Precision Recall

Module: generative_models.core.evaluation.metrics.general.precision_recall

Source: generative_models/core/evaluation/metrics/general/precision_recall.py

Overview

Precision and Recall metrics for generative models using JAX.

This module implements the improved Precision and Recall metrics for evaluating generative models as proposed in "Improved Precision and Recall Metric for Assessing Generative Models" (Kynkäänniemi et al., 2019).

All implementations use JAX for compatibility with NNX modules.

Classes

DensityPrecisionRecall

class DensityPrecisionRecall

PrecisionRecall

class PrecisionRecall

Functions

init

def __init__()

init

def __init__()

compute

def compute()

compute

def compute()

compute_f1_score

def compute_f1_score()

compute_improved_precision_recall

def compute_improved_precision_recall()

compute_manifold_radii

def compute_manifold_radii()

compute_precision

def compute_precision()

compute_recall

def compute_recall()

density_precision_recall

def density_precision_recall()

estimate_densities

def estimate_densities()

precision_recall

def precision_recall()

Module Statistics

• Classes: 2
• Functions: 12
• Imports: 6