Type ARRegressor
Namespace tensorflow.contrib.timeseries
Parent TimeSeriesRegressor
Interfaces IARRegressor
An Estimator for an (optionally non-linear) autoregressive model. ARRegressor is a window-based model, inputting fixed windows of length
`input_window_size` and outputting fixed windows of length
`output_window_size`. These two parameters must add up to the window_size
passed to the `Chunker` used to create an `input_fn` for training or
evaluation. `RandomWindowInputFn` is suggested for both training and
evaluation, although it may be seeded for deterministic evaluation.
Methods
Properties
Public static methods
ARRegressor NewDyn(object periodicities, object input_window_size, object output_window_size, object num_features, object exogenous_feature_columns, ImplicitContainer<T> num_time_buckets, ImplicitContainer<T> loss, object hidden_layer_sizes, object anomaly_prior_probability, object anomaly_distribution, object optimizer, object model_dir, object config)
Initialize the Estimator.
Parameters
-
object
periodicities - periodicities of the input data, in the same units as the time feature. Note this can be a single value or a list of values for multiple periodicities.
-
object
input_window_size - Number of past time steps of data to look at when doing the regression.
-
object
output_window_size - Number of future time steps to predict. Note that setting it to > 1 empirically seems to give a better fit.
-
object
num_features - The dimensionality of the time series (one for univariate, more than one for multivariate).
-
object
exogenous_feature_columns - A list of
tf.feature_column
s (for exampletf.feature_column.embedding_column
) corresponding to exogenous features which provide extra information to the model but are not part of the series to be predicted. Passed to `tf.compat.v1.feature_column.input_layer`. -
ImplicitContainer<T>
num_time_buckets - Number of buckets into which to divide (time % periodicity) for generating time based features.
-
ImplicitContainer<T>
loss - Loss function to use for training. Currently supported values are SQUARED_LOSS and NORMAL_LIKELIHOOD_LOSS. Note that for NORMAL_LIKELIHOOD_LOSS, we train the covariance term as well. For SQUARED_LOSS, the evaluation loss is reported based on un-scaled observations and predictions, while the training loss is computed on normalized data.
-
object
hidden_layer_sizes - list of sizes of hidden layers.
-
object
anomaly_prior_probability - If specified, constructs a mixture model under which anomalies (modeled with `anomaly_distribution`) have this prior probability. See `AnomalyMixtureARModel`.
-
object
anomaly_distribution - May not be specified unless anomaly_prior_probability is specified and is not None. Controls the distribution of anomalies under the mixture model. Currently either `ar_model.AnomalyMixtureARModel.GAUSSIAN_ANOMALY` or `ar_model.AnomalyMixtureARModel.CAUCHY_ANOMALY`. See `AnomalyMixtureARModel`. Defaults to `GAUSSIAN_ANOMALY`.
-
object
optimizer - The optimization algorithm to use when training, inheriting from tf.train.Optimizer. Defaults to Adagrad with step size 0.1.
-
object
model_dir - See `Estimator`.
-
object
config - See `Estimator`.