        return loss

def bnm_loss(out_logits):
    """
    compute batch nuclear-norm maximization loss
    """
    A = F.softmax(out_logits, dim=1)
    L_bnm = -torch.norm(A,'nuc')
    return L_bnm

def bnm_loss_improve(out_logits):
    """
    compute batch nuclear-norm maximization loss refinement
    """
    A = F.softmax(out_logits, dim=1) + 0.000001
    B = -1.0 * A *torch.log(A)
    C = B.sum(dim=1)
    index = C.argsort(descending=True)[:24]
    D = A[index]
    L_bnm = -torch.norm(D,'nuc')
    return L_bnm