Refactoring and improvement of msm covariance matrices

[AldoGl]

Proposed changes

• change None to 'identity' for the default behaviour of the covariance matrix
• add the 'inverse_variance' option for the covariance matrix of Franke (2009)
• add the 'standardise_moments' flag to only consider relative changes wrt the target moments

Types of changes

What types of changes does your code introduce?
Put an x in the boxes that apply

• Bugfix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)

Checklist

Put an x in the boxes that apply.

• I have read the CONTRIBUTING doc
• I am making a pull request against the develop branch (left side). Also you should start your branch off our develop.
• Lint and unit tests pass locally with my changes
• I have added tests that prove my fix is effective or that my feature works

[marcofavoritobi]

at this point it is probably missing a check on the type of covariance_mat? Below we assume that if covariance_mat is not a str, then it is a NDArray, but this is not enforced anywhere. On one hand, this follows the Python philosophy "we are all consenting adults" and the user is responsible of what he's doing. Perhaps a one-more check is not a bad idea though.

Suggested change

	)
	)
	if not isinstance(covariance_mat, np.ndarray):
	raise ValueError(...)

[AldoGl]

thank you, I added this check

[marcofavoritobi]

this is ok if the check in the constructor is added. It is slightly-less OK otherwise, but still good enough :)

[marcofavoritobi]

Possible improvement. Instead of this, we could have an Enum class:

class CovarianceMatrixType:
    IDENTITY = "identity"
    INVERSE_VARIANCE = "inverse_variance"

And then here (we should first check it is not a string!):

try:
    covariance_mat = CovarianceMatrixType(covariance_mat)
except ValueError:
    raise ValueError(f"expected one of {list(map(lambda x: x.value, CovarianceMatrixType))}, got {covariance_mat})

[AldoGl]

Ok thanks, I implemented this too