def __init__(
self,
cell_type: str = "LSTM",
hidden_size: int = 10,
rnn_layers: int = 2,
dropout: float = 0.1,
static_categoricals: List[str] = [],
static_reals: List[str] = [],
time_varying_categoricals_encoder: List[str] = [],
time_varying_categoricals_decoder: List[str] = [],
categorical_groups: Dict[str, List[str]] = {},
time_varying_reals_encoder: List[str] = [],
time_varying_reals_decoder: List[str] = [],
embedding_sizes: Dict[str, Tuple[int, int]] = {},
embedding_paddings: List[str] = [],
embedding_labels: Dict[str, np.ndarray] = {},
x_reals: List[str] = [],
x_categoricals: List[str] = [],
output_size: Union[int, List[int]] = 1,
target: Union[str, List[str]] = None,
target_lags: Dict[str, List[int]] = {},
loss: MultiHorizonMetric = None,
logging_metrics: nn.ModuleList = None,
**kwargs,
):
"""
Recurrent Network.
Simple LSTM or GRU layer followed by output layer
Args:
cell_type (str, optional): Recurrent cell type ["LSTM", "GRU"]. Defaults to "LSTM".
hidden_size (int, optional): hidden recurrent size - the most important hyperparameter along with
``rnn_layers``. Defaults to 10.
rnn_layers (int, optional): Number of RNN layers - important hyperparameter. Defaults to 2.
dropout (float, optional): Dropout in RNN layers. Defaults to 0.1.
static_categoricals: integer of positions of static categorical variables
static_reals: integer of positions of static continuous variables
time_varying_categoricals_encoder: integer of positions of categorical variables for encoder
time_varying_categoricals_decoder: integer of positions of categorical variables for decoder
time_varying_reals_encoder: integer of positions of continuous variables for encoder
time_varying_reals_decoder: integer of positions of continuous variables for decoder
categorical_groups: dictionary where values
are list of categorical variables that are forming together a new categorical
variable which is the key in the dictionary
x_reals: order of continuous variables in tensor passed to forward function
x_categoricals: order of categorical variables in tensor passed to forward function
embedding_sizes: dictionary mapping (string) indices to tuple of number of categorical classes and
embedding size
embedding_paddings: list of indices for embeddings which transform the zero's embedding to a zero vector
embedding_labels: dictionary mapping (string) indices to list of categorical labels
output_size (Union[int, List[int]], optional): number of outputs (e.g. number of quantiles for
QuantileLoss and one target or list of output sizes).
target (str, optional): Target variable or list of target variables. Defaults to None.
target_lags (Dict[str, Dict[str, int]]): dictionary of target names mapped to list of time steps by
which the variable should be lagged.
Lags can be useful to indicate seasonality to the models. If you know the seasonalit(ies) of your data,
add at least the target variables with the corresponding lags to improve performance.
Defaults to no lags, i.e. an empty dictionary.
loss (MultiHorizonMetric, optional): loss: loss function taking prediction and targets.
logging_metrics (nn.ModuleList, optional): Metrics to log during training.
Defaults to nn.ModuleList([SMAPE(), MAE(), RMSE(), MAPE(), MASE()]).
