build_windows(y, input_size, h)

Return [n_windows, input_size+h] rolling windows; step=1.

Parameter	Type	Default	Description
y	jnp.ndarray	-	(undocumented)
input_size	int	-	(undocumented)
h	int	-	(undocumented)

Returns: jnp.ndarray

forward_loss(model, windows, *, h, input_size, loss_fn=mae)

Forward + point loss in raw scale. windows: [B, input_size+h] -> scalar.

Parameter	Type	Default	Description
model	BiTCNNet	-	(undocumented)
windows	jnp.ndarray	-	(undocumented)
h	int	-	(undocumented)
input_size	int	-	(undocumented)
loss_fn	LossFn	mae	(undocumented)

Returns: jnp.ndarray

train(model, y, *, h, input_size, max_steps, windows_batch_size, lr, seed, loss_fn=mae)

Train model in place via a single nnx.scan. Returns per-step losses.

Window sampling replicates neuralforecast's regime-dependent scheme (_base_model.py training_step): when n_windows < windows_batch_size NF draws windows_batch_size indices WITH replacement; otherwise it takes a without-replacement permutation. The branch is chosen on concrete shapes (Python-static), so it bakes into the scan once. Sampling happens per step inside the scan to keep memory bounded to one batch.

Parameter	Type	Default	Description
model	BiTCNNet	-	(undocumented)
y	jnp.ndarray	-	(undocumented)
h	int	-	(undocumented)
input_size	int	-	(undocumented)
max_steps	int	-	(undocumented)
windows_batch_size	int	-	(undocumented)
lr	optax.ScalarOrSchedule	-	(undocumented)
seed	int	-	(undocumented)
loss_fn	LossFn	mae	(undocumented)

Returns: jnp.ndarray

predict_step(model, y, *, h, input_size)

Forecast next h steps from the final input_size of y. Returns (h,).

Parameter	Type	Default	Description
model	BiTCNNet	-	(undocumented)
y	jnp.ndarray	-	(undocumented)
h	int	-	(undocumented)
input_size	int	-	(undocumented)

Returns: jnp.ndarray