Switch inner loop to run over rows

1 files changed, 4 insertions, 3 deletions

diff --git a/src/main.rs b/src/main.rs
index 31d810b..edd202d 100644
--- a/src/main.rs
+++ b/src/main.rs
@@ -1,9 +1,10 @@
 use mdarray::{view, tensor, DSlice, Expression};
 
+// Indexing convention: C_ij <- A_ik * B_kj
 fn matmul(a: &DSlice<f64, 2>, b: &DSlice<f64, 2>, c: &mut DSlice<f64, 2>) {
-    for (mut cj, bj) in c.cols_mut().zip(b.cols()) {
-        for (ak, bkj) in a.cols().zip(bj) {
-            for (cij, aik) in cj.expr_mut().zip(ak) {
+    for (mut ci, ai) in c.rows_mut().zip(a.rows()) {
+        for (aik, bk) in ai.zip(b.rows()) {
+            for (cij, bkj) in ci.expr_mut().zip(bk) {
                 *cij = aik.mul_add(*bkj, *cij);
             }
         }