"Binary Cross Entropy" 的 Tensorflow 损失是多少？

Question

我正在尝试将 Keras 图重写为 Tensorflow 图，但想知道哪个损失函数相当于 "Binary Cross Entropy"。是 tf.nn.softmax_cross_entropy_with_logits_v2?

非常感谢！

Answer 1

不，带有 tensorflow 后端的 binary_crossentropy 的实现被定义为 here 为

@tf_export('keras.backend.binary_crossentropy')
def binary_crossentropy(target, output, from_logits=False):
    """Binary crossentropy between an output tensor and a target tensor.
    Arguments:
      target: A tensor with the same shape as `output`.
      output: A tensor.
      from_logits: Whether `output` is expected to be a logits tensor.
          By default, we consider that `output`
          encodes a probability distribution.
    Returns:
      A tensor.
    """
    # Note: nn.sigmoid_cross_entropy_with_logits
    # expects logits, Keras expects probabilities.
    if not from_logits:
        # transform back to logits
        epsilon_ = _to_tensor(epsilon(), output.dtype.base_dtype)
        output = clip_ops.clip_by_value(output, epsilon_, 1 - epsilon_)
        output = math_ops.log(output / (1 - output))
    return nn.sigmoid_cross_entropy_with_logits(labels=target, logits=output)

因此，它使用 sigmoid_crossentropy 而不是 softmax_crossentropy。