def _build_predictions(self, results, features, labels):
"""Creates the dictionary of predictions that is returned by the model."""
predictions = flatten_dict({'results': results})
# Add features and, if available, labels to predictions
predictions.update(flatten_dict({'features': features}))
if labels is not None:
predictions.update(flatten_dict({'labels': labels}))
if self._losses is not None:
predictions['losses'] = self._losses
return predictions
def _build_predictions(self, results, features, labels):
"""Creates the dictionary of predictions that is returned by the model."""
predictions = flatten_dict({'results': results})
# Add features and, if available, labels to predictions
predictions.update(flatten_dict({'features': features}))
if labels is not None:
predictions.update(flatten_dict({'labels': labels}))
if self._losses is not None:
predictions['losses'] = self._losses
return predictions
def _build_predictions(results, features, labels, losses=None):
"""Creates the dictionary of predictions that is returned by the model."""
predictions = {'results': results}
# Add features and, if available, labels to predictions
predictions.update(flatten_dict({'features': features['source_ids']
})) # TODO: source_ids ?
if labels is not None:
predictions.update(flatten_dict({'labels': labels}))
if losses is not None:
predictions[
'losses'] = losses # TODO: transpose_batch_time(losses)
return predictions