diff --git a/lib/node_modules/@stdlib/blas/ext/base/snansumpw/README.md b/lib/node_modules/@stdlib/blas/ext/base/snansumpw/README.md
index 7eb812cb24fb..c59168902d34 100644
--- a/lib/node_modules/@stdlib/blas/ext/base/snansumpw/README.md
+++ b/lib/node_modules/@stdlib/blas/ext/base/snansumpw/README.md
@@ -36,7 +36,7 @@ limitations under the License.
var snansumpw = require( '@stdlib/blas/ext/base/snansumpw' );
```
-#### snansumpw( N, x, stride )
+#### snansumpw( N, x, strideX )
Computes the sum of single-precision floating-point strided array elements, ignoring `NaN` values and using pairwise summation.
@@ -44,9 +44,8 @@ Computes the sum of single-precision floating-point strided array elements, igno
var Float32Array = require( '@stdlib/array/float32' );
var x = new Float32Array( [ 1.0, -2.0, NaN, 2.0 ] );
-var N = x.length;
-var v = snansumpw( N, x, 1 );
+var v = snansumpw( x.length, x, 1 );
// returns 1.0
```
@@ -54,9 +53,9 @@ The function has the following parameters:
- **N**: number of indexed elements.
- **x**: input [`Float32Array`][@stdlib/array/float32].
-- **stride**: index increment for `x`.
+- **strideX**: stride length for `x`.
-The `N` and stride parameters determine which elements in the strided array are accessed at runtime. For example, to compute the sum of every other element in `x`,
+The `N` and stride parameters determine which elements in the strided array are accessed at runtime. For example, to compute the sum of every other element:
```javascript
var Float32Array = require( '@stdlib/array/float32' );
@@ -81,7 +80,7 @@ var v = snansumpw( 4, x1, 2 );
// returns 5.0
```
-#### snansumpw.ndarray( N, x, stride, offset )
+#### snansumpw.ndarray( N, x, strideX, offsetX )
Computes the sum of single-precision floating-point strided array elements, ignoring `NaN` values and using pairwise summation and alternative indexing semantics.
@@ -89,17 +88,16 @@ Computes the sum of single-precision floating-point strided array elements, igno
var Float32Array = require( '@stdlib/array/float32' );
var x = new Float32Array( [ 1.0, -2.0, NaN, 2.0 ] );
-var N = x.length;
-var v = snansumpw.ndarray( N, x, 1, 0 );
+var v = snansumpw.ndarray( x.length, x, 1, 0 );
// returns 1.0
```
The function has the following additional parameters:
-- **offset**: starting index for `x`.
+- **offsetX**: starting index for `x`.
-While [`typed array`][mdn-typed-array] views mandate a view offset based on the underlying `buffer`, the `offset` parameter supports indexing semantics based on a starting index. For example, to calculate the sum of every other value in `x` starting from the second value
+While [`typed array`][mdn-typed-array] views mandate a view offset based on the underlying buffer, the offset parameter supports indexing semantics based on a starting index. For example, to calculate the sum of every other element starting from the second element:
```javascript
var Float32Array = require( '@stdlib/array/float32' );
@@ -155,6 +153,123 @@ console.log( v );
+
+
+* * *
+
+
+
+## C APIs
+
+
+
+
+
+
+
+
+
+
+
+### Usage
+
+```c
+#include "stdlib/blas/ext/base/snansumpw.h"
+```
+
+#### stdlib_strided_snansumpw( N, \*X, strideX )
+
+Computes the sum of single-precision floating-point strided array elements, ignoring `NaN` values and using pairwise summation.
+
+```c
+const float x[] = { 1.0f, -2.0f, 0.0f/0.0f, 2.0f };
+
+float v = stdlib_strided_snansumpw( 4, x, 1 );
+// returns 1.0f
+```
+
+The function accepts the following arguments:
+
+- **N**: `[in] CBLAS_INT` number of indexed elements.
+- **X**: `[in] float*` input array.
+- **strideX**: `[in] CBLAS_INT` stride length for `X`.
+
+```c
+float stdlib_strided_snansumpw( const CBLAS_INT N, const float *X, const CBLAS_INT strideX );
+```
+
+#### stdlib_strided_snansumpw_ndarray( N, \*X, strideX, offsetX )
+
+Computes the sum of single-precision floating-point strided array elements, ignoring `NaN` values and using pairwise summation and alternative indexing semantics.
+
+```c
+const float x[] = { 1.0f, -2.0f, 0.0f/0.0f, 2.0f };
+
+float v = stdlib_strided_snansumpw_ndarray( 4, x, 1, 0 );
+// returns 1.0f
+```
+
+The function accepts the following arguments:
+
+- **N**: `[in] CBLAS_INT` number of indexed elements.
+- **X**: `[in] float*` input array.
+- **strideX**: `[in] CBLAS_INT` stride length for `X`.
+- **offsetX**: `[in] CBLAS_INT` starting index for `X`.
+
+```c
+float stdlib_strided_snansumpw_ndarray( const CBLAS_INT N, const float *X, const CBLAS_INT strideX, const CBLAS_INT offsetX );
+```
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+### Examples
+
+```c
+#include "stdlib/blas/ext/base/snansumpw.h"
+#include
+
+int main( void ) {
+ // Create a strided array:
+ const float x[] = { 1.0f, 2.0f, 3.0f, 4.0f, 5.0f, 6.0f, 7.0f, 8.0f, 0.0f/0.0f, 0.0f/0.0f };
+
+ // Specify the number of elements:
+ const int N = 5;
+
+ // Specify the stride length:
+ const int strideX = 2;
+
+ // Compute the sum:
+ float v = stdlib_strided_snansumpw( N, x, strideX );
+
+ // Print the result:
+ printf( "sum: %f\n", v );
+}
+```
+
+
+
+
+
+
+
+
+
* * *
diff --git a/lib/node_modules/@stdlib/blas/ext/base/snansumpw/benchmark/c/benchmark.length.c b/lib/node_modules/@stdlib/blas/ext/base/snansumpw/benchmark/c/benchmark.length.c
index d6d332eddd62..1fea25dbff38 100644
--- a/lib/node_modules/@stdlib/blas/ext/base/snansumpw/benchmark/c/benchmark.length.c
+++ b/lib/node_modules/@stdlib/blas/ext/base/snansumpw/benchmark/c/benchmark.length.c
@@ -94,7 +94,7 @@ static float rand_float( void ) {
* @param len array length
* @return elapsed time in seconds
*/
-static double benchmark( int iterations, int len ) {
+static double benchmark1( int iterations, int len ) {
double elapsed;
float x[ len ];
float v;
@@ -111,6 +111,7 @@ static double benchmark( int iterations, int len ) {
v = 0.0f;
t = tic();
for ( i = 0; i < iterations; i++ ) {
+ // cppcheck-suppress uninitvar
v = stdlib_strided_snansumpw( len, x, 1 );
if ( v != v ) {
printf( "should not return NaN\n" );
@@ -124,6 +125,44 @@ static double benchmark( int iterations, int len ) {
return elapsed;
}
+/**
+* Runs a benchmark.
+*
+* @param iterations number of iterations
+* @param len array length
+* @return elapsed time in seconds
+*/
+static double benchmark2( int iterations, int len ) {
+ double elapsed;
+ float x[ len ];
+ float v;
+ double t;
+ int i;
+
+ for ( i = 0; i < len; i++ ) {
+ if ( rand_float() < 0.2f ) {
+ x[ i ] = 0.0f / 0.0f; // NaN
+ } else {
+ x[ i ] = ( rand_float()*20000.0f ) - 10000.0f;
+ }
+ }
+ v = 0.0f;
+ t = tic();
+ for ( i = 0; i < iterations; i++ ) {
+ // cppcheck-suppress uninitvar
+ v = stdlib_strided_snansumpw_ndarray( len, x, 1, 0 );
+ if ( v != v ) {
+ printf( "should not return NaN\n" );
+ break;
+ }
+ }
+ elapsed = tic() - t;
+ if ( v != v ) {
+ printf( "should not return NaN\n" );
+ }
+ return elapsed;
+}
+
/**
* Main execution sequence.
*/
@@ -146,7 +185,18 @@ int main( void ) {
for ( j = 0; j < REPEATS; j++ ) {
count += 1;
printf( "# c::%s:len=%d\n", NAME, len );
- elapsed = benchmark( iter, len );
+ elapsed = benchmark1( iter, len );
+ print_results( iter, elapsed );
+ printf( "ok %d benchmark finished\n", count );
+ }
+ }
+ for ( i = MIN; i <= MAX; i++ ) {
+ len = pow( 10, i );
+ iter = ITERATIONS / pow( 10, i-1 );
+ for ( j = 0; j < REPEATS; j++ ) {
+ count += 1;
+ printf( "# c::%s:ndarray:len=%d\n", NAME, len );
+ elapsed = benchmark2( iter, len );
print_results( iter, elapsed );
printf( "ok %d benchmark finished\n", count );
}
diff --git a/lib/node_modules/@stdlib/blas/ext/base/snansumpw/docs/repl.txt b/lib/node_modules/@stdlib/blas/ext/base/snansumpw/docs/repl.txt
index 9b9cdb809479..abaf94583ad7 100644
--- a/lib/node_modules/@stdlib/blas/ext/base/snansumpw/docs/repl.txt
+++ b/lib/node_modules/@stdlib/blas/ext/base/snansumpw/docs/repl.txt
@@ -1,10 +1,10 @@
-{{alias}}( N, x, stride )
+{{alias}}( N, x, strideX )
Computes the sum of single-precision floating-point strided array elements,
ignoring `NaN` values and using pairwise summation.
- The `N` and `stride` parameters determine which elements in the strided
- array are accessed at runtime.
+ The `N` and stride parameters determine which elements in the strided array
+ are accessed at runtime.
Indexing is relative to the first index. To introduce an offset, use a typed
array view.
@@ -19,8 +19,8 @@
x: Float32Array
Input array.
- stride: integer
- Index increment.
+ strideX: integer
+ Stride length.
Returns
-------
@@ -34,7 +34,7 @@
> {{alias}}( x.length, x, 1 )
1.0
- // Using `N` and `stride` parameters:
+ // Using `N` and stride parameters:
> x = new {{alias:@stdlib/array/float32}}( [ -2.0, 1.0, 1.0, -5.0, 2.0, -1.0, NaN, NaN ] );
> {{alias}}( 4, x, 2 )
1.0
@@ -46,13 +46,13 @@
-1.0
-{{alias}}.ndarray( N, x, stride, offset )
+{{alias}}.ndarray( N, x, strideX, offsetX )
Computes the sum of single-precision floating-point strided array elements,
ignoring `NaN` values and using pairwise summation and alternative indexing
semantics.
While typed array views mandate a view offset based on the underlying
- buffer, the `offset` parameter supports indexing semantics based on a
+ buffer, the offset parameter supports indexing semantics based on a
starting index.
Parameters
@@ -63,10 +63,10 @@
x: Float32Array
Input array.
- stride: integer
- Index increment.
+ strideX: integer
+ Stride length.
- offset: integer
+ offsetX: integer
Starting index.
Returns
diff --git a/lib/node_modules/@stdlib/blas/ext/base/snansumpw/docs/types/index.d.ts b/lib/node_modules/@stdlib/blas/ext/base/snansumpw/docs/types/index.d.ts
index 7af6d963af56..72820f46bdc3 100644
--- a/lib/node_modules/@stdlib/blas/ext/base/snansumpw/docs/types/index.d.ts
+++ b/lib/node_modules/@stdlib/blas/ext/base/snansumpw/docs/types/index.d.ts
@@ -27,7 +27,7 @@ interface Routine {
*
* @param N - number of indexed elements
* @param x - input array
- * @param stride - stride length
+ * @param strideX - stride length
* @returns sum
*
* @example
@@ -38,15 +38,15 @@ interface Routine {
* var v = snansumpw( x.length, x, 1 );
* // returns 1.0
*/
- ( N: number, x: Float32Array, stride: number ): number;
+ ( N: number, x: Float32Array, strideX: number ): number;
/**
* Computes the sum of single-precision floating-point strided array elements, ignoring `NaN` values and using pairwise summation and alternative indexing semantics.
*
* @param N - number of indexed elements
* @param x - input array
- * @param stride - stride length
- * @param offset - starting index
+ * @param strideX - stride length
+ * @param offsetX - starting index
* @returns sum
*
* @example
@@ -57,7 +57,7 @@ interface Routine {
* var v = snansumpw.ndarray( x.length, x, 1, 0 );
* // returns 1.0
*/
- ndarray( N: number, x: Float32Array, stride: number, offset: number ): number;
+ ndarray( N: number, x: Float32Array, strideX: number, offsetX: number ): number;
}
/**
@@ -65,7 +65,7 @@ interface Routine {
*
* @param N - number of indexed elements
* @param x - input array
-* @param stride - stride length
+* @param strideX - stride length
* @returns sum
*
* @example
diff --git a/lib/node_modules/@stdlib/blas/ext/base/snansumpw/examples/c/example.c b/lib/node_modules/@stdlib/blas/ext/base/snansumpw/examples/c/example.c
index e6f10680c330..c2cf4f3ef8cc 100644
--- a/lib/node_modules/@stdlib/blas/ext/base/snansumpw/examples/c/example.c
+++ b/lib/node_modules/@stdlib/blas/ext/base/snansumpw/examples/c/example.c
@@ -17,21 +17,20 @@
*/
#include "stdlib/blas/ext/base/snansumpw.h"
-#include
#include
int main( void ) {
// Create a strided array:
- const float x[] = { 1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0, 0.0/0.0, 0.0/0.0 };
+ const float x[] = { 1.0f, 2.0f, 3.0f, 4.0f, 5.0f, 6.0f, 7.0f, 8.0f, 0.0f/0.0f, 0.0f/0.0f };
// Specify the number of elements:
- const int64_t N = 5;
+ const int N = 5;
// Specify the stride length:
- const int64_t stride = 2;
+ const int strideX = 2;
// Compute the sum:
- float v = stdlib_strided_snansumpw( N, x, stride );
+ float v = stdlib_strided_snansumpw( N, x, strideX );
// Print the result:
printf( "sum: %f\n", v );
diff --git a/lib/node_modules/@stdlib/blas/ext/base/snansumpw/include/stdlib/blas/ext/base/snansumpw.h b/lib/node_modules/@stdlib/blas/ext/base/snansumpw/include/stdlib/blas/ext/base/snansumpw.h
index 2aa43fa48c28..ac3c7dc9107c 100644
--- a/lib/node_modules/@stdlib/blas/ext/base/snansumpw/include/stdlib/blas/ext/base/snansumpw.h
+++ b/lib/node_modules/@stdlib/blas/ext/base/snansumpw/include/stdlib/blas/ext/base/snansumpw.h
@@ -19,7 +19,7 @@
#ifndef STDLIB_BLAS_EXT_BASE_SNANSUMPW_H
#define STDLIB_BLAS_EXT_BASE_SNANSUMPW_H
-#include
+#include "stdlib/blas/base/shared.h"
/*
* If C++, prevent name mangling so that the compiler emits a binary file having undecorated names, thus mirroring the behavior of a C compiler.
@@ -31,7 +31,12 @@ extern "C" {
/**
* Computes the sum of single-precision floating-point strided array elements, ignoring `NaN` values and using pairwise summation.
*/
-float stdlib_strided_snansumpw( const int64_t N, const float *X, const int64_t stride );
+float API_SUFFIX(stdlib_strided_snansumpw)( const CBLAS_INT N, const float *X, const CBLAS_INT strideX );
+
+/**
+* Computes the sum of single-precision floating-point strided array elements, ignoring `NaN` values and using pairwise summation and alternative indexing semantics.
+*/
+float API_SUFFIX(stdlib_strided_snansumpw_ndarray)( const CBLAS_INT N, const float *X, const CBLAS_INT strideX, const CBLAS_INT offsetX );
#ifdef __cplusplus
}
diff --git a/lib/node_modules/@stdlib/blas/ext/base/snansumpw/lib/index.js b/lib/node_modules/@stdlib/blas/ext/base/snansumpw/lib/index.js
index 747491940a1f..edda01b74349 100644
--- a/lib/node_modules/@stdlib/blas/ext/base/snansumpw/lib/index.js
+++ b/lib/node_modules/@stdlib/blas/ext/base/snansumpw/lib/index.js
@@ -28,9 +28,8 @@
* var snansumpw = require( '@stdlib/blas/ext/base/snansumpw' );
*
* var x = new Float32Array( [ 1.0, -2.0, NaN, 2.0 ] );
-* var N = x.length;
*
-* var v = snansumpw( N, x, 1 );
+* var v = snansumpw( x.length, x, 1 );
* // returns 1.0
*
* @example
diff --git a/lib/node_modules/@stdlib/blas/ext/base/snansumpw/lib/ndarray.js b/lib/node_modules/@stdlib/blas/ext/base/snansumpw/lib/ndarray.js
index 48a3a93cba1a..1ccfd558901c 100644
--- a/lib/node_modules/@stdlib/blas/ext/base/snansumpw/lib/ndarray.js
+++ b/lib/node_modules/@stdlib/blas/ext/base/snansumpw/lib/ndarray.js
@@ -46,8 +46,8 @@ var BLOCKSIZE = 128;
*
* @param {PositiveInteger} N - number of indexed elements
* @param {Float32Array} x - input array
-* @param {integer} stride - stride length
-* @param {NonNegativeInteger} offset - starting index
+* @param {integer} strideX - stride length
+* @param {NonNegativeInteger} offsetX - starting index
* @returns {number} sum
*
* @example
@@ -58,7 +58,7 @@ var BLOCKSIZE = 128;
* var v = snansumpw( 5, x, 2, 1 );
* // returns 5.0
*/
-function snansumpw( N, x, stride, offset ) {
+function snansumpw( N, x, strideX, offsetX ) {
var ix;
var s0;
var s1;
@@ -76,13 +76,13 @@ function snansumpw( N, x, stride, offset ) {
if ( N <= 0 ) {
return 0.0;
}
- if ( N === 1 || stride === 0 ) {
- if ( isnanf( x[ offset ] ) ) {
+ ix = offsetX;
+ if ( strideX === 0 ) {
+ if ( isnanf( x[ ix ] ) ) {
return 0.0;
}
- return x[ offset ];
+ return N * x[ ix ];
}
- ix = offset;
if ( N < 8 ) {
// Use simple summation...
