Struct Sequential 

pub struct Sequential { /* private fields */ }

A sequential container that chains modules together.

Modules are added in the order they should be executed. The forward pass executes each module in order, passing the output of one as the input to the next.

Example

let model = Sequential::new()
    .add(Linear::new(784, 256))
    .add(ReLU)
    .add(Linear::new(256, 10));

let output = model.forward(&input);

Implementations

impl Sequential

pub fn new() -> Self

Creates a new empty Sequential container.

pub fn add<M: Module + 'static>(self, module: M) -> Self

Adds a module with an auto-generated name.

pub fn add_named<M: Module + 'static>( self, name: impl Into<String>, module: M, ) -> Self

Adds a module with a specific name.

pub fn push<M: Module + 'static>(&mut self, module: M)

Pushes a module (non-builder pattern).

pub fn push_named<M: Module + 'static>( &mut self, name: impl Into<String>, module: M, )

Pushes a named module (non-builder pattern).

pub fn len(&self) -> usize

Returns the number of modules.

pub fn is_empty(&self) -> bool

Returns true if empty.

pub fn iter(&self) -> impl Iterator<Item = (&str, &dyn Module)>

Returns an iterator over named modules.

Trait Implementations

impl Default for Sequential

fn default() -> Self

Returns the “default value” for a type.

impl Module for Sequential

fn forward(&self, input: &Variable) -> Variable

Performs the forward pass.

fn parameters(&self) -> Vec<Parameter>

Returns all parameters of this module.

fn named_parameters(&self) -> HashMap<String, Parameter>

Returns named parameters of this module.

fn set_training(&mut self, training: bool)

Sets the training mode.

fn is_training(&self) -> bool

Returns whether the module is in training mode.

fn name(&self) -> &'static str

Returns the module name for debugging.

fn num_parameters(&self) -> usize

Returns the number of trainable parameters.

fn train(&mut self)

Sets the module to training mode.

fn eval(&mut self)

Sets the module to evaluation mode.

fn zero_grad(&self)

Zeros all gradients of parameters.