Timesfm200mHyperparams

timesfm_200m_model.Timesfm200mHyperparams · inherits PydanticBaseModel

(No description provided)

__init__(self, ...)

(Pydantic default initialization)

Parameter	Type	Default	Description
...	...	...	...

Timesfm200mModel

timesfm_200m_model.Timesfm200mModel · inherits BaseModel

(No description provided)

__init__(self, params, settings)

Initialize TimesFM model.

Parameter	Type	Default	Description
params	Dict[str, Any]	-	Model parameters dictionary
settings	Dict[str, Any]	-	Settings dictionary containing device, python_version, etc.

train(self, y_context, y_target, timestamps_context, timestamps_target, x_context=None, x_target=None, **kwargs) -> "Timesfm200mModel"

Foundation model: no training needed. Mark as fitted and return self.

Parameters:

Parameter	Type	Default	Description
y_context	np.ndarray	-	(undocumented)
y_target	np.ndarray	-	(undocumented)
timestamps_context	np.ndarray	-	(undocumented)
timestamps_target	np.ndarray	-	(undocumented)
x_context	Optional[np.ndarray]	None	(undocumented)
x_target	Optional[np.ndarray]	None	(undocumented)
**kwargs		-	(undocumented)

Returns: Self (the fitted forecaster; sets self.is_fitted = True).

predict(self, y_context, timestamps_context, timestamps_target, x_context=None, x_target=None, **kwargs)

Make predictions using the trained TimesFM model.

Uses forecast_with_covariates when covariates are provided. Supports past-only, future-only, or both. TimesFM requires full (context+horizon) coverage; missing parts are padded: past-only pads future with last observed value, future-only pads past with zeros (heuristic; not guaranteed optimal).

Parameters:

Parameter	Type	Default	Description
y_context	np.ndarray	-	(undocumented)
timestamps_context	np.ndarray	-	(undocumented)
timestamps_target	np.ndarray	-	(undocumented)
x_context	Optional[np.ndarray]	None	(undocumented)
x_target	Optional[np.ndarray]	None	(undocumented)
**kwargs		-	(undocumented)

Returns: np.ndarray (Predictions array of shape (horizon, num_targets)). Raises: ValueError