[SOLVED] Selecting attribute subset specified in a "meta" exampleset

tennenrishin

Hi

Suppose we have examplesetA with a (design-time-unknown) set of attributes:
E.g.
a b c d e <-- attributes
1 2 3 4 5 <-- example 1
1 2 3 4 5 <-- example 2

We also have examplesetB holding (design-time-unknown) examples:
E.g.
X <-- attribute
a <-- example 1
d <-- example 2
e <-- example 3

The output that is needed is
a d e
1 4 5
1 4 5

The examplesetB specifies how we want to select attributes in examplesetA.

What is the best way to do this in RM?
(
Should I create a regex by looping over the examples in B?
Should I select the attributes one at a time by looping over the examples in B, and join them all together in that loop?
Is there another simpler way that I'm not thinking about?
)

Thanks for any help.

MariusHelf

Hi,

I often go for the first approach of creating a regex from the second example set.

But if your actual data does not contain missings, you could also define attribute X in Example Set B as Id, transpose it, remove all attributes (with Filter Example Range), use the Union operator to combine it with examplesetA and then remove all attributes which contain missing values via Filter Attributes.

Good luck!
~Marius