I’d like to program a differential equation that runs till it reaches a particular state (e.g. one of the space variables is zero), and then train the some parameters that define the equation so the system reaches the terminal state in minimum time. My intuition is that I could use terminate! in a …

Derivative with terminate callback

ChrisRackauckas September 11, 2020, 11:53pm 2

Sorry, that was a small bug. Fixed by https://github.com/SciML/DiffEqBase.jl/pull/584 and it should be released in about 2 hours as DiffEqBase 6.47.1

Topic		Replies	Views
How do I debug this in diffeqflux? General Usage	38	1923	November 15, 2020
No method matching terminate!(::DiffEqSensitivity.FakeIntegrator General Usage flux , sciml	3	435	August 11, 2021
Why can't I terminate an integrator in DifferentialEquations? New to Julia	1	537	January 3, 2020
Types and gradients, including Forward.gradient General Usage	20	5588	December 20, 2016
DifferentialEquations.jl + Flux.jl or Knet.jl Machine Learning diffeq	15	3654	October 18, 2018