Adriano b143c6607a feat: numpy.bitwise_count come fallback SIMD per popcount ...

NumPy 2.0+ espone np.bitwise_count: implementato in C nativo con
intrinsics SIMD (POPCNT/AVX2 vpopcnt). Aggiunto come fallback secondo
livello quando Numba non e disponibile (es. wheel constraint, env
ristretto). Numba JIT parallel resta default: misura su 1080p 0.5ms
vs 1.6ms (bitwise_count e single-thread).

AVX2 puro su _jit_score_bitmap_rescored richiederebbe C extension
con build nativa: out-of-scope per questo branch (Numba LLVM gia
autovettorizza il loop interno).

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

2026-05-04 15:36:48 +02:00

..

web

feat: upload file nella cartella IMAGES_DIR

2026-04-24 14:45:16 +02:00

__init__.py

perf: Numba JIT kernel per score_by_shift (2.1x speedup)

2026-04-24 01:30:31 +02:00

_jit_kernels.py

feat: numpy.bitwise_count come fallback SIMD per popcount

2026-05-04 15:36:48 +02:00

auto_tune.py

perf: Fase 2 speed (3x baseline) - fuse JIT + LRU + sub-pixel lazy

2026-04-24 21:21:59 +00:00

gui.py

ui: layout fisso 1600x900 con pannelli SX/DX, scena scalata

2026-04-24 08:59:06 +02:00

line_matcher.py

perf: Fase 2 speed (3x baseline) - fuse JIT + LRU + sub-pixel lazy

2026-04-24 21:21:59 +00:00

matcher.py

perf: piramide al training, refinement sub-step, multithreading

2026-04-24 01:22:56 +02:00