Merge 261d0d700749af646afbda49241f2e923c976a1c into 11330e464ba3a7f08aaf73bc918281472c503b1d

src/compute/activations.rs

@@ -17,14 +17,25 @@ pub fn drelu(x: &Matrix<f64>) -> Matrix<f64> {
     x.map(|v| if v > 0.0 { 1.0 } else { 0.0 })
 }
 
+pub fn leaky_relu(x: &Matrix<f64>) -> Matrix<f64> {
+    x.map(|v| if v > 0.0 { v } else { 0.01 * v })
+}
+
+pub fn dleaky_relu(x: &Matrix<f64>) -> Matrix<f64> {
+    x.map(|v| if v > 0.0 { 1.0 } else { 0.01 })
+}
 
 mod tests {
     use super::*;
-    
+
     // Helper function to round all elements in a matrix to n decimal places
     fn _round_matrix(mat: &Matrix<f64>, decimals: u32) -> Matrix<f64> {
         let factor = 10f64.powi(decimals as i32);
-        let rounded: Vec<f64> = mat.to_vec().iter().map(|v| (v * factor).round() / factor).collect();
+        let rounded: Vec<f64> = mat
+            .to_vec()
+            .iter()
+            .map(|v| (v * factor).round() / factor)
+            .collect();
         Matrix::from_vec(rounded, mat.rows(), mat.cols())
     }
 
@@ -36,6 +47,17 @@ mod tests {
         assert_eq!(_round_matrix(&result, 6), _round_matrix(&expected, 6));
     }
 
+    #[test]
+    fn test_sigmoid_edge_case() {
+        let x = Matrix::from_vec(vec![-1000.0, 0.0, 1000.0], 3, 1);
+        let expected = Matrix::from_vec(vec![0.0, 0.5, 1.0], 3, 1);
+        let result = sigmoid(&x);
+
+        for (r, e) in result.data().iter().zip(expected.data().iter()) {
+            assert!((r - e).abs() < 1e-6);
+        }
+    }
+
     #[test]
     fn test_relu() {
         let x = Matrix::from_vec(vec![-1.0, 0.0, 1.0], 3, 1);
@@ -43,6 +65,13 @@ mod tests {
         assert_eq!(relu(&x), expected);
     }
 
+    #[test]
+    fn test_relu_edge_case() {
+        let x = Matrix::from_vec(vec![-1e-10, 0.0, 1e10], 3, 1);
+        let expected = Matrix::from_vec(vec![0.0, 0.0, 1e10], 3, 1);
+        assert_eq!(relu(&x), expected);
+    }
+
     #[test]
     fn test_dsigmoid() {
         let y = Matrix::from_vec(vec![0.26894142, 0.5, 0.73105858], 3, 1);
@@ -50,11 +79,57 @@ mod tests {
         let result = dsigmoid(&y);
         assert_eq!(_round_matrix(&result, 6), _round_matrix(&expected, 6));
     }
-    
+
+    #[test]
+    fn test_dsigmoid_edge_case() {
+        let y = Matrix::from_vec(vec![0.0, 0.5, 1.0], 3, 1); // Assume these are outputs from sigmoid(x)
+        let expected = Matrix::from_vec(vec![0.0, 0.25, 0.0], 3, 1);
+        let result = dsigmoid(&y);
+
+        for (r, e) in result.data().iter().zip(expected.data().iter()) {
+            assert!((r - e).abs() < 1e-6);
+        }
+    }
+
     #[test]
     fn test_drelu() {
         let x = Matrix::from_vec(vec![-1.0, 0.0, 1.0], 3, 1);
         let expected = Matrix::from_vec(vec![0.0, 0.0, 1.0], 3, 1);
-        assert_eq!(drelu(&x), expected);    
+        assert_eq!(drelu(&x), expected);
     }
-}
+
+    #[test]
+    fn test_drelu_edge_case() {
+        let x = Matrix::from_vec(vec![-1e-10, 0.0, 1e10], 3, 1);
+        let expected = Matrix::from_vec(vec![0.0, 0.0, 1.0], 3, 1);
+        assert_eq!(drelu(&x), expected);
+    }
+
+    #[test]
+    fn test_leaky_relu() {
+        let x = Matrix::from_vec(vec![-1.0, 0.0, 1.0], 3, 1);
+        let expected = Matrix::from_vec(vec![-0.01, 0.0, 1.0], 3, 1);
+        assert_eq!(leaky_relu(&x), expected);
+    }
+
+    #[test]
+    fn test_leaky_relu_edge_case() {
+        let x = Matrix::from_vec(vec![-1e-10, 0.0, 1e10], 3, 1);
+        let expected = Matrix::from_vec(vec![-1e-12, 0.0, 1e10], 3, 1);
+        assert_eq!(leaky_relu(&x), expected);
+    }
+
+    #[test]
+    fn test_dleaky_relu() {
+        let x = Matrix::from_vec(vec![-1.0, 0.0, 1.0], 3, 1);
+        let expected = Matrix::from_vec(vec![0.01, 0.01, 1.0], 3, 1);
+        assert_eq!(dleaky_relu(&x), expected);
+    }
+
+    #[test]
+    fn test_dleaky_relu_edge_case() {
+        let x = Matrix::from_vec(vec![-1e-10, 0.0, 1e10], 3, 1);
+        let expected = Matrix::from_vec(vec![0.01, 0.01, 1.0], 3, 1);
+        assert_eq!(dleaky_relu(&x), expected);
+    }
+}

