Identity matrix

From Rosetta Code
Revision as of 18:37, 28 September 2024 by Tobega (talk | contribs) (→‎{{header|Tailspin}}: v0.5)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Task
Identity matrix
You are encouraged to solve this task according to the task description, using any language you may know.
Task

Build an   identity matrix   of a size known at run-time.


An identity matrix is a square matrix of size n × n,
where the diagonal elements are all 1s (ones),
and all the other elements are all 0s (zeroes).



Related tasks



11l

Translation of: Python
F identity_matrix(size)
   V matrix = [[0] * size] * size
   L(i) 0 .< size
      matrix[i][i] = 1
   R matrix

L(row) identity_matrix(3)
   print(row)
Output:
[1, 0, 0]
[0, 1, 0]
[0, 0, 1]

360 Assembly

*        Identity matrix           31/03/2017
INDENMAT CSECT
         USING  INDENMAT,R13       base register
         B      72(R15)            skip savearea
         DC     17F'0'             savearea
         STM    R14,R12,12(R13)    save previous context
         ST     R13,4(R15)         link backward
         ST     R15,8(R13)         link forward
         LR     R13,R15            set addressability
         L      R1,N               n
         MH     R1,N+2             n*n
         SLA    R1,2               *4
         ST     R1,LL              amount of storage required
         GETMAIN RU,LV=(R1)        allocate storage for matrix
         USING  DYNA,R11           make storage addressable
         LR     R11,R1             set addressability
         LA     R6,1               i=1
       DO WHILE=(C,R6,LE,N)        do i=1 to n
         LA     R7,1                 j=1
       DO WHILE=(C,R7,LE,N)          do j=1 to n
       IF CR,R6,EQ,R7 THEN             if i=j then
         LA     R2,1                     k=1
       ELSE     ,                      else
         LA     R2,0                     k=0
       ENDIF    ,                      endif
         LR     R1,R6                  i
         BCTR   R1,0                   -1
         MH     R1,N+2                 *n
         AR     R1,R7                  (i-1)*n+j
         BCTR   R1,0                   -1
         SLA    R1,2                   *4
         ST     R2,A(R1)               a(i,j)=k
         LA     R7,1(R7)               j++
       ENDDO    ,                    enddo j
         LA     R6,1(R6)             i++
       ENDDO    ,                  enddo i
         LA     R6,1               i=1
       DO WHILE=(C,R6,LE,N)        do i=1 to n
         LA     R10,PG               pgi=0
         LA     R7,1                 j=1
       DO WHILE=(C,R7,LE,N)          do j=1 to n
         LR     R1,R6                  i
         BCTR   R1,0                   -1
         MH     R1,N+2                 *n
         AR     R1,R7                  (i-1)*n+j
         BCTR   R1,0                   -1
         SLA    R1,2                   *4
         L      R2,A(R1)               a(i,j)
         XDECO  R2,XDEC                edit
         MVC    0(1,R10),XDEC+11       output
         LA     R10,1(R10)             pgi+=1
         LA     R7,1(R7)               j++
       ENDDO    ,                    enddo j
         XPRNT  PG,L'PG              print
         LA     R6,1(R6)             i++
       ENDDO    ,                  enddo i
         LA     R1,A               address to free
         LA     R2,LL              amount of storage to free
         FREEMAIN A=(R1),LV=(R2)   free allocated storage
         DROP   R11                drop register
         L      R13,4(0,R13)       restore previous savearea pointer
         LM     R14,R12,12(R13)    restore previous context
         XR     R15,R15            rc=0
         BR     R14                exit
NN       EQU    10                 parameter n  (90=>32K)
N        DC     A(NN)              n
LL       DS     F                  n*n*4
PG       DC     CL(NN)' '          buffer
XDEC     DS     CL12               temp
DYNA     DSECT
A        DS     F                  a(n,n)
         YREGS
         END    INDENMAT
Output:
1000000000
0100000000
0010000000
0001000000
0000100000
0000010000
0000001000
0000000100
0000000010
0000000001

Action!

PROC CreateIdentityMatrix(BYTE ARRAY mat,BYTE size)
  CARD pos
  BYTE i,j

  pos=0
  FOR i=1 TO size
  DO
    FOR j=1 TO size
    DO
      IF i=j THEN 
        mat(pos)=1
      ELSE
        mat(pos)=0
      FI
      pos==+1
    OD
  OD
RETURN

PROC PrintMatrix(BYTE ARRAY mat,BYTE size)
  CARD pos
  BYTE i,j,v

  pos=0
  FOR i=1 TO size
  DO
    FOR j=1 TO size
    DO
      v=mat(pos)
      IF j=size THEN
        PrintF("%I%E",v)
      ELSE
        PrintF("%I ",v)
      FI
      pos==+1
    OD
  OD
RETURN

PROC Main()
  BYTE size
  BYTE ARRAY mat(400)
  BYTE LMARGIN=$52,old

  old=LMARGIN
  LMARGIN=0 ;remove left margin on the screen

  DO
    Print("Get size of matrix [1-20] ")
    size=InputB()
  UNTIL size>=1 AND size<=20
  OD

  CreateIdentityMatrix(mat,size)
  PrintMatrix(mat,size)

  LMARGIN=old ;restore left margin on the screen
RETURN
Output:

Screenshot from Atari 8-bit computer

Get size of matrix [1-20] 13
1 0 0 0 0 0 0 0 0 0 0 0 0
0 1 0 0 0 0 0 0 0 0 0 0 0
0 0 1 0 0 0 0 0 0 0 0 0 0
0 0 0 1 0 0 0 0 0 0 0 0 0
0 0 0 0 1 0 0 0 0 0 0 0 0
0 0 0 0 0 1 0 0 0 0 0 0 0
0 0 0 0 0 0 1 0 0 0 0 0 0
0 0 0 0 0 0 0 1 0 0 0 0 0
0 0 0 0 0 0 0 0 1 0 0 0 0
0 0 0 0 0 0 0 0 0 1 0 0 0
0 0 0 0 0 0 0 0 0 0 1 0 0
0 0 0 0 0 0 0 0 0 0 0 1 0
0 0 0 0 0 0 0 0 0 0 0 0 1

Ada

When using floating point matrices in Ada 2005+ the function is defined as "Unit_Matrix" in Ada.Numerics.Generic_Real_Arrays. As a generic package it can work with user defined floating point types, or the predefined Short_Real_Arrays, Real_Arrays, and Long_Real_Arrays initializations. As seen below, the first indices of both dimensions can also be set since Ada array indices do not arbitrarily begin with a particular number.

--  As prototyped in the Generic_Real_Arrays specification:
--  function Unit_Matrix (Order : Positive; First_1, First_2 : Integer := 1) return Real_Matrix;
-- For the task:
mat : Real_Matrix := Unit_Matrix(5);

For prior versions of Ada, or non floating point types its back to basics:

type Matrix is array(Positive Range <>, Positive Range <>) of Integer;
mat : Matrix(1..5,1..5) := (others => (others => 0));
--  then after the declarative section:
for i in mat'Range(1) loop mat(i,i) := 1; end loop;

ALGOL 68

Works with: ALGOL 68 version Revision 1 - one extension to language used - PRAGMA READ - a non standard feature similar to C's #include directive.
Works with: ALGOL 68G version Any - tested with release algol68g-2.8.

Note: The generic vector and matrix code should be moved to a more generic page.

File: prelude/vector_base.a68

#!/usr/bin/a68g --script #
# -*- coding: utf-8 -*- #

# Define some generic vector initialisation and printing operations #

COMMENT REQUIRES:
  MODE SCAL = ~ # a scalar, eg REAL #;
  FORMAT scal fmt := ~;
END COMMENT

INT vec lwb := 1, vec upb := 0;
MODE VECNEW = [vec lwb:vec upb]SCAL; MODE VEC = REF VECNEW;
FORMAT vec fmt := $"("n(vec upb-vec lwb)(f(scal fmt)", ")f(scal fmt)")"$;

PRIO INIT = 1;

OP INIT = (VEC self, SCAL scal)VEC: (
  FOR col FROM LWB self TO UPB self DO self[col]:= scal OD;
  self
);

# ZEROINIT: defines the additive identity #
OP ZEROINIT = (VEC self)VEC: 
  self INIT SCAL(0);

OP REPR = (VEC self)STRING: (
  FILE f; STRING s; associate(f,s);
  vec lwb := LWB self; vec upb := UPB self;
  putf(f, (vec fmt, self)); close(f);
  s
);

SKIP

File: prelude/matrix_base.a68

# -*- coding: utf-8 -*- #

# Define some generic matrix initialisation and printing operations #

COMMENT REQUIRES:
  MODE SCAL = ~ # a scalar, eg REAL #;
  MODE VEC = []SCAL;
  FORMAT scal fmt := ~;
  et al.
END COMMENT

INT mat lwb := 1, mat upb := 0;
MODE MATNEW = [mat lwb:mat upb, vec lwb: vec upb]SCAL; MODE MAT = REF MATNEW;
FORMAT mat fmt = $"("n(vec upb-vec lwb)(f(vec fmt)","lx)f(vec fmt)")"l$;

PRIO DIAGINIT = 1;

OP INIT = (MAT self, SCAL scal)MAT: (
  FOR row FROM LWB self TO UPB self DO self[row,] INIT scal OD;
  self
);

# ZEROINIT: defines the additive identity #
OP ZEROINIT = (MAT self)MAT: 
  self INIT SCAL(0);

OP REPR = (MATNEW self)STRING: (
  FILE f; STRING s; associate(f,s);
  vec lwb := 2 LWB self; vec upb := 2 UPB self;
  mat lwb :=   LWB self; mat upb :=   UPB self;
  putf(f, (mat fmt, self)); close(f);
  s
);

OP DIAGINIT = (MAT self, VEC diag)MAT: (
  ZEROINIT self;
  FOR d FROM LWB diag TO UPB diag DO self[d,d]:= diag[d] OD;
# or alternatively using TORRIX ...
  DIAG self := diag;
#
  self
);

# ONEINIT: defines the multiplicative identity #
OP ONEINIT = (MAT self)MAT: (
  ZEROINIT self DIAGINIT (LOC[LWB self:UPB self]SCAL INIT SCAL(1))
# or alternatively using TORRIX ...
  (DIAG out) VECINIT SCAL(1)
#
);

SKIP

File: prelude/matrix_ident.a68

# -*- coding: utf-8 -*- #

PRIO IDENT = 9; # The same as I for COMPLex #

OP IDENT = (INT lwb, upb)MATNEW:
  ONEINIT LOC [lwb:upb,lwb:upb]SCAL;

OP IDENT = (INT upb)MATNEW: # default lwb is 1 #
  1 IDENT upb;

SKIP

File: prelude/matrix.a68

#!/usr/bin/a68g --script #
# -*- coding: utf-8 -*- #

PR READ "prelude/vector_base.a68" PR;
PR READ "prelude/matrix_base.a68" PR;
PR READ "prelude/matrix_ident.a68" PR;

SKIP

File: test/matrix_ident.a68

#!/usr/bin/a68g --script #
# -*- coding: utf-8 -*- #

MODE SCAL = REAL;
FORMAT scal fmt := $g(-3,1)$;

PR READ "prelude/matrix.a68" PR;

print(REPR IDENT 4)
Output:
((1.0, 0.0, 0.0, 0.0),
 (0.0, 1.0, 0.0, 0.0),
 (0.0, 0.0, 1.0, 0.0),
 (0.0, 0.0, 0.0, 1.0))

ALGOL W

begin
    % set m to an identity matrix of size s                                  %
    procedure makeIdentity( real    array m ( *, * )
                          ; integer value s
                          ) ;
        for i := 1 until s do begin
            for j := 1 until s do m( i, j ) := 0.0;
            m( i, i ) := 1.0
        end makeIdentity ;

    % test the makeIdentity procedure                                        %
    begin
        real array id5( 1 :: 5, 1 :: 5 );
        makeIdentity( id5, 5 );
        r_format := "A"; r_w := 6; r_d := 1; % set output format for reals   %
        for i := 1 until 5 do begin
            write();
            for j := 1 until 5 do writeon( id5( i, j ) )
        end for_i ;
    end text

end.

Amazing Hopper

#include <jambo.h>

Main
    Dim( 10,10 ) as eyes 'UMatrix'
    
    Printnl ' "UNIT MATRIX:\n", UMatrix '
    
    /* Classical method */
    
    Dim (10,10) as zeros (ZM)
    i=1
    Iterator ( ++i, #(i<=10), #( ZM[i,i]=1 ) )
    Printnl ' "UNIT MATRIX:\n", ZM '

    /* unit matrix non square*/
    Dim ( 5,8 ) as eyes 'rare unit matrix'
    
    Printnl ' "RARE UNIT MATRIX:\n", rare unit matrix '
End
Output:
UNIT MATRIX:
1,0,0,0,0,0,0,0,0,0
0,1,0,0,0,0,0,0,0,0
0,0,1,0,0,0,0,0,0,0
0,0,0,1,0,0,0,0,0,0
0,0,0,0,1,0,0,0,0,0
0,0,0,0,0,1,0,0,0,0
0,0,0,0,0,0,1,0,0,0
0,0,0,0,0,0,0,1,0,0
0,0,0,0,0,0,0,0,1,0
0,0,0,0,0,0,0,0,0,1

UNIT MATRIX:
1,0,0,0,0,0,0,0,0,0
0,1,0,0,0,0,0,0,0,0
0,0,1,0,0,0,0,0,0,0
0,0,0,1,0,0,0,0,0,0
0,0,0,0,1,0,0,0,0,0
0,0,0,0,0,1,0,0,0,0
0,0,0,0,0,0,1,0,0,0
0,0,0,0,0,0,0,1,0,0
0,0,0,0,0,0,0,0,1,0
0,0,0,0,0,0,0,0,0,1

RARE UNIT MATRIX:
1,0,0,0,0,0,0,0
0,1,0,0,0,0,0,0
0,0,1,0,0,0,0,0
0,0,0,1,0,0,0,0
0,0,0,0,1,0,0,0

APL

Making an identity matrix in APL involves the outer product of the equality function.

