torcheval.metrics.functional.binary_accuracy¶
- torcheval.metrics.functional.binary_accuracy(input: Tensor, target: Tensor, *, threshold: float = 0.5) Tensor ¶
计算二元准确率得分,即输入与目标匹配的频率。它的类版本是
torcheval.metrics.BinaryAccuracy
。- 参数::
input (Tensor) – 形状为 (n_sample,) 的标签预测张量。
torch.where(input < threshold, 0, 1)
将应用于输入。target (Tensor) – 形状为 (n_sample,) 的真值标签张量。
threshold (float, default 0.5) – 用于将输入转换为每个样本的预测标签的阈值。
torch.where(input < threshold, 0, 1)
将应用于input
。
示例
>>> import torch >>> from torcheval.metrics.functional import binary_accuracy >>> input = torch.tensor([0, 0, 1, 1]) >>> target = torch.tensor([1, 0, 1, 1]) >>> binary_accuracy(input, target) tensor(0.75) # 3 / 4 >>> input = torch.tensor([0, 0.2, 0.6, 0.7]) >>> target = torch.tensor([1, 0, 1, 1]) >>> binary_accuracy(input, target, threshold=0.7) tensor(0.5) # 2 / 4