Skip to content

Repository files navigation

Numerical Solvers & Analysis

A collection of numerical analysis algorithms implemented in Python alongside PDFs that explain error analysis with visualization plots.


📌 Project Overview

This repository contains implementations of fundamental numerical methods alongside theoretical documentation:

  • Linear Algebra: Iterative solvers for $Ax = b$ and spectral analysis for eigenvalues.
  • Root-Finding: One-dimensional nonlinear equation solvers.
  • Documentation: A comprehensive LaTeX PDF detailing the mathematical theory, convergence criteria, and performance comparisons.

📑 Implemented Methods (just to list a few)

1. Linear Systems & Eigenvalues

  • Gauss-Seidel Method: Iterative solver for systems of linear equations using the most recent updates at each step.
  • Successive Over-Relaxation (SOR): Accelerated variant of Gauss-Seidel using a relaxation factor ($\omega$) to speed up convergence.
  • Power Method: Iterative technique used to estimate the dominant eigenvalue and corresponding eigenvector of a matrix.
  • Inverse Power Method: Variant used to find the smallest eigenvalue (or the eigenvalue closest to a target shift) using matrix inversion/decomposition.

2. Root-Finding & Fixed-Point Iterations

  • Fixed-Point Iteration: Solves $x = g(x)$ by repeatedly evaluating the iteration function.
  • Newton's Method: Rapidly converging root-finding technique utilizing function derivatives ($x_{n+1} = x_n - \frac{f(x_n)}{f'(x_n)}$).

About

A comparative analysis of numerical methods ([e.g., Bisection, Newton, Secant]) evaluating convergence rates, performance, and mathematical error bounds. This was originally developed as part of my MAT 128B coursework at UC Davis.

Topics

Resources

Stars

0 stars

Watchers

0 watching

Forks

Releases

Packages

Contributors

Languages