"New loss" tutorial, docstring fixes. #943

AstitvaAggarwal · 2025-05-24T21:26:03Z

Checklist

Appropriate tests were added
Any code changes were done in a way that does not break public API
All documentation related to code changes were updated
The new code follows the
contributor guidelines, in particular the SciML Style Guide and
COLPRAC.
Any new documentation only uses public API

Additional context

BPINNs improvement is evident in tests, adding in tutorial made it too long imo. Will add the same for PINO in a seperate PR.

AstitvaAggarwal · 2025-05-24T21:29:53Z

@ChrisRackauckas, lmk what you think. Hope this can help with the presentation. (the current improvement has been shown for PINNs and also highly cited BPINNs papers. It completes the BPINN model formulation). And I think Inverse problems are the main thing PINNs are currently used for.

src/ode_solve.jl

ChrisRackauckas · 2025-05-25T05:42:00Z

https://buildkite.com/julialang/neuralpde-dot-jl/builds/3065#01970435-970c-4c7d-9931-ac27b5e0829f/1901-3011

ChrisRackauckas · 2025-05-25T15:24:50Z

Doctests fail

AstitvaAggarwal · 2025-05-25T16:20:11Z

@ChrisRackauckas im not sure if random initializing solvers is a good idea in tests that have some kind of fixed tolerances, but im including it anyways as it shows a clear improvement. Tests should pass now. If they fail, ill let the initialization be fixed but non informative ( [-5.0, 8.0, 5.0, -7.0] for an ideal [1.5, 1, 3.0, 1] ), this would not be a problem right?

AstitvaAggarwal · 2025-05-25T23:16:46Z

@ChrisRackauckas tests pass now, should be fine to merge.

ChrisRackauckas · 2025-05-26T02:07:04Z

https://buildkite.com/julialang/neuralpde-dot-jl/builds/3072#0197096b-97d5-48eb-9753-89779f8377eb/767-2762

AstitvaAggarwal · 2025-05-26T03:10:18Z

@ChrisRackauckas that error is not related to my tutorial. https://buildkite.com/julialang/neuralpde-dot-jl/builds/3072#0197096b-97d5-48eb-9753-89779f8377eb/767-1922

ChrisRackauckas · 2025-05-26T08:23:39Z

It's from BPINNs:

