Struct Learner 

pub struct Learner<LC: LearningComponentsTypes> { /* private fields */ }

Learner struct encapsulating all components necessary to train a Neural Network model.

Implementations

impl<B, LR, M, O> Learner<LearningComponentsMarker<B, LR, M, O>>

where B: AutodiffBackend, LR: LrScheduler + 'static, M: TrainStep + AutodiffModule<B> + Display + 'static, M::InnerModule: InferenceStep, O: Optimizer<M, B> + 'static,

pub fn new(model: M, optim: O, lr_scheduler: LR) -> Self

Create a learner.

impl<LC: LearningComponentsTypes> Learner<LC>

pub fn fork(&mut self, device: &<TrainingBackend<LC> as Backend>::Device)

Fork the learner’s model to the given device.

pub fn model(&self) -> LC::TrainingModel

Returns the current model.

pub fn lr_current(&self) -> f64

Returns the current learning rate.

pub fn lr_step(&mut self)

Executes a step of the learning rate scheduler.

pub fn train_step( &self, item: <<LC as LearningComponentsTypes>::TrainingModel as TrainStep>::Input, ) -> TrainOutput<<<LC as LearningComponentsTypes>::TrainingModel as TrainStep>::Output>

Runs a step of the model for training, which executes the forward and backward passes.

Arguments

• item - The input for the model.

Returns

The output containing the model output and the gradients.

pub fn optimizer_step(&mut self, grads: GradientsParams)

Optimize the current module with the provided gradients and learning rate.

Arguments

• optim: Optimizer used for learning.
• lr: The learning rate used for this step.
• grads: The gradients of each parameter in the current model.

pub fn optimizer_step_multi(&mut self, grads: MultiGradientsParams)

Optimize the current module with the provided gradients and learning rate.

Arguments

• optim: Optimizer used for learning.
• lr: The learning rate used for this step.
• grads: Multiple gradients associated to each parameter in the current model.

pub fn load_model(&mut self, record: LearnerModelRecord<LC>)

Load the module state from a record.

pub fn load_optim(&mut self, record: LearnerOptimizerRecord<LC>)

Load the state of the learner’s optimizer as a record.

pub fn load_scheduler(&mut self, record: LearnerSchedulerRecord<LC>)

Load the state of the learner’s scheduler as a record.

Trait Implementations

impl<LC: LearningComponentsTypes> Clone for Learner<LC>

fn clone(&self) -> Self

Returns a duplicate of the value.

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

Performs copy-assignment from source.