Main Content

rows2vars

Reorient table or timetable so that rows become variables

Description

example

T2 = rows2vars(T1) reorients the rows of T1, so that they become variables in the output table T2, as shown in the diagram. If rows2vars can concatenate the contents of the rows of T1, then the corresponding variables of T2 are arrays. Otherwise, the variables of T2 are cell arrays. rows2vars always returns a table, though T1 can be either a table or a timetable.

The rows2vars function copies the names of the variables of T1 to a new variable of T2. If T1 has row names or row times, then those names or times become the variable names of T2. Otherwise, rows2vars generates names Var1,…,VarN as the variable names of T2.

example

T2 = rows2vars(T1,Name,Value) specifies additional arguments using one or more name-value pair arguments. For example, you can use the 'VariableNamesSource' name-value pair argument to specify the source of the variable names of T2.

Examples

collapse all

Create tables, and then reorient their rows to be variables in new tables.

Load arrays of data from the patients.mat file. Create a table that contains the LastName, Gender, Age, Height, and Weight variables.

load patients
T1 = table(LastName,Gender,Age,Height,Weight);
head(T1,3)
ans=3×5 table
      LastName        Gender      Age    Height    Weight
    ____________    __________    ___    ______    ______

    {'Smith'   }    {'Male'  }    38       71       176  
    {'Johnson' }    {'Male'  }    43       69       163  
    {'Williams'}    {'Female'}    38       64       131  

Reorient the rows of T1 to be the variables of the output table.

T2 = rows2vars(T1);

Display the first four variables of T2. The first variable of T2 contains the names of the variables of T1. The remaining variables of T2 correspond to rows of T1. Since T1 did not have any row labels, the variables of T2 have default names, Var1 to VarN for N variables.

T2(:,1:4)
ans=5×4 table
    OriginalVariableNames      Var1          Var2            Var3    
    _____________________    _________    ___________    ____________

        {'LastName'}         {'Smith'}    {'Johnson'}    {'Williams'}
        {'Gender'  }         {'Male' }    {'Male'   }    {'Female'  }
        {'Age'     }         {[   38]}    {[     43]}    {[      38]}
        {'Height'  }         {[   71]}    {[     69]}    {[      64]}
        {'Weight'  }         {[  176]}    {[    163]}    {[     131]}

Create a table with row names. If a table has row names, then rows2vars turns the row names into the names of variables.

T3 = table(Gender,Age,Height,Weight,'RowNames',LastName);
head(T3,3)
ans=3×4 table
                  Gender      Age    Height    Weight
                __________    ___    ______    ______

    Smith       {'Male'  }    38       71       176  
    Johnson     {'Male'  }    43       69       163  
    Williams    {'Female'}    38       64       131  

Reorient the rows of T3.

T4 = rows2vars(T3);
T4(:,1:4)
ans=4×4 table
    OriginalVariableNames     Smith      Johnson      Williams 
    _____________________    ________    ________    __________

         {'Gender'}          {'Male'}    {'Male'}    {'Female'}
         {'Age'   }          {[  38]}    {[  43]}    {[    38]}
         {'Height'}          {[  71]}    {[  69]}    {[    64]}
         {'Weight'}          {[ 176]}    {[ 163]}    {[   131]}

Load a timetable and display it.

load bostonTT
Boston
Boston=6×3 timetable
           Time            Temp    WindSpeed    Rain
    ___________________    ____    _________    ____

    2016-06-09 06:03:00    59.5       0.1       0.05
    2016-06-09 12:00:23      63       2.3       0.08
    2016-06-09 18:02:57    61.7       3.1       0.13
    2016-06-10 06:01:47    55.4       5.7       0.15
    2016-06-10 12:06:00    62.3       2.6       0.87
    2016-06-10 18:02:57    58.8       6.2       0.33

Reorient it so that its rows become variables in the output. The rows2vars function turns the row times into names, but modifies them so that they are valid variable names. Also, the output argument returned by rows2vars is always a table, even when the input argument is a timetable.

T = rows2vars(Boston)
Warning: Table variable names that were not valid MATLAB identifiers have been modified. Since table variable names must be unique, any table variable names that happened to match the new identifiers also have been modified.
To use the original row names as new variable names, set 'PreserveVariableNames' to true.
T=3×7 table
    OriginalVariableNames    x2016_06_0906_03_00    x2016_06_0912_00_23    x2016_06_0918_02_57    x2016_06_1006_01_47    x2016_06_1012_06_00    x2016_06_1018_02_57
    _____________________    ___________________    ___________________    ___________________    ___________________    ___________________    ___________________

        {'Temp'     }               59.5                     63                   61.7                   55.4                   62.3                   58.8        
        {'WindSpeed'}                0.1                    2.3                    3.1                    5.7                    2.6                    6.2        
        {'Rain'     }               0.05                   0.08                   0.13                   0.15                   0.87                   0.33        

Starting in R2020a, you can take names from the input table or timetable and use them as variable names in the output table without modification. To preserve the original names, use the 'VariableNamingRule' name-value pair argument.