src/compute/models/dense_nn.rs

@@ -1,65 +1,278 @@
 use crate::matrix::{Matrix, SeriesOps};
-use crate::compute::activations::{relu, sigmoid, drelu};  
+use crate::compute::activations::{relu, drelu, sigmoid};
-use rand::Rng;
+use rand::prelude::*;
 
+/// Supported activation functions
+#[derive(Clone)]
+pub enum ActivationKind {
+    Relu,
+    Sigmoid,
+    Tanh,
+}
+
+impl ActivationKind {
+    /// Apply activation elementwise
+    pub fn forward(&self, z: &Matrix<f64>) -> Matrix<f64> {
+        match self {
+            ActivationKind::Relu => relu(z),
+            ActivationKind::Sigmoid => sigmoid(z),
+            ActivationKind::Tanh => z.map(|v| v.tanh()),
+        }
+    }
+
+    /// Compute elementwise derivative w.r.t. pre-activation z
+    pub fn derivative(&self, z: &Matrix<f64>) -> Matrix<f64> {
+        match self {
+            ActivationKind::Relu => drelu(z),
+            ActivationKind::Sigmoid => {
+                let s = sigmoid(z);
+                s.zip(&s, |si, sj| si * (1.0 - sj))
+            }
+            ActivationKind::Tanh => z.map(|v| 1.0 - v.tanh().powi(2)),
+        }
+    }
+}
+
+/// Weight initialization schemes
+#[derive(Clone)]
+pub enum InitializerKind {
+    /// Uniform(-limit .. limit)
+    Uniform(f64),
+    /// Xavier/Glorot uniform
+    Xavier,
+    /// He (Kaiming) uniform
+    He,
+}
+
+impl InitializerKind {
+    pub fn initialize(&self, rows: usize, cols: usize) -> Matrix<f64> {
+        let mut rng = rand::rng();
+        let fan_in = rows;
+        let fan_out = cols;
+        let limit = match self {
+            InitializerKind::Uniform(l) => *l,
+            InitializerKind::Xavier => (6.0 / (fan_in + fan_out) as f64).sqrt(),
+            InitializerKind::He => (2.0 / fan_in as f64).sqrt(),
+        };
+        let data = (0..rows * cols)
+            .map(|_| rng.random_range(-limit..limit))
+            .collect::<Vec<_>>();
+        Matrix::from_vec(data, rows, cols)
+    }
+}
+
+/// Supported losses
+#[derive(Clone)]
+pub enum LossKind {
+    /// Mean Squared Error: L = 1/m * sum((y_hat - y)^2)
+    MSE,
+    /// Binary Cross-Entropy: L = -1/m * sum(y*log(y_hat) + (1-y)*log(1-y_hat))
+    BCE,
+}
+
+impl LossKind {
+    /// Compute gradient dL/dy_hat (before applying activation derivative)
+    pub fn gradient(&self, y_hat: &Matrix<f64>, y: &Matrix<f64>) -> Matrix<f64> {
+        let m = y.rows() as f64;
+        match self {
+            LossKind::MSE => (y_hat - y) * (2.0 / m),
+            LossKind::BCE => (y_hat - y) * (1.0 / m),
+        }
+    }
+}
+
+/// Configuration for a dense neural network
+pub struct DenseNNConfig {
+    pub input_size: usize,
+    pub hidden_layers: Vec<usize>,
+    /// Must have length = hidden_layers.len() + 1
+    pub activations: Vec<ActivationKind>,
+    pub output_size: usize,
+    pub initializer: InitializerKind,
+    pub loss: LossKind,
+    pub learning_rate: f64,
+    pub epochs: usize,
+}
+
+/// A multi-layer perceptron with full configurability
 pub struct DenseNN {
-    w1: Matrix<f64>, // (n_in, n_hidden)
+    weights: Vec<Matrix<f64>>,
-    b1: Matrix<f64>, // (1,    n_hidden)
+    biases: Vec<Matrix<f64>>,
-    w2: Matrix<f64>, // (n_hidden, n_out)
+    activations: Vec<ActivationKind>,
-    b2: Matrix<f64>, // (1,    n_out)
+    loss: LossKind,
+    lr: f64,
+    epochs: usize,
 }
 
