This test evaluates your ability to estimate the condition number of a symmetric positive definite (SPD) matrix using numerical methods.
Run the following command from the root of this package:
python3 data/generate_matrix.py --n 64 --kappa 1e3 --outA A.csv --outb b.csv
This will generate A.csv and b.csv in the working directory.
- CPU: Intel / AMD x86_64
- Compiler: g++ (C++17)
- Optimization: -O3
- Libraries: Standard C++, LAPACK (for verification on small matrices)
- OS: Linux
To run the solver in the directory with eigenvalue_estimation.cpp file:
g++ -O3 eigenvalue_estimation.cpp -o eigenvalue_estimation -llapack -lblas
./eigenvalue_estimation data/A.csv data/b.csv
To run the plotting file using lambda_max.csv and lambda_min.csv log files:
python3 plot_convergence.py