MODULE REFERENCE

encoder_base

anomx.data.processing.encoders.encoder_base

Encoder Base Classes

CovariatesIndexGenerator

CovariatesIndexGenerator(input_chunk_length: int | None = None, output_chunk_length: int | None = None, lags_covariates: list[int] | None = None)

Bases: ABC

Signature parameters
Parameter	Type	Default
`input_chunk_length`	`int \| None`	`None`
`output_chunk_length`	`int \| None`	`None`
`lags_covariates`	`list[int] \| None`	`None`

anomx.data.processing.encoders.encoder_base.CovariatesIndexGenerator generates a time index for covariates at training and inference / prediction time with methods generate_train_idx(), and generate_inference_idx(). Without user covariates, it generates the minimum required covariate times spans for the corresponding scenarios described below. With user covariates, it simply copies and returns the covariates time index.

It can be used:

in combination with local forecasting estimators, or in a model agnostic scenario: All parameters can be ignored. This scenario is only supported by anomx.data.processing.encoders.encoder_base.FutureCovariatesIndexGenerator.
in combination with lag-based regression estimators: Set input_chunk_length, output_chunk_length, and lags_covariates. input_chunk_length is the absolute value of the minimum target lag abs(min(lags)) used with the lag-based regression estimator. Set output_chunk_length, and lags_covariates with the identical values used at forecasting model creation. For the covariate lags, use lags_past_covariates for anomx.data.processing.encoders.encoder_base.PastCovariatesIndexGenerator, and lags_future_covariates for anomx.data.processing.encoders.encoder_base.FutureCovariatesIndexGenerator.
in combination with chunk-based neural forecasting estimators: Set input_chunk_length, and output_chunk_length with the identical values used at forecasting model creation.

Parameters

input_chunk_length: Optionally, the number of input target time steps per chunk. Only required in scenarios B, C. Corresponds to input_chunk_length for chunk-based neural forecasting estimators, or to the absolute minimum target lag value abs(min(lags)) for lag-based regression estimators.
output_chunk_length: Optionally, the number of output target time steps per chunk. Only required in scenarios B, and C. Corresponds to output_chunk_length for both chunk-based neural forecasting estimators, and lag-based regression estimators.
lags_covariates: Optionally, a list of integers giving the covariates lags used by lag-based forecasting estimators. Only required in scenario B. Corresponds to the lag values from lags_past_covariates for past covariates, and lags_future_covariates for future covariates.

Members

generate_train_idxmethod

# CovariatesIndexGenerator.generate_train_idx Source ↗

generate_train_idx(target: TimeSeries, covariates: TimeSeries | None = None) -> tuple[TimeIndex, pd.Timestamp]

Generates/extracts time index (or integer index) for covariates at model training time.

Parameters

target: The target TimeSeries used during training.
covariates: Optionally, the covariates used for training. If given, the returned time index is equal to the covariates time index. Else, the returned time index covers the minimum required covariate time span for training a specific forecasting model. These requirements are derived from parameters set at anomx.data.processing.encoders.encoder_base.CovariatesIndexGenerator creation.

Signature parameters
Parameter	Type	Default
`target`	`TimeSeries`	`required`
`covariates`	`TimeSeries \| None`	`None`

Returns tuple[TimeIndex, pd.Timestamp]

generate_inference_idxmethod

# CovariatesIndexGenerator.generate_inference_idx Source ↗

generate_inference_idx(n: int, target: TimeSeries, covariates: TimeSeries | None = None) -> tuple[TimeIndex, pd.Timestamp]

Generates/extracts time index (or integer index) for covariates at model inference / prediction time.

Parameters

n: The forecasting horizon.
target: The target TimeSeries used during training or passed to prediction as series.
covariates: Optionally, the covariates used for prediction. If given, the returned time index is equal to the covariates time index. Else, the returned time index covers the minimum required covariate time spans for performing inference / prediction with a specific forecasting model. These requirements are derived from parameters set at anomx.data.processing.encoders.encoder_base.CovariatesIndexGenerator creation.

