Rosenbrock Methods fail when using InterpolatingAdjoint with checkpointing

[m-bossart]

Describe the bug 🐞

The sensitivity for an ODE solve using a Rosenbrock method fails with either sensealg = InterpolatingAdjoint(checkpointing=true) or sensealg = GaussAdjoint(checkpointing=true)

Expected behavior

I expect the combination of options to be compatible; the bug is related to the first call to the time gradient (which is why it only fails for the Rosenbrock methods).

Minimal Reproducible Example 👇

using Zygote, SciMLSensitivity
using OrdinaryDiffEq, Enzyme, Test
p = rand(3)
function dudt(u, p, t)
    u .* p
end
function loss(p)
    prob = ODEProblem(dudt, [3.0, 2.0, 1.0], (0.0, 1.0), p)
    sol = solve(prob, Rodas5(), dt = 0.01, saveat = 0.1, abstol = 1e-5, reltol = 1e-5)
    sum(abs2, Array(sol))
end
dp = Zygote.gradient(loss, p)[1]

function loss(p, solver, sensealg)
    prob = ODEProblem(dudt, [3.0, 2.0, 1.0], (0.0, 1.0), p)
    sol = solve(prob, solver, dt = 0.01, saveat = 0.1, sensealg = sensealg,
        abstol = 1e-5, reltol = 1e-5)
    sum(abs2, Array(sol))
end

dp1 = Zygote.gradient(p -> loss(p, Tsit5(), InterpolatingAdjoint()), p)[1]
dp2 = Zygote.gradient(p -> loss(p, Tsit5(), InterpolatingAdjoint(checkpointing = true)), p)[1]
dp3 = Zygote.gradient(p -> loss(p, Rodas4(), InterpolatingAdjoint()), p)[1]
dp4 = Zygote.gradient(p -> loss(p, Rodas4(), InterpolatingAdjoint(checkpointing = true)), p)[1]  #FAILS: ERROR: First call to automatic differentiation for time gradient failed.
dp5 = Zygote.gradient(p -> loss(p, Rodas4(), BacksolveAdjoint()), p)[1]
dp6 = Zygote.gradient(p -> loss(p, Rodas4(), BacksolveAdjoint(checkpointing = true)), p)[1]
dp7 = Zygote.gradient(p -> loss(p, Rodas4(), GaussAdjoint()), p)[1]
dp8 = Zygote.gradient(p -> loss(p, Rodas4(), GaussAdjoint(checkpointing = true)), p)[1] #FAILS: ERROR: First call to automatic differentiation for time gradient failed.
dp9 = Zygote.gradient(p -> loss(p, Tsit5(), GaussAdjoint(checkpointing = true)), p)[1]

@test dp≈dp1 rtol=1e-2
@test dp≈dp2 rtol=1e-2
@test dp≈dp3 rtol=1e-2
@test dp≈dp5 rtol=1e-2
@test dp≈dp6 rtol=1e-2
@test dp≈dp7 rtol=1e-2
@test dp≈dp9 rtol=1e-2

Error & Stacktrace ⚠️

┌ Warning: Automatic AD choice of autojacvec failed in ODE adjoint, failing back to ODE adjoint + numerical vjp
└ @ SciMLSensitivity C:\Users\Matt Bossart\.julia\packages\SciMLSensitivity\4YtYh\src\sensitivity_interface.jl:396
ERROR: First call to automatic differentiation for time gradient
failed. This means that the user `f` function is not compatible
with automatic differentiation. Methods to fix this include:

1. Turn off automatic differentiation (e.g. Rosenbrock23() becomes
   Rosenbrock23(autodiff=false)). More details can be found at
   https://docs.sciml.ai/DiffEqDocs/stable/features/performance_overloads/
2. Improving the compatibility of `f` with ForwardDiff.jl automatic
   differentiation (using tools like PreallocationTools.jl). More details
   can be found at https://docs.sciml.ai/DiffEqDocs/stable/basics/faq/#Autodifferentiation-and-Dual-Numbers
3. Defining analytical Jacobians and time gradients. More details can be
   found at https://docs.sciml.ai/DiffEqDocs/stable/types/ode_types/#SciMLBase.ODEFunction

Note 1: this failure occurred inside of the time gradient function. These
time gradients are only required by Rosenbrock methods (`Rosenbrock23`,
`Rodas4`, etc.) are are done by automatic differentiation w.r.t. the
argument `t`. If your function is compatible with automatic differentiation
w.r.t. `u`, i.e. for Jacobian generation, another way to work around this
issue is to switch to a non-Rosenbrock method.

Note 2: turning off automatic differentiation tends to have a very minimal
performance impact (for this use case, because it's forward mode for a
square Jacobian. This is different from optimization gradient scenarios).
However, one should be careful as some methods are more sensitive to
accurate gradients than others. Specifically, Rodas methods like `Rodas4`
and `Rodas5P` require accurate Jacobians in order to have good convergence,
while many other methods like BDF (`QNDF`, `FBDF`), SDIRK (`KenCarp4`),
and Rosenbrock-W (`Rosenbrock23`) do not. Thus if using an algorithm which
is sensitive to autodiff and solving at a low tolerance, please change the
algorithm as well.

