def _batch_sum_bce(x, target, name='binary_cross_entropy'):
logits = functions.binary_cross_entropy_loss_with_logits(x,
target,
name=name)
if per_output_weights is not None:
logits *= per_output_weights
return functions.reduce_batch_sum(logits)
def testBinaryCorssEntropyLossWithLogits(self):
n1 = numpy.array([2., 3., 4., 5., -6., -7.], dtype=numpy.float32)
n2 = numpy.array([1., 1., 0., 0., 0., 1.], dtype=numpy.float32)
ftensor1 = tf.constant(n1)
ftensor2 = tf.constant(n2)
out = self.Run(
functions.binary_cross_entropy_loss_with_logits(
ftensor1, ftensor2))
testing.assert_allclose(out[0],
n1 * (1 - n2) + numpy.log(1 + numpy.exp(-n1)),
rtol=TOLERANCE)
def testBinaryCorssEntropyLossWithLogits(self):
n1 = numpy.array([2., 3., 4., 5., -6., -7.], dtype=numpy.float32)
n2 = numpy.array([1., 1., 0., 0., 0., 1.], dtype=numpy.float32)
ftensor1 = tf.constant(n1)
ftensor2 = tf.constant(n2)
out = self.Run(functions.binary_cross_entropy_loss_with_logits(ftensor1,
ftensor2))
testing.assert_allclose(
out[0],
n1 * (1-n2) + numpy.log(1 + numpy.exp(-n1)),
rtol=TOLERANCE)
def _batch_sum_bce(x, target, name='binary_cross_entropy'):
return functions.reduce_batch_sum(
functions.binary_cross_entropy_loss_with_logits(x, target, name=name))
def _batch_sum_bce(x, target, name="binary_cross_entropy"):
logits = functions.binary_cross_entropy_loss_with_logits(x, target, name=name)
if per_output_weights is not None:
logits *= per_output_weights
return functions.reduce_batch_sum(logits)
