Kairos50mModel
chronax.models.kairos_50m_model.Kairos50mModel · inherits BaseForecaster
Kairos foundation model wrapper with stochastic (quantile) output. Returns 9 quantile forecasts (0.1–0.9) converted to pseudo-samples for stochastic metrics. Processes each target independently.
__init__(self, params, settings)
Initializes the Kairos 50m model wrapper.
| Parameter | Type | Default | Description |
|---|---|---|---|
params |
Dict[str, Any] |
- | (undocumented) |
settings |
Dict[str, Any] |
- | (undocumented) |
train(self, y_context, y_target, timestamps_context, timestamps_target, x_context=None, x_target=None, **kwargs)
Trains the Kairos model (loads the pretrained weights and sets up the internal structure).
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)
Generates forecasts based on the context data. The model returns pseudo-samples derived from 9 quantile forecasts.
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 (Pseudo-samples derived from quantile forecasts, shape (num_samples, forecast_horizon, num_targets)).
Raises: ValueError if the model has not been fitted (train() not called).