diff --git a/docs/src/manuals/sharpbits/stack-overflow-inference.md b/docs/src/manuals/sharpbits/stack-overflow-inference.md index d403d0b56..b2bf7e5ed 100644 --- a/docs/src/manuals/sharpbits/stack-overflow-inference.md +++ b/docs/src/manuals/sharpbits/stack-overflow-inference.md @@ -40,10 +40,7 @@ using RxInfer end end -data = (y = rand(10000), ) - -using Test #hide -@test_throws StackOverflowError infer(model = long_state_space_model(), data = data) #hide +data = (y = randn(10000), ) results = infer( model = long_state_space_model(), diff --git a/examples/.meta.jl b/examples/.meta.jl index f08a814f4..7d8c78ffc 100644 --- a/examples/.meta.jl +++ b/examples/.meta.jl @@ -153,6 +153,12 @@ return ( description = "Using Bayesian Inference and RxInfer to estimate daily litter events (adapted from https://learnableloop.com/posts/LitterModel_PORT.html)", category = :problem_specific ), + ( + filename = "Structural Dynamics with Augmented Kalman Filter.ipynb", + title = "Structural Dynamics with Augmented Kalman Filter", + description = "In this example, we estimate system states and unknown input forces for a simple **structural dynamical system** using the Augmented Kalman Filter (AKF) (https://www.sciencedirect.com/science/article/abs/pii/S0888327011003931) in **RxInfer**.", + category = :problem_specific + ), ( filename = "Tiny Benchmark.ipynb", title = "Tiny Benchmark", diff --git a/examples/pics/shear_model.png b/examples/pics/shear_model.png new file mode 100644 index 000000000..c09441b70 Binary files /dev/null and b/examples/pics/shear_model.png differ diff --git a/examples/problem_specific/Structural Dynamics with Augmented Kalman Filter.ipynb b/examples/problem_specific/Structural Dynamics with Augmented Kalman Filter.ipynb new file mode 100644 index 000000000..211a85458 --- /dev/null +++ b/examples/problem_specific/Structural Dynamics with Augmented Kalman Filter.ipynb @@ -0,0 +1,1791 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Structural Dynamics with Augmented Kalman Filter" + ] + }, + { + "cell_type": "code", + "execution_count": 1, + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "\u001b[32m\u001b[1m Activating\u001b[22m\u001b[39m project at `~/.julia/dev/RxInfer.jl/examples`\n", + "\u001b[32m\u001b[1m Updating\u001b[22m\u001b[39m registry at `~/.julia/registries/General.toml`\n", + "\u001b[32m\u001b[1m Installed\u001b[22m\u001b[39m GraphPPL ─ v4.5.1\n", + "\u001b[32m\u001b[1m Updating\u001b[22m\u001b[39m `~/.julia/dev/RxInfer.jl/examples/Project.toml`\n", + " \u001b[90m[b4ee3484] \u001b[39m\u001b[92m+ BayesBase v1.5.1\u001b[39m\n", + " \u001b[90m[6e4b80f9] \u001b[39m\u001b[92m+ BenchmarkTools v1.5.0\u001b[39m\n", + " \u001b[90m[336ed68f] \u001b[39m\u001b[92m+ CSV v0.10.15\u001b[39m\n", + " \u001b[90m[a93c6f00] \u001b[39m\u001b[92m+ DataFrames v1.7.0\u001b[39m\n", + " \u001b[90m[31c24e10] \u001b[39m\u001b[92m+ Distributions v0.25.115\u001b[39m\n", + " \u001b[90m[62312e5e] \u001b[39m\u001b[92m+ ExponentialFamily v1.6.0\u001b[39m\n", + "\u001b[32m⌃\u001b[39m \u001b[90m[587475ba] \u001b[39m\u001b[92m+ Flux v0.14.25\u001b[39m\n", + " \u001b[90m[38e38edf] \u001b[39m\u001b[92m+ GLM v1.9.0\u001b[39m\n", + " \u001b[90m[b3f8163a] \u001b[39m\u001b[92m+ GraphPPL v4.5.1\u001b[39m\n", + " \u001b[90m[34004b35] \u001b[39m\u001b[92m+ HypergeometricFunctions v0.3.25\u001b[39m\n", + " \u001b[90m[7073ff75] \u001b[39m\u001b[92m+ IJulia v1.26.0\u001b[39m\n", + " \u001b[90m[4138dd39] \u001b[39m\u001b[92m+ JLD v0.13.5\u001b[39m\n", + " \u001b[90m[b964fa9f] \u001b[39m\u001b[92m+ LaTeXStrings v1.4.0\u001b[39m\n", + " \u001b[90m[429524aa] \u001b[39m\u001b[92m+ Optim v1.10.0\u001b[39m\n", + "\u001b[33m⌅\u001b[39m \u001b[90m[3bd65402] \u001b[39m\u001b[92m+ Optimisers v0.3.4\u001b[39m\n", + " \u001b[90m[d96e819e] \u001b[39m\u001b[92m+ Parameters v0.12.3\u001b[39m\n", + " \u001b[90m[91a5bcdd] \u001b[39m\u001b[92m+ Plots v1.40.9\u001b[39m\n", + " \u001b[90m[92933f4c] \u001b[39m\u001b[92m+ ProgressMeter v1.10.2\u001b[39m\n", + " \u001b[90m[a194aa59] \u001b[39m\u001b[92m+ ReactiveMP v4.4.6\u001b[39m\n", + " \u001b[90m[37e2e3b7] \u001b[39m\u001b[92m+ ReverseDiff v1.15.3\u001b[39m\n", + " \u001b[90m[df971d30] \u001b[39m\u001b[92m+ Rocket v1.8.1\u001b[39m\n", + " \u001b[90m[86711068] \u001b[39m\u001b[92m+ RxInfer v3.8.2\u001b[39m\n", + " \u001b[90m[276daf66] \u001b[39m\u001b[92m+ SpecialFunctions v2.5.0\u001b[39m\n", + " \u001b[90m[860ef19b] \u001b[39m\u001b[92m+ StableRNGs v1.0.2\u001b[39m\n", + " \u001b[90m[4c63d2b9] \u001b[39m\u001b[92m+ StatsFuns v1.3.2\u001b[39m\n", + " \u001b[90m[f3b207a7] \u001b[39m\u001b[92m+ StatsPlots v0.15.7\u001b[39m\n", + " \u001b[90m[44d3d7a6] \u001b[39m\u001b[92m+ Weave v0.10.12\u001b[39m\n", + " \u001b[90m[fdbf4ff8] \u001b[39m\u001b[92m+ XLSX v0.10.4\u001b[39m\n", + "\u001b[32m\u001b[1m Updating\u001b[22m\u001b[39m `~/.julia/dev/RxInfer.jl/examples/Manifest.toml`\n", + " \u001b[90m[621f4979] \u001b[39m\u001b[92m+ AbstractFFTs v1.5.0\u001b[39m\n", + " \u001b[90m[7d9f7c33] \u001b[39m\u001b[92m+ Accessors v0.1.39\u001b[39m\n", + " \u001b[90m[79e6a3ab] \u001b[39m\u001b[92m+ Adapt v4.1.1\u001b[39m\n", + " \u001b[90m[66dad0bd] \u001b[39m\u001b[92m+ AliasTables v1.1.3\u001b[39m\n", + " \u001b[90m[dce04be8] \u001b[39m\u001b[92m+ ArgCheck v2.4.0\u001b[39m\n", + " \u001b[90m[ec485272] \u001b[39m\u001b[92m+ ArnoldiMethod v0.4.0\u001b[39m\n", + " \u001b[90m[7d9fca2a] \u001b[39m\u001b[92m+ Arpack v0.5.4\u001b[39m\n", + " \u001b[90m[4fba245c] \u001b[39m\u001b[92m+ ArrayInterface v7.18.0\u001b[39m\n", + " \u001b[90m[4c555306] \u001b[39m\u001b[92m+ ArrayLayouts v1.11.0\u001b[39m\n", + " \u001b[90m[a9b6321e] \u001b[39m\u001b[92m+ Atomix v1.0.1\u001b[39m\n", + " \u001b[90m[13072b0f] \u001b[39m\u001b[92m+ AxisAlgorithms v1.1.0\u001b[39m\n", + " \u001b[90m[198e06fe] \u001b[39m\u001b[92m+ BangBang v0.4.3\u001b[39m\n", + " \u001b[90m[9718e550] \u001b[39m\u001b[92m+ Baselet v0.1.1\u001b[39m\n", + " \u001b[90m[b4ee3484] \u001b[39m\u001b[92m+ BayesBase v1.5.1\u001b[39m\n", + " \u001b[90m[6e4b80f9] \u001b[39m\u001b[92m+ BenchmarkTools v1.5.0\u001b[39m\n", + " \u001b[90m[d1d4a3ce] \u001b[39m\u001b[92m+ BitFlags v0.1.9\u001b[39m\n", + " \u001b[90m[0f2f92aa] \u001b[39m\u001b[92m+ BitSetTuples v1.1.5\u001b[39m\n", + " \u001b[90m[62783981] \u001b[39m\u001b[92m+ BitTwiddlingConvenienceFunctions v0.1.6\u001b[39m\n", + " \u001b[90m[a74b3585] \u001b[39m\u001b[92m+ Blosc v0.7.3\u001b[39m\n", + " \u001b[90m[fa961155] \u001b[39m\u001b[92m+ CEnum v0.5.0\u001b[39m\n", + " \u001b[90m[2a0fbf3d] \u001b[39m\u001b[92m+ CPUSummary v0.2.6\u001b[39m\n", + " \u001b[90m[336ed68f] \u001b[39m\u001b[92m+ CSV v0.10.15\u001b[39m\n", + " \u001b[90m[082447d4] \u001b[39m\u001b[92m+ ChainRules v1.72.2\u001b[39m\n", + " \u001b[90m[d360d2e6] \u001b[39m\u001b[92m+ ChainRulesCore v1.25.1\u001b[39m\n", + " \u001b[90m[fb6a15b2] \u001b[39m\u001b[92m+ CloseOpenIntervals v0.1.13\u001b[39m\n", + " \u001b[90m[aaaa29a8] \u001b[39m\u001b[92m+ Clustering v0.15.8\u001b[39m\n", + " \u001b[90m[6309b1aa] \u001b[39m\u001b[92m+ CodecInflate64 v0.1.3\u001b[39m\n", + " \u001b[90m[944b1d66] \u001b[39m\u001b[92m+ CodecZlib v0.7.6\u001b[39m\n", + " \u001b[90m[35d6a980] \u001b[39m\u001b[92m+ ColorSchemes v3.27.1\u001b[39m\n", + " \u001b[90m[3da002f7] \u001b[39m\u001b[92m+ ColorTypes v0.12.0\u001b[39m\n", + " \u001b[90m[c3611d14] \u001b[39m\u001b[92m+ ColorVectorSpace v0.11.0\u001b[39m\n", + " \u001b[90m[5ae59095] \u001b[39m\u001b[92m+ Colors v0.13.0\u001b[39m\n", + " \u001b[90m[861a8166] \u001b[39m\u001b[92m+ Combinatorics v1.0.2\u001b[39m\n", + " \u001b[90m[bbf7d656] \u001b[39m\u001b[92m+ CommonSubexpressions v0.3.1\u001b[39m\n", + " \u001b[90m[f70d9fcc] \u001b[39m\u001b[92m+ CommonWorldInvalidations v1.0.0\u001b[39m\n", + " \u001b[90m[34da2185] \u001b[39m\u001b[92m+ Compat v4.16.0\u001b[39m\n", + " \u001b[90m[b152e2b5] \u001b[39m\u001b[92m+ CompositeTypes v0.1.4\u001b[39m\n", + " \u001b[90m[a33af91c] \u001b[39m\u001b[92m+ CompositionsBase v0.1.2\u001b[39m\n", + " \u001b[90m[f0e56b4a] \u001b[39m\u001b[92m+ ConcurrentUtilities v2.4.3\u001b[39m\n", + " \u001b[90m[8f4d0f93] \u001b[39m\u001b[92m+ Conda v1.10.2\u001b[39m\n", + " \u001b[90m[187b0558] \u001b[39m\u001b[92m+ ConstructionBase v1.5.8\u001b[39m\n", + " \u001b[90m[6add18c4] \u001b[39m\u001b[92m+ ContextVariablesX v0.1.3\u001b[39m\n", + " \u001b[90m[d38c429a] \u001b[39m\u001b[92m+ Contour v0.6.3\u001b[39m\n", + " \u001b[90m[adafc99b] \u001b[39m\u001b[92m+ CpuId v0.3.1\u001b[39m\n", + " \u001b[90m[a8cc5b0e] \u001b[39m\u001b[92m+ Crayons v4.1.1\u001b[39m\n", + " \u001b[90m[9a962f9c] \u001b[39m\u001b[92m+ DataAPI v1.16.0\u001b[39m\n", + " \u001b[90m[a93c6f00] \u001b[39m\u001b[92m+ DataFrames v1.7.0\u001b[39m\n", + " \u001b[90m[864edb3b] \u001b[39m\u001b[92m+ DataStructures v0.18.20\u001b[39m\n", + " \u001b[90m[e2d170a0] \u001b[39m\u001b[92m+ DataValueInterfaces v1.0.0\u001b[39m\n", + " \u001b[90m[244e2a9f] \u001b[39m\u001b[92m+ DefineSingletons v0.1.2\u001b[39m\n", + " \u001b[90m[8bb1440f] \u001b[39m\u001b[92m+ DelimitedFiles v1.9.1\u001b[39m\n", + " \u001b[90m[85a47980] \u001b[39m\u001b[92m+ Dictionaries v0.4.3\u001b[39m\n", + " \u001b[90m[163ba53b] \u001b[39m\u001b[92m+ DiffResults v1.1.0\u001b[39m\n", + " \u001b[90m[b552c78f] \u001b[39m\u001b[92m+ DiffRules v1.15.1\u001b[39m\n", + " \u001b[90m[b4f34e82] \u001b[39m\u001b[92m+ Distances v0.10.12\u001b[39m\n", + " \u001b[90m[31c24e10] \u001b[39m\u001b[92m+ Distributions v0.25.115\u001b[39m\n", + " \u001b[90m[ffbed154] \u001b[39m\u001b[92m+ DocStringExtensions v0.9.3\u001b[39m\n", + "\u001b[33m⌅\u001b[39m \u001b[90m[cc6bae93] \u001b[39m\u001b[92m+ DomainIntegrals v0.4.6\u001b[39m\n", + " \u001b[90m[5b8099bc] \u001b[39m\u001b[92m+ DomainSets v0.7.14\u001b[39m\n", + " \u001b[90m[460bff9d] \u001b[39m\u001b[92m+ ExceptionUnwrapping v0.1.11\u001b[39m\n", + " \u001b[90m[62312e5e] \u001b[39m\u001b[92m+ ExponentialFamily v1.6.0\u001b[39m\n", + " \u001b[90m[8f5d6c58] \u001b[39m\u001b[92m+ EzXML v1.2.0\u001b[39m\n", + " \u001b[90m[c87230d0] \u001b[39m\u001b[92m+ FFMPEG v0.4.2\u001b[39m\n", + " \u001b[90m[7a1cc6ca] \u001b[39m\u001b[92m+ FFTW v1.8.0\u001b[39m\n", + " \u001b[90m[cc61a311] \u001b[39m\u001b[92m+ FLoops v0.2.2\u001b[39m\n", + " \u001b[90m[b9860ae5] \u001b[39m\u001b[92m+ FLoopsBase v0.1.1\u001b[39m\n", + " \u001b[90m[2d5283b6] \u001b[39m\u001b[92m+ FastCholesky v1.3.1\u001b[39m\n", + "\u001b[33m⌅\u001b[39m \u001b[90m[442a2c76] \u001b[39m\u001b[92m+ FastGaussQuadrature v0.5.1\u001b[39m\n", + " \u001b[90m[5789e2e9] \u001b[39m\u001b[92m+ FileIO v1.16.6\u001b[39m\n", + " \u001b[90m[48062228] \u001b[39m\u001b[92m+ FilePathsBase v0.9.22\u001b[39m\n", + " \u001b[90m[1a297f60] \u001b[39m\u001b[92m+ FillArrays v1.13.0\u001b[39m\n", + " \u001b[90m[6a86dc24] \u001b[39m\u001b[92m+ FiniteDiff v2.26.2\u001b[39m\n", + " \u001b[90m[4130a065] \u001b[39m\u001b[92m+ FixedArguments v0.1.1\u001b[39m\n", + " \u001b[90m[53c48c17] \u001b[39m\u001b[92m+ FixedPointNumbers v0.8.5\u001b[39m\n", + "\u001b[32m⌃\u001b[39m \u001b[90m[587475ba] \u001b[39m\u001b[92m+ Flux v0.14.25\u001b[39m\n", + " \u001b[90m[1fa38f19] \u001b[39m\u001b[92m+ Format v1.3.7\u001b[39m\n", + " \u001b[90m[f6369f11] \u001b[39m\u001b[92m+ ForwardDiff v0.10.38\u001b[39m\n", + " \u001b[90m[069b7b12] \u001b[39m\u001b[92m+ FunctionWrappers v1.1.3\u001b[39m\n", + "\u001b[33m⌅\u001b[39m \u001b[90m[d9f16b24] \u001b[39m\u001b[92m+ Functors v0.4.12\u001b[39m\n", + " \u001b[90m[38e38edf] \u001b[39m\u001b[92m+ GLM v1.9.0\u001b[39m\n", + " \u001b[90m[0c68f7d7] \u001b[39m\u001b[92m+ GPUArrays v11.1.0\u001b[39m\n", + " \u001b[90m[46192b85] \u001b[39m\u001b[92m+ GPUArraysCore v0.2.0\u001b[39m\n", + " \u001b[90m[28b8d3ca] \u001b[39m\u001b[92m+ GR v0.73.10\u001b[39m\n", + " \u001b[90m[d54b0c1a] \u001b[39m\u001b[92m+ GaussQuadrature v0.5.8\u001b[39m\n", + " \u001b[90m[b3f8163a] \u001b[39m\u001b[92m+ GraphPPL v4.5.1\u001b[39m\n", + " \u001b[90m[86223c79] \u001b[39m\u001b[92m+ Graphs v1.12.0\u001b[39m\n", + " \u001b[90m[42e2da0e] \u001b[39m\u001b[92m+ Grisu v1.0.2\u001b[39m\n", + " \u001b[90m[c8ec2601] \u001b[39m\u001b[92m+ H5Zblosc v0.1.2\u001b[39m\n", + " \u001b[90m[19dc6840] \u001b[39m\u001b[92m+ HCubature v1.7.0\u001b[39m\n", + " \u001b[90m[f67ccb44] \u001b[39m\u001b[92m+ HDF5 v0.17.2\u001b[39m\n", + " \u001b[90m[cd3eb016] \u001b[39m\u001b[92m+ HTTP v1.10.15\u001b[39m\n", + " \u001b[90m[eafb193a] \u001b[39m\u001b[92m+ Highlights v0.5.3\u001b[39m\n", + " \u001b[90m[3e5b6fbb] \u001b[39m\u001b[92m+ HostCPUFeatures v0.1.17\u001b[39m\n", + " \u001b[90m[34004b35] \u001b[39m\u001b[92m+ HypergeometricFunctions v0.3.25\u001b[39m\n", + " \u001b[90m[7073ff75] \u001b[39m\u001b[92m+ IJulia v1.26.0\u001b[39m\n", + " \u001b[90m[7869d1d1] \u001b[39m\u001b[92m+ IRTools v0.4.14\u001b[39m\n", + " \u001b[90m[615f187c] \u001b[39m\u001b[92m+ IfElse v0.1.1\u001b[39m\n", + " \u001b[90m[313cdc1a] \u001b[39m\u001b[92m+ Indexing v1.1.1\u001b[39m\n", + " \u001b[90m[d25df0c9] \u001b[39m\u001b[92m+ Inflate v0.1.5\u001b[39m\n", + " \u001b[90m[22cec73e] \u001b[39m\u001b[92m+ InitialValues v0.3.1\u001b[39m\n", + " \u001b[90m[842dd82b] \u001b[39m\u001b[92m+ InlineStrings v1.4.2\u001b[39m\n", + " \u001b[90m[0c81fc1b] \u001b[39m\u001b[92m+ InputBuffers v1.0.0\u001b[39m\n", + " \u001b[90m[a98d9a8b] \u001b[39m\u001b[92m+ Interpolations v0.15.1\u001b[39m\n", + " \u001b[90m[8197267c] \u001b[39m\u001b[92m+ IntervalSets v0.7.10\u001b[39m\n", + " \u001b[90m[3587e190] \u001b[39m\u001b[92m+ InverseFunctions v0.1.17\u001b[39m\n", + " \u001b[90m[41ab1584] \u001b[39m\u001b[92m+ InvertedIndices v1.3.1\u001b[39m\n", + " \u001b[90m[92d709cd] \u001b[39m\u001b[92m+ IrrationalConstants v0.2.2\u001b[39m\n", + " \u001b[90m[82899510] \u001b[39m\u001b[92m+ IteratorInterfaceExtensions v1.0.0\u001b[39m\n", + " \u001b[90m[4138dd39] \u001b[39m\u001b[92m+ JLD v0.13.5\u001b[39m\n", + " \u001b[90m[033835bb] \u001b[39m\u001b[92m+ JLD2 v0.5.10\u001b[39m\n", + " \u001b[90m[1019f520] \u001b[39m\u001b[92m+ JLFzf v0.1.9\u001b[39m\n", + " \u001b[90m[692b3bcd] \u001b[39m\u001b[92m+ JLLWrappers v1.7.0\u001b[39m\n", + " \u001b[90m[682c06a0] \u001b[39m\u001b[92m+ JSON v0.21.4\u001b[39m\n", + " \u001b[90m[b14d175d] \u001b[39m\u001b[92m+ JuliaVariables v0.2.4\u001b[39m\n", + " \u001b[90m[63c18a36] \u001b[39m\u001b[92m+ KernelAbstractions v0.9.31\u001b[39m\n", + " \u001b[90m[5ab0869b] \u001b[39m\u001b[92m+ KernelDensity v0.6.9\u001b[39m\n", + " \u001b[90m[929cbde3] \u001b[39m\u001b[92m+ LLVM v9.1.3\u001b[39m\n", + " \u001b[90m[b964fa9f] \u001b[39m\u001b[92m+ LaTeXStrings v1.4.0\u001b[39m\n", + " \u001b[90m[23fbe1c1] \u001b[39m\u001b[92m+ Latexify v0.16.5\u001b[39m\n", + " \u001b[90m[10f19ff3] \u001b[39m\u001b[92m+ LayoutPointers v0.1.17\u001b[39m\n", + " \u001b[90m[5078a376] \u001b[39m\u001b[92m+ LazyArrays v2.3.1\u001b[39m\n", + " \u001b[90m[d3d80556] \u001b[39m\u001b[92m+ LineSearches v7.3.0\u001b[39m\n", + " \u001b[90m[2ab3a3ac] \u001b[39m\u001b[92m+ LogExpFunctions v0.3.29\u001b[39m\n", + " \u001b[90m[e6f89c97] \u001b[39m\u001b[92m+ LoggingExtras v1.1.0\u001b[39m\n", + " \u001b[90m[bdcacae8] \u001b[39m\u001b[92m+ LoopVectorization v0.12.171\u001b[39m\n", + "\u001b[32m⌃\u001b[39m \u001b[90m[7e8f7934] \u001b[39m\u001b[92m+ MLDataDevices v1.5.3\u001b[39m\n", + " \u001b[90m[d8e11817] \u001b[39m\u001b[92m+ MLStyle v0.4.17\u001b[39m\n", + " \u001b[90m[f1d291b0] \u001b[39m\u001b[92m+ MLUtils v0.4.5\u001b[39m\n", + " \u001b[90m[3da0fdf6] \u001b[39m\u001b[92m+ MPIPreferences v0.1.11\u001b[39m\n", + " \u001b[90m[1914dd2f] \u001b[39m\u001b[92m+ MacroTools v0.5.13\u001b[39m\n", + " \u001b[90m[d125e4d3] \u001b[39m\u001b[92m+ ManualMemory v0.1.8\u001b[39m\n", + " \u001b[90m[41f81499] \u001b[39m\u001b[92m+ MatrixCorrectionTools v1.2.0\u001b[39m\n", + " \u001b[90m[739be429] \u001b[39m\u001b[92m+ MbedTLS v1.1.9\u001b[39m\n", + " \u001b[90m[442fdcdd] \u001b[39m\u001b[92m+ Measures v0.3.2\u001b[39m\n", + " \u001b[90m[fa8bd995] \u001b[39m\u001b[92m+ MetaGraphsNext v0.7.1\u001b[39m\n", + " \u001b[90m[128add7d] \u001b[39m\u001b[92m+ MicroCollections v0.2.0\u001b[39m\n", + " \u001b[90m[e1d29d7a] \u001b[39m\u001b[92m+ Missings v1.2.0\u001b[39m\n", + " \u001b[90m[6f286f6a] \u001b[39m\u001b[92m+ MultivariateStats v0.10.3\u001b[39m\n", + " \u001b[90m[ffc61752] \u001b[39m\u001b[92m+ Mustache v1.0.20\u001b[39m\n", + " \u001b[90m[d41bc354] \u001b[39m\u001b[92m+ NLSolversBase v7.8.3\u001b[39m\n", + " \u001b[90m[872c559c] \u001b[39m\u001b[92m+ NNlib v0.9.26\u001b[39m\n", + " \u001b[90m[77ba4419] \u001b[39m\u001b[92m+ NaNMath v1.0.2\u001b[39m\n", + " \u001b[90m[71a1bf82] \u001b[39m\u001b[92m+ NameResolution v0.1.5\u001b[39m\n", + " \u001b[90m[d9ec5142] \u001b[39m\u001b[92m+ NamedTupleTools v0.14.3\u001b[39m\n", + " \u001b[90m[b8a86587] \u001b[39m\u001b[92m+ NearestNeighbors v0.4.21\u001b[39m\n", + " \u001b[90m[510215fc] \u001b[39m\u001b[92m+ Observables v0.5.5\u001b[39m\n", + " \u001b[90m[6fe1bfb0] \u001b[39m\u001b[92m+ OffsetArrays v1.15.0\u001b[39m\n", + " \u001b[90m[0b1bfda6] \u001b[39m\u001b[92m+ OneHotArrays v0.2.6\u001b[39m\n", + " \u001b[90m[4d8831e6] \u001b[39m\u001b[92m+ OpenSSL v1.4.3\u001b[39m\n", + " \u001b[90m[429524aa] \u001b[39m\u001b[92m+ Optim v1.10.0\u001b[39m\n", + "\u001b[33m⌅\u001b[39m \u001b[90m[3bd65402] \u001b[39m\u001b[92m+ Optimisers v0.3.4\u001b[39m\n", + " \u001b[90m[bac558e1] \u001b[39m\u001b[92m+ OrderedCollections v1.7.0\u001b[39m\n", + " \u001b[90m[90014a1f] \u001b[39m\u001b[92m+ PDMats v0.11.31\u001b[39m\n", + " \u001b[90m[d96e819e] \u001b[39m\u001b[92m+ Parameters v0.12.3\u001b[39m\n", + " \u001b[90m[69de0a69] \u001b[39m\u001b[92m+ Parsers v2.8.1\u001b[39m\n", + " \u001b[90m[b98c9c47] \u001b[39m\u001b[92m+ Pipe v1.3.0\u001b[39m\n", + " \u001b[90m[ccf2f8ad] \u001b[39m\u001b[92m+ PlotThemes v3.3.0\u001b[39m\n", + " \u001b[90m[995b91a9] \u001b[39m\u001b[92m+ PlotUtils v1.4.3\u001b[39m\n", + " \u001b[90m[91a5bcdd] \u001b[39m\u001b[92m+ Plots v1.40.9\u001b[39m\n", + " \u001b[90m[1d0040c9] \u001b[39m\u001b[92m+ PolyesterWeave v0.2.2\u001b[39m\n", + " \u001b[90m[2dfb63ee] \u001b[39m\u001b[92m+ PooledArrays v1.4.3\u001b[39m\n", + " \u001b[90m[85a6dd25] \u001b[39m\u001b[92m+ PositiveFactorizations v0.2.4\u001b[39m\n", + " \u001b[90m[aea7be01] \u001b[39m\u001b[92m+ PrecompileTools v1.2.1\u001b[39m\n", + " \u001b[90m[21216c6a] \u001b[39m\u001b[92m+ Preferences v1.4.3\u001b[39m\n", + " \u001b[90m[8162dcfd] \u001b[39m\u001b[92m+ PrettyPrint v0.2.0\u001b[39m\n", + " \u001b[90m[08abe8d2] \u001b[39m\u001b[92m+ PrettyTables v2.4.0\u001b[39m\n", + " \u001b[90m[33c8b6b6] \u001b[39m\u001b[92m+ ProgressLogging v0.1.4\u001b[39m\n", + " \u001b[90m[92933f4c] \u001b[39m\u001b[92m+ ProgressMeter v1.10.2\u001b[39m\n", + " \u001b[90m[43287f4e] \u001b[39m\u001b[92m+ PtrArrays v1.2.1\u001b[39m\n", + " \u001b[90m[1fd47b50] \u001b[39m\u001b[92m+ QuadGK v2.11.1\u001b[39m\n", + " \u001b[90m[c84ed2f1] \u001b[39m\u001b[92m+ Ratios v0.4.5\u001b[39m\n", + " \u001b[90m[a194aa59] \u001b[39m\u001b[92m+ ReactiveMP v4.4.6\u001b[39m\n", + " \u001b[90m[c1ae055f] \u001b[39m\u001b[92m+ RealDot v0.1.0\u001b[39m\n", + " \u001b[90m[3cdcf5f2] \u001b[39m\u001b[92m+ RecipesBase v1.3.4\u001b[39m\n", + " \u001b[90m[01d81517] \u001b[39m\u001b[92m+ RecipesPipeline v0.6.12\u001b[39m\n", + " \u001b[90m[189a3867] \u001b[39m\u001b[92m+ Reexport v1.2.2\u001b[39m\n", + " \u001b[90m[05181044] \u001b[39m\u001b[92m+ RelocatableFolders v1.0.1\u001b[39m\n", + " \u001b[90m[ae029012] \u001b[39m\u001b[92m+ Requires v1.3.0\u001b[39m\n", + " \u001b[90m[37e2e3b7] \u001b[39m\u001b[92m+ ReverseDiff v1.15.3\u001b[39m\n", + " \u001b[90m[79098fc4] \u001b[39m\u001b[92m+ Rmath v0.8.0\u001b[39m\n", + " \u001b[90m[df971d30] \u001b[39m\u001b[92m+ Rocket v1.8.1\u001b[39m\n", + " \u001b[90m[86711068] \u001b[39m\u001b[92m+ RxInfer v3.8.2\u001b[39m\n", + " \u001b[90m[94e857df] \u001b[39m\u001b[92m+ SIMDTypes v0.1.0\u001b[39m\n", + " \u001b[90m[476501e8] \u001b[39m\u001b[92m+ SLEEFPirates v0.6.43\u001b[39m\n", + " \u001b[90m[6c6a2e73] \u001b[39m\u001b[92m+ Scratch v1.2.1\u001b[39m\n", + " \u001b[90m[91c51154] \u001b[39m\u001b[92m+ SentinelArrays v1.4.8\u001b[39m\n", + " \u001b[90m[efcf1570] \u001b[39m\u001b[92m+ Setfield v1.1.1\u001b[39m\n", + " \u001b[90m[1277b4bf] \u001b[39m\u001b[92m+ ShiftedArrays v2.0.0\u001b[39m\n", + " \u001b[90m[605ecd9f] \u001b[39m\u001b[92m+ ShowCases v0.1.0\u001b[39m\n", + " \u001b[90m[992d4aef] \u001b[39m\u001b[92m+ Showoff v1.0.3\u001b[39m\n", + " \u001b[90m[777ac1f9] \u001b[39m\u001b[92m+ SimpleBufferStream v1.2.0\u001b[39m\n", + " \u001b[90m[699a6c99] \u001b[39m\u001b[92m+ SimpleTraits v0.9.4\u001b[39m\n", + " \u001b[90m[b85f4697] \u001b[39m\u001b[92m+ SoftGlobalScope v1.1.0\u001b[39m\n", + " \u001b[90m[a2af1166] \u001b[39m\u001b[92m+ SortingAlgorithms v1.2.1\u001b[39m\n", + " \u001b[90m[dc90abb0] \u001b[39m\u001b[92m+ SparseInverseSubset v0.1.2\u001b[39m\n", + " \u001b[90m[276daf66] \u001b[39m\u001b[92m+ SpecialFunctions v2.5.0\u001b[39m\n", + " \u001b[90m[171d559e] \u001b[39m\u001b[92m+ SplittablesBase v0.1.15\u001b[39m\n", + " \u001b[90m[860ef19b] \u001b[39m\u001b[92m+ StableRNGs v1.0.2\u001b[39m\n", + " \u001b[90m[aedffcd0] \u001b[39m\u001b[92m+ Static v1.1.1\u001b[39m\n", + " \u001b[90m[0d7ed370] \u001b[39m\u001b[92m+ StaticArrayInterface v1.8.0\u001b[39m\n", + " \u001b[90m[90137ffa] \u001b[39m\u001b[92m+ StaticArrays v1.9.10\u001b[39m\n", + " \u001b[90m[1e83bf80] \u001b[39m\u001b[92m+ StaticArraysCore v1.4.3\u001b[39m\n", + " \u001b[90m[10745b16] \u001b[39m\u001b[92m+ Statistics v1.11.1\u001b[39m\n", + " \u001b[90m[82ae8749] \u001b[39m\u001b[92m+ StatsAPI v1.7.0\u001b[39m\n", + " \u001b[90m[2913bbd2] \u001b[39m\u001b[92m+ StatsBase v0.34.4\u001b[39m\n", + " \u001b[90m[4c63d2b9] \u001b[39m\u001b[92m+ StatsFuns v1.3.2\u001b[39m\n", + " \u001b[90m[3eaba693] \u001b[39m\u001b[92m+ StatsModels v0.7.4\u001b[39m\n", + " \u001b[90m[f3b207a7] \u001b[39m\u001b[92m+ StatsPlots v0.15.7\u001b[39m\n", + " \u001b[90m[69024149] \u001b[39m\u001b[92m+ StringEncodings v0.3.7\u001b[39m\n", + " \u001b[90m[892a3eda] \u001b[39m\u001b[92m+ StringManipulation v0.4.0\u001b[39m\n", + "\u001b[32m⌃\u001b[39m \u001b[90m[09ab397b] \u001b[39m\u001b[92m+ StructArrays v0.6.21\u001b[39m\n", + " \u001b[90m[ab02a1b2] \u001b[39m\u001b[92m+ TableOperations v1.2.0\u001b[39m\n", + " \u001b[90m[3783bdb8] \u001b[39m\u001b[92m+ TableTraits v1.0.1\u001b[39m\n", + " \u001b[90m[bd369af6] \u001b[39m\u001b[92m+ Tables v1.12.0\u001b[39m\n", + " \u001b[90m[62fd8b95] \u001b[39m\u001b[92m+ TensorCore v0.1.1\u001b[39m\n", + " \u001b[90m[8290d209] \u001b[39m\u001b[92m+ ThreadingUtilities v0.5.2\u001b[39m\n", + " \u001b[90m[783c9a47] \u001b[39m\u001b[92m+ TinyHugeNumbers v1.0.2\u001b[39m\n", + " \u001b[90m[3bb67fe8] \u001b[39m\u001b[92m+ TranscodingStreams v0.11.3\u001b[39m\n", + " \u001b[90m[28d57a85] \u001b[39m\u001b[92m+ Transducers v0.4.84\u001b[39m\n", + " \u001b[90m[9d95972d] \u001b[39m\u001b[92m+ TupleTools v1.6.0\u001b[39m\n", + " \u001b[90m[5c2747f8] \u001b[39m\u001b[92m+ URIs v1.5.1\u001b[39m\n", + " \u001b[90m[3a884ed6] \u001b[39m\u001b[92m+ UnPack v1.0.2\u001b[39m\n", + " \u001b[90m[1cfade01] \u001b[39m\u001b[92m+ UnicodeFun v0.4.1\u001b[39m\n", + " \u001b[90m[1986cc42] \u001b[39m\u001b[92m+ Unitful v1.22.0\u001b[39m\n", + " \u001b[90m[45397f5d] \u001b[39m\u001b[92m+ UnitfulLatexify v1.6.4\u001b[39m\n", + " \u001b[90m[9602ed7d] \u001b[39m\u001b[92m+ Unrolled v0.1.5\u001b[39m\n", + " \u001b[90m[013be700] \u001b[39m\u001b[92m+ UnsafeAtomics v0.3.0\u001b[39m\n", + " \u001b[90m[41fe7b60] \u001b[39m\u001b[92m+ Unzip v0.2.0\u001b[39m\n", + " \u001b[90m[3d5dd08c] \u001b[39m\u001b[92m+ VectorizationBase v0.21.71\u001b[39m\n", + " \u001b[90m[81def892] \u001b[39m\u001b[92m+ VersionParsing v1.3.0\u001b[39m\n", + " \u001b[90m[ea10d353] \u001b[39m\u001b[92m+ WeakRefStrings v1.4.2\u001b[39m\n", + " \u001b[90m[44d3d7a6] \u001b[39m\u001b[92m+ Weave v0.10.12\u001b[39m\n", + " \u001b[90m[cc8bc4a8] \u001b[39m\u001b[92m+ Widgets v0.6.7\u001b[39m\n", + " \u001b[90m[efce3f68] \u001b[39m\u001b[92m+ WoodburyMatrices v1.0.0\u001b[39m\n", + " \u001b[90m[76eceee3] \u001b[39m\u001b[92m+ WorkerUtilities v1.6.1\u001b[39m\n", + " \u001b[90m[fdbf4ff8] \u001b[39m\u001b[92m+ XLSX v0.10.4\u001b[39m\n", + " \u001b[90m[ddb6d928] \u001b[39m\u001b[92m+ YAML v0.4.12\u001b[39m\n", + " \u001b[90m[c2297ded] \u001b[39m\u001b[92m+ ZMQ v1.4.0\u001b[39m\n", + " \u001b[90m[49080126] \u001b[39m\u001b[92m+ ZipArchives v2.4.0\u001b[39m\n", + " \u001b[90m[a5390f91] \u001b[39m\u001b[92m+ ZipFile v0.10.1\u001b[39m\n", + "\u001b[33m⌅\u001b[39m \u001b[90m[e88e6eb3] \u001b[39m\u001b[92m+ Zygote v0.6.75\u001b[39m\n", + " \u001b[90m[700de1a5] \u001b[39m\u001b[92m+ ZygoteRules v0.2.5\u001b[39m\n", + "\u001b[33m⌅\u001b[39m \u001b[90m[68821587] \u001b[39m\u001b[92m+ Arpack_jll v3.5.1+1\u001b[39m\n", + " \u001b[90m[0b7ba130] \u001b[39m\u001b[92m+ Blosc_jll v1.21.6+2\u001b[39m\n", + " \u001b[90m[6e34b625] \u001b[39m\u001b[92m+ Bzip2_jll v1.0.8+4\u001b[39m\n", + " \u001b[90m[83423d85] \u001b[39m\u001b[92m+ Cairo_jll v1.18.2+1\u001b[39m\n", + " \u001b[90m[ee1fde0b] \u001b[39m\u001b[92m+ Dbus_jll v1.14.10+0\u001b[39m\n", + " \u001b[90m[2702e6a9] \u001b[39m\u001b[92m+ EpollShim_jll v0.0.20230411+1\u001b[39m\n", + " \u001b[90m[2e619515] \u001b[39m\u001b[92m+ Expat_jll v2.6.4+3\u001b[39m\n", + "\u001b[33m⌅\u001b[39m \u001b[90m[b22a6f82] \u001b[39m\u001b[92m+ FFMPEG_jll v4.4.4+1\u001b[39m\n", + " \u001b[90m[f5851436] \u001b[39m\u001b[92m+ FFTW_jll v3.3.10+3\u001b[39m\n", + " \u001b[90m[a3f928ae] \u001b[39m\u001b[92m+ Fontconfig_jll v2.15.0+0\u001b[39m\n", + " \u001b[90m[d7e528f0] \u001b[39m\u001b[92m+ FreeType2_jll v2.13.3+1\u001b[39m\n", + " \u001b[90m[559328eb] \u001b[39m\u001b[92m+ FriBidi_jll v1.0.16+0\u001b[39m\n", + " \u001b[90m[0656b61e] \u001b[39m\u001b[92m+ GLFW_jll v3.4.0+2\u001b[39m\n", + " \u001b[90m[d2c73de3] \u001b[39m\u001b[92m+ GR_jll v0.73.10+0\u001b[39m\n", + " \u001b[90m[78b55507] \u001b[39m\u001b[92m+ Gettext_jll v0.21.0+0\u001b[39m\n", + " \u001b[90m[7746bdde] \u001b[39m\u001b[92m+ Glib_jll v2.82.4+0\u001b[39m\n", + " \u001b[90m[3b182d85] \u001b[39m\u001b[92m+ Graphite2_jll v1.3.14+1\u001b[39m\n", + "\u001b[32m⌃\u001b[39m \u001b[90m[0234f1f7] \u001b[39m\u001b[92m+ HDF5_jll v1.14.2+1\u001b[39m\n", + " \u001b[90m[2e76f6c2] \u001b[39m\u001b[92m+ HarfBuzz_jll v8.5.0+0\u001b[39m\n", + " \u001b[90m[e33a78d0] \u001b[39m\u001b[92m+ Hwloc_jll v2.11.2+3\u001b[39m\n", + " \u001b[90m[1d5cc7b8] \u001b[39m\u001b[92m+ IntelOpenMP_jll v2024.2.1+0\u001b[39m\n", + " \u001b[90m[aacddb02] \u001b[39m\u001b[92m+ JpegTurbo_jll v3.1.1+0\u001b[39m\n", + " \u001b[90m[c1c5ebd0] \u001b[39m\u001b[92m+ LAME_jll v3.100.2+0\u001b[39m\n", + " \u001b[90m[88015f11] \u001b[39m\u001b[92m+ LERC_jll v4.0.1+0\u001b[39m\n", + " \u001b[90m[dad2f222] \u001b[39m\u001b[92m+ LLVMExtra_jll v0.0.34+0\u001b[39m\n", + " \u001b[90m[1d63c593] \u001b[39m\u001b[92m+ LLVMOpenMP_jll v18.1.7+0\u001b[39m\n", + " \u001b[90m[dd4b983a] \u001b[39m\u001b[92m+ LZO_jll v2.10.2+3\u001b[39m\n", + "\u001b[33m⌅\u001b[39m \u001b[90m[e9f186c6] \u001b[39m\u001b[92m+ Libffi_jll v3.2.2+2\u001b[39m\n", + " \u001b[90m[d4300ac3] \u001b[39m\u001b[92m+ Libgcrypt_jll v1.11.0+0\u001b[39m\n", + " \u001b[90m[7e76a0d4] \u001b[39m\u001b[92m+ Libglvnd_jll v1.7.0+0\u001b[39m\n", + " \u001b[90m[7add5ba3] \u001b[39m\u001b[92m+ Libgpg_error_jll