rust

.gitignore

@@ -0,0 +1 @@
+/target

Cargo.toml

@@ -0,0 +1,12 @@
+[package]
+name = "barclays"
+version = "0.1.0"
+edition = "2021"
+
+# See more keys and their definitions at https://doc.rust-lang.org/cargo/reference/manifest.html
+
+[dependencies]
+lazy_static = "1.4.0"
+ndarray = "0.15.6"
+polars = { version = "0.38.3", features = ["ndarray"] }
+smartcore = { version = "0.3.2", features = ["ndarray-bindings"] }

src/lib.rs

@@ -0,0 +1 @@
+pub mod model;

src/main.rs

@@ -0,0 +1,3 @@
+fn main() {
+    println!("Hello, world!");
+}

src/model.rs

@@ -0,0 +1,53 @@
+pub mod preprocessing;
+
+use std::collections::HashMap;
+use lazy_static::lazy_static;
+use polars::prelude::*;
+use smartcore::linalg::basic::matrix::DenseMatrix;
+
+lazy_static! {
+    static ref CATEGORICAL_COLUMNS: Vec<&'static str> = vec![
+        "person_home_ownership", 
+        "loan_intent", 
+        "loan_grade", 
+        "cb_person_default_on_file",
+    ];
+    
+    static ref HOME_OWNERSHIP_VALUES: HashMap<&'static str, u32> = HashMap::from([
+        ("RENT", 1),
+        ("MORTGAGE", 2),
+        ("OWN", 3),
+        ("OTHER", 4),
+    ]);
+
+    static ref INTENT_VALUES: HashMap<&'static str, u32> = HashMap::from([
+        ("EDUCATIONAL", 1),
+        ("MEDICAL", 2),
+        ("VENTURE", 3),
+        ("PERSONAL", 4),
+        ("DEBTCONSOLIDATION", 5),
+    ]);
+
+    static ref GRADE_VALUES: HashMap<&'static str, u32> = HashMap::from([
+        ("A", 1),
+        ("B", 2),
+        ("C", 3),
+        ("D", 4),
+        ("E", 5),
+    ]);
+
+    static ref DEFAULT_FILE_VALUES: HashMap<&'static str, u32> = HashMap::from([
+        ("Y", 1),
+        ("N", 2),
+    ]);
+}
+
+pub fn init_default_predictor() {
+    let data = CsvReader::from_path("credit_risk_dataset.csv")
+        .unwrap()
+        .finish()
+        .unwrap()
+        .to_ndarray::<Float64Type>(IndexOrder::Fortran)
+        .unwrap();
+
+}

src/model/preprocessing.rs

@@ -0,0 +1,19 @@
+use smartcore::{metrics::distance::euclidian::Euclidian, neighbors::knn_regressor::{KNNRegressor, KNNRegressorParameters}};
+use ndarray::{Array1,Array2};
+
+pub struct KNNImputer {
+    model: KNNRegressor<f64,f64,Array2<f64>,Array1<f64>,Euclidian<f64>>
+}
+
+impl KNNImputer {
+    pub fn new(dataframe: ndarray::Array2<f64>, target_column: usize) -> Self {
+            let true_values: Array1<f64> = dataframe.column(target_column).to_vec().into();
+            KNNImputer {
+             model: KNNRegressor::fit(&dataframe, &true_values.into(), Default::default()).unwrap(),
+            }
+    }
+
+    pub fn impute(&self) -> bool {
+        //TODO: Predict value, 
+    }
+}