DoPredictions: Helper functions for running predictions
In cities-lab/VETravelDemandMM: Simulate Multi-Modal Travel Demand for Households

Description Usage Arguments Details Value

run predictions on Dataset_df with model objects in Model_df

1 2	DoPredictions(Model_df, Dataset_df, dataset_name, id_name, y_name, SegmentCol_vc = NULL, combine_preds = FALSE)

`Model_df`	A list-column data_frame containing a 'model' list-column, column(s) used for segmenting data, and, optionally, a 'Step' column and a 'post_func' for multi-step models and post-processing function for predictions.
`Dataset_df`	A data frame providing independent variables and segmentation column for predictions included in the specification of the model objects.
`dataset_name`	A character string for name of the Dataset_df
`id_name`	A character string for the id column of Dataset_df
`y_name`	A character string for name of the outcome variable
`SegmentCol_vc`	A vector for columns used for segmentation; if NULL assuming columns other than c("model", "Step", "post_func", "bias.adj") in Model_df is used for segmentation
`combine_preds`	A logical flag indicating whether to combine predictions for multi-step models (default FALSE)

DoPredictions predicts outcomes (y) for each observation in the Dataset_df dataset using independent variables included in Dataset_df and model objects saved in list-column data_frame Model_df.

A list containing the components specified in the Set specifications for the module along with: LENGTH: A named integer vector having a single named element, "Household", which identifies the length (number of rows) of the Household table to be created in the datastore. SIZE: A named integer vector having two elements. The first element, "Azone", identifies the size of the longest Azone name. The second element, "HhId", identifies the size of the longest HhId.

cities-lab/VETravelDemandMM documentation built on Aug. 1, 2019, 4:43 p.m.