Struct dfdx::nn::modules::BatchNorm1D
source · pub struct BatchNorm1D<const C: usize, E: Dtype, D: DeviceStorage> {
pub scale: Tensor<Rank1<C>, E, D>,
pub bias: Tensor<Rank1<C>, E, D>,
pub running_mean: Tensor<Rank1<C>, E, D>,
pub running_var: Tensor<Rank1<C>, E, D>,
pub epsilon: E,
pub momentum: E,
}
Expand description
Batch normalization for sequences as described in Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
Generics:
C
the size of the dimension to reduce. Both for 2d tensors (of the form <BATCH_SIZE, DIMENSION>) as well as 3d tensors (of the form <BATCH_SIZE, DIMENSION, SEQUENCE_LENGTH>), this is the 1st dimension.
Training vs Inference
BatchNorm1D supports the following cases (see sections below for more details):
- Training: ModuleMut and OwnedTape on the input tensor
- Inference: Module and NoneTape on the input tensor.
NOTE: ModuleMut/NoneTape, and Module/OwnedTape will fail to compile.
Examples:
type Model = BatchNorm1D<3>;
let bn = dev.build_module::<Model, f32>();
let _ = bn.forward(dev.zeros::<Rank2<4, 3>>());
let _ = bn.forward(dev.zeros::<Rank3<4, 3, 2>>());
Training
- Running statistics: updated with momentum
- Normalization: calculated using batch stats
Inference
- Running statistics: not updated
- Normalization: calculated using running stats
Fields§
§scale: Tensor<Rank1<C>, E, D>
Scale for affine transform. Defaults to 1.0
bias: Tensor<Rank1<C>, E, D>
Bias for affine transform. Defaults to 0.0
running_mean: Tensor<Rank1<C>, E, D>
Spatial mean that is updated during training. Defaults to 0.0
running_var: Tensor<Rank1<C>, E, D>
Spatial variance that is updated during training. Defaults to 1.0
epsilon: E
Added to variance before taking sqrt for numerical stability. Defaults to 1e-5
momentum: E
Controls exponential moving average of running stats.Defaults to 0.1
running_stat * (1.0 - momentum) + stat * momentum
.
Trait Implementations§
source§impl<const C: usize, E: Clone + Dtype, D: Clone + DeviceStorage> Clone for BatchNorm1D<C, E, D>
impl<const C: usize, E: Clone + Dtype, D: Clone + DeviceStorage> Clone for BatchNorm1D<C, E, D>
source§fn clone(&self) -> BatchNorm1D<C, E, D>
fn clone(&self) -> BatchNorm1D<C, E, D>
1.0.0 · source§fn clone_from(&mut self, source: &Self)
fn clone_from(&mut self, source: &Self)
source
. Read moresource§impl<const C: usize, E: Debug + Dtype, D: Debug + DeviceStorage> Debug for BatchNorm1D<C, E, D>
impl<const C: usize, E: Debug + Dtype, D: Debug + DeviceStorage> Debug for BatchNorm1D<C, E, D>
source§impl<B: Dim, const C: usize, E: Dtype, D: Device<E>> Module<Tensor<(B, Const<C>), E, D, NoneTape>> for BatchNorm1D<C, E, D>
impl<B: Dim, const C: usize, E: Dtype, D: Device<E>> Module<Tensor<(B, Const<C>), E, D, NoneTape>> for BatchNorm1D<C, E, D>
source§fn try_forward(
&self,
x: Tensor<(B, Const<C>), E, D, NoneTape>
) -> Result<Self::Output, D::Err>
fn try_forward( &self, x: Tensor<(B, Const<C>), E, D, NoneTape> ) -> Result<Self::Output, D::Err>
Inference 1d forward - does not update Self::running_mean and Self::running_var
type Error = <D as HasErr>::Err
