⊃⍋ and ⊃⍒
12 posts
• Page 2 of 2 • 1, 2
Re: ⊃⍋ and ⊃⍒
I fully understand your skepticism, do not believe me until you see it.
The problem you mention was an irritation but it *was* possible to work around it with error trapping, so it never became a high priority. Today, we see there is a need to reliably retrieve the full source of any function, and there is currently no way to do this in all scenarios. The intention is that the solution to this will also be designed to solve your original problem.
The problem you mention was an irritation but it *was* possible to work around it with error trapping, so it never became a high priority. Today, we see there is a need to reliably retrieve the full source of any function, and there is currently no way to do this in all scenarios. The intention is that the solution to this will also be designed to solve your original problem.
-
Morten|Dyalog - Posts: 453
- Joined: Tue Sep 09, 2008 3:52 pm
Re: ⊃⍋ and ⊃⍒
Regarding the original topic, I have implemented special code in 18.0 for disclose-atop-reduce, in either 2-train or Atop operator form. My timings:
- Code: Select all
cmpx'⊃⍋a' 'a⍳⌊/a' '⊃⍤⍋a'⊣a←?1e6/0
⊃⍋a → 2.7E¯1 | 0% ⎕⎕⎕⎕⎕⎕⎕⎕⎕⎕⎕⎕⎕⎕⎕⎕⎕⎕⎕⎕⎕⎕⎕⎕⎕⎕⎕⎕⎕⎕⎕⎕⎕⎕⎕⎕⎕⎕⎕⎕
a⍳⌊/a → 1.2E¯3 | -100%
⊃⍤⍋a → 6.3E¯4 | -100%
cmpx'⊃⍋a' 'a⍳⌊/a' '⊃⍤⍋a'⊣a←1e6?1e7
⊃⍋a → 2.8E¯2 | 0% ⎕⎕⎕⎕⎕⎕⎕⎕⎕⎕⎕⎕⎕⎕⎕⎕⎕⎕⎕⎕⎕⎕⎕⎕⎕⎕⎕⎕⎕⎕⎕⎕⎕⎕⎕⎕⎕⎕⎕⎕
a⍳⌊/a → 3.6E¯4 | -99% ⎕
⊃⍤⍋a → 2.0E¯4 | -100%
- Marshall|Dyalog
12 posts
• Page 2 of 2 • 1, 2
Who is online
Users browsing this forum: No registered users and 1 guest
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group