quantlab.core.config.FrozenConfig

class quantlab.core.config.FrozenConfig

Bases: object

Base of the frozen dataset, factor and model configs.

A list given to a tuple-typed field is stored as a tuple, so a config rebuilt from JSON, where every tuple was written as a list, equals the one that was saved.

Examples

>>> from quantlab.dataset.config import BaseDatasetConfig
>>> cfg = BaseDatasetConfig(zarr_file_path="stock.zarr", symbols=["AAPL"])
>>> cfg.symbols
('AAPL',)
>>> cfg.symbols = ("MSFT",)
Traceback (most recent call last):
dataclasses.FrozenInstanceError: cannot assign to field 'symbols'

to_dict()

Return the config as a plain dict, each component as its own config.

The fields declared with quantlab.core.component.component hold their components’ get_config(); see quantlab.core.component.config_to_dict.

Examples

>>> from quantlab.dataset.config import BaseDatasetConfig
>>> sorted(BaseDatasetConfig(zarr_file_path="stock.zarr").to_dict())
['end_date', 'kwargs', 'name', 'resample_freq', 'resample_how', 'start_date', 'symbols', 'zarr_file_path']