CheckpointMetadata

class openstef_foundation_models.models.checkpoint.CheckpointMetadata(**data: Any) → None

Bases: BaseConfig

Metadata describing a foundation-model checkpoint.

This document travels next to the weights file and drives the generic inference path, so that no model-family specifics are hardcoded in the backends or forecasters.

Parameters:

data (Any)

schema_version: int

model_family: str

input_names: list[str]

output_name: str

native_quantiles: list[Quantile]

context_length: int

output_patch_size: int

horizon_patches: int

resolution_minutes: int

precision: Literal['fp32', 'fp16', 'int8']

static_shapes: bool

max_covariates: int | None

property horizon_length: int

Total number of forecast timesteps the model emits.

Returns:

output_patch_size * horizon_patches.

model_config: ClassVar[ConfigDict] = {'arbitrary_types_allowed': False, 'extra': 'ignore', 'protected_namespaces': ()}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].