pub trait Module<T: Scalar, B: ComputeBackend + Default = MoiraiBackend> {
// Required methods
fn parameters(&self) -> Vec<Var<T, B>>;
fn forward(
&self,
input: &Var<T, B>,
) -> Result<Var<T, B>, ModuleError<B::Error>>;
// Provided methods
fn named_parameters(&self) -> Vec<Parameter<T, B>> { ... }
fn load_parameters(&mut self, _params: &[Var<T, B>]) { ... }
fn load_named_parameters(
&mut self,
parameters: &[Parameter<T, B>],
) -> Result<(), ParameterLoadError> { ... }
fn zero_grad(&self) -> Result<(), B::Error> { ... }
fn train(&mut self, _mode: bool) { ... }
fn eval(&mut self) { ... }
}Expand description
Trait for neural network modules.
Each module owns Parameters, can contain sub-modules, and can support train/eval modes.
Required Methods§
Sourcefn parameters(&self) -> Vec<Var<T, B>>
fn parameters(&self) -> Vec<Var<T, B>>
Collect all trainable parameters (including from sub-modules).
Provided Methods§
Sourcefn named_parameters(&self) -> Vec<Parameter<T, B>>
fn named_parameters(&self) -> Vec<Parameter<T, B>>
Collect trainable parameters with stable hierarchical names.
Leaf modules with the canonical zero-, one-, or two-parameter layout
inherit [], [weight], or [weight, bias]. Modules with a wider
layout or child modules must override this method so names express the
owned field hierarchy instead of depending on flattened ordinals.
Sourcefn load_parameters(&mut self, _params: &[Var<T, B>])
fn load_parameters(&mut self, _params: &[Var<T, B>])
Write optimizer-updated parameter values back into this module’s own
fields, consuming params in the same order parameters() enumerates
them (i.e. params.len() == parameters().len()).
This exists because Var::clone() shares gradient state but not the
value storage: coeus_optim::{SGD, Adam}::step mutates its own owned
Vec<Var<T, B>> in place (via copy-on-write on first mutation), so a
module that read parameters() to build an optimizer must call this
after each step() to see the updated values, unless the module’s own
fields are the optimizer’s Vars (as in a flat training loop that
never separately clones them into a named struct).
Default is a no-op, which is correct for parameterless modules
(activations, dropout, pooling, reshape/view layers, …). Any module
whose parameters() returns a non-empty Vec must override this.
Sourcefn load_named_parameters(
&mut self,
parameters: &[Parameter<T, B>],
) -> Result<(), ParameterLoadError>
fn load_named_parameters( &mut self, parameters: &[Parameter<T, B>], ) -> Result<(), ParameterLoadError>
Load optimizer-updated named parameters after validating the complete stable path inventory.
§Errors
Returns ParameterLoadError when count or path ordering differs from
this module’s canonical named inventory.
Dyn Compatibility§
This trait is dyn compatible.
In older versions of Rust, dyn compatibility was called "object safety".
Implementors§
impl<ScalarType: Scalar, B: ComputeBackend + Default, H: Module<ScalarType, B>, T: Module<ScalarType, B>> Module<ScalarType, B> for StaticSeq<H, T>
impl<T, B, M> Module<T, B> for Bidirectional<M>where
T: Scalar,
B: BackendOps<T> + Default,
M: Module<T, B>,
B::DeviceBuffer<T>: CpuAddressableStorage<T> + CpuAddressableStorageMut<T>,
impl<T, B: BackendOps<T> + Default, const DIM: usize> Module<T, B> for ConvTranspose<T, B, DIM>where
T: RealScalar + Float,
impl<T, B> Module<T, B> for DepthwiseConv3d<T, B>where
T: Float + RealScalar,
B: BackendOps<T> + Default,
B::DeviceBuffer<T>: CpuAddressableStorage<T> + CpuAddressableStorageMut<T>,
impl<T: AttentionScalar + FloatElement + RealScalar, B: BackendOps<T> + AttentionOps<T> + Default, const H: usize, M: AttentionMask> Module<T, B> for MultiHeadAttention<T, B, H, M>
Self-attention Module impl (Q = K = V = input).
impl<T: AttentionScalar + FloatElement, B: BackendOps<T> + AttentionOps<T> + Default, M: AttentionMask> Module<T, B> for ScaledDotProductAttention<T, B, M>
Convenience Module impl — self-attention: Q = K = V = input.
