splitvars - Split multicolumn variables in table or timetable - MATLAB (original) (raw)

Split multicolumn variables in table or timetable

Syntax

Description

T2 = splitvars([T1](#d126e1695509)) splits all multicolumn variables in T1 so that they are single-column variables in T2. All single-column variables fromT1 are unaltered.

• If a variable in T1 has multiple columns, thensplitvars makes unique names for the new variables in T2 from the name of the original variable inT1.
• If a variable in T1 is a table itself, thensplitvars uses the names of its variables (and, if necessary, the name of that table) to make unique names for the new variables in T2.

To merge variables into one multicolumn variable, use the mergevars function.

example

T2 = splitvars([T1](#d126e1695509),[vars](#d126e1695528)) splits only the table variables specified by vars. You can specify variables by name, by position, or using logical indices.

example

T2 = splitvars([T1](#d126e1695509),[vars](#d126e1695528),'NewVariableNames',[newNames](#d126e1695562)) assigns new names to the designated variables that are split out ofT1 and copied to T2.

example

Examples

collapse all

Create a table from workspace variables. Some of the variables are matrices with multiple columns.

A = (1:3)'; B = [5 11 12; 20 30 50; 0.1 3.4 5.9]'; C = {'a','XX';'b','YY';'c','ZZ'}; D = [128 256 512]'; T1 = table(A,B,C,D)

T1=3×4 table A B C D _ ________________ _______________ ___

1     5     20    0.1    {'a'}    {'XX'}    128
2    11     30    3.4    {'b'}    {'YY'}    256
3    12     50    5.9    {'c'}    {'ZZ'}    512

Split the variables B and C. All variables in the output table have one column.

T2=3×7 table A B_1 B_2 B_3 C_1 C_2 D _ ___ ___ ___ _____ ______ ___

1     5     20     0.1    {'a'}    {'XX'}    128
2    11     30     3.4    {'b'}    {'YY'}    256
3    12     50     5.9    {'c'}    {'ZZ'}    512

Create a table that contains tables, using arrays of data from the patients.mat file. Display the first three rows.

load patients Personal_Data = table(Gender,Age); BMI_Data = table(Height,Weight); BloodPressure = table(Systolic,Diastolic); T1 = table(LastName,Personal_Data,BMI_Data,BloodPressure); head(T1,3)

  LastName        Personal_Data          BMI_Data            BloodPressure    
____________    _________________    ________________    _____________________

                  Gender      Age    Height    Weight    Systolic    Diastolic
                __________    ___    ______    ______    ________    _________
                                                                              
{'Smith'   }    {'Male'  }    38       71       176        124          93    
{'Johnson' }    {'Male'  }    43       69       163        109          77    
{'Williams'}    {'Female'}    38       64       131        125          83    

Specify BloodPressure as the variable to split.

T2 = splitvars(T1,'BloodPressure'); head(T2,3)

  LastName        Personal_Data          BMI_Data        Systolic    Diastolic
____________    _________________    ________________    ________    _________

                  Gender      Age    Height    Weight                         
                __________    ___    ______    ______                         
                                                                              
{'Smith'   }    {'Male'  }    38       71       176        124          93    
{'Johnson' }    {'Male'  }    43       69       163        109          77    
{'Williams'}    {'Female'}    38       64       131        125          83    

To specify multiple variables by name, use a cell array of character vectors.

T3 = splitvars(T1,{'BMI_Data','BloodPressure'}); head(T3,3)

  LastName        Personal_Data      Height    Weight    Systolic    Diastolic
____________    _________________    ______    ______    ________    _________

                  Gender      Age                                             
                __________    ___                                             
                                                                              
{'Smith'   }    {'Male'  }    38       71       176        124          93    
{'Johnson' }    {'Male'  }    43       69       163        109          77    
{'Williams'}    {'Female'}    38       64       131        125          83    

To specify variables by position, use a numeric array.

T4 = splitvars(T1,[2 4]); head(T4,3)

  LastName        Gender      Age        BMI_Data        Systolic    Diastolic
____________    __________    ___    ________________    ________    _________

                                     Height    Weight                         
                                     ______    ______                         
                                                                              
{'Smith'   }    {'Male'  }    38       71       176        124          93    
{'Johnson' }    {'Male'  }    43       69       163        109          77    
{'Williams'}    {'Female'}    38       64       131        125          83    

Create a table that contains multi-column variables, using data from the patients.mat file. Display the first three rows.

load patients Personal_Data = [Age,Height,Weight]; BloodPressure = [Systolic,Diastolic]; T1 = table(LastName,Gender,Personal_Data,BloodPressure); head(T1,3)

  LastName        Gender       Personal_Data      BloodPressure
____________    __________    ________________    _____________

{'Smith'   }    {'Male'  }    38     71    176     124     93  
{'Johnson' }    {'Male'  }    43     69    163     109     77  
{'Williams'}    {'Female'}    38     64    131     125     83  

Split BloodPressure and specify new names for the new variables in the output table.

T2 = splitvars(T1,'BloodPressure','NewVariableNames',{'Systolic','Diastolic'}); head(T2,3)

  LastName        Gender       Personal_Data      Systolic    Diastolic
____________    __________    ________________    ________    _________

{'Smith'   }    {'Male'  }    38     71    176      124          93    
{'Johnson' }    {'Male'  }    43     69    163      109          77    
{'Williams'}    {'Female'}    38     64    131      125          83    

Split both BMI_Data and BloodPressure. For each variable being split, you must provide a cell array with the correct number of new names.

T3 = splitvars(T1,{'Personal_Data','BloodPressure'},... 'NewVariableNames',{{'Age','Height','Weight'},{'Systolic','Diastolic'}}); head(T3,3)

  LastName        Gender      Age    Height    Weight    Systolic    Diastolic
____________    __________    ___    ______    ______    ________    _________

{'Smith'   }    {'Male'  }    38       71       176        124          93    
{'Johnson' }    {'Male'  }    43       69       163        109          77    
{'Williams'}    {'Female'}    38       64       131        125          83    

Input Arguments

collapse all

Input table, specified as a table or timetable.

Variables in the input table, specified as a string array, character vector, cell array of character vectors, pattern scalar, numeric array, or logical array.

Names of the split variables, specified as a cell array of character vectors or string array.

Extended Capabilities

expand all

Thesplitvars function fully supports tall arrays. For more information, see Tall Arrays.

Usage notes and limitations:

• The vars input argument does not support pattern expressions.
• The value of the 'NewVariableNames' name-value argument must be constant.
• The variables that are split cannot have a variable number of columns.

Version History

Introduced in R2018a