Skip to content

Commit

Permalink
Automatic build\nPublished by build of: SciML/SciMLBenchmarks.jl@55403f7
Browse files Browse the repository at this point in the history
  • Loading branch information
@JuliaLangBuildkite
JuliaLangBuildkite committed Dec 20, 2024
1 parent 336c3b8 commit cc7be52
Show file tree
Hide file tree
Showing 6 changed files with 40 additions and 37 deletions.
67 changes: 30 additions & 37 deletions markdown/AutomaticDifferentiationSparse/ManualLoopDissusionSparseAD.md
View file
Original file line number Diff line number Diff line change
Expand Up @@ -9,6 +9,7 @@ using LinearAlgebra
using SparseConnectivityTracer: TracerSparsityDetector
using SparseMatrixColorings
import Enzyme,ForwardDiff,Mooncake
import Markdown, PrettyTables, Printf
```


Expand Down Expand Up @@ -113,6 +114,15 @@ scenarios = [ Scenario{:jacobian,:out}(Diffusion,u,res1=DDiffusion(u))];

```julia
df = benchmark_differentiation(bcks, scenarios)
table = PrettyTables.pretty_table(
String,
df;
backend=Val(:markdown),
header=names(df),
formatters=PrettyTables.ft_printf("%.1e"),
)

