evaluate: Evaluate performance of hierarchical classifier

In joseah/HierscPred: Supervised hierarchical classification for scPred

Description

Calculated hierarchical F1-score, median F1-score, percentage of rejected cells, percentage of internally labeled cells, confusion matrix, and population size

Usage

evaluate(y_true, y_pred, tree)

Arguments

y_true	vector with true labels
y_pred	vector with predicted labels
tree	data.tree object storing the hierarchy used during classification

Value

A list containing the different evaluation metrics (median HF1-score, HF1-score per cell population, median F1-score, F1-score per cell population, and population size)

Author(s)

Lieke Michielsen and Jose Alquicira-Hernandez

Examples

library(data.tree)

res <- evaluate(y_true, y_pred, tree)

joseah/HierscPred documentation built on Nov. 25, 2021, 12:20 p.m.