sigmoid_cross_entropy_with_logits¶

dragon.vm.tensorflow.nn.sigmoid_cross_entropy_with_logits(
  labels=None,
  logits=None,
  name=None
)[source]¶

Compute the loss of sigmoid cross entropy.

Examples:

x = tf.constant([0.1, 0.2])
y = tf.constant([0., 1.])
print(tf.nn.sigmoid_cross_entropy_with_logits(y, x))  # 0.744, 0.598

Parameters:

• labels (dragon.Tensor) – The target tensor.
• logits (dragon.Tensor) – The input tensor.
• name (str, optional) – The operation name.

Returns:

dragon.Tensor – The output tensor.