Evaluate Predictions

Evaluate predicted cell fractions against known ground truth.

Usage

EvaluatePredictions(predictions, truth)

Arguments

predictions

Data frame of predicted fractions (samples x cell types).

truth

Data frame of true fractions (same format as predictions).

Value

A list containing:

rmse

Root mean squared error overall

mae

Mean absolute error overall

correlation

Pearson correlation overall

per_celltype

Metrics per cell type

per_sample

Metrics per sample

Examples

if (FALSE) { # \dontrun{
# Evaluate on held-out test data
metrics <- EvaluatePredictions(predictions, true_fractions)
print(metrics$rmse)
} # }