dfdx 0.13.0

#![cfg_attr(feature = "nightly", feature(generic_const_exprs))]

#[cfg(feature = "nightly")]
fn main() {
    use std::time::Instant;

    use dfdx::prelude::*;

    #[cfg(feature = "cuda")]
    type Dev = Cuda;

    #[cfg(not(feature = "cuda"))]
    type Dev = Cpu;

    type Model = Conv2D<128, 256, 4>;
    type Dtype = f32;
    type InputShape = Rank4<64, 128, 28, 28>;

    println!("Benchmarking `Conv2D`");
    println!("Device {}", std::any::type_name::<Dev>());
    println!("Dtype {}", std::any::type_name::<Dtype>());
    println!("Input shape {}", std::any::type_name::<InputShape>());
    println!();

    let dev: Dev = Default::default();
    let mut m = dev.build_module::<Model, Dtype>();

    loop {
        let img: Tensor<InputShape, Dtype, _> = dev.sample_normal();

        let start = Instant::now();
        let _ = m.forward(img.clone());
        dev.synchronize();
        let infer_dur = start.elapsed();

        let start = Instant::now();
        let out = m.forward_mut(img.leaky_traced());
        let loss = out.square().mean();
        dev.synchronize();
        let fwd_dur = start.elapsed();

        let start = Instant::now();
        let _ = loss.backward();
        dev.synchronize();
        let bwd_dur = start.elapsed();

        println!("infer={infer_dur:?}, fwd={fwd_dur:?} bwd={bwd_dur:?}");
    }
}

#[cfg(not(feature = "nightly"))]
fn main() {
    panic!("Run with `cargo +nightly run ...` to run this example.");
}