Package index
-
find_label_issues() - Find examples with likely label errors
-
label_quality_scores() - Label quality scores
-
health_summary() - Summarise label health by class
Confident learning estimates
The quantities behind the label issues: per-class thresholds, the confident joint, the estimated joint distribution of given and true labels, and the noise transition matrices.
-
class_thresholds() - Per-class confidence thresholds
-
confident_joint() - The confident joint
-
estimate_joint() - Estimate the joint distribution of given and true labels
-
noise_matrices() - Noise transition matrices and class priors
-
num_label_issues() - Estimate the number of label errors
-
label_inputs() - Standardise labels and predicted probabilities
-
simulate_label_noise() - Simulate a classification dataset with label errors