For a square matrix of 3:

    ∘.=3
1 0 0
0 1 0
0 0 1

For a function that makes an identity matrix:

    ID{∘.=}
    ID 5
1 0 0 0 0
0 1 0 0 0
0 0 1 0 0
0 0 0 1 0
0 0 0 0 1

An tacit function can be defined with one of the following equivalent lines:

    ID∘.=
    ID∘.=⍳

There is a more idomatic way however:

    ID{  ρ 1, ρ0}

AppleScript

--------------------- IDENTITY MATRIX ----------------------

-- identityMatrix :: Int -> [(0|1)]
on identityMatrix(n)
    set xs to enumFromTo(1, n)
    
    script row
        on |λ|(x)
            script col
                on |λ|(i)
                    if i = x then
                        1
                    else
                        0
                    end if
                end |λ|
            end script
            map(col, xs)
        end |λ|
    end script
    
    map(row, xs)
end identityMatrix


--------------------------- TEST ---------------------------
on run
    
    unlines(map(showList, ¬
        identityMatrix(5)))
    
end run


-------------------- GENERIC FUNCTIONS ---------------------

-- enumFromTo :: Int -> Int -> [Int]
on enumFromTo(m, n)
    if m  n then
        set lst to {}
        repeat with i from m to n
            set end of lst to i
        end repeat
        lst
    else
        {}
    end if
end enumFromTo


-- intercalate :: String -> [String] -> String
on intercalate(delim, xs)
    set {dlm, my text item delimiters} to ¬
        {my text item delimiters, delim}
    set s to xs as text
    set my text item delimiters to dlm
    s
end intercalate


-- map :: (a -> b) -> [a] -> [b]
on map(f, xs)
    tell mReturn(f)
        set lng to length of xs
        set lst to {}
        repeat with i from 1 to lng
            set end of lst to |λ|(item i of xs, i, xs)
        end repeat
        return lst
    end tell
end map


-- Lift 2nd class handler function into 1st class script wrapper 
-- mReturn :: Handler -> Script
on mReturn(f)
    if class of f is script then
        f
    else
        script
            property |λ| : f
        end script
    end if
end mReturn


-- showList :: [a] -> String
on showList(xs)
    "[" & intercalate(", ", map(my str, xs)) & "]"
end showList


-- str :: a -> String
on str(x)
    x as string
end str


-- unlines :: [String] -> String
on unlines(xs)
    -- A single string formed by the intercalation
    -- of a list of strings with the newline character.
    set {dlm, my text item delimiters} to ¬
        {my text item delimiters, linefeed}
    set s to xs as text
    set my text item delimiters to dlm
    s
end unlines
Output:
[1, 0, 0, 0, 0]
[0, 1, 0, 0, 0]
[0, 0, 1, 0, 0]
[0, 0, 0, 1, 0]
[0, 0, 0, 0, 1]

Simple alternative

on indentityMatrix(n)
    set digits to {}
    set m to n - 1
    repeat (n + m) times
        set end of digits to 0
    end repeat
    set item n of digits to 1
    
    set matrix to {}
    repeat with i from n to 1 by -1
        set end of matrix to items i thru (i + m) of digits
    end repeat
    
    return matrix
end indentityMatrix

return indentityMatrix(5)
Output:
{{1, 0, 0, 0, 0}, {0, 1, 0, 0, 0}, {0, 0, 1, 0, 0}, {0, 0, 0, 1, 0}, {0, 0, 0, 0, 1}}

Arturo

identityM: function [n][
    result: array.of: @[n n] 0
    loop 0..dec n 'i -> result\[i]\[i]: 1
    return result
]

loop 4..6 'sz [
    print sz
    loop identityM sz => print
    print ""
]
Output:
4
1 0 0 0 
0 1 0 0 
0 0 1 0 
0 0 0 1 

5
1 0 0 0 0 
0 1 0 0 0 
0 0 1 0 0 
0 0 0 1 0 
0 0 0 0 1 

6
1 0 0 0 0 0 
0 1 0 0 0 0 
0 0 1 0 0 0 
0 0 0 1 0 0 
0 0 0 0 1 0 
0 0 0 0 0 1

BASIC

Applesoft BASIC

100 INPUT "MATRIX SIZE:"; SIZE%
110 GOSUB 200"IDENTITYMATRIX
120 FOR R = 0 TO SIZE%
130     FOR C = 0 TO SIZE%
140         LET S$ = CHR$(13)
150         IF C < SIZE% THEN S$ = " "
160         PRINT IM(R, C) S$; : NEXT C, R
170 END

200 REMIDENTITYMATRIX SIZE%
210 LET SIZE% = SIZE% - 1
220 DIM IM(SIZE%, SIZE%)
230 FOR I = 0 TO SIZE%
240     LET IM(I, I) = 1 : NEXT I
250 RETURN :IM

Commodore BASIC

Translation of: Applesoft BASIC
Works with: Commodore BASIC version 2.0
100 INPUT "MATRIX SIZE:"; SIZE
110 GOSUB 200: REM IDENTITYMATRIX
120 FOR R = 0 TO SIZE
130 FOR C = 0 TO SIZE
140 S$ = CHR$(13)
150 IF C < SIZE THEN S$ = ""
160 PRINT MID$(STR$(IM(R, C)),1)S$;:REM MID$ STRIPS LEADING SPACES
170 NEXT C, R
180 END
190 REM *******************************
200 REM IDENTITYMATRIX SIZE%
210 SIZE = SIZE - 1
220 DIM IM(SIZE, SIZE)
230 FOR I = 0 TO SIZE
240     IM(I, I) = 1
250 NEXT I
260 RETURN

QBasic

Works with: QBasic version 1.1
Works with: QuickBasic version 4.5
Translation of: IS-BASIC
SUB inicio(identity())
    FOR i = LBOUND(identity,1) TO UBOUND(identity,1)
        FOR j = LBOUND(identity,2) TO UBOUND(identity,2)
            LET identity(i,j) = 0
        NEXT j
        LET identity(i,i) = 1
    NEXT i
END SUB

SUB mostrar(identity())
    FOR i = LBOUND(identity,1) TO UBOUND(identity,1)
        FOR j = LBOUND(identity,2) TO UBOUND(identity,2)
            PRINT identity(i,j);
        NEXT j
        PRINT
    NEXT i
END SUB

DO
    INPUT "Enter size of matrix "; n
LOOP UNTIL n > 0

DIM identity(1 TO n, 1 TO n)

CALL inicio(identity())
CALL mostrar(identity())

BASIC256

Translation of: FreeBASIC
arraybase 1
do
    input "Enter size of matrix: ", n
until n > 0

dim identity(n, n) fill 0  #we fill everything with 0

# enter 1s in diagonal elements
for i =  1 to n
    identity[i, i] = 1
next i

# print identity matrix if n < 40
print

if n < 40 then
    for i = 1 to n
        for j = 1 to n
            print identity[i, j];
        next j
        print
    next i
else
    print "Matrix is too big to display on 80 column console"
end if
Output:
Same as FreeBASIC entry.

XBasic

Works with: Windows XBasic
Translation of: FreeBASIC
PROGRAM  "Identity matrix"
VERSION	"0.0000"

DECLARE FUNCTION  Entry ()

FUNCTION  Entry ()
  DO
    n = SBYTE(INLINE$("Enter size of matrix: "))
  LOOP UNTIL n > 0

  DIM identity[n, n] '' all zero by default

  ' enter 1s in diagonal elements
  FOR i =  1 TO n
    identity[i, i] = 1
  NEXT i

  ' print identity matrix if n < 40
  PRINT

  IF n < 40 THEN
    FOR i = 1 TO n
      FOR j = 1 TO n
        PRINT identity[i, j];
      NEXT j
        PRINT
    NEXT i
  ELSE
    PRINT "Matrix is too big to display on 80 column console"
  END IF

END FUNCTION
END PROGRAM
Output:
Same as FreeBASIC entry.

Yabasic

Translation of: FreeBASIC
repeat
    input "Enter size of matrix: " n
until n > 0 

dim identity(n, n) // all zero by default

// enter 1s in diagonal elements
for i =  1 to n
    identity(i, i) = 1
next i

// print identity matrix if n < 40 
print

if n < 40 then
    for i = 1 to n
        for j = 1 to n
            print identity(i, j); 
        next j
        print
    next i
else 
    print "Matrix is too big to display on 80 column console"
end if
Output:
Same as FreeBASIC entry.

ATS

(* ****** ****** *)
//
// How to compile:
//
// patscc -DATS_MEMALLOC_LIBC -o idmatrix idmatrix.dats
//
(* ****** ****** *)
//
#include
"share/atspre_staload.hats"
//
(* ****** ****** *)

extern
fun
idmatrix{n:nat}(n: size_t(n)): matrixref(int, n, n)
implement
idmatrix(n) =
matrixref_tabulate_cloref<int> (n, n, lam(i, j) => bool2int0(i = j))

(* ****** ****** *)

implement
main0 () =
{
//
val N = 5
//
val M = idmatrix(i2sz(N))
val () = fprint_matrixref_sep (stdout_ref, M, i2sz(N), i2sz(N), " ", "\n")
val () = fprint_newline (stdout_ref)
//
} (* end of [main0] *)

AutoHotkey

msgbox % Clipboard := I(6)
return

I(n){
    r := "--`n" , s := " "
    loop % n
    {
        k := A_index , r .= "|  "
        loop % n
            r .= A_index=k ? "1, " : "0, "
        r := RTrim(r, " ,") , r .= "  |`n"
    }
    loop % 4*n
        s .= " "
    return Rtrim(r,"`n") "`n" s "--"
}
Output:
--
|  1, 0, 0, 0, 0, 0  |
|  0, 1, 0, 0, 0, 0  |
|  0, 0, 1, 0, 0, 0  |
|  0, 0, 0, 1, 0, 0  |
|  0, 0, 0, 0, 1, 0  |
|  0, 0, 0, 0, 0, 1  |
                    --

AWK

# syntax: GAWK -f IDENTITY_MATRIX.AWK size
BEGIN {
    size = ARGV[1]
    if (size !~ /^[0-9]+$/) {
      print("size invalid or missing from command line")
      exit(1)
    }
    for (i=1; i<=size; i++) {
      for (j=1; j<=size; j++) {
        x = (i == j) ? 1 : 0
        printf("%2d",x) # print
        arr[i,j] = x # build
      }
      printf("\n")
    }
    exit(0)
}
Output:

for command

GAWK -f IDENTITY_MATRIX.AWK 5
 1 0 0 0 0
 0 1 0 0 0
 0 0 1 0 0
 0 0 0 1 0
 0 0 0 0 1

Bash

for i in `seq $1`;do printf '%*s\n' $1|tr ' ' '0'|sed "s/0/1/$i";done
Output:

for command

./scriptname 5
 1 0 0 0 0
 0 1 0 0 0
 0 0 1 0 0
 0 0 0 1 0
 0 0 0 0 1

BBC BASIC

      INPUT "Enter size of matrix: " size%
      PROCidentitymatrix(size%, im())
      FOR r% = 0 TO size%-1
        FOR c% = 0 TO size%-1
          PRINT im(r%, c%),;
        NEXT
        PRINT
      NEXT r%
      END
      
      DEF PROCidentitymatrix(s%, RETURN m())
      LOCAL i%
      DIM m(s%-1, s%-1)
      FOR i% = 0 TO s%-1
        m(i%,i%) = 1
      NEXT
      ENDPROC

Beads

beads 1 program 'Identity matrix'

var
	id : array^2 of num
	n = 5
	
calc main_init
	loop from:1 to:n index:i
		loop from:1 to:n index:j
			id[i,j] = 1 if i == j else 0
Output:
1 0 0 0 0
0 1 0 0 0
0 0 1 0 0
0 0 0 1 0
0 0 0 0 1

Burlesque

Neither very elegant nor short but it'll do

blsq ) 6 -.^^0\/r@\/'0\/.*'1+]\/{\/{rt}\/E!XX}x/+]m[sp
1 0 0 0 0 0
0 1 0 0 0 0
0 0 1 0 0 0
0 0 0 1 0 0
0 0 0 0 1 0
0 0 0 0 0 1

The example above uses strings to generate the identity matrix. If you need a matrix with real numbers (Integers) then use:

6hd0bx#a.*\[#a.*0#a?dr@{(D!)\/1\/^^bx\/[+}m[e!

Shorter alternative:

blsq ) 6 ^^^^10\/**XXcy\/co.+sp

BQN

⍝ Using table
Eye ← =⌜˜∘↕
•Show Eye 3

⍝ Using reshape
Eye1 ← {𝕩‿𝕩⥊1∾𝕩⥊0}
Eye1 5
┌─       
╵ 1 0 0  
  0 1 0  
  0 0 1  
        ┘
┌─           
╵ 1 0 0 0 0  
  0 1 0 0 0  
  0 0 1 0 0  
  0 0 0 1 0  
  0 0 0 0 1  
            ┘

Eye generates an identity matrix using a table of equality for [0,n).

Eye1 reshapes a boolean vector to generate the matrix.

Try it here!

