Add more functions implemented with [map]reduce and [map]reducedim #263

wsshin · 2017-07-31T18:10:43Z

This PR adds more functions implemented with [map]reduce and [map]reducedim to src/mapreduce.jl:

iszero
directional versions of sum, prod, all, any
mapped version of prod, count, all, any, minimum, maximum
Note that Base already has these functions.

Additionally, this PR corrects the following typos:

reducedim(op, a::StaticArray, ::Val{D}) to reducedim(op, a::StaticArray, ::Type{Val{D}})
reducedim(op, a::StaticArray, ::Val{D}, v0) to reducedim(op, a::StaticArray, ::Type{Val{D}}, v0)

coveralls · 2017-07-31T18:26:32Z

Coverage decreased (-0.3%) to 88.885% when pulling 8bd82c8 on wsshin:iszero-and-count into da1a371 on JuliaArrays:master.

codecov-io · 2017-07-31T18:29:05Z

Codecov Report

Merging #263 into master will increase coverage by 1.15%.
The diff coverage is 100%.

@@            Coverage Diff             @@
##           master     #263      +/-   ##
==========================================
+ Coverage   89.14%   90.29%   +1.15%     
==========================================
  Files          36       36              
  Lines        2625     2658      +33     
==========================================
+ Hits         2340     2400      +60     
+ Misses        285      258      -27

Impacted Files	Coverage Δ
src/StaticArrays.jl	`100% <ø> (ø)`	⬆️
src/mapreduce.jl	`100% <100%> (ø)`	⬆️
src/traits.jl	`83.78% <0%> (-2.71%)`	⬇️
src/eigen.jl	`91.32% <0%> (-2.29%)`	⬇️
src/det.jl	`100% <0%> (ø)`	⬆️
src/util.jl	`71.05% <0%> (+0.08%)`	⬆️
src/linalg.jl	`97.38% <0%> (+0.11%)`	⬆️
src/MArray.jl	`95.65% <0%> (+0.62%)`	⬆️
src/SArray.jl	`95.23% <0%> (+0.68%)`	⬆️
src/MMatrix.jl	`93.38% <0%> (+0.82%)`	⬆️
... and 7 more

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update da1a371...5172c39. Read the comment docs.

coveralls · 2017-08-01T06:25:22Z

Coverage decreased (-0.03%) to 89.116% when pulling 359e347 on wsshin:iszero-and-count into da1a371 on JuliaArrays:master.

tkoolen · 2017-08-03T02:15:19Z

Just browsing, but FYI for the Val change: the Base convention that methods should take ::Type{Val{T}} is going to be changed to ::Val{T} in 1.0 (see latest docs vs. 0.6 docs and JuliaLang/julia#22475.

wsshin · 2017-08-03T02:53:12Z

Didn't know about the discussion. Good to know. Thanks!

andyferris · 2017-08-06T06:06:49Z

src/mapreduce.jl

@@ -160,25 +160,76 @@ end
 ## reducedim ##
 ###############

-@inline reducedim(op, a::StaticArray, ::Val{D}) where {D} = mapreducedim(identity, op, a, Val{D})
-@inline reducedim(op, a::StaticArray, ::Val{D}, v0) where {D} = mapreducedim(identity, op, a, Val{D}, v0)
+@inline reducedim(op, a::StaticArray, ::Type{Val{D}}) where {D} = mapreducedim(identity, op, a, Val{D})


FYI: This possibly wasn't a typo. I had been thinking about this PR for a long time :)

However, we should probably add a if VERSION < v"0.7" to support both versions of Julia, or something (or simply support both syntaxes?)

But having to use Val(D) in reducedim and Val{D} in mapreducedim seems confusing and inconsistent. I don't see the use of Val(D) and ::Val{D} anywhere else in StaticArrays, either. Wouldn't it be better to handle this in a separate PR collectively?

OK, so what I would like is if StaticArrays used Val{D} everywhere in v0.6 and Val(D) everywhere in v0.7. (I suppose it's OK if the v0.6 version also supports Val{D}(), partly to avoid this being a breaking change)

If you were to implement that here, that would be awesome (as it seems pertinent to the change already done on this line) - however, let me know if you'd prefer to not do that at the moment, and I'll get to it later.

