F# Weekly #1, 2013

07/01/201307/01/2013F# Weekly3 Comments

Welcome to F# Weekly,

А promising start of the new year:

News

Quick F-AI update: added a Noisy-Or learner (BN EM) and introduced some “smart” log-arithmetic.
Released FunScript NuGet package.
SharpDevelop 4.3 Beta was released.
Microsoft’s ‘Orleans’ cloud programming model gets a Halo test drive.
“Add Support for F# Code-Behind ASP.NET” is currently a Top 5 ASP.NET request with > 500 votes.
The XML provider in F# data library now supports recursive documents like XHTML. See “Global inference mode“.
Luca Bolognese published “Functional programming in C – I“.
Anton Kropp presented a super simple LL(1) recursive descent parser for csv lists.
The CSV type provider now supports custom separators and “#N/A” values.
Foq: an F# mocking library v0.5 now available on Nuget with an optional Linq Expression based API for C# & F#

Blogs & Tutorials

Neil Danson shared “Mono and Mac“.
Jonathan Allen posted “Introducing the F# Software Foundation“.
Sebastian W wrote about “F# and MS CRM“.
Fabian Scherschel published “A foundation for F#“.
Steffen Forkmann blogged “F# and Microsoft Dynamics NAV 2009 Web Services“.
Jerome Laban presented “Toying around with F# Queries, Rx, Portables Libraries, Windows [Phone] 8 and the Zip operator“.
Saxon Matt posted “Adopting F#“.
Dave Thomas published “MonoGame Subdivision and Platonics“.
Colin Fang presented “A Simple Overview on How Pattern Match Compiles“.
Steffen Forkmann blogged “F# and Microsoft Dynamics NAV 2013 OData Services“.
Jack Fox posted “Semantics and List-like Data Structures“.
Darryl K. Taft published “F# Foundation: Taking Microsoft’s F# Language to a Higher Ground“.

That’s all for now. Have a great week and remember you can message me on twitter (@sergey_tihon) with any F# news.

Previous F# Weekly edition – #52

Enable Web Services Enhancements (WSE) 3.0 in Visual Studio 2008, 2010 and 2012

03/01/201303/01/2013Uncategorized1 Comment

The same steps work in Visual Studio 2012

Diganta Kumar's Blog

Web Service Enhancements 3 (WSE 3) is not officially supported since Visual Studio 2008. The reason is that Microsoft wants you to migrate your code to WCF. See below to enable Web Services Enhancements (WSE) 3.0 in VS2012, VS2010 and VS2008.

1. Download and install Web Services Enhancements (WSE) 3.0 for Microsoft .NET. Make sure you have all the files after installing WSE v3.0. In Window 7 the location is “C:\Program Files (x86)\Microsoft WSE\v3.0\Tools”. Note: Close Visual Studio before installing.

2. Go to the folder %ALLUSERSPROFILE%\Application Data\Microsoft\MSEnvShared\AddIns (notice that “Application Data” is hardcoded, which shouldn’t because Windows XP localizes that folder). If the folder is not there close Visual Studio and create the folder as show below.

Examples:
– Windows XP: “C:\Documents and Settings\All Users\Application Data\Microsoft\MSEnvShared\AddIns”
– Windows Vista / Windows 7: “C:\ProgramData\Microsoft\MSEnvShared\AddIns”. ( This is a hidden folder. Copy paste the UNC path to Windows Explorer.)

3. In the…

View original post 242 more words

F# Weekly #52, 2012 – New Year Edition

31/12/201231/12/2012F# Weekly3 Comments

Welcome to F# Weekly,

This is 10th anniversary edition of F# Weekly and last in this year. I want to say thank you to those who were with me all the time. I wish you Happy New Year full of functional happiness, love and new achievements.

Last portion of news from 2012:

News

Microsoft Dynamics CRM type provider preview have been presented.
Completed work on a command-line compiler and nuget package for FunScript.

Blogs & Tutorials

Neil Danson blogged “The Roslyn Incident“.
Mathias Brandewinder presented “Support Vector Machine in F#: getting there“.
Steve Horsfield published “F#: Composing Functions“.
Sergey Tihon blogged “F#/.NET function minimization (optimization)“.
Jack Fox posted “Engineering Random Bits: F# Generics“.
Sergey Tihon wrote about “Rhythm of the F# Сommunity heartbeat“.
Jonathan Allen posted “Building Type Providers for F#“.
Mathias Brandewinder released “AdaBoost in F#“.
Adam Granicz wrote about “Running the jQuery Mobile WebSharper sample“.
Ryan Riley posted “F# Manifest“.

That’s all for now. See you in the New 2013 Year.

Previous F# Weekly edition – #51

Rhythm of the F# Сommunity heartbeat

29/12/201201/06/2013F#, F# Weekly3 Comments

The new year is getting closer. It is time of magic and miracles, time to sum up and to lift the veil of secrecy over F# Community.

I have been collecting all tweets (and retweets) with hashtag #fsharp since the beginning of November. It is time to show up some statistics. 😉

When are we tweeting? – Always!

F# Community Heartbeat. London time zone (UTC)

Do we have a rest from F#? – A bit on Saturday.