impl<T: Float + FloatElement + Neg<Output = T> + RealScalar, B: BackendOps<T> + Default> Module<T, B> for LocalResponseNorm
impl<T: Float + FloatElement + RealScalar, B: BackendOps<T> + Default> Module<T, B> for AlphaDropout
impl<T: Float + FloatElement + RealScalar, B: BackendOps<T> + Default> Module<T, B> for FeatureAlphaDropout
impl<T: Float + FloatElement + RealScalar, B: BackendOps<T> + Default> Module<T, B> for GaussianNoise
impl<T: Float + FloatElement, B: BackendOps<T> + Default> Module<T, B> for Dropout
impl<T: Float + RealScalar, B: BackendOps<T> + Default> Module<T, B> for Bilinear<T, B>
impl<T: Float + RealScalar, B: BackendOps<T> + Default> Module<T, B> for FeedForward<T, B>
impl<T: Float + RealScalar, B: BackendOps<T> + Default> Module<T, B> for GLU
impl<T: Float + RealScalar, B: BackendOps<T> + Default> Module<T, B> for GRUCell<T, B>
impl<T: Float + RealScalar, B: BackendOps<T> + Default> Module<T, B> for Gru<T, B>
impl<T: Float + RealScalar, B: BackendOps<T> + Default> Module<T, B> for LSTMCell<T, B>
impl<T: Float + RealScalar, B: BackendOps<T> + Default> Module<T, B> for Lstm<T, B>
impl<T: Float + RealScalar, B: BackendOps<T> + Default> Module<T, B> for PReLU<T, B>
impl<T: Float + RealScalar, B: BackendOps<T> + Default> Module<T, B> for RNNCell<T, B>
impl<T: Float + RealScalar, B: BackendOps<T> + Default> Module<T, B> for Rnn<T, B>
impl<T: Float + RealScalar, B: BackendOps<T> + Default> Module<T, B> for Softmax
impl<T: Float + RealScalar, B: BackendOps<T> + Default> Module<T, B> for SwiGlu<T, B>
impl<T: Float + RealScalar, B: BackendOps<T> + Default> Module<T, B> for Tanhshrink
Implements the crate::module::Module interface for Tanhshrink.
impl<T: Float + RealScalar, B: ElementwiseOps<T> + ReductionOps<T> + Default> Module<T, B> for SinusoidalEncoding<T, B>
impl<T: Float + RealScalar, B> Module<T, B> for RotaryEmbedding<T, B>
impl<T: Float, B: BackendOps<T> + Default, D: ConvDim> Module<T, B> for Conv<T, B, D>
impl<T: Float, B: BackendOps<T> + Default> Module<T, B> for AvgPool1d<T, B>
impl<T: Float, B: BackendOps<T> + Default> Module<T, B> for AvgPool2d<T, B>
impl<T: Float, B: BackendOps<T> + Default> Module<T, B> for AvgPool3d<T, B>
impl<T: Float, B: BackendOps<T> + Default> Module<T, B> for Celu
impl<T: Float, B: BackendOps<T> + Default> Module<T, B> for ELU
impl<T: Float, B: BackendOps<T> + Default> Module<T, B> for GeLU
Implements the crate::module::Module interface for GeLU.
impl<T: Float, B: BackendOps<T> + Default> Module<T, B> for GeLUTanh
impl<T: Float, B: BackendOps<T> + Default> Module<T, B> for Hardshrink
impl<T: Float, B: BackendOps<T> + Default> Module<T, B> for Hardsigmoid
Implements the crate::module::Module interface for Hardsigmoid.
impl<T: Float, B: BackendOps<T> + Default> Module<T, B> for Hardswish
Implements the crate::module::Module interface for Hardswish.
impl<T: Float, B: BackendOps<T> + Default> Module<T, B> for Hardtanh
impl<T: Float, B: BackendOps<T> + Default> Module<T, B> for LeakyReLU
impl<T: Float, B: BackendOps<T> + Default> Module<T, B> for LogSigmoid
Implements the crate::module::Module interface for LogSigmoid.
impl<T: Float, B: BackendOps<T> + Default> Module<T, B> for MaxPool1d<T, B>
impl<T: Float, B: BackendOps<T> + Default> Module<T, B> for MaxPool2d<T, B>
impl<T: Float, B: BackendOps<T> + Default> Module<T, B> for MaxPool3d<T, B>
impl<T: Float, B: BackendOps<T> + Default> Module<T, B> for Mish
Implements the crate::module::Module interface for Mish.
