add validation using assert sample and rtype

[sbidari]

This PR aims to adds validation of RandomVariables using _assert_sample_and_rtype function. The validation methods currently implemented varies between checking something is a random variable using isinstance and checking that the random variable is valid (i.e check tuple return type).

Based on the comments here #341 (comment), I think @damonbayer would like to keep the current implementation of assert isinstance and maybe add further check with _assert_sample_and_rtype? I wanted to confirm my understanding before adding more to this PR

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 93.47%. Comparing base (87a5960) to head (81665bf).
Report is 24 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #356      +/-   ##
==========================================
+ Coverage   93.24%   93.47%   +0.23%     
==========================================
  Files          39       39              
  Lines         918      920       +2     
==========================================
+ Hits          856      860       +4     
+ Misses         62       60       -2     

Flag	Coverage Δ
unittests	93.47% <100.00%> (+0.23%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[damonbayer]

I would like it so that any time we have some function f(x: RandomVariable), we validate that x is a RandomVariable using _assert_type.

Separately, any time we instantiate a RandomVariable, we check that it is valid via something like _assert_sample_and_rtype.

So if you have code like this

my_non_rv = y
f(x=my_non_rv)

f will error because x is not a RandomVariable.

But if you have code like this

my_rv = some_way_to_make_an_rv_with_the_wrong_return_type()
f(x=my_rv)

some_way_to_make_an_rv_with_the_wrong_return_type will error because it has the wrong return type and we won't even try to call f.

cc @dylanhmorris and @gvegayon for thoughts.

[sbidari]

I would like it so that any time we have some function f(x: RandomVariable), we validate that x is a RandomVariable using _assert_type.

This can be done in #354

Separately, any time we instantiate a RandomVariable, we check that it is valid via something like _assert_sample_and_rtype.

I don't think _assert_sample_and_rtype is useful for this. But happy to hear if you have ideas for it.

I think this PR and associated issue can be closed? @damonbayer

[sbidari]

Note for clarification: shelving this for now after discussion with Damon. Potential design changes may affect this in the future.

[damonbayer]

Note for clarification: shelving this for now after discussion with Damon. Potential design changes may affect this in the future.

Closing for now.