MethodError: no method matching Float64(::ForwardDiff.Dual{ForwardDiff.Tag{DiffEqBase.OrdinaryDiffEqTag, Float64}, Float64, 1})

Closest candidates are:
  (::Type{T})(::Real, ::RoundingMode) where T<:AbstractFloat
   @ Base rounding.jl:207
  (::Type{T})(::T) where T<:Number
   @ Core boot.jl:792
  Float64(::IrrationalConstants.Twoπ)
   @ IrrationalConstants C:\Users\Matt Bossart\.julia\packages\IrrationalConstants\vp5v4\src\macro.jl:112
  ...

Stacktrace:
  [1] derivative!(df::Vector{…}, f::SciMLBase.TimeGradientWrapper{…}, x::Float64, fx::Vector{…}, integrator::OrdinaryDiffEq.ODEIntegrator{…}, grad_config::Vector{…})
    @ OrdinaryDiffEq C:\Users\Matt Bossart\.julia\packages\OrdinaryDiffEq\HQ92J\src\derivative_wrappers.jl:97
  [2] calc_tderivative!(integrator::OrdinaryDiffEq.ODEIntegrator{…}, cache::OrdinaryDiffEq.Rodas4Cache{…}, dtd1::Float64, repeat_step::Bool)
    @ OrdinaryDiffEq C:\Users\Matt Bossart\.julia\packages\OrdinaryDiffEq\HQ92J\src\derivative_utils.jl:33
  [3] calc_rosenbrock_differentiation!
    @ C:\Users\Matt Bossart\.julia\packages\OrdinaryDiffEq\HQ92J\src\derivative_utils.jl:818 [inlined]
  [4] perform_step!(integrator::OrdinaryDiffEq.ODEIntegrator{…}, cache::OrdinaryDiffEq.Rodas4Cache{…}, repeat_step::Bool)
    @ OrdinaryDiffEq C:\Users\Matt Bossart\.julia\packages\OrdinaryDiffEq\HQ92J\src\perform_step\rosenbrock_perform_step.jl:1891
  [5] perform_step!
    @ C:\Users\Matt Bossart\.julia\packages\OrdinaryDiffEq\HQ92J\src\perform_step\rosenbrock_perform_step.jl:1856 [inlined]
  [6] solve!(integrator::OrdinaryDiffEq.ODEIntegrator{…})
    @ OrdinaryDiffEq C:\Users\Matt Bossart\.julia\packages\OrdinaryDiffEq\HQ92J\src\solve.jl:557
  [7] #__solve#560
    @ C:\Users\Matt Bossart\.julia\packages\OrdinaryDiffEq\HQ92J\src\solve.jl:7 [inlined]
  [8] __solve
    @ C:\Users\Matt Bossart\.julia\packages\OrdinaryDiffEq\HQ92J\src\solve.jl:1 [inlined]
  [9] solve_call(_prob::ODEProblem{…}, args::Rodas4{…}; merge_callbacks::Bool, kwargshandle::Nothing, kwargs::@Kwargs{…})
    @ DiffEqBase C:\Users\Matt Bossart\.julia\packages\DiffEqBase\c8MAQ\src\solve.jl:612
 [10] solve_up(prob::ODEProblem{…}, sensealg::Nothing, u0::Vector{…}, p::Vector{…}, args::Rodas4{…}; kwargs::@Kwargs{…})
    @ DiffEqBase C:\Users\Matt Bossart\.julia\packages\DiffEqBase\c8MAQ\src\solve.jl:1080
 [11] solve_up
    @ C:\Users\Matt Bossart\.julia\packages\DiffEqBase\c8MAQ\src\solve.jl:1066 [inlined]
 [12] solve(prob::ODEProblem{…}, args::Rodas4{…}; sensealg::Nothing, u0::Nothing, p::Nothing, wrap::Val{…}, kwargs::@Kwargs{…})
    @ DiffEqBase C:\Users\Matt Bossart\.julia\packages\DiffEqBase\c8MAQ\src\solve.jl:1003
 [13] _adjoint_sensitivities(sol::ODESolution{…}, sensealg::InterpolatingAdjoint{…}, alg::Rodas4{…}; t::StepRangeLen{…}, dgdu_discrete::Function, dgdp_discrete::Nothing, dgdu_continuous::Nothing, dgdp_continuous::Nothing, g::Nothing, abstol::Float64, reltol::Float64, checkpoints::Vector{…}, corfunc_analytical::Nothing, callback::Nothing, kwargs::@Kwargs{…})