C

#include <stdlib.h>
#include <stdio.h>
int main(int argc, char** argv) {
   if (argc < 2) {
      printf("usage: identitymatrix <number of rows>\n");
      exit(EXIT_FAILURE);
   }
   int rowsize = atoi(argv[1]);
   if (rowsize < 0) {
      printf("Dimensions of matrix cannot be negative\n");
      exit(EXIT_FAILURE);
   }
   int numElements = rowsize * rowsize;
   if (numElements < rowsize) {
      printf("Squaring %d caused result to overflow to %d.\n", rowsize, numElements);
      abort();
   }
   int** matrix = calloc(numElements, sizeof(int*));
   if (!matrix) {
      printf("Failed to allocate %d elements of %ld bytes each\n", numElements, sizeof(int*));
      abort();
   }
   for (unsigned int row = 0;row < rowsize;row++) {
      matrix[row] = calloc(numElements, sizeof(int));
      if (!matrix[row]) {
         printf("Failed to allocate %d elements of %ld bytes each\n", numElements, sizeof(int));
         abort();
      }
      matrix[row][row] = 1;
   }
   printf("Matrix is: \n");
   for (unsigned int row = 0;row < rowsize;row++) {
      for (unsigned int column = 0;column < rowsize;column++) {
         printf("%d ", matrix[row][column]);
      }
      printf("\n");
   }
}

C#

using System;
using System.Linq;

namespace IdentityMatrix
{
    class Program
    {
        static void Main(string[] args)
        {
            if (args.Length != 1)
            {
                Console.WriteLine("Requires exactly one argument");
                return;
            }
            int n;
            if (!int.TryParse(args[0], out n))
            {
                Console.WriteLine("Requires integer parameter");
                return;
            }

            var identity =
                Enumerable.Range(0, n).Select(i => Enumerable.Repeat(0, n).Select((z,j) => j == i ? 1 : 0).ToList()).ToList();
            foreach (var row in identity)
            {
                foreach (var elem in row)
                {
                    Console.Write(" " + elem);
                }
                Console.WriteLine();
            }
            Console.ReadKey();
        }
    }
}
Output:
 1 0 0 0 0 0
 0 1 0 0 0 0
 0 0 1 0 0 0
 0 0 0 1 0 0
 0 0 0 0 1 0
 0 0 0 0 0 1

C++

Library: STL
template<class T>
class matrix
{
public:
    matrix( unsigned int nSize ) : 
      m_oData(nSize * nSize, 0), m_nSize(nSize) {}

      inline T& operator()(unsigned int x, unsigned int y)
      {
          return m_oData[x+m_nSize*y];
      }

      void identity()
      {
          int nCount = 0;
          int nStride = m_nSize + 1;
          std::generate( m_oData.begin(), m_oData.end(), 
              [&]() { return !(nCount++%nStride); } );
      }

      inline unsigned int size() { return m_nSize; }

private:
    std::vector<T>    m_oData;
    unsigned int      m_nSize;
};

int main()
{
    int nSize;
    std::cout << "Enter matrix size (N): ";
    std::cin >> nSize;

    matrix<int> oMatrix( nSize );

    oMatrix.identity();

    for ( unsigned int y = 0; y < oMatrix.size(); y++ )
    {
        for ( unsigned int x = 0; x < oMatrix.size(); x++ )
        {
            std::cout << oMatrix(x,y) << " ";
        }
        std::cout << std::endl;
    }
    return 0;
}
Library: boost
#include <boost/numeric/ublas/matrix.hpp>

int main()
{
    using namespace boost::numeric::ublas;
    
    int nSize;
    std::cout << "Enter matrix size (N): ";
    std::cin >> nSize;

    identity_matrix<int> oMatrix( nSize );

    for ( unsigned int y = 0; y < oMatrix.size2(); y++ )
    {
        for ( unsigned int x = 0; x < oMatrix.size1(); x++ )
        {
            std::cout << oMatrix(x,y) << " ";
        }
        std::cout << std::endl;
    }

    return 0;
}
Output:
Enter matrix size (N): 5
1 0 0 0 0
0 1 0 0 0
0 0 1 0 0
0 0 0 1 0
0 0 0 0 1

Clio

fn identity-matrix n:
  [0:n] -> * fn i:
    [0:n] -> * if = i: 1
                 else: 0

5 -> identity-matrix -> * print

Clojure

Translation of: PicoLisp

The (vec ) function in the following solution is with respect to vector matrices. If dealing with normal lists matrices (e.g.

 '( (0 1) (2 3) )

, then care to remove the vec function.

(defn identity-matrix [n]
  (let [row (conj (repeat (dec n) 0) 1)]
    (vec
      (for [i (range 1 (inc n))]
        (vec 
          (reduce conj (drop i row ) (take i row)))))))
Output:
=> (identity-matrix 5)
[[1 0 0 0 0] [0 1 0 0 0] [0 0 1 0 0] [0 0 0 1 0] [0 0 0 0 1]]

The following is a more idomatic definition that utilizes infinite lists and cycling.

(defn identity-matrix [n]
  (take n 
    (partition n (dec n) 
                         (cycle (conj (repeat (dec n) 0) 1)))))

Common Lisp

Common Lisp provides multi-dimensional arrays.

(defun make-identity-matrix (n)
  (let ((array (make-array (list n n) :initial-element 0)))
    (loop for i below n do (setf (aref array i i) 1))
    array))
Output:
* (make-identity-matrix 5)
#2A((1 0 0 0 0) (0 1 0 0 0) (0 0 1 0 0) (0 0 0 1 0) (0 0 0 0 1))
(defun identity-matrix (n)
  (loop for a from 1 to n
        collect (loop for e from 1 to n 
                      if (= a e) collect 1
                      else collect 0)))
Output:
> (identity-matrix 5)
((1 0 0 0 0) (0 1 0 0 0) (0 0 1 0 0) (0 0 0 1 0) (0 0 0 0 1))

Component Pascal

BlackBox Component Builder

MODULE Algebras;
IMPORT StdLog,Strings;

TYPE
	Matrix = POINTER TO ARRAY OF ARRAY OF INTEGER;
	
PROCEDURE NewIdentityMatrix(n: INTEGER): Matrix;
VAR
	m: Matrix;
	i: INTEGER;
BEGIN
	NEW(m,n,n);
	FOR i := 0 TO n - 1 DO
		m[i,i] := 1;
	END;
	RETURN m;
END NewIdentityMatrix;

PROCEDURE Show(m: Matrix);
VAR
	i,j: INTEGER;
BEGIN
	FOR i := 0 TO LEN(m,0) - 1 DO
		FOR j := 0 TO LEN(m,1) - 1 DO
			StdLog.Int(m[i,j]);
		END;
		StdLog.Ln
	END
END Show;

PROCEDURE Do*;
BEGIN
	Show(NewIdentityMatrix(5));
END Do;
END Algebras.

Execute: ^Q Algebras.Do

Output:
 1 0 0 0 0
 0 1 0 0 0
 0 0 1 0 0
 0 0 0 1 0
 0 0 0 0 1

D

import std.traits;

T[][] matId(T)(in size_t n) pure nothrow if (isAssignable!(T, T)) {
    auto Id = new T[][](n, n);

    foreach (r, row; Id) {
        static if (__traits(compiles, {row[] = 0;})) {
            row[] = 0; // vector op doesn't work with T = BigInt
            row[r] = 1;
        } else {
            foreach (c; 0 .. n)
                row[c] = (c == r) ? 1 : 0;
        }
    }

    return Id;
}

void main() {
    import std.stdio, std.bigint;
    enum form = "[%([%(%s, %)],\n %)]]";

    immutable id1 = matId!real(5);
    writefln(form ~ "\n", id1);

    immutable id2 = matId!BigInt(3);
    writefln(form ~ "\n", id2);

    // auto id3 = matId!(const int)(2); // cant't compile
}
Output:
[[1, 0, 0, 0, 0],
 [0, 1, 0, 0, 0],
 [0, 0, 1, 0, 0],
 [0, 0, 0, 1, 0],
 [0, 0, 0, 0, 1]]

[[1, 0, 0],
 [0, 1, 0],
 [0, 0, 1]]

Delphi

program IdentityMatrix;

// Modified from the Pascal version

{$APPTYPE CONSOLE}

var
  matrix: array of array of integer;
  n, i, j: integer;
  
begin
  write('Size of matrix: ');
  readln(n);
  setlength(matrix, n, n);

  for i := 0 to n - 1 do
    matrix[i,i] := 1;
    
  for i := 0 to n - 1 do
  begin
    for j := 0 to n - 1 do
      write (matrix[i,j], ' ');
    writeln;
  end;
end.
Output:
Size of matrix: 5
1 0 0 0 0 
0 1 0 0 0 
0 0 1 0 0 
0 0 0 1 0 
0 0 0 0 1 

EasyLang

proc idmat lng . mat[][] .
   len mat[][] lng
   for i to lng
      len mat[i][] lng
      mat[i][i] = 1
   .
.
idmat 4 m[][]
print m[][]

Eiffel

class
	APPLICATION

inherit
	ARGUMENTS

create
	make

feature {NONE} -- Initialization

	make
			-- Run application.
		local
		    dim : INTEGER -- Dimension of the identity matrix
		do
		    from dim := 1 until dim > 10 loop
		    	print_matrix( identity_matrix(dim) )
				dim := dim + 1
				io.new_line
			end

		end

feature -- Access

	identity_matrix(dim : INTEGER) : ARRAY2[REAL_64]

		require
			dim > 0
		local
			matrix : ARRAY2[REAL_64]
			i : INTEGER
		do

			create matrix.make_filled (0.0, dim, dim)
			from i := 1 until i > dim loop
				matrix.put(1.0, i, i)
				i := i + 1
			end

			Result := matrix
		end

	print_matrix(matrix : ARRAY2[REAL_64])
		local
			i, j : INTEGER
		do
			from i := 1 until i > matrix.height loop
				print("[ ")
				from j := 1 until j > matrix.width loop
					print(matrix.item (i, j))
					print(" ")
					j := j + 1
				end
				print("]%N")
				i := i + 1
			end
		end

end
Output:
[ 1 0 0 0 0 0 0 0 0 0 ]
[ 0 1 0 0 0 0 0 0 0 0 ]
[ 0 0 1 0 0 0 0 0 0 0 ]
[ 0 0 0 1 0 0 0 0 0 0 ]
[ 0 0 0 0 1 0 0 0 0 0 ]
[ 0 0 0 0 0 1 0 0 0 0 ]
[ 0 0 0 0 0 0 1 0 0 0 ]
[ 0 0 0 0 0 0 0 1 0 0 ]
[ 0 0 0 0 0 0 0 0 1 0 ]
[ 0 0 0 0 0 0 0 0 0 1 ]

Elena

ELENA 6.x :

import extensions;
import system'routines;
import system'collections;
 
public program()
{
    var n := console.write("Enter the matrix size:").readLine().toInt();
 
    var identity := new Range(0, n).selectBy::(i => new Range(0,n).selectBy::(j => (i == j).iif(1,0) ).summarize(new ArrayList())) 
                         .summarize(new ArrayList());
 
    identity.forEach::
        (row) { console.printLine(row.asEnumerable()) }
}
Output:
Enter the matrix size:3
1,0,0
0,1,0
0,0,1

Elixir

defmodule Matrix do
  def identity(n) do
    Enum.map(0..n-1, fn i ->
      for j <- 0..n-1, do: (if i==j, do: 1, else: 0)
    end)
  end
end

IO.inspect Matrix.identity(5)
Output:
[[1, 0, 0, 0, 0], [0, 1, 0, 0, 0], [0, 0, 1, 0, 0], [0, 0, 0, 1, 0],
 [0, 0, 0, 0, 1]]

Erlang

%% Identity Matrix in Erlang for the Rosetta Code Wiki.
%% Implemented by Arjun Sunel

-module(identity_matrix).
-export([square_matrix/2 , identity/1]).

square_matrix(Size, Elements) ->
    [[Elements(Column, Row) || Column <- lists:seq(1, Size)] || Row <- lists:seq(1, Size)].

identity(Size) ->
    square_matrix(Size, fun(Column, Row) -> case Column of Row -> 1; _ -> 0 end end).

ERRE

PROGRAM IDENTITY

!$DYNAMIC
DIM A[0,0]

BEGIN
  PRINT(CHR$(12);) ! CLS
  INPUT("Matrix size",N%)
  !$DIM A[N%,N%]
  FOR I%=1 TO N% DO
    A[I%,I%]=1
  END FOR
! print matrix
  FOR I%=1 TO N% DO
    FOR J%=1 TO N% DO
      WRITE("###";A[I%,J%];)
    END FOR
    PRINT
  END FOR
END PROGRAM

Euler Math Toolbox

function IdentityMatrix(n)
  $  X:=zeros(n,n);
  $  for i=1 to n 
  $    X[i,i]:=1;
  $  end;
  $  return X;
  $endfunction
>function IdentityMatrix (n:index)
$  return setdiag(zeros(n,n),0,1);
$endfunction
>id(5)

Excel

LAMBDA

Excel can lift functions over scalar values to functions over two-dimensional arrays.

Here we bind the name IDMATRIX to a lambda expression in the Name Manager of the Excel WorkBook:

(See LAMBDA: The ultimate Excel worksheet function)

IDMATRIX
=LAMBDA(n,
    LET(
        ixs, SEQUENCE(n, n, 0, 1),
        x, MOD(ixs, n),
        y, QUOTIENT(ixs, n),

        IF(x = y,
            1,
            0
        )
    )
)
Output:

The formula in cell B2 below populates the B2:F6 grid:

fx =IDMATRIX(A2)
A B C D E F
1 N Identity matrix
2 5 1 0 0 0 0
3 0 1 0 0 0
4 0 0 1 0 0
5 0 0 0 1 0
6 0 0 0 0 1
7
8 3 1 0 0
9 0 1 0
10 0 0 1

F#

Builds a 2D matrix with the given square size.

let ident n = Array2D.init n n (fun i j -> if i = j then 1 else 0)
Output:
ident 10;;
val it : int [,] = [[1; 0; 0; 0; 0; 0; 0; 0; 0; 0]
                    [0; 1; 0; 0; 0; 0; 0; 0; 0; 0]
                    [0; 0; 1; 0; 0; 0; 0; 0; 0; 0]
                    [0; 0; 0; 1; 0; 0; 0; 0; 0; 0]
                    [0; 0; 0; 0; 1; 0; 0; 0; 0; 0]
                    [0; 0; 0; 0; 0; 1; 0; 0; 0; 0]
                    [0; 0; 0; 0; 0; 0; 1; 0; 0; 0]
                    [0; 0; 0; 0; 0; 0; 0; 1; 0; 0]
                    [0; 0; 0; 0; 0; 0; 0; 0; 1; 0]
                    [0; 0; 0; 0; 0; 0; 0; 0; 0; 1]]

Factor

Works with: Factor version 0.99 2020-07-03
USING: math.matrices prettyprint ;

6 <identity-matrix> .
Output:
{
    { 1 0 0 0 0 0 }
    { 0 1 0 0 0 0 }
    { 0 0 1 0 0 0 }
    { 0 0 0 1 0 0 }
    { 0 0 0 0 1 0 }
    { 0 0 0 0 0 1 }
}

FBSL

FBSL's BASIC layer can easily manipulate square matrices of arbitrary sizes and data types in ways similar to e.g. BBC BASIC or OxygenBasic as shown elsewhere on this page. But FBSL has also an extremely fast built-in single-precision vector2f/3f/4f, plane4f, quaternion4f, and matrix4f math library totaling 150 functions and targeting primarily 3D rendering tasks:

#APPTYPE CONSOLE

TYPE M4F ' Matrix 4F

m11 AS SINGLE
m12 AS SINGLE
m13 AS SINGLE
m14 AS SINGLE
m21 AS SINGLE
m22 AS SINGLE
m23 AS SINGLE
m24 AS SINGLE
m31 AS SINGLE
m32 AS SINGLE
m33 AS SINGLE
m34 AS SINGLE
m41 AS SINGLE
m42 AS SINGLE
m43 AS SINGLE
m44 AS SINGLE

END TYPE

DIM m AS M4F ' DIM zeros out any variable automatically

PRINT "Matrix 'm' is identity: ", IIF(MATRIXISIDENTITY(@m), "TRUE", "FALSE") ' is matrix an identity?
MATRIXIDENTITY(@m) ' set matrix to identity
PRINT "Matrix 'm' is identity: ", IIF(MATRIXISIDENTITY(@m), "TRUE", "FALSE") ' is matrix an identity?

