[Joss review] tests with real-life data, and compare to other implementations
Closed this issue · 4 comments
This is more relevant for testing regression models. Since real-life data never fits neatly into a "box", it is useful to compare results against other implementations (lifelines uses R's survival and flexsurvreg libraries).
I only completed the regression model fitters after writing the paper. And in the paper I make no claims about regression models. If I remove all references to regression models, and submit a major change in future, can we exclude the regression features from this review?
I'm fine with dropping #20, but then I would still suggest testing real-life data in non-regression tests and comparing vs R and/or lifelines. This can also test the robustness of your _parameter_initialiser methods (since we'll be testing it against different-distributed data points)
Removed reference to regression in docs. 457c77e
Added tests to lifelines data sets. See e3da8d7