v1.51.1+0\u001b[39m\n", + " \u001b[90m[94ce4f54] \u001b[39m\u001b[92m+ Libiconv_jll v1.17.0+1\u001b[39m\n", + " \u001b[90m[4b2f31a3] \u001b[39m\u001b[92m+ Libmount_jll v2.40.2+2\u001b[39m\n", + " \u001b[90m[89763e89] \u001b[39m\u001b[92m+ Libtiff_jll v4.7.1+0\u001b[39m\n", + " \u001b[90m[38a345b3] \u001b[39m\u001b[92m+ Libuuid_jll v2.40.2+2\u001b[39m\n", + " \u001b[90m[5ced341a] \u001b[39m\u001b[92m+ Lz4_jll v1.10.1+0\u001b[39m\n", + " \u001b[90m[856f044c] \u001b[39m\u001b[92m+ MKL_jll v2024.2.0+0\u001b[39m\n", + " \u001b[90m[7cb0a576] \u001b[39m\u001b[92m+ MPICH_jll v4.2.3+0\u001b[39m\n", + " \u001b[90m[f1f71cc9] \u001b[39m\u001b[92m+ MPItrampoline_jll v5.5.1+2\u001b[39m\n", + " \u001b[90m[9237b28f] \u001b[39m\u001b[92m+ MicrosoftMPI_jll v10.1.4+3\u001b[39m\n", + " \u001b[90m[e7412a2a] \u001b[39m\u001b[92m+ Ogg_jll v1.3.5+1\u001b[39m\n", + " \u001b[90m[fe0851c0] \u001b[39m\u001b[92m+ OpenMPI_jll v5.0.6+0\u001b[39m\n", + " \u001b[90m[458c3c95] \u001b[39m\u001b[92m+ OpenSSL_jll v3.0.15+3\u001b[39m\n", + " \u001b[90m[efe28fd5] \u001b[39m\u001b[92m+ OpenSpecFun_jll v0.5.6+0\u001b[39m\n", + " \u001b[90m[91d4177d] \u001b[39m\u001b[92m+ Opus_jll v1.3.3+0\u001b[39m\n", + " \u001b[90m[36c8627f] \u001b[39m\u001b[92m+ Pango_jll v1.55.5+0\u001b[39m\n", + "\u001b[33m⌅\u001b[39m \u001b[90m[30392449] \u001b[39m\u001b[92m+ Pixman_jll v0.43.4+0\u001b[39m\n", + "\u001b[33m⌅\u001b[39m \u001b[90m[c0090381] \u001b[39m\u001b[92m+ Qt6Base_jll v6.7.1+1\u001b[39m\n", + " \u001b[90m[629bc702] \u001b[39m\u001b[92m+ Qt6Declarative_jll v6.7.1+2\u001b[39m\n", + " \u001b[90m[ce943373] \u001b[39m\u001b[92m+ Qt6ShaderTools_jll v6.7.1+1\u001b[39m\n", + " \u001b[90m[e99dba38] \u001b[39m\u001b[92m+ Qt6Wayland_jll v6.7.1+1\u001b[39m\n", + " \u001b[90m[f50d1b31] \u001b[39m\u001b[92m+ Rmath_jll v0.5.1+0\u001b[39m\n", + " \u001b[90m[a44049a8] \u001b[39m\u001b[92m+ Vulkan_Loader_jll v1.3.243+0\u001b[39m\n", + " \u001b[90m[a2964d1f] \u001b[39m\u001b[92m+ Wayland_jll v1.21.0+2\u001b[39m\n", + " \u001b[90m[2381bf8a] \u001b[39m\u001b[92m+ Wayland_protocols_jll v1.36.0+0\u001b[39m\n", + " \u001b[90m[02c8fc9c] \u001b[39m\u001b[92m+ XML2_jll v2.13.5+0\u001b[39m\n", + " \u001b[90m[aed1982a] \u001b[39m\u001b[92m+ XSLT_jll v1.1.42+0\u001b[39m\n", + " \u001b[90m[ffd25f8a] \u001b[39m\u001b[92m+ XZ_jll v5.6.4+0\u001b[39m\n", + " \u001b[90m[f67eecfb] \u001b[39m\u001b[92m+ Xorg_libICE_jll v1.1.1+0\u001b[39m\n", + " \u001b[90m[c834827a] \u001b[39m\u001b[92m+ Xorg_libSM_jll v1.2.4+0\u001b[39m\n", + " \u001b[90m[4f6342f7] \u001b[39m\u001b[92m+ Xorg_libX11_jll v1.8.6+3\u001b[39m\n", + " \u001b[90m[0c0b7dd1] \u001b[39m\u001b[92m+ Xorg_libXau_jll v1.0.11+3\u001b[39m\n", + " \u001b[90m[935fb764] \u001b[39m\u001b[92m+ Xorg_libXcursor_jll v1.2.3+0\u001b[39m\n", + " \u001b[90m[a3789734] \u001b[39m\u001b[92m+ Xorg_libXdmcp_jll v1.1.4+3\u001b[39m\n", + " \u001b[90m[1082639a] \u001b[39m\u001b[92m+ Xorg_libXext_jll v1.3.6+3\u001b[39m\n", + " \u001b[90m[d091e8ba] \u001b[39m\u001b[92m+ Xorg_libXfixes_jll v6.0.0+0\u001b[39m\n", + " \u001b[90m[a51aa0fd] \u001b[39m\u001b[92m+ Xorg_libXi_jll v1.8.2+0\u001b[39m\n", + " \u001b[90m[d1454406] \u001b[39m\u001b[92m+ Xorg_libXinerama_jll v1.1.5+0\u001b[39m\n", + " \u001b[90m[ec84b674] \u001b[39m\u001b[92m+ Xorg_libXrandr_jll v1.5.4+0\u001b[39m\n", + " \u001b[90m[ea2f1a96] \u001b[39m\u001b[92m+ Xorg_libXrender_jll v0.9.11+1\u001b[39m\n", + " \u001b[90m[14d82f49] \u001b[39m\u001b[92m+ Xorg_libpthread_stubs_jll v0.1.1+3\u001b[39m\n", + " \u001b[90m[c7cfdc94] \u001b[39m\u001b[92m+ Xorg_libxcb_jll v1.17.0+3\u001b[39m\n", + " \u001b[90m[cc61e674] \u001b[39m\u001b[92m+ Xorg_libxkbfile_jll v1.1.2+1\u001b[39m\n", + " \u001b[90m[e920d4aa] \u001b[39m\u001b[92m+ Xorg_xcb_util_cursor_jll v0.1.4+0\u001b[39m\n", + " \u001b[90m[12413925] \u001b[39m\u001b[92m+ Xorg_xcb_util_image_jll v0.4.0+1\u001b[39m\n", + " \u001b[90m[2def613f] \u001b[39m\u001b[92m+ Xorg_xcb_util_jll v0.4.0+1\u001b[39m\n", + " \u001b[90m[975044d2] \u001b[39m\u001b[92m+ Xorg_xcb_util_keysyms_jll v0.4.0+1\u001b[39m\n", + " \u001b[90m[0d47668e] \u001b[39m\u001b[92m+ Xorg_xcb_util_renderutil_jll v0.3.9+1\u001b[39m\n", + " \u001b[90m[c22f9ab0] \u001b[39m\u001b[92m+ Xorg_xcb_util_wm_jll v0.4.1+1\u001b[39m\n", + " \u001b[90m[35661453] \u001b[39m\u001b[92m+ Xorg_xkbcomp_jll v1.4.6+1\u001b[39m\n", + " \u001b[90m[33bec58e] \u001b[39m\u001b[92m+ Xorg_xkeyboard_config_jll v2.39.0+0\u001b[39m\n", + " \u001b[90m[c5fb5394] \u001b[39m\u001b[92m+ Xorg_xtrans_jll v1.5.0+3\u001b[39m\n", + " \u001b[90m[8f1865be] \u001b[39m\u001b[92m+ ZeroMQ_jll v4.3.5+3\u001b[39m\n", + " \u001b[90m[3161d3a3] \u001b[39m\u001b[92m+ Zstd_jll v1.5.7+0\u001b[39m\n", + " \u001b[90m[35ca27e7] \u001b[39m\u001b[92m+ eudev_jll v3.2.9+0\u001b[39m\n", + " \u001b[90m[214eeab7] \u001b[39m\u001b[92m+ fzf_jll v0.56.3+0\u001b[39m\n", + " \u001b[90m[1a1c6b14] \u001b[39m\u001b[92m+ gperf_jll v3.1.1+1\u001b[39m\n", + " \u001b[90m[477f73a3] \u001b[39m\u001b[92m+ libaec_jll v1.1.2+2\u001b[39m\n", + " \u001b[90m[a4ae2306] \u001b[39m\u001b[92m+ libaom_jll v3.11.0+0\u001b[39m\n", + " \u001b[90m[0ac62f75] \u001b[39m\u001b[92m+ libass_jll v0.15.2+0\u001b[39m\n", + " \u001b[90m[1183f4f0] \u001b[39m\u001b[92m+ libdecor_jll v0.2.2+0\u001b[39m\n", + " \u001b[90m[2db6ffa8] \u001b[39m\u001b[92m+ libevdev_jll v1.11.0+0\u001b[39m\n", + " \u001b[90m[f638f0a6] \u001b[39m\u001b[92m+ libfdk_aac_jll v2.0.3+0\u001b[39m\n", + " \u001b[90m[36db933b] \u001b[39m\u001b[92m+ libinput_jll v1.18.0+0\u001b[39m\n", + " \u001b[90m[b53b4c65] \u001b[39m\u001b[92m+ libpng_jll v1.6.45+0\u001b[39m\n", + " \u001b[90m[a9144af2] \u001b[39m\u001b[92m+ libsodium_jll v1.0.20+3\u001b[39m\n", + " \u001b[90m[f27f6e37] \u001b[39m\u001b[92m+ libvorbis_jll v1.3.7+2\u001b[39m\n", + " \u001b[90m[009596ad] \u001b[39m\u001b[92m+ mtdev_jll v1.1.6+0\u001b[39m\n", + " \u001b[90m[1317d2d5] \u001b[39m\u001b[92m+ oneTBB_jll v2021.12.0+0\u001b[39m\n", + "\u001b[33m⌅\u001b[39m \u001b[90m[1270edf5] \u001b[39m\u001b[92m+ x264_jll v2021.5.5+0\u001b[39m\n", + "\u001b[33m⌅\u001b[39m \u001b[90m[dfaa095f] \u001b[39m\u001b[92m+ x265_jll v3.5.0+0\u001b[39m\n", + " \u001b[90m[d8fb68d0] \u001b[39m\u001b[92m+ xkbcommon_jll v1.4.1+2\u001b[39m\n", + " \u001b[90m[0dad84c5] \u001b[39m\u001b[92m+ ArgTools v1.1.2\u001b[39m\n", + " \u001b[90m[56f22d72] \u001b[39m\u001b[92m+ Artifacts v1.11.0\u001b[39m\n", + " \u001b[90m[2a0f44e3] \u001b[39m\u001b[92m+ Base64 v1.11.0\u001b[39m\n", + " \u001b[90m[ade2ca70] \u001b[39m\u001b[92m+ Dates v1.11.0\u001b[39m\n", + " \u001b[90m[8ba89e20] \u001b[39m\u001b[92m+ Distributed v1.11.0\u001b[39m\n", + " \u001b[90m[f43a241f] \u001b[39m\u001b[92m+ Downloads v1.6.0\u001b[39m\n", + " \u001b[90m[7b1f6079] \u001b[39m\u001b[92m+ FileWatching v1.11.0\u001b[39m\n", + " \u001b[90m[9fa8497b] \u001b[39m\u001b[92m+ Future v1.11.0\u001b[39m\n", + " \u001b[90m[b77e0a4c] \u001b[39m\u001b[92m+ InteractiveUtils v1.11.0\u001b[39m\n", + " \u001b[90m[4af54fe1] \u001b[39m\u001b[92m+ LazyArtifacts v1.11.0\u001b[39m\n", + " \u001b[90m[b27032c2] \u001b[39m\u001b[92m+ LibCURL v0.6.4\u001b[39m\n", + " \u001b[90m[76f85450] \u001b[39m\u001b[92m+ LibGit2 v1.11.0\u001b[39m\n", + " \u001b[90m[8f399da3] \u001b[39m\u001b[92m+ Libdl v1.11.0\u001b[39m\n", + " \u001b[90m[37e2e46d] \u001b[39m\u001b[92m+ LinearAlgebra v1.11.0\u001b[39m\n", + " \u001b[90m[56ddb016] \u001b[39m\u001b[92m+ Logging v1.11.0\u001b[39m\n", + " \u001b[90m[d6f4376e] \u001b[39m\u001b[92m+ Markdown v1.11.0\u001b[39m\n", + " \u001b[90m[a63ad114] \u001b[39m\u001b[92m+ Mmap v1.11.0\u001b[39m\n", + " \u001b[90m[ca575930] \u001b[39m\u001b[92m+ NetworkOptions v1.2.0\u001b[39m\n", + " \u001b[90m[44cfe95a] \u001b[39m\u001b[92m+ Pkg v1.11.0\u001b[39m\n", + " \u001b[90m[de0858da] \u001b[39m\u001b[92m+ Printf v1.11.0\u001b[39m\n", + " \u001b[90m[9abbd945] \u001b[39m\u001b[92m+ Profile v1.11.0\u001b[39m\n", + " \u001b[90m[3fa0cd96] \u001b[39m\u001b[92m+ REPL v1.11.0\u001b[39m\n", + " \u001b[90m[9a3f8284] \u001b[39m\u001b[92m+ Random v1.11.0\u001b[39m\n", + " \u001b[90m[ea8e919c] \u001b[39m\u001b[92m+ SHA v0.7.0\u001b[39m\n", + " \u001b[90m[9e88b42a] \u001b[39m\u001b[92m+ Serialization v1.11.0\u001b[39m\n", + " \u001b[90m[1a1011a3] \u001b[39m\u001b[92m+ SharedArrays v1.11.0\u001b[39m\n", + " \u001b[90m[6462fe0b] \u001b[39m\u001b[92m+ Sockets v1.11.0\u001b[39m\n", + " \u001b[90m[2f01184e] \u001b[39m\u001b[92m+ SparseArrays v1.11.0\u001b[39m\n", + " \u001b[90m[f489334b] \u001b[39m\u001b[92m+ StyledStrings v1.11.0\u001b[39m\n", + " \u001b[90m[4607b0f0] \u001b[39m\u001b[92m+ SuiteSparse\u001b[39m\n", + " \u001b[90m[fa267f1f] \u001b[39m\u001b[92m+ TOML v1.0.3\u001b[39m\n", + " \u001b[90m[a4e569a6] \u001b[39m\u001b[92m+ Tar v1.10.0\u001b[39m\n", + " \u001b[90m[8dfed614] \u001b[39m\u001b[92m+ Test v1.11.0\u001b[39m\n", + " \u001b[90m[cf7118a7] \u001b[39m\u001b[92m+ UUIDs v1.11.0\u001b[39m\n", + " \u001b[90m[4ec0a83e] \u001b[39m\u001b[92m+ Unicode v1.11.0\u001b[39m\n", + " \u001b[90m[e66e0078] \u001b[39m\u001b[92m+ CompilerSupportLibraries_jll v1.1.1+0\u001b[39m\n", + " \u001b[90m[deac9b47] \u001b[39m\u001b[92m+ LibCURL_jll v8.6.0+0\u001b[39m\n", + " \u001b[90m[e37daf67] \u001b[39m\u001b[92m+ LibGit2_jll v1.7.2+0\u001b[39m\n", + " \u001b[90m[29816b5a] \u001b[39m\u001b[92m+ LibSSH2_jll v1.11.0+1\u001b[39m\n", + " \u001b[90m[c8ffd9c3] \u001b[39m\u001b[92m+ MbedTLS_jll v2.28.6+0\u001b[39m\n", + " \u001b[90m[14a3606d] \u001b[39m\u001b[92m+ MozillaCACerts_jll v2023.12.12\u001b[39m\n", + " \u001b[90m[4536629a] \u001b[39m\u001b[92m+ OpenBLAS_jll v0.3.27+1\u001b[39m\n", + " \u001b[90m[05823500] \u001b[39m\u001b[92m+ OpenLibm_jll v0.8.1+2\u001b[39m\n", + " \u001b[90m[efcefdf7] \u001b[39m\u001b[92m+ PCRE2_jll v10.42.0+1\u001b[39m\n", + " \u001b[90m[bea87d4a] \u001b[39m\u001b[92m+ SuiteSparse_jll v7.7.0+0\u001b[39m\n", + " \u001b[90m[83775a58] \u001b[39m\u001b[92m+ Zlib_jll v1.2.13+1\u001b[39m\n", + " \u001b[90m[8e850b90] \u001b[39m\u001b[92m+ libblastrampoline_jll v5.11.0+0\u001b[39m\n", + " \u001b[90m[8e850ede] \u001b[39m\u001b[92m+ nghttp2_jll v1.59.0+0\u001b[39m\n", + " \u001b[90m[3f19e933] \u001b[39m\u001b[92m+ p7zip_jll v17.4.0+2\u001b[39m\n", + "\u001b[36m\u001b[1m Info\u001b[22m\u001b[39m Packages marked with \u001b[32m⌃\u001b[39m and \u001b[33m⌅\u001b[39m have new versions available. Those with \u001b[32m⌃\u001b[39m may be upgradable, but those with \u001b[33m⌅\u001b[39m are restricted by compatibility constraints from upgrading. To see why use `status --outdated -m`\n", + "\u001b[92m\u001b[1mPrecompiling\u001b[22m\u001b[39m project...\n", + " 1266.4 ms\u001b[32m ✓ \u001b[39m\u001b[90mMetaGraphsNext\u001b[39m\n", + " 2032.4 ms\u001b[32m ✓ \u001b[39mGraphPPL\n", + " 1403.1 ms\u001b[32m ✓ \u001b[39mGraphPPL → GraphPPLDistributionsExt\n", + " 4029.6 ms\u001b[32m ✓ \u001b[39mRxInfer\n", + " 4 dependencies successfully precompiled in 11 seconds. 481 already precompiled.\n" + ] + } + ], + "source": [ + "# Activate local environment, see `Project.toml`\n", + "import Pkg; Pkg.activate(\"..\"); Pkg.instantiate();" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "This example demonstrates state and input force estimation for structural dynamical systems with [Augmented Kalman Filter (AKF)](https://www.sciencedirect.com/science/article/abs/pii/S0888327011003931) implemented in RxInfer.\n", + "\n", + "**NOTE**: This example was originally featured in [this blog post](https://vflores-io.github.io/). Check it out for additional insights! The notebook has been prepared by [Víctor Flores](https://vflores-io.github.io/) and adapted by [Dmitry Bagaev](https://github.com/bvdmitri)." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## State and Input Estimation\n", + "\n", + "State-space models are fundamental tools in control theory and signal processing that allow us to analyze complex dynamical systems by breaking them down into first-order differential equations. They are particularly important for structural dynamics problems because they can capture both the internal states (like position and velocity) and external influences (like forces) in a unified mathematical framework. A typical **state-space model** formulation might look like this:\n", + "\n", + "$$\n", + "x[k+1] \\sim \\mathcal{N}(A x[k] + B p[k], Q),\n", + "$$\n", + "$$\n", + "y[k] \\sim \\mathcal{N}(G x[k] + J p[k], R),\n", + "$$\n", + "\n", + "where:\n", + "- $x[k]$ represents the system states at time-step $k$\n", + "- $p[k]$ represents the unknown input forces at time-step $k$\n", + "- $y[k]$ represents our noisy measurements at time-step $k$\n", + "- $A$ is the state transition matrix that describes how the system evolves from one time step to the next\n", + "- $B$ is the input matrix that maps the external forces to their effects on the states\n", + "- $Q$ is the process noise covariance matrix that captures uncertainties in the system dynamics\n", + "- $R$ is the measurement noise covariance matrix that represents uncertainties in sensor measurements\n" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## 4-floor shear building model\n", + "\n", + "For this example, we consider a simplified **4-floor shear building model** with **4 degrees of freedom (DOF)**. This system is depicted below:\n", + "\n", + "![](../pics/shear_model.png)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "In this example, the dynamics of a structural system are governed by its **mass** ($ M $), **stiffness** ($ K $), and **damping** ($ C $) matrices, leading to the equation of motion:\n", + "\n", + "$$\n", + "M \\ddot{x}(t) + C \\dot{x}(t) + K x(t) = p(t),\n", + "$$\n", + "\n", + "where $ x(t) $ represents the displacements at each degree of freedom, and $ (t) $ is the external force applied to the system.\n", + "\n", + "This model captures the essential dynamics of a multi-story structure while remaining computationally manageable. The system matrices are defined as follows:\n", + "\n", + "- $ M $ is the diagonal mass matrix representing the lumped masses at each floor, \n", + "- $ K $ is the stiffness matrix representing inter-floor lateral stiffness, and \n", + "- $ C $ is the proportional damping matrix reflecting energy dissipation." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Lets begin the experiment! To start, we import the necessary packages." + ] + }, + { + "cell_type": "code", + "execution_count": 2, + "metadata": {}, + "outputs": [], + "source": [ + "using LinearAlgebra, Statistics, Random, Plots" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "To keep our analysis organized, we'll use a custom `StructuralModelData` data structure. This structure serves as a central repository for **model parameters**, **simulation settings**, **system matrices**, **results**, and **outputs**." + ] + }, + { + "cell_type": "code", + "execution_count": 3, + "metadata": {}, + "outputs": [], + "source": [ + "# define a data structure for the structural model environment\n", + "struct StructuralModelData\n", + " t::Union{Nothing, Any}\n", + " ndof::Union{Nothing, Int64}\n", + " nf::Union{Nothing, Int64}\n", + " N_data::Union{Nothing, Int64}\n", + " y_meas::Union{Nothing, Vector{Vector{Float64}}}\n", + " A_aug::Union{Nothing, Matrix{Float64}}\n", + " G_aug::Union{Nothing, Matrix{Float64}}\n", + " G_aug_fullfield::Union{Nothing, Matrix{Float64}}\n", + " Q_akf::Union{Nothing, Matrix{Float64}}\n", + " R::Union{Nothing, LinearAlgebra.Diagonal{Float64, Vector{Float64}}}\n", + " x_real::Union{Nothing, Matrix{Float64}}\n", + " y_real::Union{Nothing, Matrix{Float64}}\n", + " p_real::Union{Nothing, Matrix{Float64}}\n", + "end" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "We also define a structure for the system matrices." + ] + }, + { + "cell_type": "code", + "execution_count": 4, + "metadata": {}, + "outputs": [], + "source": [ + "# define the structural system matrices\n", + "struct StructuralMatrices\n", + " M::Union{Nothing, Matrix{Float64}}\n", + " K::Union{Nothing, Matrix{Float64}}\n", + " C::Union{Nothing, Matrix{Float64}}\n", + "end\n", + "\n", + "\n", + "M = I(4)\n", + "\n", + "\n", + "K = [\n", + " 2 -1 0 0;\n", + " -1 2 -1 0;\n", + " 0 -1 2 -1;\n", + " 0 0 -1 1\n", + "] * 1e3\n", + "\n", + "C = [\n", + " 2 -1 0 0;\n", + " -1 2 -1 0;\n", + " 0 -1 2 -1;\n", + " 0 0 -1 1\n", + "]\n", + "\n", + "StructuralModel = StructuralMatrices(M, K, C);" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "#### Constructing the State-Space Model\n", + "\n", + "We convert the structural system into its **discrete-time state-space form** for numerical simulation. Starting from the equation of motion:\n", + "\n", + "$$\n", + "M \\ddot{x}(t) + C \\dot{x}(t) + K x(t) = F(t),\n", + "$$\n", + "\n", + "we introduce the state variable:\n", + "$$\n", + "z(t) = \\begin{bmatrix} x(t) \\\\ \\dot{x}(t) \\end{bmatrix},\n", + "$$\n", + "which allows us to express the system as:\n", + "$$\n", + "\\dot{z}(t) = A_{\\text{c}} z(t) + B_{\\text{c}} p(t),\n", + "$$\n", + "where:\n", + "- $ A_{\\text{c}} = \\begin{bmatrix} 0 & I \\\\ -(M^{-1} K) & -(M^{-1} C) \\end{bmatrix} $\n", + "- $ B_{\\text{c}} = \\begin{bmatrix} 0 \\\\ M^{-1} S_p \\end{bmatrix} $\n", + "- $ S_p $ is the input selection matrix that determines where the external forces $ p(t) $ are applied.\n", + "\n", + "To perform simulations, the system is discretized using a time step $ \\Delta t $ as:\n", + "$$\n", + "z[k+1] = A z[k] + B p[k],\n", + "$$\n", + "where:\n", + "- $ A = e^{A_{\\text{c}} \\Delta t} $ is the **state transition matrix**.\n", + "- $ B = (A - I) A_{\\text{c}}^{-1} B_{\\text{c}} $ is the **input matrix**, obtained by integrating the continuous-time system.\n", + "\n", + "This state-space representation forms the basis for propagating the system states during simulation.\n" + ] + }, + { + "cell_type": "code", + "execution_count": 5, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "construct_ssm (generic function with 1 method)" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "# function to construct the state space model\n", + "function construct_ssm(StructuralModel,dt, ndof, nf)\n", + " # unpack the structural model\n", + " M = StructuralModel.M\n", + " K = StructuralModel.K\n", + " C = StructuralModel.C\n", + " \n", + " \n", + " Sp = zeros(ndof, nf)\n", + " Sp[4, 1] = 1\n", + "\n", + " Z = zeros(ndof, ndof)\n", + " Id = I(ndof)\n", + "\n", + " A_continuous = [Z Id;\n", + " -(M \\ K) -(M \\ C)]\n", + " B_continuous = [Z; Id \\ M] * Sp\n", + "\n", + " A = exp(dt * A_continuous)\n", + " B = (A - I(2*ndof)) * A_continuous \\ B_continuous\n", + "\n", + " return A, B, Sp\n", + "end" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Generating Input Forces\n", + "\n", + "External forces $ p[k] $ acting on the system are modeled as **Gaussian white noise**:\n", + "\n", + "$$\n", + "p[k] \\sim \\mathcal{N}(\\mu, \\sigma^2),\n", + "$$\n", + "where $ \\mu $ is the mean and $ \\sigma $ controls the intensity of the force.\n", + "\n", + "In this example, the inputs are generated independently at each time step $ k $ and across input channels to simulate random excitations, such as wind or seismic forces.\n", + "\n" + ] + }, + { + "cell_type": "code", + "execution_count": 6, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "generate_input (generic function with 1 method)" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "# function to generate random input noise\n", + "function generate_input(N_data::Int, nf::Int; input_mu::Float64, input_std::Float64)\n", + " Random.seed!(42)\n", + " p_real = input_mu .+ randn(N_data, nf) .* input_std\n", + " return p_real\n", + "end" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Observation Model\n", + "\n", + "System responses, such as accelerations, are often measured at specific locations using sensors. The measurements are simulated using the equation:\n", + "\n", + "$$\n", + "y[k] = G x[k] + J p[k] + v[k],\n", + "$$\n", + "where:\n", + "- $ G $ maps the system states $ x[k] $ to measured outputs.\n", + "- $ J $ maps the input forces $ p[k] $ to the measurements.\n", + "- $ v[k] \\sim \\mathcal{N}(0, \\sigma_y^2 I) $ is Gaussian noise representing sensor inaccuracies.\n", + "\n", + "The noise variance $ \\sigma_y^2 $ is chosen as a fraction of the true system response variance for realism.\n", + "\n", + "In this example, **accelerations** are measured at selected degrees of freedom (e.g., nodes 1 and 4).\n" + ] + }, + { + "cell_type": "code", + "execution_count": 7, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "generate_measurements (generic function with 1 method)" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "# function to generate the measurements and noise\n", + "function generate_measurements(ndof, na, nv, nd, N_data, x_real, y_real, p_real, StructuralModel, Sp)\n", + " # unpack the structural model\n", + " M = StructuralModel.M\n", + " K = StructuralModel.K\n", + " C = StructuralModel.C\n", + " \n", + " Sa = zeros(na, ndof) # selection matrix\n", + " Sa[1, 1] = 1 # acceleration at node 1\n", + " Sa[2, 4] = 1 # acceleration at node 4\n", + " G = Sa * [-(M \\ K) -(M \\ C)] \n", + " J = Sa * (I \\ M) * Sp\n", + "\n", + " ry = Statistics.var(y_real[2*ndof+1, :], ) * (0.1^2) # simulate noise as 1% RMS of the noise-free acceleration response\n", + "\n", + " nm = na + nv + nd\n", + "\n", + " R = I(nm) .* ry\n", + "\n", + " y_meas = zeros(nm, N_data)\n", + " y_noise = sqrt(ry) .* randn(nm, N_data)\n", + "\n", + " # reconstruct the measurements\n", + " y_meas = Vector{Vector{Float64}}(undef, N_data)\n", + " for i in 1:N_data\n", + " y_meas[i] = G * x_real[:, i] + J * p_real[i, :] + y_noise[:, i]\n", + " end\n", + "\n", + " return y_meas, G, J, R\n", + "end" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Simulating the Structural Response\n", + "\n", + "The structural response under applied forces is governed by the state-space equations:\n", + "\n", + "$$\n", + "\\begin{aligned}\n", + "x[k+1] & = A x[k] + B p[k], \\\\\n", + "y[k] & = G_{\\text{full}} x[k] + J_{\\text{full}} p[k],\n", + "\\end{aligned}\n", + "$$\n", + "where $ x[k] $ are the system states, $ p[k] $ are the input forces, and $ y[k] $ are the **full-field responses**, i.e., the response at every degree of freedom in our structure.\n", + "\n", + "\n", + "The function below returns:\n", + "- **True States**: $ x_{\\text{real}} $, propagated using $ A $ and $ B $.\n", + "- **Full-Field Responses**: $ y_{\\text{real}} $, incorporating both states and inputs.\n", + "- **Input Forces**: $ p_{\\text{real}} $, generated as stochastic excitations.\n", + "- **Response Matrices**: $ G_{\\text{full}} $ (state-to-response) and $ J_{\\text{full}} $ (input-to-response).\n", + "\n", + "These outputs simulate the physical behavior of the system and serve as the basis for inference. We keep the matrices because they will be used later when analyzing our results.\n" + ] + }, + { + "cell_type": "code", + "execution_count": 8, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "simulate_response (generic function with 1 method)" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "# function to simulate the structural response\n", + "function simulate_response(A, B, StructuralModel, Sp, nf, ndof, N_data)\n", + " # unpack the structural model\n", + " M = StructuralModel.M\n", + " K = StructuralModel.K\n", + " C = StructuralModel.C\n", + " \n", + " p_real = generate_input(N_data, nf, input_mu = 0.0, input_std = 0.05)\n", + "\n", + " Z = zeros(ndof, ndof)\n", + " Id = I(ndof)\n", + " \n", + " G_full = [\n", + " Id Z;\n", + " Z Id;\n", + " -(M \\ K) -(M \\ C)\n", + " ]\n", + "\n", + " J_full = [\n", + " Z;\n", + " Z;\n", + " Id \\ M\n", + " ] * Sp\n", + " \n", + " # preallocate matrices\n", + " x_real = zeros(2 * ndof, N_data)\n", + " y_real = zeros(3 * ndof, N_data)\n", + "\n", + " for i in 2:N_data\n", + " x_real[:, i] = A * x_real[:, i-1] + B * p_real[i-1, :]\n", + " y_real[:, i] = G_full * x_real[:, i-1] + J_full * p_real[i-1, :]\n", + " end\n", + "\n", + " return x_real, y_real, p_real, G_full, J_full\n", + "end " + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Augmented State-Space Model\n", + "\n", + "In structural health monitoring, external input forces $ p[k] $ acting on a structure, such as environmental loads or unknown excitations, are often not directly measurable. To estimate both the system states $ x[k] $ and these unknown input forces, we **augment the state vector** as follows:\n", + "\n", + "$$\n", + "\\tilde{x}[k] = \n", + "\\begin{bmatrix}\n", + "x[k] \\\\\n", + "p[k]\n", + "\\end{bmatrix}.\n", + "$$\n", + "\n", + "This approach allows us to simultaneously infer the internal system states (e.g., displacements and velocities) and the unknown inputs using available measurements.\n", + "\n", + "\n", + "The augmented system dynamics are then expressed as:\n", + "\n", + "$$\n", + "\\begin{aligned}\n", + "\\tilde{x}[k+1] & = A_{\\text{aug}} \\tilde{x}[k] + w[k], \\\\\n", + "y[k] & = G_{\\text{aug}} \\tilde{x}[k] + v[k],\n", + "\\end{aligned}\n", + "$$\n", + "\n", + "where:\n", + "- $ A_{\\text{aug}} $: Augmented state transition matrix. \n", + "- $ G_{\\text{aug}} $: Augmented measurement matrix. \n", + "- $ Q_{\\text{akf}} $: Augmented process noise covariance, capturing uncertainties in both states and inputs. \n", + "- $ w[k] $, $ v[k] $: Process and measurement noise. \n", + "\n", + "##### Full-Field vs. Measurement Space \n", + "\n", + "To avoid confusion, we define two augmented measurement matrices: \n", + "- $ G_{\\text{aug}} $: Projects the augmented state vector $ \\tilde{x}[k] $ to the observed **sensor measurements** (e.g., accelerations at specific nodes). \n", + "- $ G^* $: The **augmented full-field measurement matrix**, which projects the augmented state vector to the full-field **system response**. This includes all degrees of freedom (displacements, velocities, and accelerations). \n", + "\n", + "The distinction is critical:\n", + "- $ G_{\\text{aug}} $ is used directly in the smoother to estimate states and inputs from limited measurements. \n", + "- $ G^* $ is used later to reconstruct the full response field for visualization and validation.\n", + "\n", + "For clarity, we will refer to the **augmented full-field matrix** as $ G^* $ throughout the rest of this example, whereas, in the code, this will be the `G_aug_fullfield` object.\n", + "\n", + "#### Noise Covariances \n", + "In this step, the process and measurement noise covariances are assumed to be **known** or **pre-calibrated**. For example:\n", + "- The input force uncertainty $ Q_p $ is set to reflect significant variability. \n", + "- State noise covariance $ Q_x $ is chosen to reflect uncertainty in the model. \n", + "\n", + "The augmented noise covariance matrix $ Q_{\\text{akf}} $ combines these quantities:\n", + "\n", + "$$\n", + "Q_{\\text{akf}} =\n", + "\\begin{aligned}\n", + "\\begin{bmatrix}\n", + "Q_x & 0 \\\\\n", + "0 & Q_p\n", + "\\end{bmatrix}\n", + "\\end{aligned}.\n", + "$$\n", + "\n" + ] + }, + { + "cell_type": "code", + "execution_count": 9, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "construct_augmented_model (generic function with 1 method)" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "# function to construct the augmented model\n", + "function construct_augmented_model(A, B, G, J, G_full, J_full, nf, ndof)\n", + " Z_aug = zeros(nf, 2*ndof)\n", + " A_aug = [\n", + " A B;\n", + " Z_aug I(nf)\n", + " ]\n", + " G_aug = [G J]\n", + "\n", + " G_aug_fullfield = [G_full J_full] # full-field augmented matrix\n", + "\n", + " Qp_aug = I(nf) * 1e-2 # assumed known or pre-callibrated\n", + " \n", + " Qx_aug = zeros(2*ndof, 2*ndof)\n", + " Qx_aug[(ndof+1):end, (ndof+1):end] = I(ndof) * 1e-1 # assumed known or pre-callibrated\n", + "\n", + " Q_akf = [\n", + " Qx_aug Z_aug';\n", + " Z_aug Qp_aug\n", + " ]\n", + "\n", + " return A_aug, G_aug, Q_akf, G_aug_fullfield\n", + "end" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "Finally, we combine all the key steps into a single workflow to generate the system dynamics, responses, measurements, and the augmented state-space model.\n", + "\n", + "The results are stored in a `StructuralModelData` object for convenient access:\n" + ] + }, + { + "cell_type": "code", + "execution_count": 10, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "get_structural_model (generic function with 1 method)" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "function get_structural_model(StructuralModel, simulation_time, dt)\n", + "\n", + " # intialize\n", + " ndof = size(StructuralModel.M)[1] # number of degrees of freedom\n", + " nf = 1 # number of inputs\n", + " na, nv, nd = 2, 0, 0 # number of oberved accelerations, velocities, and displacements\n", + " N_data = Int(simulation_time / dt) + 1\n", + " t = range(0, stop=simulation_time, length=N_data)\n", + "\n", + " # construct state-space model from structural matrices\n", + " A, B, Sp = construct_ssm(StructuralModel, dt, ndof, nf)\n", + "\n", + " # Generate input and simulate response\n", + " x_real, y_real, p_real, G_full, J_full = simulate_response(A, B, StructuralModel, Sp, nf, ndof, N_data)\n", + "\n", + " # Generate measurements\n", + " y_meas, G, J, R = generate_measurements(ndof, na, nv, nd, N_data, x_real, y_real, p_real, StructuralModel, Sp)\n", + "\n", + " # Construct augmented model\n", + " A_aug, G_aug, Q_akf, G_aug_fullfield = construct_augmented_model(A, B, G, J, G_full, J_full, nf, ndof)\n", + "\n", + " return StructuralModelData(t, ndof, nf, N_data, y_meas, A_aug, G_aug, G_aug_fullfield, Q_akf, R, x_real, y_real, p_real)\n", + "end" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "We define the simulation time and time step, then run the workflow to generate the structural model:\n" + ] + }, + { + "cell_type": "code", + "execution_count": 11, + "metadata": {}, + "outputs": [], + "source": [ + "simulation_time = 5.0\n", + "dt = 0.001\n", + "\n", + "model_data = get_structural_model(StructuralModel, simulation_time, dt);" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## State and Input Estimation with RxInfer\n", + "\n", + "In this section, we use **RxInfer** to estimate the system states and unknown input forces from the simulated noisy measurements using the Augmented State Space Model discussed.\n" + ] + }, + { + "cell_type": "code", + "execution_count": 12, + "metadata": {}, + "outputs": [], + "source": [ + "using RxInfer" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Defining the AKF Smoother Model\n", + "\n", + "Here, we define our **Augmented Kalman Filter (AKF)** smoother using RxInfer. This probabilistic model estimates the system states and unknown input forces based on the measurements.\n", + "- **State Prior**: We start with a prior belief about the initial state, `x0`. \n", + "- **State Transition**: At each time step, the system state evolves based on the transition matrix $ A $ and process noise covariance $ Q $:\n", + " $$\n", + " x[k] \\sim \\mathcal{N}(A x[k-1], Q).\n", + " $$\n", + "- **Measurements**: The observations (sensor data) are modeled as noisy measurements of the states:\n", + " $$\n", + " y[k] \\sim \\mathcal{N}(G x[k], R),\n", + " $$\n", + " where $ G $ maps the states to the measurements, and $ R $ is the measurement noise covariance.\n" + ] + }, + { + "cell_type": "code", + "execution_count": 13, + "metadata": {}, + "outputs": [], + "source": [ + "@model function smoother_model(y, x0, A, G, Q, R)\n", + "\n", + " x_prior ~ x0\n", + " x_prev = x_prior # initialize previous state with x_prior\n", + "\n", + " for i in 1:length(y)\n", + " x[i] ~ MvNormal(mean = A * x_prev, cov = Q)\n", + " y[i] ~ MvNormal(mean = G * x[i], cov = R)\n", + " x_prev = x[i]\n", + " end\n", + "\n", + "end" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Running the AKF Smoother\n", + "\n", + "Now that we have our system set up, it's time to estimate the system states and unknown input forces using RxInfer. We'll run the **Augmented Kalman Filter (AKF) smoother** to make sense of the noisy measurements.\n", + "\n", + "Here’s the game plan:\n", + "\n", + "1. **Unpack the Data**: \n", + " We grab everything we need from the `model_data` object – time, matrices, measurements, and noise covariances.\n", + "\n", + "2. **Set the Initial State**: \n", + " We start with a prior belief about the first state, assuming it's zero with some process noise: \n", + " $$\n", + " x_0 \\sim \\mathcal{N}(0, Q_{\\text{akf}}).\n", + " $$\n", + "\n", + "3. **Run the Smoother**: \n", + " We define a helper function to keep things tidy. This function calls RxInfer’s `infer` method, which does the heavy lifting for us.\n", + "\n", + "4. **Extract and Reconstruct**: \n", + " - RxInfer gives us **state marginals**, which are the posterior estimates of the states. \n", + " - Using a helper function, we reconstruct the **full-field responses** (displacements, velocities, and accelerations). \n", + " - We also extract the estimated input forces, which are part of the augmented state.\n", + "\n", + "That’s it! With just a few lines of code, RxInfer takes care of the math behind the scenes and delivers smooth, reliable estimates of what’s happening inside the system.\n" + ] + }, + { + "cell_type": "code", + "execution_count": 14, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "run_smoother (generic function with 1 method)" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "# RxInfer returns the result in its own structure. \n", + "# Here we wrap the results in a different struct for the example's convenience\n", + "struct InferenceResults\n", + " state_marginals\n", + " y_full_means\n", + " y_full_stds\n", + " p_means\n", + " p_stds\n", + "end\n", + "\n", + "function run_smoother(model_data)\n", + " # unpack the model data\n", + " t = model_data.t;\n", + " N_data = model_data.N_data\n", + " A_aug = model_data.A_aug;\n", + " G_aug = model_data.G_aug;\n", + " G_aug_fullfield = model_data.G_aug_fullfield;\n", + " Q_akf = model_data.Q_akf;\n", + " R = model_data.R;\n", + " y_meas = model_data.y_meas;\n", + " \n", + " # initialize the state - required when doing smoothing\n", + " x0 = MvNormalMeanCovariance(zeros(size(A_aug, 1)), Q_akf);\n", + "\n", + " # define the smoother engine\n", + " function smoother_engine(y_meas, A, G, Q, R)\n", + " # run the akf smoother\n", + " result_smoother = infer(\n", + " model = smoother_model(x0 = x0, A = A, G = G, Q = Q, R = R),\n", + " data = (y = y_meas,),\n", + " options = (limit_stack_depth = 500, ) # This setting is required for large models\n", + " )\n", + "\n", + " # return posteriors as this inference task returns the results as posteriors\n", + " # because inference is done over the full graph\n", + " return result_smoother.posteriors[:x]\n", + " end\n", + "\n", + " # get the marginals of x\n", + " state_marginals = smoother_engine(y_meas, A_aug, G_aug, Q_akf, R)\n", + " \n", + " # reconstructing the full-field response:\n", + " # use helper function to reconstruct the full-field response\n", + " y_full_means, y_full_stds = reconstruct_full_field(state_marginals, G_aug_fullfield, N_data)\n", + "\n", + " # extract the estimated input (input modeled as an augmentation state)\n", + " p_results_means = getindex.(mean.(state_marginals), length(state_marginals[1]))\n", + " p_results_stds = getindex.(std.(state_marginals), length(state_marginals[1]))\n", + " \n", + " return InferenceResults(state_marginals, y_full_means, y_full_stds, p_results_means, p_results_stds)\n", + "end" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Mapping States to Full-Field Responses \n", + "\n", + "In the `run_smoother` function, we used a helper function to map the **state estimates** from the AKF smoother back to the **full-field responses** (e.g., displacements, velocities, and accelerations). \n", + "\n", + "_Why is this important?_\n", + "\n", + "While the smoother estimates the system states, we often care about physical quantities like accelerations or displacements across the entire structure.\n", + "\n", + "Using the **augmented full-field matrix** $ G^* $, we compute:\n", + "- **Response means** from state means: \n", + " $$\n", + " \\mu_y[i] = G^* \\mu_x[i].\n", + " $$ \n", + "- **Response uncertainties** from state covariances: \n", + " $$\n", + " \\sigma_y[i] = \\sqrt{\\text{diag}(G^* \\Sigma_x[i] {G^*}^\\top)}.