PAUSE

Output:

Matrix 'm' is identity: FALSE
Matrix 'm' is identity: TRUE

Press any key to continue...

Fermat

Func Identity(n)=Array id[n,n];[id]:=[1].

Identity(7)
[id]
Output:
[[  1,  0,  0,  0,  0,  0,  0, `
    0,  1,  0,  0,  0,  0,  0, `
    0,  0,  1,  0,  0,  0,  0, `
    0,  0,  0,  1,  0,  0,  0, `
    0,  0,  0,  0,  1,  0,  0, `
    0,  0,  0,  0,  0,  1,  0, `
    0,  0,  0,  0,  0,  0,  1   ]]

Forth

Works with: gforth version 0.7.9_20170308
S" fsl-util.fs" REQUIRED

: build-identity ( 'p n -- 'p )  \ make an NxN identity matrix
  0 DO
    I 1+ 0 DO
      I J = IF  1.0E0 DUP I J }} F!
      ELSE
        0.0E0 DUP J I }} F!
        0.0E0 DUP I J }} F!
      THEN
    LOOP
  LOOP ;

6 6 float matrix a{{
a{{ 6 build-identity
6 6 a{{ }}fprint

Fortran

Works with: Fortran version 95
program identitymatrix

  real, dimension(:, :), allocatable :: I
  character(len=8) :: fmt
  integer :: ms, j

  ms = 10   ! the desired size

  allocate(I(ms,ms))
  I = 0                           ! Initialize the array.
  forall(j = 1:ms) I(j,j) = 1     ! Set the diagonal.

  ! I is the identity matrix, let's show it:

  write (fmt, '(A,I2,A)') '(', ms, 'F6.2)'
  ! if you consider to have used the (row, col) convention, 
  ! the following will print the transposed matrix (col, row)
  ! but I' = I, so it's not important here  
  write (*, fmt) I(:,:)

  deallocate(I)

end program identitymatrix

Notorious trick

The objective is to do the assignment in one fell swoop, rather than separately setting the 0 values and the 1 values. It works because, with integer arithmetic, the only way that both i/j and j/i are one is when they are equal - thus one on the diagonal elements, and zero elsewhere because either i < j so that i/j = 0, or i > j so that j/i = 0. While this means two divides and a multiply per element instead of simply transferring a constant, the constraint on speed is likely to be the limited bandwidth from cpu to memory. The expression's code would surely fit in the cpu's internal memory, and registers would be used for the variables.

      Program Identity
      Integer N
      Parameter (N = 666)
      Real A(N,N)
      Integer i,j

      ForAll(i = 1:N, j = 1:N) A(i,j) = (i/j)*(j/i)
      
      END

The ForAll statement is a feature of F90, and carries the implication that the assignments may be done in any order, even "simultaneously" (as with multiple cpus), plus that all RHS values are calculated before any LHS part receives a value - not relevant here since the RHS makes no reference to items altered in the LHS. Earlier Fortran compilers lack this statement and so one must use explicit DO-loops:

      DO 1 I = 1,N
        DO 1 J = 1,N
    1     A(I,J) = (I/J)*(J/I)

Array assignment statements are also a feature of F90 and later.

An alternative might be a simpler logical expression testing i = j except that the numerical values for true and false on a particular system may well not be 1 and 0 but (for instance, via Compaq F90/95 on Windows XP) 0 and -1 instead. On an IBM 390 mainframe, pl/i and Fortran used different values. The Burroughs 6700 inspected the low-order bit only, with the intriguing result that odd integers would be deemed true and even false. Integer arithmetic can't be relied upon across languages either, because in pl/i, integer division doesn't truncate.

FreeBASIC

' FB 1.05.0 Win64

Dim As Integer n

Do
  Input "Enter size of matrix "; n
Loop Until n > 0 

Dim identity(1 To n, 1 To n) As Integer '' all zero by default

' enter 1s in diagonal elements
For i As Integer =  1 To n
  identity(i, i) = 1
Next

' print identity matrix if n < 40 
Print

If n < 40 Then
  For i As Integer = 1 To n
    For j As Integer = 1 To n
      Print identity(i, j); 
    Next j
    Print
  Next i
Else 
  Print "Matrix is too big to display on 80 column console"
End If

Print
Print "Press any key to quit"
Sleep

Sample input/output

Output:
Enter size of matrix ? 5

 1 0 0 0 0
 0 1 0 0 0
 0 0 1 0 0
 0 0 0 1 0
 0 0 0 0 1

Frink

This does not use the Matrix.frink library, which has methods to create an identity matrix, but shows how to build a "raw" identity matrix as a two-dimensional array, and shows how to nicely format it using built-in routines.

n = parseInt[input["Enter matrix dimension as an integer: "]]
println[formatMatrix[makeArray[[n, n], {|a,b| a==b ? 1 : 0}]]]
Output:
Enter matrix dimension as an integer:  3
┌       ┐
│1  0  0│
│       │
│0  1  0│
│       │
│0  0  1│
└       ┘

FunL

def identity( n ) = vector( n, n, \r, c -> if r == c then 1 else 0 )

println( identity(3) )
Output:
((1, 0, 0), (0, 1, 0), (0, 0, 1))

Fōrmulæ

Fōrmulæ programs are not textual, visualization/edition of programs is done showing/manipulating structures but not text. Moreover, there can be multiple visual representations of the same program. Even though it is possible to have textual representation —i.e. XML, JSON— they are intended for storage and transfer purposes more than visualization and edition.

Programs in Fōrmulæ are created/edited online in its website.

In this page you can see and run the program(s) related to this task and their results. You can also change either the programs or the parameters they are called with, for experimentation, but remember that these programs were created with the main purpose of showing a clear solution of the task, and they generally lack any kind of validation.

Solution

Test cases

FutureBasic

include "NSLog.incl"

local fn IdentityMatrix( n as NSInteger ) as CFStringRef
NSInteger i, j
CFMutableArrayRef tempArr = fn MutableArrayWithCapacity( n )
CFMutableStringRef mutStr = fn MutableStringWithCapacity( 0 )

for i = 0 to n - 1
MutableArrayRemoveAllObjects( tempArr )
for j = 0 to n - 1
MutableArrayInsertObjectAtIndex( tempArr, @"0", j )
next
MutableArrayReplaceObjectAtIndex( tempArr, @"1", i )
MutableStringAppendString( mutStr, fn ArrayComponentsJoinedByString( tempArr, @" " ) )
MutableStringAppendString( mutStr, @"\n" )
next
end fn = fn StringWithString( mutStr )

NSLog( @"3:\n%@", fn IdentityMatrix( 3 ) )
NSLog( @"5:\n%@", fn IdentityMatrix( 5 ) )
NSLog( @"7:\n%@", fn IdentityMatrix( 7 ) )
NSLog( @"9:\n%@", fn IdentityMatrix( 9 ) )

HandleEvents
Output:
3:
1 0 0
0 1 0
0 0 1

5:
1 0 0 0 0
0 1 0 0 0
0 0 1 0 0
0 0 0 1 0
0 0 0 0 1

7:
1 0 0 0 0 0 0
0 1 0 0 0 0 0
0 0 1 0 0 0 0
0 0 0 1 0 0 0
0 0 0 0 1 0 0
0 0 0 0 0 1 0
0 0 0 0 0 0 1

9:
1 0 0 0 0 0 0 0 0
0 1 0 0 0 0 0 0 0
0 0 1 0 0 0 0 0 0
0 0 0 1 0 0 0 0 0
0 0 0 0 1 0 0 0 0
0 0 0 0 0 1 0 0 0
0 0 0 0 0 0 1 0 0
0 0 0 0 0 0 0 1 0
0 0 0 0 0 0 0 0 1


GAP

# Built-in
IdentityMat(3);

# One can also specify the base ring
IdentityMat(3, Integers mod 10);


Go

Library gonum/mat

package main

import (
    "fmt"

    "gonum.org/v1/gonum/mat"
)

func eye(n int) *mat.Dense {
    m := mat.NewDense(n, n, nil)
    for i := 0; i < n; i++ {
        m.Set(i, i, 1)
    }
    return m
}

func main() {
    fmt.Println(mat.Formatted(eye(3)))
}
Output:
⎡1  0  0⎤
⎢0  1  0⎥
⎣0  0  1⎦

Library go.matrix

A somewhat earlier matrix library for Go.

package main

import (
    "fmt"

    mat "github.com/skelterjohn/go.matrix"
)

func main() {
    fmt.Println(mat.Eye(3))
}
Output:
{1, 0, 0,
 0, 1, 0,
 0, 0, 1}

From scratch

Simplest: A matrix as a slice of slices, allocated separately.

package main

import "fmt"

func main() {
    fmt.Println(I(3))
}

func I(n int) [][]float64 {
    m := make([][]float64, n)
    for i := 0; i < n; i++ {
        a := make([]float64, n)
        a[i] = 1
        m[i] = a
    }
    return m
}
Output:

No special formatting method used.

[[1 0 0] [0 1 0] [0 0 1]]

2D, resliced: Representation as a slice of slices still, but with all elements based on single underlying slice. Might save a little memory management, might have a little better locality.

package main

import "fmt"

func main() {
    fmt.Println(I(3))
}

func I(n int) [][]float64 {
    m := make([][]float64, n)
    a := make([]float64, n*n)
    for i := 0; i < n; i++ {
        a[i] = 1
        m[i] = a[:n]
        a = a[n:]
    }
    return m
}
Output:

Same as previous.

Flat: Representation as a single flat slice. You just have to know to handle it as a square matrix. In many cases that's not a problem and the code is simpler this way. If you want to add a little bit of type checking, you can define a matrix type as shown here.

package main

import "fmt"

type matrix []float64

func main() {
    n := 3
    m := I(n)
    // dump flat represenation
    fmt.Println(m)

    // function x turns a row and column into an index into the
    // flat representation.
    x := func(r, c int) int { return r*n + c }

    // access m by row and column.
    for r := 0; r < n; r++ {
        for c := 0; c < n; c++ {
            fmt.Print(m[x(r, c)], " ")
        }
        fmt.Println()
    }
}

func I(n int) matrix {
    m := make(matrix, n*n)
    // a fast way to initialize the flat representation
    n++
    for i := 0; i < len(m); i += n {
        m[i] = 1
    }
    return m
}
Output:
[1 0 0 0 1 0 0 0 1]
1 0 0 
0 1 0 
0 0 1 

Groovy

Solution:

def makeIdentityMatrix = { n ->
    (0..<n).collect { i -> (0..<n).collect { j -> (i == j) ? 1 : 0 } }
}

Test:

(2..6).each { order ->
    def iMatrix = makeIdentityMatrix(order)
    iMatrix.each { println it }
    println()
}
Output:
[1, 0]
[0, 1]

[1, 0, 0]
[0, 1, 0]
[0, 0, 1]

[1, 0, 0, 0]
[0, 1, 0, 0]
[0, 0, 1, 0]
[0, 0, 0, 1]

[1, 0, 0, 0, 0]
[0, 1, 0, 0, 0]
[0, 0, 1, 0, 0]
[0, 0, 0, 1, 0]
[0, 0, 0, 0, 1]

[1, 0, 0, 0, 0, 0]
[0, 1, 0, 0, 0, 0]
[0, 0, 1, 0, 0, 0]
[0, 0, 0, 1, 0, 0]
[0, 0, 0, 0, 1, 0]
[0, 0, 0, 0, 0, 1]

Haskell

matI n = [ [fromEnum $ i == j | i <- [1..n]] | j <- [1..n]]

And a function to show matrix pretty:

showMat :: [[Int]] -> String
showMat = unlines . map (unwords . map show)


*Main> putStr $ showMat $ matId 9
1 0 0 0 0 0 0 0 0
0 1 0 0 0 0 0 0 0
0 0 1 0 0 0 0 0 0
0 0 0 1 0 0 0 0 0
0 0 0 0 1 0 0 0 0
0 0 0 0 0 1 0 0 0
0 0 0 0 0 0 1 0 0
0 0 0 0 0 0 0 1 0
0 0 0 0 0 0 0 0 1

We could alternatively bypassing the syntactic sugaring of list comprehension notation, and use a bind function directly:

idMatrix :: Int -> [[Int]]
idMatrix n =
  let xs = [1 .. n]
  in xs >>= \x -> [xs >>= \y -> [fromEnum (x == y)]]

or reduce the number of terms a little to:

idMatrix :: Int -> [[Int]]
idMatrix n =
  let xs = [1 .. n]
  in (\x -> fromEnum . (x ==) <$> xs) <$> xs

main :: IO ()
main = (putStr . unlines) $ unwords . fmap show <$> idMatrix 5
Output:
1 0 0 0 0
0 1 0 0 0
0 0 1 0 0
0 0 0 1 0
0 0 0 0 1

Icon and Unicon

This code works for Icon and Unicon.

link matrix
procedure main(argv)
    if not (integer(argv[1]) > 0) then stop("Argument must be a positive integer.")
    matrix1 := identity_matrix(argv[1], argv[1])
    write_matrix(&output,matrix1)
end
Output:
->im 6
1 0 0 0 0 0 
0 1 0 0 0 0 
0 0 1 0 0 0 
0 0 0 1 0 0 
0 0 0 0 1 0 
0 0 0 0 0 1
->

