{"id":2984,"date":"2018-07-05T08:58:33","date_gmt":"2018-07-05T13:58:33","guid":{"rendered":"https:\/\/blogs.mathworks.com\/loren\/?p=2984"},"modified":"2018-07-05T08:58:33","modified_gmt":"2018-07-05T13:58:33","slug":"memoize-functions-in-matlab","status":"publish","type":"post","link":"https:\/\/blogs.mathworks.com\/loren\/2018\/07\/05\/memoize-functions-in-matlab\/","title":{"rendered":"Memoize Functions in MATLAB"},"content":{"rendered":"

Very early on in this blog (2006!), I wrote a post on memoizing functions<\/a>, i.e., caching results so outputs that have already been calculated don't need to be calculated again (code included at the end of this post). Memoization can provide a significant performance boost especially if the function in question is expensive to calculate, and is likely to have inputs repeated. <\/p>\r\n

It wasn't until recently that I realized this functionality (memoize<\/a>) was added to MATLAB in R2017a. Needless to say, the shipping function is different than the solution I presented over 10 years ago. And without the limitations that mine had (limited to elementwise functions that had a single input). <\/p>\r\n

Table of Contents<\/h3><\/p>\r\n

What is Memoization? <\/a>
Let's Try It <\/a>
Is That All? <\/a>
Do You Use Memoization? <\/a>
From Reference from My 2006 Post<\/a>
<\/p>\r\n

What is Memoization?<\/h3>\r\n

The idea of memoization is to cache function results from specific inputs so if these same inputs are used again, the function can simply return the values computed earlier, without rerunning the computation. This can be useful if you have a function that is very expensive to compute. <\/p>\r\n

Of course, if you run the memoized function a lot, it will take up increasing amounts of memory as unique inputs get added to the list, unless we do something to limit the cache size. That's what MATLAB does now with the function memoize<\/inline>. <\/p>\r\n

Let's Try It<\/h3>\r\n

As in my earlier post, let's try something simple, the function sin<\/inline>. <\/p>

fmem = memoize(@sin)\r\n<\/pre>
fmem = \r\n  MemoizedFunction with properties:\r\n\r\n     Function: @sin\r\n      Enabled: 1\r\n    CacheSize: 10\r\n<\/pre>
y = fmem(pi.\/(1:5)')\r\n<\/pre>
y = 5\u00d71<\/em>\r\n\r\n   1.2246e-16\r\n            1\r\n      0.86603br\r\n      0.70711\r\n      0.58779\r\n\r\n<\/pre>
So, we still get the answers we expect.<\/p>\r\n      
Now let's compute some more values, some already in the cache and others not.<\/p>
ymore = fmem(pi.\/(1:10)')\r\n<\/pre>
ymore = 10\u00d71<\/em>\r\n\r\n   1.2246e-16\r\n            1\r\n      0.86603\r\n      0.70711\r\n      0.58779\r\n          0.5\r\n      0.43388\r\n      0.38268\r\n      0.34202\r\n      0.30902\r\n\r\n<\/pre>
Again, no surprises on the out. The values are the ones we expect. I am not doing enough computation here for you to see the benefit of reduced time from caching, however. <\/p>\r\n      
Is That All?<\/h3>\r\n      
Of course not! There are a bunch of choices you can use to control how much information gets cached, etc. Here's some links for more information. <\/p>\r\n