Parameter estimation with one data point

[moesphere]

It seems that parameter estimation with one data point does not work as the corresponding loop through the solution starts at index = 2: here

Partially related to this issue

using DiffEqParamEstim, DifferentialEquations
function f(du, u, p, t)
    du[1] = dx = p[1] * u[1] - u[1] * u[2]
    du[2] = dy = -3 * u[2] + u[1] * u[2]
end
  
u0 = [1.0; 1.0]
tspan = (0.0, 10.0)
p = [1.5]
prob = ODEProblem(f, u0, tspan, p)
loss_objective = build_loss_objective(prob, Tsit5(), L2Loss([1.], ones(2)))
loss_objective(p) # returns zero
sum((solve(prob)(1.0)-ones(2)).^2) # should return this instead

[CedricLhoussaine]

I confirm this issue: the parameter estimation always skip the first data point.

[ChrisRackauckas]

oh interesting, it would be good to just change that to one. If the initial condition matches, then it's just zero so it won't change that case.

[ChrisRackauckas]

Fixed by #157