Models

systematica.portfolio.composers.models

Composer

Composer(
    model_name: str,
    signal_params=_Nothing.NOTHING,
    portfolio_params: Dict[str, Any] = _Nothing.NOTHING,
    model_params: Dict[str, Any] = _Nothing.NOTHING,
)

Run single pipeline from trial.

A composer that executes a single pipeline using the model class.

Method generated by attrs for class BaseComposer.

Ancestors

• systematica.portfolio.composers.base.BaseComposer

Methods

run

run(
    self,
    data: vectorbtpro.data.base.Data,
    s1: str,
    s2: str,
    use_close: bool = True,
) ‑> systematica.portfolio.analyzers.base.PortfolioAnalyzer

Execute the pipeline with the given data and symbols.

Parameters:

Name	Type	Default	Description
data	vbt.Data	--	Input data for the pipeline.
s1	str	--	First symbol.
s2	str	--	Second symbol.
use_close	bool	True	Whether to use ‘Close’ prices. Defaults to True.

Returns:

Type	Description
PortfolioAnalyzer	Portfolio analyzer object with the pipeline results.

SignalComposer

SignalComposer(
    model_name: str,
    signal_params=_Nothing.NOTHING,
    portfolio_params: Dict[str, Any] = _Nothing.NOTHING,
    model_params: Dict[str, Any] = _Nothing.NOTHING,
)

Combine trial strategies via concat signals.

A composer that combines multiple trial strategies by concatenating their signals.

Method generated by attrs for class BaseComposer.

Ancestors

• systematica.portfolio.composers.base.BaseComposer

Methods

run

run(
    self,
    data: vectorbtpro.data.base.Data,
    s1: str,
    s2: str,
    use_close: bool = True,
) ‑> systematica.portfolio.analyzers.base.PortfolioAnalyzer

Execute and combine multiple strategies by concatenating their signals.

Parameters:

Name	Type	Default	Description
data	vbt.Data	--	Input data for the pipeline.
s1	str	--	First symbol.
s2	str	--	Second symbol.
use_close	bool	True	Whether to use ‘Close’ prices (default is True).

Returns:

Type	Description
PortfolioAnalyzer	Portfolio analyzer object with combined strategy results.

ColumnStackComposer

ColumnStackComposer(
    model_name: str,
    signal_params=_Nothing.NOTHING,
    portfolio_params: Dict[str, Any] = _Nothing.NOTHING,
    model_params: Dict[str, Any] = _Nothing.NOTHING,
)

Combine trial strategies via colum_stack method.

A composer that combines multiple trial strategies using the column_stack method from vectorbtpro.

Method generated by attrs for class BaseComposer.

Ancestors

• systematica.portfolio.composers.base.BaseComposer

Methods

run

run(
    self,
    data: vectorbtpro.data.base.Data,
    s1: str,
    s2: str,
    group_by: bool = True,
    use_close: bool = True,
) ‑> systematica.portfolio.analyzers.base.PortfolioAnalyzer

Execute and combine multiple strategies using column stacking.

Parameters:

Name	Type	Default	Description
data	vbt.Data	--	Input data for the pipeline.
s1	str	--	First symbol.
s2	str	--	Second symbol.
group_by	bool	True	Whether to group the results, by default True.
use_close	bool	True	Whether to use ‘Close’ prices (default is True).

Returns:

Type	Description
PortfolioAnalyzer	Portfolio analyzer object with column-stacked strategy results.

PortfolioComposer

PortfolioComposer(
    feature_config: Dict[str, Any],
    model_params: Dict[str, Any],
)

Compose portfolio(s) from trials.

Method generated by attrs for class PortfolioComposer.

Class variables

• feature_config: Dict[str, Any]: Configuration for a feature calculation.

• model_params: Dict[str, Any]: A dictionary of optional keyword arguments needed to compose the portfolio.

Instance variables

• all_params: Dict[str, Any]: All parameters.

Methods

get_data

get_data(
    self,
    loader: vectorbtpro.data.base.Data | Callable = None,
    loader_kwargs: Dict[str, Any] = None,
) ‑> vectorbtpro.data.base.Data

Get data.

Parameters:

Name	Type	Default	Description
loader	vbt.Data	--	| tp.Callable, default None Data loader function or object. If None, uses load_clean_data.
loader_kwargs	tp.Kwargs	None	Additional keyword arguments for the loader function. If None, uses config.LOADER_PARAMS.

Returns:

Type	Description
tp.Kwargs	Data params.

get_compose_params

get_compose_params(
    self,
) ‑> Dict[str, Any]

Get compose parameters.

Returns:

Type	Description
tp.Kwargs	Compose params.

get_symbol_params

get_symbol_params(
    self,
) ‑> Dict[str, Any]

Get symbol params.

Returns:

Type	Description
tp.Kwargs	Symbol params..

run

run(
    self,
    column_stack: bool = True,
    group_by: bool = True,
    loader: vectorbtpro.data.base.Data | Callable = None,
    loader_kwargs: Dict[str, Any] = None,
) ‑> systematica.portfolio.analyzers.base.PortfolioAnalyzer

Execute composers.

THe composer strategy is selected based on model_params.

Parameters:

Name	Type	Default	Description
column_stack	bool	True	Whether to use column stacking for the results.
group_by	bool	True	Whether to group the results by the trial keys.
loader	vbt.Data	--	| tp.Callable, default None Data loader function or object. If None, uses load_clean_data.
loader_kwargs	tp.Kwargs	None	Additional keyword arguments for the loader function. If None, uses config.LOADER_PARAMS.

Returns:

Type	Description
PortfolioAnalyzer	Portfolio analyzer object with column-stacked strategy results.