Yes.. I should have been reading the updated code, rather than the commits.

It's been too long since I did anything serious with github.

Sorry about the noise.

Thanks,

-- 
Raul

On Wed, May 19, 2021 at 6:12 AM Ric Sherlock <[email protected]> wrote:
>
> Hi Raul,
> The pull request is named "Update histogram to correctly count items in
> last bin" and links to Issue #4 (
> https://github.com/jsoftware/stats_base/issues/4) that was filed in the
> repository called "histogram doesn't count all the data". That mentions the
> problems to be addressed and also links to the forum thread where the issue
> was raised and discussed. Not sure that we need to revisit all that again
> in the commit message?
>
> I think the comment you're referring to is one I added to the original
> version of histogram which I renamed to histogram0 in the first commit. My
> thinking was that if for whatever reason we decided we needed to keep that
> implementation, then we should at least note that it undercounts the final
> bin. You'll notice in the 2nd commit of the pull request I remove that verb
> along with the comment.
>
> Hope that makes sense?
>
> On Wed, May 19, 2021 at 9:29 PM Raul Miller <[email protected]> wrote:
>
> > Details would be nice in the commit message. "corrected" doesn't
> > really explain the issue. But I guess that's just a suggestion for
> > later.
> >
> > Looking at the implementation itself, I don't remember what situation
> > the 'NB. The number of points in the last interval is undercounted by
> > 1'  is good for. I could look it up, but I think comments should focus
> > on purpose rather than just mention implementation details. Is this
> > something you feel comfortable elaborating on?
> >
> > Thanks,
> >
> > --
> > Raul
> >
> > On Wed, May 19, 2021 at 4:04 AM Ric Sherlock <[email protected]> wrote:
> > >
> > > I've made some changes to stats/base in this Pull request
> > > https://github.com/jsoftware/stats_base/pull/5
> > > Any suggestions/objections?
> > >
> > > Changes are basically:
> > >  * the current version of histogram is replaced with histogram2 (also
> > > called histogramL)
> > >  * histogram1 is added as histogramR
> > >  * added Idotr
> > >  * added binnedData
> > >
> > >
> > > On Wed, Apr 14, 2021 at 3:59 PM Gilles Kirouac <[email protected]> wrote:
> > >
> > > >
> > > > Will histogram1 and histogram2 be added to stats/base?
> > > >
> > > >
> > > > ~ Gilles
> > > > ----------------------------------------------------------------------
> > > > For information about J forums see http://www.jsoftware.com/forums.htm
> > > >
> > > ----------------------------------------------------------------------
> > > For information about J forums see http://www.jsoftware.com/forums.htm
> > ----------------------------------------------------------------------
> > For information about J forums see http://www.jsoftware.com/forums.htm
> >
> ----------------------------------------------------------------------
> For information about J forums see http://www.jsoftware.com/forums.htm
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

Reply via email to