XPSConfidenceMetricsEntry

import type { XPSConfidenceMetricsEntry } from "https://googleapis.deno.dev/v1/language:v2.ts";

ConfidenceMetricsEntry includes generic precision, recall, f1 score etc. Next tag: 16.

interface XPSConfidenceMetricsEntry {

confidenceThreshold?: number;

f1Score?: number;

f1ScoreAt1?: number;

falseNegativeCount?: bigint;

falsePositiveCount?: bigint;

falsePositiveRate?: number;

falsePositiveRateAt1?: number;

positionThreshold?: number;

precision?: number;

precisionAt1?: number;

recall?: number;

recallAt1?: number;

trueNegativeCount?: bigint;

truePositiveCount?: bigint;

}

§Properties

confidenceThreshold?: number

[src]

Metrics are computed with an assumption that the model never return predictions with score lower than this value.

f1Score?: number

[src]

The harmonic mean of recall and precision.

f1ScoreAt1?: number

[src]

The harmonic mean of recall_at1 and precision_at1.

falseNegativeCount?: bigint

[src]

The number of ground truth labels that are not matched by a model created label.

falsePositiveCount?: bigint

[src]

The number of model created labels that do not match a ground truth label.

falsePositiveRate?: number

[src]

False Positive Rate for the given confidence threshold.

falsePositiveRateAt1?: number

[src]

The False Positive Rate when only considering the label that has the highest prediction score and not below the confidence threshold for each example.

positionThreshold?: number

[src]

Metrics are computed with an assumption that the model always returns at most this many predictions (ordered by their score, descendingly), but they all still need to meet the confidence_threshold.

precision?: number

[src]

Precision for the given confidence threshold.

precisionAt1?: number

[src]

The precision when only considering the label that has the highest prediction score and not below the confidence threshold for each example.

recall?: number

[src]

Recall (true positive rate) for the given confidence threshold.

recallAt1?: number

[src]

The recall (true positive rate) when only considering the label that has the highest prediction score and not below the confidence threshold for each example.

trueNegativeCount?: bigint

[src]

The number of labels that were not created by the model, but if they would, they would not match a ground truth label.

truePositiveCount?: bigint

[src]

The number of model created labels that match a ground truth label.