How giancarloantonucci/nsdemovingwindow.jl works
NSDEMovingWindow.jl solves a dynamical-system problem across overlapping time windows, using Parareal as the local solver. The README documents window-to-window seeding, retained window solutions and diagnostics, solution lookup, and ensemble statistics, plus configurable window strategies and optional plotting. The graph reflects only those documented relationships; source excerpts were unavailable, so internal call details and links to unsampled capabilities are omitted.
Main entry points: mowi/solve.jl Window execution · src/solve.jl Solve entry · solution.jl Moving-window solution · adaptive.jl Window strategies
Components 11
Solver configuration
- MoWi configuration
src/mowi/constructor.jl - Window strategies
src/mowi/adaptive.jl
Window runtime
- Solve entry
src/solve.jl - Window execution
src/mowi/solve.jl - Window state and cache
src/mowi/cache.jl
Solutions and analysis
- Moving-window solution
src/mowi/solution.jl - Ensemble statistics
src/mowi/solution.jl - Plot recipes
src/plots_recipes.jl
Other components
- User
- Dynamical problem
- Parareal local solver
Connections 12
- User → Solve entry: calls solve
README.md - User → MoWi configuration: configures windows
README.md - User → Parareal local solver: selects local solver
README.md - User → Dynamical problem: provides problem
README.md - Solve entry → Window execution: runs windows
- Window execution → Parareal local solver: uses local solver
README.md - Window execution → Window state and cache: seeds next window
README.md - Window execution → Moving-window solution: retains results
README.md - User → Moving-window solution: receives result
README.md - User → Ensemble statistics: computes ensemble mean
README.md - User → Plot recipes: uses plotting
README.md - Window strategies → Window execution: configures windows