Comments (3)
I pushed it along and got pretty far:
using DynamicQuantities, OrdinaryDiffEq, RecursiveArrayTools
function RecursiveArrayTools.recursive_unitless_bottom_eltype(a::Type{
<:DynamicQuantities.Quantity{T}
}) where T
T
end
function RecursiveArrayTools.recursive_unitless_eltype(a::Type{<:DynamicQuantities.Quantity{T}}) where T
T
end
DiffEqBase.value(x::DynamicQuantities.Quantity) = x.value
@inline function DiffEqBase.UNITLESS_ABS2(x::AbstractArray)
mapreduce(DiffEqBase.UNITLESS_ABS2, DiffEqBase.abs2_and_sum, x, init = zero(real(first(DiffEqBase.value(x)))))
end
@inline function DiffEqBase.UNITLESS_ABS2(x::DynamicQuantities.Quantity)
abs(DiffEqBase.value(x))
end
function DiffEqBase.abs2_and_sum(x::DynamicQuantities.Quantity, y::Float64)
reduce(Base.add_sum, DiffEqBase.value(x), init = zero(real(DiffEqBase.value(x)))) +
reduce(Base.add_sum, y, init = zero(real(DiffEqBase.value(eltype(y)))))
end
DiffEqBase.recursive_length(u::Array) = length(u)
Base.sign(x::DynamicQuantities.Quantity) = Base.sign(DiffEqBase.value(x))
function DiffEqBase.prob2dtmin(prob; use_end_time = true)
DiffEqBase.prob2dtmin(prob.tspan, oneunit(first(prob.tspan)), use_end_time)
end
DiffEqBase.NAN_CHECK(x::DynamicQuantities.Quantity) = isnan(x)
Base.zero(x::Array{T}) where {T<:DynamicQuantities.Quantity} = zero.(x)
@inline function DiffEqBase.calculate_residuals(ũ, u₀, u₁, α, ρ, internalnorm, t)
@. DiffEqBase.calculate_residuals(ũ, u₀, u₁, α, ρ, internalnorm, t)
end
f(u, p, t) = u / t;
problem = ODEProblem(f, [1.0u"km/s"], (0.0u"s", 1.0u"s"));
sol = solve(problem, Tsit5(), dt = 0.1u"s")
with just one internal modification. Two interface breaks are weird though:
First one:
julia> typeof(one(0.0u"s"))
Quantity{Float64, Dimensions{DynamicQuantities.FixedRational{Int32, 25200}}}
that should just be Float64
?
Second there's something odd in brodcasting I haven't isolated yet.
from differentialequations.jl.
Thanks, nice work!
Regarding one
, see the discussion here: SymbolicML/DynamicQuantities.jl#40. This resulted in the package BaseType.jl for specifically getting the base numeric type. But maybe an interim is to allow Float64
return value, I’m not sure.
Also one alternative to this sort of modification is some of the ideas in SymbolicML/DynamicQuantities.jl#76
from differentialequations.jl.
Here is the PR to implement these changes: SymbolicML/DynamicQuantities.jl#74
So I think the missing part is switching to oneunit(::T)
and one(::T)
in OrdinaryDiffEq.jl?
julia> sol = solve(problem, Tsit5(), dt = 0.1u"s")
ERROR: Cannot create a dimensionful 1 for a `UnionAbstractQuantity` type without knowing the dimensions. Please use `oneunit(::UnionAbstractQuantity)` instead.
