dlls <\/em>references:<\/p>\nopen Accord.Statistics.Models.Regression.Linear\r\nopen FSharp.Data<\/pre>\nIn F#, the CSV model can be implemented as follows:<\/p>\n
type CupcakeSale = \r\n {CupcakeId : int;\r\n Type : string;\r\n Price : float;\r\n CustomerId : int;\r\n Rating : float;\r\n Total: float}<\/pre>\nNote:F# does not use any kind of code separator, instead it takes into consideration the whitespaces as indentation.<\/em><\/div>\nA type<\/em> in F# represents a class in C# or other object-oriented languages. This way, you can define the name and types of your object attributes (which is also a variable).<\/p>\nOnce you\u2019re dealing with a large amount of records, you\u2019ll create a resizable list, import the local CSV file, iterate over its content and add its elements to the list just to simulate the data a real application would supply:<\/p>\n
let sales = ResizeArray<CupcakeSale>()\r\ntype CupcakeSalesCSV = CsvProvider<\"cupcake_sales.csv\", \",\">\r\nlet csv = new CupcakeSalesCSV()\r\nfor row in csv.Rows do\r\n sales.Add({CupcakeId = row.CupcakeId;\r\n Type = row.Type;\r\n Price = (float) row.Price;\r\n CustomerId = row.CustomerId;\r\n Rating = (float) row.Rating;\r\n Total = (float) row.Total})<\/pre>\nFurther, you can increase the number of inputs in this list to achieve more accuracy with the algorithm result. Or, if you prefer, you can get the data from a data source or even a web service.<\/p>\n
The next step is to group the data in a way to have the rating average values (as the x<\/em> in the formula) vs the average of the total of times each cupcake was ordered (the y<\/em>, the most important information).<\/p>\nlet filterByRating = [(fun (x : CupcakeSale) -> x.Rating)]\r\nlet filterByTotal = [(fun (x : CupcakeSale) -> x.Total)]\r\nlet groupedByRatings = \r\n sales\r\n |> Seq.groupBy(fun x -> x.CupcakeId)\r\n |> Seq.map(fun (key, values) -> \r\n (key, [for prop in filterByRating -> values |> Seq.averageBy prop]))\r\n |> Seq.toArray\r\nlet groupedByTotals = \r\n sales\r\n |> Seq.groupBy(fun x -> x.CupcakeId)\r\n |> Seq.map(fun (key, values) -> \r\n (key, [for prop in filterByTotal -> values |> Seq.averageBy prop]))\r\n |> Seq.toArray<\/pre>\nThe variables filterBy-Something<\/strong> are only a condition attached to the grouping + filtering made upon groupedBy-Something<\/strong> variables. The operator |><\/em><\/strong> (pipe-forward) allows you to pass an intermediate\/temporary result (on the leftmost side) to the next function (on the closest right side).<\/p>\nThe function groupBy <\/strong>gets the given list (sales<\/strong>) and returns a sequence of tuples containing the unique key (the cupcake id<\/strong>) and all the elements that key matches within another sequence. The next function, map<\/strong>, receives these tuples mapped to key\/value variables. Then, the lambda expression iterates over the values to apply an averageBy <\/strong>function against the filterBy-(Rating\/Total)<\/strong> condition you\u2019ve seen before and finally returns an array containing each tuple of id\/(rating and total averages). <\/strong>Check the following REPL result as shown in Figure 8.<\/p>\n![](\"https:\/\/www.red-gate.com\/simple-talk\/wp-content\/uploads\/2018\/07\/word-image-23.png\")
<\/p>\n
Figure 8: Calculating the averages and totals<\/p>\n
Tuples are represented by parentheses and the first element is called key<\/em> (or left<\/em>), the second is value<\/em> (or right<\/em>).<\/p>\nThe linear regression object is created from the x<\/em> and y<\/em> values which, in turn, must be represented as arrays with each index of each array corresponding to the respective values<\/em> and total<\/em>. Because of that, the code iterates over the array of tuples (of both groupedBySomething <\/strong>variables) to extract this correlated data:<\/p>\nlet cupcakeIds = Array.create groupedByRatings.Length 0\r\nlet x_averages = Array.create groupedByRatings.Length 0.0\r\nlet y_totals = Array.create groupedByRatings.Length 0.0\r\nfor i in 0..groupedByRatings.Length - 1 do\r\n let averageTuple = Array.get groupedByRatings i\r\n let totalTuple = Array.get groupedByTotals i\r\n let _id = fst averageTuple\r\n let _average = (snd averageTuple).[0]\r\n let _total = (snd totalTuple).[0]\r\n Array.set cupcakeIds i _id\r\n Array.set x_averages i _average\r\n Array.set y_totals i _total<\/pre>\nArray.create<\/strong> is a function that creates a new array with the given length and fills it with the given values. The reserved words fst<\/strong> and snd<\/strong> are used to access the first and second values of a tuple in F#, respectively.<\/p>\nFinally, you can review the linear regression code:<\/p>\n