s = 0.0;
@@ -90,47 +90,47 @@ function snansumpw( N, x, stride, offset ) {
if ( isnanf( x[ ix ] ) === false ) {
s = float64ToFloat32( s + x[ ix ] );
}
- ix += stride;
+ ix += strideX;
}
return s;
}
if ( N <= BLOCKSIZE ) {
// Sum a block with 8 accumulators (by loop unrolling, we lower the effective blocksize to 16)...
s0 = ( isnanf( x[ ix ] ) ) ? 0.0 : x[ ix ];
- ix += stride;
+ ix += strideX;
s1 = ( isnanf( x[ ix ] ) ) ? 0.0 : x[ ix ];
- ix += stride;
+ ix += strideX;
s2 = ( isnanf( x[ ix ] ) ) ? 0.0 : x[ ix ];
- ix += stride;
+ ix += strideX;
s3 = ( isnanf( x[ ix ] ) ) ? 0.0 : x[ ix ];
- ix += stride;
+ ix += strideX;
s4 = ( isnanf( x[ ix ] ) ) ? 0.0 : x[ ix ];
- ix += stride;
+ ix += strideX;
s5 = ( isnanf( x[ ix ] ) ) ? 0.0 : x[ ix ];
- ix += stride;
+ ix += strideX;
s6 = ( isnanf( x[ ix ] ) ) ? 0.0 : x[ ix ];
- ix += stride;
+ ix += strideX;
s7 = ( isnanf( x[ ix ] ) ) ? 0.0 : x[ ix ];
- ix += stride;
+ ix += strideX;
M = N % 8;
for ( i = 8; i < N-M; i += 8 ) {
s0 = ( isnanf( x[ ix ] ) ) ? s0 : float64ToFloat32( s0 + x[ ix ] );
- ix += stride;
+ ix += strideX;
s1 = ( isnanf( x[ ix ] ) ) ? s1 : float64ToFloat32( s1 + x[ ix ] );
- ix += stride;
+ ix += strideX;
s2 = ( isnanf( x[ ix ] ) ) ? s2 : float64ToFloat32( s2 + x[ ix ] );
- ix += stride;
+ ix += strideX;
s3 = ( isnanf( x[ ix ] ) ) ? s3 : float64ToFloat32( s3 + x[ ix ] );
- ix += stride;
+ ix += strideX;
s4 = ( isnanf( x[ ix ] ) ) ? s4 : float64ToFloat32( s4 + x[ ix ] );
- ix += stride;
+ ix += strideX;
s5 = ( isnanf( x[ ix ] ) ) ? s5 : float64ToFloat32( s5 + x[ ix ] );
- ix += stride;
+ ix += strideX;
s6 = ( isnanf( x[ ix ] ) ) ? s6 : float64ToFloat32( s6 + x[ ix ] );
- ix += stride;
+ ix += strideX;
s7 = ( isnanf( x[ ix ] ) ) ? s7 : float64ToFloat32( s7 + x[ ix ] );
- ix += stride;
+ ix += strideX;
}
// Pairwise sum the accumulators:
s = float64ToFloat32( float64ToFloat32( float64ToFloat32(s0+s1) + float64ToFloat32(s2+s3) ) + float64ToFloat32( float64ToFloat32(s4+s5) + float64ToFloat32(s6+s7) ) ); // eslint-disable-line max-len
@@ -140,14 +140,14 @@ function snansumpw( N, x, stride, offset ) {
if ( isnanf( x[ ix ] ) === false ) {
s = float64ToFloat32( s + x[ ix ] );
}
- ix += stride;
+ ix += strideX;
}
return s;
}
// Recurse by dividing by two, but avoiding non-multiples of unroll factor...
n = floor( N/2 );
n -= n % 8;
- return float64ToFloat32( snansumpw( n, x, stride, ix ) + snansumpw( N-n, x, stride, ix+(n*stride) ) ); // eslint-disable-line max-len
+ return float64ToFloat32( snansumpw( n, x, strideX, ix ) + snansumpw( N-n, x, strideX, ix+(n*strideX) ) ); // eslint-disable-line max-len
}
diff --git a/lib/node_modules/@stdlib/blas/ext/base/snansumpw/lib/ndarray.native.js b/lib/node_modules/@stdlib/blas/ext/base/snansumpw/lib/ndarray.native.js
index 1c8a9ae57c4e..f2d003cd33f0 100644
--- a/lib/node_modules/@stdlib/blas/ext/base/snansumpw/lib/ndarray.native.js
+++ b/lib/node_modules/@stdlib/blas/ext/base/snansumpw/lib/ndarray.native.js
@@ -20,9 +20,7 @@
// MODULES //
-var minViewBufferIndex = require( '@stdlib/strided/base/min-view-buffer-index' );
-var offsetView = require( '@stdlib/strided/base/offset-view' );
-var addon = require( './snansumpw.native.js' );
+var addon = require( './../src/addon.node' );
// MAIN //
@@ -32,8 +30,8 @@ var addon = require( './snansumpw.native.js' );
*
* @param {PositiveInteger} N - number of indexed elements
* @param {Float32Array} x - input array
-* @param {integer} stride - stride length
-* @param {NonNegativeInteger} offset - starting index
+* @param {integer} strideX - stride length
+* @param {NonNegativeInteger} offsetX - starting index
* @returns {number} sum
*
* @example
@@ -44,11 +42,8 @@ var addon = require( './snansumpw.native.js' );
* var v = snansumpw( 5, x, 2, 1 );
* // returns 5.0
*/
-function snansumpw( N, x, stride, offset ) {
- var view;
- offset = minViewBufferIndex( N, stride, offset );
- view = offsetView( x, offset );
- return addon( N, view, stride );
+function snansumpw( N, x, strideX, offsetX ) {
+ return addon.ndarray( N, x, strideX, offsetX );
}
diff --git a/lib/node_modules/@stdlib/blas/ext/base/snansumpw/lib/snansumpw.js b/lib/node_modules/@stdlib/blas/ext/base/snansumpw/lib/snansumpw.js
index 545a972e98a2..926732b90e52 100644
--- a/lib/node_modules/@stdlib/blas/ext/base/snansumpw/lib/snansumpw.js
+++ b/lib/node_modules/@stdlib/blas/ext/base/snansumpw/lib/snansumpw.js
@@ -20,9 +20,8 @@
// MODULES //
-var float64ToFloat32 = require( '@stdlib/number/float64/base/to-float32' );
-var isnanf = require( '@stdlib/math/base/assert/is-nanf' );
