Performance tips#

using DataFrames
using BenchmarkTools
using CategoricalArrays
using PooledArrays
using Random

Access by column number is faster than by name#

x = DataFrame(rand(5, 1000), :auto)
@btime $x[!, 500]; ## Faster
  3.095 ns (0 allocations: 0 bytes)
@btime $x.x500;  ## Slower
  11.482 ns (0 allocations: 0 bytes)

When working with data DataFrame use barrier functions or type annotation#

function f_bad() ## this function will be slow
    Random.seed!(1)
    x = DataFrame(rand(1000000, 2), :auto)
    y, z = x[!, 1], x[!, 2]
    p = 0.0
    for i in 1:nrow(x)
        p += y[i] * z[i]
    end
    p
end

@btime f_bad();
# if you run @code_warntype f_bad() then you notice
# that Julia does not know column types of `DataFrame`
  100.285 ms (5999022 allocations: 122.06 MiB)

solution 1 is to use barrier function (it should be possible to use it in almost any code) for the calculation. You will notice much less memopry allocations and faster performance.

function f_inner(y, z)
    p = 0.0
    for i in eachindex(y, z)
        p += y[i] * z[i]
    end
    p
end

function f_barrier()
    Random.seed!(1)
    x = DataFrame(rand(1000000, 2), :auto)
    f_inner(x[!, 1], x[!, 2])
end

@btime f_barrier();
  3.905 ms (44 allocations: 30.52 MiB)

or use inbuilt function if possible

using LinearAlgebra

function f_inbuilt()
    Random.seed!(1)
    x = DataFrame(rand(1000000, 2), :auto)
    dot(x[!, 1], x[!, 2])
end

@btime f_inbuilt();
  3.347 ms (44 allocations: 30.52 MiB)

solution 2 is to provide the types of extracted columns. However, there are cases in which you will not know these types.

function f_typed()
    Random.seed!(1)
    x = DataFrame(rand(1000000, 2), :auto)
    y::Vector{Float64}, z::Vector{Float64} = x[!, 1], x[!, 2]
    p = 0.0
    for i in 1:nrow(x)
        p += y[i] * z[i]
    end
    p
end

@btime f_typed();
  3.850 ms (44 allocations: 30.52 MiB)

In general for tall and narrow tables it is often useful to use Tables.rowtable, Tables.columntable or Tables.namedtupleiterator for intermediate processing of data in a type-stable way.

Consider using delayed DataFrame creation technique#

also notice the difference in performance between copying vs non-copying data frame creation

function f1()
    x = DataFrame([Vector{Float64}(undef, 10^4) for i in 1:100], :auto, copycols=false) ## we work with a DataFrame directly
    for c in 1:ncol(x)
        d = x[!, c]
        for r in 1:nrow(x)
            d[r] = rand()
        end
    end
    x
end

function f1a()
    x = DataFrame([Vector{Float64}(undef, 10^4) for i in 1:100], :auto) ## we work with a DataFrame directly
    for c in 1:ncol(x)
        d = x[!, c]
        for r in 1:nrow(x)
            d[r] = rand()
        end
    end
    x
end

function f2()
    x = Vector{Any}(undef, 100)
    for c in 1:length(x)
        d = Vector{Float64}(undef, 10^4)
        for r in eachindex(d)
            d[r] = rand()
        end
        x[c] = d
    end
    DataFrame(x, :auto, copycols=false) ## we delay creation of DataFrame after we have our job done
end

function f2a()
    x = Vector{Any}(undef, 100)
    for c in eachindex(x)
        d = Vector{Float64}(undef, 10^4)
        for r in eachindex(d)
            d[r] = rand()
        end
        x[c] = d
    end
    DataFrame(x, :auto) ## we delay creation of DataFrame after we have our job done
end

@btime f1();
@btime f1a();
@btime f2();
@btime f2a();
  27.487 ms (1949728 allocations: 37.40 MiB)
  28.199 ms (1950028 allocations: 45.03 MiB)
  1.146 ms (728 allocations: 7.66 MiB)
  1.599 ms (1028 allocations: 15.29 MiB)

You can add rows to a DataFrame in place and it is fast#

x = DataFrame(rand(10^6, 5), :auto)
y = DataFrame(transpose(1.0:5.0), :auto)
z = [1.0:5.0;]

@btime vcat($x, $y); ## creates a new DataFrame - slow
@btime append!($x, $y); ## in place - fast

x = DataFrame(rand(10^6, 5), :auto) ## reset to the same starting point
@btime push!($x, $z); ## add a single row in place - fast
  2.825 ms (212 allocations: 38.16 MiB)
  1.116 μs (29 allocations: 1.50 KiB)
  418.724 ns (16 allocations: 256 bytes)

Allowing missing as well as categorical slows down computations#

using StatsBase

function test(data) ## uses countmap function to test performance
    println(eltype(data))
    x = rand(data, 10^6)
    y = categorical(x)
    println(" raw:")
    @btime countmap($x)
    println(" categorical:")
    @btime countmap($y)
    nothing
end

test(1:10)
test([randstring() for i in 1:10])
test(allowmissing(1:10))
test(allowmissing([randstring() for i in 1:10]))
Int64
 raw:
  1.751 ms (8 allocations: 7.63 MiB)
 categorical:
  15.555 ms (1000004 allocations: 30.52 MiB)
String
 raw:
  16.500 ms (4 allocations: 448 bytes)
 categorical:
  26.715 ms (1000004 allocations: 30.52 MiB)
Union{Missing, Int64}
 raw:
  6.382 ms (4 allocations: 464 bytes)
 categorical:
  15.813 ms (1000004 allocations: 30.52 MiB)
Union{Missing, String}
 raw:
  19.140 ms (4 allocations: 448 bytes)
 categorical:
  29.456 ms (1000004 allocations: 30.52 MiB)

When aggregating use column selector and prefer integer, categorical, or pooled array grouping variable#

df = DataFrame(x=rand('a':'d', 10^7), y=1);
gdf = groupby(df, :x)

GroupedDataFrame with 4 groups based on key: x

First Group (2502360 rows): x = 'a': ASCII/Unicode U+0061 (category Ll: Letter, lowercase)
2502335 rows omitted
Rowxy
CharInt64
1a1
2a1
3a1
4a1
5a1
6a1
7a1
8a1
9a1
10a1
11a1
12a1
13a1
2502349a1
2502350a1
2502351a1
2502352a1
2502353a1
2502354a1
2502355a1
2502356a1
2502357a1
2502358a1
2502359a1
2502360a1

Last Group (2498457 rows): x = 'd': ASCII/Unicode U+0064 (category Ll: Letter, lowercase)
2498432 rows omitted
Rowxy
CharInt64
1d1
2d1
3d1
4d1
5d1
6d1
7d1
8d1
9d1
10d1
11d1
12d1
13d1
2498446d1
2498447d1
2498448d1
2498449d1
2498450d1
2498451d1
2498452d1
2498453d1
2498454d1
2498455d1
2498456d1
2498457d1

traditional syntax, slow

@btime combine(v -> sum(v.y), $gdf)
  17.556 ms (332 allocations: 19.11 MiB)
4×2 DataFrame
Rowxx1
CharInt64
1a2502360
2b2499998
3c2499185
4d2498457

use column selector

@btime combine($gdf, :y => sum)
  7.011 ms (198 allocations: 10.14 KiB)
4×2 DataFrame
Rowxy_sum
CharInt64
1a2502360
2b2499998
3c2499185
4d2498457
transform!(df, :x => categorical => :x);
gdf = groupby(df, :x)

GroupedDataFrame with 4 groups based on key: x

First Group (2502360 rows): x = CategoricalArrays.CategoricalValue{Char, UInt32} 'a'
2502335 rows omitted
Rowxy
Cat…Int64
1a1
2a1
3a1
4a1
5a1
6a1
7a1
8a1
9a1
10a1
11a1
12a1
13a1
2502349a1
2502350a1
2502351a1
2502352a1
2502353a1
2502354a1
2502355a1
2502356a1
2502357a1
2502358a1
2502359a1
2502360a1

