work/smartcore
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Patch to version 0.4.0 (#257)

Browse Source 
* uncomment test

* Add random test for logistic regression

* linting

* Bump version

* Add test for logistic regression

* linting

* initial commit

* final

* final-clean

* Bump to 0.4.0

* Fix linter

* cleanup

* Update CHANDELOG with breaking changes

* Update CHANDELOG date

* Add functional methods to DenseMatrix implementation

* linting

* add type declaration in test

* Fix Wasm tests failing

* linting

* fix tests

* linting

* Add type annotations on BBDTree constructor

* fix clippy

* fix clippy

* fix tests

* bump version

* run fmt. fix changelog

---------

Co-authored-by: Edmund Cape <edmund@Edmunds-MacBook-Pro.local>
This commit is contained in:
Lorenzo
2024-03-04 13:51:27 +00:00
committed by GitHub
parent 80a93c1a0e
commit 239c00428f
45 changed files with 759 additions and 406 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/linear/bg_solver.rs
+4 -2
View File
@@ -12,7 +12,8 @@
//! pub struct BGSolver {}
//! impl<'a, T: FloatNumber, X: Array2<T>> BiconjugateGradientSolver<'a, T, X> for BGSolver {}
//!
//! let a = DenseMatrix::from_2d_array(&[&[25., 15., -5.], &[15., 18., 0.], &[-5., 0., 11.]]);
//! let a = DenseMatrix::from_2d_array(&[&[25., 15., -5.], &[15., 18., 0.], &[-5., 0.,
//! 11.]]).unwrap();
//! let b = vec![40., 51., 28.];
//! let expected = vec![1.0, 2.0, 3.0];
//! let mut x = Vec::zeros(3);
@@ -158,7 +159,8 @@ mod tests {
#[test]
fn bg_solver() {
let a = DenseMatrix::from_2d_array(&[&[25., 15., -5.], &[15., 18., 0.], &[-5., 0., 11.]]);
let a = DenseMatrix::from_2d_array(&[&[25., 15., -5.], &[15., 18., 0.], &[-5., 0., 11.]])
.unwrap();
let b = vec![40., 51., 28.];
let expected = [1.0, 2.0, 3.0];
src/linear/elastic_net.rs
+6 -4
View File
@@ -38,7 +38,7 @@
//! &[502.601, 393.1, 251.4, 125.368, 1960., 69.564],
//! &[518.173, 480.6, 257.2, 127.852, 1961., 69.331],
//! &[554.894, 400.7, 282.7, 130.081, 1962., 70.551],
//! ]);
//! ]).unwrap();
//!
//! let y: Vec<f64> = vec![83.0, 88.5, 88.2, 89.5, 96.2, 98.1, 99.0,
//! 100.0, 101.2, 104.6, 108.4, 110.8, 112.6, 114.2, 115.7, 116.9];
@@ -511,7 +511,8 @@ mod tests {
&[502.601, 393.1, 251.4, 125.368, 1960., 69.564],
&[518.173, 480.6, 257.2, 127.852, 1961., 69.331],
&[554.894, 400.7, 282.7, 130.081, 1962., 70.551],
]);
])
.unwrap();
let y: Vec<f64> = vec![
83.0, 88.5, 88.2, 89.5, 96.2, 98.1, 99.0, 100.0, 101.2, 104.6, 108.4, 110.8, 112.6,
@@ -562,7 +563,8 @@ mod tests {
&[17.0, 1918.0, 1.4054969025700674],
&[18.0, 1929.0, 1.3271699396384906],
&[19.0, 1915.0, 1.1373332337674806],
]);
])
.unwrap();
let y: Vec<f64> = vec![
1.48, 2.72, 4.52, 5.72, 5.25, 4.07, 3.75, 4.75, 6.77, 4.72, 6.78, 6.79, 8.3, 7.42,
@@ -627,7 +629,7 @@ mod tests {
// &[502.601, 393.1, 251.4, 125.368, 1960., 69.564],
// &[518.173, 480.6, 257.2, 127.852, 1961., 69.331],
// &[554.894, 400.7, 282.7, 130.081, 1962., 70.551],
// ]);
// ]).unwrap();
// let y = vec![
// 83.0, 88.5, 88.2, 89.5, 96.2, 98.1, 99.0, 100.0, 101.2, 104.6, 108.4, 110.8, 112.6,
src/linear/lasso.rs
+2 -1
View File
@@ -418,7 +418,8 @@ mod tests {
&[502.601, 393.1, 251.4, 125.368, 1960., 69.564],
&[518.173, 480.6, 257.2, 127.852, 1961., 69.331],
&[554.894, 400.7, 282.7, 130.081, 1962., 70.551],
]);
])
.unwrap();
let y: Vec<f64> = vec![