<html>
<body>
<!--StartFragment-->
┌ Error: Hint-handler #93 for MethodError in AbstractMCMC caused an error
--
  | │   exception =
  | │    1-element ExceptionStack:
  | │    MethodError: no method matching parentmodule(::Int64)
  | │    The function `parentmodule` exists, but no method is defined for this combination of argument types.
  | │
  | │    Closest candidates are:
  | │      parentmodule(::Any, ::Any)
  | │       @ Base reflection.jl:2292
  | │      parentmodule(::Base.StackTraces.StackFrame)
  | │       @ Base stacktraces.jl:286
  | │      parentmodule(::Module)
  | │       @ Base reflection.jl:21
  | │      ...
  | │
  | │    Stacktrace:
  | │      [1] (::AbstractMCMC.var"#93#95")(io::IOContext{IOBuffer}, exc::MethodError, argtypes::Vector{Any}, ::Vector{Any})
  | │        @ AbstractMCMC ~/.cache/julia-buildkite-plugin/depots/f8da2e12-18ea-4414-879b-afc071467714/packages/AbstractMCMC/kwj9g/src/AbstractMCMC.jl:126
  | │      [2] invokelatest(::Any, ::Any, ::Vararg{Any}; kwargs::@Kwargs{})
  | │        @ Base ./essentials.jl:1055
  | │      [3] invokelatest(::Any, ::Any, ::Vararg{Any})
  | │        @ Base ./essentials.jl:1052
  | │      [4] show_error_hints(::IOContext{IOBuffer}, ::MethodError, ::Vector{Any}, ::Vararg{Vector{Any}})
  | │        @ Base.Experimental ./experimental.jl:321
  | │      [5] showerror(io::IO, ex::MethodError)
  | │        @ Base ./errorshow.jl:360
  | │      [6] showerror(io::IOContext{IOBuffer}, ex::MethodError, bt::Vector{Union{Ptr{Nothing}, Base.InterpreterIP}}; backtrace::Bool)
  | │        @ Base ./errorshow.jl:97
  | │      [7] showerror
  | │        @ ./errorshow.jl:95 [inlined]
  | │      [8] showvalue(io::IOContext{IOBuffer}, e::Tuple{MethodError, Vector{Union{Ptr{Nothing}, Base.InterpreterIP}}})
  | │        @ Base.CoreLogging ./logging/ConsoleLogger.jl:56
  | │      [9] handle_message(logger::Base.CoreLogging.ConsoleLogger, level::Base.CoreLogging.LogLevel, message::Any, _module::Any, group::Any, id::Any, filepath::Any, line::Any; kwargs...)
  | │        @ Base.CoreLogging ./logging/ConsoleLogger.jl:143
  | │     [10] invokelatest(::Any, ::Any, ::Vararg{Any}; kwargs::@Kwargs{exception::Tuple{MethodError, Vector{Union{Ptr{Nothing}, Base.InterpreterIP}}}})
  | │        @ Base ./essentials.jl:1057
  | │     [11] macro expansion
  | │        @ ./logging/logging.jl:388 [inlined]
  | │     [12] macro expansion
  | │        @ ~/.cache/julia-buildkite-plugin/depots/f8da2e12-18ea-4414-879b-afc071467714/packages/Documenter/iRt2s/src/utilities/utilities.jl:47 [inlined]
  | │     [13] runner(::Type{Documenter.Expanders.ExampleBlocks}, node::MarkdownAST.Node{Nothing}, page::Documenter.Page, doc::Documenter.Document)
