def test_weighted(self):
msle_obj = metrics.MeanSquaredLogarithmicError()
self.evaluate(variables.variables_initializer(msle_obj.variables))
y_true = constant_op.constant(((0, 1, 0, 1, 0), (0, 0, 1, 1, 1),
(1, 1, 1, 1, 0), (0, 0, 0, 0, 1)))
y_pred = constant_op.constant(((0, 0, 1, 1, 0), (1, 1, 1, 1, 1),
(0, 1, 0, 1, 0), (1, 1, 1, 1, 1)))
sample_weight = constant_op.constant((1., 1.5, 2., 2.5))
result = msle_obj(y_true, y_pred, sample_weight=sample_weight)
self.assertAllClose(0.26082, self.evaluate(result), atol=1e-5)
def test_config(self):
msle_obj = metrics.MeanSquaredLogarithmicError(name='my_msle',
dtype=dtypes.int32)
self.assertEqual(msle_obj.name, 'my_msle')
self.assertEqual(msle_obj._dtype, dtypes.int32)
# Check save and restore config
msle_obj2 = metrics.MeanSquaredLogarithmicError.from_config(
msle_obj.get_config())
self.assertEqual(msle_obj2.name, 'my_msle')
self.assertEqual(msle_obj2._dtype, dtypes.int32)
def test_unweighted(self):
msle_obj = metrics.MeanSquaredLogarithmicError()
self.evaluate(variables.variables_initializer(msle_obj.variables))
y_true = constant_op.constant(((0, 1, 0, 1, 0), (0, 0, 1, 1, 1),
(1, 1, 1, 1, 0), (0, 0, 0, 0, 1)))
y_pred = constant_op.constant(((0, 0, 1, 1, 0), (1, 1, 1, 1, 1),
(0, 1, 0, 1, 0), (1, 1, 1, 1, 1)))
update_op = msle_obj.update_state(y_true, y_pred)
self.evaluate(update_op)
result = msle_obj.result()
self.assertAllClose(0.24022, result, atol=1e-5)