Balanced Meta-Softmax for Long-Tailed Visual Recognition
发布时间
阅读量：
阅读量
def balanced_softmax_loss(labels, logits, sample_per_class, reduction):
"""Compute the Balanced Softmax Loss between `logits` and the ground truth `labels`.
Args:
labels: A int tensor of size [batch].
logits: A float tensor of size [batch, no_of_classes].
sample_per_class: A int tensor of size [no of classes].
reduction: string. One of "none", "mean", "sum"
Returns:
loss: A float tensor. Balanced Softmax Loss.
"""
spc = sample_per_class.type_as(logits)
spc = spc.unsqueeze(0).expand(logits.shape[0], -1)
logits = logits + spc.log()
loss = F.cross_entropy(input=logits, target=labels, reduction=reduction)
return loss这种策略能够更加有效地聚焦于少数类别。对于少数类样本而言，其取对数后的值较小会导致对应的logits输出较大。大致如此。文章中还提到了一种叫做'元采样器'的技术细节尚不明确。
全部评论 (0)
还没有任何评论哟~