DayOfWeek — Tweets distribution by day of week.

Who to follow? – Choose yourself.

Top 20 most active users (based on the tweets and RT)

Who writes more of unique tweets?

Top 20 users that generates unique tweets.

Who is the best retweeter?

Top 20 best retweeters. — Top 20 the best retweeters.

F#/.NET function minimization (optimization)

25/12/201214/11/2013F#, Machine Learning and NLP1 Comment

I have done some research on function minimization algorithms implemented on .NET. Short summary can be found below.

Gradient descent

Gradient descent is one of the simplest function optimization algorithms. You can implement it by yourself or using one of the following articles:

“Gradient descent” by Jon Harrop
“Gradient descent algorithm. F#” by Igor Shapiro

DotNumerics

DotNumerics is a Numerical Library for .NET. The library is written in pure C# and has more than 100,000 lines of code with the most advanced algorithms for Linear Algebra, Differential Equations and Optimization problems.

Unfortunately, dotNumerics does not have a detailed documentation. Let’s go through all minimization algorithms implemented in dotNumerics. First of all, we implement banana function from simplex method example available on the library site.

#r @"DotNumerics.dll"
open System
open DotNumerics.Optimization

//f(a,b) = 100*(b-a^2)^2 + (1-a)^2
let BananaFunction (x: float array) =
    100.0 * Math.Pow((x.[1] - x.[0] * x.[0]), 2.0) + Math.Pow((1.0 - x.[0]), 2.0)

Downhill Simplex

Downhill Simplex method of Nelder and Mead

The key advantage of Downhill Simplex method is that it does not require the gradient function. All you need is a function and an initial guess.

let initialGuess = [|0.1; 2.0|]

let simplexMin =
    let simplex = Simplex();
    simplex.ComputeMin(BananaFunction,initialGuess);

We have a bit of control over the evaluation model. We can restrict MaxFunEvaluations and specify custom Tolerance in Simplex model. In this case, model instantiation looks like below.

    let simplex = Simplex(MaxFunEvaluations=10000, Tolerance=1e-5);

Truncated Newton

“A Survey of Truncated-Newton Methods”, Journal of Computational and Applied Mathematics.

All other algorithms require gradient function to make calculation.


//f'a(a,b) = (100*(b-a^2)^2 + (1-a)^2)'a = 100*2*(b-a^2)*(-2a) - 2*(1-a)
//f'b(a,b) = (100*(b-a^2)^2 + (1-a)^2)'b = 100*2*(b-a^2)
let BananaFunctionGradient (x: float array) =
    [|100.0 * 2.0 * (x.[1] - x.[0] * x.[0]) * (-2.0 * x.[0]) - 2.0 * (1.0 - x.[0]);
      100.0 * 2.0 * (x.[1] - x.[0] * x.[0])|]

let newtonMin =
    let newton = TruncatedNewton()
    newton.ComputeMin(BananaFunction,BananaFunctionGradient,initialGuess);

Truncated Newton algorithm has three more configuration parameters than Downhill Simplex: Accuracy, MaximunStep and SearchSeverity.

L-BFGS-B

Limited memory Broyden–Fletcher–Goldfarb–Shanno method

let bfgsMin =
    let lbfgsb = L_BFGS_B()
    lbfgsb.ComputeMin(BananaFunction, BananaFunctionGradient, initialGuess);

L-BFGS-B has one more configuration parameters than Downhill Simplex – it is AccuracyFactor.

Results

Below you can find evaluation results received from models with default parameters.

Real: 00:00:00.024, CPU: 00:00:00.062, GC gen0: 0, gen1: 0, gen2: 0
val simplexMin : float [] = [|0.999999998; 0.9999999956|]
Real: 00:00:00.074, CPU: 00:00:00.078, GC gen0: 0, gen1: 0, gen2: 0
val newtonMin : float [] = [|0.9999999999; 0.9999999999|]
Real: 00:00:00.137, CPU: 00:00:00.140, GC gen0: 0, gen1: 0, gen2: 0
val bfgsMin : float [] = [|1.0; 1.0|]

F# Weekly #51, 2012

24/12/2012F# Weekly1 Comment

Welcome to F# Weekly,

A roundup of F# content from this past week:

News

New video! “An Informal Deep Dive With Don Syme: The Freebase Type Provider“.
New F# book coming soon! 😉 “F# Deep Dives” by Tomas Petricek, Phillip Trelford and Co.
Vote to make F# first-class language in the ASP.NET stack.
Announced public CI for F# Community projects.
Natallie Baikevich announced “fquanty” – a collection of common performance functions.
Tomas Petricek proposed to use “F# Formatting” to document F# projects.

Blogs & Tutorials

Daniel Mohl published “F# PowerPack NuGet Packages for F# 3.0, .NET 4.x, and Silverlight 5“.
Tomas Petricek blogged “The Building Blocks of a F# Markdown Parser“.
Reto Matter posted “Functional Feed-forward Neural Networks Part I: Setting it up“.
Tomas Petricek posted “Manning: F# Deep Dives deal of the day“.
Tomas Petricek blogged “Processing trees with F# zipper computation“.
Robert Jeppesen presented “Sql optimization trick in F#“.
Examples and slides from Functional Programming Day in Oslo.
Tao Liu presented “Functional Programming and/or F# Conferences“.
Sergey Tihon proposed to use “F# for heating“.
Daniel Mohl blogged “Buiilding Web, Cloud, and Mobile Solutions with F#“.

