gluonts.model.deepar package

class gluonts.model.deepar.DeepAREstimator(freq: str, prediction_length: int, trainer: =, batch_size=None, callbacks=None, clip_gradient=10.0, ctx=None, epochs=100, hybridize=True, init="xavier", learning_rate=0.001, learning_rate_decay_factor=0.5, minimum_learning_rate=5e-05, num_batches_per_epoch=50, patience=10, weight_decay=1e-08), context_length: Optional[int] = None, num_layers: int = 2, num_cells: int = 40, cell_type: str = 'lstm', dropoutcell_type: str = 'ZoneoutCell', dropout_rate: float = 0.1, use_feat_dynamic_real: bool = False, use_feat_static_cat: bool = False, use_feat_static_real: bool = False, cardinality: Optional[List[int]] = None, embedding_dimension: Optional[List[int]] = None, distr_output: =, scaling: bool = True, lags_seq: Optional[List[int]] = None, time_features: Optional[List[gluonts.time_feature._base.TimeFeature]] = None, num_parallel_samples: int = 100, imputation_method: Optional[gluonts.transform.feature.MissingValueImputation] = None, train_sampler: Optional[gluonts.transform.sampler.InstanceSampler] = None, validation_sampler: Optional[gluonts.transform.sampler.InstanceSampler] = None, dtype: gluonts.core.component.DType = <class 'numpy.float32'>, alpha: float = 0.0, beta: float = 0.0, batch_size: int = 32, default_scale: Optional[float] = None, minimum_scale: float = 1e-10, impute_missing_values: bool = False, num_imputation_samples: int = 1)[source]


Construct a DeepAR estimator.

This implements an RNN-based model, close to the one described in [SFG17].

Note: the code of this model is unrelated to the implementation behind SageMaker’s DeepAR Forecasting Algorithm.

  • freq – Frequency of the data to train on and predict

  • prediction_length – Length of the prediction horizon

  • trainer – Trainer object to be used (default: Trainer())

  • context_length – Number of steps to unroll the RNN for before computing predictions (default: None, in which case context_length = prediction_length)

  • num_layers – Number of RNN layers (default: 2)

  • num_cells – Number of RNN cells for each layer (default: 40)

  • cell_type – Type of recurrent cells to use (available: ‘lstm’ or ‘gru’; default: ‘lstm’)

  • dropoutcell_type – Type of dropout cells to use (available: ‘ZoneoutCell’, ‘RNNZoneoutCell’, ‘VariationalDropoutCell’ or ‘VariationalZoneoutCell’; default: ‘ZoneoutCell’)

  • dropout_rate – Dropout regularization parameter (default: 0.1)

  • use_feat_dynamic_real – Whether to use the feat_dynamic_real field from the data (default: False)

  • use_feat_static_cat – Whether to use the feat_static_cat field from the data (default: False)

  • use_feat_static_real – Whether to use the feat_static_real field from the data (default: False)

  • cardinality – Number of values of each categorical feature. This must be set if use_feat_static_cat == True (default: None)

  • embedding_dimension – Dimension of the embeddings for categorical features (default: [min(50, (cat+1)//2) for cat in cardinality])

  • distr_output – Distribution to use to evaluate observations and sample predictions (default: StudentTOutput())

  • scaling – Whether to automatically scale the target values (default: true)

  • lags_seq – Indices of the lagged target values to use as inputs of the RNN (default: None, in which case these are automatically determined based on freq)

  • time_features – Time features to use as inputs of the RNN (default: None, in which case these are automatically determined based on freq)

  • num_parallel_samples – Number of evaluation samples per time series to increase parallelism during inference. This is a model optimization that does not affect the accuracy (default: 100)

  • imputation_method – One of the methods from ImputationStrategy

  • train_sampler – Controls the sampling of windows during training.

  • validation_sampler – Controls the sampling of windows during validation.

  • alpha – The scaling coefficient of the activation regularization

  • beta – The scaling coefficient of the temporal activation regularization

  • batch_size – The size of the batches to be used training and prediction.

  • minimum_scale – The minimum scale that is returned by the MeanScaler

  • default_scale – Default scale that is applied if the context length window is completely unobserved. If not set, the scale in this case will be the mean scale in the batch.

  • impute_missing_values – Whether to impute the missing values during training by using the current model parameters. Recommended if the dataset contains many missing values. However, this is a lot slower than the default mode.

  • num_imputation_samples – How many samples to use to impute values when impute_missing_values=True

create_predictor(transformation: gluonts.transform._base.Transformation, trained_network: mxnet.gluon.block.HybridBlock) → gluonts.model.predictor.Predictor[source]

Create and return a predictor object.


A predictor wrapping a HybridBlock used for inference.

Return type


create_training_data_loader(data: gluonts.dataset.common.Dataset, **kwargs) → Iterable[Dict[str, Any]][source]
create_training_network() → gluonts.model.deepar._network.DeepARTrainingNetwork[source]

Create and return the network used for training (i.e., computing the loss).


The network that computes the loss given input data.

Return type


create_transformation() → gluonts.transform._base.Transformation[source]

Create and return the transformation needed for training and inference.


The transformation that will be applied entry-wise to datasets, at training and inference time.

Return type


create_validation_data_loader(data: gluonts.dataset.common.Dataset, **kwargs) → Iterable[Dict[str, Any]][source]
classmethod derive_auto_fields(train_iter)[source]
freq = None
lead_time = None
prediction_length = None