Signature parameters
Parameter	Type	Default
`n`	`int`	`required`
`target`	`TimeSeries`	`required`
`covariates`	`TimeSeries \| None`	`None`

Returns tuple[TimeIndex, pd.Timestamp]

generate_train_inference_idxmethod

# CovariatesIndexGenerator.generate_train_inference_idx Source ↗

generate_train_inference_idx(n: int, target: TimeSeries, covariates: TimeSeries | None = None) -> tuple[TimeIndex, pd.Timestamp]

Generates/extracts time index (or integer index) for covariates for training and inference / prediction.

Parameters

n: The forecasting horizon.
target: The target TimeSeries used for training and inference / prediction as series.
covariates: Optionally, the covariates used for training and inference / prediction. If given, the returned time index is equal to the covariates time index. Else, the returned time index covers the minimum required covariate time spans for performing training and inference / prediction with a specific forecasting model. These requirements are derived from parameters set at anomx.data.processing.encoders.encoder_base.CovariatesIndexGenerator creation.

Signature parameters
Parameter	Type	Default
`n`	`int`	`required`
`target`	`TimeSeries`	`required`
`covariates`	`TimeSeries \| None`	`None`

Returns tuple[TimeIndex, pd.Timestamp]

base_component_nameproperty

# CovariatesIndexGenerator.base_component_name Source ↗

base_component_name: str

Returns the index generator base component name. - "pc": past covariates - "fc": future covariates

Returns str

classView source ↗

Encoder

Encoder()

Bases: ABC

Abstract class for all encoders

Members

encode_trainmethod

# Encoder.encode_train Source ↗

encode_train(target: TimeSeries, covariates: TimeSeries | None = None, merge_covariates: bool = True, **kwargs) -> TimeSeries

Each subclass must implement a method to encode the covariates index for training.

Parameters

target: The target TimeSeries used during training or passed to prediction as series.
covariates: Optionally, the past or future covariates used for training.
merge_covariates: Whether to merge the encoded TimeSeries with covariates.

Signature parameters
Parameter	Type	Default
`target`	`TimeSeries`	`required`
`covariates`	`TimeSeries \| None`	`None`
`merge_covariates`	`bool`	`True`
`**kwargs`	`—`	`—`

Returns TimeSeries

encode_inferencemethod

# Encoder.encode_inference Source ↗

encode_inference(n: int, target: TimeSeries, covariates: TimeSeries | None = None, merge_covariates: bool = True, **kwargs) -> TimeSeries

Each subclass must implement a method to encode the covariates index for prediction.

Parameters

n: The forecast horizon
target: The target TimeSeries used during training or passed to prediction as series
covariates: Optionally, the past or future covariates used for prediction.
merge_covariates: Whether to merge the encoded TimeSeries with covariates.

Signature parameters
Parameter	Type	Default
`n`	`int`	`required`
`target`	`TimeSeries`	`required`
`covariates`	`TimeSeries \| None`	`None`
`merge_covariates`	`bool`	`True`
`**kwargs`	`—`	`—`

Returns TimeSeries

encode_train_inferencemethod

# Encoder.encode_train_inference Source ↗

encode_train_inference(n: int, target: TimeSeries, covariates: TimeSeries | None = None, merge_covariates: bool = True, **kwargs) -> TimeSeries

Each subclass must implement a method to encode the covariates index for training and prediction.

Parameters

n: The forecast horizon
target: The target TimeSeries used during training and prediction.
covariates: Optionally, the past or future covariates used for training and prediction.
merge_covariates: Whether to merge the encoded TimeSeries with covariates.

Signature parameters
Parameter	Type	Default
`n`	`int`	`required`
`target`	`TimeSeries`	`required`
`covariates`	`TimeSeries \| None`	`None`
`merge_covariates`	`bool`	`True`
`**kwargs`	`—`	`—`

Returns TimeSeries

fit_calledproperty

# Encoder.fit_called Source ↗

fit_called: bool

Returns whether the Encoder object has been fitted.

Returns bool

requires_fitproperty

# Encoder.requires_fit Source ↗

requires_fit: bool

Whether the Encoder sub class must be fit with Encoder.encode_train() before inference with Encoder.encode_inference().

Returns bool

classView source ↗

FutureCovariatesIndexGenerator

FutureCovariatesIndexGenerator(input_chunk_length: int | None = None, output_chunk_length: int | None = None, lags_covariates: list[int] | None = None)

Bases: CovariatesIndexGenerator

Generates index for future covariates on train and inference datasets.