 impl DenseNN {
-    pub fn new(n_in: usize, n_hidden: usize, n_out: usize) -> Self {
+    /// Build a new DenseNN from the given configuration
-        let mut rng = rand::rng();
+    pub fn new(config: DenseNNConfig) -> Self {
-        let mut init = |rows, cols| {
+        let mut sizes = vec![config.input_size];
-            let data = (0..rows * cols)
+        sizes.extend(&config.hidden_layers);
-                .map(|_| rng.random_range(-1.0..1.0))
+        sizes.push(config.output_size);
-                .collect::<Vec<_>>();
+
-            Matrix::from_vec(data, rows, cols)
+        assert_eq!(
-        };
+            config.activations.len(),
-        Self {
+            sizes.len() - 1,
-            w1: init(n_in, n_hidden),
+            "Number of activation functions must match number of layers"
-            b1: Matrix::zeros(1, n_hidden),
+        );
-            w2: init(n_hidden, n_out),
+
-            b2: Matrix::zeros(1, n_out),
+        let mut weights = Vec::with_capacity(sizes.len() - 1);
+        let mut biases  = Vec::with_capacity(sizes.len() - 1);
+
+        for i in 0..sizes.len() - 1 {
+            let w = config.initializer.initialize(sizes[i], sizes[i + 1]);
+            let b = Matrix::zeros(1, sizes[i + 1]);
+            weights.push(w);
+            biases.push(b);
+        }
+
+        DenseNN {
+            weights,
+            biases,
+            activations: config.activations,
+            loss: config.loss,
+            lr: config.learning_rate,
+            epochs: config.epochs,
         }
     }
 
