Margaret Duff
Margaret Duff
As discussed in the software meeting today, we are resurrecting this PR. @paskino @jakobsj - I would be grateful for a brief look
In https://github.com/TomographicImaging/CIL/pull/1645 we found that we needed a random seed for the power method tests! It would be good to be able to set the random seed in the argument...
Discussed with @zeljkozeljko Example ```python m, n = 900, 1000 A = np.random.normal(0, 1, (m, n)) U, S, VT = np.linalg.svd(A, full_matrices=False) smallest_eigenval = 1e-2 largest_eigenval = 1. Anew =...
Suggestion: Perhaps when calculating the Lipschitz constant of least squares, we could use a larger iteration number for the power method?
Some open questions: - This BlockDiagonalOperator is an Operator but not a BlockOperator because it doesn't have the full list of operators needed for the BlockOperator. I wonder if this...
@jakobsj You mentioned that you had thought about this before - any thoughts?
Need to also test with strong convexity constant >0
Replaced by #1498
So we can't move the tolerance check to a callback, as that is passed to the 'run' argument and not in the algorithm set-up. The next question is: should early...
Thanks @epapoutsellis! It looks good. I added a few todos for myself around testing and documentation. Feel free to answer them if you have time before me but I will...