Markdown.parse(table)
```

```
Expand All @@ -125,22 +135,22 @@ Testing benchmarks
| 12 12 1m32.6s
| 12 12 1m32.4s
ADTypes.AutoEnzyme(mode=EnzymeCore.ReverseMode{false, false, EnzymeCore.F
FIABI, false, false}())
| 2 2 23.8s
| 2 2 23.7s
ADTypes.AutoEnzyme(mode=EnzymeCore.ForwardMode{false, EnzymeCore.FFIABI,
false, false}())
| 2 2 22.7s
| 2 2 22.5s
ADTypes.AutoMooncake{Nothing}(nothing)
| 2 2 34.5s
| 2 2 34.4s
ADTypes.AutoForwardDiff()
Expand All @@ -156,39 +166,22 @@ e{false, EnzymeCore.FFIABI, false, false}()), sparsity_detector=SparseConne
ctivityTracer.TracerSparsityDetector(), coloring_algorithm=SparseMatrixColo
rings.GreedyColoringAlgorithm{:direct, SparseMatrixColorings.NaturalOrder}(
SparseMatrixColorings.NaturalOrder())) | 2 2 3.2s
12×12 DataFrame
Row │ backend scenario
o ⋯
│ Abstract… Scenario…
S ⋯
─────┼─────────────────────────────────────────────────────────────────────
─────
1 │ AutoEnzyme(mode=ReverseMode{fals… Scenario{:jacobian,:out} Diffusi…
v ⋯
2 │ AutoEnzyme(mode=ReverseMode{fals… Scenario{:jacobian,:out} Diffusi…
j
3 │ AutoEnzyme(mode=ForwardMode{fals… Scenario{:jacobian,:out} Diffusi…
v
4 │ AutoEnzyme(mode=ForwardMode{fals… Scenario{:jacobian,:out} Diffusi…
j
5 │ AutoMooncake{Nothing}(nothing) Scenario{:jacobian,:out} Diffusi…
v ⋯
6 │ AutoMooncake{Nothing}(nothing) Scenario{:jacobian,:out} Diffusi…
j
7 │ AutoForwardDiff() Scenario{:jacobian,:out} Diffusi…
v
8 │ AutoForwardDiff() Scenario{:jacobian,:out} Diffusi…
j
9 │ AutoSparse(dense_ad=AutoForwardD… Scenario{:jacobian,:out} Diffusi…
v ⋯
10 │ AutoSparse(dense_ad=AutoForwardD… Scenario{:jacobian,:out} Diffusi…
j
11 │ AutoSparse(dense_ad=AutoEnzyme(m… Scenario{:jacobian,:out} Diffusi…
v
12 │ AutoSparse(dense_ad=AutoEnzyme(m… Scenario{:jacobian,:out} Diffusi…
j
10 columns om
itted
```



| **backend** | **scenario** | **operator** | **prepared** | **calls** | **samples** | **evals** | **time** | **allocs** | **bytes** | **gc_fraction** | **compile_fraction** |
| ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------:| -----------------------------------------------------------------------:| ------------------:| ------------:| ---------:| -----------:| ---------:| --------:| ----------:| ---------:| ---------------:| --------------------:|
| AutoEnzyme(mode=ReverseMode{false, false, FFIABI, false, false}()) | Scenario{:jacobian,:out} Diffusion : Vector{Float64} -> Vector{Float64} | value_and_jacobian | 1.0e+00 | 6.4e+01 | 2.0e+01 | 1.0e+00 | 4.3e-02 | 2.3e+03 | 2.8e+08 | 5.5e-02 | 0.0e+00 |
| AutoEnzyme(mode=ReverseMode{false, false, FFIABI, false, false}()) | Scenario{:jacobian,:out} Diffusion : Vector{Float64} -> Vector{Float64} | jacobian | 1.0e+00 | 6.3e+01 | 2.3e+01 | 1.0e+00 | 4.3e-02 | 2.3e+03 | 2.8e+08 | 5.9e-02 | 0.0e+00 |
| AutoEnzyme(mode=ForwardMode{false, FFIABI, false, false}()) | Scenario{:jacobian,:out} Diffusion : Vector{Float64} -> Vector{Float64} | value_and_jacobian | 1.0e+00 | 6.3e+01 | 1.3e+02 | 1.0e+00 | 6.5e-03 | 3.2e+03 | 1.8e+07 | 0.0e+00 | 0.0e+00 |
| AutoEnzyme(mode=ForwardMode{false, FFIABI, false, false}()) | Scenario{:jacobian,:out} Diffusion : Vector{Float64} -> Vector{Float64} | jacobian | 1.0e+00 | 6.3e+01 | 1.2e+02 | 1.0e+00 | 7.0e-03 | 3.2e+03 | 1.8e+07 | 0.0e+00 | 0.0e+00 |
| AutoMooncake{Nothing}(nothing) | Scenario{:jacobian,:out} Diffusion : Vector{Float64} -> Vector{Float64} | value_and_jacobian | 1.0e+00 | 1.0e+00 | 2.0e+00 | 1.0e+00 | 6.0e-01 | 1.0e+04 | 4.0e+09 | 5.9e-02 | 0.0e+00 |
| AutoMooncake{Nothing}(nothing) | Scenario{:jacobian,:out} Diffusion : Vector{Float64} -> Vector{Float64} | jacobian | 1.0e+00 | 0.0e+00 | 2.0e+00 | 1.0e+00 | 6.1e-01 | 1.0e+04 | 4.0e+09 | 6.1e-02 | 0.0e+00 |
| AutoForwardDiff() | Scenario{:jacobian,:out} Diffusion : Vector{Float64} -> Vector{Float64} | value_and_jacobian | 1.0e+00 | 8.5e+01 | 2.4e+02 | 1.0e+00 | 3.7e-03 | 1.7e+02 | 1.7e+07 | 0.0e+00 | 0.0e+00 |
| AutoForwardDiff() | Scenario{:jacobian,:out} Diffusion : Vector{Float64} -> Vector{Float64} | jacobian | 1.0e+00 | 8.4e+01 | 2.4e+02 | 1.0e+00 | 3.7e-03 | 1.7e+02 | 1.7e+07 | 0.0e+00 | 0.0e+00 |
| AutoSparse(dense_ad=AutoForwardDiff(), sparsity_detector=TracerSparsityDetector(), coloring_algorithm=GreedyColoringAlgorithm{:direct, NaturalOrder}(NaturalOrder())) | Scenario{:jacobian,:out} Diffusion : Vector{Float64} -> Vector{Float64} | value_and_jacobian | 1.0e+00 | 2.0e+00 | 4.5e+04 | 1.0e+00 | 1.3e-05 | 8.0e+00 | 9.6e+04 | 0.0e+00 | 0.0e+00 |
| AutoSparse(dense_ad=AutoForwardDiff(), sparsity_detector=TracerSparsityDetector(), coloring_algorithm=GreedyColoringAlgorithm{:direct, NaturalOrder}(NaturalOrder())) | Scenario{:jacobian,:out} Diffusion : Vector{Float64} -> Vector{Float64} | jacobian | 1.0e+00 | 1.0e+00 | 4.9e+04 | 1.0e+00 | 1.3e-05 | 7.0e+00 | 8.8e+04 | 0.0e+00 | 0.0e+00 |
| AutoSparse(dense_ad=AutoEnzyme(mode=ForwardMode{false, FFIABI, false, false}()), sparsity_detector=TracerSparsityDetector(), coloring_algorithm=GreedyColoringAlgorithm{:direct, NaturalOrder}(NaturalOrder())) | Scenario{:jacobian,:out} Diffusion : Vector{Float64} -> Vector{Float64} | value_and_jacobian | 1.0e+00 | 2.0e+00 | 4.9e+04 | 1.0e+00 | 1.2e-05 | 1.0e+01 | 9.7e+04 | 0.0e+00 | 0.0e+00 |
| AutoSparse(dense_ad=AutoEnzyme(mode=ForwardMode{false, FFIABI, false, false}()), sparsity_detector=TracerSparsityDetector(), coloring_algorithm=GreedyColoringAlgorithm{:direct, NaturalOrder}(NaturalOrder())) | Scenario{:jacobian,:out} Diffusion : Vector{Float64} -> Vector{Float64} | jacobian | 1.0e+00 | 1.0e+00 | 5.5e+04 | 1.0e+00 | 9.0e-06 | 9.0e+00 | 8.9e+04 | 0.0e+00 | 0.0e+00 |

Binary file modified markdown/AutomaticDifferentiationSparse/figures/BrusselatorSparseAD_10_1.png
View file
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file modified markdown/AutomaticDifferentiationSparse/figures/BrusselatorSparseAD_11_1.png
View file
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file modified markdown/AutomaticDifferentiationSparse/figures/BrusselatorSparseAD_8_1.png
View file
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file modified markdown/AutomaticDifferentiationSparse/figures/BrusselatorSparseAD_9_1.png
View file
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
10 changes: 10 additions & 0 deletions script/AutomaticDifferentiationSparse/ManualLoopDissusionSparseAD.jl
View file
Original file line number Diff line number Diff line change
Expand Up @@ -5,6 +5,7 @@ using LinearAlgebra
using SparseConnectivityTracer: TracerSparsityDetector
using SparseMatrixColorings
import Enzyme,ForwardDiff,Mooncake
import Markdown, PrettyTables, Printf


bcks = [
Expand Down Expand Up @@ -60,4 +61,13 @@ scenarios = [ Scenario{:jacobian,:out}(Diffusion,u,res1=DDiffusion(u))];


df = benchmark_differentiation(bcks, scenarios)
table = PrettyTables.pretty_table(
String,
df;
backend=Val(:markdown),
header=names(df),
formatters=PrettyTables.ft_printf("%.1e"),
)

Markdown.parse(table)

0 comments on commit cc7be52

Please sign in to comment.