    @ SciMLSensitivity C:\Users\Matt Bossart\.julia\packages\SciMLSensitivity\4YtYh\src\sensitivity_interface.jl:448
 [14] _adjoint_sensitivities
    @ C:\Users\Matt Bossart\.julia\packages\SciMLSensitivity\4YtYh\src\sensitivity_interface.jl:405 [inlined]
 [15] adjoint_sensitivities(sol::ODESolution{…}, args::Rodas4{…}; sensealg::InterpolatingAdjoint{…}, verbose::Bool, kwargs::@Kwargs{…})
    @ SciMLSensitivity C:\Users\Matt Bossart\.julia\packages\SciMLSensitivity\4YtYh\src\sensitivity_interface.jl:397
 [16] (::SciMLSensitivity.var"#adjoint_sensitivity_backpass#308"{…})(Δ::ODESolution{…})
    @ SciMLSensitivity C:\Users\Matt Bossart\.julia\packages\SciMLSensitivity\4YtYh\src\concrete_solve.jl:625
 [17] ZBack
    @ C:\Users\Matt Bossart\.julia\packages\Zygote\nsBv0\src\compiler\chainrules.jl:211 [inlined]
 [18] (::Zygote.var"#kw_zpullback#53"{…})(dy::ODESolution{…})
    @ Zygote C:\Users\Matt Bossart\.julia\packages\Zygote\nsBv0\src\compiler\chainrules.jl:237
 [19] #291
    @ C:\Users\Matt Bossart\.julia\packages\Zygote\nsBv0\src\lib\lib.jl:206 [inlined]
 [20] (::Zygote.var"#2169#back#293"{…})(Δ::ODESolution{…})
    @ Zygote C:\Users\Matt Bossart\.julia\packages\ZygoteRules\M4xmc\src\adjoint.jl:72
 [21] #solve#51
    @ C:\Users\Matt Bossart\.julia\packages\DiffEqBase\c8MAQ\src\solve.jl:1003 [inlined]
 [22] (::Zygote.Pullback{…})(Δ::ODESolution{…})
    @ Zygote C:\Users\Matt Bossart\.julia\packages\Zygote\nsBv0\src\compiler\interface2.jl:0
 [23] #291
    @ C:\Users\Matt Bossart\.julia\packages\Zygote\nsBv0\src\lib\lib.jl:206 [inlined]
 [24] (::Zygote.var"#2169#back#293"{…})(Δ::ODESolution{…})
    @ Zygote C:\Users\Matt Bossart\.julia\packages\ZygoteRules\M4xmc\src\adjoint.jl:72
 [25] solve
    @ C:\Users\Matt Bossart\.julia\packages\DiffEqBase\c8MAQ\src\solve.jl:993 [inlined]
 [26] (::Zygote.Pullback{…})(Δ::ODESolution{…})
    @ Zygote C:\Users\Matt Bossart\.julia\packages\Zygote\nsBv0\src\compiler\interface2.jl:0
 [27] loss
    @ c:\Users\Matt Bossart\OneDrive - UCB-O365\Desktop\temp_for_mwe\mwe_checkpointing\mwe.jl:18 [inlined]
 [28] (::Zygote.Pullback{Tuple{…}, Tuple{…}})(Δ::Float64)
    @ Zygote C:\Users\Matt Bossart\.julia\packages\Zygote\nsBv0\src\compiler\interface2.jl:0
 [29] #63
    @ c:\Users\Matt Bossart\OneDrive - UCB-O365\Desktop\temp_for_mwe\mwe_checkpointing\mwe.jl:26 [inlined]
 [30] (::Zygote.Pullback{Tuple{…}, Tuple{…}})(Δ::Float64)
    @ Zygote C:\Users\Matt Bossart\.julia\packages\Zygote\nsBv0\src\compiler\interface2.jl:0
 [31] (::Zygote.var"#75#76"{Zygote.Pullback{Tuple{…}, Tuple{…}}})(Δ::Float64)
    @ Zygote C:\Users\Matt Bossart\.julia\packages\Zygote\nsBv0\src\compiler\interface.jl:91
 [32] gradient(f::Function, args::Vector{Float64})
    @ Zygote C:\Users\Matt Bossart\.julia\packages\Zygote\nsBv0\src\compiler\interface.jl:148
 [33] top-level scope
    @ c:\Users\Matt Bossart\OneDrive - UCB-O365\Desktop\temp_for_mwe\mwe_checkpointing\mwe.jl:26

