csvToWx

1.'writePatternfile_____2.'aMemoryRAW'________3.'aOsZilloskop1RAW'_____4.'UNDERDEVELOPMENT'

modes of Files: 1- WX 2- TXT

Setup: Three modes

• 1. A GHC GUI
• 2. Automatic (the same as above avoiding GUI's UNDERCONSTRUCTION )
• 3. Easy Acces (quickly read a csv file and process...)

i. will open a simple Graphical User Interface (GUI) in the Glasgow Haskell Compiler (GHC) the aim is to automate reading of csv data and of long number streams converting them and changing them via predefined functions. The whole program is derived from four main functions.

ii. Runs all functions as i. but avoiding the GUI...(realisation semi construction 28-09-19) of 4 main functions some can be completly controlled just via Global-Variables*.

iii. Quick and easy access to read csv -> write patternfile.txt

Main

• 1. A GHC GUI enter> main

  The advantage of using Global Variables, the stream-crypt program can be fully automated which helps overseing bigger longer files and keeping an eye on the real aim of the Wx-Maxima application.

  The disadvantage of using Global Variables is that we can loose track of which functions we just applied to our data, getting no visiual feedback, both sides can be avoided be directly calling the "PatternfileRAW' functions that need more variables but can be easily be called via your own batch/bat files or any other program.

Four functions to use

The whole stream-crypt revolves around four functions that do the 'dirty' work. The easiest and almost laziest way is by utiliesing commandline and directly typ in the function plus variabes.

1. 'writePatternfile'

via 'Patternfile.hs' read a csv file from disc and export it as aPatternfile.txt e.g enter: 'writePatternfile x nF'

x="fileTorRead.csv" ; nF = "fileToWrite.txt"

   writePatternfile x nF => format [val,year,month,day]

fire up your editor and jump to line :858 below the mentioned code and the GHC GUI are pictured to compare Enable 'writePatternfile' to read other csv formats!

2. 'aMemoryRAW'

writes screen1.txt statistical data about the data set autoInputII:String; e.g "1" if ==1 -> hide output ?! => opens menu: Options : POINTCLOUD 1..4,6: Pointcloud CLOSE 5

    => maybe writes: "lala.wxm" filled with  1 or 2 ..or 4 or 6
             content: if 1 or 2 or 3 or 4 or 6 == True -> do P.aMemoryRAW autoInputII file22writ file33writ
             with: autoInputII = "1"; file22writ; file33writ;
                      else do P.aMemoryRAW "2" file22writ file33writ
                      
    => writes: "screenI.txt" --a crude statistical overview of the set 
    => proceed: start a second Computation from the same source
    => reads: fileToread.txt (same as step above)
    => order: applies predefiend function to the value 
    =>        [MQ3,MQ4,MQ5,MQ6] -- group of sin functions (periodic)
    => do Wx.WriteMQScreenI && Wx.WriteMQ6ScreenII
    => load: if  then 'WriteWXmaximafile.hs* -> do writeMQ6SCREENI
             else 'WriteWXmaximafile.hs* -> do writeMQ6SCREENII

select any String you to import as as line into Wx-Maxima, 'aMemory' will convert it to WX.maxima format. Set to a random pointcloud while a myriad other settings are out there.

    => function: aMemoryRAW outputOrnot fileToread fileTowrite 
                              String     String    String
    => :*Patternfile>aMemoryRAW "1" "seniot.txt" "seniot7.txt"             

3. 'aOsZilloskop1RAW'

will prepare real vals and simu vals, will exportDATA.txt in first run (simu vals for re-import) same function will exportDATA2.txt cleaned just vals [String]

=> writes: exportDatat1,exportData2,HTML
=> function: aOsZilosko1 outputOrnot fileToread   fileTowrite   howMAny   crit   foHans   fun1 fun2
=> types:                  String     String        String      String   String   String   Int  Int
=> *Patternfile>aMemoryRAW "1" "seniot.txt" "seniot7.txt"

First and formost yields a Clean.txt file:[String]

that by now has been sorted. This file can be reimported as aPatternfile again but thus the

Overview: snd Computation in aOsZilloskop1

1.choose Simulation (not fully functional yet) will simulate but too much noise

2. call every single statistical function used in Patternfile and write to file.txt

3. calls aOszilloskop again and applies our Gobal Var selectFunction (Patternfile.hs)

5.CLOSES

6. Help UNDERCONSTRCTION

Is set to use MQ functions, which are simple sin functions, nice periodic functions.

       => in 'Patternfile.hs' see global variables 
       : set function: selectFuncL = ( show [1,2,3,4,5]) 
       : control which functions to select of 

MQ-Functions

The name stems from 'Magisches Quadrat' magic square in English. The Oszilloskop uses [MQ3,MQ4,MQ5,MQ6]

The 3x3.

"There is just one 3x3 magic square...".

    => in WriteWXmaximaFile.hs
       fofourier1 = [4,9,2]
       fofourier2 = [3,5,7]
       fofourier3 = [8,1,6]
       
       fourier1 (x) = (sin((head fofourier1)*x)+ sin ((last (take 2 fofourier1))*x) + sin ((last fofourier1)*x)) 
       fourier2 (x) = (sin((head fofourier2)*x)+ sin ((last (take 2 fofourier2))*x) + sin ((last fofourier2)*x))
       fourier3 (x) = (sin((head fofourier3)*x)+ sin ((last (take 2 fofourier3))*x) + sin ((last fofourier3)*x))
       fourier123 (x) = (fourier1 x + fourier2 x + fourier3 x)*(1/90)

Remarkably fourier123 (x) Reihen und spalten are the same function!? which as a argument the thing behaves a little bit like a magic square as plottedn below MQ3Rows = MQ3Collums ,a periodic function to explore.

Example: Stream-Crypt at work

The intention was to 'bend' Data sets improving upon my first crude statitical analysis- I could say let the whole set end lower than the input set depending on the highest/lowest points and where I will prope the data, e.g

picMQ3:

A simple sin would have done it as well :) but just for curiosity reasons I kept that one, or did it ???. Depending on how an 'interval' is defined it may bring some differences. I wanted to be able to visiualize one of one MQ3, one of the four pandiagonal 4x4 and .... 5x5 , ..6x6 mixed with a given set of vals, source and simulated and just gaze into the abyss :) of data science. Hence if I take one interval of the MQ3 function and a let it loose on the data, naturally the data Set will end up lower than the original one. In 'aOsZilloskop1' the vals are 'befriended' with a MQ function. changing the vals of a given data set: I can say:

            => length (ofSet) = someNumber `:= Ls1
            => with: ls1 = (one Interval of MQ3)

As it turns out the input vals in MQ3 do resemble the original function picMQ3, here plotted with the original input data. but is not 'attuned' to the original vals by 'attuned' i mean that the y1 and y2 values of the interval of one function selected for plotting, should be really close to the (y1..y2 ...interval) of another of these functions.

e.g Patternfile.hs :1723 in OsZilloskop1

One value of the function close to zero is estimated : Such as the simuVals y1 y2 interval relate to the realVals y1 y2 interval as above.

    => let mixWithMQ6 --x nF crit --openB file to open ; -openA file to write ; forD criterium

e.g Patternfile.hs :1723 in OsZilloskop1

          ----------------OSZILLOSKOP SHOWS INPUT AND RANDOM
        One value of the function close to zero is estimated :
          =  let ay1= aOsZilloskop1 output x nF criterium (read rnd) -- nF crit
   
             in let ay2 p = fourier1MQ6 p + fourier2MQ6 p + fourier3MQ6 p + fourier4MQ6 p + fourier5MQ6 p + fourier6MQ6 p 
             in let ay3 =  (6.28319901) -- NullStelle Des Intervals
             in let ay4 = (ay3/ (read xX)) -- takes forC howmany lines  
             in let soMany2take fk = ay4 * fk -- determines which value of fourier123 2 select
             in let forThis fk = fourierMQ6NOPAN123 (soMany2take fk)
             in let fofoThis = map forThis [1..(read xX)]
             in fofoThis --intoConsider

picAbyss:

Alright MQ3 did not work(yet) the others did :) and now the red simuVal problem reveals itself, while green being the real vals. So by building a thieve that sums up all real and simulated val within OsZilloskop1 and around we probe to 'tune' into the data to find variables and the right ones

What is good about this

The functions now tuned in can be used in simulations about the data set if using the various MQ functions.

The drawback

This is all but visualisation there is no problem solved by making this a plot. If the generated data may contain any information what sort ever it must be carried out by another domain that is capable to 'weigh' or 'rate' each newly generated function

good about the drawback

Using the dit rnd function we can make each generated function 'unique' and that possibly could help that other 'weigh' rate' domain.

4.'aCrunchlist1RAW'

UNDERDEVELOPMENT most basic HTA/HTML plotter can be easily changed from hta to html by Haskell generated VB.script that cannot be displayed in (firefox,..) just typing differet format afile.hta <=> afile.html,

Hello Microsoft ;), the hta runs => writes: TESTGuiVal.hta Haskell generated VB.script that cannot be displayed in (firefox,..)

writes senio, writes Html, using global variables
=> writes: TESTGuiVal.hta

Patternfile.hs 'writePatternFile "fileToRead.csv" "fileToWrite.txt"' Reads a CSV file e.g: "durble.csv" Writes a a file with line format (val,year,month,day) a file called 'Patternfile.txt' 'aMemory ' Reads 'Patternfile.txt' Turns it into a String will plot some statistic about the given DATA Set. Write a WX.Maxima file

Nexus: Armchair-phillosohy ,

What is good about the code?

The stuff is without any blows and whistles, like 'sophisticated Haskell'

One of the most remarkable sentences about Haskell to me, is to keep it rather simple more functional with less fancy code that will blow number theorisists minds but not be easyly handled for simple programming tasks. The main Menu helps to automate tasks.

The drawback

Especially Patternfile.txt has lengthy repetitive code passages that could be easily shortend up and compacted with purer code. The first aim of the Stream-Crypt 1.1.19 was to setup a working program that can handle longer files with a new Main module and many adjustments from 'older' (three years) code. By just focusing on the overall program content and how to automate wxm file's intention and writing
less on the brilliance of just a single function. .

What is good about the drawback

repetitive and very similar code may point to use a temptiningly different approach of coding which is a task for the future. https://github.com/simonmar/par-tutorial .