-    pub fn forward(&self, x: &Matrix<f64>) -> (Matrix<f64>, Matrix<f64>, Matrix<f64>) {
+    /// Perform a full forward pass, returning pre-activations (z) and activations (a)
-        // z1 = X·W1 + b1 ; a1 = ReLU(z1)
+    fn forward_full(&self, x: &Matrix<f64>) -> (Vec<Matrix<f64>>, Vec<Matrix<f64>>) {
-        let z1 = x.dot(&self.w1) + &self.b1;
+        let mut zs = Vec::with_capacity(self.weights.len());
-        let a1 = relu(&z1);
+        let mut activs = Vec::with_capacity(self.weights.len() + 1);
-        // z2 = a1·W2 + b2 ; a2 = softmax(z2) (here binary => sigmoid)
+        activs.push(x.clone());
-        let z2 = a1.dot(&self.w2) + &self.b2;
+
-        let a2 = sigmoid(&z2); // binary output
+        let mut a = x.clone();
-        (a1, z2, a2)           // keep intermediates for back-prop
+        for (i, (w, b)) in self.weights.iter().zip(self.biases.iter()).enumerate() {
+            let z = &a.dot(w) + &Matrix::repeat_rows(b, a.rows());
+            let a_next = self.activations[i].forward(&z);
+            zs.push(z);
+            activs.push(a_next.clone());
+            a = a_next;
+        }
+
+        (zs, activs)
     }
 
-    pub fn train(&mut self, x: &Matrix<f64>, y: &Matrix<f64>, lr: f64, epochs: usize) {
+    /// Train the network on inputs X and targets Y
+    pub fn train(&mut self, x: &Matrix<f64>, y: &Matrix<f64>) {
         let m = x.rows() as f64;
-        for _ in 0..epochs {
+        for _ in 0..self.epochs {
-            let (a1, _z2, y_hat) = self.forward(x);
+            let (zs, activs) = self.forward_full(x);
+            let y_hat = activs.last().unwrap().clone();
 
-            // -------- backwards ----------
+            // Initial delta (dL/dz) on output
-            // dL/da2 = y_hat - y  (BCE derivative)
+            let mut delta = match self.loss {
-            let dz2 = &y_hat - y;                           // (m, n_out)
+                LossKind::BCE => self.loss.gradient(&y_hat, y),
-            let dw2 = a1.transpose().dot(&dz2) / m;         // (n_h, n_out)
+                LossKind::MSE => {
-            // let db2 = dz2.sum_vertical() * (1.0 / m);       // broadcast ok
+                    let grad = self.loss.gradient(&y_hat, y);
-            let db2 = Matrix::from_vec(dz2.sum_vertical(), 1, dz2.cols()) * (1.0 / m); // (1, n_out)
+                    let dz = self.activations.last().unwrap().derivative(zs.last().unwrap());
-            let da1 = dz2.dot(&self.w2.transpose());        // (m,n_h)
+                    grad.zip(&dz, |g, da| g * da)
-            let dz1 = da1.zip(&a1, |g, act| g * drelu(&Matrix::from_cols(vec![vec![act]])).data()[0]); // (m,n_h)
+                }
-            
+            };
-            // real code: drelu returns Matrix, broadcasting needed; you can optimise.
 
-            let dw1 = x.transpose().dot(&dz1) / m;          // (n_in,n_h)
+            // Backpropagate through layers
-            let db1 = Matrix::from_vec(dz1.sum_vertical(), 1, dz1.cols()) * (1.0 / m); // (1, n_h)
+            for l in (0..self.weights.len()).rev() {
+                let a_prev = &activs[l];
+                let dw = a_prev.transpose().dot(&delta) / m;
+                let db = Matrix::from_vec(delta.sum_vertical(), 1, delta.cols()) / m;
 
