MeanAbsoluteErrorLoss<T>
Models & Types · AiDotNet.LossFunctions
Implements the Mean Absolute Error (MAE) loss function.
Inheritance: LossFunctionBase<T> → MeanAbsoluteErrorLoss<T>
For Beginners
Mean Absolute Error measures the average absolute difference between predicted and actual values.
The formula is: MAE = (1/n) * ?|predicted - actual|
MAE has these key properties:
- It treats all errors linearly (unlike MSE which squares errors)
- It’s less sensitive to outliers than MSE
- It’s simple to understand as the average magnitude of errors
- It’s always positive, with perfect predictions giving a value of zero
MAE is ideal for problems where:
- You’re predicting continuous values
- You want all errors to be treated equally (not emphasizing large errors)
- The prediction errors follow a Laplace distribution
- Outliers should not have a disproportionate influence on the model
Example
using AiDotNet.LossFunctions;
using AiDotNet.Tensors.LinearAlgebra;
var loss = new MeanAbsoluteErrorLoss<float>();
var predicted = new Vector<float>(new[] { 0.9f, 0.2f, 0.7f });
var actual = new Vector<float>(new[] { 1.0f, 0.0f, 1.0f });
float value = loss.CalculateLoss(predicted, actual);
Console.WriteLine($"MeanAbsoluteErrorLoss = {value:F4}");
Constructors
| Constructor | Summary |
|---|---|
MeanAbsoluteErrorLoss() |
Methods
| Method | Summary |
|---|---|
T CalculateLoss(Vector<T> predicted, Vector<T> actual) |
Calculates the Mean Absolute Error between predicted and actual values. |
Tensor<T> ComputeTapeLoss(Tensor<T> predicted, Tensor<T> target) |
|
ValueTuple<T, Tensor<T>> CalculateLossAndGradientGpu(Tensor<T> predicted, Tensor<T> actual) |
Calculates both MAE loss and gradient on GPU in a single efficient pass. |