IS-BASIC

100 PROGRAM "Identity.bas"
110 INPUT PROMPT "Enter size of matrix: ":N
120 NUMERIC A(1 TO N,1 TO N)
130 CALL INIT(A)
140 CALL WRITE(A)
150 DEF INIT(REF T)
160   FOR I=LBOUND(T,1) TO UBOUND(T,1)
170     FOR J=LBOUND(T,2) TO UBOUND(T,2)
180       LET T(I,J)=0
190     NEXT
200     LET T(I,I)=1
210   NEXT
220 END DEF
230 DEF WRITE(REF T)
240   FOR I=LBOUND(T,1) TO UBOUND(T,1)
250     FOR J=LBOUND(T,2) TO UBOUND(T,2)
260       PRINT T(I,J);
270     NEXT
280     PRINT
290   NEXT
300 END DEF

J

   =i.4        NB. create an Identity matrix of size 4
1 0 0 0
0 1 0 0
0 0 1 0
0 0 0 1
   Id=: =@i.   NB. define as a verb with a user-defined name
   Id 5        NB. create an Identity matrix of size 5
1 0 0 0 0
0 1 0 0 0
0 0 1 0 0
0 0 0 1 0
0 0 0 0 1

Java

public class PrintIdentityMatrix {

    public static void main(String[] args) {
        int n = 5;
        int[][] array = new int[n][n];

        IntStream.range(0, n).forEach(i -> array[i][i] = 1);

        Arrays.stream(array)
                .map((int[] a) -> Arrays.toString(a))
                .forEach(System.out::println);
    }
}
Output:
[1, 0, 0, 0, 0]
[0, 1, 0, 0, 0]
[0, 0, 1, 0, 0]
[0, 0, 0, 1, 0]
[0, 0, 0, 0, 1]

JavaScript

ES5

function idMatrix(n) {
    return Array.apply(null, new Array(n))
        .map(function (x, i, xs) {
            return xs.map(function (_, k) {
                return i === k ? 1 : 0;
            })
        });
}

ES6

(() => {

    // identityMatrix :: Int -> [[Int]]
    const identityMatrix = n =>
        Array.from({
            length: n
        }, (_, i) => Array.from({
            length: n
        }, (_, j) => i !== j ? 0 : 1));


    // ----------------------- TEST ------------------------
    return identityMatrix(5)
        .map(JSON.stringify)
        .join('\n');
})();
Output:
[1,0,0,0,0]
[0,1,0,0,0]
[0,0,1,0,0]
[0,0,0,1,0]
[0,0,0,0,1]

jq

Construction

def identity(n):
  [range(0;n) | 0] as $row
  | reduce range(0;n) as $i ([]; . + [ $row | .[$i] = 1 ] );

Example:

identity(4)

produces:

[[1,0,0,0],[0,1,0,0],[0,0,1,0],[0,0,0,1]]

Using matrix/2

Using the definition of matrix/2 at Create_a_two-dimensional_array_at_runtime#jq:

def identity(n): 
  reduce range(0;n) as $i
    (0 | matrix(n;n); .[$i][$i] = 1);

Jsish

/* Identity matrix, in Jsish */
function identityMatrix(n) {
    var mat = new Array(n).fill(0);
    for (var r in mat) {
        mat[r] = new Array(n).fill(0);
        mat[r][r] = 1;
    }
    return mat;
}

provide('identityMatrix', 1);

if (Interp.conf('unitTest')) {
;    identityMatrix(0);
;    identityMatrix(1);
;    identityMatrix(2);
;    identityMatrix(3);
    var mat = identityMatrix(4);
    for (var r in mat) puts(mat[r]);
}

/*
=!EXPECTSTART!=
identityMatrix(0) ==> []
identityMatrix(1) ==> [ [ 1 ] ]
identityMatrix(2) ==> [ [ 1, 0 ], [ 0, 1 ] ]
identityMatrix(3) ==> [ [ 1, 0, 0 ], [ 0, 1, 0 ], [ 0, 0, 1 ] ]
[ 1, 0, 0, 0 ]
[ 0, 1, 0, 0 ]
[ 0, 0, 1, 0 ]
[ 0, 0, 0, 1 ]
=!EXPECTEND!=
*/
Output:
promt$ jsish -u identityMatrix.jsi
[PASS] identityMatrix.jsi

Julia

I is an object of type UniformScaling, representing an identity matrix of any size, boolean by default, that can be multiplied by a scalar

using LinearAlgebra
unitfloat64matrix = 1.0I

UniformScaling object can be used as a function to construct a Diagonal matrix of given size, that can be converted to a full matrix using collect

using LinearAlgebra
diagI3 = 1.0I(3)
fullI3 = collect(diagI3)

The function I(3) is not defined in Julia-1.0.5. Other ways to construct a full matrix of given size are

using LinearAlgebra
fullI3 = Matrix{Float64}(I, 3, 3)
fullI3 = Array{Float64}(I, 3, 3)
fullI3 = Array{Float64,2}(I, 3, 3)
fullI3 = zeros(3,3) + I

K

  =4
(1 0 0 0
 0 1 0 0
 0 0 1 0
 0 0 0 1)
  =5
(1 0 0 0 0
 0 1 0 0 0
 0 0 1 0 0
 0 0 0 1 0
 0 0 0 0 1)

Kotlin

fun main() {
    print("Enter size of matrix : ")
    val n = readln().toInt()
    println()
    val identity = Array(n) { i ->
        IntArray(n) { j -> 
            if (i == j) 1 else 0
        }
    }

    // print identity matrix if n <= 40
    if (n <= 40) 
        for (row in identity) println(row.joinToString(" "))
    else 
        println("Matrix is too big to display on 80 column console")
}

Sample input/output

Output:
Enter size of matrix : 5

1 0 0 0 0
0 1 0 0 0
0 0 1 0 0
0 0 0 1 0
0 0 0 0 1

Lambdatalk

{def identity
 {lambda {:n}
  {A.new {S.map {{lambda {:n :i} 
   {A.new {S.map {{lambda {:i :j}
                          {if {= :i :j} then 1 else 0} } :i}  
          {S.serie 0 :n}}}} :n} 
         {S.serie 0 :n}} }}}
-> identity

{identity 2}
-> [[1,0],[0,1]]

{identity 5}
-> [[1,0,0,0,0],[0,1,0,0,0],[0,0,1,0,0],[0,0,0,1,0],[0,0,0,0,1]]


Lang5

: identity-matrix
    dup iota 'A set

    : i.(*) A in ;
    [1] swap append reverse A swap reshape 'i. apply
    ;

5 identity-matrix .
Output:
[
  [    1     0     0     0     0  ]
  [    0     1     0     0     0  ]
  [    0     0     1     0     0  ]
  [    0     0     0     1     0  ]
  [    0     0     0     0     1  ]
]

LFE

(defun identity
  ((`(,m ,n))
   (identity m n))
  ((m)
   (identity m m)))

(defun identity (m n)
  (lists:duplicate m (lists:duplicate n 1)))

From the LFE REPL; note that the last two usage examples demonstrate how identify could be used when composed with functions that get the dimension of a matrix:

> (identity 3)
((1 1 1) (1 1 1) (1 1 1))
> (identity 3 3)
((1 1 1) (1 1 1) (1 1 1))
> (identity '(3 3))
((1 1 1) (1 1 1) (1 1 1))

LSL

To test it yourself; rez a box on the ground, and add the following as a New Script.

default {
	state_entry() {
		llListen(PUBLIC_CHANNEL, "", llGetOwner(), "");
		llOwnerSay("Please Enter a Dimension for an Identity Matrix.");
	}
	listen(integer iChannel, string sName, key kId, string sMessage) {
		llOwnerSay("You entered "+sMessage+".");
		list lMatrix = [];
		integer x = 0;
		integer n = (integer)sMessage;
		for(x=0 ; x<n*n ; x++) {
			lMatrix += [(integer)(((x+1)%(n+1))==1)];
		}
		//llOwnerSay("["+llList2CSV(lMatrix)+"]");
		for(x=0 ; x<n ; x++) {
			llOwnerSay("["+llList2CSV(llList2ListStrided(lMatrix, x*n, (x+1)*n-1, 1))+"]");
		}
	}
}
Output:
You: 0
Identity_Matrix: You entered 0.
You: 1
Identity_Matrix: You entered 1.
Identity_Matrix: [1]
You: 3
Identity_Matrix: You entered 3.
Identity_Matrix: [1, 0, 0]
Identity_Matrix: [0, 1, 0]
Identity_Matrix: [0, 0, 1]
You: 5
Identity_Matrix: You entered 5.
Identity_Matrix: [1, 0, 0, 0, 0]
Identity_Matrix: [0, 1, 0, 0, 0]
Identity_Matrix: [0, 0, 1, 0, 0]
Identity_Matrix: [0, 0, 0, 1, 0]
Identity_Matrix: [0, 0, 0, 0, 1]

Lua

function identity_matrix (size)
        local m = {}
        for i = 1, size do
                m[i] = {}
                for j = 1, size do
                        m[i][j] = i == j and 1 or 0
                end
        end
        return m
end

function print_matrix (m)
        for i = 1, #m do
                print(table.concat(m[i], " "))
        end
end

print_matrix(identity_matrix(5))
Output:
1 0 0 0 0
0 1 0 0 0
0 0 1 0 0
0 0 0 1 0
0 0 0 0 1

Maple

One of a number of ways to do this:

> LinearAlgebra:-IdentityMatrix( 4 );
                           [1    0    0    0]
                           [                ]
                           [0    1    0    0]
                           [                ]
                           [0    0    1    0]
                           [                ]
                           [0    0    0    1]

Here, for instance, is another, in which the entries are (4-byte) floats.

> Matrix( 4, shape = scalar[1], datatype = float[4] );
                         [1.    0.    0.    0.]
                         [                    ]
                         [0.    1.    0.    0.]
                         [                    ]
                         [0.    0.    1.    0.]
                         [                    ]
                         [0.    0.    0.    1.]

Yet another, with 2-byte integer entries:

> Matrix( 4, shape = identity, datatype = integer[ 2 ] );
                           [1    0    0    0]
                           [                ]
                           [0    1    0    0]
                           [                ]
                           [0    0    1    0]
                           [                ]
                           [0    0    0    1]

MathCortex

I = eye(10)

Mathematica / Wolfram Language

IdentityMatrix[4]

MATLAB / Octave

The eye function create the identity (I) matrix, e.g.:

I = eye(10)

Maxima

ident(4);
/* matrix([1, 0, 0, 0],
          [0, 1, 0, 0],
          [0, 0, 1, 0],
          [0, 0, 0, 1]) */

NetRexx

Using int Array

Translation of: REXX
/* NetRexx ************************************************************
* show identity matrix of size n
* I consider m[i,j] to represent the matrix
* 09.07.2013 Walter Pachl (translated from REXX Version 2)
**********************************************************************/
options replace format comments java crossref symbols binary

Parse Arg n .
If n='' then n=5
Say 'Identity Matrix of size' n '(m[i,j] IS the Matrix)'
m=int[n,n] -- Allocate 2D square array at run-time
Loop i=0 To n-1 -- Like Java, arrays in NetRexx start at 0
  ol=''
  Loop j=0 To n-1
    m[i,j]=(i=j)
    ol=ol m[i,j]
    End
  Say ol
  End

Using Indexed String

/* NetRexx */
options replace format comments java crossref symbols nobinary

runSample(arg)
return

-- ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
method createIdMatrix(n) public static
  DIM_ = 'DIMENSION'
  m = 0 -- Indexed string to hold matrix; default value for all elements is zero
  m[DIM_] = n
  loop i = 1 to n -- NetRexx indexed strings don't have to start at zero
    m[i, i] = 1   -- set this diagonal element to 1
    end i
  return m
  
-- ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
method displayIdMatrix(m) public static
  DIM_ = 'DIMENSION'
  if \m.exists(DIM_) then signal RuntimeException('Matrix dimension not set')
  n = m[DIM_]
  loop i = 1 to n
    ol = ''
    loop j = 1 To n
      ol = ol m[i, j]
      end j
    say ol
    end i
  return

-- ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
method runSample(arg) public static
  parse arg n .
  if n = '' then n = 5
  say 'Identity Matrix of size' n
  displayIdMatrix(createIdMatrix(n))
  return

Nim

proc identityMatrix(n: Positive): auto =
  result = newSeq[seq[int]](n)
  for i in 0 ..< result.len:
    result[i] = newSeq[int](n)
    result[i][i] = 1

Nu

Works with: Nushell version 0.97.1
def identity-matrix [n: int] {
  ..<$n | each {|i| ..<$n | each { $in == $i | into int } }
}

..5 | each { {k: $in v: (identity-matrix $in)} } | transpose -r | table -i false -e --flatten
Output:
╭───┬───────┬─────────┬───────────┬─────────────┬───────────────╮
│ 0 │   1   │    2    │     3     │      4      │       5       │
├───┼───────┼─────────┼───────────┼─────────────┼───────────────┤
│   │ ╭───╮ │ ╭─────╮ │ ╭───────╮ │ ╭─────────╮ │ ╭───────────╮ │
│   │ │ 1 │ │ │ 1 0 │ │ │ 1 0 0 │ │ │ 1 0 0 0 │ │ │ 1 0 0 0 0 │ │
│   │ ╰───╯ │ │ 0 1 │ │ │ 0 1 0 │ │ │ 0 1 0 0 │ │ │ 0 1 0 0 0 │ │
│   │       │ ╰─────╯ │ │ 0 0 1 │ │ │ 0 0 1 0 │ │ │ 0 0 1 0 0 │ │
│   │       │         │ ╰───────╯ │ │ 0 0 0 1 │ │ │ 0 0 0 1 0 │ │
│   │       │         │           │ ╰─────────╯ │ │ 0 0 0 0 1 │ │
│   │       │         │           │             │ ╰───────────╯ │
╰───┴───────┴─────────┴───────────┴─────────────┴───────────────╯

Objeck

class IdentityMatrix {
  function : Matrix(n : Int) ~ Int[,] {
    array := Int->New[n,n];
    
    for(row:=0; row<n; row+=1;){
      for(col:=0; col<n; col+=1;){
        if(row = col){
          array[row, col] := 1;
        }
        else{
          array[row,col] := 0;
        };
      };
    };
    return array;
  }
  
  function : PrintMatrix(array : Int[,]) ~ Nil {
    sizes := array->Size();
    for(row:=0; row<sizes[0]; row+=1;){
      for(col:=0; col<sizes[1]; col+=1;){
        value := array[row,col];
        "{$value} \t"->Print();
      };
      '\n'->PrintLine();
    };
  }
  
  function : Main(args : String[]) ~ Nil {
    PrintMatrix(Matrix(5));
  }
}

OCaml

From the interactive loop (that we call the "toplevel"):

$ ocaml

# let make_id_matrix n =
    let m = Array.make_matrix n n 0.0 in
    for i = 0 to pred n do
      m.(i).(i) <- 1.0
    done;
    (m)
  ;;
val make_id_matrix : int -> float array array = <fun>

# make_id_matrix 4 ;;
- : float array array =
[| [|1.; 0.; 0.; 0.|];
   [|0.; 1.; 0.; 0.|];
   [|0.; 0.; 1.; 0.|];
   [|0.; 0.; 0.; 1.|] |]

another way:

# let make_id_matrix n =
    Array.init n (fun i ->
      Array.init n (fun j ->
        if i = j then 1.0 else 0.0))
  ;;
val make_id_matrix : int -> float array array = <fun>

# make_id_matrix 4 ;;
- : float array array =
[| [|1.; 0.; 0.; 0.|];
   [|0.; 1.; 0.; 0.|];
   [|0.; 0.; 1.; 0.|];
   [|0.; 0.; 0.; 1.|] |]

When we write a function in the toplevel, it returns us its signature (the prototype), and when we write a variable (or a function call), it returns its type and its value.