Last Group (2498457 rows): x = CategoricalArrays.CategoricalValue{Char, UInt32} 'd'
2498432 rows omitted
Rowxy
Cat…Int64
1d1
2d1
3d1
4d1
5d1
6d1
7d1
8d1
9d1
10d1
11d1
12d1
13d1
2498446d1
2498447d1
2498448d1
2498449d1
2498450d1
2498451d1
2498452d1
2498453d1
2498454d1
2498455d1
2498456d1
2498457d1
@btime combine($gdf, :y => sum)
  6.874 ms (206 allocations: 10.62 KiB)
4×2 DataFrame
Rowxy_sum
Cat…Int64
1a2502360
2b2499998
3c2499185
4d2498457
transform!(df, :x => PooledArray{Char} => :x)
10000000×2 DataFrame
9999975 rows omitted
Rowxy
CharInt64
1a1
2b1
3c1
4c1
5a1
6a1
7d1
8d1
9a1
10b1
11b1
12b1
13a1
9999989a1
9999990a1
9999991d1
9999992d1
9999993d1
9999994c1
9999995b1
9999996b1
9999997b1
9999998a1
9999999c1
10000000a1
gdf = groupby(df, :x)

GroupedDataFrame with 4 groups based on key: x

First Group (2502360 rows): x = 'a': ASCII/Unicode U+0061 (category Ll: Letter, lowercase)
2502335 rows omitted
Rowxy
CharInt64
1a1
2a1
3a1
4a1
5a1
6a1
7a1
8a1
9a1
10a1
11a1
12a1
13a1
2502349a1
2502350a1
2502351a1
2502352a1
2502353a1
2502354a1
2502355a1
2502356a1
2502357a1
2502358a1
2502359a1
2502360a1

Last Group (2498457 rows): x = 'd': ASCII/Unicode U+0064 (category Ll: Letter, lowercase)
2498432 rows omitted
Rowxy
CharInt64
1d1
2d1
3d1
4d1
5d1
6d1
7d1
8d1
9d1
10d1
11d1
12d1
13d1
2498446d1
2498447d1
2498448d1
2498449d1
2498450d1
2498451d1
2498452d1
2498453d1
2498454d1
2498455d1
2498456d1
2498457d1
@btime combine($gdf, :y => sum)
  7.070 ms (200 allocations: 10.20 KiB)
4×2 DataFrame
Rowxy_sum
CharInt64
1a2502360
2b2499998
3c2499185
4d2498457

