Closed jarrodmillman closed 1 month ago
Still needs work (but it runs now):
$ python advanced/scipy_sparse/examples/lobpcg_sakurai.py /home/jarrod/src/scientific-python-lectures/advanced/scipy_sparse/examples/lobpcg_sakurai.py:47: UserWarning: Exited at iteration 500 with accuracies [1158100.72206623 8047.95297708 6596.98347933] not reaching the requested tolerance 1e-06. Use iteration 0 instead with accuracy 2.0908717846394285. eigs, vecs, resnh = sp.sparse.linalg.lobpcg( /home/jarrod/src/scientific-python-lectures/advanced/scipy_sparse/examples/lobpcg_sakurai.py:47: UserWarning: Exited postprocessing with accuracies [5.4496576 0.42115509 0.40180266] not reaching the requested tolerance 1e-06. eigs, vecs, resnh = sp.sparse.linalg.lobpcg( Results by LOBPCG for n=2500 [1.6384843 0.17288616 0.16788489] Exact eigenvalues [0.06250005 0.0625002 0.06250044] Elapsed time 1.311448997 /home/jarrod/src/scientific-python-lectures/advanced/scipy_sparse/examples/lobpcg_sakurai.py:64: UserWarning: FigureCanvasAgg is non-interactive, and thus cannot be shown plt.show()
According to the notes, we should be seeing:
$ python examples/lobpcg_sakurai.py Results by LOBPCG for n=2500 [ 0.06250083 0.06250028 0.06250007] Exact eigenvalues [ 0.06250005 0.0625002 0.06250044] Elapsed time 7.01
@stefanv Any idea how to fix the code?
Should be better now.
Still needs work (but it runs now):
According to the notes, we should be seeing: