test: migrate math/base/special/binomcoeff to ULP-based testing

PR-URL: #11720
Co-authored-by: Athan Reines <kgryte@gmail.com>
Reviewed-by: Athan Reines <kgryte@gmail.com> 
Ref: #11352

Author: MANDEep22332

lib/node_modules/@stdlib/math/base/special/binomcoeff/test/test.js

@@ -22,8 +22,7 @@
 
 var tape = require( 'tape' );
 var isnanf = require( '@stdlib/math/base/assert/is-nanf' );
-var absf = require( '@stdlib/math/base/special/absf' );
-var EPS = require( '@stdlib/constants/float32/eps' );
+var isAlmostSameValue = require( '@stdlib/number/float32/base/assert/is-almost-same-value' );
 var float64ToFloat32 = require( '@stdlib/number/float64/base/to-float32' );
 var binomcoeff = require( './../lib' );
 
@@ -54,8 +53,6 @@ tape( 'the function returns `NaN` if provided `NaN` for any parameter', function
 
 tape( 'the function evaluates the binomial coefficient for integers `n` and `k`', function test( t ) {
 	var expected;
-	var delta;
-	var tol;
 	var n;
 	var k;
 	var v;
@@ -67,15 +64,9 @@ tape( 'the function evaluates the binomial coefficient for integers `n` and `k`'
 	for ( i = 0; i < n.length; i++ ) {
 		v = binomcoeff( n[ i ], k[ i ] );
 		expected[ i ] = float64ToFloat32( expected[ i ] );
-		if ( expected[ i ] === v ) {
-			t.strictEqual( v, expected[ i ], 'returns expected value' );
-			continue;
-		}
-		delta = absf( v - expected[ i ] );
 
 		// NOTE: Exact comparison fails for large values due to single-precision floating-point rounding errors when intermediate results exceed the maximum safe integer for a 32-bit float.
-		tol = 1.25 * EPS * absf( expected[ i ] );
-		t.ok( delta <= tol, 'within tolerance. n: ' + n[ i ] + '. k: ' + k[ i ] + '. actual: ' + v + '. expected: ' + expected[ i ] + '. tol: ' + tol + '. Δ: ' + delta + '.' );
+		t.strictEqual( isAlmostSameValue( v, expected[ i ], 2 ), true, 'returns expected value' );
 	}
 	t.end();
 });
@@ -93,7 +84,7 @@ tape( 'the function evaluates the binomial coefficient for integers `n` and `k`
 	for ( i = 0; i < n.length; i++ ) {
 		v = binomcoeff( n[ i ], k[ i ] );
 		expected[ i ] = float64ToFloat32( expected[ i ] );
-		t.strictEqual( v, expected[ i ], 'returns expected value. actual: '+v+'. expected: '+expected[i]+'. n: '+n[i]+'. k: '+k[i]+'.' );
+		t.strictEqual( v, expected[ i ], 'returns expected value' );
 	}
 	t.end();
 });