Trait Metric 

pub trait Metric:
    Send
    + Sync
    + Clone {
    type Input;

    // Required methods
    fn name(&self) -> Arc<String>;
    fn update(
        &mut self,
        item: &Self::Input,
        metadata: &MetricMetadata,
    ) -> SerializedEntry;
    fn clear(&mut self);

    // Provided methods
    fn description(&self) -> Option<String> { ... }
    fn attributes(&self) -> MetricAttributes { ... }
}

Metric trait.

Notes

Implementations should define their own input type only used by the metric. This is important since some conflict may happen when the model output is adapted for each metric’s input type.

Required Associated Types

type Input

The input type of the metric.

Required Methods

fn name(&self) -> Arc<String>

The parameterized name of the metric.

This should be unique, so avoid using short generic names, prefer using the long name.

For a metric that can exist at different parameters (e.g., top-k accuracy for different values of k), the name should be unique for each instance.

fn update( &mut self, item: &Self::Input, metadata: &MetricMetadata, ) -> SerializedEntry

Update the metric state and returns the current metric entry.

fn clear(&mut self)

Clear the metric state.

Provided Methods

fn description(&self) -> Option<String>

A short description of the metric.

fn attributes(&self) -> MetricAttributes

Attributes of the metric.

By default, metrics have no attributes.

Dyn Compatibility

This trait is not dyn compatible.

In older versions of Rust, dyn compatibility was called "object safety", so this trait is not object safe.

Implementors

impl Metric for CpuMemory

type Input = ()

impl Metric for CpuTemperature

type Input = ()

impl Metric for CpuUse

type Input = ()

impl Metric for CudaMetric

type Input = ()

impl Metric for IterationSpeedMetric

type Input = ()

impl Metric for LearningRateMetric

type Input = ()

impl<B> Metric for AccuracyMetric<B>

where B: Backend,

type Input = AccuracyInput<B>

impl<B> Metric for AurocMetric<B>

where B: Backend,

type Input = AurocInput<B>

impl<B> Metric for CharErrorRate<B>

where B: Backend,

The character error rate metric implementation.

type Input = CerInput<B>

impl<B> Metric for FBetaScoreMetric<B>

where B: Backend,

type Input = ConfusionStatsInput<B>

impl<B> Metric for HammingScore<B>

where B: Backend,

type Input = HammingScoreInput<B>

impl<B> Metric for LossMetric<B>

where B: Backend,

type Input = LossInput<B>

impl<B> Metric for PerplexityMetric<B>

where B: Backend,

type Input = PerplexityInput<B>

impl<B> Metric for PrecisionMetric<B>

where B: Backend,

type Input = ConfusionStatsInput<B>

impl<B> Metric for RecallMetric<B>

where B: Backend,

type Input = ConfusionStatsInput<B>

impl<B> Metric for TopKAccuracyMetric<B>

where B: Backend,

type Input = TopKAccuracyInput<B>

impl<B> Metric for WordErrorRate<B>

where B: Backend,

type Input = WerInput<B>

impl<B, const D: usize> Metric for DiceMetric<B, D>

where B: Backend,

type Input = DiceInput<B, D>