Vahaduo Online tool for G25

[anglesqueville]

I had a short debate with Ruderico on the French subforum about Vahaduo. I rather agree with him that for individual analyses this new tool doesn't bring anything new. It's certainly a bit quicker, that's all, and the lack of a penalty can be a problem. BUT if you work with large target files, it's a real benediction. I don't know who is this Vahaduo guy, but I'm ready to become president of his fan club. Don't miss to try also https://vahaduo.github.io/g25views/#

[JMcB]

Personally, I love the format and the presentation. Unfortunately, the lack of a penalty option makes it problematic in my case.

Here is a run against the population averages with penalty, using Poi’s FreeForm Runner.

JMcB_scaled
1000 cycles
500 batches

Fit: 1.4058

English: 40.4
French_Brittany: 26.8
Swedish: 4.2
Norwegian: 3.8
German: 3
Basque_Spanish: 2
English_Cornwall: 2
Basque_French: 1.6
French_East: 1.6
Spanish_Pais_Vasco: 1.6
Swiss_French: 1


And here’s a run on Vahaduo:


Attachment 32817


My ancestry is in my signature.

Hopefully, he’ll be adding a penalty option down the road.

[Trelvern]

Let us compare this new tool and R

moderns averaged

V



0.02312409 English_Cornwall
0.02315099 Belgian
0.02338748 French_Brittany
0.02461122 English
0.02629482 Scottish
0.02632177 Welsh
0.02771466 Orcadian
0.02776520 Irish
0.02846279 French
0.03064985 Swiss_German
0.03068600 Dutch
0.03085576 German
0.03131455 French_East
0.03161421 Shetlandic
0.03553250 Norwegian
0.03793997 Icelandic
0.04104319 Austrian
0.04112135 Swiss_French
0.04464463 Swedish
0.04914864 German_East
0.04964553 Spanish_Penedes
0.05118265 Hungarian
0.05158701 Spanish_Girona
0.05175398 Spanish_Barcelones
0.05208923 Spanish_Camp_de_Tarragona


Target: Trelvern_scaled
Distance: 1.3381% / 0.01338051
Aggregated
48.8 Welsh
20.4 Basque_Spanish
18.8 English_Cornwall
5.2 Kaitag
4.0 Spanish_Navarra
1.2 Ket
1.2 Yemenite_Jew
0.2 Shetlandic
0.2 Wichi



R

"1. CLOSEST SINGLE ITEM DISTANCE%"
English_Cornwall Belgian French_Brittany English
2.312409 2.315099 2.338748 2.461122
Scottish Welsh Orcadian Irish
2.629482 2.632177 2.771466 2.776520

pen=0.001


Welsh,31.6
English_Cornwall,27.6
Scottish,4.8
Irish,3.8
Belgian,3.6
French,3.2
French_Brittany,2.6
Basque_Spanish,1.6
Shetlandic,1.4
Spanish_Camp_de_Tarragona,1.4
Swiss_French,1.4
Basque_French,1
Spanish_Cantabria,1
Spanish_Peri-Barcelona,1
Spanish_Aragon,0.8
Spanish_Castilla_La_Mancha,0.8
Spanish_Catalunya_Central,0.8
Spanish_Pais_Vasco,0.8
Spanish_Soria,0.8
Spanish_Terres_de_l'Ebre,0.8
French_South,0.6
Norwegian,0.6
Spanish_Alacant,0.6
Spanish_Barcelones,0.6
Spanish_Mallorca,0.6
Spanish_Valencia,0.6
....

pen=0

Trelvern_scaled

Welsh,49
Basque_Spanish,19.8
English_Cornwall,18.8
Kaitag,4.8
Spanish_Navarra,4.6
Ket,1.2
Yemenite_Jew,1.2
Darginian,0.2
Shetlandic,0.2
Wichi,0.2


Pros:

very easy to use
very fast
closest distances more than 8

Cons:

Only pen=0

So for an individual (as Anglesqueville said) a moderate interest
but very good at processing a lot of data.

[Finn]

I had a short debate with Ruderico on the French subforum about Vahaduo. I rather agree with him that for individual analyses this new tool doesn't bring anything new. It's certainly a bit quicker, that's all, and the lack of a penalty can be a problem. BUT if you work with large target files, it's a real benediction. I don't know who is this Vahaduo guy, but I'm ready to become president of his fan club. Don't miss to try also https://vahaduo.github.io/g25views/#

Thanks looks nice but get only errors when I fill in the coordinates....Do you have a solution?

[JMcB]

Thanks looks nice but get only errors when I fill in the coordinates....Do you have a solution?

Did you removed the PC numbers:

,PC1,PC2,PC3,PC4,PC5,PC6,PC7,PC8,PC9,PC10,PC11,PC1 2,PC13,PC14,PC15,PC16,PC17,PC18,PC19,PC20,PC21,PC2 2,PC23,PC24,PC25?

If not, try doing so.

[anglesqueville]

Thanks looks nice but get only errors when I fill in the coordinates....Do you have a solution?

I'm quite sure that the answer is what JMcB said. The blanks are automatically removed (nice). I've used Vahaduo a lot since I got it (thanks to Eurogens), and found no problem. Use it offline, it's far quicker. You only have to save the complete html page, you'll get an icon (depending on the internet platform you are currently using).

[Trelvern]

I'm quite sure that the answer is what JMcB said. The blanks are automatically removed (nice). I've used Vahaduo a lot since I got it (thanks to Eurogens), and found no problem. Use it offline, it's far quicker. You only have to save the complete html page, you'll get an icon (depending on the internet platform you are currently using).

"The blanks are automatically removed"
i noticed
Very useful indeed
Is there any other tool that does it?
Would be helpful with some

[JMcB]

"The blanks are automatically removed"
i noticed
Very useful indeed
Is there any other tool that does it?
Would be helpful with some

I believe Poi’s FreeForm Runner does that, too.

[anglesqueville]

I didn't use the "fast mode" so far. I guess that it has to do with the number of iterations in the Monte-Carlo algorithm. Try it and compare the outputs. Anyway, it may be useful only for huge source or target files, as the program is really a speedy one (I repeat because it's really important: run it offline if you use large files).

[Finn]

Did you removed the PC numbers:

,PC1,PC2,PC3,PC4,PC5,PC6,PC7,PC8,PC9,PC10,PC11,PC1 2,PC13,PC14,PC15,PC16,PC17,PC18,PC19,PC20,PC21,PC2 2,PC23,PC24,PC25?

If not, try doing so.

thanks...strange my input was this 0.134311267881738,0.142174118601101,0.064864755299 0066,0.0607242910209745,0.0424694859870001,0.02314 7924744996,0.00681530850365557,0.00899962499218717 ,0.00777190581517815,-0.00309801387989144,-0.00617076008284231,0.00569493107947761,-0.0114468729354353,-0.0013762267255071,0.0196794435896953,0.0068946588 0229036,0.0,0.00658780691884636,0.0049022239850908 5,0.00562769935231085,0.0059894306908086,0.0051934 1506140227,-0.00308119295079033,0.014098378630183,-0.00191599582463011

than with one space,
no effect.....
may be due to safari?