Octave

The eye function create the identity (I) matrix, e.g.:

I = eye(10)

Ol

(define (make-identity-matrix n)
   (map (lambda (i)
         (append (repeat 0 i) '(1) (repeat 0 (- n i 1))))
      (iota n)))

(for-each print (make-identity-matrix 3))
(for-each print (make-identity-matrix 17))
Output:
(1 0 0)
(0 1 0)
(0 0 1)
(1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0)
(0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0)
(0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0)
(0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0)
(0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0)
(0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0)
(0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0)
(0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0)
(0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0)
(0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0)
(0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0)
(0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0)
(0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0)
(0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0)
(0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0)
(0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0)
(0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1)

ooRexx

ooRexx doesn't have a proper matrix class, but it does have multidimensional arrays.

say "a 3x3 identity matrix"
say
call printMatrix createIdentityMatrix(3)
say
say "a 5x5 identity matrix"
say
call printMatrix createIdentityMatrix(5)

::routine createIdentityMatrix
  use arg size
  matrix = .array~new(size, size)
  loop i = 1 to size
      loop j = 1 to size
          if i == j then matrix[i, j] = 1
          else matrix[i, j] = 0
      end j
  end i
  return matrix

::routine printMatrix
  use arg matrix

  loop i = 1 to matrix~dimension(1)
      line = ""
      loop j = 1 to matrix~dimension(2)
          line = line matrix[i, j]
      end j
      say line
  end i
Output:
a 3x3 identity matrix

 1 0 0
 0 1 0
 0 0 1

a 5x5 identity matrix

 1 0 0 0 0
 0 1 0 0 0
 0 0 1 0 0
 0 0 0 1 0
 0 0 0 0 1

OxygenBasic

Class SquareMatrix
'=================

  double *Cell
  sys    size

  method SetIdentity()
  indexbase 0
  sys e,i,j
  e=size*size
  for i=0 to <size
    cell(i*size+j)=1 : j++
  next
  end method

  method constructor(sys n)
  @cell=getmemory n*n*sizeof double
  size=n
  end method

  method destructor()
  freememory @cell
  end method

end class

new SquareMatrix M(8)
M.SetIdentity
'...
del M

PARI/GP

Built-in:

matid(9)

Custom:

matrix(9,9,i,j,i==j)

Pascal

program IdentityMatrix(input, output);

var
  matrix: array of array of integer;
  n, i, j: integer;
  
begin
  write('Size of matrix: ');
  readln(n);
  setlength(matrix, n, n);

  for i := 0 to n - 1 do
    matrix[i,i] := 1;
    
  for i := 0 to n - 1 do
  begin
    for j := 0 to n - 1 do
      write (matrix[i,j], ' ');
    writeln;
  end;
end.
Output:
% ./IdentityMatrix
Size of matrix: 5
1 0 0 0 0 
0 1 0 0 0 
0 0 1 0 0 
0 0 0 1 0 
0 0 0 0 1 

PascalABC.NET

begin
  var n := ReadInteger;
  var matrix: array [,] of integer := MatrGen(n,n,(i,j) -> i = j ? 1 : 0);
  matrix.Println
end.
Output:
5
   1   0   0   0   0
   0   1   0   0   0
   0   0   1   0   0
   0   0   0   1   0
   0   0   0   0   1


Perl

use strict;
use warnings;
use feature 'say';

sub identity_matrix {
    my($n) = shift() - 1;
    map { [ (0) x $_, 1, (0) x ($n - $_) ] } 0..$n
}

for (<4 5 6>) {
  say "\n$_:";
  say join ' ', @$_ for identity_matrix $_;
}
Output:
4:
1 0 0 0 
0 1 0 0 
0 0 1 0 
0 0 0 1 

5:
1 0 0 0 0 
0 1 0 0 0 
0 0 1 0 0 
0 0 0 1 0 
0 0 0 0 1 

6:
1 0 0 0 0 0 
0 1 0 0 0 0 
0 0 1 0 0 0 
0 0 0 1 0 0 
0 0 0 0 1 0 
0 0 0 0 0 1

Phix

function identity(integer n)
sequence res = repeat(repeat(0,n),n)
    for i=1 to n do
        res[i][i] = 1
    end for
    return res
end function
 
ppOpt({pp_Nest,1})
pp(identity(3))
pp(identity(5))
pp(identity(7))
pp(identity(9))
Output:
{{1,0,0},
 {0,1,0},
 {0,0,1}}
{{1,0,0,0,0},
 {0,1,0,0,0},
 {0,0,1,0,0},
 {0,0,0,1,0},
 {0,0,0,0,1}}
{{1,0,0,0,0,0,0},
 {0,1,0,0,0,0,0},
 {0,0,1,0,0,0,0},
 {0,0,0,1,0,0,0},
 {0,0,0,0,1,0,0},
 {0,0,0,0,0,1,0},
 {0,0,0,0,0,0,1}}
{{1,0,0,0,0,0,0,0,0},
 {0,1,0,0,0,0,0,0,0},
 {0,0,1,0,0,0,0,0,0},
 {0,0,0,1,0,0,0,0,0},
 {0,0,0,0,1,0,0,0,0},
 {0,0,0,0,0,1,0,0,0},
 {0,0,0,0,0,0,1,0,0},
 {0,0,0,0,0,0,0,1,0},
 {0,0,0,0,0,0,0,0,1}}

PHP

function identity($length) {
	return array_map(function($key, $value) {$value[$key] = 1; return $value;}, range(0, $length-1),
	array_fill(0, $length, array_fill(0,$length, 0)));
}
function print_identity($identity) {
	echo implode(PHP_EOL, array_map(function ($value) {return implode(' ', $value);}, $identity));
}
print_identity(identity(10));
Output:
1 0 0 0 0 0 0 0 0 0
0 1 0 0 0 0 0 0 0 0
0 0 1 0 0 0 0 0 0 0
0 0 0 1 0 0 0 0 0 0
0 0 0 0 1 0 0 0 0 0
0 0 0 0 0 1 0 0 0 0
0 0 0 0 0 0 1 0 0 0
0 0 0 0 0 0 0 1 0 0
0 0 0 0 0 0 0 0 1 0
0 0 0 0 0 0 0 0 0 1

PicoLisp

(de identity (Size)
   (let L (need Size (1) 0)
      (make
         (do Size
            (link (copy (rot L))) ) ) ) )

Test:

: (identity 3)
-> ((1 0 0) (0 1 0) (0 0 1))

: (mapc println (identity 5))
(1 0 0 0 0)
(0 1 0 0 0)
(0 0 1 0 0)
(0 0 0 1 0)
(0 0 0 0 1)

PL/I

identity: procedure (A, n);
   declare A(n,n) fixed controlled;
   declare (i,n) fixed binary;
   allocate A; A = 0;
   do i = 1 to n; A(i,i) = 1; end;
end identity;

PostScript

% n  ident  [identity-matrix]
% create an identity matrix of dimension n*n.
% Uses a local dictionary for its one parameter, perhaps overkill.
% Constructs arrays of arrays of integers using [], for loops, and stack manipulation.
/ident { 1 dict begin /n exch def
    [
    1 1 n {                              % [ i
        [ exch                           % [ [ i
        1 1 n {                          % [ [ i j
            1 index eq { 1 }{ 0 } ifelse % [ [ i b
            exch                         % [ [ b i
        } for                            % [ [ b+ i
        pop ]                            % [ [ b+ ]
    } for                                % [ [b+]+ ]
    ]
end } def

PowerShell

function identity($n) {
    0..($n-1) | foreach{$row = @(0) * $n; $row[$_] = 1; ,$row}
}
function show($a) { $a | foreach{ "$_"} }
$array = identity 4
show $array

Output:

 
1 0 0 0
0 1 0 0
0 0 1 0
0 0 0 1
$array[0][0]
$array[0][1]

Output:

 
1
0


Prolog

Works with: SWi-Prolog
%rotates one list clockwise by one integer
rotate(Int,List,Rotated) :-
	integer(Int),
	length(Suff,Int),
	append(Pre,Suff,List),
	append(Suff,Pre,Rotated).
%rotates a list of lists by a list of integers
rotate(LoInts,LoLists,Rotated) :-
	is_list(LoInts),
	maplist(rotate,LoInts,LoLists,Rotated).

%helper function
append_(Suff,Pre,List) :-
	append([Pre],Suff,List).	
idmatrix(N,IdMatrix):-
	%make an N length list of 1s and append with N-1 0s
	length(Ones,N),
	maplist(=(1),Ones),
	succ(N0,N),
	length(Zeros,N0),
	maplist(=(0),Zeros),
	maplist(append_(Zeros),Ones,M),
	%create the offsets at rotate each row
	numlist(0,N0,Offsets),
	rotate(Offsets,M,IdMatrix).

main :-
	idmatrix(5,I),
	maplist(writeln,I).
Output:
?- main.
[1,0,0,0,0]
[0,1,0,0,0]
[0,0,1,0,0]
[0,0,0,1,0]
[0,0,0,0,1]
true .

PureBasic

>Procedure identityMatrix(Array i(2), size) ;valid only for size >= 0
  ;formats array i() as an identity matrix of size x size
  Dim i(size - 1, size - 1)

  Protected j
  For j = 0 To size - 1
    i(j, j) = 1
  Next 
EndProcedure


Procedure displayMatrix(Array a(2))
  Protected rows = ArraySize(a(), 2), columns = ArraySize(a(), 1)
  Protected i, j
  
  For i = 0 To rows
    For j = 0 To columns
      Print(RSet(Str(a(i, j)), 3, " "))
    Next
    PrintN("")
  Next
EndProcedure

If OpenConsole()
  Dim i3(0, 0)
  Dim i4(0, 0)
  
  identityMatrix(i3(), 3)
  identityMatrix(i4(), 4)
  
  displayMatrix(i3())
  PrintN("")
  displayMatrix(i4())
  
  Print(#CRLF$ + #CRLF$ + "Press ENTER to exit"): Input()
  CloseConsole()
EndIf
Output:
  1  0  0
  0  1  0
  0  0  1

  1  0  0  0
  0  1  0  0
  0  0  1  0
  0  0  0  1

Python

Nested lists

A simple solution, using nested lists to represent the matrix.

def identity(size):
    matrix = [[0]*size for i in range(size)]
    #matrix = [[0] * size] * size    #Has a flaw. See http://stackoverflow.com/questions/240178/unexpected-feature-in-a-python-list-of-lists

    for i in range(size):
        matrix[i][i] = 1
    
    for rows in matrix:
        for elements in rows:
            print elements,
        print ""

Nested maps and comprehensions

Works with: Python version 3.7
'''Identity matrices by maps and equivalent list comprehensions'''

import operator


# idMatrix :: Int -> [[Int]]
def idMatrix(n):
    '''Identity matrix of order n,
       expressed as a nested map.
    '''
    eq = curry(operator.eq)
    xs = range(0, n)
    return list(map(
        lambda x: list(map(
            compose(int)(eq(x)),
            xs
        )),
        xs
    ))


# idMatrix3 :: Int -> [[Int]]
def idMatrix2(n):
    '''Identity matrix of order n,
       expressed as a nested comprehension.
    '''
    xs = range(0, n)
    return ([int(x == y) for x in xs] for y in xs)


# TEST ----------------------------------------------------
def main():
    '''
        Identity matrix of dimension five,
        by two different routes.
    '''
    for f in [idMatrix, idMatrix2]:
        print(
            '\n' + f.__name__ + ':',
            '\n\n' + '\n'.join(map(str, f(5))),
        )


# GENERIC -------------------------------------------------

# compose (<<<) :: (b -> c) -> (a -> b) -> a -> c
def compose(g):
    '''Right to left function composition.'''
    return lambda f: lambda x: g(f(x))


# curry :: ((a, b) -> c) -> a -> b -> c
def curry(f):
    '''A curried function derived
       from an uncurried function.'''
    return lambda a: lambda b: f(a, b)


# MAIN ---
if __name__ == '__main__':
    main()
Output:
idMatrix: 

[1, 0, 0, 0, 0]
[0, 1, 0, 0, 0]
[0, 0, 1, 0, 0]
[0, 0, 0, 1, 0]
[0, 0, 0, 0, 1]

idMatrix2: 

[1, 0, 0, 0, 0]
[0, 1, 0, 0, 0]
[0, 0, 1, 0, 0]
[0, 0, 0, 1, 0]
[0, 0, 0, 0, 1]

Dict of points

A dict of tuples of two ints (x, y) are used to represent the matrix.