Signature parameters
Parameter	Type	Default
`input_chunk_length`	`int \| None`	`None`
`output_chunk_length`	`int \| None`	`None`
`lags_covariates`	`list[int] \| None`	`None`

It can be used:

in combination with local forecasting estimators, or in a model agnostic scenario: All parameters can be ignored. This scenario is only supported by anomx.data.processing.encoders.encoder_base.FutureCovariatesIndexGenerator.
in combination with lag-based regression estimators: Set input_chunk_length, output_chunk_length, and lags_covariates. input_chunk_length is the absolute value of the minimum target lag abs(min(lags)) used with the lag-based regression estimator. Set output_chunk_length, and lags_covariates with the identical values used at forecasting model creation. For the covariate lags, use lags_past_covariates for anomx.data.processing.encoders.encoder_base.PastCovariatesIndexGenerator, and lags_future_covariates for anomx.data.processing.encoders.encoder_base.FutureCovariatesIndexGenerator.
in combination with chunk-based neural forecasting estimators: Set input_chunk_length, and output_chunk_length with the identical values used at forecasting model creation.

Parameters

input_chunk_length: Optionally, the number of input target time steps per chunk. Only required in scenarios B, C. Corresponds to input_chunk_length for chunk-based neural forecasting estimators, or to the absolute minimum target lag value abs(min(lags)) for lag-based regression estimators.
output_chunk_length: Optionally, the number of output target time steps per chunk. Only required in scenarios B, and C. Corresponds to output_chunk_length for both chunk-based neural forecasting estimators, and lag-based regression estimators.
lags_covariates: Optionally, a list of integers giving the covariates lags used by lag-based forecasting estimators. Only required in scenario B. Corresponds to the lag values from lags_past_covariates for past covariates, and lags_future_covariates for future covariates.

Members

generate_train_idxmethod

# FutureCovariatesIndexGenerator.generate_train_idx Source ↗

generate_train_idx(target: TimeSeries, covariates: TimeSeries | None = None) -> tuple[TimeIndex, pd.Timestamp]

Signature parameters
Parameter	Type	Default
`target`	`TimeSeries`	`required`
`covariates`	`TimeSeries \| None`	`None`

Returns tuple[TimeIndex, pd.Timestamp]

generate_inference_idxmethod

# FutureCovariatesIndexGenerator.generate_inference_idx Source ↗

generate_inference_idx(n: int, target: TimeSeries, covariates: TimeSeries | None = None) -> tuple[TimeIndex, pd.Timestamp]

Signature parameters
Parameter	Type	Default
`n`	`int`	`required`
`target`	`TimeSeries`	`required`
`covariates`	`TimeSeries \| None`	`None`

Returns tuple[TimeIndex, pd.Timestamp]

base_component_nameproperty

# FutureCovariatesIndexGenerator.base_component_name Source ↗

base_component_name: str

Returns str

Inherited members

generate_train_inference_idxCovariatesIndexGenerator

classView source ↗

PastCovariatesIndexGenerator

PastCovariatesIndexGenerator(input_chunk_length: int | None = None, output_chunk_length: int | None = None, lags_covariates: list[int] | None = None)

Bases: CovariatesIndexGenerator

Generates index for past covariates on train and inference datasets

Signature parameters
Parameter	Type	Default
`input_chunk_length`	`int \| None`	`None`
`output_chunk_length`	`int \| None`	`None`
`lags_covariates`	`list[int] \| None`	`None`

It can be used:

in combination with local forecasting estimators, or in a model agnostic scenario: All parameters can be ignored. This scenario is only supported by anomx.data.processing.encoders.encoder_base.FutureCovariatesIndexGenerator.
in combination with lag-based regression estimators: Set input_chunk_length, output_chunk_length, and lags_covariates. input_chunk_length is the absolute value of the minimum target lag abs(min(lags)) used with the lag-based regression estimator. Set output_chunk_length, and lags_covariates with the identical values used at forecasting model creation. For the covariate lags, use lags_past_covariates for anomx.data.processing.encoders.encoder_base.PastCovariatesIndexGenerator, and lags_future_covariates for anomx.data.processing.encoders.encoder_base.FutureCovariatesIndexGenerator.
in combination with chunk-based neural forecasting estimators: Set input_chunk_length, and output_chunk_length with the identical values used at forecasting model creation.

