Struct ClassificationOutput 

pub struct ClassificationOutput {
    pub loss: Tensor<1>,
    pub output: Tensor<2>,
    pub targets: Tensor<1, Int>,
}

Simple classification output adapted for multiple metrics.

Supported metrics:

• Accuracy
• AUROC
• TopKAccuracy
• Perplexity
• Precision (via ConfusionStatsInput)
• Recall (via ConfusionStatsInput)
• FBetaScore (via ConfusionStatsInput)
• Loss.

Fields

loss: Tensor<1>

The loss.

output: Tensor<2>

The class logits or probabilities. Shape: [batch_size, num_classes].

targets: Tensor<1, Int>

The ground truth class index for each sample. Shape: [batch_size].

Implementations

impl ClassificationOutput

pub fn new(loss: Tensor<1>, output: Tensor<2>, targets: Tensor<1, Int>) -> Self

Constructs a new ClassificationOutput.

Trait Implementations

impl Adaptor<AccuracyInput> for ClassificationOutput

fn adapt(&self) -> AccuracyInput

Adapt the type to be passed to a metric.

impl Adaptor<ConfusionStatsInput> for ClassificationOutput

fn adapt(&self) -> ConfusionStatsInput

Adapt the type to be passed to a metric.

impl Adaptor<LossInput> for ClassificationOutput

fn adapt(&self) -> LossInput

Adapt the type to be passed to a metric.

impl Adaptor<PerplexityInput> for ClassificationOutput

fn adapt(&self) -> PerplexityInput

Adapt the type to be passed to a metric.

impl Adaptor<TopKAccuracyInput> for ClassificationOutput

fn adapt(&self) -> TopKAccuracyInput

Adapt the type to be passed to a metric.

impl ItemLazy for ClassificationOutput

fn sync(self) -> Result<Self, ExecutionError>

Sync the item.