.NET: Convert a Binary Loss to a Percent Accuracy

1. For perfect predictions, the loss approaches 0, which should correspond to ~100% accuracy (~1.0)
2. For random predictions, the loss approaches e (e=2.71828), which should correspond to ~0% accuracy (~0.0)

/// <summary>
/// Converts a binary cross-entropy loss value to an accuracy percentage (0-1).
/// </summary>
/// <param name="loss">The binary cross-entropy loss value from the network</param>
/// <returns>Accuracy value between 0 and 1 (0% to 100%)</returns>
public static float ConvertBinaryLossToAccuracy(float loss)
{
	return MathF.Exp(-loss)
}

How to use this static method:

float loss = 0.693f; // roughly 50% accuracy
float percentAccuracy = ConvertBinaryLossToAccuracy(loss);
Console.WriteLine($"{percentAccuracy:P1}");
// --- OR ---
float percentAccuracy2 = ConvertBinaryLossToAccuracy(loss) * 100;
Console.WriteLine($"{percentAccuracy2:F1}%");

Comment