Timeline for Reshaping associations, generalization of GroupBy
Current License: CC BY-SA 3.0
17 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Feb 22, 2018 at 7:26 | vote | accept | Kuba | ||
| Feb 16, 2016 at 23:48 | answer | added | Ronald Monson | timeline score: 2 | |
| Feb 16, 2016 at 21:56 | answer | added | Mr.Wizard | timeline score: 6 | |
| Feb 15, 2016 at 18:01 | answer | added | WReach | timeline score: 15 | |
| Feb 15, 2016 at 15:02 | comment | added | Kuba |
@J.M. because it won't work in general: GroupBy[dataset, {#type &, #subtype &, #type &, #subtype &}]. However, GroupBy[dataset, {#type &, #subtype &, #type &, #subtype &}, Map[f, #, {3}] &] would be kind of ok, since 3 is 4-1 :P (take a look at gwr's answer)
|
|
| Feb 15, 2016 at 14:57 | history | tweeted | twitter.com/StackMma/status/699246074704371712 | ||
| Feb 15, 2016 at 14:54 | answer | added | gwr | timeline score: 5 | |
| Feb 15, 2016 at 14:51 | comment | added | J. M.'s missing motivation♦ |
Why is GroupBy[dataset, {#type &, #subtype &}, Map[First]] not suitable?
|
|
| Feb 15, 2016 at 14:25 | history | edited | Kuba | CC BY-SA 3.0 |
edited title
|
| Feb 15, 2016 at 14:09 | answer | added | Leonid Shifrin | timeline score: 9 | |
| Feb 15, 2016 at 13:52 | comment | added | Kuba |
@gwr I didn't, this is quite nice. The "full nested" approach is "better" in that sense that you can still take nested["a"] or nested[[All, "I"]].
|
|
| Feb 15, 2016 at 13:44 | comment | added | gwr |
Did you note GroupBy[ dataset, {#type, #subtype}& ] which leads to a more controlled levelspec (this is also robust for your mimic extended case)? Maybe nested[ {"a", "I" }, "value"] is an option?
|
|
| Feb 15, 2016 at 13:30 | answer | added | Edmund | timeline score: 10 | |
| Feb 15, 2016 at 13:12 | comment | added | Kuba |
@gwr unfortunately, it is not general enough. We may have more levels, you can mimic it with: GroupBy[dataset, {#type &, #subtype &, #type &, #subtype &}], now you would have to adjust levelspec of Apply, and using negative levels may be problematic because the input may be more complex than a list of 1-lvl associations.
|
|
| Feb 15, 2016 at 13:09 | comment | added | gwr |
What about something like GroupBy[ dataset, {#type &, #subtype &}, Apply[ Association, #, 1 ] & ]?
|
|
| Feb 15, 2016 at 12:53 | history | edited | Kuba | CC BY-SA 3.0 |
edited title
|
| Feb 15, 2016 at 12:46 | history | asked | Kuba | CC BY-SA 3.0 |