mae

chronax.tft_losses.mae

Mean absolute error.

Parameter	Type	Default	Description
pred	jnp.ndarray	-	(undocumented)
target	jnp.ndarray	-	(undocumented)

Returns: jnp.ndarray (undocumented)

mse

chronax.tft_losses.mse

Mean squared error.

Parameter	Type	Default	Description
pred	jnp.ndarray	-	(undocumented)
target	jnp.ndarray	-	(undocumented)

Returns: jnp.ndarray (undocumented)

huber

chronax.tft_losses.huber

Huber loss, delta = 1.0.

Parameter	Type	Default	Description
pred	jnp.ndarray	-	(undocumented)
target	jnp.ndarray	-	(undocumented)

Returns: jnp.ndarray (undocumented)

MultiQuantileLoss

chronax.tft_losses.MultiQuantileLoss

Multi-quantile (pinball) loss. __call__(pred[...,h,Q], target[...,h]). QL(y, y_hat, q) = q*(y-y_hat)+ + (1-q)*(y_hat-y)+, averaged over quantiles and all elements. Quantiles are sorted, must lie in (0, 1), and must include 0.5 (the median / "mean" head). Picklable (holds a plain tuple).

__init__(self, quantiles: Sequence[float] = (0.1, 0.5, 0.9))

(undocumented)

Parameter	Type	Default	Description
quantiles	Sequence[float]	(0.1, 0.5, 0.9)	(undocumented)

__call__(self, pred: jnp.ndarray, target: jnp.ndarray) -> jnp.ndarray

(undocumented)

Parameter	Type	Default	Description
pred	jnp.ndarray	-	(undocumented)
target	jnp.ndarray	-	(undocumented)

Returns: jnp.ndarray (undocumented)

outputsize_multiplier

chronax.tft_losses.outputsize_multiplier

Output-head width implied by loss (1 for point losses).

Parameter	Type	Default	Description
loss	-	-	(undocumented)

Returns: int (undocumented)

resolve

chronax.tft_losses.resolve

Return a callable loss from a registry string, a callable, or an MQ instance.

Parameter	Type	Default	Description
loss	str | LossFn | MultiQuantileLoss	-	(undocumented)

Returns: (undocumented)