I have a dataset which are keyed on {"Item", "Hour of Day", "Value"}.

In the examples, I noticed that it is possible to apply functions using such as

Select[(#a > 1 &)]

But what would the slot name be for my string values above? Is there a way to interrogate a dataset for its slots?

Should I simply avoid the issue and rename them if I wish to perform such operations?

  • $\begingroup$ Did you try #["Item"] and #Item as described in the docu? $\endgroup$ – Karsten 7. Jul 19 '14 at 18:01
  • 1
    $\begingroup$ @Karsten7. Thanks, that does it. If you had put this as an answer, then I would be happy to close the question. $\endgroup$ – Eric Brown Jul 19 '14 at 18:11

The functionality of Slot(#) got extended in Mma version 10. As described in the documentation you can use




to pick out the elements with the key "Item".

As pointed out in the comment by @alancalvitti


also works, as

#"Item" & // InputForm

#Item &

and can be used for your second key:

#"Hour of Day"
| improve this answer | |
  • 3
    $\begingroup$ For keys with protected characters need quotes in the 2nd form above: #"SUBJECT_ID" $\endgroup$ – alancalvitti Jul 19 '14 at 20:44
  • $\begingroup$ Quoted names are not documented for special characters (spaced cases for example). Nice Tip. +1 $\endgroup$ – Murta Jul 19 '14 at 21:24

If your keys are not strings, then it is a bit trickier, as Slot[keyname] does not seem to work. The only form I have managed to get working is, for example,

Select[Key[keyname][#] > 1 &][dataset]
| improve this answer | |

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.