Add Gaussian Noise Process submodule #62
Conversation
|
more on submodules here: https://docs.julialang.org/en/v1/manual/modules/ |
Codecov Report
@@ Coverage Diff @@
## master #62 +/- ##
=========================================
- Coverage 92.3% 79.5% -12.8%
=========================================
Files 15 23 +8
Lines 1165 1679 +514
=========================================
+ Hits 1075 1334 +259
- Misses 90 345 +255
Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here. |
src/IonSim.jl
Outdated
| @@ -28,6 +28,17 @@ include("hamiltonians.jl") | |||
| include("time_evolution.jl") | |||
| include("species/include_species.jl") | |||
|
|
|||
| module GaussianNoiseProcess | |||
There was a problem hiding this comment.
I recommend calling the module "noise" so you can use it like "IonSim.noise"
|
|
||
| @inline wiener_randn(rng::AbstractRNG,::Type{T}) where T = randn(rng,T) | ||
|
|
||
| # == Generators of Discrete AR/MA time series == # |
There was a problem hiding this comment.
These could be docstrings, correct?
src/noise/ARMA_NoiseProcess.jl
Outdated
|
|
||
| # AR(p) Processes Step | ||
| n = length(W.u) | ||
| test = X.p >= n |
There was a problem hiding this comment.
What is test supposed to mean here?
src/noise/ARMA_NoiseProcess.jl
Outdated
| return new_val - W.u[end] | ||
| end | ||
|
|
||
| @inline function (X::MA_Process)(dW,W,dt,u,p,t,rng) |
There was a problem hiding this comment.
What does this function do that the previous one doesn't do?
src/noise/ARMA_NoiseProcess.jl
Outdated
| @inline function sinc_bridge(X,dW,W,W0,Wh,q,h,u,p,t,rng) | ||
|
|
||
| # Brownian bridge correction term (added during interpolation) | ||
| # iscontinuous(W) = true <- should fix to false |
src/noise/CARMA_NoiseProcess.jl
Outdated
| using PyCall | ||
| so = pyimport("scipy.optimize") | ||
|
|
||
| @inline wiener_randn(rng::AbstractRNG,::Type{T}) where T = randn(rng,T) |
There was a problem hiding this comment.
Do you have to define this function every time?
src/noise/CARMA_NoiseProcess.jl
Outdated
|
|
||
| @inline wiener_randn(rng::AbstractRNG,::Type{T}) where T = randn(rng,T) | ||
|
|
||
| # == ARMAProcess: Generates ARMA Coefficients and Sinc-interpolated NoiseProcess == # |
There was a problem hiding this comment.
Are there wikipedia or research paper links for how these noise processes are generated? Could we link to them in the code?
src/noise/CARMA_NoiseProcess.jl
Outdated
| # end | ||
|
|
||
| function ReImCoeffSolver(x, α, κ, σ) | ||
| t = x[1:length(x)÷2] |
src/noise/CARMA_NoiseProcess.jl
Outdated
| acf::Vector | ||
| end | ||
|
|
||
| struct CARMAProcess |
There was a problem hiding this comment.
Why is CARMA only in one file but ARMA in three files?
src/noise/CARMA_NoiseProcess.jl
Outdated
|
|
||
| # acf_iir = discrete.arma.sigma^2 .* [dot(discrete.arma.psi[1:end-i], discrete.arma.psi[1+i:end]) for i in 0:(length(discrete.arma.psi)-1)] | ||
| # ys = discrete.arma.ma_acf[ts .+ 1] | ||
| carma_α, _ = so.curve_fit((x,d...)->ReImCoeffSolver(x,collect(d),carma_κ, discrete.arma.sigma), [ts;ts], [ys;zeros(length(ys))], |
|
Goal after today: Make non-adaptive time-stepping the default |
|
Two things from Joe:
|
TODO::