Use views instead of materializing a new DataFrame#

x = DataFrame(rand(100, 1000), :auto)
@btime $x[1:1, :]
  176.249 μs (3993 allocations: 159.07 KiB)
1×1000 DataFrame
900 columns omitted
Rowx1x2x3x4x5x6x7x8x9x10x11x12x13x14x15x16x17x18x19x20x21x22x23x24x25x26x27x28x29x30x31x32x33x34x35x36x37x38x39x40x41x42x43x44x45x46x47x48x49x50x51x52x53x54x55x56x57x58x59x60x61x62x63x64x65x66x67x68x69x70x71x72x73x74x75x76x77x78x79x80x81x82x83x84x85x86x87x88x89x90x91x92x93x94x95x96x97x98x99x100
Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64
10.2879390.5729060.662740.1888510.8813060.6887950.8098680.3056280.6282870.4316360.4601460.2872030.9178270.5941920.9845540.8740090.8596150.301060.9079560.5818310.9252270.6683580.710560.4224620.7287010.193350.6200840.5949480.136570.6901620.5800430.7848960.1895720.9705890.04378690.210710.8894120.2224320.8335840.8868110.8461980.6156340.9077510.2802030.8662020.4637180.5181480.4505850.8586220.4874580.4873880.4152810.2538440.8902050.4393330.2123260.4275540.1316950.04335520.8069750.1194660.2103520.2902220.7687520.3992320.6535290.6403820.9271220.6370530.910590.5227250.5430980.7925620.8564690.4853310.9246050.3157050.838930.05803390.03607730.4544640.0363760.5448740.2852350.3334790.4505410.2101890.1642950.1625840.4693410.543090.3892030.8688570.8201070.951390.6521920.7552780.5758020.861420.882691
@btime $x[1, :]
  23.447 ns (0 allocations: 0 bytes)
DataFrameRow (1000 columns)
900 columns omitted
Rowx1x2x3x4x5x6x7x8x9x10x11x12x13x14x15x16x17x18x19x20x21x22x23x24x25x26x27x28x29x30x31x32x33x34x35x36x37x38x39x40x41x42x43x44x45x46x47x48x49x50x51x52x53x54x55x56x57x58x59x60x61x62x63x64x65x66x67x68x69x70x71x72x73x74x75x76x77x78x79x80x81x82x83x84x85x86x87x88x89x90x91x92x93x94x95x96x97x98x99x100
Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64
10.2879390.5729060.662740.1888510.8813060.6887950.8098680.3056280.6282870.4316360.4601460.2872030.9178270.5941920.9845540.8740090.8596150.301060.9079560.5818310.9252270.6683580.710560.4224620.7287010.193350.6200840.5949480.136570.6901620.5800430.7848960.1895720.9705890.04378690.210710.8894120.2224320.8335840.8868110.8461980.6156340.9077510.2802030.8662020.4637180.5181480.4505850.8586220.4874580.4873880.4152810.2538440.8902050.4393330.2123260.4275540.1316950.04335520.8069750.1194660.2103520.2902220.7687520.3992320.6535290.6403820.9271220.6370530.910590.5227250.5430980.7925620.8564690.4853310.9246050.3157050.838930.05803390.03607730.4544640.0363760.5448740.2852350.3334790.4505410.2101890.1642950.1625840.4693410.543090.3892030.8688570.8201070.951390.6521920.7552780.5758020.861420.882691
@btime view($x, 1:1, :)
  17.899 ns (0 allocations: 0 bytes)
1×1000 SubDataFrame
900 columns omitted
Rowx1x2x3x4x5x6x7x8x9x10x11x12x13x14x15x16x17x18x19x20x21x22x23x24x25x26x27x28x29x30x31x32x33x34x35x36x37x38x39x40x41x42x43x44x45x46x47x48x49x50x51x52x53x54x55x56x57x58x59x60x61x62x63x64x65x66x67x68x69x70x71x72x73x74x75x76x77x78x79x80x81x82x83x84x85x86x87x88x89x90x91x92x93x94x95x96x97x98x99x100
Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64
10.2879390.5729060.662740.1888510.8813060.6887950.8098680.3056280.6282870.4316360.4601460.2872030.9178270.5941920.9845540.8740090.8596150.301060.9079560.5818310.9252270.6683580.710560.4224620.7287010.193350.6200840.5949480.136570.6901620.5800430.7848960.1895720.9705890.04378690.210710.8894120.2224320.8335840.8868110.8461980.6156340.9077510.2802030.8662020.4637180.5181480.4505850.8586220.4874580.4873880.4152810.2538440.8902050.4393330.2123260.4275540.1316950.04335520.8069750.1194660.2103520.2902220.7687520.3992320.6535290.6403820.9271220.6370530.910590.5227250.5430980.7925620.8564690.4853310.9246050.3157050.838930.05803390.03607730.4544640.0363760.5448740.2852350.3334790.4505410.2101890.1642950.1625840.4693410.543090.3892030.8688570.8201070.951390.6521920.7552780.5758020.861420.882691
@btime $x[1:1, 1:20]
  3.729 μs (70 allocations: 3.09 KiB)
1×20 DataFrame
Rowx1x2x3x4x5x6x7x8x9x10x11x12x13x14x15x16x17x18x19x20
Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64
10.2879390.5729060.662740.1888510.8813060.6887950.8098680.3056280.6282870.4316360.4601460.2872030.9178270.5941920.9845540.8740090.8596150.301060.9079560.581831
@btime $x[1, 1:20]
  18.510 ns (0 allocations: 0 bytes)
DataFrameRow (20 columns)
Rowx1x2x3x4x5x6x7x8x9x10x11x12x13x14x15x16x17x18x19x20
Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64
10.2879390.5729060.662740.1888510.8813060.6887950.8098680.3056280.6282870.4316360.4601460.2872030.9178270.5941920.9845540.8740090.8596150.301060.9079560.581831
@btime view($x, 1:1, 1:20)
  18.510 ns (0 allocations: 0 bytes)
1×20 SubDataFrame
Rowx1x2x3x4x5x6x7x8x9x10x11x12x13x14x15x16x17x18x19x20
Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64Float64
10.2879390.5729060.662740.1888510.8813060.6887950.8098680.3056280.6282870.4316360.4601460.2872030.9178270.5941920.9845540.8740090.8596150.301060.9079560.581831

This notebook was generated using Literate.jl.