✨ Add NeuraNetwork

src/algebra.rs

@@ -0,0 +1,32 @@
+/// An extension of `std::ops::AddAssign`
+pub trait NeuraAddAssign {
+    fn add_assign(&mut self, other: &Self);
+}
+
+impl<Left: NeuraAddAssign, Right: NeuraAddAssign> NeuraAddAssign for (Left, Right) {
+    fn add_assign(&mut self, other: &Self) {
+        NeuraAddAssign::add_assign(&mut self.0, &other.0);
+        NeuraAddAssign::add_assign(&mut self.1, &other.1);
+    }
+}
+
+impl<const N: usize, T: NeuraAddAssign> NeuraAddAssign for [T; N] {
+    fn add_assign(&mut self, other: &[T; N]) {
+        for i in 0..N {
+            NeuraAddAssign::add_assign(&mut self[i], &other[i]);
+        }
+    }
+}
+
+macro_rules! base {
+    ( $type:ty ) => {
+        impl NeuraAddAssign for $type {
+            fn add_assign(&mut self, other: &Self) {
+                std::ops::AddAssign::add_assign(self, other);
+            }
+        }
+    }
+}
+
+base!(f32);
+base!(f64);

src/derivable/activation.rs

@@ -1,27 +1,11 @@
-pub trait Activation {
+use super::NeuraDerivable;
-    fn eval(&self, input: f64) -> f64;
-
-    fn eval_f32(&self, input: f32) -> f32 {
-        self.eval(input as f64) as f32
-    }
-
-    fn derivate(&self, at: f64) -> f64;
-
-    fn derivate_f32(&self, at: f32) -> f32 {
-        self.derivate(at as f64) as f32
-    }
-}
 
 #[derive(Clone, Copy, Debug, PartialEq)]
 pub struct Relu;
-impl Activation for Relu {
-    #[inline(always)]
-    fn eval(&self, input: f64) -> f64 {
-        input.max(0.0)
-    }
 
+impl NeuraDerivable<f64> for Relu {
     #[inline(always)]
-    fn eval_f32(&self, input: f32) -> f32 {
+    fn eval(&self, input: f64) -> f64 {
         input.max(0.0)
     }
 
@@ -33,9 +17,16 @@ impl Activation for Relu {
             0.0
         }
     }
+}
+
+impl NeuraDerivable<f32> for Relu {
+    #[inline(always)]
+    fn eval(&self, input: f32) -> f32 {
+        input.max(0.0)
+    }
 
     #[inline(always)]
-    fn derivate_f32(&self, input: f32) -> f32 {
+    fn derivate(&self, input: f32) -> f32 {
         if input > 0.0 {
             1.0
         } else {

src/derivable/loss.rs

@@ -0,0 +1,22 @@
+use super::NeuraLoss;
+
+#[derive(Clone, Copy, Debug, PartialEq)]
+pub struct Euclidean;
+impl<const N: usize> NeuraLoss<[f64; N]> for Euclidean {
+    type Out = f64;
+    type Target = [f64; N];
+
+    fn eval(&self, target: [f64; N], actual: [f64; N]) -> f64 {
+        let mut sum_squared = 0.0;
+
+        for i in 0..N {
+            sum_squared += (target[i] - actual[i]) * (target[i] - actual[i]);
+        }
+
+        sum_squared * 0.5
+    }
+
+    fn nabla(&self, target: [f64; N], actual: [f64; N]) -> [f64; N] {
+        todo!()
+    }
+}

src/derivable/mod.rs

@@ -0,0 +1,20 @@
+pub mod activation;
+pub mod loss;
+
+pub trait NeuraDerivable<F> {
+    fn eval(&self, input: F) -> F;
+
+    /// Should return the derivative of `self.eval(input)`
+    fn derivate(&self, at: F) -> F;
+}
+
+pub trait NeuraLoss<F> {
+    type Out;
+    type Target;
+
+    fn eval(&self, target: Self::Target, actual: F) -> Self::Out;
+
+    /// Should return the gradient of the loss function according to `actual`
+    /// ($\nabla_{\texttt{actual}} \texttt{self.eval}(\texttt{target}, \texttt{actual})$).
+    fn nabla(&self, target: Self::Target, actual: F) -> F;
+}

src/layer.rs

@@ -1,9 +0,0 @@
-mod dense;
-pub use dense::NeuraDenseLayer;
-
-pub trait NeuraLayer {
-    type Input;
-    type Output;
-
-    fn eval(&self, input: &Self::Input) -> Self::Output;
-}

src/layer/dense.rs

@@ -1,14 +1,18 @@
 use super::NeuraLayer;
-use crate::{activation::Activation, utils::multiply_matrix_vector};
+use crate::{derivable::NeuraDerivable, utils::multiply_matrix_vector};
 use rand::Rng;
 