-            // -------- update ----------
+                // Update weights & biases
-            self.w2 = &self.w2 - &(dw2 * lr);
+                self.weights[l] = &self.weights[l] - &(dw * self.lr);
-            self.b2 = &self.b2 - &(db2 * lr);
+                self.biases[l]  = &self.biases[l]  - &(db * self.lr);
-            self.w1 = &self.w1 - &(dw1 * lr);
+
-            self.b1 = &self.b1 - &(db1 * lr);
+                // Propagate delta to previous layer
+                if l > 0 {
+                    let w_t = self.weights[l].transpose();
+                    let da = self.activations[l - 1].derivative(&zs[l - 1]);
+                    delta = delta.dot(&w_t).zip(&da, |d, a| d * a);
+                }
+            }
         }
     }
+
+    /// Run a forward pass and return the network's output
+    pub fn predict(&self, x: &Matrix<f64>) -> Matrix<f64> {
+        let mut a = x.clone();
+        for (i, (w, b)) in self.weights.iter().zip(self.biases.iter()).enumerate() {
+            let z = &a.dot(w) + &Matrix::repeat_rows(b, a.rows());
+            a = self.activations[i].forward(&z);
+        }
+        a
+    }
+}
+
+// ------------------------------
+// Simple tests
+// ------------------------------
+
+#[cfg(test)]
+mod tests {
+    use super::*;
+    use crate::matrix::Matrix;
+
+    #[test]
+    fn test_predict_shape() {
+        let config = DenseNNConfig {
+            input_size: 1,
+            hidden_layers: vec![2],
+            activations: vec![ActivationKind::Relu, ActivationKind::Sigmoid],
+            output_size: 1,
+            initializer: InitializerKind::Uniform(0.1),
+            loss: LossKind::MSE,
+            learning_rate: 0.01,
+            epochs: 0,
+        };
+        let model = DenseNN::new(config);
+        let x = Matrix::from_vec(vec![1.0, 2.0, 3.0], 3, 1);
+        let preds = model.predict(&x);
+        assert_eq!(preds.rows(), 3);
+        assert_eq!(preds.cols(), 1);
+    }
+
+    #[test]
+    fn test_train_no_epochs() {
+        let config = DenseNNConfig {
+            input_size: 1,
+            hidden_layers: vec![2],
+            activations: vec![ActivationKind::Relu, ActivationKind::Sigmoid],
+            output_size: 1,
+            initializer: InitializerKind::Uniform(0.1),
+            loss: LossKind::MSE,
+            learning_rate: 0.01,
+            epochs: 0,
+        };
+        let mut model = DenseNN::new(config);
+        let x = Matrix::from_vec(vec![1.0, 2.0], 2, 1);
+        let before = model.predict(&x);
+        model.train(&x, &before);
+        let after = model.predict(&x);
+        for i in 0..before.rows() {
+            assert!((before[(i, 0)] - after[(i, 0)]).abs() < 1e-12);
+        }
+    }
+
+    #[test]
+    fn test_dense_nn_step() {
+        let config = DenseNNConfig {
+            input_size: 1,
+            hidden_layers: vec![2],
+            activations: vec![ActivationKind::Relu, ActivationKind::Sigmoid],
+            output_size: 1,
+            initializer: InitializerKind::He,
+            loss: LossKind::BCE,
+            learning_rate: 0.01,
+            epochs: 5000,
+        };
+        let mut model = DenseNN::new(config);
+        let x = Matrix::from_vec(vec![1.0, 2.0, 3.0, 4.0], 4, 1);
+        let y = Matrix::from_vec(vec![0.0, 0.0, 1.0, 1.0], 4, 1);
+        model.train(&x, &y);
+        let preds = model.predict(&x);
+        assert!((preds[(0, 0)] - 0.0).abs() < 0.5);
+        assert!((preds[(1, 0)] - 0.0).abs() < 0.5);
+        assert!((preds[(2, 0)] - 1.0).abs() < 0.5);
+        assert!((preds[(3, 0)] - 1.0).abs() < 0.5);
+    }
 }

src/compute/models/linreg.rs

@@ -34,10 +34,10 @@ impl LinReg {
     }
 }
 
+#[cfg(test)]
 mod tests {
 
-    use super::LinReg;
+    use super::*;
-    use crate::matrix::{Matrix};
 
