I think the decision to not require a specific value, is it allows mutating the only the nullability of an element without touching the value which in some contexts can be useful.
As an implementation specific details this might be something that can be added to metadata and propagated accordingly within the C++ implementation/compute kernels. On Tue, Apr 7, 2020 at 3:04 AM Felix Benning <felix.benn...@gmail.com> wrote: > I guess it would be helpful, when trying to achieve zero-modification > between R and another language, if the standard used for communication > would allow for that. Or when setting all nulls to zero for an algorithm > and then saving it to a database for later use. But at the same time, I > only know about this project since a couple of days ago, so my opinion > on this is likely uneducated at best. I am mostly curious about how this > pans out and what the trade offs are, since I went down this rabbit hole > of "how to handle nulls" this far, I might as well bottom out ;-). > > - Felix > > On 06.04.20 22:26, Wes McKinney wrote: > > For the sake of others reading, this discussion might be a bit > > confusing to happen upon because the scope isn't clear. It seems that > > we are discussing the C++ implementation and not the columnar format, > > is that right? > > > > Adding any additional metadata about this to the columnar format / > > Flatbuffers files / C interface is probably a non-starter. We've > > discussed the contents of data "underneath" a null and consistently > > the consensus is that is is unspecified. > > > > Applications (as well as internal details of some implementations and > > their interactions with external libraries) are free to set > > custom_metadata fields in schemas to indicate otherwise. However, one > > must take care to not propagate this metadata inappropriately from one > > realization of a schema (as an Array or RecordBatch) where it is true > > to another where it is not true. Similarly, one should also be careful > > not to use such metadata on data whose provenance is unknown. > > > > - Wes > > > > On Mon, Apr 6, 2020 at 11:37 AM Felix Benning <felix.benn...@gmail.com> > wrote: > >> In that case it is probably necessary to have a "has_sentinel" flag and > a > >> "sentinel_value" variable. Since other algorithms might benefit from not > >> having to set these values to zero. Which is probably the reason why the > >> value "underneath" was set to unspecified in the first place. > Alternatively > >> a "sentinel_enum" could specify whether the sentinel is 0, or the R > >> sentinel value is used. This would sacrifice flexibility for size. > Although > >> size probably does not matter, when meta data for entire columns are > >> concerned. So the first approach is probably better. > >> > >> Felix > >> > >> On Mon, 6 Apr 2020 at 17:59, Francois Saint-Jacques < > fsaintjacq...@gmail.com> > >> wrote: > >> > >>> It does make sense, I would go a little further and make this > >>> field/property a single value of the same type than the array. This > >>> would allow using any arbitrary sentinel value for unknown values (0 > >>> in your suggested case). The end result is zero-copy for R bindings > >>> (if stars are aligned). I created ARROW-8348 [1] for this. > >>> > >>> François > >>> > >>> [1] https://jira.apache.org/jira/browse/ARROW-8348 > >>> > >>> On Mon, Apr 6, 2020 at 11:02 AM Felix Benning <felix.benn...@gmail.com > > > >>> wrote: > >>>> Would it make sense to have an `na_are_zero` flag? Since null > checking is > >>>> not without cost, it might be helpful to some algorithms, if the > content > >>>> "underneath" the nulls is zero. For example in means, or scalar > products > >>>> and thus matrix multiplication, knowing that the array has zeros where > >>> the > >>>> na's are, would allow these algorithms to pretend that there are no > na's. > >>>> Since setting all nulls to zero in a matrix of n columns and n rows > costs > >>>> O(n^2), it would make sense to set them all to zero before matrix > >>>> multiplication i.e. O(n^3) and similarly expensive algorithms. If > there > >>> was > >>>> a `na_are_zero` flag, other algorithms could later utilize this work > >>>> already being done. Algorithms which change the data and violate this > >>>> contract, would only need to reset the flag. And in some use cases, it > >>>> might be possible to use idle time of the computer to "clean up" the > >>> na's, > >>>> preparing for the next query. > >>>> > >>>> Felix > >>>> > >>>> ---------- Forwarded message --------- > >>>> From: Wes McKinney <wesmck...