rUv 4ec8a86f5d Audit activity labels and evaluate native CSI resolution ...

Preserve native source precision and exact pooled controls under the original
training and development split. Add an 18-configuration resolution ablation,
input provenance and artifact tests, and measured results retaining the
existing ACF model. Document unresolved semantic labels and confirmation limits.

2026-10-12 17:11:10 -07:00

..

__init__.py

feat(training): add reproducible CSI encoder harness and evidence contracts

2026-10-12 14:42:31 -07:00

data.py

feat(training): add reproducible CSI encoder harness and evidence contracts

2026-10-12 14:42:31 -07:00

espfi_blend.py

perf(training): encode fixed RFF maps by seed and record rejected blend

2026-10-12 15:49:41 -07:00

espfi_context.py

Explore RuField and WorldGraph with bounded CSI context experiments

2026-10-12 16:50:20 -07:00

espfi_generalize.py

feat(training): improve development macro F1 with isolated spectral model search

2026-10-12 15:00:22 -07:00

espfi_kernel.py

perf(training): halve seeded RFF model state with verified float32 inference

2026-10-12 16:00:45 -07:00

espfi_motion.py

Retain motion ACF research candidate with worst-room selection and measured gates

2026-10-12 16:29:27 -07:00

espfi_resolution_data.py

Audit activity labels and evaluate native CSI resolution

2026-10-12 17:11:10 -07:00

espfi_resolution.py

Audit activity labels and evaluate native CSI resolution

2026-10-12 17:11:10 -07:00

espfi_runtime_benchmark.py

perf(csi): add verified buffered inference runtime and parity benchmark

2026-10-12 16:11:27 -07:00

espfi_runtime.py

perf(csi): add verified buffered inference runtime and parity benchmark

2026-10-12 16:11:27 -07:00

espfi_seeded.py

perf(csi): add verified buffered inference runtime and parity benchmark

2026-10-12 16:11:27 -07:00

espfi_train.py

feat(training): public CSI activity optimization with leakage guards and measured inference folding

2026-10-12 14:53:12 -07:00

espfi.py

feat(training): improve development macro F1 with isolated spectral model search

2026-10-12 15:00:22 -07:00

model.py

feat(training): add reproducible CSI encoder harness and evidence contracts

2026-10-12 14:42:31 -07:00

train.py

feat(training): add reproducible CSI encoder harness and evidence contracts

2026-10-12 14:42:31 -07:00