This repository contains implementations of numerical methods developed as part of the "Metodi Numerici per la Chimica Fisica" course.
- Cholesky Decomposition with linear systems solver.
- Jacobi method.
- Conjugate gradient with preconditioning.
- Davidson method eigensolver.
Compiled with GNU Fortran (gfortran) compiler
BLAS and LAPACK libraries needed
# Navigate to desired algorithm directory
cd Cholesky/ # or Jacobi/ or Conjugate_gradient/ or Davidson/
# Compile using the provided Makefile
make
# Run the executablek
./main.exe
All implementations use a common test matrix bulid with:
- Dominant diagonal elements defined as A_(ii)=i+1,
- and off-diagonal elements given by A_(ij)=1/(i+j).
This matrix structure mimics the properties usually found in quantum chemistry problems such as being symmetric, positive definite, diagonally dominant, and sparse.