The Altair Community is migrating to a new platform to provide a better experience for you. In preparation for the migration, the Altair Community is on read-only mode from October 28 - November 6, 2024. Technical support via cases will continue to work as is. For any urgent requests from Students/Faculty members, please submit the form linked here
"Algorithm run over new data - id label???"
Contributor I
Hello all,
I am rather new to RapidMiner and to get an idea I found the page neuralmarkettrends.com. Here, Tom gives a few lessons on how to use RM, which I find very useful. The example he illustrates is how to predict the trend of Gold. I am sure a few of you know that page, so maybe you can help me on this problem.
I have data in excel (an equity index, bond index interest rates, etc.) and I define the corresponding direction of Gold. I run RM over this and it gives me the "best" weights on the input to explain the development of Gold (doing this I had to specify as label ID the Gold trend (up, down, range, or something like that) and an create ID which I chose to be the date.
RM creates a .mod-file which I then want to run over new data to predict the direction of Gold tomorrow. In the new data file I do not have a label ID (because the Gold trend is what I want to predict). What Tom does in his tutorial is putting the label ID to zero (he uses RM 4.0). I cannot put in to Zeri though (RM 4.4) because it jumps back to 1. 1 however is the date column. Do you know how to solve this problem? Did you ever experience something like this?
And, in general, do you know any book or tutorial that gives assistance in getting better in RM?
Thank you,
el_martin
I am rather new to RapidMiner and to get an idea I found the page neuralmarkettrends.com. Here, Tom gives a few lessons on how to use RM, which I find very useful. The example he illustrates is how to predict the trend of Gold. I am sure a few of you know that page, so maybe you can help me on this problem.
I have data in excel (an equity index, bond index interest rates, etc.) and I define the corresponding direction of Gold. I run RM over this and it gives me the "best" weights on the input to explain the development of Gold (doing this I had to specify as label ID the Gold trend (up, down, range, or something like that) and an create ID which I chose to be the date.
RM creates a .mod-file which I then want to run over new data to predict the direction of Gold tomorrow. In the new data file I do not have a label ID (because the Gold trend is what I want to predict). What Tom does in his tutorial is putting the label ID to zero (he uses RM 4.0). I cannot put in to Zeri though (RM 4.4) because it jumps back to 1. 1 however is the date column. Do you know how to solve this problem? Did you ever experience something like this?
And, in general, do you know any book or tutorial that gives assistance in getting better in RM?
Thank you,
el_martin
Tagged:
0
Answers
Firstly, welcome to the asylum! Now to biz... 4.0 is very ancient and much has moved on since then, but do not be alarmed, the principles remain the same. The ID column is just that, it carries a unique identifier for each row. Using dates as ID's is most natural, I do it for my site which is also in the financial domain.
My guess is that in the bad old days column 1 was referred to by 0, and that in the intervening years that has been cleaned up. So if your spreadsheet has the date in column 1, stuff in 1 as the ID column and press on to glory!
As to learning more... Unfortunately the RM documentation is pretty unhelpful. I gave up and went on a course. Working, and reworking through the examples is a must, and this forum has loads of good stuff.
Good luck amigo !