Parameters

input_chunk_length: Optionally, the number of input target time steps per chunk. Only required in scenarios B, C. Corresponds to input_chunk_length for chunk-based neural forecasting estimators, or to the absolute minimum target lag value abs(min(lags)) for lag-based regression estimators.
output_chunk_length: Optionally, the number of output target time steps per chunk. Only required in scenarios B, and C. Corresponds to output_chunk_length for both chunk-based neural forecasting estimators, and lag-based regression estimators.
lags_covariates: Optionally, a list of integers giving the covariates lags used by lag-based forecasting estimators. Only required in scenario B. Corresponds to the lag values from lags_past_covariates for past covariates, and lags_future_covariates for future covariates.

Members

generate_train_idxmethod

# PastCovariatesIndexGenerator.generate_train_idx Source ↗

generate_train_idx(target: TimeSeries, covariates: TimeSeries | None = None) -> tuple[TimeIndex, pd.Timestamp]

Signature parameters
Parameter	Type	Default
`target`	`TimeSeries`	`required`
`covariates`	`TimeSeries \| None`	`None`

Returns tuple[TimeIndex, pd.Timestamp]

generate_inference_idxmethod

# PastCovariatesIndexGenerator.generate_inference_idx Source ↗

generate_inference_idx(n: int, target: TimeSeries, covariates: TimeSeries | None = None) -> tuple[TimeIndex, pd.Timestamp]

Signature parameters
Parameter	Type	Default
`n`	`int`	`required`
`target`	`TimeSeries`	`required`
`covariates`	`TimeSeries \| None`	`None`

Returns tuple[TimeIndex, pd.Timestamp]

base_component_nameproperty

# PastCovariatesIndexGenerator.base_component_name Source ↗

base_component_name: str

Returns str

Inherited members

generate_train_inference_idxCovariatesIndexGenerator

classView source ↗

SequentialEncoderTransformer

SequentialEncoderTransformer(transformer: FittableDataTransformer, transform_mask: list[bool])

SequentialEncoderTransformer applies transformation to the non-transformed encoded covariates output of SequentialEncoder.encode_train() and SequentialEncoder.encode_inference(). The transformer is fitted when transform() is called for the first time. This ensures proper transformation of train, validation and inference dataset covariates. User-supplied covariates are not transformed.

Signature parameters
Parameter	Type	Default
`transformer`	`FittableDataTransformer`	`required`
`transform_mask`	`list[bool]`	`required`

Parameters

transformer: A FittableDataTransformer object with a fit_transform() and transform() method.
transform_mask: A boolean 1-D mask specifying which of the input covariates to transform() must be transformed.

Members

transformmethod

# SequentialEncoderTransformer.transform Source ↗

transform(covariates: list[TimeSeries]) -> list[TimeSeries]

This method applies transformation to the non-transformed encoded covariates output of SequentialEncoder._encode_sequence() after being merged with user-defined covariates. The transformer is fitted when transform() is called for the first time. This ensures proper transformation of train, validation and inference dataset covariates. The masks ensure that no covariates are transformed that user explicitly supplied to estimator.fit() and estimator.predict()

Parameters

covariates: The non-transformed encoded covariates output of SequentialEncoder._encode_sequence() before merging with user-defined covariates.

Signature parameters
Parameter	Type	Default
`covariates`	`list[TimeSeries]`	`required`

Returns list[TimeSeries]

fit_calledproperty

# SequentialEncoderTransformer.fit_called Source ↗

fit_called: bool

Return whether the transformer has been fitted.

Returns bool

classView source ↗

SingleEncoder

SingleEncoder(index_generator: CovariatesIndexGenerator)

Bases: Encoder, ABC

SingleEncoder: Abstract class for single index encoders. Single encoders can be used to implement new encoding techniques. Each single encoder must implement an _encode() method that carries the encoding logic.

The _encode() method must take an index as input and generate a encoded single TimeSeries as output.

Signature parameters
Parameter	Type	Default
`index_generator`	`CovariatesIndexGenerator`	`required`

Single encoders take an index_generator to generate the required index for encoding past and future covariates. See anomx._series_utils.data.covariate_index_generators.py for the CovariatesIndexGenerator subclasses. For past covariates encoders, use a PastCovariatesIndexGenerator. For future covariates encoders use a FutureCovariatesIndexGenerator.

