[MNT] lack of test cases covering all functionality for multiple regressors

[Ramana-Raja]

Describe the issue

many regressors, such as convolution-based and feature-based models, are missing test cases which actually test all the functionality of the regressors

Suggest a potential alternative/fix

it would be good to add test cases which cover up all functionality of the regressors

Additional context

No response

[MatthewMiddlehurst]

This issue needs more detail. Every estimator is covered by the general testing.

[Ramana-Raja]

This issue needs more detail. Every estimator is covered by the general testing.

I meant that there is a lack of test cases that check whether the tests actually generate correct or accurate outputs.For example,using ARI scores for clusterers etc, and also making sure the tests cover all functionality of regressors.

[MatthewMiddlehurst]

Please edit the base issue. Also we do test expected output, so there is no need to test scores for the default parameters if the results are already stored.

[Ramana-Raja]

Please edit the base issue. Also we do test expected output, so there is no need to test scores for the default parameters if the results are already stored.

Alright, understood. Do you have any suggestions on what else could be added to the test cases