Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
* Include modellable_params argument in Model class. This can be used if you do not want rough heuristics to be used to decide which params have modellable values * rerun black formatting * - seperate functions for laplace smoothing - store intermediate non laplace smoothed counts in model class * use deepcopy instead of copy to prevent original variable from being edited * try change the way laplace smoothing is done for param counts * undo experimental change to laplace smoothing for params * change how the param probabilities are computed. (changed from categorical mistake to bernoulli) * update tests based on changes on how to calculate param probabilities * fix one docstring issue in laplace_smooth module * Add module for sessionizing data * handle nans when sessionizing and aggregating * fix bug in create_session_col function when we have 0 rows in df * write tests for the new sessionize module * include notebook showing how to use anomalous_sequence sub package * add kql magic temp files to gitignore * add AnomalousSequence rst file to DataAnalysis section * Some formatting fixes Co-authored-by: Ian Hellen <ianhelle@microsoft.com>
- Loading branch information