These are some of my incomplete personal notes based on the book Numerical Algorithms by Justin Solomon.
- prelims.ipynb: Matrix storage patterns (row-major vs column-major), sparse matrices, floating-point representation
- LU_decomp_Gaussian_elimination.ipynb: Gaussian elimination, forward/backward substitution, LU factorization
- Cholesky_ConditionNumer.ipynb: Cholesky decomposition (LL^T), LDLT factorization, condition numbers
- QR_decomposition.ipynb: QR factorization via Gram-Schmidt and Householder transformations, least squares
- EigenVecs.ipynb: Eigenvalue/eigenvector computation (power iteration, inverse iteration, QR iteration, deflation)
- IterativeSolvers.ipynb: Iterative methods for linear systems (gradient descent, conjugate gradient(TODO))
- RootFinding.ipynb: Root finding algorithms (bisection, fixed-point iteration, Newton's method, secant method)
- UnconstrainedOptim.ipynb: Unconstrained optimization (Newton's method(TODO), gradient descent)
- Constrained_optimization.ipynb: Constrained optimization (Lagrange multipliers, KKT conditions, ADMM)
- NonlinearOptimzations.ipynb: Nonlinear least squares (Gauss-Newton, Levenberg-Marquardt), ICP implementation
- InterpolationandDifferentiation.ipynb: Polynomial interpolation, numerical integration (quadrature), finite differences, automatic differentiation
- ODEs.ipynb: ODE solvers (Euler methods, Runge-Kutta, leapfrog, exponential integrators)
- PDEs.ipynb: PDE solvers (finite difference method, finite element method) for Laplace/Poisson equations
Each notebook includes:
- My personal notes about the methods
- Some simple Python implementations that I felt necessary to understand the content (still some TODOs)
- Few Examples
- NumPy, SciPy
- Matplotlib, PyVista (for visualizations)
- Jupyter Notebook