When making this change, I would like to have things like

default_similar_type{T,S,D}(::Type{T}, s::Size{S}, ::Type{Val{D}})

in abstractarray.jl change to

default_similar_type{T,S,D}(::Type{T}, s::Size{S}, ::Val{D})

as well. Because such a change is outside mapreduce.jl, I think creating a separate PR dedicated to ::Type{Val{D}} -> ::Val{D} would be cleaner... So yes, I would prefer not to make this change at the moment.

Fair enough

andyferris · 2017-08-06T06:21:36Z

src/mapreduce.jl

+# with an initial value v0 = true and false.
+#
+# 4. Some implementations (e.g., count(a, Val{D})) are commented out because corresponding
+# Base functions (e.g., count(a, D)) do not exist yet.


Yet? Meaning they are mooted for v0.7?

(We are allowed to be a bit more agile and move a bit beyond Base wherever it makes sense, so long as we don't break anything that tends to work for AbstractArray)

5172c39 uncomments these functions!

andyferris · 2017-08-06T06:24:38Z

Basically looks good to me. Thanks @wsshin.

I guess we may as well try and support the syntax for v0.6 and v1.0? (Ideally, I'd like to support multiple versions of Julia - in the past we've been hampered by internal compiler changes meaning different approaches are valid/optimal, but so far I am not really expecting this for v0.7/v1.0).

wsshin · 2017-08-06T12:50:48Z

Wait. While trying to uncomment count(a, Val{D}), I found some bugs. I will report it back later.

andyferris · 2017-08-06T12:56:15Z

OK

wsshin · 2017-08-07T04:34:18Z

The bug I was referring to was ignorance of the possibility of element type change in mapreducedim. Because of this, count(f::Function, a::StaticArray, ::Type{Val{D}}) was returning an array of Float64, rather than Int64, when eltype(a) was Float64. Now properly promoted element types are used in mapreducedim and its friends.

One question, though. I used Base.promote_op to deduce the proper element types after mapping and operation, but in _map I find that Core.Inference.return_type is used for the same purpose. I find that Base.promote_op wraps around Core.Inference.return_type (https://github.com/JuliaLang/julia/blob/master/base/promotion.jl#L321), so I guessed that promote_op takes care of some difficult situations better than return_type, but I'm not sure which one of the two is better. Any suggestions?

coveralls · 2017-08-07T04:40:17Z

Coverage increased (+1.2%) to 90.293% when pulling 5172c39 on wsshin:iszero-and-count into da1a371 on JuliaArrays:master.

andyferris · 2017-08-08T03:19:46Z

Feel free to use either. I think there are historic reasons surrounding this... something about one being inferable and the other not (but these days they should both be fine).

BTW, I'd be most happy with the idea of just letting the natural types fall out, rather than using inference. At the moment we can't do this everywhere since we'd need a "eltypeless" constructor for every StaticArray, but unlike Base.Array, we can construct the result before wrapping it in an array so technically we don't need eltype prediction (in such a system, inference will still work to make everything fast, but only in the background).

wsshin · 2017-08-08T03:27:22Z

I think I know what you mean by "letting the natural types fall out", but just in case, could you point me to a couple of examples of this practice in StaticArrays, if any?

andyferris · 2017-08-08T03:33:07Z

Constructors are one case, things get promoted to a common type and an output is generated.

I think complete reductions also might do this?

wsshin · 2017-08-08T03:47:33Z

Hmmm... Not sure if I understood what you meant. Could you be more specific about how to apply your idea in _mapreducedim?

andyferris · 2017-08-08T05:23:38Z

I don't think you can implement this idea (at this stage) for partial reductions because the infrastructure isn't available. I was just expressing an idea of where I'd like to get to.

wsshin · 2017-08-08T05:24:49Z

I see. Then can we merge this now?

andyferris · 2017-08-08T05:29:30Z

Done!

