Repository navigation
Conversation
bench.jl --device N runs the GPU libraries on the N-th GPU (KernelAbstractions' device! for CTOperators, CuPy for parallelproj), e.g. a free one on a shared machine. The guide "Run on a GPU" explains selecting the GPU before creating a projector. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
tune_launch.jl times a projector on one GPU (CUDA or ROCm, device selectable) for workgroup sizes and tiles of detector cells, the device's defaults first, and lists the fastest configurations. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A new explanation page builds a scan up from one ray to a helical cone-beam scan and introduces the parameters of the trajectory functions where they appear (angles, dso, dsd, nu, nv, du, dv, pitch, z0), with a figure per step. Getting Started lists the parameters and links to it. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Kernel and launch comments keep the reasons and invariants; how they were found and the measured numbers are left to the commit log and benchmark/README.md. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Projectors apply M_rays A M_image and its exact adjoint. The generic layer applies the masks around forward!/backward!, so every operator gets these semantics; Siddon and Joseph skip masked rays in their kernels (trait handles_raymask) and clip rays to the bounding box of the image mask (VoxelGrid lo/hi). subset and adapt keep the masks. Re-exports the mask types of ImageGeoms. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
How-to "Use Masks" (circular field of view, defective detector cells, region of interest) and explanation "Masks" (semantics, adjointness, where masks save time, masks for other operators), a section on masks in the implementation guide, and GPU timings with masks in the benchmark README. The navigation collapses to the first level. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Codecov Report❌ Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## main #10 +/- ##
==========================================
- Coverage 98.15% 96.30% -1.85%
==========================================
Files 18 19 +1
Lines 489 542 +53
==========================================
+ Hits 480 522 +42
- Misses 9 20 +11 ☔ View full report in Codecov by Harness. 🚀 New features to boost your workflow:
|
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Adds masks to the projectors: an image mask (the voxels that are reconstructed) and a ray mask (the rays whose data are used).