-pub struct NeuraDenseLayer<Act: Activation, const INPUT_LEN: usize, const OUTPUT_LEN: usize> {
+pub struct NeuraDenseLayer<
+    Act: NeuraDerivable<f64>,
+    const INPUT_LEN: usize,
+    const OUTPUT_LEN: usize,
+> {
     weights: [[f64; INPUT_LEN]; OUTPUT_LEN],
     bias: [f64; OUTPUT_LEN],
     activation: Act,
 }
 
-impl<Act: Activation, const INPUT_LEN: usize, const OUTPUT_LEN: usize>
+impl<Act: NeuraDerivable<f64>, const INPUT_LEN: usize, const OUTPUT_LEN: usize>
     NeuraDenseLayer<Act, INPUT_LEN, OUTPUT_LEN>
 {
     pub fn new(
@@ -43,7 +47,7 @@ impl<Act: Activation, const INPUT_LEN: usize, const OUTPUT_LEN: usize>
     }
 }
 
-impl<Act: Activation, const INPUT_LEN: usize, const OUTPUT_LEN: usize> NeuraLayer
+impl<Act: NeuraDerivable<f64>, const INPUT_LEN: usize, const OUTPUT_LEN: usize> NeuraLayer
     for NeuraDenseLayer<Act, INPUT_LEN, OUTPUT_LEN>
 {
     type Input = [f64; INPUT_LEN];
@@ -64,7 +68,7 @@ impl<Act: Activation, const INPUT_LEN: usize, const OUTPUT_LEN: usize> NeuraLaye
 #[cfg(test)]
 mod test {
     use super::*;
-    use crate::activation::Relu;
+    use crate::derivable::activation::Relu;
 
     #[test]
     fn test_from_rng() {

src/layer/mod.rs

@@ -0,0 +1,22 @@
+mod dense;
+pub use dense::NeuraDenseLayer;
+
+pub trait NeuraLayer {
+    type Input;
+    type Output;
+
+    fn eval(&self, input: &Self::Input) -> Self::Output;
+}
+
+#[macro_export]
+macro_rules! neura_layer {
+    ( "dense", $activation:expr, $output:expr ) => {
+        NeuraDenseLayer::from_rng(&mut rand::thread_rng(), $activation)
+            as NeuraDenseLayer<_, _, $output>
+    };
+
+    ( "dense", $activation:expr, $output:expr, $input:expr ) => {
+        NeuraDenseLayer::from_rng(&mut rand::thread_rng(), $activation)
+            as NeuraDenseLayer<_, $input, $output>
+    };
+}

src/lib.rs

@@ -1,3 +1,8 @@
-pub mod activation;
+#![feature(generic_arg_infer)]
+
+pub mod derivable;
 pub mod layer;
+pub mod network;
+pub mod algebra;
+
 mod utils;

