Add clean workflow documentation with proper base

Author: hnlearndev

Cargo.lock

@@ -11,3 +11,5 @@ rayon = "1.10.0"
 spreadsheet-ods = "0.25.0"
 uuid = "1.17.0"
 serde_json = "1.0.140"
+itertools = "0.14.0"
+chrono = "0.4.41"

Cargo.toml

@@ -1,66 +1,21 @@
+use chrono::NaiveDate;
 use ndarray::prelude::*;
 use ndarray_rand::RandomExt;
 use ndarray_rand::rand::SeedableRng;
 use ndarray_rand::rand::rngs::StdRng;
 use ndarray_rand::rand_distr::Uniform;
 use polars::prelude::*;
 
-//---------------------------------------------------------------------------------------------------------
-// STRUCTS
-//---------------------------------------------------------------------------------------------------------
-pub struct ModelPoint {
-    pub model: String,
-    pub id: i32,
-    pub entry_age: i32,
-    pub gender: String,
-    pub term: i32,
-    pub policy_count: f64,
-    pub sum_insured: f64,
-    pub duration_mth: Option<i32>,  // For SE model
-    pub issue_date: Option<String>, // For ASL_SE model
-    pub payment_freq: Option<i32>,  // For ASL_SE model - 1, 4, 6, 12
-    pub payment_term: Option<i32>,  // For ASL_SE model
-}
+pub mod asl_se_mp_gen;
+pub mod pricing_mp_gen;
+pub mod s_mp_gen;
+pub mod se_mp_gen;
+
+
 
 //---------------------------------------------------------------------------------------------------------
 // PUBLIC
 //---------------------------------------------------------------------------------------------------------
-pub fn generate_s_model_points(mp_size: usize, seed: usize) -> PolarsResult<DataFrame> {
-    // Get seed for random number generation
-    let mut rng = StdRng::seed_from_u64(seed as u64);
-
-    // Issue Age (Integer): Random 20 - 59 year old
-    let entry_age = Array1::random_using(mp_size, Uniform::new(20, 60), &mut rng); // 60 is exclusive, so range is 20-59
-
-    // Gender (String): Random "M" and "F"
-    let gender_binary = Array1::random_using(mp_size, Uniform::new(0, 2), &mut rng); // 0 or 1
-    let gender: Vec<&str> = gender_binary
-        .iter()
-        .map(|&x| if x == 0 { "M" } else { "F" }) // map 0 to "M" and 1 to "F"
-        .collect();
-
-    // Policy term (Integer): Random 10, 15 or 20
-    let term = (Array1::random_using(mp_size, Uniform::new(2, 5), &mut rng)) * 5;
-
-    // Policy count
-    let policy_count = Array1::<f64>::ones(mp_size);
-
-    // Sum insured (Float): Random values between 100,000 and 1,000,000 (multiple of 1000)
-    let sum_insured = Array1::random_using(mp_size, Uniform::new(0.0f64, 1.0f64), &mut rng) // Random floats between 0 and 1
-        .mapv(|x| (((900_000.0 * x + 100_000.0) / 1000.0).round() * 1000.0));
-
-    // Create a DataFrame with the generated data
-    let model_points_df = df![
-        "id"  => (1..(mp_size+1) as i32).collect::<Vec<i32>>(),
-        "entry_age" => entry_age.to_vec(),
-        "gender" => gender,
-        "term" => term.to_vec(),
-        "policy_count" => policy_count.to_vec().into_iter().collect::<Vec<f64>>(),
-        "sum_insured" => sum_insured.to_vec(),
-    ]?;
-
-    Ok(model_points_df)
-}
 
 pub fn convert_model_points_df_to_vector(df: &DataFrame) -> PolarsResult<Vec<ModelPoint>> {
     let id = df.column("id")?.i32()?;