"""
if loss is None:
loss = MAE()
if logging_metrics is None:
logging_metrics = nn.ModuleList(
[SMAPE(), MAE(), RMSE(),
MAPE(), MASE()])
self.save_hyperparameters()
# store loss function separately as it is a module
super().__init__(loss=loss, logging_metrics=logging_metrics, **kwargs)
self.embeddings = MultiEmbedding(
embedding_sizes=embedding_sizes,
embedding_paddings=embedding_paddings,
categorical_groups=categorical_groups,
x_categoricals=x_categoricals,
)
lagged_target_names = [
l for lags in target_lags.values() for l in lags
]
assert set(self.encoder_variables) - set(
to_list(target)
) - set(lagged_target_names) == set(self.decoder_variables) - set(
lagged_target_names
), "Encoder and decoder variables have to be the same apart from target variable"
for targeti in to_list(target):
assert (
targeti in time_varying_reals_encoder
), f"target {targeti} has to be real" # todo: remove this restriction
assert (
isinstance(target, str) and isinstance(loss, MultiHorizonMetric)
) or (
isinstance(target, (list, tuple)) and isinstance(loss, MultiLoss)
and len(loss) == len(target)
), "number of targets should be equivalent to number of loss metrics"
rnn_class = get_rnn(cell_type)
cont_size = len(self.reals)
cat_size = sum(
[size[1] for size in self.hparams.embedding_sizes.values()])
input_size = cont_size + cat_size
self.rnn = rnn_class(
input_size=input_size,
hidden_size=self.hparams.hidden_size,
num_layers=self.hparams.rnn_layers,
dropout=self.hparams.dropout if self.hparams.rnn_layers > 1 else 0,
batch_first=True,
)
# add linear layers for argument projects
if isinstance(target, str): # single target
self.output_projector = nn.Linear(self.hparams.hidden_size,
self.hparams.output_size)
assert not isinstance(
self.loss, QuantileLoss
), "QuantileLoss does not work with recurrent network"
else: # multi target
self.output_projector = nn.ModuleList([
nn.Linear(self.hparams.hidden_size, size)
for size in self.hparams.output_size
])
for l in self.loss:
assert not isinstance(
l, QuantileLoss
), "QuantileLoss does not work with recurrent network"
def __init__(
self,
cell_type: str = "LSTM",
hidden_size: int = 10,
rnn_layers: int = 2,
dropout: float = 0.1,
static_categoricals: List[str] = [],
static_reals: List[str] = [],
time_varying_categoricals_encoder: List[str] = [],
time_varying_categoricals_decoder: List[str] = [],
categorical_groups: Dict[str, List[str]] = {},
time_varying_reals_encoder: List[str] = [],
time_varying_reals_decoder: List[str] = [],
embedding_sizes: Dict[str, Tuple[int, int]] = {},
embedding_paddings: List[str] = [],
embedding_labels: Dict[str, np.ndarray] = {},
x_reals: List[str] = [],
x_categoricals: List[str] = [],
n_validation_samples: int = None,
n_plotting_samples: int = None,
target: Union[str, List[str]] = None,
loss: DistributionLoss = None,
logging_metrics: nn.ModuleList = None,
**kwargs,
):
"""
DeepAR Network.
The code is based on the article `DeepAR: Probabilistic forecasting with autoregressive recurrent networks
<https://www.sciencedirect.com/science/article/pii/S0169207019301888>`_.
Args:
cell_type (str, optional): Recurrent cell type ["LSTM", "GRU"]. Defaults to "LSTM".
hidden_size (int, optional): hidden recurrent size - the most important hyperparameter along with
``rnn_layers``. Defaults to 10.
rnn_layers (int, optional): Number of RNN layers - important hyperparameter. Defaults to 2.
dropout (float, optional): Dropout in RNN layers. Defaults to 0.1.
static_categoricals: integer of positions of static categorical variables
static_reals: integer of positions of static continuous variables
time_varying_categoricals_encoder: integer of positions of categorical variables for encoder
time_varying_categoricals_decoder: integer of positions of categorical variables for decoder
time_varying_reals_encoder: integer of positions of continuous variables for encoder
time_varying_reals_decoder: integer of positions of continuous variables for decoder
categorical_groups: dictionary where values
are list of categorical variables that are forming together a new categorical
variable which is the key in the dictionary
x_reals: order of continuous variables in tensor passed to forward function
x_categoricals: order of categorical variables in tensor passed to forward function
embedding_sizes: dictionary mapping (string) indices to tuple of number of categorical classes and
embedding size
embedding_paddings: list of indices for embeddings which transform the zero's embedding to a zero vector
embedding_labels: dictionary mapping (string) indices to list of categorical labels
n_validation_samples (int, optional): Number of samples to use for calculating validation metrics.
