feat: add input validation for non-private methods of approximation.py

[pc532627]

Refs: 251

PR Type

• Feature

Description

Input validation added for approximation.py as a check before proceeding with other modules.

How Has This Been Tested?

Test A: test_kernel_mean_approximator_creation_invalid_types
Test B: test_random_approximator_creation_invalid_types
Test C: test_annchor_approximator_creation_invalid_types
Test D: test_nystrom_approximator_creation_invalid_types

Checklist before requesting a review

• I have made sure that my PR is not a duplicate.
• My code follows the style guidelines of this project.
• I have commented my code, particularly in hard-to-understand areas.
• I have performed a self-review of my code.
• I have made corresponding changes to the documentation.
• My changes generate no new warnings.
• I have added tests that prove my fix is effective or that my feature works.
• New and existing unit tests pass locally with my changes.
• Any dependent changes have been merged and published in downstream modules.

[pc532627]

@tp832944 first module covered for an initial check of the approach. I suggest we don't verify inputs of private functions, as these are not intended to be accessed by users. We need to clarify exactly what is an acceptable input if something is just 'array like' and how to check this. Also, we should verify what to do with JIT decorated functions - if they check only on first call, then we aren't validating on future calls when compiled. I suggest we don't input validate in these cases.

[tp832944]

Your validation function idea is good @pc532627. I've suggested several changes to keep it more general.

[tp832944]

Thank you for your changes @pc532627. Here are a few more please.

[tp832944]

All good @pc532627. Now for someone else to extend it to the rest of the codebase (new PR).