Lines Matching refs:features
79 def _dnn_linear_combined_model_fn(features,
95 features: dict of `Tensor`.
125 or features has the wrong type.
127 if not isinstance(features, dict):
128 raise ValueError('features should be a dictionary of `Tensor`s. '
129 'Given type: {}'.format(type(features)))
158 values=tuple(six.itervalues(features)),
168 dnn_logits = dnn_logit_fn(features=features, mode=mode)
181 values=tuple(six.itervalues(features)),
186 linear_logits = logit_fn(features=features)
222 features=features,
281 Input of `train` and `evaluate` should have following features,
287 - if `column` is a `_WeightedCategoricalColumn`, two features: the first
289 name. Both features' `value` must be a `SparseTensor`.
344 used as a key to fetch weight tensor from the `features`. If it is a
388 def _model_fn(features, labels, mode, config):
391 features=features,
462 Input of `train` and `evaluate` should have following features,
468 - if `column` is a `_WeightedCategoricalColumn`, two features: the first
470 name. Both features' `value` must be a `SparseTensor`.
525 used as a key to fetch weight tensor from the `features`. If it is a
551 def _model_fn(features, labels, mode, config):
554 features=features,
