mirror of
https://github.com/Brandon-Rozek/website.git
synced 2024-12-24 04:51:25 +00:00
39 lines
No EOL
1.3 KiB
Markdown
39 lines
No EOL
1.3 KiB
Markdown
---
|
|
date: 2022-11-12 11:49:51-05:00
|
|
draft: false
|
|
math: false
|
|
medium_enabled: true
|
|
medium_post_id: 2558d6abd9b8
|
|
tags:
|
|
- Scala
|
|
title: Memoization in Scala
|
|
---
|
|
|
|
In a [recent post](/blog/corecursion-unfold-infinite-sequences/), I talked about how corecursion is a great solution for removing redundant calculations. However if we're sticking to a recursive approach, one way we can reduce redundancies is to use memoization. The idea here is that we save prior computations in some data structure and refer to them if requested.
|
|
|
|
[Pathikrit on StackOverflow](https://stackoverflow.com/a/36960228) provided a great solution for Scala using hashmaps:
|
|
|
|
```scala
|
|
import scala.collection.mutable
|
|
def memoize[I, O](f: I => O): I => O = new mutable.HashMap[I, O]() {self =>
|
|
override def apply(key: I) = self.synchronized(getOrElseUpdate(key, f(key)))
|
|
}
|
|
```
|
|
|
|
If the input is already a key in the hashmap, then we return its value. Otherwise we calculate the output via `f` and update the hashmap.
|
|
|
|
Now lets memoize say the Fibonacci sequence
|
|
|
|
```scala
|
|
val fib : Int => Int = memoize {
|
|
case 0 => 0
|
|
case 1 => 1
|
|
case n => fib(n - 1) + fib(n - 2)
|
|
}
|
|
```
|
|
|
|
Calling `fib(5)` returns `5`. However, we can also see the saved computations by analyzing the hashmap `fib`.
|
|
|
|
```
|
|
HashMap(0 -> 0, 1 -> 1, 2 -> 1, 3 -> 2, 4 -> 3, 5 -> 5)
|
|
``` |