eztune_table: Makes a table of performance results

In jillbo1000/EZtuneTest:

Description

eztune_table creates a table that contains the average error rates and run times for a series of eztune runs and compares them to the best results obtained from a grid search.

Usage

eztune_table(dat, value = "loss_mse_acc_10", best_grid = NULL)

Arguments

dat	Dataset created from load_opt_data or with several datasets from load_opt_data combined with rbind. Information from this dataset will be used to populate the majority of the table.
value	the value from dat that should be used as the error metric. This will typically be either loss_mse_acc_10 or loss_mae_auc_10
best_grid	A dataset that contains the best error rate obtained from the grid search. The first column must contain the names of the datasets used to populate the tables. The names must be identical to those in the dat object. The second column is numeric and contains the best error measure obtained by the grid search for that dataset. If it is omitted the table will be made without the grid information. The function get_best_grid can be used to create

Value

Returns a data.frame with the performance metrics from the EZtune tests. Each column is a character vector. This data.frame can be passed to the function eztune_table_latex to create a latex version of the table.

See Also

load_opt_data, get_best_grid, get_grid_data, eztune_table_latex

jillbo1000/EZtuneTest documentation built on Oct. 5, 2021, 4:16 p.m.