-var sum = require( './ndarray.js' );
+var stride2offset = require( '@stdlib/strided/base/stride2offset' );
+var ndarray = require( './ndarray.js' );
// MAIN //
@@ -40,49 +39,19 @@ var sum = require( './ndarray.js' );
*
* @param {PositiveInteger} N - number of indexed elements
* @param {Float32Array} x - input array
-* @param {integer} stride - stride length
+* @param {integer} strideX - stride length
* @returns {number} sum
*
* @example
* var Float32Array = require( '@stdlib/array/float32' );
*
* var x = new Float32Array( [ 1.0, -2.0, NaN, 2.0 ] );
-* var N = x.length;
*
-* var v = snansumpw( N, x, 1 );
+* var v = snansumpw( x.length, x, 1 );
* // returns 1.0
*/
-function snansumpw( N, x, stride ) {
- var ix;
- var s;
- var i;
-
- if ( N <= 0 ) {
- return 0.0;
- }
- if ( N === 1 || stride === 0 ) {
- if ( isnanf( x[ 0 ] ) ) {
- return 0.0;
- }
- return x[ 0 ];
- }
- if ( stride < 0 ) {
- ix = (1-N) * stride;
- } else {
- ix = 0;
- }
- if ( N < 8 ) {
- // Use simple summation...
- s = 0.0;
- for ( i = 0; i < N; i++ ) {
- if ( isnanf( x[ ix ] ) === false ) {
- s = float64ToFloat32( s + x[ ix ] );
- }
- ix += stride;
- }
- return s;
- }
- return sum( N, x, stride, ix );
+function snansumpw( N, x, strideX ) {
+ return ndarray( N, x, strideX, stride2offset( N, strideX ) );
}
diff --git a/lib/node_modules/@stdlib/blas/ext/base/snansumpw/lib/snansumpw.native.js b/lib/node_modules/@stdlib/blas/ext/base/snansumpw/lib/snansumpw.native.js
index bedd15436359..622dbc406aad 100644
--- a/lib/node_modules/@stdlib/blas/ext/base/snansumpw/lib/snansumpw.native.js
+++ b/lib/node_modules/@stdlib/blas/ext/base/snansumpw/lib/snansumpw.native.js
@@ -30,20 +30,19 @@ var addon = require( './../src/addon.node' );
*
* @param {PositiveInteger} N - number of indexed elements
* @param {Float32Array} x - input array
-* @param {integer} stride - stride length
+* @param {integer} strideX - stride length
* @returns {number} sum
*
* @example
* var Float32Array = require( '@stdlib/array/float32' );
*
* var x = new Float32Array( [ 1.0, -2.0, NaN, 2.0 ] );
-* var N = x.length;
*
-* var v = snansumpw( N, x, 1 );
+* var v = snansumpw( x.length, x, 1 );
* // returns 1.0
*/
-function snansumpw( N, x, stride ) {
- return addon( N, x, stride );
+function snansumpw( N, x, strideX ) {
+ return addon( N, x, strideX );
}
diff --git a/lib/node_modules/@stdlib/blas/ext/base/snansumpw/manifest.json b/lib/node_modules/@stdlib/blas/ext/base/snansumpw/manifest.json
index af3ddac3e5f2..16bf93a5a65d 100644
--- a/lib/node_modules/@stdlib/blas/ext/base/snansumpw/manifest.json
+++ b/lib/node_modules/@stdlib/blas/ext/base/snansumpw/manifest.json
@@ -28,14 +28,12 @@
{
"task": "build",
"src": [
- "./src/snansumpw.c"
+ "./src/main.c"
],
"include": [
"./include"
],
- "libraries": [
- "-lm"
- ],
+ "libraries": [],
"libpath": [],
"dependencies": [
"@stdlib/math/base/assert/is-nanf",
@@ -43,39 +41,42 @@
"@stdlib/napi/argv",
"@stdlib/napi/argv-float",
"@stdlib/napi/argv-int64",
- "@stdlib/napi/argv-strided-float32array"
+ "@stdlib/napi/argv-strided-float32array",
+ "@stdlib/napi/create-double",
+ "@stdlib/blas/base/shared",
+ "@stdlib/strided/base/stride2offset"
]
},
{
"task": "benchmark",
"src": [
- "./src/snansumpw.c"
+ "./src/main.c"
],
"include": [
"./include"
],
- "libraries": [
- "-lm"
- ],
+ "libraries": [],
"libpath": [],
"dependencies": [
- "@stdlib/math/base/assert/is-nanf"
+ "@stdlib/math/base/assert/is-nanf",
+ "@stdlib/blas/base/shared",
+ "@stdlib/strided/base/stride2offset"
]
},
{
"task": "examples",
"src": [
- "./src/snansumpw.c"
+ "./src/main.c"
],
"include": [
"./include"
],
- "libraries": [
- "-lm"
- ],
+ "libraries": [],
"libpath": [],
"dependencies": [
- "@stdlib/math/base/assert/is-nanf"
+ "@stdlib/math/base/assert/is-nanf",
+ "@stdlib/blas/base/shared",
+ "@stdlib/strided/base/stride2offset"
]
}
]
diff --git a/lib/node_modules/@stdlib/blas/ext/base/snansumpw/src/addon.c b/lib/node_modules/@stdlib/blas/ext/base/snansumpw/src/addon.c
index dc9b86e54bbe..4149c3d24c42 100644
--- a/lib/node_modules/@stdlib/blas/ext/base/snansumpw/src/addon.c
+++ b/lib/node_modules/@stdlib/blas/ext/base/snansumpw/src/addon.c
@@ -17,12 +17,13 @@
*/
#include "stdlib/blas/ext/base/snansumpw.h"
+#include "stdlib/blas/base/shared.h"
#include "stdlib/napi/export.h"
#include "stdlib/napi/argv.h"
#include "stdlib/napi/argv_int64.h"
#include "stdlib/napi/argv_strided_float32array.h"
+#include "stdlib/napi/create_double.h"
#include
-#include
/**
* Receives JavaScript callback invocation data.
