Explanation of L2 norm of an error
When solving for the sparsest possible set of DEs, it is likely our found model will not describe the data exactly - there will be an error
Therefore we can measure the error and give the user it's $l^2$-norm
- The error is a vector of errors at each time-step
- More information regarding the $l^2$-norm is here
When working with 2 or more dimensional data, the $l^2$-norm returned will be vector of $l^2$ norms in each coordinate