Trait candle_core::Module

pub trait Module: Debug {
    // Required method
    fn forward(&self, xs: &Tensor) -> Result<Tensor>;

    // Provided method
    fn set_training(&mut self, _training: bool) { ... }
}

Required Methods

fn forward(&self, xs: &Tensor) -> Result<Tensor>

Provided Methods

fn set_training(&mut self, _training: bool)

Change the module to use training mode vs eval mode.

The default implementation does nothing as this is only used for a couple modules such as dropout or batch-normalization.

Implementors

impl Module for QMatMul