pytroll / pyresample
94%

DEFAULT BRANCH: main
Repo Added
Files 99
Badge
Embed ▾
README BADGES
x

If you need to use a raster PNG badge, change the '.svg' to '.png' in the link

Markdown

Textile

RDoc

HTML

Rst

LAST BUILD ON BRANCH main
branch: SELECT
CHANGE BRANCH
x

coverage: 93.915%. Remained the same
35048142149

push

github

djhoese 
Optimize the EWA fornav inner loops without changing results

Restructure compute_ewa() and compute_ewa_single() so the per-grid-cell loop
only does the weight lookup and the two grid updates. The ellipse parameters
and weight table fields are now copied into locals and the array arguments are
declared restrict, so the compiler no longer has to assume every grid store may
have modified them and reload them each iteration. The input pixel's
fill/NaN check is hoisted out of the bounding box loops (once per swath pixel,
with a compact list of valid channels in the multi-channel path),
maximum_weight_mode is a template parameter so each mode gets its own
specialised loop, the weight table is over-allocated by one entry so the index
clamp can be dropped, and the bounding box test is evaluated on the float
bounds before any int cast so pixels far outside the grid on either side (the
common case for dask chunks) are rejected with four comparisons and NaN
coordinates fall out without isnan(). -fno-math-errno is added to the compile
flags so sqrt()/exp() can inline. No floating point operation changes order, so
outputs are bit-for-bit identical to the previous build; this was verified with
np.array_equal across both wrappers, all coordinate/image dtypes and both weight
modes. Synthetic timings show 1.1-1.3x on the single-channel path, most for
large ellipses and for chunks where most source pixels lie outside the grid.

13149 of 14001 relevant lines covered (93.91%)

2.82 hits per line

Relevant lines Covered
14001 RELEVANT LINES 13149 COVERED LINES
2.82 HITS PER LINE
Source Files on main

Recent builds

Builds Branch Commit Type Ran Committer Via Coverage
35048142149 main Optimize the EWA fornav inner loops without changing results Restructure compute_ewa() and compute_ewa_single() so the per-grid-cell loop only does the weight lookup and the two grid updates. The ellipse parameters and weight table fields are now... push djhoese github
93.91
35046994174 opt-ewa-inner-loop Merge 7454aee87 into 1b16aa4b1 Pull #752 web-flow github
93.91
35026971419 main Fix memory leaks and fill bug in EWA resampling push djhoese github
93.91
35022110896 bugfix-ewa-leaks Merge 9e2bb3d86 into 41ab83deb Pull #751 web-flow github
93.91
35019463328 main Merge pull request #750 from djhoese/ewa-persist-prune Optimize Dask EWA persist and prune fornav tasks push web-flow github
93.91
35010529628 ewa-persist-prune Merge 82b485caa into 2140cedcc Pull #750 web-flow github
93.91
34864394940 feat-agents-guide Merge 71f7a5b7a into 2140cedcc Pull #744 web-flow github
93.75
34752808737 main Make arguments lowercase push djhoese github
93.75
34690504502 main Fix bilinear cache type decoding push djhoese github
93.72
34635202024 fix-returns-docstring-xrefs Merge f4a479f45 into 1ef650389 Pull #749 web-flow github
93.75
See All Builds (2095)