83.0, 88.5, 88.2, 89.5, 96.2, 98.1, 99.0, 100.0, 101.2, 104.6, 108.4, 110.8, 112.6,
src/linear/linear_regression.rs
+4 -3
View File
@@ -40,7 +40,7 @@
//! &[502.601, 393.1, 251.4, 125.368, 1960., 69.564],
//! &[518.173, 480.6, 257.2, 127.852, 1961., 69.331],
//! &[554.894, 400.7, 282.7, 130.081, 1962., 70.551],
//! ]);
//! ]).unwrap();
//!
//! let y: Vec<f64> = vec![83.0, 88.5, 88.2, 89.5, 96.2, 98.1, 99.0,
//! 100.0, 101.2, 104.6, 108.4, 110.8, 112.6, 114.2, 115.7, 116.9];
@@ -341,7 +341,8 @@ mod tests {
&[502.601, 393.1, 251.4, 125.368, 1960., 69.564],
&[518.173, 480.6, 257.2, 127.852, 1961., 69.331],
&[554.894, 400.7, 282.7, 130.081, 1962., 70.551],
]);
])
.unwrap();
let y: Vec<f64> = vec![
83.0, 88.5, 88.2, 89.5, 96.2, 98.1, 99.0, 100.0, 101.2, 104.6, 108.4, 110.8,
@@ -393,7 +394,7 @@ mod tests {
// &[502.601, 393.1, 251.4, 125.368, 1960., 69.564],
// &[518.173, 480.6, 257.2, 127.852, 1961., 69.331],
// &[554.894, 400.7, 282.7, 130.081, 1962., 70.551],
// ]);
// ]).unwrap();
// let y = vec![
// 83.0, 88.5, 88.2, 89.5, 96.2, 98.1, 99.0, 100.0, 101.2, 104.6, 108.4, 110.8, 112.6,
src/linear/logistic_regression.rs
+84 -34
View File
@@ -35,7 +35,7 @@
//! &[4.9, 2.4, 3.3, 1.0],
//! &[6.6, 2.9, 4.6, 1.3],
//! &[5.2, 2.7, 3.9, 1.4],
//! ]);
//! ]).unwrap();
//! let y: Vec<i32> = vec![
//! 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
//! ];
@@ -416,7 +416,7 @@ impl<TX: Number + FloatNumber + RealNumber, TY: Number + Ord, X: Array2<TX>, Y:
/// Fits Logistic Regression to your data.
/// * `x` - _NxM_ matrix with _N_ observations and _M_ features in each observation.
/// * `y` - target class values
/// * `parameters` - other parameters, use `Default::default()` to set parameters to default values.
/// * `parameters` - other parameters, use `Default::default()` to set parameters to default values.
pub fn fit(
x: &X,
y: &Y,
@@ -611,7 +611,8 @@ mod tests {
&[10., -2.],
&[8., 2.],
&[9., 0.],
]);
])
.unwrap();
let y = vec![0, 0, 1, 1, 2, 1, 1, 0, 0, 2, 1, 1, 0, 0, 1];
@@ -671,7 +672,8 @@ mod tests {
&[10., -2.],
&[8., 2.],
&[9., 0.],
]);
])
.unwrap();
let y = vec![0, 0, 1, 1, 1, 1, 1, 0, 0, 1, 1, 1, 0, 0, 1];
@@ -733,7 +735,8 @@ mod tests {
&[10., -2.],
&[8., 2.],
&[9., 0.],
]);
])
.unwrap();
let y: Vec<i32> = vec![0, 0, 1, 1, 2, 1, 1, 0, 0, 2, 1, 1, 0, 0, 1];
let lr = LogisticRegression::fit(&x, &y, Default::default()).unwrap();
@@ -818,37 +821,41 @@ mod tests {
assert!(reg_coeff_sum < coeff);
}
// TODO: serialization for the new DenseMatrix needs to be implemented
// #[cfg_attr(all(target_arch = "wasm32", not(target_os = "wasi")), wasm_bindgen_test::wasm_bindgen_test)]
// #[test]
// #[cfg(feature = "serde")]
// fn serde() {
// let x = DenseMatrix::from_2d_array(&[
// &[1., -5.],
// &[2., 5.],
// &[3., -2.],
// &[1., 2.],
// &[2., 0.],
// &[6., -5.],
// &[7., 5.],
// &[6., -2.],
// &[7., 2.],
// &[6., 0.],
// &[8., -5.],
// &[9., 5.],
// &[10., -2.],
// &[8., 2.],
// &[9., 0.],
// ]);
// let y: Vec<i32> = vec![0, 0, 1, 1, 2, 1, 1, 0, 0, 2, 1, 1, 0, 0, 1];
//TODO: serialization for the new DenseMatrix needs to be implemented
#[cfg_attr(
all(target_arch = "wasm32", not(target_os = "wasi")),
wasm_bindgen_test::wasm_bindgen_test
)]
#[test]
#[cfg(feature = "serde")]
fn serde() {
let x: DenseMatrix<f64> = DenseMatrix::from_2d_array(&[
&[1., -5.],
&[2., 5.],
&[3., -2.],
&[1., 2.],