\n", + " $$ \n", + "\n", + "This gives us both the expected **responses** and their **uncertainties** at each time step. \n", + "\n", + "In other words, this function connects the smoother’s internal state estimates to meaningful, physical quantities, making it easy to visualize the system’s behavior. \n" + ] + }, + { + "cell_type": "code", + "execution_count": 15, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "reconstruct_full_field (generic function with 1 method)" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "# helper function to reconstruct the full field response from the state posteriors\n", + "function reconstruct_full_field(\n", + " x_marginals,\n", + " G_aug_fullfield,\n", + " N_data::Int\n", + " )\n", + " \n", + " # preallocate the full field response\n", + " y_means = Vector{Vector{Float64}}(undef, N_data) # vector of vectors\n", + " y_stds = Vector{Vector{Float64}}(undef, N_data)\n", + "\n", + " # reconstruct the full-field response using G_aug_fullfield\n", + " for i in 1:N_data\n", + " # extract the mean and covariance of the state posterior\n", + " state_mean = mean(x_marginals[i]) # each index is a vector\n", + " state_cov = cov(x_marginals[i])\n", + "\n", + " # project mean and covariance onto the full-field response space\n", + " y_means[i] = G_aug_fullfield * state_mean\n", + " y_stds[i] = sqrt.(diag(G_aug_fullfield * state_cov * G_aug_fullfield'))\n", + " end\n", + "\n", + " return y_means, y_stds\n", + "\n", + "end" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "We now run the AKF smoother using the structural model data to estimate the system states, reconstruct the full-field responses, and extract the input forces along with their uncertainties.\n", + "\n", + "Let’s fire up that RxInfer!\n" + ] + }, + { + "cell_type": "code", + "execution_count": 16, + "metadata": {}, + "outputs": [], + "source": [ + "# run the smoother\n", + "smoother_results = run_smoother(model_data);" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "We first write a helper function and then plot the **true states**, **full-field response**, **input**, their **estimates**, and the associated **uncertainty**:" + ] + }, + { + "cell_type": "code", + "execution_count": 17, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "plot_with_uncertainty (generic function with 2 methods)" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "# helper function\n", + "function plot_with_uncertainty(\n", + " t,\n", + " true_values,\n", + " estimated_means,\n", + " estimated_uncertainties,\n", + " ylabel_text,\n", + " title_text,\n", + " label_suffix=\"\";\n", + " plot_size = (700,300),\n", + " \n", + ")\n", + " # plot true values\n", + " plt = plot(\n", + " t,\n", + " true_values,\n", + " label=\"true ($label_suffix)\",\n", + " lw=2,\n", + " color=:blue,\n", + " size=plot_size,\n", + " left_margin = 5Plots.mm,\n", + " top_margin = 5Plots.mm, \n", + " bottom_margin = 5Plots.mm \n", + " )\n", + "\n", + " # plot estimated values with uncertainty ribbon\n", + " plot!(\n", + " plt,\n", + " t,\n", + " estimated_means,\n", + " ribbon=estimated_uncertainties,\n", + " fillalpha=0.3,\n", + " label=\"estimated ($label_suffix)\",\n", + " lw=2,\n", + " color=:orange,\n", + " linestyle=:dash\n", + " )\n", + "\n", + " # add labels and title\n", + " xlabel!(\"time (s)\")\n", + " ylabel!(ylabel_text)\n", + " title!(title_text)\n", + " \n", + " return plt\n", + "end" + ] + }, + { + "cell_type": "code", + "execution_count": 18, + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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", + "image/svg+xml": [ + "\n", + "\n", + "\n", + " \n", + " \n", + " \n", + "\n", + "\n", + "\n", + " \n", + " \n", + " \n", + "\n", + "\n", + "\n", + " \n", + " \n", + " \n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n" + ], + "text/html": [ + "\n", + "\n", + "\n", + " \n", + " \n", + " \n", + "\n", + "\n", + "\n", + " \n", + " \n", + " \n", + "\n", + "\n", + "\n", + " \n", + " \n", + " \n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "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", + "image/svg+xml": [ + "\n", + "\n", + "\n", + " \n", + " \n", + " \n", + "\n", + "\n", + "\n", + " \n", + " \n", + " \n", + "\n", + "\n", + "\n", + " \n", + " \n", + " \n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n" + ], + "text/html": [ + "\n", + "\n", + "\n", + " \n", + " \n", + " \n", + "\n", + "\n", + "\n", + " \n", + " \n", + " \n", + "\n", + "\n", + "\n", + " \n", + " \n", + " \n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "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", + "image/svg+xml": [ + "\n", + "\n", + "\n", + " \n", + " \n", + " \n", + "\n", + "\n", + "\n", + " \n", + " \n", + " \n", + "\n", + "\n", + "\n", + " \n", + " \n", + " \n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n" + ], + "text/html": [ + "\n", + "\n", + "\n", + " \n", + " \n", + " \n", + "\n", + "\n", + "\n", + " \n", + " \n", + " \n", + "\n", + "\n", + "\n", + " \n", + " \n", + " \n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "# select some DOFs to plot\n", + "ndof = size(StructuralModel.M)[1]\n", + "\n", + "display_state_dof = 4 # dof 1:4 displacements, dof 5:8 velocities\n", + "display_response_dof = 2*ndof + 1 # dof 1:4 displacements, dof 5:8 velocities, dof 9:12 accelerations\n", + "display_input_dof = 1 # the only one really\n", + "\n", + "# plot the states\n", + "state_plot = plot_with_uncertainty(\n", + " model_data.t,\n", + " model_data.x_real[display_state_dof, :],\n", + " getindex.(mean.(smoother_results.state_marginals), display_state_dof),\n", + " getindex.(std.(smoother_results.state_marginals), display_state_dof),\n", + " \"state value\",\n", + " \"state estimate (dof $(display_state_dof))\",\n", + " \"state dof $(display_state_dof)\"\n", + ");\n", + "\n", + "# plot the responses\n", + "response_plot = plot_with_uncertainty(\n", + " model_data.t,\n", + " model_data.y_real[display_response_dof, :],\n", + " getindex.(smoother_results.y_full_means, display_response_dof),\n", + " getindex.(smoother_results.y_full_stds, display_response_dof),\n", + " \"response value\",\n", + " \"reconstructed response (dof $(display_response_dof))\",\n", + " \"response dof $(display_response_dof)\"\n", + ");\n", + "\n", + "# plot the inputs\n", + "input_plot = plot_with_uncertainty(\n", + " model_data.t,\n", + " model_data.p_real[:, display_input_dof],\n", + " smoother_results.p_means,\n", + " smoother_results.p_stds,\n", + " \"force value\",\n", + " \"input estimate (applied at dof $(display_input_dof))\",\n", + " \"input force $(display_input_dof)\"\n", + ");\n", + "\n", + "display(state_plot)\n", + "display(response_plot)\n", + "display(input_plot)\n" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Let's quickly go over these results now:\n", + "1. State estimation: The **true state** and the **estimated state** show excellent agreement, demonstrating the accuracy of the smoother model implemented via `RxInfer`. The **uncertainty bounds** around the estimated states are noticeable, especially early in the domain. This reflects the natural uncertainty in state estimation since only **accelerations** are observed, whereas displacements and velocities are inferred through integration.\n", + "2. Reconstructed response: the **real response** and the **reconstructed response** align well across the domain, confirming that the filter captures the dynamics quite nicely. The uncertainty bounds here are narrower, showing that the confidence improves as the filter incorporates observations of these quantities of interest (i.e. accelerations).\n", + "3. Input force reconstruction: The **input force** and its **reconstructed counterpart** show significant high frequency variations with very narrow uncertainty bounds. This is expected because accelerations, being the directly observed quantities, are estimated with higher confidence. Plus, we gave ourselves a small advantage by using a well-calibrated prior on this quantity of interest ($Q_p$)." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "The results demonstrate how well the smoother model, implemented with RxInfer, performs in capturing the system dynamics and reconstructing hidden states and inputs. Notably, setting up the probabilistic model was straightforward and intuitive—much easier than dealing with the rest of the structural modeling! This highlights the power of RxInfer for quickly building and solving complex inference problems while keeping the implementation clean and efficient.\n", + "\n", + "With just a few lines of code, we were able to estimate states, reconstruct responses, and confidently quantify uncertainties—a win for both accuracy and usability. 🚀" + ] + } + ], + "metadata": { + "kernelspec": { + "display_name": "Julia 1.11.2", + "language": "julia", + "name": "julia-1.11" + }, + "language_info": { + "file_extension": ".jl", + "mimetype": "application/julia", + "name": "julia", + "version": "1.11.2" + } + }, + "nbformat": 4, + "nbformat_minor": 4 +}