     #[test]
     fn test_linreg_fit_predict() {

src/compute/models/logreg.rs

@@ -1,5 +1,5 @@
-use crate::matrix::{Matrix, SeriesOps};
 use crate::compute::activations::sigmoid;
+use crate::matrix::{Matrix, SeriesOps};
 
 pub struct LogReg {
     w: Matrix<f64>,
@@ -15,14 +15,14 @@ impl LogReg {
     }
 
     pub fn predict_proba(&self, x: &Matrix<f64>) -> Matrix<f64> {
-        sigmoid(&(x.dot(&self.w) + self.b))          // σ(Xw + b)
+        sigmoid(&(x.dot(&self.w) + self.b)) // σ(Xw + b)
     }
 
     pub fn fit(&mut self, x: &Matrix<f64>, y: &Matrix<f64>, lr: f64, epochs: usize) {
         let m = x.rows() as f64;
         for _ in 0..epochs {
-            let p  = self.predict_proba(x);          // shape (m,1)
+            let p = self.predict_proba(x); // shape (m,1)
-            let err = &p - y;                        // derivative of BCE wrt pre-sigmoid
+            let err = &p - y; // derivative of BCE wrt pre-sigmoid
             let grad_w = x.transpose().dot(&err) / m;
             let grad_b = err.sum_vertical().iter().sum::<f64>() / m;
             self.w = &self.w - &(grad_w * lr);
@@ -31,6 +31,25 @@ impl LogReg {
     }
 
     pub fn predict(&self, x: &Matrix<f64>) -> Matrix<f64> {
-        self.predict_proba(x).map(|p| if p >= 0.5 { 1.0 } else { 0.0 })
+        self.predict_proba(x)
+            .map(|p| if p >= 0.5 { 1.0 } else { 0.0 })
+    }
+}
+
+#[cfg(test)]
+mod tests {
+    use super::*;
+
+    #[test]
+    fn test_logreg_fit_predict() {
+        let x = Matrix::from_vec(vec![1.0, 2.0, 3.0, 4.0], 4, 1);
+        let y = Matrix::from_vec(vec![0.0, 0.0, 1.0, 1.0], 4, 1);
+        let mut model = LogReg::new(1);
+        model.fit(&x, &y, 0.01, 10000);
+        let preds = model.predict(&x);
+        assert_eq!(preds[(0, 0)], 0.0);
+        assert_eq!(preds[(1, 0)], 0.0);
+        assert_eq!(preds[(2, 0)], 1.0);
+        assert_eq!(preds[(3, 0)], 1.0);
     }
 }

src/matrix/mat.rs

@@ -89,6 +89,10 @@ impl<T: Clone> Matrix<T> {
         self.cols
     }
 
+    pub fn shape(&self) -> (usize, usize) {
+        (self.rows, self.cols)
+    }
+
     /// Get element reference (immutable). Panics on out-of-bounds.
     pub fn get(&self, r: usize, c: usize) -> &T {
         &self[(r, c)]
@@ -323,6 +327,21 @@ impl<T: Clone> Matrix<T> {
         self.data = new_data;
         self.rows = new_rows;
     }
+
+    /// Return a new matrix where row 0 of `self` is repeated `n` times.
+    pub fn repeat_rows(&self, n: usize) -> Matrix<T>
+    where
+        T: Clone,
+    {
+        let mut data = Vec::with_capacity(n * self.cols());
+        let zeroth_row = self.row(0);
+        for value in &zeroth_row {
+            for _ in 0..n {
+                data.push(value.clone()); // Clone each element
+            }
+        }
+        Matrix::from_vec(data, n, self.cols)
+    }
 }
 
 impl Matrix<f64> {
@@ -1153,6 +1172,25 @@ mod tests {
         assert_eq!(ma.row(2), &[3, 6, 9]);
     }
 
+    #[test]
+    fn test_shape() {
+        let ma = static_test_matrix_2x4();
+        assert_eq!(ma.shape(), (2, 4));
+        assert_eq!(ma.rows(), 2);
+        assert_eq!(ma.cols(), 4);
+    }
+
+    #[test]
+    fn test_repeat_rows() {
+        let ma = static_test_matrix();
+        // Returns a new matrix where row 0 of `self` is repeated `n` times.
+        let repeated = ma.repeat_rows(3);
+        // assert all rows are equal to the first row
+        for r in 0..repeated.rows() {
+            assert_eq!(repeated.row(r), ma.row(0));
+        }
+    }
+
     #[test]
     #[should_panic(expected = "row index 3 out of bounds for 3 rows")]
     fn test_row_out_of_bounds() {