caused by: MethodError: no method matching Float64(::ForwardDiff.Dual{ForwardDiff.Tag{DiffEqBase.OrdinaryDiffEqTag, Float64}, Float64, 1})

Closest candidates are:
  (::Type{T})(::Real, ::RoundingMode) where T<:AbstractFloat
   @ Base rounding.jl:207
  (::Type{T})(::T) where T<:Number
   @ Core boot.jl:792
  Float64(::IrrationalConstants.Twoπ)
   @ IrrationalConstants C:\Users\Matt Bossart\.julia\packages\IrrationalConstants\vp5v4\src\macro.jl:112
  ...

Stacktrace:
  [1] convert(::Type{Float64}, x::ForwardDiff.Dual{ForwardDiff.Tag{DiffEqBase.OrdinaryDiffEqTag, Float64}, Float64, 1})
    @ Base .\number.jl:7
  [2] setindex!(A::Vector{Float64}, x::ForwardDiff.Dual{ForwardDiff.Tag{…}, Float64, 1}, i1::Int64)
    @ Base .\array.jl:1021
  [3] macro expansion
    @ C:\Users\Matt Bossart\.julia\packages\FastBroadcast\lSD0E\src\FastBroadcast.jl:162 [inlined]
  [4] macro expansion
    @ .\simdloop.jl:77 [inlined]
  [5] __fast_materialize!
    @ C:\Users\Matt Bossart\.julia\packages\FastBroadcast\lSD0E\src\FastBroadcast.jl:161 [inlined]
  [6] _fast_materialize!
    @ C:\Users\Matt Bossart\.julia\packages\FastBroadcast\lSD0E\src\FastBroadcast.jl:195 [inlined]
  [7] fast_materialize!
    @ C:\Users\Matt Bossart\.julia\packages\FastBroadcast\lSD0E\src\FastBroadcast.jl:276 [inlined]
  [8] _ode_interpolant!(out::Vector{…}, Θ::ForwardDiff.Dual{…}, dt::Float64, y₀::Vector{…}, y₁::Vector{…}, k::Vector{…}, cache::OrdinaryDiffEq.Rodas4ConstantCache{…}, idxs::Nothing, T::Type{…}, differential_vars::Nothing)
    @ OrdinaryDiffEq C:\Users\Matt Bossart\.julia\packages\OrdinaryDiffEq\HQ92J\src\dense\rosenbrock_interpolants.jl:167
  [9] ode_interpolant!
    @ C:\Users\Matt Bossart\.julia\packages\OrdinaryDiffEq\HQ92J\src\dense\generic_dense.jl:961 [inlined]
 [10] ode_interpolation!(out::Vector{…}, tval::ForwardDiff.Dual{…}, id::OrdinaryDiffEq.InterpolationData{…}, idxs::Nothing, deriv::Type{…}, p::Vector{…}, continuity::Symbol)
    @ OrdinaryDiffEq C:\Users\Matt Bossart\.julia\packages\OrdinaryDiffEq\HQ92J\src\dense\generic_dense.jl:897
 [11] InterpolationData
    @ C:\Users\Matt Bossart\.julia\packages\OrdinaryDiffEq\HQ92J\src\interp_func.jl:172 [inlined]
 [12] #_#474
    @ C:\Users\Matt Bossart\.julia\packages\SciMLBase\rR75x\src\solutions\ode_solutions.jl:181 [inlined]
 [13] AbstractODESolution (repeats 2 times)
    @ C:\Users\Matt Bossart\.julia\packages\SciMLBase\rR75x\src\solutions\ode_solutions.jl:179 [inlined]
 [14] split_states(du::Vector{…}, u::Vector{…}, t::ForwardDiff.Dual{…}, S::SciMLSensitivity.ODEInterpolatingAdjointSensitivityFunction{…}; update::Bool)
    @ SciMLSensitivity C:\Users\Matt Bossart\.julia\packages\SciMLSensitivity\4YtYh\src\interpolating_adjoint.jl:237
 [15] split_states
    @ C:\Users\Matt Bossart\.julia\packages\SciMLSensitivity\4YtYh\src\interpolating_adjoint.jl:162 [inlined]