  | │        @ Documenter ~/.cache/julia-buildkite-plugin/depots/f8da2e12-18ea-4414-879b-afc071467714/packages/Documenter/iRt2s/src/expander_pipeline.jl:864
  | │     [14] dispatch(::Type{Documenter.Expanders.ExpanderPipeline}, ::MarkdownAST.Node{Nothing}, ::Vararg{Any})
  | │        @ Documenter.Selectors ~/.cache/julia-buildkite-plugin/depots/f8da2e12-18ea-4414-879b-afc071467714/packages/Documenter/iRt2s/src/utilities/Selectors.jl:170
  | │     [15] expand(doc::Documenter.Document)
  | │        @ Documenter ~/.cache/julia-buildkite-plugin/depots/f8da2e12-18ea-4414-879b-afc071467714/packages/Documenter/iRt2s/src/expander_pipeline.jl:59
  | │     [16] runner(::Type{Documenter.Builder.ExpandTemplates}, doc::Documenter.Document)
  | │        @ Documenter ~/.cache/julia-buildkite-plugin/depots/f8da2e12-18ea-4414-879b-afc071467714/packages/Documenter/iRt2s/src/builder_pipeline.jl:224
  | │     [17] dispatch(::Type{Documenter.Builder.DocumentPipeline}, x::Documenter.Document)
  | │        @ Documenter.Selectors ~/.cache/julia-buildkite-plugin/depots/f8da2e12-18ea-4414-879b-afc071467714/packages/Documenter/iRt2s/src/utilities/Selectors.jl:170
  | │     [18] #88
  | │        @ ~/.cache/julia-buildkite-plugin/depots/f8da2e12-18ea-4414-879b-afc071467714/packages/Documenter/iRt2s/src/makedocs.jl:275 [inlined]
  | │     [19] withenv(::Documenter.var"#88#90"{Documenter.Document}, ::Pair{String, Nothing}, ::Vararg{Pair{String, Nothing}})
  | │        @ Base ./env.jl:265
  | │     [20] #87
  | │        @ ~/.cache/julia-buildkite-plugin/depots/f8da2e12-18ea-4414-879b-afc071467714/packages/Documenter/iRt2s/src/makedocs.jl:274 [inlined]
  | │     [21] cd(f::Documenter.var"#87#89"{Documenter.Document}, dir::String)
  | │        @ Base.Filesystem ./file.jl:112
  | │     [22] makedocs(; debug::Bool, format::Documenter.HTMLWriter.HTML, kwargs::@Kwargs{sitename::String, authors::String, modules::Vector{Module}, clean::Bool, doctest::Bool, linkcheck::Bool, warnonly::Vector{Symbol}, pages::Vector{Any}})
  | │        @ Documenter ~/.cache/julia-buildkite-plugin/depots/f8da2e12-18ea-4414-879b-afc071467714/packages/Documenter/iRt2s/src/makedocs.jl:273
  | │     [23] top-level scope
  | │        @ /var/lib/buildkite-agent/builds/gpuci-13/julialang/neuralpde-dot-jl/docs/make.jl:12
  | │     [24] include(fname::String)
  | │        @ Main ./sysimg.jl:38
  | │     [25] top-level scope
  | │        @ none:5
  | │     [26] eval
  | │        @ ./boot.jl:430 [inlined]
  | │     [27] exec_options(opts::Base.JLOptions)
  | │        @ Base ./client.jl:296
  | │     [28] _start()
  | │        @ Base ./client.jl:531

<!--EndFragment-->
</body>
</html>

and then yours:

<html>
<body>
<!--StartFragment-->
┌ Error: failed to run `@example` block in docs/src/tutorials/ode_parameter_estimation.md:62-65
--
  | │ ```@example param_estim_lv
  | │ sol = solve(prob, alg, verbose = true, abstol = 1e-8, maxiters = 5000, saveat = t_)
  | │ @test sol.k.u.p≈true_p rtol=1e-2 norm=Base.Fix1(maximum, abs) # hide
  | │ ```
  | │   exception =
  | │    MethodError: objects of type Int64 are not callable
  | │    The object of type `Int64` exists, but no method is defined for this combination of argument types when trying to treat it as a callable object.
  | │    Maybe you forgot to use an operator such as *, ^, %, / etc. ?
  | │    Stacktrace:
  | │      [1] macro expansion
  | │        @ ~/.cache/julia-buildkite-plugin/depots/f8da2e12-18ea-4414-879b-afc071467714/packages/Zygote/zowwZ/src/compiler/interface2.jl:0 [inlined]
  | │      [2] _pullback(::Zygote.Context{false}, ::Int64, ::ComponentArrays.ComponentVector{Float64, Vector{Float64}, Tuple{ComponentArrays.Axis{(depvar = ViewAxis(1:542, Axis(layer_1 = ViewAxis(1:30, Axis(weight = ViewAxis(1:15, ShapedAxis((15, 1))), bias = ViewAxis(16:30, Shaped1DAxis((15,))))), layer_2 = ViewAxis(31:270, Axis(weight = ViewAxis(1:225, ShapedAxis((15, 15))), bias = ViewAxis(226:240, Shaped1DAxis((15,))))), layer_3 = ViewAxis(271:510, Axis(weight = ViewAxis(1:225, ShapedAxis((15, 15))), bias = ViewAxis(226:240, Shaped1DAxis((15,))))), layer_4 = ViewAxis(511:542, Axis(weight = ViewAxis(1:30, ShapedAxis((2, 15))), bias = ViewAxis(31:32, Shaped1DAxis((2,))))))), p = ViewAxis(543:546, Shaped1DAxis((4,))))}}}, ::NeuralPDE.ODEPhi{Vector{Float64}, Float64, Lux.StatefulLuxLayer{Static.True, Lux.Chain{@NamedTuple{layer_1::Lux.Dense{typeof(NNlib.σ), Int64, Int64, Nothing, Nothing, Static.True}, layer_2::Lux.Dense{typeof(NNlib.σ), Int64, Int64, Nothing, Nothing, Static.True}, layer_3::Lux.Dense{typeof(NNlib.σ), Int64, Int64, Nothing, Nothing, Static.True}, layer_4::Lux.Dense{typeof(identity), Int64, Int64, Nothing, Nothing, Static.True}}, Nothing}, Nothing, @NamedTuple{layer_1::@NamedTuple{}, layer_2::@NamedTuple{}, layer_3::@NamedTuple{}, layer_4::@NamedTuple{}}}})
  | │        @ Zygote ~/.cache/julia-buildkite-plugin/depots/f8da2e12-18ea-4414-879b-afc071467714/packages/Zygote/zowwZ/src/compiler/interface2.jl:91
  | │      [3] total_loss
  | │        @ /var/lib/buildkite-agent/builds/gpuci-13/julialang/neuralpde-dot-jl/src/ode_solve.jl:410 [inlined]
  | │      [4] _pullback(::Zygote.Context{false}, ::NeuralPDE.var"#total_loss#190"{Nothing, Int64, Int64, NeuralPDE.var"#157#160"{NeuralPDE.ODEPhi{Vector{Float64}, Float64, Lux.StatefulLuxLayer{Static.True, Lux.Chain{@NamedTuple{layer_1::Lux.Dense{typeof(NNlib.σ), Int64, Int64, Nothing, Nothing, Static.True}, layer_2::Lux.Dense{typeof(NNlib.σ), Int64, Int64, Nothing, Nothing, Static.True}, layer_3::Lux.Dense{typeof(NNlib.σ), Int64, Int64, Nothing, Nothing, Static.True}, layer_4::Lux.Dense{typeof(identity), Int64, Int64, Nothing, Nothing, Static.True}}, Nothing}, Nothing, @NamedTuple{layer_1::@NamedTuple{}, layer_2::@NamedTuple{}, layer_3::@NamedTuple{}, layer_4::@NamedTuple{}}}}, ODEFunction{false, SciMLBase.AutoSpecialize, typeof(Main.__atexample__named__param_estim_lv.lv), LinearAlgebra.UniformScaling{Bool}, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, typeof(SciMLBase.DEFAULT_OBSERVED), Nothing, Nothing, Nothing, Nothing}, Bool, Vector{Float64}, Bool, Vector{Float64}}, WeightedIntervalTraining{Float64}, NeuralPDE.ODEPhi{Vector{Float64}, Float64, Lux.StatefulLuxLayer{Static.True, Lux.Chain{@NamedTuple{layer_1::Lux.Dense{typeof(NNlib.σ), Int64, Int64, Nothing, Nothing, Static.True}, layer_2::Lux.Dense{typeof(NNlib.σ), Int64, Int64, Nothing, Nothing, Static.True}, layer_3::Lux.Dense{typeof(NNlib.σ), Int64, Int64, Nothing, Nothing, Static.True}, layer_4::Lux.Dense{typeof(identity), Int64, Int64, Nothing, Nothing, Static.True}}, Nothing}, Nothing, @NamedTuple{layer_1::@NamedTuple{}, layer_2::@NamedTuple{}, layer_3::@NamedTuple{}, layer_4::@NamedTuple{}}}}, typeof(Main.__atexample__named__param_estim_lv.additional_loss), Bool, Bool, Bool, Vector{Float64}, ODEFunction{false, SciMLBase.AutoSpecialize, typeof(Main.__atexample__named__param_estim_lv.lv), LinearAlgebra.UniformScaling{Bool}, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, typeof(SciMLBase.DEFAULT_OBSERVED), Nothing, Nothing, Nothing, Nothing}, Tuple{Float64, Float64}}, ::ComponentArrays.ComponentVector{Float64, Vector{Float64}, Tuple{ComponentArrays.Axis{(depvar = ViewAxis(1:542, Axis(layer_1 = ViewAxis(1:30, Axis(weight = ViewAxis(1:15, ShapedAxis((15, 1))), bias = ViewAxis(16:30, Shaped1DAxis((15,))))), layer_2 = ViewAxis(31:270, Axis(weight = ViewAxis(1:225, ShapedAxis((15, 15))), bias = ViewAxis(226:240, Shaped1DAxis((15,))))), layer_3 = ViewAxis(271:510, Axis(weight = ViewAxis(1:225, ShapedAxis((15, 15))), bias = ViewAxis(226:240, Shaped1DAxis((15,))))), layer_4 = ViewAxis(511:542, Axis(weight = ViewAxis(1:30, ShapedAxis((2, 15))), bias = ViewAxis(31:32, Shaped1DAxis((2,))))))), p = ViewAxis(543:546, Shaped1DAxis((4,))))}}}, ::SciMLBase.NullParameters)
  | │        @ Zygote ~/.cache/julia-buildkite-plugin/depots/f8da2e12-18ea-4414-879b-afc071467714/packages/Zygote/zowwZ/src/compiler/interface2.jl:0
  | │      [5] pullback(::Function, ::Zygote.Context{false}, ::ComponentArrays.ComponentVector{Float64, Vector{Float64}, Tuple{ComponentArrays.Axis{(depvar = ViewAxis(1:542, Axis(layer_1 = ViewAxis(1:30, Axis(weight = ViewAxis(1:15, ShapedAxis((15, 1))), bias = ViewAxis(16:30, Shaped1DAxis((15,))))), layer_2 = ViewAxis(31:270, Axis(weight = ViewAxis(1:225, ShapedAxis((15, 15))), bias = ViewAxis(226:240, Shaped1DAxis((15,))))), layer_3 = ViewAxis(271:510, Axis(weight = ViewAxis(1:225, ShapedAxis((15, 15))), bias = ViewAxis(226:240, Shaped1DAxis((15,))))), layer_4 = ViewAxis(511:542, Axis(weight = ViewAxis(1:30, ShapedAxis((2, 15))), bias = ViewAxis(31:32, Shaped1DAxis((2,))))))), p = ViewAxis(543:546, Shaped1DAxis((4,))))}}}, ::Vararg{Any})
  | │        @ Zygote ~/.cache/julia-buildkite-plugin/depots/f8da2e12-18ea-4414-879b-afc071467714/packages/Zygote/zowwZ/src/compiler/interface.jl:90
  | │      [6] pullback
  | │        @ ~/.cache/julia-buildkite-plugin/depots/f8da2e12-18ea-4414-879b-afc071467714/packages/Zygote/zowwZ/src/compiler/interface.jl:88 [inlined]
  | │      [7] gradient(::Function, ::ComponentArrays.ComponentVector{Float64, Vector{Float64}, Tuple{ComponentArrays.Axis{(depvar = ViewAxis(1:542, Axis(layer_1 = ViewAxis(1:30, Axis(weight = ViewAxis(1:15, ShapedAxis((15, 1))), bias = ViewAxis(16:30, Shaped1DAxis((15,))))), layer_2 = ViewAxis(31:270, Axis(weight = ViewAxis(1:225, ShapedAxis((15, 15))), bias = ViewAxis(226:240, Shaped1DAxis((15,))))), layer_3 = ViewAxis(271:510, Axis(weight = ViewAxis(1:225, ShapedAxis((15, 15))), bias = ViewAxis(226:240, Shaped1DAxis((15,))))), layer_4 = ViewAxis(511:542, Axis(weight = ViewAxis(1:30, ShapedAxis((2, 15))), bias = ViewAxis(31:32, Shaped1DAxis((2,))))))), p = ViewAxis(543:546, Shaped1DAxis((4,))))}}}, ::SciMLBase.NullParameters)
  | │        @ Zygote ~/.cache/julia-buildkite-plugin/depots/f8da2e12-18ea-4414-879b-afc071467714/packages/Zygote/zowwZ/src/compiler/interface.jl:147
  | │      [8] gradient
  | │        @ ~/.cache/julia-buildkite-plugin/depots/f8da2e12-18ea-4414-879b-afc071467714/packages/DifferentiationInterface/zJHX8/ext/DifferentiationInterfaceZygoteExt/DifferentiationInterfaceZygoteExt.jl:123 [inlined]
  | │      [9] gradient!(f::Function, grad::ComponentArrays.ComponentVector{Float64, Vector{Float64}, Tuple{ComponentArrays.Axis{(depvar = ViewAxis(1:542, Axis(layer_1 = ViewAxis(1:30, Axis(weight = ViewAxis(1:15, ShapedAxis((15, 1))), bias = ViewAxis(16:30, Shaped1DAxis((15,))))), layer_2 = ViewAxis(31:270, Axis(weight = ViewAxis(1:225, ShapedAxis((15, 15))), bias = ViewAxis(226:240, Shaped1DAxis((15,))))), layer_3 = ViewAxis(271:510, Axis(weight = ViewAxis(1:225, ShapedAxis((15, 15))), bias = ViewAxis(226:240, Shaped1DAxis((15,))))), layer_4 = ViewAxis(511:542, Axis(weight = ViewAxis(1:30, ShapedAxis((2, 15))), bias = ViewAxis(31:32, Shaped1DAxis((2,))))))), p = ViewAxis(543:546, Shaped1DAxis((4,))))}}}, prep::DifferentiationInterface.NoGradientPrep{Nothing}, backend::ADTypes.AutoZygote, x::ComponentArrays.ComponentVector{Float64, Vector{Float64}, Tuple{ComponentArrays.Axis{(depvar = ViewAxis(1:542, Axis(layer_1 = ViewAxis(1:30, Axis(weight = ViewAxis(1:15, ShapedAxis((15, 1))), bias = ViewAxis(16:30, Shaped1DAxis((15,))))), layer_2 = ViewAxis(31:270, Axis(weight = ViewAxis(1:225, ShapedAxis((15, 15))), bias = ViewAxis(226:240, Shaped1DAxis((15,))))), layer_3 = ViewAxis(271:510, Axis(weight = ViewAxis(1:225, ShapedAxis((15, 15))), bias = ViewAxis(226:240, Shaped1DAxis((15,))))), layer_4 = ViewAxis(511:542, Axis(weight = ViewAxis(1:30, ShapedAxis((2, 15))), bias = ViewAxis(31:32, Shaped1DAxis((2,))))))), p = ViewAxis(543:546, Shaped1DAxis((4,))))}}}, contexts::DifferentiationInterface.Constant{SciMLBase.NullParameters})
  | │        @ DifferentiationInterfaceZygoteExt ~/.cache/julia-buildkite-plugin/depots/f8da2e12-18ea-4414-879b-afc071467714/packages/DifferentiationInterface/zJHX8/ext/DifferentiationInterfaceZygoteExt/DifferentiationInterfaceZygoteExt.jl:139
  | │     [10] (::OptimizationZygoteExt.var"#grad#14"{SciMLBase.NullParameters, OptimizationFunction{true, ADTypes.AutoZygote, NeuralPDE.var"#total_loss#190"{Nothing, Int64, Int64, NeuralPDE.var"#157#160"{NeuralPDE.ODEPhi{Vector{Float64}, Float64, Lux.StatefulLuxLayer{Static.True, Lux.Chain{@NamedTuple{layer_1::Lux.Dense{typeof(NNlib.σ), Int64, Int64, Nothing, Nothing, Static.True}, layer_2::Lux.Dense{typeof(NNlib.σ), Int64, Int64, Nothing, Nothing, Static.True}, layer_3::Lux.Dense{typeof(NNlib.σ), Int64, Int64, Nothing, Nothing, Static.True}, layer_4::Lux.Dense{typeof(identity), Int64, Int64, Nothing, Nothing, Static.True}}, Nothing}, Nothing, @NamedTuple{layer_1::@NamedTuple{}, layer_2::@NamedTuple{}, layer_3::@NamedTuple{}, layer_4::@NamedTuple{}}}}, ODEFunction{false, SciMLBase.AutoSpecialize, typeof(Main.__atexample__named__param_estim_lv.lv), LinearAlgebra.UniformScaling{Bool}, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, typeof(SciMLBase.DEFAULT_OBSERVED), Nothing, Nothing, Nothing, Nothing}, Bool, Vector{Float64}, Bool, Vector{Float64}}, WeightedIntervalTraining{Float64}, NeuralPDE.ODEPhi{Vector{Float64}, Float64, Lux.StatefulLuxLayer{Static.True, Lux.Chain{@NamedTuple{layer_1::Lux.Dense{typeof(NNlib.σ), Int64, Int64, Nothing, Nothing, Static.True}, layer_2::Lux.Dense{typeof(NNlib.σ), Int64, Int64, Nothing, Nothing, Static.True}, layer_3::Lux.Dense{typeof(NNlib.σ), Int64, Int64, Nothing, Nothing, Static.True}, layer_4::Lux.Dense{typeof(identity), Int64, Int64, Nothing, Nothing, Static.True}}, Nothing}, Nothing, @NamedTuple{layer_1::@NamedTuple{}, layer_2::@NamedTuple{}, layer_3::@NamedTuple{}, layer_4::@NamedTuple{}}}}, typeof(Main.__atexample__named__param_estim_lv.additional_loss), Bool, Bool, Bool, Vector{Float64}, ODEFunction{false, SciMLBase.AutoSpecialize, typeof(Main.__atexample__named__param_estim_lv.lv), LinearAlgebra.UniformScaling{Bool}, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, typeof(SciMLBase.DEFAULT_OBSERVED), Nothing, Nothing, Nothing, Nothing}, Tuple{Float64, Float64}}, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, typeof(SciMLBase.DEFAULT_OBSERVED_NO_TIME), Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing}, ADTypes.AutoZygote})(res::ComponentArrays.ComponentVector{Float64, Vector{Float64}, Tuple{ComponentArrays.Axis{(depvar = ViewAxis(1:542, Axis(layer_1 = ViewAxis(1:30, Axis(weight = ViewAxis(1:15, ShapedAxis((15, 1))), bias = ViewAxis(16:30, Shaped1DAxis((15,))))), layer_2 = ViewAxis(31:270, Axis(weight = ViewAxis(1:225, ShapedAxis((15, 15))), bias = ViewAxis(226:240, Shaped1DAxis((15,))))), layer_3 = ViewAxis(271:510, Axis(weight = ViewAxis(1:225, ShapedAxis((15, 15))), bias = ViewAxis(226:240, Shaped1DAxis((15,))))), layer_4 = ViewAxis(511:542, Axis(weight = ViewAxis(1:30, ShapedAxis((2, 15))), bias = ViewAxis(31:32, Shaped1DAxis((2,))))))), p = ViewAxis(543:546, Shaped1DAxis((4,))))}}}, θ::ComponentArrays.ComponentVector{Float64, Vector{Float64}, Tuple{ComponentArrays.Axis{(depvar = ViewAxis(1:542, Axis(layer_1 = ViewAxis(1:30, Axis(weight = ViewAxis(1:15, ShapedAxis((15, 1))), bias = ViewAxis(16:30, Shaped1DAxis((15,))))), layer_2 = ViewAxis(31:270, Axis(weight = ViewAxis(1:225, ShapedAxis((15, 15))), bias = ViewAxis(226:240, Shaped1DAxis((15,))))), layer_3 = ViewAxis(271:510, Axis(weight = ViewAxis(1:225, ShapedAxis((15, 15))), bias = ViewAxis(226:240, Shaped1DAxis((15,))))), layer_4 = ViewAxis(511:542, Axis(weight = ViewAxis(1:30, ShapedAxis((2, 15))), bias = ViewAxis(31:32, Shaped1DAxis((2,))))))), p = ViewAxis(543:546, Shaped1DAxis((4,))))}}})
  | │        @ OptimizationZygoteExt ~/.cache/julia-buildkite-plugin/depots/f8da2e12-18ea-4414-879b-afc071467714/packages/OptimizationBase/UXLhR/ext/OptimizationZygoteExt.jl:35

<!--EndFragment-->
</body>
</html>

both are yours.

ChrisRackauckas · 2025-05-27T07:36:24Z

docs/src/tutorials/ode_parameter_estimation.md

@@ -54,7 +54,7 @@ Next we define the optimizer and [`NNODE`](@ref) which is then plugged into the
 ```@example param_estim_lv
 opt = LBFGS(linesearch = BackTracking())
 alg = NNODE(chain, opt, ps; strategy = WeightedIntervalTraining([0.7, 0.2, 0.1], 500),
-    param_estim = true, additional_loss)
+    param_estim = true, additional_loss = additional_loss)


odd that this matters, this is perfectly fine syntax, i.e. a kwarg of ;a where the value is a automatically expands to a=a.

Actually it dint, it was working fine. The issue was somewhere else.

added tutorial, docstring fixes

af58a51

ChrisRackauckas reviewed May 24, 2025

View reviewed changes

src/ode_solve.jl Show resolved Hide resolved

typos

c26280d

ChrisRackauckas and others added 4 commits May 25, 2025 05:42

Update Project.toml

abf11a4

Update index.md

76af4f3

format

739da79

Update data_collocation_Inverse.md

a278088

doc tests

23b4339

AstitvaAggarwal and others added 2 commits May 25, 2025 23:36

Update data_collocation_Inverse.md

936c0e5

relax tests

34acd3d

AstitvaAggarwal added 3 commits May 26, 2025 14:51

my code is innocent!

b3e48c5

retrigger tests, prev-param estim doctut kwarg fix

0b34c64

fix: ode_solve.jl

428e0ad

ChrisRackauckas merged commit 0fa677e into SciML:master May 27, 2025
42 of 60 checks passed

ChrisRackauckas reviewed May 27, 2025

View reviewed changes

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

Uh oh!

"New loss" tutorial, docstring fixes. #943

"New loss" tutorial, docstring fixes. #943

Uh oh!

AstitvaAggarwal commented May 24, 2025

Uh oh!

AstitvaAggarwal commented May 24, 2025

Uh oh!

Uh oh!

ChrisRackauckas commented May 25, 2025

Uh oh!

ChrisRackauckas commented May 25, 2025

Uh oh!

AstitvaAggarwal commented May 25, 2025 •

edited

Loading

Uh oh!

AstitvaAggarwal commented May 25, 2025

Uh oh!

ChrisRackauckas commented May 26, 2025

Uh oh!

AstitvaAggarwal commented May 26, 2025

Uh oh!

ChrisRackauckas commented May 26, 2025

Uh oh!

Uh oh!

ChrisRackauckas May 27, 2025

Uh oh!

AstitvaAggarwal May 27, 2025

Uh oh!

Uh oh!

Uh oh!

"New loss" tutorial, docstring fixes. #943

"New loss" tutorial, docstring fixes. #943

Uh oh!

Conversation

AstitvaAggarwal commented May 24, 2025

Checklist

Additional context

Uh oh!

AstitvaAggarwal commented May 24, 2025

Uh oh!

Uh oh!

ChrisRackauckas commented May 25, 2025

Uh oh!

ChrisRackauckas commented May 25, 2025

Uh oh!

AstitvaAggarwal commented May 25, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

AstitvaAggarwal commented May 25, 2025

Uh oh!

ChrisRackauckas commented May 26, 2025

Uh oh!

AstitvaAggarwal commented May 26, 2025

Uh oh!

ChrisRackauckas commented May 26, 2025

Uh oh!

Uh oh!

ChrisRackauckas May 27, 2025

Choose a reason for hiding this comment

Uh oh!

AstitvaAggarwal May 27, 2025

Choose a reason for hiding this comment

Uh oh!

Uh oh!

AstitvaAggarwal commented May 25, 2025 •

edited

Loading