问题陈述：编写一个函数，该函数返回使用给定的硬币面额对目标金额进行找零所需的最小硬币数。
我试图将下面的 i32integer-basedsolution转换为基于floatf32 或 f64 的解决方案，以便它可以接受十进制输入，如硬币面额0.5、1.5、2.0等。

use std::cmp;

fn min_number_of_change(n: i32, denoms: Vec<u32>) -> i32 {
    let mut ways: Vec<i32> = vec![i32::MAX;n as usize + 1];
    ways[0] = 0;

    for denom in denoms.iter() {
        for current in 0..ways.len() {
            if *denom <= current as u32 {
                ways[current as usize] = cmp::min(ways[current as usize], 1 + ways[current as usize - *denom as usize]) 
                                  
            }
        }
    }
    if ways[n as usize] != i32::MAX {
        ways[n as usize]
    } else {
          -1
    }
}
fn main() {
    let denoms: Vec<u32> = vec![1, 5, 10, 2, 3];
    let n: i32 = 6;

    let result: i32 = min_number_of_change(n, denoms);
    println!("Result: {}", result);
}

Play Ground for above code
我很天真地尝试用f32和min函数替换i32来进行浮点数比较。运行时，编译器抱怨类型不匹配，无法从usize中减去f32，类型[f32]无法被f32索引。我想我遗漏了一些非常基本的要点。

fn min_number_of_change(n: f32, denoms: Vec<f32>) -> f32 {
    let mut ways: Vec<f32> = vec![f32::INFINITY; n + 1.0];
    ways[0] = 0.0;

    for denom in denoms.iter() {
        for current in 0..ways.len() {
            if *denom <= current {
                ways[current] = (ways[current].min(1 + ways[current  - *denom ]), 1 + ways[current  - *denom ])
            }
        }
    }
    if ways[n] != f32::INFINITY {
        ways[n]
    } else {
        -1.0
    }
}

fn main() {
    let denoms: Vec<f32> = vec![2.00, 1.00, 0.50, 0.20, 0.10, 0.05, 0.02, 0.01];
    let n: f32 = 4.55;

    let result: f32 = min_number_of_change(n, denoms);
    println!("Result: {}", result);
}

Playground for above code