* Add tests for functions
* Define `Model` object to incorporate the training set and the formula. This will get rid of `train` argument in the following functions:
+ `produce_roc_table(mod, train)`
+ `count_cases(mod, data, train, ...)`
+ `model_by_lrt(mod, train, ...)`
+ `model_by_vif(mod, train, ...)`
+ `terbin_model(mod, train)`
* Generalize functions to work with `sm.GLM` of any family and features of any df, as well as `sm.OLS`:
+ `anova`
+ `drop1`
+ `model_by_lrt`
+ `plot_rl`
ver 0.0.1 (GitHub)
* Added the following plotting functions:
+ `plot_lm`
+ `plot_qq`
+ `plot_rf`
+ `plot_rlev`
+ `plot_sl`
ver 0.0.1 (PyPI)
* Created a repository
* Uploaded to PyPI
* Wrote a README.md and setup.py
* Wrote functions in statspark/r_inspired.py