Defaults to None, i.e. no sampling at validation stage and using "mean" of distribution for logging
metrics calculation.
n_plotting_samples (int, optional): Number of samples to generate for plotting predictions
during training. Defaults to ``n_validation_samples`` if not None or 100 otherwise.
target (str, optional): Target variable or list of target variables. Defaults to None.
loss (DistributionLoss, optional): Distribution loss function. Keep in mind that each distribution
loss function might have specific requirements for target normalization.
Defaults to :py:class:`~pytorch_forecasting.metrics.NormalDistributionLoss`.
logging_metrics (nn.ModuleList, optional): Metrics to log during training.
Defaults to nn.ModuleList([SMAPE(), MAE(), RMSE(), MAPE(), MASE()]).
"""
if loss is None:
loss = NormalDistributionLoss()
if logging_metrics is None:
logging_metrics = nn.ModuleList(
[SMAPE(), MAE(), RMSE(),
MAPE(), MASE()])
if n_plotting_samples is None:
if n_validation_samples is None:
n_plotting_samples = n_validation_samples
else:
n_plotting_samples = 100
self.save_hyperparameters()
# store loss function separately as it is a module
super().__init__(loss=loss, logging_metrics=logging_metrics, **kwargs)
self.embeddings = MultiEmbedding(
embedding_sizes=embedding_sizes,
embedding_paddings=embedding_paddings,
categorical_groups=categorical_groups,
x_categoricals=x_categoricals,
)
assert set(self.encoder_variables) - set(to_list(target)) == set(
self.decoder_variables
), "Encoder and decoder variables have to be the same apart from target variable"
for targeti in to_list(target):
assert (
targeti in time_varying_reals_encoder
), f"target {targeti} has to be real" # todo: remove this restriction
assert (
isinstance(target, str) and isinstance(loss, DistributionLoss)
) or (
isinstance(target,
(list, tuple)) and isinstance(loss, MultiLoss) and
len(loss)
== len(target)
), "number of targets should be equivalent to number of loss metrics"
time_series_rnn = get_cell(cell_type)
self.rnn = time_series_rnn(
input_size=self.input_size,
hidden_size=self.hparams.hidden_size,
num_layers=self.hparams.rnn_layers,
dropout=self.hparams.dropout if self.hparams.rnn_layers > 1 else 0,
batch_first=True,
)
# add linear layers for argument projects
if isinstance(loss, MultiLoss): # multi target
self.distribution_projector = nn.ModuleList([
nn.Linear(self.hparams.hidden_size, len(args))
for args in self.loss.distribution_arguments
])
else:
self.distribution_projector = nn.Linear(
self.hparams.hidden_size,
len(self.loss.distribution_arguments))
def __init__(
self,
stack_types: List[str] = ["trend", "seasonality"],
num_blocks=[3, 3],
num_block_layers=[3, 3],
widths=[32, 512],
sharing: List[int] = [True, True],
expansion_coefficient_lengths: List[int] = [3, 7],
prediction_length: int = 1,
context_length: int = 1,
dropout: float = 0.1,
learning_rate: float = 1e-2,
log_interval: int = -1,
log_gradient_flow: bool = False,
log_val_interval: int = None,
weight_decay: float = 1e-3,
loss: MultiHorizonMetric = None,
reduce_on_plateau_patience: int = 1000,
backcast_loss_ratio: float = 0.0,
logging_metrics: nn.ModuleList = None,
**kwargs,
):
"""
Initialize NBeats Model - use its :py:meth:`~from_dataset` method if possible.
Based on the article
`N-BEATS: Neural basis expansion analysis for interpretable time series
forecasting <http://arxiv.org/abs/1905.10437>`_. The network has (if used as ensemble) outperformed all
other methods
including ensembles of traditional statical methods in the M4 competition. The M4 competition is arguably
the most
important benchmark for univariate time series forecasting.
Args:
stack_types: One of the following values: “generic”, “seasonality" or “trend". A list of strings
of length 1 or ‘num_stacks’. Default and recommended value
for generic mode: [“generic”] Recommended value for interpretable mode: [“trend”,”seasonality”]
num_blocks: The number of blocks per stack. A list of ints of length 1 or ‘num_stacks’.
