Special code seems to be involved because usually replacing 2&| by 2 | ]
should not make much difference in performance but it makes a difference
in this case:
Cut=. -.&a: @: (<;._2@,~)
st=. (, */&.:>@:(1 2&{))@:(] ; 7!:2@:] ; 6!:2)
stp2=. [ ((;:'Sentence Space Time Product') , st&>) LF Cut ]
stp=. 1&stp2 : stp2 f.
I=. i.100000
11 stp noun define
+:^:(2&| )"0 I
+:^:(2 | ])"0 I
]`+:@.(2&| )"0 I
]`+:@.(2 | ])"0 I
)
┌──────────────────────┬───────┬──────────┬───────┐
│Sentence │Space │Time │Product│
├──────────────────────┼───────┼──────────┼───────┤
│ +:^:(2&| )"0 I│1052160│0.046733 │49170.6│
├──────────────────────┼───────┼──────────┼───────┤
│ +:^:(2 | ])"0 I│1052160│0.0580962 │61126.5│
├──────────────────────┼───────┼──────────┼───────┤
│ ]`+:@.(2&| )"0 I│6032768│0.00553589│33396.8│
├──────────────────────┼───────┼──────────┼───────┤
│ ]`+:@.(2 | ])"0 I│6033280│0.0204798 │123560 │
└──────────────────────┴───────┴──────────┴───────┘
Notice also,
11 stp noun define
(2&| )"0 I
(2 | ])"0 I
)
┌──────────────────────┬──────┬───────────┬───────┐
│Sentence │Space │Time │Product│
├──────────────────────┼──────┼───────────┼───────┤
│ (2&| )"0 I│133504│0.000196446│26.2264│
├──────────────────────┼──────┼───────────┼───────┤
│ (2 | ])"0 I│133888│0.0142925 │1913.59│
└──────────────────────┴──────┴───────────┴───────┘
versus
11 stp noun define
(2&| ) I
(2 | ]) I
)
┌──────────────────────┬──────┬───────────┬───────┐
│Sentence │Space │Time │Product│
├──────────────────────┼──────┼───────────┼───────┤
│ (2&| ) I│132864│0.000195543│25.9807│
├──────────────────────┼──────┼───────────┼───────┤
│ (2 | ]) I│132864│0.000195243│25.9407│
└──────────────────────┴──────┴───────────┴───────┘
The following is a summary of the performance of the solutions suggested so
far including the 2 | ] variants:
11 stp noun define
+:^:(2&| )"0 I NB. Pascal
+:^:(2 | ])"0 I NB. Pascal
]`+:@.(2&| )"0 I NB. Pascal
]`+:@.(2 | ])"0 I NB. Pascal
(* 1 + 2&| ) I NB. Dan
(* 1 + 2 | ]) I NB. Dan
(32 b.~ 1&(17 b.)) I NB. Henry
)
┌────────────────────────────────────┬───────┬──────────┬───────┐
│Sentence │Space │Time │Product│
├────────────────────────────────────┼───────┼──────────┼───────┤
│ +:^:(2&| )"0 I NB. Pascal │1052672│0.0490038 │51585 │
├────────────────────────────────────┼───────┼──────────┼───────┤
│ +:^:(2 | ])"0 I NB. Pascal │1052672│0.055311 │58224.4│
├────────────────────────────────────┼───────┼──────────┼───────┤
│ ]`+:@.(2&| )"0 I NB. Pascal │6033280│0.00488106│29448.8│
├────────────────────────────────────┼───────┼──────────┼───────┤
│ ]`+:@.(2 | ])"0 I NB. Pascal │6033792│0.0199239 │120216 │
├────────────────────────────────────┼───────┼──────────┼───────┤
│ (* 1 + 2&| ) I NB. Dan │2100096│0.00116601│2448.73│
├────────────────────────────────────┼───────┼──────────┼───────┤
│ (* 1 + 2 | ]) I NB. Dan │2100096│0.00116002│2436.15│
├────────────────────────────────────┼───────┼──────────┼───────┤
│(32 b.~ 1&(17 b.)) I NB. Henry │2100992│0.00107981│2268.67│
└────────────────────────────────────┴───────┴──────────┴───────┘
(Notice also the extra space used due to the NB.)
This is a clean version of the script:
Cut=. -.&a: @: (<;._2@,~)
st=. (, */&.:>@:(1 2&{))@:(] ; 7!:2@:] ; 6!:2)
stp2=. [ ((;:'Sentence Space Time Product') , st&>) LF Cut ]
stp=. 1&stp2 : stp2 f.
I=. i.100000
11 stp noun define
+:^:(2&| )"0 I
+:^:(2 | ])"0 I
]`+:@.(2&| )"0 I
]`+:@.(2 | ])"0 I
)
11 stp noun define
(2&| )"0 I
(2 | ])"0 I
)
11 stp noun define
(2&| ) I
(2 | ]) I
)
11 stp noun define
+:^:(2&| )"0 I NB. Pascal
+:^:(2 | ])"0 I NB. Pascal
]`+:@.(2&| )"0 I NB. Pascal
]`+:@.(2 | ])"0 I NB. Pascal
(* 1 + 2&| ) I NB. Dan
(* 1 + 2 | ]) I NB. Dan
(32 b.~ 1&(17 b.)) I NB. Henry
)
On Mon, Dec 28, 2015 at 1:48 PM, 'Pascal Jasmin' via Programming <
[email protected]> wrote:
> timespacex '+:^:(2&|)"0 i.100000'
> 0.0292787 2.10074e6
> timespacex ']`+:@.(2&|)"0 i.100000'
> 0.00561727 7.08134e6
>
> above code doubles odd numbers, and returns even numbers unchanged.
>
>
> I've previously noticed that ^: seemd faster than ]`u@. ... but I guess
> not?
> ----------------------------------------------------------------------
> For information about J forums see http://www.jsoftware.com/forums.htm
>
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm
