-
Notifications
You must be signed in to change notification settings - Fork 15
NEW: Solve the lamino-alignment problem #94
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
base: main
Are you sure you want to change the base?
Conversation
Because map_coordinates is not implemented for higher orders in CuPy.
In theory, this should improve runtimes for long-lived Lamino operators.
Make the CUDA scatter gather kernels for the Lamino operator work for any number of dimensions because we will reuse it for the Tomo operator.
Conflicts: src/tike/operators/cupy/flow.py src/tike/operators/cupy/lamino.py src/tike/operators/numpy/flow.py src/tike/operators/numpy/lamino.py tests/test_align.py
Hello @carterbox! Thanks for updating this PR. We checked the lines you've touched for PEP 8 issues, and found:
Comment last updated at 2021-07-27 16:34:15 UTC |
Conflicts: src/tike/align/solvers/cross_correlation.py src/tike/operators/cupy/alignment.py src/tike/operators/cupy/gradient.py src/tike/ptycho/ptycho.py
Purpose
Provide a function for solving the lamino-alignment problem with ADMM.
Approach
Solve the laminography + alignment problem using the Laminography operators, Lanczos Interpolation operator, and Flow solver which wraps OpenCV's Farneback's method.
Also includes some changes to the ThreadPool and scatter/gather CUDA kernels which should be merged separately.
Pre-Merge Checklists
Submitter
yapf
to format python code.Reviewer