Model Evaluation Measures (Median Absolute Error, R^2 Score)

Explore how to evaluate regression models using Median Absolute Error, a robust error measure against outliers, and the R squared score, which shows how well the model explains data variance. Understand when to use these metrics and how they help compare model performance effectively.

We'll cover the following...

- Median absolute error
  - Formula
  - Code
- R² score, the coefficient of determination
  - Formula
  - Code
- Key Take Aways

Median absolute error

Median is a Descriptive Statistics Measure and is robust to outliers in the dataset. The rate of error is calculated by taking the median of all absolute differences between the target and the prediction. Lower values means the model is more accurate.

Formula

If $\hat{y}$ is the predicted target real valued output, then $y$ is the corresponding target real valued output and $m$ is the total number of instances. Then the median absolute error is calculated as follows:

$MedAE(\hat{y}, y) = median(|\hat{y}^1 - y^1|,...,|\hat{y}^m - y^m|)$ ...

1.What is Data Science ?

2.Applications of Data Science

3.Overview of Libraries

4.Probability and Statistics

5.Machine Learning Part-1

6.Machine Learning Part-2

7.Machine Learning Part-3

8.Deep Learning

9.Machine Learning Tools and Libraries

10.Big Data Tools and Technologies

11.Where to go next ?

Mock Interview

Mock Interview

Model Evaluation Measures (Median Absolute Error, R^2 Score)

Median absolute error

Formula