SikongJueluo b5a40819cc feat: vectorize CAM retrieval with NumPy and add multi-worker support ...

- Replace scalar hamming distance with NumPy bitwise_count for batch retrieval
- Add ThreadPoolExecutor-based multi-worker query parallelism
- Improve missing dataset error message with generation command hint
- Increase DEFAULT_MAX_QUERIES from 128 to 8192 for meaningful throughput tests

2026-06-07 20:45:20 +08:00

..

common.py

refactor(scripts): extract shared utilities into common module

2026-05-03 19:49:45 +08:00

prepare_cam_retrieval_dataset.py

feat: vectorize CAM retrieval with NumPy and add multi-worker support

2026-06-07 20:45:20 +08:00

remote_docker_run.py

refactor(scripts): extract shared utilities into common module

2026-05-03 19:49:45 +08:00

run_cam_correctness.py

refactor(cam): rename read-noise path to read-pass-through and reorganize test module structure

2026-05-27 16:34:34 +08:00

run_retrieval_noise_sweep.py

feat(benchmarks): add noise injection experiment support to CAM retrieval benchmark

2026-05-27 19:04:25 +08:00

sw_retrieval_benchmark.py

feat: vectorize CAM retrieval with NumPy and add multi-worker support

2026-06-07 20:45:20 +08:00

sweep_cam_perf.py

feat(sim): add CAM performance sweep test infrastructure

2026-05-16 15:42:29 +08:00