 [16] (::SciMLSensitivity.ODEInterpolatingAdjointSensitivityFunction{…})(du::Vector{…}, u::Vector{…}, p::Vector{…}, t::ForwardDiff.Dual{…})
    @ SciMLSensitivity C:\Users\Matt Bossart\.julia\packages\SciMLSensitivity\4YtYh\src\interpolating_adjoint.jl:125
 [17] ODEFunction
    @ C:\Users\Matt Bossart\.julia\packages\SciMLBase\rR75x\src\scimlfunctions.jl:2297 [inlined]
 [18] TimeGradientWrapper
    @ C:\Users\Matt Bossart\.julia\packages\SciMLBase\rR75x\src\function_wrappers.jl:17 [inlined]
 [19] derivative!(df::Vector{…}, f::SciMLBase.TimeGradientWrapper{…}, x::Float64, fx::Vector{…}, integrator::OrdinaryDiffEq.ODEIntegrator{…}, grad_config::Vector{…})
    @ OrdinaryDiffEq C:\Users\Matt Bossart\.julia\packages\OrdinaryDiffEq\HQ92J\src\derivative_wrappers.jl:95
 [20] calc_tderivative!(integrator::OrdinaryDiffEq.ODEIntegrator{…}, cache::OrdinaryDiffEq.Rodas4Cache{…}, dtd1::Float64, repeat_step::Bool)
    @ OrdinaryDiffEq C:\Users\Matt Bossart\.julia\packages\OrdinaryDiffEq\HQ92J\src\derivative_utils.jl:33
 [21] calc_rosenbrock_differentiation!
    @ C:\Users\Matt Bossart\.julia\packages\OrdinaryDiffEq\HQ92J\src\derivative_utils.jl:818 [inlined]
 [22] perform_step!(integrator::OrdinaryDiffEq.ODEIntegrator{…}, cache::OrdinaryDiffEq.Rodas4Cache{…}, repeat_step::Bool)
    @ OrdinaryDiffEq C:\Users\Matt Bossart\.julia\packages\OrdinaryDiffEq\HQ92J\src\perform_step\rosenbrock_perform_step.jl:1891
 [23] perform_step!
    @ C:\Users\Matt Bossart\.julia\packages\OrdinaryDiffEq\HQ92J\src\perform_step\rosenbrock_perform_step.jl:1856 [inlined]
 [24] solve!(integrator::OrdinaryDiffEq.ODEIntegrator{…})
    @ OrdinaryDiffEq C:\Users\Matt Bossart\.julia\packages\OrdinaryDiffEq\HQ92J\src\solve.jl:557
 [25] #__solve#560
    @ C:\Users\Matt Bossart\.julia\packages\OrdinaryDiffEq\HQ92J\src\solve.jl:7 [inlined]
 [26] __solve
    @ C:\Users\Matt Bossart\.julia\packages\OrdinaryDiffEq\HQ92J\src\solve.jl:1 [inlined]
 [27] solve_call(_prob::ODEProblem{…}, args::Rodas4{…}; merge_callbacks::Bool, kwargshandle::Nothing, kwargs::@Kwargs{…})
    @ DiffEqBase C:\Users\Matt Bossart\.julia\packages\DiffEqBase\c8MAQ\src\solve.jl:612
 [28] solve_up(prob::ODEProblem{…}, sensealg::Nothing, u0::Vector{…}, p::Vector{…}, args::Rodas4{…}; kwargs::@Kwargs{…})
    @ DiffEqBase C:\Users\Matt Bossart\.julia\packages\DiffEqBase\c8MAQ\src\solve.jl:1080
 [29] solve_up
    @ C:\Users\Matt Bossart\.julia\packages\DiffEqBase\c8MAQ\src\solve.jl:1066 [inlined]
 [30] solve(prob::ODEProblem{…}, args::Rodas4{…}; sensealg::Nothing, u0::Nothing, p::Nothing, wrap::Val{…}, kwargs::@Kwargs{…})
    @ DiffEqBase C:\Users\Matt Bossart\.julia\packages\DiffEqBase\c8MAQ\src\solve.jl:1003
 [31] _adjoint_sensitivities(sol::ODESolution{…}, sensealg::InterpolatingAdjoint{…}, alg::Rodas4{…}; t::StepRangeLen{…}, dgdu_discrete::Function, dgdp_discrete::Nothing, dgdu_continuous::Nothing, dgdp_continuous::Nothing, g::Nothing, abstol::Float64, reltol::Float64, checkpoints::Vector{…}, corfunc_analytical::Nothing, callback::Nothing, kwargs::@Kwargs{…})