Parameters

index_generator: An instance of CovariatesIndexGenerator with methods generate_train_idx() and generate_inference_idx(). Used to generate the index for encoders.

Members

encode_trainmethod

# SingleEncoder.encode_train Source ↗

encode_train(target: TimeSeries, covariates: TimeSeries | None = None, merge_covariates: bool = True, **kwargs) -> TimeSeries

Returns encoded index for training.

Parameters

target: The target TimeSeries used during training or passed to prediction as series
covariates: Optionally, the covariates used for training: past covariates if self.index_generator is a PastCovariatesIndexGenerator, future covariates if self.index_generator is a FutureCovariatesIndexGenerator
merge_covariates: Whether to merge the encoded TimeSeries with covariates.

Signature parameters
Parameter	Type	Default
`target`	`TimeSeries`	`required`
`covariates`	`TimeSeries \| None`	`None`
`merge_covariates`	`bool`	`True`
`**kwargs`	`—`	`—`

Returns TimeSeries

encode_inferencemethod

# SingleEncoder.encode_inference Source ↗

encode_inference(n: int, target: TimeSeries, covariates: TimeSeries | None = None, merge_covariates: bool = True, **kwargs) -> TimeSeries

Returns encoded index for inference/prediction.

Parameters

n: The forecast horizon
target: The target TimeSeries used during training or passed to prediction as series
covariates: Optionally, the covariates used for prediction: past covariates if self.index_generator is a PastCovariatesIndexGenerator, future covariates if self.index_generator is a FutureCovariatesIndexGenerator
merge_covariates: Whether to merge the encoded TimeSeries with covariates.

Signature parameters
Parameter	Type	Default
`n`	`int`	`required`
`target`	`TimeSeries`	`required`
`covariates`	`TimeSeries \| None`	`None`
`merge_covariates`	`bool`	`True`
`**kwargs`	`—`	`—`

Returns TimeSeries

encode_train_inferencemethod

# SingleEncoder.encode_train_inference Source ↗

encode_train_inference(n: int, target: TimeSeries, covariates: TimeSeries | None = None, merge_covariates: bool = True, **kwargs) -> TimeSeries

Returns encoded index for inference/prediction.

Parameters

n: The forecast horizon
target: The target TimeSeries used during training and prediction.
covariates: Optionally, the covariates used for training and prediction: past covariates if self.index_generator is a PastCovariatesIndexGenerator, future covariates if self.index_generator is a FutureCovariatesIndexGenerator
merge_covariates: Whether to merge the encoded TimeSeries with covariates.

Signature parameters
Parameter	Type	Default
`n`	`int`	`required`
`target`	`TimeSeries`	`required`
`covariates`	`TimeSeries \| None`	`None`
`merge_covariates`	`bool`	`True`
`**kwargs`	`—`	`—`

Returns TimeSeries

accept_transformerproperty

# SingleEncoder.accept_transformer Source ↗

accept_transformer: list[bool]

Whether the SingleEncoder sub class accepts to be transformed.

Returns list[bool]

encoding_n_componentsproperty

# SingleEncoder.encoding_n_components Source ↗

encoding_n_components: int

The number of components in the SingleEncoder output.

Returns int

componentsproperty

# SingleEncoder.components Source ↗

components: pd.Index

Returns the encoded component names. Only available after Encoder.encode_train() or Encoder.encode_inference() have been called.

Returns pd.Index

base_component_nameproperty

# SingleEncoder.base_component_name Source ↗

base_component_name: str

Returns the base encoder base component name. The string follows the given format: "anomx_enc_{covariates_temp}_{encoder}_{attribute}", where the elements are:

covariates_temp: "pc" or "fc" for past, or future covariates respectively.
encoder: the SingleEncoder type used: * "cyc" (cyclic temporal encoder), * "dta" (datetime attribute encoder), * "pos" (positional integer index encoder), * "cus" (custom callable index encoder)
attribute: the attribute used for the underlying encoder. Some examples: * "month_sin", "month_cos" (for "cyc") * "month" (for "dta") * "relative" (for "pos") * "custom" (for "cus")

Returns str

Inherited members

fit_calledEncoder requires_fitEncoder