Skip to contents

GluonTS DeepAR (Torch) Modeling Function (Bridge)

Usage

deepar_torch_fit_impl(
  x,
  y,
  freq,
  prediction_length,
  id,
  epochs = 5,
  context_length = NULL,
  num_layers = 2,
  hidden_size = 40,
  num_cells = 40,
  dropout_rate = 0.1,
  num_feat_dynamic_real = 0,
  num_feat_static_cat = 0,
  num_feat_static_real = 0,
  cardinality = NULL,
  embedding_dimension = NULL,
  distr_output = "default",
  scaling = TRUE,
  lags_seq = NULL,
  time_features = NULL,
  num_parallel_samples = 100,
  batch_size = 32,
  scale_by_id = FALSE,
  ...
)

Arguments

x

A dataframe of xreg (exogenous regressors)

y

A numeric vector of values to fit

freq

A pandas timeseries frequency such as "5min" for 5-minutes or "D" for daily. Refer to Pandas Offset Aliases.

prediction_length

Numeric value indicating the length of the prediction horizon

id

A quoted column name that tracks the GluonTS FieldName "item_id"

epochs

Number of epochs that the network will train (default: 5).

context_length

Number of steps to unroll the RNN for before computing predictions (default: NULL, in which case context_length = prediction_length)

num_layers

Number of RNN layers (default: 2)

hidden_size

Hidden units

num_cells

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

dropout_rate

Dropout regularization parameter (default: 0.1)

num_feat_dynamic_real

Number of dynamic numeric features

num_feat_static_cat

Number of static categorical features

num_feat_static_real

Number of static numeric features

cardinality

Number of values of each categorical feature. This must be set if use_feat_static_cat == TRUE (default: NULL)

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: NULL, 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)

batch_size

Number of examples in each batch (default: 32).

scale_by_id

Scales numeric data by id group using mean = 0, standard deviation = 1 transformation. (default: FALSE)

...

Parameters passed to pytorch_lightning.trainer.Trainer()