    @ SciMLSensitivity C:\Users\Matt Bossart\.julia\packages\SciMLSensitivity\4YtYh\src\sensitivity_interface.jl:448
 [32] _adjoint_sensitivities
    @ C:\Users\Matt Bossart\.julia\packages\SciMLSensitivity\4YtYh\src\sensitivity_interface.jl:405 [inlined]
 [33] adjoint_sensitivities(sol::ODESolution{…}, args::Rodas4{…}; sensealg::InterpolatingAdjoint{…}, verbose::Bool, kwargs::@Kwargs{…})
    @ SciMLSensitivity C:\Users\Matt Bossart\.julia\packages\SciMLSensitivity\4YtYh\src\sensitivity_interface.jl:397
 [34] (::SciMLSensitivity.var"#adjoint_sensitivity_backpass#308"{…})(Δ::ODESolution{…})
    @ SciMLSensitivity C:\Users\Matt Bossart\.julia\packages\SciMLSensitivity\4YtYh\src\concrete_solve.jl:625
 [35] ZBack
    @ C:\Users\Matt Bossart\.julia\packages\Zygote\nsBv0\src\compiler\chainrules.jl:211 [inlined]
 [36] (::Zygote.var"#kw_zpullback#53"{…})(dy::ODESolution{…})
    @ Zygote C:\Users\Matt Bossart\.julia\packages\Zygote\nsBv0\src\compiler\chainrules.jl:237
 [37] #291
    @ C:\Users\Matt Bossart\.julia\packages\Zygote\nsBv0\src\lib\lib.jl:206 [inlined]
 [38] (::Zygote.var"#2169#back#293"{…})(Δ::ODESolution{…})
    @ Zygote C:\Users\Matt Bossart\.julia\packages\ZygoteRules\M4xmc\src\adjoint.jl:72
 [39] #solve#51
    @ C:\Users\Matt Bossart\.julia\packages\DiffEqBase\c8MAQ\src\solve.jl:1003 [inlined]
 [40] (::Zygote.Pullback{…})(Δ::ODESolution{…})
    @ Zygote C:\Users\Matt Bossart\.julia\packages\Zygote\nsBv0\src\compiler\interface2.jl:0
 [41] #291
    @ C:\Users\Matt Bossart\.julia\packages\Zygote\nsBv0\src\lib\lib.jl:206 [inlined]
 [42] (::Zygote.var"#2169#back#293"{…})(Δ::ODESolution{…})
    @ Zygote C:\Users\Matt Bossart\.julia\packages\ZygoteRules\M4xmc\src\adjoint.jl:72
 [43] solve
    @ C:\Users\Matt Bossart\.julia\packages\DiffEqBase\c8MAQ\src\solve.jl:993 [inlined]
 [44] (::Zygote.Pullback{…})(Δ::ODESolution{…})
    @ Zygote C:\Users\Matt Bossart\.julia\packages\Zygote\nsBv0\src\compiler\interface2.jl:0
 [45] loss
    @ c:\Users\Matt Bossart\OneDrive - UCB-O365\Desktop\temp_for_mwe\mwe_checkpointing\mwe.jl:18 [inlined]
 [46] (::Zygote.Pullback{Tuple{…}, Tuple{…}})(Δ::Float64)
    @ Zygote C:\Users\Matt Bossart\.julia\packages\Zygote\nsBv0\src\compiler\interface2.jl:0
 [47] #63
    @ c:\Users\Matt Bossart\OneDrive - UCB-O365\Desktop\temp_for_mwe\mwe_checkpointing\mwe.jl:26 [inlined]
 [48] (::Zygote.Pullback{Tuple{…}, Tuple{…}})(Δ::Float64)
    @ Zygote C:\Users\Matt Bossart\.julia\packages\Zygote\nsBv0\src\compiler\interface2.jl:0
 [49] (::Zygote.var"#75#76"{Zygote.Pullback{Tuple{…}, Tuple{…}}})(Δ::Float64)
    @ Zygote C:\Users\Matt Bossart\.julia\packages\Zygote\nsBv0\src\compiler\interface.jl:91
 [50] gradient(f::Function, args::Vector{Float64})
    @ Zygote C:\Users\Matt Bossart\.julia\packages\Zygote\nsBv0\src\compiler\interface.jl:148