Stacktrace:
[1] error(s::String)
@ Base ./error.jl:35
[2] oneunit(::Type{Quantity{Float64, Dimensions{DynamicQuantities.FixedRational{Int32, 25200}}}})
@ DynamicQuantities ~/Documents/DynamicQuantities.jl/src/utils.jl:191
[3] __init(prob::ODEProblem{…}, alg::Tsit5{…}, timeseries_init::Tuple{}, ts_init::Tuple{}, ks_init::Tuple{}, recompile::Type{…}; saveat::Tuple{}, tstops::Tuple{}, d_discontinuities::Tuple{}, save_idxs::Nothing, save_everystep::Bool, save_on::Bool, save_start::Bool, save_end::Nothing, callback::Nothing, dense::Bool, calck::Bool, dt::Quantity{…}, dtmin::Nothing, dtmax::Quantity{…}, force_dtmin::Bool, adaptive::Bool, gamma::Rational{…}, abstol::Nothing, reltol::Nothing, qmin::Rational{…}, qmax::Int64, qsteady_min::Int64, qsteady_max::Int64, beta1::Nothing, beta2::Nothing, qoldinit::Rational{…}, controller::Nothing, fullnormalize::Bool, failfactor::Int64, maxiters::Int64, internalnorm::typeof(DiffEqBase.ODE_DEFAULT_NORM), internalopnorm::typeof(LinearAlgebra.opnorm), isoutofdomain::typeof(DiffEqBase.ODE_DEFAULT_ISOUTOFDOMAIN), unstable_check::typeof(DiffEqBase.ODE_DEFAULT_UNSTABLE_CHECK), verbose::Bool, timeseries_errors::Bool, dense_errors::Bool, advance_to_tstop::Bool, stop_at_next_tstop::Bool, initialize_save::Bool, progress::Bool, progress_steps::Int64, progress_name::String, progress_message::typeof(DiffEqBase.ODE_DEFAULT_PROG_MESSAGE), progress_id::Symbol, userdata::Nothing, allow_extrapolation::Bool, initialize_integrator::Bool, alias_u0::Bool, alias_du0::Bool, initializealg::OrdinaryDiffEq.DefaultInit, kwargs::@Kwargs{})
@ OrdinaryDiffEq ~/.julia/packages/OrdinaryDiffEq/qxpST/src/solve.jl:220
[4] __solve(::ODEProblem{…}, ::Tsit5{…}; kwargs::@Kwargs{…})
@ OrdinaryDiffEq ~/.julia/packages/OrdinaryDiffEq/qxpST/src/solve.jl:5
[5] solve_call(_prob::ODEProblem{…}, args::Tsit5{…}; merge_callbacks::Bool, kwargshandle::Nothing, kwargs::@Kwargs{…})
@ DiffEqBase ~/.julia/packages/DiffEqBase/NYLhl/src/solve.jl:557
[6] solve_up(prob::ODEProblem{…}, sensealg::Nothing, u0::Vector{…}, p::SciMLBase.NullParameters, args::Tsit5{…}; kwargs::@Kwargs{…})
@ DiffEqBase ~/.julia/packages/DiffEqBase/NYLhl/src/solve.jl:1006
[7] solve(prob::ODEProblem{…}, args::Tsit5{…}; sensealg::Nothing, u0::Nothing, p::Nothing, wrap::Val{…}, kwargs::@Kwargs{…})
@ DiffEqBase ~/.julia/packages/DiffEqBase/NYLhl/src/solve.jl:929
[8] top-level scope
@ REPL[21]:1
Some type information was truncated. Use `show(err)` to see complete types.
from differentialequations.jl.
Related Issues (20)
- LoadError: TypeError: in setfield! HOT 1
- oneunit(::Type{Any} HOT 1
- Failed to precompile LoadError: UndefVarError: `OperatorAssumptions` not defined HOT 2
- ImplicitEuler fails to solve a nonlinear spring-damper problem HOT 5
- Plotting ODESolution using plot(sol) fails when using ODEFunction in combination with 'syms' keyword HOT 4
- SciML UDE tutorial code breaks in Julia 1.10 HOT 1
- StackOverflowError when passing tuple of parameters containing an `ODESolution` HOT 3
- Type-instability when using nested structs HOT 2
- I/O: Example of interface to `IterableTables` errors when trying to predefine column names HOT 1
- Unable to get full matrix whose columns is the solution to an ODE with newer version of RecursiveArrayTools. HOT 2
- Unexpected behavior with `PresetTimeCallback`
- Type-instability arising from initial condition HOT 5
- Error in ContinuousCallback event handling HOT 1
- autodiff in default algorithm HOT 5
- Seems idxs not working for EnsembleSummary plot HOT 2
- DAE with continuous callback and `DImplicitEuler` results in incorrect `integrator.u` in `affect!()` HOT 1
- Failed to precompile BoundaryValueDiffEq HOT 3
- Alignment between “Recommended Methods” and the polyalgorithms HOT 3
- tstops should give meaningful error if given duals
- Permit setting `seed` for multiple stochastic ensemble simulations
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from differentialequations.jl.