Also known as: MAE
Mean Absolute Error (MAE) is a regression metric that measures the average absolute difference between predicted and actual values. Because it uses absolute errors rather than squared errors, it is less sensitive to large outliers than mean squared error.
In regression , an evaluation metric that is the average of all the model errors, where model error is the distance between the predicted label value and the correct label value.
The average loss per example when L~1~ loss is used. Calculate Mean Absolute Error as follows: 1. Calculate the L~1~ loss for a batch. 2. Divide the L~1~ loss by the number of examples in the batch. where: - $n$ is the number of examples. - $y$ is the actual value of the label. - $\\\\y\$ is the value that the model predicts for $y$. For example, consider the calculation of L~1~ loss on the following batch of five examples: Model's predicted value --- 6 4 11 6 8 So, L~1~ loss is 8 and the number of examples is 5. Therefore, the Mean Absolute Error is: Contrast Mean Absolute Error with Mean Squared Error and Root Mean Squared Error.