Sam Maurer
Sam Maurer
OLSRegressionStep uses the dependent variable from the model expression as the `out_column` (destination for predicted values) if none is specified. There's a bug where we're not stripping inline transformations, causing...
Better reporting about missing values for MNL -- When data tables have missing values, those rows are automatically filtered out (I think by Patsy) before models are estimated or predicted...
This issue is to discuss managing Orca table and column registrations, similarly to how the ModelManager module handles model step registrations. The objectives are similar to the model step management:...
Writing up a column ordering glitch. I've only half looked into this, so some of the details may be wrong. Calling `DataFrameWrapper.to_frame(column_list)`, if the columns you ask for are in...
I'm having trouble merging sets of tables with overlapping broadcast relationships. For example, these combinations run: - broadcasts from A -> B and A -> C, merge tables A, B,...
References: - https://doi.org/10.1177%2F0081175018793460 - https://infoscience.epfl.ch/record/117096/files/Frej07.pdf
The `test_parallel_lottery_choices` unit test is failing in certain environments. My local machine, MacOS 10.15: - py36 and py37: works - py38 and py39: error GitHub Actions ([link](https://github.com/UDST/choicemodels/runs/1921592875?check_suite_focus=true)): - ubuntu: all...
It looks like the [MultinomialLogitResults](https://github.com/UDST/choicemodels/blob/4e89877626b661334c4e04bda17e2f7aa81bfc4e/choicemodels/mnl.py#L228) class can only provide fitted probabilities for MergedChoiceTables that are generated with sampling of alternatives. This is a bug; there's no reason we can't calculate...
This is a new issue to discuss performance optimization for sampling of alternatives in the `MergedChoiceTable` utility. Earlier discussion is in issues #4, #5, and #11, now closed. Core functionality...
PyLogit estimation for MNL is more flexible than the native ChoiceModels implementation (originally brought over from `urbansim.urbanchoice.mnl`), but much slower. See initial experiments here: [Destination-choice-models-02.ipynb](https://github.com/UDST/choicemodels/blob/master/notebooks/Destination-choice-models-02.ipynb) What's causing the performance gap?...