![]() |
Perform filtering using an exported saved model.
tf.contrib.timeseries.saved_model_utils.filter_continuation(
continue_from,
signatures,
session,
features
)
Filtering refers to updating model state based on new observations. Predictions based on the returned model state will be conditioned on these observations.
Args:
continue_from
: A dictionary containing the results of either an Estimator's evaluate method or a previous filter step (cold start or continuation). Used to determine the model state to start filtering from.signatures
: TheMetaGraphDef
protocol buffer returned fromtf.compat.v1.saved_model.loader.load
. Used to determine the names of Tensors to feed and fetch. Must be from the same model ascontinue_from
.session
: The session to use. The session's graph must be the one into whichtf.compat.v1.saved_model.loader.load
loaded the model.features
: A dictionary mapping keys to Numpy arrays, with several possible shapes (requires keysFilteringFeatures.TIMES
andFilteringFeatures.VALUES
): Single example;TIMES
is a scalar andVALUES
is either a scalar or a vector of length [number of features]. Sequence;TIMES
is a vector of shape [series length],VALUES
either has shape series length or series length x number of features. Batch of sequences;TIMES
is a vector of shape [batch size x series length],VALUES
has shape [batch size x series length] or [batch size x series length x number of features]. In any case,VALUES
and any exogenous features must have their shapes prefixed by the shape of the value corresponding to theTIMES
key.
Returns:
A dictionary containing model state updated to account for the observations
in features
.