@gmail.com> > >>>> Date: Sun, 5 Apr 2020 at 22:31 > >>>> Subject: Re: Attn: Wes, Re: Masked Arrays > >>>> To: <u...@arrow.apache.org> > >>>> > >>>> > >>>> As I recall the contents "underneath" have been discussed before and > >>>> the consensus was that the contents are not specified. If you'e like > >>>> to make a proposal to change something I would suggest raising it on > >>>> dev@arrow.apache.org > >>>> > >>>> On Sun, Apr 5, 2020 at 1:56 PM Felix Benning <felix.benn...@gmail.com > > > >>>> wrote: > >>>>> Follow up: Do you think it would make sense to have an `na_are_zero` > >>>> flag? Since it appears that the baseline (naively assuming there are > no > >>>> null values) is still a bit faster than equally optimized null value > >>>> handling algorithms. So you might want to make the assumption, that > all > >>>> null values are set to zero in the array (instead of undefined). This > >>> would > >>>> allow for very fast means, scalar products and thus matrix > multiplication > >>>> which ignore nas. And in case of matrix multiplication, you might > prefer > >>>> sacrificing an O(n^2) effort to set all null entries to zero before > >>>> multiplying. And assuming you do not overwrite this data, you would be > >>> able > >>>> to reuse that assumption in later computations with such a flag. > >>>>> In some use cases, you might even be able to utilize unused computing > >>>> resources for this task. I.e. clean up the nulls while the computer is > >>> not > >>>> used, preparing for the next query. > >>>>> > >>>>> On Sun, 5 Apr 2020 at 18:34, Felix Benning <felix.benn...@gmail.com> > >>>> wrote: > >>>>>> Awesome, that was exactly what I was looking for, thank you! > >>>>>> > >>>>>> On Sun, 5 Apr 2020 at 00:40, Wes McKinney <wesmck...@gmail.com> > >>> wrote: > >>>>>>> I wrote a blog post a couple of years about this > >>>>>>> > >>>>>>> https://wesmckinney.com/blog/bitmaps-vs-sentinel-values/ > >>>>>>> > >>>>>>> Pasha Stetsenko did a follow-up analysis that showed that my > >>>>>>> "sentinel" code could be significantly improved, see: > >>>>>>> > >>>>>>> https://github.com/st-pasha/microbench-nas/blob/master/README.md > >>>>>>> > >>>>>>> Generally speaking in Apache Arrow we've been happy to have a > uniform > >>>>>>> representation of nullness across all types, both primitive > >>> (booleans, > >>>>>>> numbers, or strings) and nested (lists, structs, unions, etc.). > Many > >>>>>>> computational operations (like elementwise functions) need not > >>> concern > >>>>>>> themselves with the nulls at all, for example, since the bitmap > from > >>>>>>> the input array can be passed along (with zero copy even) to the > >>>>>>> output array. > >>>>>>> > >>>>>>> On Sat, Apr 4, 2020 at 4:39 PM Felix Benning < > >>> felix.benn...@gmail.com> > >>>> wrote: > >>>>>>>> Does anyone have an opinion (or links) about Bitpattern vs Masked > >>>> Arrays for NA implementations? There seems to have been a discussion > >>> about > >>>> that in the numpy community in 2012 > >>>> https://numpy.org/neps/nep-0026-missing-data-summary.html without an > >>>> apparent result. > >>>>>>>> Summary of the Summary: > >>>>>>>> - The Bitpattern approach reserves one bitpattern of any type as > >>> na, > >>>> the only type not having spare bitpatterns are integers which means > this > >>>> decreases their range by one. This approach is taken by R and was > >>> regarded > >>>> as more performant in 2012. > >>>>>>>> - The Mask approach was deemed more flexible, since it would allow > >>>> "degrees of missingness", and also cleaner/easier implementation. > >>>>>>>> Since bitpattern checks would probably disrupt SIMD, I feel like > >>> some > >>>> calculations (e.g. mean) would actually benefit more, from setting na > >>>> values to zero, proceeding as if they were not there, and using the > >>> number > >>>> of nas in the metadata to adjust the result. This of course does not > work > >>>> if two columns are used (e.g. scalar product), which is probably more > >>>> important. > >>>>>>>> Was using Bitmasks in Arrow a conscious performance decision? Or > >>> was > >>>> the decision only based on the fact, that R and Bitpattern > >>> implementations > >>>> in general are a niche, which means that Bitmasks are more compatible > >>> with > >>>> other languages? > >>>>>>>> I am curious about this topic, since the "lack of proper na > >>> support" > >>>> was cited as the reason, why Python would never replace R in > statistics. > >>>>>>>> Thanks, > >>>>>>>> > >>>>>>>> Felix > >>>>>>>> > >>>>>>>> > >>>>>>>> On 31.03.20 14:52, Joris Van den Bossche wrote: > >>>>>>>> > >>>>>>>> Note