@@ -34,14 +35,27 @@
static napi_value addon( napi_env env, napi_callback_info info ) {
STDLIB_NAPI_ARGV( env, info, argv, argc, 3 );
STDLIB_NAPI_ARGV_INT64( env, N, argv, 0 );
- STDLIB_NAPI_ARGV_INT64( env, stride, argv, 2 );
- STDLIB_NAPI_ARGV_STRIDED_FLOAT32ARRAY( env, X, N, stride, argv, 1 );
-
- napi_value v;
- napi_status status = napi_create_double( env, stdlib_strided_snansumpw( N, X, stride ), &v );
- assert( status == napi_ok );
+ STDLIB_NAPI_ARGV_INT64( env, strideX, argv, 2 );
+ STDLIB_NAPI_ARGV_STRIDED_FLOAT32ARRAY( env, X, N, strideX, argv, 1 );
+ STDLIB_NAPI_CREATE_DOUBLE( env, (double)API_SUFFIX(stdlib_strided_snansumpw)( N, X, strideX ), v );
+ return v;
+}
+/**
+* Receives JavaScript callback invocation data.
+*
+* @param env environment under which the function is invoked
+* @param info callback data
+* @return Node-API value
+*/
+static napi_value addon_method( napi_env env, napi_callback_info info ) {
+ STDLIB_NAPI_ARGV( env, info, argv, argc, 4 );
+ STDLIB_NAPI_ARGV_INT64( env, N, argv, 0 );
+ STDLIB_NAPI_ARGV_INT64( env, strideX, argv, 2 );
+ STDLIB_NAPI_ARGV_INT64( env, offsetX, argv, 3 );
+ STDLIB_NAPI_ARGV_STRIDED_FLOAT32ARRAY( env, X, N, strideX, argv, 1 );
+ STDLIB_NAPI_CREATE_DOUBLE( env, (double)API_SUFFIX(stdlib_strided_snansumpw_ndarray)( N, X, strideX, offsetX ), v );
return v;
}
-STDLIB_NAPI_MODULE_EXPORT_FCN( addon )
+STDLIB_NAPI_MODULE_EXPORT_FCN_WITH_METHOD( addon, "ndarray", addon_method );
diff --git a/lib/node_modules/@stdlib/blas/ext/base/snansumpw/src/snansumpw.c b/lib/node_modules/@stdlib/blas/ext/base/snansumpw/src/main.c
similarity index 61%
rename from lib/node_modules/@stdlib/blas/ext/base/snansumpw/src/snansumpw.c
rename to lib/node_modules/@stdlib/blas/ext/base/snansumpw/src/main.c
index fbede2c45824..fe2fe34f8f4e 100644
--- a/lib/node_modules/@stdlib/blas/ext/base/snansumpw/src/snansumpw.c
+++ b/lib/node_modules/@stdlib/blas/ext/base/snansumpw/src/main.c
@@ -18,7 +18,8 @@
#include "stdlib/blas/ext/base/snansumpw.h"
#include "stdlib/math/base/assert/is_nanf.h"
-#include
+#include "stdlib/blas/base/shared.h"
+#include "stdlib/strided/base/stride2offset.h"
/**
* Computes the sum of single-precision floating-point strided array elements, ignoring `NaN` values and using pairwise summation.
@@ -31,18 +32,38 @@
*
* - Higham, Nicholas J. 1993. "The Accuracy of Floating Point Summation." _SIAM Journal on Scientific Computing_ 14 (4): 783–99. doi:[10.1137/0914050](https://doi.org/10.1137/0914050).
*
-* @param N number of indexed elements
-* @param X input array
-* @param stride stride length
-* @return output value
+* @param N number of indexed elements
+* @param X input array
+* @param strideX stride length
+* @return output value
*/
-float stdlib_strided_snansumpw( const int64_t N, const float *X, const int64_t stride ) {
- float *xp1;
- float *xp2;
- int64_t ix;
- int64_t M;
- int64_t n;
- int64_t i;
+float API_SUFFIX(stdlib_strided_snansumpw)( const CBLAS_INT N, const float *X, const CBLAS_INT strideX ) {
+ CBLAS_INT ox = stdlib_strided_stride2offset( N, strideX );
+ return API_SUFFIX(stdlib_strided_snansumpw_ndarray)( N, X, strideX, ox );
+}
+
+/**
+* Computes the sum of single-precision floating-point strided array elements, ignoring `NaN` values and using pairwise summation and alternative indexing semantics.
+*
+* ## Method
+*
+* - This implementation uses pairwise summation, which accrues rounding error `O(log2 N)` instead of `O(N)`. The recursion depth is also `O(log2 N)`.
+*
+* ## References
+*
+* - Higham, Nicholas J. 1993. "The Accuracy of Floating Point Summation." _SIAM Journal on Scientific Computing_ 14 (4): 783–99. doi:[10.1137/0914050](https://doi.org/10.1137/0914050).