>>> def identity(size):
...     return {(x, y):int(x == y) for x in range(size) for y in range(size)}
... 
>>> size = 4
>>> matrix = identity(size)
>>> print('\n'.join(' '.join(str(matrix[(x, y)]) for x in range(size)) for y in range(size)))
1 0 0 0
0 1 0 0
0 0 1 0
0 0 0 1
>>>

Numpy

A solution using the numpy library

np.mat(np.eye(size))

Quackery

[ [] swap times
    [ 0 i^ of 1 join 0 i of
      join nested join ] ]  is identity ( n --> [ )

5 identity echo
Output:
[ [ 1 0 0 0 0 ] [ 0 1 0 0 0 ] [ 0 0 1 0 0 ] [ 0 0 0 1 0 ] [ 0 0 0 0 1 ] ]

R

When passed a single scalar argument, diag produces an identity matrix of size given by the scalar. For example:

diag(3)

produces:

     [,1] [,2] [,3]
[1,]    1    0    0
[2,]    0    1    0
[3,]    0    0    1

Or you can also use the method that is shown below

Identity_matrix=function(size){
  x=matrix(0,size,size)
  for (i in 1:size) {
    x[i,i]=1
  }
  return(x)
}

Racket

#lang racket
(require math)
(identity-matrix 5)
Output:
(array #[#[1 0 0 0 0] 
         #[0 1 0 0 0]
         #[0 0 1 0 0]
         #[0 0 0 1 0]
         #[0 0 0 0 1]])

Raku

(formerly Perl 6)

Works with: rakudo version 2015-09-15
sub identity-matrix($n) {
    my @id;
    for flat ^$n X ^$n -> $i, $j {
        @id[$i][$j] = +($i == $j);
    }
    @id;
}

.say for identity-matrix(5);
Output:
[1 0 0 0 0]
[0 1 0 0 0]
[0 0 1 0 0]
[0 0 0 1 0]
[0 0 0 0 1]

On the other hand, this may be clearer and/or faster:

sub identity-matrix($n) {
    my @id = [0 xx $n] xx $n;
    @id[$_][$_] = 1 for ^$n;
    @id;
}

Here is yet an other way to do it:

sub identity-matrix($n) {
    [1, |(0 xx $n-1)], *.rotate(-1) ... *[*-1]
}

Red

Red[]

identity-matrix: function [size][
    matrix: copy []
    repeat i size [
        append/only matrix append/dup copy [] 0 size
        matrix/:i/:i: 1
    ]
    matrix
]

probe identity-matrix 5
Output:
[[1 0 0 0 0] [0 1 0 0 0] [0 0 1 0 0] [0 0 0 1 0] [0 0 0 0 1]]

REXX

version 1

The REXX language doesn't have matrices as such, so the problem is largely how to display the "matrix".

The code to display the matrices was kept as a stand-alone general-purpose (square) matrix display
subroutine,   which, in part,   determines if the square matrix is indeed a square matrix based on the
number of elements given.

It also finds the maximum widths of the integer and decimal fraction parts   (if any)   and uses those widths
to align   (right-justify according to the [possibly implied] decimal point)   the columns of the square matrix.

It also tries to display a centered (and easier to read) matrix,   along with a title.

/*REXX program  creates and displays any sized  identity matrix  (centered, with title).*/
           do k=3  to 6                          /* [↓]  build and display a sq. matrix.*/
           call ident_mat  k                     /*build & display a KxK square matrix. */
           end   /*k*/                           /* [↑]  use general─purpose display sub*/
exit                                             /*stick a fork in it,  we're all done. */
/*──────────────────────────────────────────────────────────────────────────────────────*/
ident_mat: procedure;  parse arg n; $=
              do    r=1  for n                   /*build identity matrix, by row and col*/
                 do c=1  for n;     $= $ (r==c)  /*append  zero  or  one  (if on diag). */
                 end   /*c*/
              end      /*r*/
           call showMat  'identity matrix of size'   n,   $
           return
/*──────────────────────────────────────────────────────────────────────────────────────*/
showMat: procedure; parse arg hdr,x;  #=words(x) /*#  is the number of matrix elements. */
         dp= 0                                   /*DP:  max width of decimal fractions. */
         w= 0                                    /*W:   max width of integer part.      */
                 do n=1  until n*n>=#;  _= word(x,n)      /*determine the matrix order. */
                 parse var _ y '.' f;   w= max(w, length(y));      dp= max(dp, length(f) )
                 end   /*n*/                     /* [↑]  idiomatically find the widths. */
         w= w +1
         say;  say center(hdr, max(length(hdr)+8, (w+1)*#%n), '─');  say
         #= 0                                                            /*#: element #.*/
                 do   row=1  for n;     _= left('', n+w)                 /*indentation. */
                   do col=1  for n;     #= # + 1                         /*bump element.*/
                   _=_ right(format(word(x, #), , dp)/1, w)
                   end   /*col*/                 /* [↑]  division by unity normalizes #.*/
                 say _                           /*display a single line of the matrix. */
                 end     /*row*/
         return
output   when using the default sizes   (3 ──► 6)   for generating four matrices:
────identity matrix of size 3────

       1  0  0
       0  1  0
       0  0  1

────identity matrix of size 4────

        1  0  0  0
        0  1  0  0
        0  0  1  0
        0  0  0  1

────identity matrix of size 5────

         1  0  0  0  0
         0  1  0  0  0
         0  0  1  0  0
         0  0  0  1  0
         0  0  0  0  1

────identity matrix of size 6────

          1  0  0  0  0  0
          0  1  0  0  0  0
          0  0  1  0  0  0
          0  0  0  1  0  0
          0  0  0  0  1  0
          0  0  0  0  0  1

version 2

An alternative?!

/* REXX ***************************************************************
* show identity matrix of size n
* I consider m.i.j to represent the matrix (not needed for showing)
* 06.07.2012 Walter Pachl
**********************************************************************/
Parse Arg n
Say 'Identity Matrix of size' n '(m.i.j IS the Matrix)'
m.=0
Do i=1 To n
  ol=''
  Do j=1 To n
    m.i.j=(i=j)
    ol=ol''format(m.i.j,2) /* or ol=ol (i=j)                         */
    End
  Say ol
  End
Output:
Identity Matrix of size 3  (m.i.j IS the Matrix)
 1 0 0
 0 1 0
 0 0 1    

This could be a 3-dimensional sparse matrix with one element set:

m.=0
m.0=1000 /* the matrix' size */
m.4.17.333='Walter'

Ring

size = 5
im = newlist(size, size)
identityMatrix(size, im)
for r = 1 to size
    for c = 1 to size
        see im[r][c]
    next
    see nl
next 
 
func identityMatrix s, m
     m = newlist(s, s)
     for i = 1 to s
         m[i][i] = 1
     next
     return m

func newlist x, y
     if isstring(x) x=0+x ok
     if isstring(y) y=0+y ok
     alist = list(x)
     for t in alist
         t = list(y)
     next
     return alist

Output:

10000
01000
00100
00010
00001

Gui version

# Project : Identity Matrix
# Date    : 2022/16/02
# Author  : Gal Zsolt (~ CalmoSoft ~)
# Email   : <calmosoft@gmail.com>

load "stdlib.ring"
load "guilib.ring"

size = 8
C_Spacing = 1

C_ButtonBlueStyle   = 'border-radius:6px;color:black; background-color: blue'
C_ButtonOrangeStyle = 'border-radius:6px;color:black; background-color: orange'

Button = newlist(size,size)
LayoutButtonRow = list(size)

app = new qApp 
{
      win = new qWidget() {
	    setWindowTitle('Identity Matrix')
	    move(500,100)
	    reSize(600,600)
	    winheight = win.height()
	    fontSize = 18 + (winheight / 100)

 	    LayoutButtonMain = new QVBoxLayout()			
	    LayoutButtonMain.setSpacing(C_Spacing)
	    LayoutButtonMain.setContentsmargins(0,0,0,0)

	    for Row = 1 to size
		LayoutButtonRow[Row] = new QHBoxLayout() {
				       setSpacing(C_Spacing)
				       setContentsmargins(0,0,0,0)
				       } 
         	 for Col = 1 to size
		     Button[Row][Col] = new QPushButton(win) {
                                        setSizePolicy(1,1)                                                
					}
					
		     LayoutButtonRow[Row].AddWidget(Button[Row][Col])	
		 next
		 LayoutButtonMain.AddLayout(LayoutButtonRow[Row])			
	      next
              LayoutDataRow1 = new QHBoxLayout() { setSpacing(C_Spacing) setContentsMargins(0,0,0,0) }
              LayoutButtonMain.AddLayout(LayoutDataRow1)
              setLayout(LayoutButtonMain)
              show()
   }
   pBegin()
   exec()
   }

func pBegin()
     for Row = 1 to size
         for Col = 1 to size 
             if Row = Col
                Button[Row][Col].setStyleSheet(C_ButtonOrangeStyle)
                Button[Row][Col].settext("1")
             else
                Button[Row][Col].setStyleSheet(C_ButtonBlueStyle)
                Button[Row][Col].settext("0")
             ok
	 next
     next
     score = 0

Output image:

Identity Matrix

RPL

Input:
3 IDN
Output:
1:          [[ 1 0 0 ]
             [ 0 1 0 ]
            [ 0 0 1 ]]

Ruby

Using Array

def identity(size)
  Array.new(size){|i| Array.new(size){|j| i==j ? 1 : 0}}
end

[4,5,6].each do |size|
  puts size, identity(size).map {|r| r.to_s}, ""
end
Output:
4
[1, 0, 0, 0]
[0, 1, 0, 0]
[0, 0, 1, 0]
[0, 0, 0, 1]

5
[1, 0, 0, 0, 0]
[0, 1, 0, 0, 0]
[0, 0, 1, 0, 0]
[0, 0, 0, 1, 0]
[0, 0, 0, 0, 1]

6
[1, 0, 0, 0, 0, 0]
[0, 1, 0, 0, 0, 0]
[0, 0, 1, 0, 0, 0]
[0, 0, 0, 1, 0, 0]
[0, 0, 0, 0, 1, 0]
[0, 0, 0, 0, 0, 1]

Using Matrix

require 'matrix'
p Matrix.identity(5)
 # => Matrix[[1, 0, 0, 0, 0], [0, 1, 0, 0, 0], [0, 0, 1, 0, 0], [0, 0, 0, 1, 0], [0, 0, 0, 0, 1]]

Run BASIC

' formats array im() of size ims
for ims = 4 to 6

print :print "--- Size: ";ims;" ---"
 Dim im(ims,ims)

 For i = 1 To ims
   im(i,i) = 1
 next 

 For row = 1 To ims
   print "[";
   cma$ = ""
     For col = 1 To ims
       print cma$;im(row, col);
       cma$ = ", "
    next
   print "]"
 next
next ims
Output:
--- Size: 4 ---
[1, 0, 0, 0]
[0, 1, 0, 0]
[0, 0, 1, 0]
[0, 0, 0, 1]

--- Size: 5 ---
[1, 0, 0, 0, 0]
[0, 1, 0, 0, 0]
[0, 0, 1, 0, 0]
[0, 0, 0, 1, 0]
[0, 0, 0, 0, 1]

--- Size: 6 ---
[1, 0, 0, 0, 0, 0]
[0, 1, 0, 0, 0, 0]
[0, 0, 1, 0, 0, 0]
[0, 0, 0, 1, 0, 0]
[0, 0, 0, 0, 1, 0]
[0, 0, 0, 0, 0, 1]

Rust

Run with command-line containing the matrix size.

extern crate num;
struct Matrix<T> {
    data: Vec<T>,
    size: usize,
}

impl<T> Matrix<T>
where
    T: num::Num + Clone + Copy,
{
    fn new(size: usize) -> Self {
        Self {
            data: vec![T::zero(); size * size],
            size: size,
        }
    }
    fn get(&mut self, x: usize, y: usize) -> T {
        self.data[x + self.size * y]
    }
    fn identity(&mut self) {
        for (i, item) in self.data.iter_mut().enumerate() {
            *item = if i % (self.size + 1) == 0 {
                T::one()
            } else {
                T::zero()
            }
        }
    }
}

fn main() {
    let size = std::env::args().nth(1).unwrap().parse().unwrap();
    let mut matrix = Matrix::<i32>::new(size);
    matrix.identity();
    for y in 0..size {
        for x in 0..size {
            print!("{} ", matrix.get(x, y));
        }
        println!();
    }
}

Scala

def identityMatrix(n:Int)=Array.tabulate(n,n)((x,y) => if(x==y) 1 else 0)
def printMatrix[T](m:Array[Array[T]])=m map (_.mkString("[", ", ", "]")) mkString "\n"

printMatrix(identityMatrix(5))
Output:
[1, 0, 0, 0, 0]
[0, 1, 0, 0, 0]
[0, 0, 1, 0, 0]
[0, 0, 0, 1, 0]
[0, 0, 0, 0, 1]

Scheme

When representing a matrix as a collection of nested lists:

(define (identity n)
  (letrec
      ((uvec 
	(lambda (m i acc)
	  (if (= i n)
	      acc
	      (uvec m (+ i 1)
		    (cons (if (= i m) 1 0) acc)))))
       (idgen
	(lambda (i acc)
	  (if (= i n)
	      acc
	      (idgen (+ i 1)
		     (cons (uvec i 0 '()) acc))))))
       (idgen 0 '())))

Test program:

(display (identity 4))
Output:
((1 0 0 0) (0 1 0 0) (0 0 1 0) (0 0 0 1))

Seed7

$ include "seed7_05.s7i";

const type: matrix is array array integer;

const func matrix: identity (in integer: size) is func
  result
    var matrix: identity is matrix.value;
  local
    var integer: index is 0;
  begin
    identity := size times size times 0;
    for index range 1 to size do
      identity[index][index] := 1;
    end for;
  end func;

const proc: writeMat (in matrix: a) is func
  local
    var integer: i is 0;
    var integer: num is 0;
  begin
    for key i range a do
      for num range a[i] do
        write(num lpad 2);
      end for;
      writeln;
    end for;
  end func;

const proc: main is func
  begin
    writeMat(identity(6));
  end func;
Output:
 1 0 0 0 0 0
 0 1 0 0 0 0
 0 0 1 0 0 0
 0 0 0 1 0 0
 0 0 0 0 1 0
 0 0 0 0 0 1

