Neural Net Lottery Picker

MillionsWanted,

Congratulations with your win!!!

"Please tell more about "Fixed Seed Input Weights." 

Based on the following information I added the fixed seed input weights:

• neural networks tend to get stuck in local minima, so it's a good idea to give them many different starting values. You can't do that if they all start at zero.
• if the neurons start with the same weights, then all the neurons will follow the same gradient, and will always end up doing the same thing as one another.
• Many processes are sensitive to their initial conditions (Chaos theory)
• ...

Version 1.2 had some minor bugs. A new release (version 1.2.1) of the tool is available for download.

Thanks. Looks like it's more than luck involved.

Everything has a pattern to the game.

Hi stoopendaal,

Great little utility!

When saving results, is there any way to just save the picks and not the other columns?

Thanks for the program.

This is a question for stoopendaal:

I have a file of a lottery game drawing history that I am ready to import into the Neural Net Lottery Picker.

A line from my file looks like this:

3745 08/20/2015 08-09-25-30-33

This line includes the drawing number followed by the date of the drawing followed by the numbers themselves as they were drawn on that day.

According to the instructions in your software,  I am supposed to delete the drawing number and the date from each line of the numbers that I import into your software so that each line looks like this:

08-09-25-30-33       for example.

Is this correct?  Please confirm.  Thanks.

Sincerely,

Kosh

It still wont open. I trying the net frame work repair.

  I just opened the files to notepad and put on the desktop. I got it to work!

Hi,
I can not download the new versão1.2.1
how to ?

Hi Kosh,

Yes, that's correct.

I use Excel to separate the draw results from the other data like drawing number, date etc. (In Excel: Data => Text To Columns).

Examples:

1. numbers separated by -

5-12-23-25-28-43
9-22-28-30-36-40
11-15-16-19-28-31
10-17-26-37-42-44
6-17-22-37-41-42
1-5-29-32-39-43
11-12-17-18-31-38
14-18-24-32-43-44

2. numbers separated with whitespace

5 12 23 25 28 43
9 22 28 30 36 40
11 15 16 19 28 31
10 17 26 37 42 44
6 17 22 37 41 42
1 5 29 32 39 43
11 12 17 18 31 38
14 18 24 32 43 44
2 4 9 16 25 45
1 8 20 26 28 42
3 9 15 19 30 38
10 17 23 24 27 41
11 15 17 26 36 41
7 14 16 29 43 45

Hi sandnan,

It's the same link you can find here in this topic (see my response to lakerben).

Hi goodlukman,

"When saving results, is there any way to just save the picks and not the other columns?"  ==> In the next version of the tool this option will be available.

Thank you.

By the way, can you add a config file or some other method to save previous settings?

im trying for p3.

nm data with spaces:

9 1 8
2 3 0
7 5 4
4 0 6
2 9 2
7 9 9
6 0 7
2 5 9
4 4 7
1 4 3
1 6 1
8 0 5
0 3 6
2 5 5
0 4 5
6 0 5

results

01-03-00-;5.37690892432576;0.579704069133864;0.473329718665528;10:52:02.1406250;Alpha:9E-05| Noise:0.25|Threshold Error:0.01|Hidden Neurons:50|Test Data Range:50
05-02-03-;4.00376595182496;0.50023531661515;0.884440918571907;10:52:02.6562500;Alpha:9E-05| Noise:0.25|Threshold Error:0.01|Hidden Neurons:50|Test Data Range:50
02-01-04-;3.98175736113642;0.49885853212211;0.699059834828399;10:52:04.6875000;Alpha:9E-05| Noise:0.25|Threshold Error:0.01|Hidden Neurons:50|Test Data Range:50
03-01-04-;3.90601470357413;0.494091002724583;0.837280760786149;10:52:13.1718750;Alpha:9E-05| Noise:0.25|Threshold Error:0.01|Hidden Neurons:50|Test Data Range:50
06-02-00-;3.90151190661012;0.493806130139281;0.758613462889047;10:52:17.1718750;Alpha:9E-05| Noise:0.25|Threshold Error:0.01|Hidden Neurons:50|Test Data Range:50
04-00-03-;3.87122114091556;0.4918854758043;0.761455749016093;10:52:43.5625000;Alpha:9E-05| Noise:0.25|Threshold Error:0.01|Hidden Neurons:50|Test Data Range:50
03-04-01-;3.84848166023655;0.49043868502065;0.795372762511657;10:54:35.0625000;Alpha:9E-05| Noise:0.25|Threshold Error:0.01|Hidden Neurons:50|Test Data Range:50
04-02-01-;3.84515306826676;0.490226546370831;0.81004200121261;10:55:06.6875000;Alpha:9E-05| Noise:0.25|Threshold Error:0.01|Hidden Neurons:50|Test Data Range:50

stoopendaal,

I don't believe that the threshold(stop training) parameter is working.   The results "average errors" as well as "last error" have exceeded my threshold settings and the program never stops the training process. 

Here is an example:

If I set the threshold(stop training) to .68 and the last error is .617846400703044 shouldn't the program stop training? Or maybe I don't understand how the program works?