+*
+* @param N number of indexed elements
+* @param X input array
+* @param strideX stride length
+* @param offsetX starting index
+* @return output value
+*/
+float API_SUFFIX(stdlib_strided_snansumpw_ndarray)( const CBLAS_INT N, const float *X, const CBLAS_INT strideX, const CBLAS_INT offsetX ) {
+ CBLAS_INT ix;
+ CBLAS_INT M;
+ CBLAS_INT n;
+ CBLAS_INT i;
float sum;
float s0;
float s1;
@@ -56,16 +77,12 @@ float stdlib_strided_snansumpw( const int64_t N, const float *X, const int64_t s
if ( N <= 0 ) {
return 0.0f;
}
- if ( N == 1 || stride == 0 ) {
- if ( stdlib_base_is_nanf( X[ 0 ] ) ) {
+ ix = offsetX;
+ if ( strideX == 0 ) {
+ if ( stdlib_base_is_nanf( X[ ix ] ) ) {
return 0.0f;
}
- return X[ 0 ];
- }
- if ( stride < 0 ) {
- ix = (1-N) * stride;
- } else {
- ix = 0;
+ return N * X[ ix ];
}
if ( N < 8 ) {
// Use simple summation...
@@ -74,7 +91,7 @@ float stdlib_strided_snansumpw( const int64_t N, const float *X, const int64_t s
if ( !stdlib_base_is_nanf( X[ ix ] ) ) {
sum += X[ ix ];
}
- ix += stride;
+ ix += strideX;
}
return sum;
}
@@ -82,62 +99,55 @@ float stdlib_strided_snansumpw( const int64_t N, const float *X, const int64_t s
if ( N <= 128 ) {
// Sum a block with 8 accumulators (by loop unrolling, we lower the effective blocksize to 16)...
s0 = ( stdlib_base_is_nanf( X[ ix ] ) ) ? 0.0 : X[ ix ];
- ix += stride;
+ ix += strideX;
s1 = ( stdlib_base_is_nanf( X[ ix ] ) ) ? 0.0 : X[ ix ];
- ix += stride;
+ ix += strideX;
s2 = ( stdlib_base_is_nanf( X[ ix ] ) ) ? 0.0 : X[ ix ];
- ix += stride;
+ ix += strideX;
s3 = ( stdlib_base_is_nanf( X[ ix ] ) ) ? 0.0 : X[ ix ];
- ix += stride;
+ ix += strideX;
s4 = ( stdlib_base_is_nanf( X[ ix ] ) ) ? 0.0 : X[ ix ];
- ix += stride;
+ ix += strideX;
s5 = ( stdlib_base_is_nanf( X[ ix ] ) ) ? 0.0 : X[ ix ];
- ix += stride;
+ ix += strideX;
s6 = ( stdlib_base_is_nanf( X[ ix ] ) ) ? 0.0 : X[ ix ];
- ix += stride;
+ ix += strideX;
s7 = ( stdlib_base_is_nanf( X[ ix ] ) ) ? 0.0 : X[ ix ];
- ix += stride;
+ ix += strideX;
M = N % 8;
for ( i = 8; i < N-M; i += 8 ) {
s0 += ( stdlib_base_is_nanf( X[ ix ] ) ) ? 0.0 : X[ ix ];
- ix += stride;
+ ix += strideX;
s1 += ( stdlib_base_is_nanf( X[ ix ] ) ) ? 0.0 : X[ ix ];
- ix += stride;
+ ix += strideX;
s2 += ( stdlib_base_is_nanf( X[ ix ] ) ) ? 0.0 : X[ ix ];
- ix += stride;
+ ix += strideX;
s3 += ( stdlib_base_is_nanf( X[ ix ] ) ) ? 0.0 : X[ ix ];
- ix += stride;
+ ix += strideX;
s4 += ( stdlib_base_is_nanf( X[ ix ] ) ) ? 0.0 : X[ ix ];
- ix += stride;
+ ix += strideX;
s5 += ( stdlib_base_is_nanf( X[ ix ] ) ) ? 0.0 : X[ ix ];
- ix += stride;
+ ix += strideX;
s6 += ( stdlib_base_is_nanf( X[ ix ] ) ) ? 0.0 : X[ ix ];
- ix += stride;
+ ix += strideX;
s7 += ( stdlib_base_is_nanf( X[ ix ] ) ) ? 0.0 : X[ ix ];
- ix += stride;
+ ix += strideX;
}
// Pairwise sum the accumulators:
- sum = ((s0+s1) + (s2+s3)) + ((s4+s5) + (s6+s7));
+ sum = ( (s0+s1) + (s2+s3) ) + ( (s4+s5) + (s6+s7) );
// Clean-up loop...
for (; i < N; i++ ) {
if ( !stdlib_base_is_nanf( X[ ix ] ) ) {
sum += X[ ix ];
}
- ix += stride;
+ ix += strideX;
}
return sum;
}
// Recurse by dividing by two, but avoiding non-multiples of unroll factor...