SenseTalk

set matrix to buildIdentityMatrix(3)

repeat for each item in matrix
	put it
end repeat

set matrix to buildIdentityMatrix(17)

repeat for each item in matrix
	put it
end repeat

function buildIdentityMatrix matrixSize
	set matrixList to ()
	repeat matrixSize times
		set rowMatrixIndex to the counter
		set rowMatrix to ()
		repeat matrixSize times
			if the counter equals rowMatrixIndex
				insert 1 after rowMatrix
			else
				insert 0 after rowMatrix
			end if
		end repeat
		insert rowMatrix nested after matrixList
	end repeat
	return matrixList
end buildIdentityMatrix

Output for n 3

(1,0,0)
(0,1,0)
(0,0,1)

Output for n 17

(1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0)
(0,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0)
(0,0,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0)
(0,0,0,1,0,0,0,0,0,0,0,0,0,0,0,0,0)
(0,0,0,0,1,0,0,0,0,0,0,0,0,0,0,0,0)
(0,0,0,0,0,1,0,0,0,0,0,0,0,0,0,0,0)
(0,0,0,0,0,0,1,0,0,0,0,0,0,0,0,0,0)
(0,0,0,0,0,0,0,1,0,0,0,0,0,0,0,0,0)
(0,0,0,0,0,0,0,0,1,0,0,0,0,0,0,0,0)
(0,0,0,0,0,0,0,0,0,1,0,0,0,0,0,0,0)
(0,0,0,0,0,0,0,0,0,0,1,0,0,0,0,0,0)
(0,0,0,0,0,0,0,0,0,0,0,1,0,0,0,0,0)
(0,0,0,0,0,0,0,0,0,0,0,0,1,0,0,0,0)
(0,0,0,0,0,0,0,0,0,0,0,0,0,1,0,0,0)
(0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,0,0)
(0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,0)
(0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1)

Sidef

func identity_matrix(n) {
    n.of { |i|
        n.of { |j|
            i == j ? 1 : 0
        }
    }
}

for n (ARGV ? ARGV.map{.to_i} : [4, 5, 6]) {
  say "\n#{n}:"
  for row (identity_matrix(n)) {
    say row.join(' ')
  }
}
Output:
4:
1 0 0 0
0 1 0 0
0 0 1 0
0 0 0 1

5:
1 0 0 0 0
0 1 0 0 0
0 0 1 0 0
0 0 0 1 0
0 0 0 0 1

6:
1 0 0 0 0 0
0 1 0 0 0 0
0 0 1 0 0 0
0 0 0 1 0 0
0 0 0 0 1 0
0 0 0 0 0 1

Sinclair ZX81 BASIC

Works with 1k of RAM, but for a larger matrix you'll want at least 2k.

 10 INPUT S
 20 DIM M(S,S)
 30 FOR I=1 TO S
 40 LET M(I,I)=1
 50 NEXT I
 60 FOR I=1 TO S
 70 SCROLL
 80 FOR J=1 TO S
 90 PRINT M(I,J);
100 NEXT J
110 PRINT
120 NEXT I
Input:
10
Output:
1000000000
0100000000
0010000000
0001000000
0000100000
0000010000
0000001000
0000000100
0000000010
0000000001

Smalltalk

Works with: Pharo Smalltalk
(Array2D identity: (UIManager default request: 'Enter size of the matrix:') asInteger) asString
Output:
'(1 0 0 0
0 1 0 0
0 0 1 0
0 0 0 1 )'

Sparkling

function unitMatrix(n) {
	return map(range(n), function(k1, v1) {
		return map(range(n), function(k2, v2) {
			return v2 == v1 ? 1 : 0;
		});
	});
}

Standard ML

 val eye= fn n => List.tabulate( n, fn i => List.tabulate( n, fn j=> if j=i then 1.0 else 0.0));

Stata

Stata matrix

. mat a = I(3)
. mat list a

symmetric a[3,3]
    c1  c2  c3
r1   1
r2   0   1
r3   0   0   1

Mata

: I(3)
[symmetric]
       1   2   3
    +-------------+
  1 |  1          |
  2 |  0   1      |
  3 |  0   0   1  |
    +-------------+

Swift

Translation of: Elixir
func identityMatrix(size: Int) -> [[Int]] {
  return (0..<size).map({i in
    return (0..<size).map({ $0 == i ? 1 : 0})
  })
}

print(identityMatrix(size: 5))
Output:
[[1, 0, 0, 0, 0], [0, 1, 0, 0, 0], [0, 0, 1, 0, 0], [0, 0, 0, 1, 0], [0, 0, 0, 0, 1]]

Tailspin

templates identityMatrix
  def n: $;
  [1..$n -> [1..~$ -> 0, 1, $~..$n -> 0]] !
end identityMatrix

def identity: 5 -> identityMatrix;
$identity... -> '|$(1);$(2..last)... -> ', $;';|
' -> !OUT::write

v0.5

identityMatrix templates
  n is $;
  [1..$n -> [1..~$ -> 0, 1, $~..$n -> 0]] !
end identityMatrix

identity is 5 -> identityMatrix;
$identity... -> '|$(1);$(2..)... -> ', $;';|
' !
Output:
|1, 0, 0, 0, 0|
|0, 1, 0, 0, 0|
|0, 0, 1, 0, 0|
|0, 0, 0, 1, 0|
|0, 0, 0, 0, 1|

Tcl

When representing a matrix as a collection of nested lists:

proc I {rank {zero 0.0} {one 1.0}} {
    set m [lrepeat $rank [lrepeat $rank $zero]]
    for {set i 0} {$i < $rank} {incr i} {
	lset m $i $i $one
    }
    return $m
}

Or alternatively with the help of the tcllib package for rectangular data structures:

Library: Tcllib (Package: struct::matrix)
package require struct::matrix

proc I {rank {zero 0.0} {one 1.0}} {
    set m [struct::matrix]
    $m add columns $rank
    $m add rows $rank
    for {set i 0} {$i < $rank} {incr i} {
	for {set j 0} {$j < $rank} {incr j} {
	    $m set cell $i $j [expr {$i==$j ? $one : $zero}]
	}
    }
    return $m
}

Demonstrating the latter:

set m [I 5 0 1]    ;# Integer 0/1 for clarity of presentation
puts [$m format 2string]
Output:
1 0 0 0 0
0 1 0 0 0
0 0 1 0 0
0 0 0 1 0
0 0 0 0 1

TypeScript

function identity(n) {
    if (n < 1) return "Not defined";
    else if (n == 1) return 1;
    else {
        var idMatrix:number[][];
        for (var i: number = 0; i < n; i++) {
            for (var j: number = 0; j < n; j++) {
                if (i != j) idMatrix[i][j] = 0;
                else idMatrix[i][j] = 1;
            }            
        }
        return idMatrix;
    }
}

Uiua

Works with: Uiua version 0.11.1
IdMatrix ← ⊞=.⇡
IdMatrix 7
Output:
╭─               
╷ 1 0 0 0 0 0 0  
  0 1 0 0 0 0 0  
  0 0 1 0 0 0 0  
  0 0 0 1 0 0 0  
  0 0 0 0 1 0 0  
  0 0 0 0 0 1 0  
  0 0 0 0 0 0 1  
                ╯

Vala

int main (string[] args) {
	if (args.length < 2) {
		print ("Please, input an integer > 0.\n");
		return 0;
	}
	var n = int.parse (args[1]);
	if (n <= 0) {
		print ("Please, input an integer > 0.\n");
		return 0;
	}
	int[,] array = new int[n, n];
	for (var i = 0; i < n; i ++) {
		for (var j = 0; j < n; j ++) {
			if (i == j) array[i,j] = 1;
			else array[i,j] = 0;
		}
	}
	for (var i = 0; i < n; i ++) {
		for (var j = 0; j < n; j ++) {
			print ("%d ", array[i,j]);
		}
		print ("\b\n");
	}
	return 0;
}

VBA

Private Function Identity(n As Integer) As Variant
    Dim I() As Integer
    ReDim I(n - 1, n - 1)
    For j = 0 To n - 1
        I(j, j) = 1
    Next j
    Identity = I
End Function

VBScript

build_matrix(7)

Sub build_matrix(n)
	Dim matrix()
	ReDim matrix(n-1,n-1)
	i = 0
	'populate the matrix
	For row = 0 To n-1
		For col = 0 To n-1
			If col = i Then
				matrix(row,col) = 1
			Else
				matrix(row,col) = 0
			End If	
		Next
		i = i + 1
	Next
	'display the matrix
	For row = 0 To n-1
		For col = 0 To n-1
			If col < n-1 Then
				WScript.StdOut.Write matrix(row,col) & " "
			Else
				WScript.StdOut.Write matrix(row,col)
			End If
		Next
		WScript.StdOut.WriteLine
	Next
End Sub
Output:
1 0 0 0 0 0 0
0 1 0 0 0 0 0
0 0 1 0 0 0 0
0 0 0 1 0 0 0
0 0 0 0 1 0 0
0 0 0 0 0 1 0
0 0 0 0 0 0 1


Alternate version

n = 8

arr = Identity(n)

for i = 0 to n-1
    for j = 0 to n-1
        wscript.stdout.Write arr(i,j) & " "
    next
    wscript.stdout.writeline
next

Function Identity (size)
    Execute Replace("dim a(#,#):for i=0 to #:for j=0 to #:a(i,j)=0:next:a(i,i)=1:next","#",size-1)
    Identity = a
End Function
Output:
1 0 0 0 0 0 0 0
0 1 0 0 0 0 0 0
0 0 1 0 0 0 0 0
0 0 0 1 0 0 0 0
0 0 0 0 1 0 0 0
0 0 0 0 0 1 0 0
0 0 0 0 0 0 1 0
0 0 0 0 0 0 0 1

Visual Basic

Works with: Visual Basic version 6
Option Explicit
'------------
Public Function BuildIdentityMatrix(ByVal Size As Long) As Byte()
Dim i As Long
Dim b() As Byte

  Size = Size - 1
  ReDim b(0 To Size, 0 To Size)
  'at this point, the matrix is allocated and
  'all elements are initialized to 0 (zero)
  For i = 0 To Size
    b(i, i) = 1   'set diagonal elements to 1
  Next i
  BuildIdentityMatrix = b
  
End Function
'------------
Sub IdentityMatrixDemo(ByVal Size As Long)
Dim b() As Byte
Dim i As Long, j As Long

  b() = BuildIdentityMatrix(Size)
  For i = LBound(b(), 1) To UBound(b(), 1)
    For j = LBound(b(), 2) To UBound(b(), 2)
      Debug.Print CStr(b(i, j));
    Next j
  Debug.Print
  Next i

End Sub
'------------
Sub Main()

  IdentityMatrixDemo 5
  Debug.Print
  IdentityMatrixDemo 10

End Sub
Output:
10000
01000
00100
00010
00001

1000000000
0100000000
0010000000
0001000000
0000100000
0000010000
0000001000
0000000100
0000000010
0000000001

Wortel

@let {
  im ^(%^\@table ^(@+ =) @to)

  !im 4
}

Returns:

[[1 0 0 0]
 [0 1 0 0]
 [0 0 1 0]
 [0 0 0 1]]

Wren

Library: Wren-matrix
Library: Wren-fmt
import "./matrix" for Matrix
import "./fmt" for Fmt

var numRows = 10 // say
Fmt.mprint(Matrix.identity(numRows), 2, 0)
Output:
| 1  0  0  0  0  0  0  0  0  0|
| 0  1  0  0  0  0  0  0  0  0|
| 0  0  1  0  0  0  0  0  0  0|
| 0  0  0  1  0  0  0  0  0  0|
| 0  0  0  0  1  0  0  0  0  0|
| 0  0  0  0  0  1  0  0  0  0|
| 0  0  0  0  0  0  1  0  0  0|
| 0  0  0  0  0  0  0  1  0  0|
| 0  0  0  0  0  0  0  0  1  0|
| 0  0  0  0  0  0  0  0  0  1|

XPL0

include c:\cxpl\codes;
def IntSize = 4;                        \number of bytes in an integer
int Matrix, Size, I, J;

[Text(0, "Size: ");  Size:= IntIn(0);
Matrix:= Reserve(Size*IntSize);         \reserve memory for 2D integer array
for I:= 0 to Size-1 do
        Matrix(I):= Reserve(Size*IntSize);
for J:= 0 to Size-1 do                  \make array an identity matrix
    for I:= 0 to Size-1 do
        Matrix(I,J):= if I=J then 1 else 0;
for J:= 0 to Size-1 do                  \display the result
    [for I:= 0 to Size-1 do
        [IntOut(0, Matrix(I,J));  ChOut(0, ^ )];
    CrLf(0);
    ];
]
Output:
Size: 5
1 0 0 0 0 
0 1 0 0 0 
0 0 1 0 0 
0 0 0 1 0 
0 0 0 0 1 

zkl

Using lists of lists:

fcn idMatrix(n){
   m:=(0).pump(n,List.createLong(n).write,0)*n;
   m.apply2(fcn(row,rc){ row[rc.inc()]=1 },Ref(0));
   m
}
idMatrix(5).println();
idMatrix(5).pump(Console.println);
Output:
L(L(1,0,0,0,0),L(0,1,0,0,0),L(0,0,1,0,0),L(0,0,0,1,0),L(0,0,0,0,1))
L(1,0,0,0,0)
L(0,1,0,0,0)
L(0,0,1,0,0)
L(0,0,0,1,0)
L(0,0,0,0,1)

ZX Spectrum Basic

Translation of: Applesoft_BASIC
10 INPUT "Matrix size: ";size
20 GO SUB 200: REM Identity matrix
30 FOR r=1 TO size
40 FOR c=1 TO size
50 LET s$=CHR$ 13
60 IF c<size THEN LET s$=" "
70 PRINT i(r,c);s$;
80 NEXT c
90 NEXT r
100 STOP 
200 REM Identity matrix size
220 DIM i(size,size)
230 FOR i=1 TO size
240 LET i(i,i)=1
250 NEXT i
260 RETURN