3 ms·
The thing is that anymore anything short of a deep net is equally difficult to implement as a linear regression, if not easier due to nulls and categoricals. A
by oneoff786 4y ago
The thing is that anymore anything short of a deep net is equally difficult to implement as a linear regression, if not easier due to nulls and categoricals.
Also, modern tooling makes a lot of these models more than explainable enough for a lot of cases… 10% is a lot