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.

Sign up for GitHub

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

Run findall(rows) only if rows are not all true #2727

Merged
merged 17 commits into from
May 16, 2021
Merged

Run findall(rows) only if rows are not all true #2727

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
17 commits
Select commit Hold shift + click to select a range
f0a5ea7
Run `findall(rows)` only if `rows` are not .==true
pstorozenko Apr 18, 2021
92f943e
Custom, optimized `_findall(B::BitVector)`
pstorozenko May 9, 2021
577dc85
Test for type added
pstorozenko May 9, 2021
704e2c7
Some more random tests added
pstorozenko May 9, 2021
01ba69c
Merge branch 'main' into ps/sdf_continuous
pstorozenko May 9, 2021
e3ea0be
Added empty test and some tweeks to random tests
pstorozenko May 9, 2021
5c60e2c
B.chunks is always `Vector{UInt64}` even on x86
pstorozenko May 10, 2021
422eb0b
Ret Vector{Int32} on x86 and Vector{Int64} on x64
pstorozenko May 10, 2021
6c91fa7
Update src/other/utils.jl
pstorozenko May 10, 2021
2fbcfc5
Apply suggestions from code review, add tests
pstorozenko May 10, 2021
2aec918
Trim whitespaces and wait with `I`
pstorozenko May 10, 2021
ae25f99
Adding assert before return
pstorozenko May 11, 2021
7af8058
Minor tests fix
pstorozenko May 11, 2021
196d5d4
Whitespaces
pstorozenko May 11, 2021
0ed431c
B.chunks is always of UInt64
pstorozenko May 11, 2021
3d3bffe
Apply suggestions from code review
pstorozenko May 14, 2021
37114a7
`~UInt64(0)` put in all places
pstorozenko May 14, 2021
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
8 changes: 8 additions & 0 deletions src/other/utils.jl
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -168,3 +168,11 @@ function _nt_like_hash(v, h::UInt)

return xor(objectid(Tuple(propertynames(v))), h)
end

function _findall(v::AbstractVector{Bool})
if all(v)
return keys(v)
else
return findall(v)
end
end
2 changes: 1 addition & 1 deletion src/subdataframe/subdataframe.jl
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -96,7 +96,7 @@ Base.@propagate_inbounds function SubDataFrame(parent::DataFrame, rows::Abstract
throw(ArgumentError("invalid length of `AbstractVector{Bool}` row index " *
"(got $(length(rows)), expected $(nrow(parent)))"))
end
return SubDataFrame(parent, findall(rows), cols)
return SubDataFrame(parent, _findall(rows), cols)
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Can you also test performance and corrected of view (it should be OK, but just to be sure as this is the point of the change we make so I prefer to double check). Thank you!

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Sure, I'll check it with some more scenario than in the beginning of this PR

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Fortunately it is faster :)

df = DataFrame(
    a = rand(1000000),
    b = allowmissing(rand(1000000)),
    c = [allowmissing(rand(1000000 - 1)); missing],
    d = repeat([1:9; missing], 100000),
    e = repeat([rand(50000); missings(50000)], 10),
    f = [missings(300000); ones(400000); missings(300000)],
    g = missings(1000000)
)

Main:

julia> @btime dropmissing(df, :a, view=true);
  1.872 ms (9 allocations: 7.75 MiB)

julia> @btime dropmissing(df, :b, view=true);
  2.075 ms (14 allocations: 7.75 MiB)

julia> @btime dropmissing(df, :c, view=true);
  2.111 ms (14 allocations: 7.75 MiB)

julia> @btime dropmissing(df, :d, view=true);
  1.913 ms (14 allocations: 6.99 MiB)

julia> @btime dropmissing(df, :e, view=true);
  1.117 ms (14 allocations: 3.94 MiB)

julia> @btime dropmissing(df, :f, view=true);
  909.713 μs (14 allocations: 3.18 MiB)
 
julia> @btime dropmissing(df, :g, view=true);
  152.216 μs (13 allocations: 126.67 KiB)

julia> @btime dropmissing(df, view=true);
  1.432 ms (38 allocations: 1.63 MiB)

Main + this PR:

julia> @btime dropmissing(df, :a, view=true);
  17.093 μs (7 allocations: 122.34 KiB)

julia> @btime dropmissing(df, :b, view=true);
  162.656 μs (12 allocations: 126.59 KiB)

julia> @btime dropmissing(df, :c, view=true);
  168.988 μs (12 allocations: 126.59 KiB)

julia> @btime dropmissing(df, :d, view=true);
  1.890 ms (14 allocations: 6.99 MiB)

julia> @btime dropmissing(df, :e, view=true);
  1.100 ms (14 allocations: 3.94 MiB)

julia> @btime dropmissing(df, :f, view=true);
  166.613 μs (12 allocations: 126.59 KiB)

julia> @btime dropmissing(df, :g, view=true);
  188.653 μs (13 allocations: 126.67 KiB)
  
julia> @btime dropmissing(df, view=true);
  1.410 ms (38 allocations: 1.63 MiB)

end

Base.@propagate_inbounds function SubDataFrame(parent::DataFrame, rows::AbstractVector, cols)
Expand Down