n = N / 2;
n -= n % 8;
- if ( stride < 0 ) {
- xp1 = (float *)X + ( (n-N)*stride );
- xp2 = (float *)X;
- } else {
- xp1 = (float *)X;
- xp2 = (float *)X + ( n*stride );
- }
- return stdlib_strided_snansumpw( n, xp1, stride ) + stdlib_strided_snansumpw( N-n, xp2, stride );
+ return stdlib_strided_snansumpw_ndarray( n, X, strideX, ix ) + stdlib_strided_snansumpw_ndarray( N-n, X, strideX, ix+(n*strideX) );
}
diff --git a/lib/node_modules/@stdlib/blas/ext/base/snansumpw/test/test.ndarray.js b/lib/node_modules/@stdlib/blas/ext/base/snansumpw/test/test.ndarray.js
index 56e17b02ae34..f31f475bbac2 100644
--- a/lib/node_modules/@stdlib/blas/ext/base/snansumpw/test/test.ndarray.js
+++ b/lib/node_modules/@stdlib/blas/ext/base/snansumpw/test/test.ndarray.js
@@ -178,14 +178,26 @@ tape( 'the function supports a negative `stride` parameter', function test( t )
t.end();
});
-tape( 'if provided a `stride` parameter equal to `0`, the function returns the first indexed element', function test( t ) {
+tape( 'if provided a `stride` parameter equal to `0`, the function returns the sum of the first element repeated N times', function test( t ) {
var x;
var v;
x = new Float32Array( [ 1.0, -2.0, -4.0, 5.0, 3.0 ] );
v = snansumpw( x.length, x, 0, 0 );
- t.strictEqual( v, 1.0, 'returns expected value' );
+ t.strictEqual( v, 5.0, 'returns expected value' );
+
+ t.end();
+});
+
+tape( 'if provided a `stride` parameter equal to `0` and the first element is `NaN`, the function returns 0.0', function test( t ) {
+ var x;
+ var v;
+
+ x = new Float32Array( [ NaN, -2.0, -4.0, 5.0, 3.0 ] );
+
+ v = snansumpw( x.length, x, 0, 0 );
+ t.strictEqual( v, 0.0, 'returns expected value' );
t.end();
});
diff --git a/lib/node_modules/@stdlib/blas/ext/base/snansumpw/test/test.ndarray.native.js b/lib/node_modules/@stdlib/blas/ext/base/snansumpw/test/test.ndarray.native.js
index 98667801d815..713dee042832 100644
--- a/lib/node_modules/@stdlib/blas/ext/base/snansumpw/test/test.ndarray.native.js
+++ b/lib/node_modules/@stdlib/blas/ext/base/snansumpw/test/test.ndarray.native.js
@@ -187,14 +187,26 @@ tape( 'the function supports a negative `stride` parameter', opts, function test
t.end();
});
-tape( 'if provided a `stride` parameter equal to `0`, the function returns the first indexed element', opts, function test( t ) {
+tape( 'if provided a `stride` parameter equal to `0`, the function returns the sum of the first element repeated N times', opts, function test( t ) {
var x;
var v;
x = new Float32Array( [ 1.0, -2.0, -4.0, 5.0, 3.0 ] );
v = snansumpw( x.length, x, 0, 0 );
- t.strictEqual( v, 1.0, 'returns expected value' );
+ t.strictEqual( v, 5.0, 'returns expected value' );
+
+ t.end();
+});
+
+tape( 'if provided a `stride` parameter equal to `0` and the first element is `NaN`, the function returns 0.0', opts, function test( t ) {
+ var x;
+ var v;
+
+ x = new Float32Array( [ NaN, -2.0, -4.0, 5.0, 3.0 ] );
+
+ v = snansumpw( x.length, x, 0, 0 );
+ t.strictEqual( v, 0.0, 'returns expected value' );
t.end();
});
diff --git a/lib/node_modules/@stdlib/blas/ext/base/snansumpw/test/test.snansumpw.js b/lib/node_modules/@stdlib/blas/ext/base/snansumpw/test/test.snansumpw.js
index e1acc00fca8a..e8a87705129b 100644
--- a/lib/node_modules/@stdlib/blas/ext/base/snansumpw/test/test.snansumpw.js
+++ b/lib/node_modules/@stdlib/blas/ext/base/snansumpw/test/test.snansumpw.js
@@ -34,7 +34,7 @@ tape( 'main export is a function', function test( t ) {
});
tape( 'the function has an arity of 3', function test( t ) {
- t.strictEqual( snansumpw.length, 3, 'reutrn expected value' );
+ t.strictEqual( snansumpw.length, 3, 'returns expected value' );
t.end();
});
@@ -178,14 +178,26 @@ tape( 'the function supports a negative `stride` parameter', function test( t )
t.end();
});
-tape( 'if provided a `stride` parameter equal to `0`, the function returns the first element', function test( t ) {
+tape( 'if provided a `stride` parameter equal to `0`, the function returns the sum of the first element repeated N times', function test( t ) {
var x;
var v;
x = new Float32Array( [ 1.0, -2.0, -4.0, 5.0, 3.0 ] );
v = snansumpw( x.length, x, 0 );
- t.strictEqual( v, 1.0, 'returns expected value' );
+ t.strictEqual( v, 5.0, 'returns expected value' );
+
+ t.end();
+});
+
+tape( 'if provided a `stride` parameter equal to `0` and the first element is `NaN`, the function returns 0.0', function test( t ) {
+ var x;
+ var v;
+
+ x = new Float32Array( [ NaN, -2.0, -4.0, 5.0, 3.0 ] );
+
+ v = snansumpw( x.length, x, 0 );
+ t.strictEqual( v, 0.0, 'returns expected value' );
t.end();
});
diff --git a/lib/node_modules/@stdlib/blas/ext/base/snansumpw/test/test.snansumpw.native.js b/lib/node_modules/@stdlib/blas/ext/base/snansumpw/test/test.snansumpw.native.js
index 256931d154cb..6bd665bf3335 100644
--- a/lib/node_modules/@stdlib/blas/ext/base/snansumpw/test/test.snansumpw.native.js
+++ b/lib/node_modules/@stdlib/blas/ext/base/snansumpw/test/test.snansumpw.native.js
@@ -43,7 +43,7 @@ tape( 'main export is a function', opts, function test( t ) {
});
tape( 'the function has an arity of 3', opts, function test( t ) {
- t.strictEqual( snansumpw.length, 3, 'reutrn expected value' );
+ t.strictEqual( snansumpw.length, 3, 'returns expected value' );
t.end();
});
@@ -269,14 +269,26 @@ tape( 'the function supports a negative `stride` parameter', opts, function test
t.end();
});
-tape( 'if provided a `stride` parameter equal to `0`, the function returns the first element', opts, function test( t ) {
+tape( 'if provided a `stride` parameter equal to `0`, the function returns the sum of the first element repeated N times', opts, function test( t ) {
var x;
var v;
x = new Float32Array( [ 1.0, -2.0, -4.0, 5.0, 3.0 ] );
v = snansumpw( x.length, x, 0 );
- t.strictEqual( v, 1.0, 'returns expected value' );
+ t.strictEqual( v, 5.0, 'returns expected value' );
+
+ t.end();
+});
+
+tape( 'if provided a `stride` parameter equal to `0` and the first element is `NaN`, the function returns 0.0', opts, function test( t ) {
+ var x;
+ var v;
+
+ x = new Float32Array( [ NaN, -2.0, -4.0, 5.0, 3.0 ] );
+
+ v = snansumpw( x.length, x, 0 );
+ t.strictEqual( v, 0.0, 'returns expected value' );
t.end();
});