tf.nn.sigmoid_cross_entropy_with_logits

最新推荐文章于 2021-03-13 16:49:54 发布

懦夫的抉择

最新推荐文章于 2021-03-13 16:49:54 发布

阅读量1.9k

点赞数

分类专栏： tensorflowandbiye

本文链接：https://blog.youkuaiyun.com/li872149120/article/details/79253871

版权

该博客详细介绍了TensorFlow中tf.nn.sigmoid_cross_entropy_with_logits的用法，主要应用于独立且非互斥的分类任务，如多标签分类。它计算给定logits的sigmoid交叉熵损失，讨论了函数的稳定实现以及防止溢出的方法，并提供了操作的输入参数和返回值的说明。

摘要生成于 C知道，由 DeepSeek-R1 满血版支持，前往体验 >

tf.nn.sigmoid_cross_entropy_with_logits

sigmoid_cross_entropy_with_logits(
    _sentinel=None,
    labels=None,
    logits=None,
    name=None
)

Defined in tensorflow/python/ops/nn_impl.py.

See the guide: Neural Network > Classification

Computes sigmoid cross entropy given logits.

Measures the probability error in discrete classification tasks in which eachclass is independent and not mutually exclusive. For instance, one couldperform multilabel classification where a picture can contain both an elephantand a dog at the same time.

For brevity, let x = logits, z = labels. The logistic loss is