Default and recommended value for generic mode: [1] Recommended value for interpretable mode: [3]
num_block_layers: Number of fully connected layers with ReLu activation per block. A list of ints of length
1 or ‘num_stacks’.
Default and recommended value for generic mode: [4] Recommended value for interpretable mode: [4]
width: Widths of the fully connected layers with ReLu activation in the blocks.
A list of ints of length 1 or ‘num_stacks’. Default and recommended value for generic mode: [512]
Recommended value for interpretable mode: [256, 2048]
sharing: Whether the weights are shared with the other blocks per stack.
A list of ints of length 1 or ‘num_stacks’. Default and recommended value for generic mode: [False]
Recommended value for interpretable mode: [True]
expansion_coefficient_length: If the type is “G” (generic), then the length of the expansion
coefficient.
If type is “T” (trend), then it corresponds to the degree of the polynomial. If the type is “S”
(seasonal) then this is the minimum period allowed, e.g. 2 for changes every timestep.
A list of ints of length 1 or ‘num_stacks’. Default value for generic mode: [32] Recommended value for
interpretable mode: [3]
prediction_length: Length of the prediction. Also known as 'horizon'.
context_length: Number of time units that condition the predictions. Also known as 'lookback period'.
Should be between 1-10 times the prediction length.
backcast_loss_ratio: weight of backcast in comparison to forecast when calculating the loss.
A weight of 1.0 means that forecast and backcast loss is weighted the same (regardless of backcast and
forecast lengths). Defaults to 0.0, i.e. no weight.
loss: loss to optimize. Defaults to MASE().
log_gradient_flow: if to log gradient flow, this takes time and should be only done to diagnose training
failures
reduce_on_plateau_patience (int): patience after which learning rate is reduced by a factor of 10
logging_metrics (nn.ModuleList[MultiHorizonMetric]): list of metrics that are logged during training.
Defaults to nn.ModuleList([SMAPE(), MAE(), RMSE(), MAPE(), MASE()])
**kwargs: additional arguments to :py:class:`~BaseModel`.
"""
if logging_metrics is None:
logging_metrics = nn.ModuleList(
[SMAPE(), MAE(), RMSE(),
MAPE(), MASE()])
if loss is None:
loss = MASE()
self.save_hyperparameters()
super().__init__(loss=loss, logging_metrics=logging_metrics, **kwargs)
# setup stacks
self.net_blocks = nn.ModuleList()
for stack_id, stack_type in enumerate(stack_types):
for _ in range(num_blocks[stack_id]):
if stack_type == "generic":
net_block = NBEATSGenericBlock(
units=self.hparams.widths[stack_id],
thetas_dim=self.hparams.
expansion_coefficient_lengths[stack_id],
num_block_layers=self.hparams.
num_block_layers[stack_id],
backcast_length=context_length,
forecast_length=prediction_length,
dropout=self.hparams.dropout,
)
elif stack_type == "seasonality":
net_block = NBEATSSeasonalBlock(
units=self.hparams.widths[stack_id],
num_block_layers=self.hparams.
num_block_layers[stack_id],
backcast_length=context_length,
forecast_length=prediction_length,
min_period=self.hparams.
expansion_coefficient_lengths[stack_id],
dropout=self.hparams.dropout,
)
elif stack_type == "trend":
net_block = NBEATSTrendBlock(
units=self.hparams.widths[stack_id],
thetas_dim=self.hparams.
expansion_coefficient_lengths[stack_id],
num_block_layers=self.hparams.