Environment (please complete the following information):

• Output of using Pkg; Pkg.status()

  [7da242da] Enzyme v0.12.21
  [1dea7af3] OrdinaryDiffEq v6.85.0
  [1ed8b502] SciMLSensitivity v7.62.0
  [e88e6eb3] Zygote v0.6.70

• Output of using Pkg; Pkg.status(; mode = PKGMODE_MANIFEST)

  [47edcb42] ADTypes v1.5.3
  [621f4979] AbstractFFTs v1.5.0
  [7d9f7c33] Accessors v0.1.36
  [79e6a3ab] Adapt v4.0.4
  [66dad0bd] AliasTables v1.1.3
  [ec485272] ArnoldiMethod v0.4.0
  [4fba245c] ArrayInterface v7.11.0
  [4c555306] ArrayLayouts v1.10.0
  [a9b6321e] Atomix v0.1.0
  [62783981] BitTwiddlingConvenienceFunctions v0.1.6
  [fa961155] CEnum v0.5.0
  [2a0fbf3d] CPUSummary v0.2.6
  [49dc2e85] Calculus v0.5.1
  [7057c7e9] Cassette v0.3.13
  [082447d4] ChainRules v1.69.0
  [d360d2e6] ChainRulesCore v1.24.0
  [fb6a15b2] CloseOpenIntervals v0.1.13
  [38540f10] CommonSolve v0.2.4
  [bbf7d656] CommonSubexpressions v0.3.0
  [f70d9fcc] CommonWorldInvalidations v1.0.0
  [34da2185] Compat v4.15.0
  [a33af91c] CompositionsBase v0.1.2
  [2569d6c7] ConcreteStructs v0.2.3
  [187b0558] ConstructionBase v1.5.5
  [adafc99b] CpuId v0.3.1
  [9a962f9c] DataAPI v1.16.0
  [864edb3b] DataStructures v0.18.20
  [e2d170a0] DataValueInterfaces v1.0.0
  [2b5f629d] DiffEqBase v6.151.5
  [459566f4] DiffEqCallbacks v3.6.2
  [77a26b50] DiffEqNoiseProcess v5.21.0
  [163ba53b] DiffResults v1.1.0
  [b552c78f] DiffRules v1.15.1
  [a0c0ee7d] DifferentiationInterface v0.5.7
  [b4f34e82] Distances v0.10.11
  [31c24e10] Distributions v0.25.109
  [ffbed154] DocStringExtensions v0.9.3
  [fa6b7ba4] DualNumbers v0.6.8
  [da5c29d0] EllipsisNotation v1.8.0
  [4e289a0a] EnumX v1.0.4
  [7da242da] Enzyme v0.12.21
  [f151be2c] EnzymeCore v0.7.6
  [d4d017d3] ExponentialUtilities v1.26.1
  [e2ba6199] ExprTools v0.1.10
  [7034ab61] FastBroadcast v0.3.4
  [9aa1b823] FastClosures v0.3.2
  [29a986be] FastLapackInterface v2.0.4
  [1a297f60] FillArrays v1.11.0
  [6a86dc24] FiniteDiff v2.23.1
  [f6369f11] ForwardDiff v0.10.36
  [f62d2435] FunctionProperties v0.1.2
  [069b7b12] FunctionWrappers v1.1.3
  [77dc65aa] FunctionWrappersWrappers v0.1.3
  [d9f16b24] Functors v0.4.11
  [0c68f7d7] GPUArrays v10.2.3
  [46192b85] GPUArraysCore v0.1.6
  [61eb1bfa] GPUCompiler v0.26.6
  [c145ed77] GenericSchur v0.5.4
  [86223c79] Graphs v1.11.2
  [3e5b6fbb] HostCPUFeatures v0.1.17
  [34004b35] HypergeometricFunctions v0.3.23
  [7869d1d1] IRTools v0.4.14
  [615f187c] IfElse v0.1.1
  [d25df0c9] Inflate v0.1.5
  [3587e190] InverseFunctions v0.1.14
  [92d709cd] IrrationalConstants v0.2.2
  [82899510] IteratorInterfaceExtensions v1.0.0
  [692b3bcd] JLLWrappers v1.5.0
  [ccbc3e58] JumpProcesses v9.11.1
  [ef3ab10e] KLU v0.6.0
  [63c18a36] KernelAbstractions v0.9.22
  [ba0b0d4f] Krylov v0.9.6
  [929cbde3] LLVM v8.0.0
  [10f19ff3] LayoutPointers v0.1.17
  [5078a376] LazyArrays v2.1.2
  [2d8b4e74] LevyArea v1.0.0
  [d3d80556] LineSearches v7.2.0
  [7ed4a6bd] LinearSolve v2.30.2
  [2ab3a3ac] LogExpFunctions v0.3.28
  [bdcacae8] LoopVectorization v0.12.171
  [1914dd2f] MacroTools v0.5.13
  [d125e4d3] ManualMemory v0.1.8
  [bb5d69b7] MaybeInplace v0.1.3
  [e1d29d7a] Missings v1.2.0
  [46d2c3a1] MuladdMacro v0.2.4
  [d41bc354] NLSolversBase v7.8.3
  [2774e3e8] NLsolve v4.5.1
  [872c559c] NNlib v0.9.18
  [77ba4419] NaNMath v1.0.2
  [8913a72c] NonlinearSolve v3.13.1
  [d8793406] ObjectFile v0.4.1
  [6fe1bfb0] OffsetArrays v1.14.0
  [429524aa] Optim v1.9.4
  [3bd65402] Optimisers v0.3.3
  [bac558e1] OrderedCollections v1.6.3
  [1dea7af3] OrdinaryDiffEq v6.85.0
  [90014a1f] PDMats v0.11.31
  [65ce6f38] PackageExtensionCompat v1.0.2
  [d96e819e] Parameters v0.12.3
  [e409e4f3] PoissonRandom v0.4.4
  [f517fe37] Polyester v0.7.15
  [1d0040c9] PolyesterWeave v0.2.2
  [85a6dd25] PositiveFactorizations v0.2.4
  [d236fae5] PreallocationTools v0.4.22
  [aea7be01] PrecompileTools v1.2.1
  [21216c6a] Preferences v1.4.3
  [43287f4e] PtrArrays v1.2.0
  [1fd47b50] QuadGK v2.9.4