src/network.rs

@@ -0,0 +1,103 @@
+use crate::{layer::NeuraLayer, train::NeuraTrainable};
+
+pub struct NeuraNetwork<Layer: NeuraLayer, ChildNetwork> {
+    layer: Layer,
+    child_network: ChildNetwork,
+}
+
+impl<Layer: NeuraLayer, ChildNetwork> NeuraNetwork<Layer, ChildNetwork> {
+    pub fn new(layer: Layer, child_network: ChildNetwork) -> Self {
+        Self {
+            layer,
+            child_network,
+        }
+    }
+
+    pub fn new_match_output(layer: Layer, child_network: ChildNetwork) -> Self
+    where
+        ChildNetwork: NeuraLayer<Input = Layer::Output>,
+    {
+        Self::new(layer, child_network)
+    }
+
+    pub fn child_network(&self) -> &ChildNetwork {
+        &self.child_network
+    }
+}
+
+impl<Layer: NeuraLayer> From<Layer> for NeuraNetwork<Layer, ()> {
+    fn from(layer: Layer) -> Self {
+        Self {
+            layer,
+            child_network: (),
+        }
+    }
+}
+
+impl<Layer: NeuraLayer> NeuraLayer for NeuraNetwork<Layer, ()> {
+    type Input = Layer::Input;
+    type Output = Layer::Output;
+
+    fn eval(&self, input: &Self::Input) -> Self::Output {
+        self.layer.eval(input)
+    }
+}
+
+impl<Layer: NeuraLayer, ChildNetwork: NeuraLayer<Input = Layer::Output>> NeuraLayer
+    for NeuraNetwork<Layer, ChildNetwork>
+{
+    type Input = Layer::Input;
+
+    type Output = ChildNetwork::Output;
+
+    fn eval(&self, input: &Self::Input) -> Self::Output {
+        self.child_network.eval(&self.layer.eval(input))
+    }
+}
+
+#[macro_export]
+macro_rules! neura_network {
+    [] => {
+        ()
+    };
+
+    [ $layer:expr $(,)? ] => {
+        NeuraNetwork::from($layer)
+    };
+
+    [ $first:expr, $($rest:expr),+ $(,)? ] => {
+        NeuraNetwork::new_match_output($first, neura_network![$($rest),+])
+    };
+}
+
+#[cfg(test)]
+mod test {
+    use crate::{derivable::activation::Relu, layer::NeuraDenseLayer, neura_layer};
+
+    use super::*;
+
+    #[test]
+    fn test_neura_network_macro() {
+        let mut rng = rand::thread_rng();
+
+        let _ = neura_network![
+            NeuraDenseLayer::from_rng(&mut rng, Relu) as NeuraDenseLayer<_, 8, 16>,
+            NeuraDenseLayer::from_rng(&mut rng, Relu) as NeuraDenseLayer<_, _, 12>,
+            NeuraDenseLayer::from_rng(&mut rng, Relu) as NeuraDenseLayer<_, _, 2>
+        ];
+
+        let _ =
+            neura_network![NeuraDenseLayer::from_rng(&mut rng, Relu) as NeuraDenseLayer<_, 8, 16>,];
+
+        let _ = neura_network![
+            NeuraDenseLayer::from_rng(&mut rng, Relu) as NeuraDenseLayer<_, 8, 16>,
+            NeuraDenseLayer::from_rng(&mut rng, Relu) as NeuraDenseLayer<_, _, 12>,
+        ];
+
+        let _ = neura_network![
+            neura_layer!("dense", Relu, 16, 8),
+            neura_layer!("dense", Relu, 12),
+            neura_layer!("dense", Relu, 2)
+        ];
+    }
+}

src/utils.rs

@@ -1,4 +1,4 @@
-pub fn multiply_matrix_vector<const WIDTH: usize, const HEIGHT: usize>(
+pub(crate) fn multiply_matrix_vector<const WIDTH: usize, const HEIGHT: usize>(
     matrix: &[[f64; WIDTH]; HEIGHT],
     vector: &[f64; WIDTH],
 ) -> [f64; HEIGHT] {
@@ -14,3 +14,43 @@ pub fn multiply_matrix_vector<const WIDTH: usize, const HEIGHT: usize>(
 
     result
 }
+
+pub(crate) fn assign_add_vector<const N: usize>(sum: &mut [f64; N], operand: &[f64; N]) {
+    for i in 0..N {
+        sum[i] += operand[i];
+    }
+}
+
+pub(crate) fn chunked<I: Iterator>(
+    iter: I,
+    chunk_size: usize,
+) -> impl Iterator<Item = Vec<I::Item>> {
+    struct Chunked<J: Iterator> {
+        iter: J,
+        chunk_size: usize,
+    }
+
+    impl<J: Iterator> Iterator for Chunked<J> {
+        type Item = Vec<J::Item>;
+
+        fn next(&mut self) -> Option<Self::Item> {
+            let mut result = Vec::with_capacity(self.chunk_size);
+
+            for _ in 0..self.chunk_size {
+                if let Some(item) = self.iter.next() {
+                    result.push(item);
+                } else {
+                    break;
+                }
+            }
+
+            if result.len() > 0 {
+                Some(result)
+            } else {
+                None
+            }
+        }
+    }
+
+    Chunked { iter, chunk_size }
+}