&[2., 0.],
&[6., -5.],
&[7., 5.],
&[6., -2.],
&[7., 2.],
&[6., 0.],
&[8., -5.],
&[9., 5.],
&[10., -2.],
&[8., 2.],
&[9., 0.],
])
.unwrap();
let y: Vec<i32> = vec![0, 0, 1, 1, 2, 1, 1, 0, 0, 2, 1, 1, 0, 0, 1];
// let lr = LogisticRegression::fit(&x, &y, Default::default()).unwrap();
let lr = LogisticRegression::fit(&x, &y, Default::default()).unwrap();
// let deserialized_lr: LogisticRegression<f64, i32, DenseMatrix<f64>, Vec<i32>> =
// serde_json::from_str(&serde_json::to_string(&lr).unwrap()).unwrap();
let deserialized_lr: LogisticRegression<f64, i32, DenseMatrix<f64>, Vec<i32>> =
serde_json::from_str(&serde_json::to_string(&lr).unwrap()).unwrap();
// assert_eq!(lr, deserialized_lr);
// }
assert_eq!(lr, deserialized_lr);
}
#[cfg_attr(
all(target_arch = "wasm32", not(target_os = "wasi")),
@@ -877,7 +884,8 @@ mod tests {
&[4.9, 2.4, 3.3, 1.0],
&[6.6, 2.9, 4.6, 1.3],
&[5.2, 2.7, 3.9, 1.4],
]);
])
.unwrap();
let y: Vec<i32> = vec![0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1];
let lr = LogisticRegression::fit(&x, &y, Default::default()).unwrap();
@@ -899,4 +907,46 @@ mod tests {
assert!(reg_coeff_sum < coeff);
}
#[cfg_attr(
all(target_arch = "wasm32", not(target_os = "wasi")),
wasm_bindgen_test::wasm_bindgen_test
)]
#[test]
fn lr_fit_predict_random() {
let x: DenseMatrix<f32> = DenseMatrix::rand(52181, 94);
let y1: Vec<i32> = vec![1; 2181];
let y2: Vec<i32> = vec![0; 50000];
let y: Vec<i32> = y1.into_iter().chain(y2.into_iter()).collect();
let lr = LogisticRegression::fit(&x, &y, Default::default()).unwrap();
let lr_reg = LogisticRegression::fit(
&x,
&y,
LogisticRegressionParameters::default().with_alpha(1.0),
)
.unwrap();
let y_hat = lr.predict(&x).unwrap();
let y_hat_reg = lr_reg.predict(&x).unwrap();
assert_eq!(y.len(), y_hat.len());
assert_eq!(y.len(), y_hat_reg.len());
}
#[test]
fn test_logit() {
let x: &DenseMatrix<f64> = &DenseMatrix::rand(52181, 94);
let y1: Vec<u32> = vec![1; 2181];
let y2: Vec<u32> = vec![0; 50000];
let y: &Vec<u32> = &(y1.into_iter().chain(y2.into_iter()).collect());
println!("y vec height: {:?}", y.len());
println!("x matrix shape: {:?}", x.shape());
let lr = LogisticRegression::fit(x, y, Default::default()).unwrap();
let y_hat = lr.predict(&x).unwrap();
println!("y_hat shape: {:?}", y_hat.shape());
assert_eq!(y_hat.shape(), 52181);
}
}
src/linear/ridge_regression.rs
+4 -3
View File
@@ -40,7 +40,7 @@
//! &[502.601, 393.1, 251.4, 125.368, 1960., 69.564],
//! &[518.173, 480.6, 257.2, 127.852, 1961., 69.331],
//! &[554.894, 400.7, 282.7, 130.081, 1962., 70.551],
//! ]);
//! ]).unwrap();
//!
//! let y: Vec<f64> = vec![83.0, 88.5, 88.2, 89.5, 96.2, 98.1, 99.0,
//! 100.0, 101.2, 104.6, 108.4, 110.8, 112.6, 114.2, 115.7, 116.9];
@@ -455,7 +455,8 @@ mod tests {
&[502.601, 393.1, 251.4, 125.368, 1960., 69.564],
&[518.173, 480.6, 257.2, 127.852, 1961., 69.331],
&[554.894, 400.7, 282.7, 130.081, 1962., 70.551],
]);
])
.unwrap();
let y: Vec<f64> = vec![
83.0, 88.5, 88.2, 89.5, 96.2, 98.1, 99.0, 100.0, 101.2, 104.6, 108.4, 110.8, 112.6,
@@ -513,7 +514,7 @@ mod tests {
// &[502.601, 393.1, 251.4, 125.368, 1960., 69.564],
// &[518.173, 480.6, 257.2, 127.852, 1961., 69.331],
// &[554.894, 400.7, 282.7, 130.081, 1962., 70.551],
// ]);
// ]).unwrap();
// let y = vec![
// 83.0, 88.5, 88.2, 89.5, 96.2, 98.1, 99.0, 100.0, 101.2, 104.6, 108.4, 110.8, 112.6,