From 84fe0157e408a414e35822e94116b67dcc37b3c0 Mon Sep 17 00:00:00 2001 From: musvaage Date: Wed, 15 Nov 2023 09:36:12 -0600 Subject: [PATCH] typos --- docs/src/solvers/NonlinearSystemSolvers.md | 2 +- docs/src/tutorials/getting_started.md | 6 +++--- src/dfsane.jl | 4 ++-- src/trustRegion.jl | 4 ++-- 4 files changed, 8 insertions(+), 8 deletions(-) diff --git a/docs/src/solvers/NonlinearSystemSolvers.md b/docs/src/solvers/NonlinearSystemSolvers.md index 8aa7c0b1a..749468de5 100644 --- a/docs/src/solvers/NonlinearSystemSolvers.md +++ b/docs/src/solvers/NonlinearSystemSolvers.md @@ -10,7 +10,7 @@ Solves for ``f(u)=0`` in the problem defined by `prob` using the algorithm The default method `FastShortcutNonlinearPolyalg` is a good choice for most problems. It is a polyalgorithm that attempts to use a fast algorithm (Klement, Broyden) and if that fails it falls back to a more robust -algorithm (`NewtonRaphson`) before falling back the most robust varient of +algorithm (`NewtonRaphson`) before falling back the most robust variant of `TrustRegion`. For basic problems this will be very fast, for harder problems it will make sure to work. diff --git a/docs/src/tutorials/getting_started.md b/docs/src/tutorials/getting_started.md index 9f1685557..13ae48029 100644 --- a/docs/src/tutorials/getting_started.md +++ b/docs/src/tutorials/getting_started.md @@ -11,13 +11,13 @@ There are three types of nonlinear systems: 1. The "standard nonlinear system", i.e. the `NonlinearProblem`. This is a system of equations with an initial condition where you want to satisfy - all equations simultaniously. + all equations simultaneously. 2. The "interval rootfinding problem", i.e. the `IntervalNonlinearProblem`. This is the case where you're given an interval `[a,b]` and need to find where `f(u) = 0` for `u` inside the bounds. 3. The "steady state problem", i.e. find the `u` such that `u' = f(u) = 0`. While related to (1), it's not entirely the same because there's a uniquely - defined privledged root. + defined privileged root. 4. The nonlinear least squares problem, which is an overconstrained nonlinear system (i.e. more equations than states) which might not be satisfiable, i.e. there may be no `u` such that `f(u) = 0`, and thus we find the `u` which @@ -77,7 +77,7 @@ There are multiple return codes which can mean the solve was successful, and thu general command `SciMLBase.successful_retcode` to check whether the solution process exited as intended: -```@exmaple +```@example SciMLBase.successful_retcode(sol) ``` diff --git a/src/dfsane.jl b/src/dfsane.jl index e158b61c6..5648f8af9 100644 --- a/src/dfsane.jl +++ b/src/dfsane.jl @@ -6,8 +6,8 @@ A low-overhead and allocation-free implementation of the df-sane method for solving large-scale nonlinear systems of equations. For in depth information about all the parameters and the algorithm, -see the paper: [W LaCruz, JM Martinez, and M Raydan (2006), Spectral residual mathod without -gradient information for solving large-scale nonlinear systems of equations, Mathematics of +see the paper: [W LaCruz, JM Martinez, and M Raydan (2006), Spectral Residual Method without +Gradient Information for Solving Large-Scale Nonlinear Systems of Equations, Mathematics of Computation, 75, 1429-1448.](https://www.researchgate.net/publication/220576479_Spectral_Residual_Method_without_Gradient_Information_for_Solving_Large-Scale_Nonlinear_Systems_of_Equations) ### Keyword Arguments diff --git a/src/trustRegion.jl b/src/trustRegion.jl index 0070694ff..651591e84 100644 --- a/src/trustRegion.jl +++ b/src/trustRegion.jl @@ -600,7 +600,7 @@ function dogleg!(cache::TrustRegionCache{true}) return end - # Take the intersection of dogled with trust region if Cauchy point lies inside the trust region + # Take the intersection of dogleg with trust region if Cauchy point lies inside the trust region @. u_cauchy = -(d_cauchy / l_grad) * cache.g # compute Cauchy point @. u_tmp = u_gauss_newton - u_cauchy # calf of the dogleg -- use u_tmp to avoid allocation @@ -630,7 +630,7 @@ function dogleg!(cache::TrustRegionCache{false}) return end - # Take the intersection of dogled with trust region if Cauchy point lies inside the trust region + # Take the intersection of dogleg with trust region if Cauchy point lies inside the trust region u_cauchy = -(d_cauchy / l_grad) * cache.g # compute Cauchy point u_tmp = u_gauss_newton - u_cauchy # calf of the dogleg a = dot(u_tmp, u_tmp)