Use StaticArrays' reduce and mapreduce for iszero and count(f, a)

8bd82c8

Add more functions implemented with [map]reduce and [map]reducedim

359e347

wsshin changed the title ~~Use StaticArrays' reduce and mapreduce for iszero and count(f, a)~~ Add more functions implemented with [map]reduce and [map]reducedim Aug 1, 2017

andyferris reviewed Aug 6, 2017

View reviewed changes

andyferris approved these changes Aug 6, 2017

View reviewed changes

Handle possibility of eltype change in mapreducedim and diff

5172c39

andyferris merged commit 7d711ab into JuliaArrays:master Aug 8, 2017

wsshin deleted the iszero-and-count branch August 8, 2017 06:47

wsshin mentioned this pull request Feb 7, 2018

Fix several broadcast issues (fixes #197, #199, #200, #242) #274

Merged

pablosanjose mentioned this pull request May 1, 2020

iszero(::SVector{0}) broken #778

Closed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Add more functions implemented with [map]reduce and [map]reducedim #263

Add more functions implemented with [map]reduce and [map]reducedim #263

wsshin commented Jul 31, 2017 •

edited

Loading

coveralls commented Jul 31, 2017

codecov-io commented Jul 31, 2017 •

edited

Loading

coveralls commented Aug 1, 2017

tkoolen commented Aug 3, 2017

wsshin commented Aug 3, 2017

andyferris Aug 6, 2017 •

edited

Loading

wsshin Aug 6, 2017

andyferris Aug 6, 2017 •

edited

Loading

wsshin Aug 6, 2017

andyferris Aug 6, 2017

andyferris Aug 6, 2017

wsshin Aug 8, 2017

andyferris commented Aug 6, 2017

wsshin commented Aug 6, 2017 •

edited

Loading

andyferris commented Aug 6, 2017

wsshin commented Aug 7, 2017 •

edited

Loading

coveralls commented Aug 7, 2017

andyferris commented Aug 8, 2017

wsshin commented Aug 8, 2017

andyferris commented Aug 8, 2017

wsshin commented Aug 8, 2017

andyferris commented Aug 8, 2017

wsshin commented Aug 8, 2017

andyferris commented Aug 8, 2017

Add more functions implemented with [map]reduce and [map]reducedim #263

Add more functions implemented with [map]reduce and [map]reducedim #263

Conversation

wsshin commented Jul 31, 2017 • edited Loading

coveralls commented Jul 31, 2017

codecov-io commented Jul 31, 2017 • edited Loading

Codecov Report

coveralls commented Aug 1, 2017

tkoolen commented Aug 3, 2017

wsshin commented Aug 3, 2017

andyferris Aug 6, 2017 • edited Loading

Choose a reason for hiding this comment

wsshin Aug 6, 2017

Choose a reason for hiding this comment

andyferris Aug 6, 2017 • edited Loading

Choose a reason for hiding this comment

wsshin Aug 6, 2017

Choose a reason for hiding this comment

andyferris Aug 6, 2017

Choose a reason for hiding this comment

andyferris Aug 6, 2017

Choose a reason for hiding this comment

wsshin Aug 8, 2017

Choose a reason for hiding this comment

andyferris commented Aug 6, 2017

wsshin commented Aug 6, 2017 • edited Loading

andyferris commented Aug 6, 2017

wsshin commented Aug 7, 2017 • edited Loading

coveralls commented Aug 7, 2017

andyferris commented Aug 8, 2017

wsshin commented Aug 8, 2017

andyferris commented Aug 8, 2017

wsshin commented Aug 8, 2017

andyferris commented Aug 8, 2017

wsshin commented Aug 8, 2017

andyferris commented Aug 8, 2017

wsshin commented Jul 31, 2017 •

edited

Loading

codecov-io commented Jul 31, 2017 •

edited

Loading

andyferris Aug 6, 2017 •

edited

Loading

andyferris Aug 6, 2017 •

edited

Loading

wsshin commented Aug 6, 2017 •

edited

Loading

wsshin commented Aug 7, 2017 •

edited

Loading