misc minor corrections

Author: ercbk

README.Rmd

@@ -35,13 +35,13 @@ Various elements of the technique can be altered to improve performance. These i
 3. Inner-Loop CV strategy  
 4. Grid search strategy  
 
-For the performance experiemnt (question 2), I'll be varying the repeats of the outer-loop cv strategy for each method. The fastest implementation of each method will be tuned with different sizes of data ranging from 100 to 5000 observations. The mean absolute error will be calculated for each combination of repeat, data size, and method. 
+For the performance experiment (question 2), I'll be varying the repeats of the outer-loop cv strategy for each method. The fastest implementation of each method will be tuned with different sizes of data ranging from 100 to 5000 observations. The mean absolute error will be calculated for each combination of repeat, data size, and method. 
 
 I'm using a 4 core, 16 GB RAM machine.
 
 Progress (duration in seconds)  
 
-![](duration-experiment/outputs/0224-results.png)  
+![](duration-experiment/outputs/0225-results.png)  
 
 References  
 

README.md

@@ -50,7 +50,7 @@ These include:
 3\. Inner-Loop CV strategy  
 4\. Grid search strategy
 
-For the performance experiemnt (question 2), I’ll be varying the repeats
+For the performance experiment (question 2), I’ll be varying the repeats
 of the outer-loop cv strategy for each method. The fastest
 implementation of each method will be tuned with different sizes of data
 ranging from 100 to 5000 observations. The mean absolute error will be
@@ -60,7 +60,7 @@ I’m using a 4 core, 16 GB RAM machine.
 
 Progress (duration in seconds)
 
-![](duration-experiment/outputs/0224-results.png)
+![](duration-experiment/outputs/0225-results.png)
 
 References
 

duration-experiment/kuhn-johnson/nested-cv-h2o-kj.R

@@ -70,7 +70,7 @@ ncv_dat_10 <- rsample::nested_cv(small_dat,
                               inside = bootstraps(times = 25))
 
 
-
+# Start h2o cluster
 h2o.init()
 
 
@@ -86,14 +86,15 @@ error_FUN <- function(model){
 }
 
 
+# Distributed Random Forest
 
 rf_FUN <- function(x, y, anal_h2o, ass_h2o, params) {
 
    mtries <- params$mtries[[1]]
    ntrees <- params$ntrees[[1]]
 
    # h20 ususally needs unique ids or loops will return exact same values over and over
-   modelId <- as.character(Sys.time())
+   gridId <- as.character(dqrng::dqrnorm(1))
 
    h2o.show_progress()
 
@@ -107,6 +108,8 @@ rf_FUN <- function(x, y, anal_h2o, ass_h2o, params) {
 }
 
 
+# Elastic Net Regression
+
 glm_FUN <- function(x, y, anal_h2o, ass_h2o, params) {
 
    alpha <- params$alpha[[1]]
@@ -154,6 +157,7 @@ params_list <- list(glm = list(alpha = c(0, 0.25, 0.5, 0.75, 1),
 #####################################################
 
 
+# inputs params, model, and resample, calls model and error functions, outputs error
 mod_error <- function(params, mod_FUN, dat) {
    anal_df <- rsample::analysis(dat)
    ass_df <- rsample::assessment(dat)
@@ -297,4 +301,5 @@ tic.clearlog()
 # MLflow uses waitress for Windows. Killing it also kills mlflow.exe, python.exe, console window host processes
 installr::kill_process(process = c("waitress-serve.exe"))
 
-
+# shutdown cluster
+h2o.shutdown(prompt = FALSE)

duration-experiment/kuhn-johnson/nested-cv-parsnip-kj.R

@@ -101,7 +101,7 @@ pars_ranger_FUN <- function(params, analysis_set) {
 
 
 
-# Regularized Regression
+# Elastic Net Regression
 
 glm_FUN <- function(params, analysis_set) {
    alpha <- params$mixture[[1]]

duration-experiment/kuhn-johnson/nested-cv-ranger-kj.R

@@ -102,7 +102,7 @@ ranger_FUN <- function(params, analysis_set) {
 }
 
 
-# Regularized Regression
+# Elastic Net Regression
 
 glm_FUN <- function(params, analysis_set) {
    alpha <- params$mixture[[1]]

duration-experiment/kuhn-johnson/nested-cv-sklearn-kj.R

@@ -92,6 +92,8 @@ error_FUN <- function(y_obs, y_hat){
 #####################################
 
 
+# Random Forest
+
 sklearn_rf_FUN <- function(params, analysis_set) {
    sklearn_e <- import("sklearn.ensemble")
    max_features <- r_to_py(params$mtry[[1]])
@@ -112,7 +114,7 @@ sklearn_rf_FUN <- function(params, analysis_set) {
 }
 
 
-# Regularized Regression
+# Elastic Net Regression
 
 glm_FUN <- function(params, analysis_set) {
       alpha <- params$mixture[[1]]

duration-experiment/kuhn-johnson/nested-cv-tune-kj.R

@@ -157,6 +157,7 @@ params_list <- list(glm = glm_params, rf = rf_params)
 ################################
 
 
+# inputs params, model, and resample, calls model and error functions, outputs error
 mod_error <- function(params, mod_FUN, dat) {
    y_col <- ncol(dat$data)
    y_obs <- assessment(dat)[y_col]