Releases: SciML/DiffEqFlux.jl
Releases · SciML/DiffEqFlux.jl
v0.10.1
v0.10.0
v0.10.0 (2019-12-14)
Merged pull requests:
- CompatHelper: bump compat for "Requires" to "1.0" (#95) (github-actions[bot])
v0.9.0
v0.9.0 (2019-12-10)
Merged pull requests:
- Update DiffEqSensitivity bound (#94) (ChrisRackauckas)
- CompatHelper: bump compat for "DiffResults" to "1.0" (#92) (github-actions[bot])
v0.8.1
v0.8.1 (2019-11-30)
Closed issues:
- Preprocessing Layer in Neural ODE causes Error (maybe Tracker related). (#90)
- Compilation problem (#89)
Merged pull requests:
- Added separate Callbacks for the forward and adjoint solves (#84) (PiotrSokol)
v0.8.0
v0.8.0 (2019-11-18)
Closed issues:
- Forward solve fails with LinearAlgebra.Adjoint initial values (#85)
Merged pull requests:
- Remove plotting from tests (#87) (ChrisRackauckas)
v0.7.0
v0.7.0 (2019-10-20)
Closed issues:
- MethodError: no method matching Float64(::Tracker.TrackedReal{Float64}) when using Array instead of Vector as parameters (#82)
- ERROR: LoadError: MethodError: objects of type Tracker.Params are not callable - partial_neural.jl (#77)
- pictures (#67)
- Derivative w.r.t initial conditions with saveat for end point (#25)
- Scalar saveat with neural ODE has a type issue (#12)
Merged pull requests:
- allow arbitrary parameter sizing (#83) (ChrisRackauckas)
- fixed typo in Mixed Neural DEs and minor editing (#78) (scheidan)
- Replace Flux with Tracker (#76) (ChrisRackauckas)
- Update README.md (#72) (ndalton12)
- Fix neural DDE in README (#71) (devmotion)
- add GDP regression test (#69) (ChrisRackauckas)
v0.6.1
v0.6.1 (2019-07-21)
Merged pull requests:
- make save_end=true by default in adjoint (#68) (ChrisRackauckas)
v0.6.0
v0.5.2
v0.5.2 (2019-07-17)
Closed issues:
- "MethodError: no method matching gemm!" error when using Rosenbrock23 with diffeq_rd. (#55)
Merged pull requests:
- add gpu tests (#60) (ChrisRackauckas)
- Minor edits to neural ode example in Readme (#59) (Vaibhavdixit02)
- greatly improve tests (#58) (ChrisRackauckas)