num_block_layers[stack_id],
backcast_length=context_length,
forecast_length=prediction_length,
dropout=self.hparams.dropout,
)
else:
raise ValueError(f"Unknown stack type {stack_type}")
self.net_blocks.append(net_block)
def __init__(
self,
activation_class: str = "ReLU",
hidden_size: int = 300,
n_hidden_layers: int = 3,
dropout: float = 0.1,
norm: bool = True,
static_categoricals: List[str] = [],
static_reals: List[str] = [],
time_varying_categoricals_encoder: List[str] = [],
time_varying_categoricals_decoder: List[str] = [],
categorical_groups: Dict[str, List[str]] = {},
time_varying_reals_encoder: List[str] = [],
time_varying_reals_decoder: List[str] = [],
embedding_sizes: Dict[str, Tuple[int, int]] = {},
embedding_paddings: List[str] = [],
embedding_labels: Dict[str, np.ndarray] = {},
x_reals: List[str] = [],
x_categoricals: List[str] = [],
output_size: Union[int, List[int]] = 1,
target: Union[str, List[str]] = None,
loss: MultiHorizonMetric = None,
logging_metrics: nn.ModuleList = None,
**kwargs,
):
"""
Args:
activation_class (str, optional): PyTorch activation class. Defaults to "ReLU".
hidden_size (int, optional): hidden recurrent size - the most important hyperparameter along with
``n_hidden_layers``. Defaults to 10.
n_hidden_layers (int, optional): Number of hidden layers - important hyperparameter. Defaults to 2.
dropout (float, optional): Dropout. Defaults to 0.1.
norm (bool, optional): if to use normalization in the MLP. Defaults to True.
static_categoricals: integer of positions of static categorical variables
static_reals: integer of positions of static continuous variables
time_varying_categoricals_encoder: integer of positions of categorical variables for encoder
time_varying_categoricals_decoder: integer of positions of categorical variables for decoder
time_varying_reals_encoder: integer of positions of continuous variables for encoder
time_varying_reals_decoder: integer of positions of continuous variables for decoder
categorical_groups: dictionary where values
are list of categorical variables that are forming together a new categorical
variable which is the key in the dictionary
x_reals: order of continuous variables in tensor passed to forward function
x_categoricals: order of categorical variables in tensor passed to forward function
embedding_sizes: dictionary mapping (string) indices to tuple of number of categorical classes and
embedding size
embedding_paddings: list of indices for embeddings which transform the zero's embedding to a zero vector
embedding_labels: dictionary mapping (string) indices to list of categorical labels
output_size (Union[int, List[int]], optional): number of outputs (e.g. number of quantiles for
QuantileLoss and one target or list of output sizes).
target (str, optional): Target variable or list of target variables. Defaults to None.
loss (MultiHorizonMetric, optional): loss: loss function taking prediction and targets.
Defaults to QuantileLoss.
logging_metrics (nn.ModuleList, optional): Metrics to log during training.
Defaults to nn.ModuleList([SMAPE(), MAE(), RMSE(), MAPE(), MASE()]).
"""
if loss is None:
loss = QuantileLoss()
if logging_metrics is None:
logging_metrics = nn.ModuleList(
[SMAPE(), MAE(), RMSE(),
MAPE(), MASE()])
self.save_hyperparameters()
# store loss function separately as it is a module
super().__init__(loss=loss, logging_metrics=logging_metrics, **kwargs)
self.input_embeddings = MultiEmbedding(
embedding_sizes={
name: val
for name, val in embedding_sizes.items()
if name in self.decoder_variables + self.static_variables
},
embedding_paddings=embedding_paddings,
categorical_groups=categorical_groups,
x_categoricals=x_categoricals,
)
# define network
if isinstance(self.hparams.output_size, int):
mlp_output_size = self.hparams.output_size
else:
mlp_output_size = sum(self.hparams.output_size)
cont_size = len(self.decoder_reals_positions)
cat_size = sum(self.input_embeddings.output_size.values())
input_size = cont_size + cat_size
self.mlp = FullyConnectedModule(
dropout=dropout,
norm=self.hparams.norm,
activation_class=getattr(nn, self.hparams.activation_class),
input_size=input_size,
output_size=mlp_output_size,
hidden_size=self.hparams.hidden_size,
n_hidden_layers=self.hparams.n_hidden_layers,
)
