outer array product

vecxt/js-native/src/array.scala

@@ -71,6 +71,23 @@ object JsNativeDoubleArrays:
 
   extension (vec: NArray[Double])
 
+    inline def outer(other: NArray[Double])(using ClassTag[Double]): Matrix[Double] =
+      val n = vec.length
+      val m = other.length
+      val out: NArray[Double] = NArray.ofSize[Double](n * m)
+
+      var i = 0
+      while i < n do
+        var j = 0
+        while j < m do
+          out(j * n + i) = vec(i) * other(j)
+          j = j + 1
+        end while
+        i = i + 1
+      end while
+      Matrix[Double](out, (n, m))(using BoundsCheck.DoBoundsCheck.no)
+    end outer
+
     inline def <(num: Double): NArray[Boolean] =
       logicalIdx((a, b) => a < b, num)
 

vecxt/jvm/src/arrays.scala

@@ -18,6 +18,7 @@ package vecxt
 import scala.util.chaining.*
 
 import vecxt.BoundsCheck.BoundsCheck
+import vecxt.matrix.Matrix
 
 import dev.ludovic.netlib.blas.JavaBLAS.getInstance as blas
 import jdk.incubator.vector.ByteVector
@@ -473,6 +474,29 @@ object arrays:
       ranks
     end elementRanks
 
+    inline def outer(other: Array[Double])(using ClassTag[Double]): Matrix[Double] =
+      val n = vec.length
+      val m = other.length
+      val out = new Array[Double](n * m)
+
+      var j = 0
+      while j < m do
+        var i = 0
+        val tmp = DoubleVector.broadcast(spd, other(j))
+        while i < spd.loopBound(n) do
+          DoubleVector.fromArray(spd, vec, i).mul(tmp).intoArray(out, j * n + i)
+          i = i + spdl
+        end while
+
+        while i < n do
+          out(j * n + i) = vec(i) * other(j)
+          i = i + 1
+        end while
+        j = j + 1
+      end while
+      Matrix(out, (n, m))(using BoundsCheck.DoBoundsCheck.no)
+    end outer
+
     def variance: Double =
       // https://www.cuemath.com/sample-variance-formula/
       val μ = vec.mean

vecxt/test/src/array.test.scala

@@ -223,6 +223,24 @@ class ArrayExtensionSuite extends munit.FunSuite:
     assertEquals(v_idx3.trues, 1)
   }
 
+  test("outer product".only) {
+    val v1 = NArray[Double](1.0, 2.0, 3.0)
+    val v2 = NArray[Double](4.0, 5.0)
+    val outer = v1.outer(v2)
+    val shouldBe = Matrix.fromRows[Double](
+      NArray(
+        NArray(4.0, 5.0),
+        NArray(8.0, 10.0),
+        NArray(12.0, 15.0)
+      )
+    )
+    assertEquals(outer.rows, 3)
+    assertEquals(outer.cols, 2)
+
+    assertVecEquals(outer.raw, shouldBe.raw)
+
+  }
+
   test("norm") {
     assertEqualsDouble(v_fill.norm, Math.sqrt(1 + 4 + 9 + 16), 0.00001)
   }