training: Estimate an Agent Model with Supervised Learning
In JohnNay/eat: Empirical Agent Training Software for Data-Driven Modeling

training uses a caret::train or stats::glm algorithm to estimate an individual-level model and return a list where each element is either a list of with element of class "train" (if caret::train was used) or a list with each element of class "glm" if (if stat::glm was used).

training(trainData, features, Formula, sampling = FALSE,
  sampling_size = 1000, package = c("caretglm", "caretglmnet", "glm",
  "caretnnet", "caretdnn"), tune_length = 10, parallel = FALSE,
  cv_type = c("cv", "repeatedcv"))

`trainData`	`data.frame` with each row (obervational unit) being an individual agent decision. With a column called "group" specifying which group of `agg_patterns` each obseravtion is in.
`features`	`list` of the variables (columns in `data`) to be used in the prediction `Formula`. As many elements in the `list` as we want discrete models for different times. Each element of the `list` is a `character vector`, with each element of the `character vector` being a feature to use for training an individual-level model.
`Formula`	`list` where each element is a length one character vector that specifies a formula, e.g. `"y ~ x"`. The character vector makes sense in the context of the `features` and `data`. There are as many elements in the list as there are discrete models for different times.
`sampling`	optional logical vector length one, default is `FALSE`. If `sampling == TRUE`, we sample equal numbers of observations from each 'group' to reduce potential problems with the final estimated model being too affected by groups with more observations.
`sampling_size`	optional numeric vector length one specifying how many observations from each group that `training` should sample to train the model, default is 1000. Only applicable when `sampling` argument is set to `TRUE`.
`package`	optional character vector length one, default is `"caretglm", "caretglmnet", "glm", "caretnnet", "caretdnn"`.
`tune_length`	optional numeric vector length one specifying how many rows for caret::train to create in design matrix of hyper-parameter sets.
`parallel`	optional logical vector length one, default is `FALSE`. Uses `doParallel::registerDoParallel()`. Should be `FALSE` when training() is being called from inside cv_abm(), which, by default, is already running in parallel. @param cores optional numeric vector length one.
`cv_type`	optional character vector length one, default is `c("cv", "repeatedcv")`. Passed on to `caret::trainControl()`.