Enum LossReduction 

pub enum LossReduction {
    Mean,
    Sum,
}

How the per-sample losses are combined into a single scalar for backpropagation.

• Mean - divide by the number of elements (scale-invariant, recommended for most use cases).
• Sum - sum without normalisation (sensitive to batch size; may require a lower learning rate).

Variants

Mean

Average the loss over all contributing pairs.

Sum

Sum the loss over all contributing pairs without normalisation.

Trait Implementations

impl Clone for LossReduction

fn clone(&self) -> LossReduction

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for LossReduction

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

Formats the value using the given formatter.

impl<'de> Deserialize<'de> for LossReduction

fn deserialize<__D>(__deserializer: __D) -> Result<Self, __D::Error>

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl Serialize for LossReduction

fn serialize<__S>(&self, __serializer: __S) -> Result<__S::Ok, __S::Error>

where __S: Serializer,

Serialize this value into the given Serde serializer.