Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

t/kmeans.t crashes with PDL-2.014 #19

Closed
kmx opened this issue Oct 12, 2015 · 7 comments
Closed

t/kmeans.t crashes with PDL-2.014 #19

kmx opened this issue Oct 12, 2015 · 7 comments

Comments

@kmx
Copy link
Contributor

kmx commented Oct 12, 2015

I have narrowed down the crash to this line: https://github.com/kmx/PDL-Stats/blob/master/Kmeans/kmeans.pd#L702

  $clus_this = $clus_this->mv(-1,2)->clump(2..$clus_this->ndims-1)->dice_axis(2,\@i_best)->reshape( @{ $shapes[1] } )->sever,

Probably during reshape but honestly my knowledge of pdl-stats guts is very low so it is more a guess from me.

@devel-chm
Copy link
Member

Taking a look at kmeans.pd, the problem is that int is being used everywhere instead of PDL_Indx.
It looks like PDL::Stats won't work with PDL_Indx data type as 64bit ints until ported.

@maggiexyz
Copy link
Contributor

Thanks so much for looking into this guys! The earliest I may be able to
take a look will be 2 weeks from now...

Best,
Maggie
On Oct 12, 2015 6:30 PM, "devel-chm" notifications@github.com wrote:

Taking a look at kmeans.pd, the problem is that int is being used
everywhere instead of PDL_Indx.
It looks like PDL::Stats won't work with PDL_Indx data type as 64bit ints
until ported.


Reply to this email directly or view it on GitHub
#19 (comment)
.

@kmx
Copy link
Contributor Author

kmx commented Oct 13, 2015

Chris is right that PDL::Stats needs a revision as for the PDL_indx vs. int (it is IMO worth a separate ticket). However my guess is that the crash reported here is triggered by another issue in PDL core http://sourceforge.net/p/pdl/bugs/403/

So before creating some workarounds in PDL::Stats to avoid this crash I'd wait for some progress on the above-mentioned PDL core bug #403

@devel-chm
Copy link
Member

I concur. Please wait until bug no. 403 is resolved.

--Chris

On Tue, Oct 13, 2015 at 5:09 PM, kmx notifications@github.com wrote:

Chris is right that PDL::Stats needs a revision as for the PDL_indx vs.
int (it is IMO worth a separate ticket). However my guess is that the
crash reported here is triggered by another issue in PDL core
http://sourceforge.net/p/pdl/bugs/403/

So before creating some workarounds in PDL::Stats to avoid this crash I'd
wait for some progress on the above-mentioned PDL core bug #403


Reply to this email directly or view it on GitHub
#19 (comment)
.

maggiexyz added a commit that referenced this issue Oct 20, 2015
workaround for issue #19 (t/kmeans.t crashes with PDL-2.014)
@kmx
Copy link
Contributor Author

kmx commented Nov 27, 2015

Do you think it would be possible to have at least a devel version on CPAN with this fix?

@maggiexyz
Copy link
Contributor

Release 0.73_1 is now on cpan.
On Nov 27, 2015 10:08 AM, "kmx" notifications@github.com wrote:

Do you think it would be possible to have at least a devel version on CPAN
with this fix?


Reply to this email directly or view it on GitHub
#19 (comment)
.

@mohawk2
Copy link
Member

mohawk2 commented Mar 13, 2021

I believe all issues raised here have now been fixed.

@mohawk2 mohawk2 closed this as completed Mar 13, 2021
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

4 participants