## Goal Implement element-wise and linear algebra primitives on `Tensor`. ## Tasks - [ ] Element-wise: `+`, `-`, `*`, `/` (with broadcasting, simple cases first) - [ ] Reductions: `sum`, `mean` - [ ] `matmul` for 2D tensors (naive triple-loop is fine to start) - [ ] Operator overloading for clean syntax ## Acceptance criteria - Each op has a unit test comparing against hand-computed values - API feels natural: `auto z = x + y;` and `auto z = matmul(a, b);` ## Depends on - #<tensor-issue>
Goal
Implement element-wise and linear algebra primitives on
Tensor.Tasks
+,-,*,/(with broadcasting, simple cases first)sum,meanmatmulfor 2D tensors (naive triple-loop is fine to start)Acceptance criteria
auto z = x + y;andauto z = matmul(a, b);Depends on