src/assumptions/assumptions.ods

@@ -0,0 +1,34 @@
+use super::*;
+use itertools::iproduct;
+
+pub fn generate_s_model_points() -> PolarsResult<DataFrame> {
+    let age: Vec<i32> = (20..60).collect();
+    let term: Vec<i32> = vec![10, 15, 20];
+    let gender: Vec<&str> = vec!["M", "F"];
+
+    // Generate all combinations of age, term and gender
+    let combinations: Vec<(i32, i32, &str)> = iproduct!(age.iter(), term.iter(), gender.iter())
+        .map(|(&a, &t, &g)| (a, t, g))
+        .collect();
+
+    let model: Vec<&str> = vec!["s_model"; combinations.len()];
+    let id: Vec<i32> = (1..=combinations.len() as i32).collect();
+    let age: Vec<i32> = combinations.iter().map(|(a, _, _)| *a).collect();
+    let term: Vec<i32> = combinations.iter().map(|(_, t, _)| *t).collect();
+    let gender: Vec<&str> = combinations.iter().map(|(_, _, g)| *g).collect();
+    // If more information on demographics is needed, it can be added here.
+    let policy_count: Vec<f64> = vec![1.0; combinations.len()];
+    let sum_insured: Vec<f64> = vec![1000.0; combinations.len()];
+
+    let df = df![
+        "model" => model,
+        "id" => id,
+        "age" => age,
+        "term" => term,
+        "gender" => gender,
+        "policy_count" => policy_count,
+        "sum_insured" => sum_insured,
+    ]?;
+
+    Ok(df)
+}

src/model_points.rs

@@ -0,0 +1,38 @@
+use super::*;
+
+pub fn generate_s_model_points(mp_size: usize, seed: usize) -> PolarsResult<DataFrame> {
+    // Get seed for random number generation
+    let mut rng = StdRng::seed_from_u64(seed as u64);
+
+    // Issue Age (Integer): Random 20 - 59 year old
+    let entry_age = Array1::random_using(mp_size, Uniform::new(20, 60), &mut rng); // 60 is exclusive, so range is 20-59
+
+    // Gender (String): Random "M" and "F"
+    let gender_binary = Array1::random_using(mp_size, Uniform::new(0, 2), &mut rng); // 0 or 1
+    let gender: Vec<&str> = gender_binary
+        .iter()
+        .map(|&x| if x == 0 { "M" } else { "F" }) // map 0 to "M" and 1 to "F"
+        .collect();
+
+    // Policy term (Integer): Random 10, 15 or 20
+    let term = (Array1::random_using(mp_size, Uniform::new(2, 5), &mut rng)) * 5;
+
+    // Policy count
+    let policy_count = Array1::<f64>::ones(mp_size);
+
+    // Sum insured (Float): Random values between 100,000 and 1,000,000 (multiple of 1000)
+    let sum_insured = Array1::random_using(mp_size, Uniform::new(0.0f64, 1.0f64), &mut rng) // Random floats between 0 and 1
+        .mapv(|x| (((900_000.0 * x + 100_000.0) / 1000.0).round() * 1000.0));
+
+    // Create a DataFrame with the generated data
+    let model_points_df = df![
+        "id"  => (1..(mp_size+1) as i32).collect::<Vec<i32>>(),
+        "entry_age" => entry_age.to_vec(),
+        "gender" => gender,
+        "term" => term.to_vec(),
+        "policy_count" => policy_count.to_vec().into_iter().collect::<Vec<f64>>(),
+        "sum_insured" => sum_insured.to_vec(),
+    ]?;
+
+    Ok(model_points_df)
+}

src/projections/projection_mp/asl_se_model.rs renamed to src/model_points/asl_se_mp_gen.rs

@@ -1,6 +1,9 @@
 use crate::assumptions::assumption_scenario::AssumptionScenario;
-use polars::prelude::*; // Add this line or adjust the path as needed
+use chrono::NaiveDate;
+use ndarray::Array1;
+use polars::prelude::*;
 
-pub mod asl_se_model;
-pub mod s_model;
-pub mod se_model;
+pub mod asl_se;
+pub mod structs;
+pub mod s;
+pub mod se;