splitit to splitit2

README.md

@@ -8,7 +8,7 @@ your needs.  If, however, you need to implement something a bit different from
 generic use cases, the package also includes lower level commands that can save 
 you time from having to code the entire cross-validation process.  These 
 commands are named after the four-steps found in all cross-validation work:
-`splitit`, `fitit`, `predictit`, and `validateit`.  There are also a few utility 
+`splitit2`, `fitit`, `predictit`, and `validateit`.  There are also a few utility 
 commands that take care of the metaprogramming tasks needed to allow these 
 commands to be applied to the correct fold/split of the data.
 
@@ -250,9 +250,9 @@ them.
 
 # Phase Specific Commands 
 
-## splitit
+## splitit2
 ```
-splitit # [#] [if] [in] [, Uid(varlist) TPoint(string asis) KFold(integer 1) 
+splitit2 # [#] [if] [in] [, Uid(varlist) TPoint(string asis) KFold(integer 1) 
 						   SPLit(string asis) loo ]
 ```						   
 

cmdmod.sthlp

@@ -46,7 +46,7 @@ predictions are made on the appropriate out-of-sample portion of the data set.
 {phang}
 {opt spl:it} must contain the name of the variable that stores the test, 
 validation, and test splits.  There will only be a single variable if the splits 
-were created using {help splitit}.  
+were created using {help splitit2}.  
 
 {dlgtab:Optional}
 

crossvalidate.pkg

@@ -5,7 +5,7 @@ d implement multiple cross-validation procedures with estiamtion commands in
 d Stata.  There are two prefix commands -xv- and -xvloo- that are used to 
 d implement the entire cross-validation pipeline with the estimation command 
 d supplied by the user.  For users that require more fine-grained control over 
-d the cross-validation process, the commands -splitit-, -fitit-, -predictit-, &
+d the cross-validation process, the commands -splitit2-, -fitit-, -predictit-, &
 d -validateit- implement each phase of cross-validation discretely.  There are 
 d also several utility commands included as well.  Lastly, the included Mata 
 d library contains over 40 different evaluation functions covering binary and 
@@ -47,8 +47,8 @@ f libxv.mlib
 f libxv.sthlp
 f predictit.ado
 f predictit.sthlp
-f splitit.ado
-f splitit.sthlp
+f splitit2.ado
+f splitit2.sthlp
 f state.ado
 f state.sthlp
 f validateit.ado

crossvalidate.sthlp

@@ -48,7 +48,7 @@ individual commands.
 {synopt :{opt {help xv}}}Cross-Validation{p_end}
 {synopt :{opt {help xvloo}}}Leave-One-Out Cross-Validation{p_end}
 {syntab:Lower Level Commands}
-{synopt :{opt {help splitit}}}Splits the dataset into train/test or train/validation/test splits{p_end}
+{synopt :{opt {help splitit2}}}Splits the dataset into train/test or train/validation/test splits{p_end}
 {synopt :{opt {help fitit}}}Calls the estimation command on the appropriate split{p_end}
 {synopt :{opt {help predictit}}}Predicts the outcome on the appropriate split{p_end}
 {synopt :{opt {help validateit}}}Computes {p_end}
@@ -84,7 +84,7 @@ able to use this prefix.
 {dlgtab:Lower Level Commands}
 
 {phang}
-{help splitit} is a command called by the prefix commands to create the splits 
+{help splitit2} is a command called by the prefix commands to create the splits 
 in the data in memory.  As mentioned above, you can create train/test and 
 train/validation/test splits with or without K-Folds, using simple random 
 sampling or clustered sampling (which includes sampling panel units).  This 

fitit.sthlp

@@ -47,7 +47,7 @@ INCLUDE help xvphase-fit
 {phang}
 {opt spl:it} must contain the name of the variable that stores the test, 
 validation, and test splits.  There will only be a single variable if the splits 
-were created using {help splitit}.  
+were created using {help splitit2}.  
 
 {phang}
 {opt res:ults} is used to {help estimates_store:estimates store} the estimation 

libxv.sthlp

@@ -101,7 +101,7 @@ the classes being predicted cannot be handled.  Functions that might otherwise
 allow the specification of a parameter to adjust the computation are also 
 unable to be accommodated.  If you find the need to use that type of a 
 validation/test metric, you could still avoid having to write an entire 
-cross-validation pipeline by using {help splitit} and {help fitit} to do that 
+cross-validation pipeline by using {help splitit2} and {help fitit} to do that 
 work for you.  Then, predict the values of interest and compute your metric.  If 
 what we've provided in this package already meets your needs, feel free to use 
 the {help xv} prefix instead to let the computer do the work.

predictit.sthlp

@@ -63,7 +63,7 @@ fitting the model to all of the training data.
 {phang}
 {opt spl:it} must contain the name of the variable that stores the test, 
 validation, and test splits.  There will only be a single variable if the splits 
-were created using {help splitit}.  Additionally, if you are passing an 
+were created using {help splitit2}.  Additionally, if you are passing an 
 estimation command string as the first argument to this command, you must 
 pass the split variable name to this option.
 

splitit.ado → splitit2.ado

@@ -5,15 +5,16 @@
 *                                                                              *
 *******************************************************************************/
 
-*! splitit
+*! splitit2
+*! splitit -> splitit2 to prevent collision with existing SSC routine
 *! v 0.0.11
 *! 28FEB2024
 
 // Drop program from memory if already loaded
-cap prog drop splitit
+cap prog drop splitit2
 