source§impl<B: Dim, const C: usize, L: Dim, E: Dtype, D: Device<E>> Module<Tensor<(B, Const<C>, L), E, D, NoneTape>> for BatchNorm1D<C, E, D>
impl<B: Dim, const C: usize, L: Dim, E: Dtype, D: Device<E>> Module<Tensor<(B, Const<C>, L), E, D, NoneTape>> for BatchNorm1D<C, E, D>
source§fn try_forward(
&self,
x: Tensor<(B, Const<C>, L), E, D, NoneTape>
) -> Result<Self::Output, D::Err>
fn try_forward( &self, x: Tensor<(B, Const<C>, L), E, D, NoneTape> ) -> Result<Self::Output, D::Err>
Inference 2d forward - does not update Self::running_mean and Self::running_var
§type Output = Tensor<(B, Const<C>, L), E, D, NoneTape>
type Output = Tensor<(B, Const<C>, L), E, D, NoneTape>
Input
.type Error = <D as HasErr>::Err
source§impl<B: Dim, const C: usize, E: Dtype, D: Device<E>> ModuleMut<Tensor<(B, Const<C>), E, D, OwnedTape<E, D>>> for BatchNorm1D<C, E, D>
impl<B: Dim, const C: usize, E: Dtype, D: Device<E>> ModuleMut<Tensor<(B, Const<C>), E, D, OwnedTape<E, D>>> for BatchNorm1D<C, E, D>
source§fn try_forward_mut(
&mut self,
x: Tensor<(B, Const<C>), E, D, OwnedTape<E, D>>
) -> Result<Self::Output, D::Err>
fn try_forward_mut( &mut self, x: Tensor<(B, Const<C>), E, D, OwnedTape<E, D>> ) -> Result<Self::Output, D::Err>
Training 2d forward - updates Self::running_mean and Self::running_var
§type Output = Tensor<(B, Const<C>), E, D, OwnedTape<E, D>>
type Output = Tensor<(B, Const<C>), E, D, OwnedTape<E, D>>
Input
.type Error = <D as HasErr>::Err
source§fn forward_mut(&mut self, input: Input) -> Self::Output
fn forward_mut(&mut self, input: Input) -> Self::Output
source§impl<B: Dim, const C: usize, L: Dim, E: Dtype, D: Device<E>> ModuleMut<Tensor<(B, Const<C>, L), E, D, OwnedTape<E, D>>> for BatchNorm1D<C, E, D>
impl<B: Dim, const C: usize, L: Dim, E: Dtype, D: Device<E>> ModuleMut<Tensor<(B, Const<C>, L), E, D, OwnedTape<E, D>>> for BatchNorm1D<C, E, D>
source§fn try_forward_mut(
&mut self,
x: Tensor<(B, Const<C>, L), E, D, OwnedTape<E, D>>
) -> Result<Self::Output, D::Err>
fn try_forward_mut( &mut self, x: Tensor<(B, Const<C>, L), E, D, OwnedTape<E, D>> ) -> Result<Self::Output, D::Err>
Training 1d forward - updates Self::running_mean and Self::running_var
§type Output = Tensor<(B, Const<C>, L), E, D, OwnedTape<E, D>>
type Output = Tensor<(B, Const<C>, L), E, D, OwnedTape<E, D>>
Input
.type Error = <D as HasErr>::Err
source§fn forward_mut(&mut self, input: Input) -> Self::Output
fn forward_mut(&mut self, input: Input) -> Self::Output
source§impl<const C: usize, E: Dtype, D: Device<E>> TensorCollection<E, D> for BatchNorm1D<C, E, D>
impl<const C: usize, E: Dtype, D: Device<E>> TensorCollection<E, D> for BatchNorm1D<C, E, D>
§type To<E2: Dtype, D2: Device<E2>> = BatchNorm1D<C, E2, D2>
type To<E2: Dtype, D2: Device<E2>> = BatchNorm1D<C, E2, D2>
source§fn iter_tensors<V: ModuleVisitor<Self, E, D>>(
visitor: &mut V
) -> Result<Option<Self::To<V::E2, V::D2>>, V::Err>
fn iter_tensors<V: ModuleVisitor<Self, E, D>>( visitor: &mut V ) -> Result<Option<Self::To<V::E2, V::D2>>, V::Err>
Err(_)
to indicate an error,
Ok(None)
to indicate that there is no error and a module has not been built, and
Ok(Some(_))
contains Self::Output<E2, D2>