def testToScalar(self):
self.assertEqual(1, metric_util.to_scalar(np.array([1])))
self.assertEqual(1.0, metric_util.to_scalar(np.array(1.0)))
self.assertEqual('string', metric_util.to_scalar(np.array([['string']])))
sparse_tensor = tf.compat.v1.SparseTensorValue(
indices=np.array([0]), values=np.array([1]), dense_shape=(1,))
self.assertEqual(1, metric_util.to_scalar(sparse_tensor))
def key_by_query_key(extracts: types.Extracts,
query_key: Text) -> Tuple[Text, types.Extracts]:
"""Extract the query key from the extract and key by that."""
value = metric_util.to_scalar(util.get_by_keys(
extracts, [constants.FEATURES_KEY, query_key], optional=True),
tensor_name=query_key)
if value is None:
missing_query_key_counter.inc()
return ('', extracts)
return ('{}'.format(value), extracts)
def _gain(self, i: metric_types.StandardMetricInputs) -> float:
gain = util.get_by_keys(i.features, [self._gain_key])
if gain.size == 1:
scalar = metric_util.to_scalar(gain)
if scalar is not None:
return scalar
raise ValueError('expected {} to be scalar, but instead it has size = {}: '
'value={}, metric_keys={}, '
'StandardMetricInputs={}'.format(self._gain_key, gain.size,
gain, self._metric_keys,
i))
def _query(
self, i: metric_types.StandardMetricInputs
) -> Optional[Union[float, int, Text]]:
return metric_util.to_scalar(
util.get_by_keys(i.features, [self._query_key]))