that pandas is starting to use a notion of "masked arrays" as > >>>> well, for example for its nullable integer data type, but also not > using > >>>> the np.ma masked array, but a custom implementation (for technical > >>> reasons > >>>> in pandas this was easier). > >>>>>>>> Also, there has been quite some discussion last year in numpy > >>> about a > >>>> possible re-implementation of a MaskedArray, but using numpy's > protocols > >>>> (`__array_ufunc__`, `__array_function__` etc), instead of being a > >>> subclass > >>>> like np.ma now is. See eg > >>>> > https://mail.python.org/pipermail/numpy-discussion/2019-June/079681.html > >>> . > >>>>>>>> Joris > >>>>>>>> > >>>>>>>> On Mon, 30 Mar 2020 at 18:57, Daniel Nugent <nug...@gmail.com> > >>> wrote: > >>>>>>>>> Ok. That actually aligns closely to what I'm familiar with. Good > >>> to > >>>> know. > >>>>>>>>> Thanks again for taking the time to respond, > >>>>>>>>> > >>>>>>>>> -Dan Nugent > >>>>>>>>> > >>>>>>>>> > >>>>>>>>> On Mon, Mar 30, 2020 at 12:38 PM Wes McKinney < > >>> wesmck...@gmail.com> > >>>> wrote: > >>>>>>>>>> Social and technical reasons I guess. Empirically it's just not > >>>> used much. > >>>>>>>>>> You can see my comments about numpy.ma in my 2010 paper about > >>> pandas > >>>>>>>>>> > >>> https://conference.scipy.org/proceedings/scipy2010/pdfs/mckinney.pdf > >>>>>>>>>> At least in 2010, there were notable performance problems when > >>> using > >>>>>>>>>> MaskedArray for computations > >>>>>>>>>> > >>>>>>>>>> "We chose to use NaN as opposed to using NumPy MaskedArrays for > >>>>>>>>>> performance reasons (which are beyond the scope of this paper), > >>> as > >>>> NaN > >>>>>>>>>> propagates in floating-point operations in a natural way and can > >>> be > >>>>>>>>>> easily detected in algorithms." > >>>>>>>>>> > >>>>>>>>>> On Mon, Mar 30, 2020 at 11:20 AM Daniel Nugent < > nug...@gmail.com > >>>> wrote: > >>>>>>>>>>> Thanks! Since I'm just using it to jump to Arrow, I think I'll > >>>> stick with it. > >>>>>>>>>>> Do you have any feelings about why Numpy's masked arrays didn't > >>>> gain favor when many data representation formats explicitly support > >>> nullity > >>>> (including Arrow)? Is it just that not carrying nulls in computations > >>>> forward is preferable (that is, early filtering/value filling was > >>> easier)? > >>>>>>>>>>> -Dan Nugent > >>>>>>>>>>> > >>>>>>>>>>> > >>>>>>>>>>> On Mon, Mar 30, 2020 at 11:40 AM Wes McKinney < > >>> wesmck...@gmail.com> > >>>> wrote: > >>>>>>>>>>>> On Mon, Mar 30, 2020 at 8:31 AM Daniel Nugent < > >>> nug...@gmail.com> > >>>> wrote: > >>>>>>>>>>>>> Didn’t want to follow up on this on the Jira issue earlier > >>>> since it's sort of tangential to that bug and more of a usage > question. > >>> You > >>>> said: > >>>>>>>>>>>>>> I wouldn't recommend building applications based on them > >>>> nowadays since the level of support / compatibility in other projects > is > >>>> low. > >>>>>>>>>>>>> In my case, I am using them since it seemed like a > >>>> straightforward representation of my data that has nulls, the format > I’m > >>>> converting from has zero cost numpy representations, and converting > from > >>> an > >>>> internal format into Arrow in memory structures appears zero cost (or > >>> close > >>>> to it) as well. I guess I can just provide the mask as an explicit > >>>> argument, but my original desire to use it came from being able to > >>> exploit > >>>> numpy.ma.concatenate in a way that saved some complexity in > >>> implementation. > >>>>>>>>>>>>> Since Arrow itself supports masking values with a bitfield, > >>> is > >>>> there something intrinsic to the notion of array masks that is not > well > >>>> supported? Or do you just mean the specific numpy MaskedArray class? > >>>>>>>>>>>> I mean just the numpy.ma module. Not many Python computing > >>>> projects > >>>>>>>>>>>> nowadays treat MaskedArray objects as first class citizens. > >>>> Depending > >>>>>>>>>>>> on what you need it may or may not be a problem. pyarrow > >>> supports > >>>>>>>>>>>> ingesting from MaskedArray as a convenience, but it would not > >>> be > >>>>>>>>>>>> common in my experience for a library's APIs to return > >>>> MaskedArrays. > >>>>>>>>>>>>> If this is too much of a numpy question rather than an arrow > >>>> question, could you point me to where I can read up on masked array > >>> support > >>>> or maybe what the right place to ask the numpy community about whether > >>> what > >>>> I'm doing is appropriate or not. > >>>>>>>>>>>>> Thanks, > >>>>>>>>>>>>> > >>>>>>>>>>>>> > >>>>>>>>>>>>> -Dan Nugent >
