On Sat, 26 Nov 2022 at 14:27, Henry Rich wrote:
> This problem, which Elijah has found can be attributed to my insouciance
> about the robustness of CRC-32 as a hash function, afflicts the entire
> i.-family and has no good workaround.
>
> However, I think it will be easy to fix, and I hope we wi
I figured something like this, but my insouciance concerning &. &: and &.:
had prevented a successful implementation (now I think I understand).
Thanks again!
Ben
On Sat, 26 Nov 2022 at 16:21, Elijah Stone wrote:
> For dyad i. use i.&:(1&|."1). Similar applies to other dyadic family
> members.
Great, thanks in advance
Ben
On Sat, 26 Nov 2022 at 14:27, Henry Rich wrote:
> This problem, which Elijah has found can be attributed to my insouciance
> about the robustness of CRC-32 as a hash function, afflicts the entire
> i.-family and has no good workaround.
>
> However, I think it will be
For dyad i. use i.&:(1&|."1). Similar applies to other dyadic family members.
On Sat, 26 Nov 2022, Ben Gorte wrote:
Hi Henry, could you be a bit more explicit about these 32 bits +
workaround?
Eliahs suggestion helps, but doesn't seem to solve the entire issue.
For example, something similar o
This problem, which Elijah has found can be attributed to my insouciance
about the robustness of CRC-32 as a hash function, afflicts the entire
i.-family and has no good workaround.
However, I think it will be easy to fix, and I hope we will get the fix
out in the next beta, which should be co
Hi Henry, could you be a bit more explicit about these 32 bits +
workaround?
Eliahs suggestion helps, but doesn't seem to solve the entire issue.
For example, something similar occurs in dyadic i.
Thanks,
Ben
On Fri, 25 Nov 2022 at 02:57, Henry Rich wrote:
> It's a data dependency:
>
> xxx1
> can duplicate on windows, both of these are fast:
> ~. 1 ,.~ ?. 100 2$100
> ~. 1 ,.~ ?. 100 4$100
That is the same here. It gets slow when the fourth column in the second
case is (almost) a constant
A =: ?100 3$100
6!:2 'echo $~.A,.1234'
632325 4
25.7533
Ben
On Fri, 25 Nov 2
> I cannot duplicate your experience.
> A3=:?1 3$100
That's the same here, it only occurs with larger arrays:
A =: ?200 3$100
6!:2 'echo $~.A'
864712 3
0.080995
6!:2 'echo $~.A,.1'
864712 4
67.7995
Fun fact: In J Playground,
https://jsoftware.github.io/j-playground/bin/html2/#
It's a data dependency:
xxx1 =. 1 ,"1~ 1 3?@$ 100
6!:2'~. xxx1'
0.0068845
xxx1 =. 1 ,"1~ 1 3?@$ 1
6!:2'~. xxx1'
0.000578
Repetitions at lengths of 32 bits seem to be trouble (not surprising
when you think about it). One kludgy-smelling workaround would be to
rotate
can duplicate on windows, both of these are fast:
~. 1 ,.~ ?. 100 2$100
~. 1 ,.~ ?. 100 4$100
On Thursday, November 24, 2022 at 12:40:33 a.m. EST, Elijah Stone
wrote:
That looks rather serious. It seems some problem with the hashing function is
causing an unreasonably high
I cannot duplicate your experience.
A3=:?1 3$100
A31=: A3,.1
A4=:?1 4$100
6!:2'echo $~.A3'
9951 3
0.007586
6!:2'echo $~.A31'
9951 4
0.008745
6!:2'echo $~.A34'
2.8e_5
6!:2'echo $~.A4'
1 4
0.009025
JVERSION
Engine: j903/j64/iOS
beta-k: GPL3/2021-10-15T01:07:05
Pla
Yeah, great workaround, thanks!
Ben
On Thu, 24 Nov 2022 at 16:40, Elijah Stone wrote:
> That looks rather serious. It seems some problem with the hashing
> function is
> causing an unreasonably high rate of collisions. As a temporary
> workaround,
> you can try using ~.&.:(1&|."1), or use the
That looks rather serious. It seems some problem with the hashing function is
causing an unreasonably high rate of collisions. As a temporary workaround,
you can try using ~.&.:(1&|."1), or use the 32-bit version.
On Thu, 24 Nov 2022, Ben Gorte wrote:
G'day,
Still J-ing along, I believe I
13 matches
Mail list logo