  [74087812] Random123 v1.7.0
  [e6cf234a] RandomNumbers v1.5.3
  [c1ae055f] RealDot v0.1.0
  [3cdcf5f2] RecipesBase v1.3.4
  [731186ca] RecursiveArrayTools v3.23.1
  [f2c3362d] RecursiveFactorization v0.2.23
  [189a3867] Reexport v1.2.2
  [ae029012] Requires v1.3.0
  [ae5879a3] ResettableStacks v1.1.1
  [37e2e3b7] ReverseDiff v1.15.3
  [79098fc4] Rmath v0.7.1
  [7e49a35a] RuntimeGeneratedFunctions v0.5.13
  [94e857df] SIMDTypes v0.1.0
  [476501e8] SLEEFPirates v0.6.43
  [0bca4576] SciMLBase v2.42.0
  [c0aeaf25] SciMLOperators v0.3.8
  [1ed8b502] SciMLSensitivity v7.62.0
  [53ae85a6] SciMLStructures v1.4.1
  [6c6a2e73] Scratch v1.2.1
  [efcf1570] Setfield v1.1.1
  [727e6d20] SimpleNonlinearSolve v1.10.1
  [699a6c99] SimpleTraits v0.9.4
  [ce78b400] SimpleUnPack v1.1.0
  [a2af1166] SortingAlgorithms v1.2.1
  [47a9eef4] SparseDiffTools v2.19.0
  [dc90abb0] SparseInverseSubset v0.1.2
  [0a514795] SparseMatrixColorings v0.3.3
  [e56a9233] Sparspak v0.3.9
  [276daf66] SpecialFunctions v2.4.0
  [aedffcd0] Static v1.1.0
  [0d7ed370] StaticArrayInterface v1.5.1
  [90137ffa] StaticArrays v1.9.7
  [1e83bf80] StaticArraysCore v1.4.3
  [82ae8749] StatsAPI v1.7.0
  [2913bbd2] StatsBase v0.34.3
  [4c63d2b9] StatsFuns v1.3.1
  [789caeaf] StochasticDiffEq v6.66.0
  [7792a7ef] StrideArraysCore v0.5.7
  [09ab397b] StructArrays v0.6.18
  [53d494c1] StructIO v0.3.0
  [2efcf032] SymbolicIndexingInterface v0.3.22
  [3783bdb8] TableTraits v1.0.1
  [bd369af6] Tables v1.11.1
  [8290d209] ThreadingUtilities v0.5.2
  [a759f4b9] TimerOutputs v0.5.24
  [9f7883ad] Tracker v0.2.34
  [d5829a12] TriangularSolve v0.2.1
  [410a4b4d] Tricks v0.1.8
  [781d530d] TruncatedStacktraces v1.4.0
  [3a884ed6] UnPack v1.0.2
  [013be700] UnsafeAtomics v0.2.1
  [d80eeb9a] UnsafeAtomicsLLVM v0.1.5
  [3d5dd08c] VectorizationBase v0.21.70
  [19fa3120] VertexSafeGraphs v0.2.0
  [e88e6eb3] Zygote v0.6.70
  [700de1a5] ZygoteRules v0.2.5
⌅ [7cc45869] Enzyme_jll v0.0.128+0
  [1d5cc7b8] IntelOpenMP_jll v2024.1.0+0
  [dad2f222] LLVMExtra_jll v0.0.30+0
  [856f044c] MKL_jll v2024.1.0+0
  [efe28fd5] OpenSpecFun_jll v0.5.5+0
  [f50d1b31] Rmath_jll v0.4.2+0
  [1317d2d5] oneTBB_jll v2021.12.0+0
  [0dad84c5] ArgTools v1.1.1
  [56f22d72] Artifacts
  [2a0f44e3] Base64
  [ade2ca70] Dates
  [8ba89e20] Distributed
  [f43a241f] Downloads v1.6.0
  [7b1f6079] FileWatching
  [9fa8497b] Future
  [b77e0a4c] InteractiveUtils
  [4af54fe1] LazyArtifacts
  [b27032c2] LibCURL v0.6.4
  [76f85450] LibGit2
  [8f399da3] Libdl
  [37e2e46d] LinearAlgebra
  [56ddb016] Logging
  [d6f4376e] Markdown
  [a63ad114] Mmap
  [ca575930] NetworkOptions v1.2.0
  [44cfe95a] Pkg v1.10.0
  [de0858da] Printf
  [3fa0cd96] REPL
  [9a3f8284] Random
  [ea8e919c] SHA v0.7.0
  [9e88b42a] Serialization
  [1a1011a3] SharedArrays
  [6462fe0b] Sockets
  [2f01184e] SparseArrays v1.10.0
  [10745b16] Statistics v1.10.0
  [4607b0f0] SuiteSparse
  [fa267f1f] TOML v1.0.3
  [a4e569a6] Tar v1.10.0
  [8dfed614] Test
  [cf7118a7] UUIDs
  [4ec0a83e] Unicode
  [e66e0078] CompilerSupportLibraries_jll v1.1.1+0
  [deac9b47] LibCURL_jll v8.4.0+0
  [e37daf67] LibGit2_jll v1.6.4+0
  [29816b5a] LibSSH2_jll v1.11.0+1
  [c8ffd9c3] MbedTLS_jll v2.28.2+1
  [14a3606d] MozillaCACerts_jll v2023.1.10
  [4536629a] OpenBLAS_jll v0.3.23+4
  [05823500] OpenLibm_jll v0.8.1+2
  [bea87d4a] SuiteSparse_jll v7.2.1+1
  [83775a58] Zlib_jll v1.2.13+1
  [8e850b90] libblastrampoline_jll v5.8.0+1
  [8e850ede] nghttp2_jll v1.52.0+1
  [3f19e933] p7zip_jll v17.4.0+2

• Output of versioninfo()

Julia Version 1.10.4
Commit 48d4fd4843 (2024-06-04 10:41 UTC)
Build Info:
  Official https://julialang.org/ release
Platform Info:
  OS: Windows (x86_64-w64-mingw32)
  CPU: 16 × 11th Gen Intel(R) Core(TM) i7-11800H @ 2.30GHz
  WORD_SIZE: 64
  LIBM: libopenlibm
  LLVM: libLLVM-15.0.7 (ORCJIT, tigerlake)
Threads: 1 default, 0 interactive, 1 GC (on 16 virtual cores)
Environment:
  JULIA_EDITOR = code
  JULIA_NUM_THREADS =