Struct SyncBatchNorm2d 

#[non_exhaustive]
pub struct SyncBatchNorm2d<T: Float> {
    pub num_features: usize,
    pub eps: f64,
    pub momentum: f64,
    pub affine: bool,
    pub weight: Option<Parameter<T>>,
    pub bias: Option<Parameter<T>>,
    /* private fields */
}

2-D synchronized batch normalization.

Same API surface as BatchNorm2d plus an Arc<dyn Backend> for cross-rank communication. When world_size == 1 (or no backend is provided), behaves exactly like a plain BatchNorm2d.

Marked #[non_exhaustive] so future configuration knobs (e.g., running-stats sync mode, momentum schedule, fused fwd/bwd flags) can be added without breaking external struct-literal construction. Construct via SyncBatchNorm2d::new / SyncBatchNorm2d::with_backend.

Fields (Non-exhaustive)

Non-exhaustive structs could have additional fields added in future. Therefore, non-exhaustive structs cannot be constructed in external crates using the traditional Struct { .. } syntax; cannot be matched against without a wildcard ..; and struct update syntax will not work.

num_features: usize

eps: f64

momentum: f64

affine: bool

weight: Option<Parameter<T>>

bias: Option<Parameter<T>>

Implementations

impl<T: Float> SyncBatchNorm2d<T>

pub fn new( num_features: usize, eps: f64, momentum: f64, affine: bool, ) -> FerrotorchResult<Self>

Create a new SyncBatchNorm2d with no backend (acts as a plain BatchNorm2d). Use with_backend to attach the process group.

pub fn with_backend(self, backend: Arc<dyn Backend>) -> Self

Attach a backend so the layer synchronizes its statistics across the distributed process group.

pub fn running_mean(&self) -> Vec<f64>

Snapshot of the current per-channel running mean.

pub fn running_var(&self) -> Vec<f64>

Snapshot of the current per-channel running variance.

pub fn num_batches_tracked(&self) -> usize

Number of training batches processed so far.

Trait Implementations

impl<T: Float> Debug for SyncBatchNorm2d<T>

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

Formats the value using the given formatter.

impl<T: Float> Module<T> for SyncBatchNorm2d<T>

fn forward(&self, input: &Tensor<T>) -> FerrotorchResult<Tensor<T>>

Forward pass. Takes input tensor, returns output tensor.

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

Iterate over all learnable parameters.

fn parameters_mut(&mut self) -> Vec<&mut Parameter<T>>

Iterate over all learnable parameters mutably.

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

Named parameters for state dict serialization.

fn train(&mut self)

Set training mode. Affects dropout, batchnorm, etc.

fn eval(&mut self)

Set evaluation mode.

fn is_training(&self) -> bool

Whether the module is in training mode.

fn to_device(&mut self, device: Device) -> Result<(), FerrotorchError>

Move all parameters and buffers to a device.

fn state_dict(&self) -> HashMap<String, Tensor<T>>

Export parameters and buffers as a state dict (torch parity).

fn buffers(&self) -> Vec<&Buffer<T>>

Iterate over all non-trainable buffers (e.g. running mean / variance in BatchNorm). Default returns empty — concrete modules with buffers override.

fn buffers_mut(&mut self) -> Vec<&mut Buffer<T>>

Mutable iteration over all buffers. Default returns empty.

fn named_buffers(&self) -> Vec<(String, &Buffer<T>)>

Named buffers (dot-separated paths for nested modules). Default returns empty.

fn as_any(&self) -> Option<&(dyn Any + 'static)>

Downcast hook for type-erased buffer-loader dispatch. (#984)

fn children(&self) -> Vec<&dyn Module<T>>

Direct child modules. Default returns empty (leaf module).

fn named_children(&self) -> Vec<(String, &dyn Module<T>)>

Direct child modules with their attribute names. Default returns empty.

fn modules(&self) -> Vec<&dyn Module<T>>

where Self: Sized,

All modules in this subtree, depth-first (self first, then each child’s descendants in order).

fn descendants_dyn(&self) -> Vec<&dyn Module<T>>

All strict descendants of self in depth-first order. Object-safe.

fn named_modules(&self) -> Vec<(String, &dyn Module<T>)>

where Self: Sized,

All modules in this subtree with dot-separated path names. The root is named ""; children paths are joined with ..

fn named_descendants_dyn(&self) -> Vec<(String, &dyn Module<T>)>

Strict descendants with dot-paths. Object-safe.

fn with_forward_hook( self, hook: Box<dyn Fn(&Tensor<T>, &Tensor<T>) + Sync + Send>, ) -> (HookedModule<Self, T>, HookHandle)

where Self: Sized,

Wrap this module in a HookedModule and register a forward hook. Returns the wrapper paired with a HookHandle that can be used to remove the hook later. The wrapper implements Module<T> itself, so it slots into any place the original module did. Mirrors torch.nn.Module.register_forward_hook.

fn with_forward_pre_hook( self, hook: Box<dyn Fn(&Tensor<T>) -> Result<Tensor<T>, FerrotorchError> + Sync + Send>, ) -> (HookedModule<Self, T>, HookHandle)

where Self: Sized,

Wrap this module in a HookedModule and register a forward pre-hook. See Self::with_forward_hook. Mirrors torch.nn.Module.register_forward_pre_hook.

fn with_backward_hook( self, hook: Box<dyn Fn(&Tensor<T>, &Tensor<T>) + Sync + Send>, ) -> (HookedModule<Self, T>, HookHandle)

where Self: Sized,

Wrap this module in a HookedModule and register a backward hook. See Self::with_forward_hook. Mirrors torch.nn.Module.register_backward_hook.

fn zero_grad(&self) -> Result<(), FerrotorchError>

Set the gradient of every parameter to None.

fn requires_grad_(&mut self, requires_grad: bool)

Toggle requires_grad on every parameter (freeze / unfreeze the module). Mirrors torch.nn.Module.requires_grad_.

fn apply_to_parameters(&mut self, f: &mut dyn FnMut(&mut Parameter<T>))

Apply a function to every parameter in this module. Mirrors torch.nn.Module.apply for the parameter case (true apply recurses over all submodules; the recursive form requires &mut dyn Module which conflicts with this trait’s &mut self borrow).

fn load_state_dict( &mut self, state: &HashMap<String, Tensor<T>>, strict: bool, ) -> Result<(), FerrotorchError>

Load parameters from a state dict.