Load a timetable and display it.

load bostonTT
Boston
Boston=6×3 timetable
           Time            Temp    WindSpeed    Rain
    ___________________    ____    _________    ____

    2016-06-09 06:03:00    59.5       0.1       0.05
    2016-06-09 12:00:23      63       2.3       0.08
    2016-06-09 18:02:57    61.7       3.1       0.13
    2016-06-10 06:01:47    55.4       5.7       0.15
    2016-06-10 12:06:00    62.3       2.6       0.87
    2016-06-10 18:02:57    58.8       6.2       0.33

Reorient the timetable so that its row times become variable names in the output table. Convert the datetime values to strings and preserve the resulting names using the 'VariableNamingRule' name-value pair.

T = rows2vars(Boston,'VariableNamingRule','preserve')
T=3×7 table
    OriginalVariableNames    2016-06-09 06:03:00    2016-06-09 12:00:23    2016-06-09 18:02:57    2016-06-10 06:01:47    2016-06-10 12:06:00    2016-06-10 18:02:57
    _____________________    ___________________    ___________________    ___________________    ___________________    ___________________    ___________________

        {'Temp'     }               59.5                     63                   61.7                   55.4                   62.3                   58.8        
        {'WindSpeed'}                0.1                    2.3                    3.1                    5.7                    2.6                    6.2        
        {'Rain'     }               0.05                   0.08                   0.13                   0.15                   0.87                   0.33        

The variable names in T are not valid MATLAB identifiers because the dates start with a number. However, you can use dot notation to refer to such variables, using parentheses.

T.('2016-06-09 06:03:00')
ans = 3×1

   59.5000
    0.1000
    0.0500

Read data from a spreadsheet into a table. Display the first three rows.

T1 = readtable('patients.xls');
head(T1,3)
ans=3×10 table
      LastName        Gender      Age              Location               Height    Weight    Smoker    Systolic    Diastolic    SelfAssessedHealthStatus
    ____________    __________    ___    _____________________________    ______    ______    ______    ________    _________    ________________________

    {'Smith'   }    {'Male'  }    38     {'County General Hospital'  }      71       176      true        124          93             {'Excellent'}      
    {'Johnson' }    {'Male'  }    43     {'VA Hospital'              }      69       163      false       109          77             {'Fair'     }      
    {'Williams'}    {'Female'}    38     {'St. Mary's Medical Center'}      64       131      false       125          83             {'Good'     }      

Reorient the rows of T1 to be variables of a new table, T2. Specify that the LastName variable from T1 is the source of the names of the variables of T2.

T2 = rows2vars(T1,'VariableNamesSource','LastName');

Display the first four variables of T2. The first variable of T2 contains the names of the variables of T1. The remaining variables of T2 correspond to rows of T1.

T2(:,1:4)
ans=9×4 table
       OriginalVariableNames                   Smith                   Johnson                  Williams           
    ____________________________    ___________________________    _______________    _____________________________

    {'Gender'                  }    {'Male'                   }    {'Male'       }    {'Female'                   }
    {'Age'                     }    {[                     38]}    {[         43]}    {[                       38]}
    {'Location'                }    {'County General Hospital'}    {'VA Hospital'}    {'St. Mary's Medical Center'}
    {'Height'                  }    {[                     71]}    {[         69]}    {[                       64]}
    {'Weight'                  }    {[                    176]}    {[        163]}    {[                      131]}
    {'Smoker'                  }    {[                      1]}    {[          0]}    {[                        0]}
    {'Systolic'                }    {[                    124]}    {[        109]}    {[                      125]}
    {'Diastolic'               }    {[                     93]}    {[         77]}    {[                       83]}
    {'SelfAssessedHealthStatus'}    {'Excellent'              }    {'Fair'       }    {'Good'                     }

Display the data in T2.Smith. In this example, every variable of T2 is a 9-by-1 cell array, because the values in the rows of T1 cannot be concatenated into arrays.

T2.Smith
ans=9×1 cell array
    {'Male'                   }
    {[                     38]}
    {'County General Hospital'}
    {[                     71]}
    {[                    176]}
    {[                      1]}
    {[                    124]}
    {[                     93]}
    {'Excellent'              }

Read data from a spreadsheet into a table. Use the first column of the spreadsheet as the row names of the table. Display the first three rows.

