chronax.models.informer.informer

`Scaler`

informer_scaler.Scaler

Defines the interface for per-window scalers, exposing methods to calculate statistics, transform data, and inverse transform data.

`stats(self, x: jnp.ndarray, axis: int = 1) -> tuple[jnp.ndarray, jnp.ndarray]`

Calculates the shift and scale parameters from the input array x.

Parameter	Type	Default	Description
`x`	`jnp.ndarray`	-	Input data array.
`axis`	`int`	`1`	Axis along which statistics are calculated.

Returns: tuple[jnp.ndarray, jnp.ndarray] (shift, scale).

`transform(self, x: jnp.ndarray, shift: jnp.ndarray, scale: jnp.ndarray) -> jnp.ndarray`

Applies the scaling transformation.

Parameter	Type	Default	Description
`x`	`jnp.ndarray`	-	Input data array.
`shift`	`jnp.ndarray`	-	Shift parameter calculated by `stats`.
`scale`	`jnp.ndarray`	-	Scale parameter calculated by `stats`.

Returns: jnp.ndarray (transformed data).

`inverse(self, z: jnp.ndarray, shift: jnp.ndarray, scale: jnp.ndarray) -> jnp.ndarray`

Applies the inverse scaling transformation.

Parameter	Type	Default	Description
`z`	`jnp.ndarray`	-	Scaled data array.
`shift`	`jnp.ndarray`	-	Shift parameter calculated by `stats`.
`scale`	`jnp.ndarray`	-	Scale parameter calculated by `stats`.

Returns: jnp.ndarray (original scale data).

`IdentityScaler`

informer_scaler.IdentityScaler

No-op scaler (shift=0, scale=1). Matches neuralforecast's scaler_type='identity'.

`stats(self, x: jnp.ndarray, axis: int = 1) -> tuple[jnp.ndarray, jnp.ndarray]`

Calculates shift (zeros) and scale (ones) based on the input shape.

Parameter	Type	Default	Description
`x`	`jnp.ndarray`	-	(undocumented)
`axis`	`int`	`1`	(undocumented)

Returns: tuple[jnp.ndarray, jnp.ndarray] (shift, scale).

`transform(self, x: jnp.ndarray, shift: jnp.ndarray, scale: jnp.ndarray) -> jnp.ndarray`

Returns the input x unchanged.

Parameter	Type	Default	Description
`x`	`jnp.ndarray`	-	(undocumented)
`shift`	`jnp.ndarray`	-	(undocumented)
`scale`	`jnp.ndarray`	-	(undocumented)

Returns: jnp.ndarray (unscaled data x).

`inverse(self, z: jnp.ndarray, shift: jnp.ndarray, scale: jnp.ndarray) -> jnp.ndarray`

Returns the input z unchanged.

Parameter	Type	Default	Description
`z`	`jnp.ndarray`	-	(undocumented)
`shift`	`jnp.ndarray`	-	(undocumented)
`scale`	`jnp.ndarray`	-	(undocumented)

Returns: jnp.ndarray (unscaled data z).

`RobustScaler`

informer_scaler.RobustScaler

Median + MAD scaler with 0.6745*std fallback when MAD=0.

`stats(self, x: jnp.ndarray, axis: int = 1) -> tuple[jnp.ndarray, jnp.ndarray]`

Calculates median (shift) and MAD (scale), applying a standard deviation fallback if MAD is zero.

Parameter	Type	Default	Description
`x`	`jnp.ndarray`	-	(undocumented)
`axis`	`int`	`1`	(undocumented)

Returns: tuple[jnp.ndarray, jnp.ndarray] (median/shift, robust scale).

`transform(self, x: jnp.ndarray, shift: jnp.ndarray, scale: jnp.ndarray) -> jnp.ndarray`

Applies robust scaling: (x - shift) / scale.

Parameter	Type	Default	Description
`x`	`jnp.ndarray`	-	(undocumented)
`shift`	`jnp.ndarray`	-	(undocumented)
`scale`	`jnp.ndarray`	-	(undocumented)

Returns: jnp.ndarray (scaled data).

`inverse(self, z: jnp.ndarray, shift: jnp.ndarray, scale: jnp.ndarray) -> jnp.ndarray`

Applies inverse robust scaling: z * scale + shift.

Parameter	Type	Default	Description
`z`	`jnp.ndarray`	-	(undocumented)
`shift`	`jnp.ndarray`	-	(undocumented)
`scale`	`jnp.ndarray`	-	(undocumented)

Returns: jnp.ndarray (original scale data).

`resolve_scaler`

informer_scaler.resolve_scaler

Resolve a scaler from a name ('identity'/'robust') or a Scaler instance.

Parameter	Type	Default	Description
`scaler`	`str \\| Scaler`	-	Name ('identity'/'robust') or an existing Scaler instance.

Returns: Scaler (The resolved scaler instance). Raises: ValueError (If an unknown string name is provided).