Sequence

class vulkpy.nn.Sequence

Bases: object

Sequential Model

All layers sequentially connceted.

Methods Summary

predict(x[, y])	Predict Label/Target
train(x, y)	Train model

Methods Documentation

predict(x: Array, y: Array | None = None) → Array | Tuple[Array, Array]

Predict Label/Target

Parameters:

• x (vulkpy.Array) – Features

• y (vulkpy.Array, optional) – Labels/Targets.

Returns:

• pred_y (vulkpy.Array) – Predicted Labels/Targets

• loss (vulkpy.Array) – Loss. Return only if y is specified.

train(x: Array, y: Array) → Tuple[Array, Array]

Train model

Parameters:

• x (vulkpy.Array) – Features and Labels/Targets

• y (vulkpy.Array) – Features and Labels/Targets

Returns:

• y (vulkpy.Array) – Predicted Labels/Targets

• loss (vulkpy.Array) – Loss

__init__(layers: Iterable[Module], loss: Loss)

Initialize Sequence

Parameters:

• layers (iterable of vulkpy.nn.Module) – Layers to be called sequentially

• loss (vulkpy.nn.Loss) – Loss layer