tensorlogic_trustformers::normalization_variants

Struct BatchNorm

pub struct BatchNorm {
    pub num_channels: usize,
    pub eps: f64,
    pub momentum: f64,
    pub gamma: ArrayD<f64>,
    pub beta: ArrayD<f64>,
    pub affine: bool,
    pub running_mean: ArrayD<f64>,
    pub running_var: ArrayD<f64>,
    pub training: bool,
    pub num_batches_tracked: u64,
}

Expand description

Batch Normalization: normalizes across the batch dimension.

Tracks running mean/variance for evaluation mode. Input shape: [batch, channels, ...spatial_dims...]

Fields§

§num_channels: usize

Number of channels (features).

§eps: f64

Small constant for numerical stability.

§momentum: f64

Momentum for running statistics update (EMA coefficient).

§gamma: ArrayD<f64>

Learnable scale parameter.

§beta: ArrayD<f64>

Learnable shift parameter.

§affine: bool

Whether to apply learnable affine transformation.

§running_mean: ArrayD<f64>

Running mean used during evaluation.

§running_var: ArrayD<f64>

Running variance used during evaluation.

§training: bool

Whether the module is in training mode.

§num_batches_tracked: u64

Number of mini-batches tracked.

Implementations§

Source §

impl BatchNorm

Source

pub fn new( num_channels: usize, eps: f64, momentum: f64, affine: bool, ) -> Result<Self, NormalizationError>

Create a new BatchNorm layer.

Source

pub fn forward( &mut self, input: &ArrayD<f64>, ) -> Result<ArrayD<f64>, NormalizationError>

Forward pass: normalize across batch (and spatial) dimensions per channel.

In training mode: compute batch statistics and update running stats via EMA. In eval mode: use running statistics.

Source

pub fn eval_mode(&mut self)

Switch to evaluation mode (use running statistics).

Source

pub fn train_mode(&mut self)

Switch to training mode (compute batch statistics).

Source

pub fn is_training(&self) -> bool

Check whether the module is in training mode.

Source

pub fn reset_running_stats(&mut self)

Reset running statistics to initial values.

Trait Implementations§

Source §

impl Clone for BatchNorm

Source §

fn clone(&self) -> BatchNorm

Returns a duplicate of the value. Read more

1.0.0 · Source§

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source. Read more

Source §

impl Debug for BatchNorm

Source §

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter. Read more

Auto Trait Implementations§

§

impl UnwindSafe for BatchNorm

Blanket Implementations§

Source §

impl<T> Any for T
where T: 'static + ?Sized,

Source §

fn type_id(&self) -> TypeId

Gets the TypeId of self. Read more

Source §

impl<T> Borrow<T> for T
where T: ?Sized,

Source §

fn borrow(&self) -> &T

Immutably borrows from an owned value. Read more

Source §

impl<T> BorrowMut<T> for T
where T: ?Sized,

Source §

fn borrow_mut(&mut self) -> &mut T

Mutably borrows from an owned value. Read more

Source §

impl<T> CloneToUninit for T
where T: Clone,

Source §

unsafe fn clone_to_uninit(&self, dest: *mut u8)

🔬This is a nightly-only experimental API. (clone_to_uninit)

Performs copy-assignment from self to dest. Read more

Source §

impl<T> From<T> for T

Source §

fn from(t: T) -> T

Returns the argument unchanged.

Source §

impl<T, U> Into for T
where U: From<T>,

Source §

fn into(self) -> U

Calls U::from(self).

That is, this conversion is whatever the implementation of From<T> for U chooses to do.

Source §

impl<T> IntoEither for T

Source §

fn into_either(self, into_left: bool) -> Either<Self, Self>

Converts self into a Left variant of Either<Self, Self> if into_left is true. Converts self into a Right variant of Either<Self, Self> otherwise. Read more

Source §

fn into_either_with<F>(self, into_left: F) -> Either<Self, Self>
where F: FnOnce(&Self) -> bool,

Converts self into a Left variant of Either<Self, Self> if into_left(&self) returns true. Converts self into a Right variant of Either<Self, Self> otherwise. Read more

Source §