let ols = OrdinaryLeastSquares()\r\nlet regression = ols.Learn(x_averages, y_totals)\r\n\/\/ Let's say you received this value from the web form\r\nlet purchasedCupcakeId = 1;\r\n\/\/ A shortcut to find the corresponding sale of the provided cupcake id\r\nlet findIndex arr elem = arr |> Array.findIndex((=) elem)\r\nlet i = findIndex cupcakeIds purchasedCupcakeId\r\nlet result = regression.Transform(Array.get x_averages i)\r\nSystem.Console.WriteLine(\"Suggested quantity: {0}\", System.Math.Round(result))<\/pre>\nOrdinaryLeastSquares<\/em> (OLS) is the standard linear regression approach used by the Accord.NET library. It is based on the famous least-squares procedure that, in turn, models a single and continuous variable response (y) predicted by another variable recorded on an interval (x) through a linear relationship. In other words, OLS is the most basic and common mathematical representation of the linear regression formula: y = f(x) + e. Its function called Learn<\/strong> is responsible for receiving the x\/y arrays and so learn by them (more and more every time you increase the arrays) resulting in the regression itself. Once you have seen the model digested by the regression function, all you need to do is to provide a cupcake rating average of a specific cupcake type and the suggested value will be generated as result in the beginning of the F# Interactive output shown in Figure 9. Note that the value was rounded since the user won\u2019t be able to buy a fraction of a cupcake :<\/p>\n![](\"https:\/\/www.red-gate.com\/simple-talk\/wp-content\/uploads\/2018\/07\/word-image-24.png\")
<\/p>\n
Figure 9: The suggested quantity<\/p>\n
Now, try it yourself. Feed your algorithm by increasing the number of cupcake sales to the list and analyzing how the linear regression calculates the best fit for (low) user reviews versus the total of number of orders.<\/p>\n
Conclusion<\/h2>\n
Using .NET libraries to accomplish machine learning algorithms not only makes it possible to use the power of data science to help predict different sort of things, but also allows you to integrate code easily in all .NET projects\/technologies. Of course, it goes further: It provides fast ways to export code to dll\u2019s<\/em> that can be imported into other languages that know how to translate one type of object into another.<\/p>\nThe code in this article is, perhaps, the simplest type of linear regression that can be built with machine learning. Fortunately, its official website<\/a> is full of awesome content, more complex examples of regression, classification, tests of hypotheses and even image\/audio manipulation. With regard to F#, both the official project<\/a> and the .NET support documentation<\/a> pages are well documented in a way that is extremely easy to understand. All of that is available to help you to take bigger steps as you learn.<\/p>\n","protected":false},"excerpt":{"rendered":"The .NET F# language can be used for machine learning. In this article, Diogo Souza explains what is needed in Visual Studio to take advantage of this feature and walks you through a simple regression example. …<\/p>\n","protected":false},"author":320401,"featured_media":0,"comment_status":"open","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"footnotes":""},"categories":[143538,47],"tags":[],"coauthors":[60461],"class_list":["post-79591","post","type-post","status-publish","format-standard","hentry","category-dotnet-development","category-data-science"],"acf":[],"_links":{"self":[{"href":"https:\/\/www.red-gate.com\/simple-talk\/wp-json\/wp\/v2\/posts\/79591","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.red-gate.com\/simple-talk\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.red-gate.com\/simple-talk\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.red-gate.com\/simple-talk\/wp-json\/wp\/v2\/users\/320401"}],"replies":[{"embeddable":true,"href":"https:\/\/www.red-gate.com\/simple-talk\/wp-json\/wp\/v2\/comments?post=79591"}],"version-history":[{"count":10,"href":"https:\/\/www.red-gate.com\/simple-talk\/wp-json\/wp\/v2\/posts\/79591\/revisions"}],"predecessor-version":[{"id":79621,"href":"https:\/\/www.red-gate.com\/simple-talk\/wp-json\/wp\/v2\/posts\/79591\/revisions\/79621"}],"wp:attachment":[{"href":"https:\/\/www.red-gate.com\/simple-talk\/wp-json\/wp\/v2\/media?parent=79591"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.red-gate.com\/simple-talk\/wp-json\/wp\/v2\/categories?post=79591"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.red-gate.com\/simple-talk\/wp-json\/wp\/v2\/tags?post=79591"},{"taxonomy":"author","embeddable":true,"href":"https:\/\/www.red-gate.com\/simple-talk\/wp-json\/wp\/v2\/coauthors?post=79591"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}
![](\"https:\/\/www.red-gate.com\/simple-talk\/wp-content\/uploads\/2018\/07\/word-image-16.png\")
<\/p>\n![](\"https:\/\/www.red-gate.com\/simple-talk\/wp-content\/uploads\/2018\/07\/word-image-17.png\")
<\/p>\n![](\"https:\/\/www.red-gate.com\/simple-talk\/wp-content\/uploads\/2018\/07\/word-image-18.png\")
<\/p>\n![](\"https:\/\/www.red-gate.com\/simple-talk\/wp-content\/uploads\/2018\/07\/word-image-19.png\")
<\/p>\n![](\"https:\/\/www.red-gate.com\/simple-talk\/wp-content\/uploads\/2018\/07\/word-image-20.png\")
<\/p>\n![](\"https:\/\/www.red-gate.com\/simple-talk\/wp-content\/uploads\/2018\/07\/word-image-21.png\")
<\/p>\n![](\"https:\/\/www.red-gate.com\/simple-talk\/wp-content\/uploads\/2018\/07\/word-image-22.png\")
<\/p>\n