High Order SVD and ACA project of the Low Rank approximation lecture
- Install all requirements '''pip install -r requirements.txt''' in your favorite virtualenv
- Run (within the '''lra''' folder) '''python python setup.py build_ext --inplace'''
- Task 1 / HOSVD: run (within the '''lra''' folder) python task1.py domension --acc rel_error with the demanded dimension and relative error. --plot creates plots of the singular values
- Task 4 / ACA: run (within the '''lra''' folder) python task4.py domension --acc rel_error with the demanded dimension and relative error. --full runs full pivoting. --speed deactiveates calculation of full error, which requires the full matrix
- Benchmarks and tests are in LRA_tests.py and have to be run with pytest with its benchmarking module. Run with --slowrun to run all benchmarks (takes a lot of time)