This repository was ispired by awesome Tutorial: OpenCL SGEMM tuning for Kepler.

Goals

1. Implement matrix multiplication using wgpu-rs
2. Build a simple nerual network using wgpu-rs

Code examples

• examples/simple.rs contains spagetti-code of matmul with performance mesuarement.
• examples/linear.rs contains a usage of "Tensor" multiplication, where Tensor is just a container for data: Vec<f32> and dim: [usize;2] which helps to build a wgpu Buffer