T1 = readtable('patients.xls','ReadRowNames',true);
head(T1,3)
ans=3×9 table
                  Gender      Age              Location               Height    Weight    Smoker    Systolic    Diastolic    SelfAssessedHealthStatus
                __________    ___    _____________________________    ______    ______    ______    ________    _________    ________________________

    Smith       {'Male'  }    38     {'County General Hospital'  }      71       176      true        124          93             {'Excellent'}      
    Johnson     {'Male'  }    43     {'VA Hospital'              }      69       163      false       109          77             {'Fair'     }      
    Williams    {'Female'}    38     {'St. Mary's Medical Center'}      64       131      false       125          83             {'Good'     }      

Reorient specified variables from T1 and discard the rest. To specify data variables by name, use a cell array of character vectors.

T2 = rows2vars(T1,'DataVariables',{'Gender','Age','Height','Weight'});
T2(:,1:4)
ans=4×4 table
    OriginalVariableNames     Smith      Johnson      Williams 
    _____________________    ________    ________    __________

         {'Gender'}          {'Male'}    {'Male'}    {'Female'}
         {'Age'   }          {[  38]}    {[  43]}    {[    38]}
         {'Height'}          {[  71]}    {[  69]}    {[    64]}
         {'Weight'}          {[ 176]}    {[ 163]}    {[   131]}

You also can specify data variables by position in the input table. To specify positions of variables, use a numeric array.

T3 = rows2vars(T1,'DataVariables',[1 2 6:9]);
T3(:,1:4)
ans=6×4 table
       OriginalVariableNames            Smith        Johnson      Williams 
    ____________________________    _____________    ________    __________

    {'Gender'                  }    {'Male'     }    {'Male'}    {'Female'}
    {'Age'                     }    {[       38]}    {[  43]}    {[    38]}
    {'Smoker'                  }    {[        1]}    {[   0]}    {[     0]}
    {'Systolic'                }    {[      124]}    {[ 109]}    {[   125]}
    {'Diastolic'               }    {[       93]}    {[  77]}    {[    83]}
    {'SelfAssessedHealthStatus'}    {'Excellent'}    {'Fair'}    {'Good'  }

Input Arguments

collapse all

Input table, specified as a table or timetable.

Name-Value Pair Arguments

Specify optional comma-separated pairs of Name,Value arguments. Name is the argument name and Value is the corresponding value. Name must appear inside quotes. You can specify several name and value pair arguments in any order as Name1,Value1,...,NameN,ValueN.

Example: 'VariableNamingRule','preserve' preserves original names taken from T1, without modifying them to be valid MATLAB® identifiers.

Variable in T1 that contains variable names, specified as the comma-separated pair consisting of 'VariableNamesSource' and a character vector, string scalar, positive integer, or logical vector. The rows2vars function interprets the value of 'VariableNamesSource' as shown in the table.

Value of 'VariableNamesSource'

Meaning

Character vector or string scalar

Name of a variable in the input table T1.

Integer n

The nth variable in T1.

Logical vector, whose length equals the number of variables in T1

The nth element is 1 (true), corresponding to the nth variable of T1. All other elements are 0 (false).

While the value of 'VariableNamesSource' must be a name, number, or logical array that specifies a table variable, the variable itself can have any data type, with these limitations.

  • The values contained in the specified table variable must have a data type that allows the values to be converted to strings. For example, the value of 'VariableNamesSource' can be the name of a table variable that contains a datetime array, because datetime values can be converted to strings.

  • The number of names taken from the specified table variable must match the number of rows of the input table.

Selected variables from T1, specified as the comma-separated pair consisting of 'DataVariables' and a character vector, cell array of character vectors, string array, positive integer, vector of positive integers, logical vector, or subscript object. The rows2vars function selects the variables specified by the value of 'DataVariables' and reorients only those variables to become the rows of T2. The remaining variables of T1 are discarded.

Rule for naming variables in T2, specified as the comma-separated pair consisting of 'VariableNamingRule' and either the value 'modify' or 'preserve'.

The values of 'VariableNamingRule' specify the following rules for naming variable in the output table or timetable.

Value of 'VariableNamingRule'

Rule

'modify' (default)

Modify names taken from the input table or timetable so that the corresponding variable names in the output are also valid MATLAB identifiers.

'preserve'

Preserve original names taken from the input table or timetable. The corresponding variable names in the output can have any Unicode® characters, including spaces and non-ASCII characters.

Note: In some cases, rows2vars must modify original names even when 'preserve' is the rule. Such cases include:

  • Duplicate names

  • Names that conflict with table dimension names

  • Names that conflict with a reserved name.

  • Names whose lengths exceed the value of namelengthmax.

Introduced in R2018a