All issues

base.py suggests nonexistent poetry groups; AUTO_PRIORITY's benchmark harness is absent from the repo; assorted dead code and stale paths

#311

· Ramdam17 opened

on Jul 30, 2026

compute_nmPLV applies the n:m exponent to the time axis instead of the channel axis, averages the wrong axis, and calls a deprecated helper

#310

· Ramdam17 opened

on Jul 30, 2026

CUDA: free_all_blocks() on every call defeats the CuPy memory pool; pair indices rebuilt per call in all four dispatchers

#309

· Ramdam17 opened

on Jul 30, 2026

Sync test suite: tolerances ~10^5-10^6 looser than measured error, a duplicate class silently drops 5 tests, and #279 has no regression test

#308

· Ramdam17 opened

on Jul 30, 2026

CI exercises zero accelerated backends: all 49 skipif-gated sync tests skip

#307

· Ramdam17 opened

on Jul 30, 2026

pair_connectivity cannot forward optimization/priority, and compute_sync masks all compute errors as "metric not supported"

#306

· Ramdam17 opened

on Jul 30, 2026

'numba' inside a priority= list is silently ignored, contradicting three documented examples

#305

· Ramdam17 opened

on Jul 30, 2026

optimization='auto' cannot select metal or cuda_kernel unless torch is installed; availability probes never check for a device

#304

· Ramdam17 opened

on Jul 30, 2026

Metal backend returns float32 while every other backend returns float64

#303

· Ramdam17 opened

on Jul 30, 2026

ImCoh is documented signed but returns |Im|; ccorr and accorr use opposite sign conventions

#302

· Ramdam17 opened

on Jul 30, 2026

Coherence is documented as magnitude-squared in 4 places but implemented unsquared

#301

· Ramdam17 opened

on Jul 30, 2026

Six Metal tests are tautological (numpy vs numpy) and one docstring describes a kernel that never existed

#300

· Ramdam17 opened

on Jul 30, 2026