That’s all for now. Have a great week and remember you can message me on twitter (@sergey_tihon) with any F# news.

Previous F# Weekly edition – #50

F# for heating

22/12/201201/06/2013F#, Tips and Tricks8 Comments

Sometimes, when you feel really alone and want a bit of heat then F# can help you here.

All you need is a laptop and F#. Just type the following snippet into FSI and wait for a minute. =)

[|1..999|] |> Array.Parallel.iter (fun _ ->
    while true do ignore())

Wish you warm Christmas!

P.S. The same solution works when you are freezing.

F# Weekly #50, 2012

17/12/201217/12/2012F# Weekly1 Comment

Welcome to F# Weekly,

A roundup of F# content from this past week:

News

Infer.NET 2.5 turns the simple succinct syntax of F# into a probabilistic modeling language for Bayesian machine learning. (video presentation)
Tomas Petricek presented “F# Data: Library for Data Access“.
David Siegel shared printf-style function for calling external programs with typesafe commandlines.
Paulmichael Blasucci pushed new version of fs-zmq (an F# binding for ØMQ)

Blogs & Tutorials

Daniel Mohl published “Using the New ASP.NET MVC 4 Template in MonoDevelop“.
Richard Warburton blogged “Are all Map Functions the same?“.
Steffen Forkmann presented “Testing, proving, checking – How to verify code properties“.
Steffen Forkmann presented “Using Footloose from F# – part I – Setting up the infrastructure“.
Srdjan Strbanovic blogged “Simple Continuous Integration with FAKE and windows scheduler F#“.
Jack Fox shared “Twitter and the F# Programming Community“.
Michael Newton posted a new blog post “Anyone Using FsCheck for Business Code?“.
Luca Bolognese blogged “LLite : language friendly literate programming“.
Kevin Berridge wrote about “Neat F#: Custom Operators“.
Tao Liu presented “Debug F# Type Provider“.
Faisal Waris blogged “Rich / Mobile Web Development – Without Losing Your Sanity“.

Upcoming events

That’s all for now. Have a great week and remember you can message me on twitter (@sergey_tihon) with any F# news.

Previous F# Weekly edition – #49

F# Weekly #49, 2012

10/12/2012F# Weekly1 Comment

Welcome to F# Weekly,

A roundup of F# content from this past week:

News

Tomas Petricek shared examples of “Charting World Bank data with F#“.
F# 3.0 is now part of Mono 3.0.2 bundle on Mac.
New section “Math Stacks for F#” was published on The F# Software Foundation.
Keith Sheppard announced “An F# binding for LLVM“.

Blogs & Tutorials

Tao Liu published “F# Computational Expression Sample – File System + XML“.
James Litsios blogged “Introduction to stateful monads“.
Ted Neward on The Advantages of F#.
Tomas Petricek shared a presentation “How F# Learned to Stop Worrying and Love the Data“.
Luca Bolognese blogged “Exceptions vs. Return Values to represent errors (in F#) – IV – Implementation“.
Don Syme presented “QuickLinq Helpers“.
Don Syme published “F# Community Update“.
Daniel Mohl wrote of “Using the New ASP.NET MVC 4 Template in MonoDevelop“.
Richard Warburton blogged “Are all Map Functions the same?“.
Zach Bray published “Building an “actor” in F# with higher throughput than Akka and Erlang actors“.
Tao Liu presented “Remove F# snippet shortcut“.

That’s all for now. Have a great week and remember you can message me on twitter (@sergey_tihon) with any F# news.

Previous F# Weekly edition – #48

Sergey Tihon's Blog

Running Home Assistant & Matter Server on a UGREEN NAS: A Deep Dive into Thread Device Commissioning

Pi-hole DNS on UGOS PRO 1.6.0.2917

F# Weekly | Behind The Scenes

New IKVM 8.2 & MavenReference for .NET projects

Announcing OpenXML Package Explorer for VS Code

`dotnet watch` with Microsoft.Identity.Web or custom IDistributedCache

ML.NET Recommendation Engine: Pitfall of One-Class Matrix Factorization

Introducing Clippit, get your slides out of PPTX.

HashiCorp Vault and TLS Certificate Authentication for .NET Applications (Comprehensive guide)

Be better WPF / MvvmLight developer in 2018

F# Weekly #1, 2013

Enable Web Services Enhancements (WSE) 3.0 in Visual Studio 2008, 2010 and 2012

F# Weekly #52, 2012 – New Year Edition

Rhythm of the F# Сommunity heartbeat

F#/.NET function minimization (optimization)

Gradient descent

DotNumerics

Downhill Simplex

Truncated Newton

L-BFGS-B

Results

F# Weekly #51, 2012

F# for heating

F# Weekly #50, 2012

F# Weekly #49, 2012