 // Define program
-prog def splitit, rclass sortpreserve
+prog def splitit2, rclass sortpreserve
 
 	// Version statement 
 	version 15

splitit.sthlp → splitit2.sthlp

@@ -1,19 +1,19 @@
 {smcl}
 {* *! version 0.0.7 06mar2024}{...}
-{viewerjumpto "Syntax" "splitit##syntax"}{...}
-{viewerjumpto "Description" "splitit##description"}{...}
-{viewerjumpto "Options" "splitit##options"}{...}
-{viewerjumpto "Examples" "splitit##examples"}{...}
-{viewerjumpto "Returned Values" "splitit##retvals"}{...}
-{viewerjumpto "Additional Information" "splitit##additional"}{...}
-{viewerjumpto "Contact" "splitit##contact"}{...}
+{viewerjumpto "Syntax" "splitit2##syntax"}{...}
+{viewerjumpto "Description" "splitit2##description"}{...}
+{viewerjumpto "Options" "splitit2##options"}{...}
+{viewerjumpto "Examples" "splitit2##examples"}{...}
+{viewerjumpto "Returned Values" "splitit2##retvals"}{...}
+{viewerjumpto "Additional Information" "splitit2##additional"}{...}
+{viewerjumpto "Contact" "splitit2##contact"}{...}
 {title:Dataset Splitting and Folding for Cross-Validation in Stata}
 
 {marker syntax}{...}
 {title:Syntax}
 
 {p 8 32 2}
-{cmd:splitit} # [#] {ifin} [{cmd:,} {cmdab:u:id(}{it:varlist}{cmd:)} 
+{cmd:splitit2} # [#] {ifin} [{cmd:,} {cmdab:u:id(}{it:varlist}{cmd:)} 
 {cmdab:tp:oint(}{it:string asis}{cmd:)} {cmdab:k:fold(}{it:integer}{cmd:)} 
 {cmdab:spl:it(}{it:string asis}{cmd:)} {cmd:loo}]{p_end}
 
@@ -68,7 +68,7 @@ set.
 {phang}
 {opt spl:it} is used to specify the name of a new variable that will store the 
 identifiers for the splits in the data.  If no argument is passed to this option 
-{cmd splitit} will store the result in a variable named {it:_xvsplit}.
+{cmd splitit2} will store the result in a variable named {it:_xvsplit}.
 
 {phang}
 {opt loo} is an option used to alter the underlying logic used to compute the 
@@ -87,27 +87,27 @@ the predicted and observed values for all of the training set units in aggregate
 
 {p 4 4 2}Simple Random Sampling{p_end}
 {p 6 4 2}Test/Train Split{p_end}
-{p 8 4 2}{stata splitit .8, ret(splitvar)}{p_end}
+{p 8 4 2}{stata splitit2 .8, ret(splitvar)}{p_end}
 {p 6 4 2}Train/Validation/Test Split{p_end}
-{p 8 4 2}{stata splitit .6 .2, ret(splitvar)}{p_end}
+{p 8 4 2}{stata splitit2 .6 .2, ret(splitvar)}{p_end}
 
 {p 4 4 2}K-Fold Simple Random Sampling{p_end}
 {p 6 4 2}Test/Train Split{p_end}
-{p 8 4 2}{stata splitit .8, ret(splitvar) k(5)}{p_end}
+{p 8 4 2}{stata splitit2 .8, ret(splitvar) k(5)}{p_end}
 {p 6 4 2}Train/Validation/Test Split{p_end}
-{p 8 4 2}{stata splitit .6 .2, ret(splitvar) k(5)}{p_end}
+{p 8 4 2}{stata splitit2 .6 .2, ret(splitvar) k(5)}{p_end}
 
 {p 4 4 2}Clustered Random Sampling{p_end}
 {p 6 4 2}Test/Train Split{p_end}
-{p 8 4 2}{stata splitit .8, ret(splitvar) uid(foreign)}{p_end}
+{p 8 4 2}{stata splitit2 .8, ret(splitvar) uid(foreign)}{p_end}
 {p 6 4 2}Train/Validation/Test Split{p_end}
-{p 8 4 2}{stata splitit .6 .2, ret(splitvar) uid(foreign)}{p_end}
+{p 8 4 2}{stata splitit2 .6 .2, ret(splitvar) uid(foreign)}{p_end}
 
 {p 4 4 2}K-Fold Clustered Random Sampling{p_end}
 {p 6 4 2}Test/Train Split{p_end}
-{p 8 4 2}{stata splitit .8, ret(splitvar) k(5) uid(foreign)}{p_end}
+{p 8 4 2}{stata splitit2 .8, ret(splitvar) k(5) uid(foreign)}{p_end}
 {p 6 4 2}Train/Validation/Test Split{p_end}
-{p 8 4 2}{stata splitit .6 .2, ret(splitvar) k(5) uid(foreign)}{p_end}
+{p 8 4 2}{stata splitit2 .6 .2, ret(splitvar) k(5) uid(foreign)}{p_end}
 
 
 {marker retvals}{...}

test/crossvalidatetest.do

@@ -2,7 +2,7 @@ di "Test Suite for the Cross-Validation Suite"
 do test/libxvtests.do
 do test/classifytests.do
 do test/cmdmodtests.do
-do test/splitittests.do
+do test/splitit2tests.do
 do test/fitittests.do
 do test/predictittests.do
 do test/validateittests.do