impl<T: Float, B: BackendOps<T> + Default> Module<T, B> for SiLU
Implements the crate::module::Module interface for SiLU.
impl<T: Float, B: BackendOps<T> + Default> Module<T, B> for Sigmoid
Implements the crate::module::Module interface for Sigmoid.
impl<T: Float, B: BackendOps<T> + Default> Module<T, B> for Softplus
Implements the crate::module::Module interface for Softplus.
impl<T: Float, B: BackendOps<T> + Default> Module<T, B> for Softshrink
impl<T: Float, B: BackendOps<T> + Default> Module<T, B> for Softsign
Implements the crate::module::Module interface for Softsign.
impl<T: Float, B: BackendOps<T> + Default> Module<T, B> for Tanh
Implements the crate::module::Module interface for Tanh.
impl<T: Float, B: BackendOps<T> + Default> Module<T, B> for Threshold
impl<T: FloatElement + AttentionScalar + RealScalar, B: BackendOps<T> + AttentionOps<T> + Default, const H: usize, M: AttentionMask> Module<T, B> for TransformerEncoderLayer<T, B, H, M>
impl<T: FloatElement + AttentionScalar + RealScalar, B: BackendOps<T> + AttentionOps<T> + Default, const H: usize, const N: usize, M: AttentionMask> Module<T, B> for TransformerEncoder<T, B, H, N, M>
impl<T: FloatElement + Float + RealScalar, B: BackendOps<T> + Default, const DIM: usize> Module<T, B> for BatchNorm<T, B, DIM>
impl<T: FloatElement + Float + RealScalar, B: BackendOps<T> + Default, const DIM: usize> Module<T, B> for InstanceNorm<T, B, DIM>
Implements the Module interface for every InstanceNorm rank.
impl<T: FloatElement + Float + RealScalar, B: BackendOps<T> + Default, const G: usize> Module<T, B> for GroupNorm<T, B, G>
Implements the crate::module::Module interface for GroupNorm.
impl<T: FloatElement + Float + RealScalar, B: BackendOps<T> + Default> Module<T, B> for AdaptiveAvgPool1d<T, B>
impl<T: FloatElement + Float + RealScalar, B: BackendOps<T> + Default> Module<T, B> for AdaptiveAvgPool2d<T, B>
impl<T: FloatElement + Float + RealScalar, B: BackendOps<T> + Default> Module<T, B> for AdaptiveMaxPool1d<T, B>
impl<T: FloatElement + Float + RealScalar, B: BackendOps<T> + Default> Module<T, B> for AdaptiveMaxPool2d<T, B>
impl<T: FloatElement + Float + RealScalar, B: BackendOps<T> + Default> Module<T, B> for EmbeddingBag<T, B>
impl<T: FloatElement + Float + RealScalar, B: BackendOps<T> + Default> Module<T, B> for GlobalAvgPool1d<T, B>
impl<T: FloatElement + Float + RealScalar, B: BackendOps<T> + Default> Module<T, B> for GlobalAvgPool2d<T, B>
impl<T: FloatElement + Float + RealScalar, B: BackendOps<T> + Default> Module<T, B> for GlobalAvgPool3d<T, B>
impl<T: FloatElement + Float + RealScalar, B: BackendOps<T> + Default> Module<T, B> for GlobalMaxPool2d<T, B>
impl<T: FloatElement + Float + RealScalar, B: BackendOps<T> + Default> Module<T, B> for GlobalMaxPool3d<T, B>
impl<T: FloatElement + Float + RealScalar, B: BackendOps<T> + Default> Module<T, B> for LayerNorm<T, B>
Implements the crate::module::Module interface for LayerNorm.
impl<T: FloatElement + Float + RealScalar, B: BackendOps<T> + Default> Module<T, B> for RMSNorm<T, B>
Implements the crate::module::Module interface for RMSNorm.
impl<T: Scalar + RealScalar, B: BackendOps<T> + Default> Module<T, B> for Linear<T, B>
impl<T: Scalar, B: BackendOps<T> + Default> Module<T, B> for Embedding<T, B>
impl<T: Scalar, B: BackendOps<T> + Default> Module<T, B> for Fold1d<T, B>
impl<T: Scalar, B: BackendOps<T> + Default> Module<T, B> for Fold2d<T, B>
impl<T: Scalar, B: BackendOps<T> + Default> Module<T, B> for ReLU
Implements the crate::module::Module interface for ReLU.