looping_cv_shell: GBM Cross Validation Function

In skirmer/skpack: Personal package of functions I use regularly.

Description

Notes: the "looping_cv_shell" function runs the entire cross validation procedure. Steps before you run: 1. Check that the params are what you want- if you don't need to test all the variations, then by all means set single values. 2. This runs all the outcomes currently available - if more or different have been built, change outcome_list and omit_list to reflect. 3. The metrics produced are: AUC, RMSE, Brier Score, Log Loss, Avg Precision at 100, Avg Precision at 500, and Avg Precision at 1000. If you want others, add them at the end and ensure that the results output c(), label set, and the unlist statements are adjusted accordingly.

Usage

looping_cv_shell(nfold = 5, colsamplebt = 1, eval_metric = "rmse",
  nthread = 2, splitpct = 0.8)

Arguments

No	actual inputs, just make sure your dataset containing all desired variables is located in the gloval environment titled final_data_dt. And make sure it's a data.table.

Examples

cv_test <- looping_cv_shell()
cv_test1 <- data.frame(matrix(unlist(cv_test), ncol = 12, byrow=T), stringsAsFactors = FALSE)
cv_test1 <- rbind(cv_test1, data.frame(matrix(unlist(looping_cv_shell()), ncol = 12, byrow=T), stringsAsFactors = FALSE))
colnames(cv_test1) <- c("AUC", "RMSE", "Brier","Log Loss","Avg Precision at 100","Avg Precision at 500",
"Avg Precision at 1000","Outcome", "Trees", "Max Depth", "Eta", "Subsample")

skirmer/skpack documentation built on May 30, 2019, 1:06 a.m.