28.50 mAP
YOLOv8n · COCO val2017 · 320 × 320
Exact report value: 0.284969Results with scope
What does the evidence prove?
01 / Software accuracy
How accurate is the calibrated proxy?
The pinned 320×320 YOLOv8n FLOAT export and signed-symmetric INT8 ONNX Runtime proxy were evaluated over 5,000 COCO val2017 images. The proxy loses 0.887 mAP50–95 points, within the project’s one-point budget. A deterministic 512-image train2017 subset supplied calibration data.
27.61 mAP
Signed-symmetric INT8 proxy · 320 × 320
Exact report value: 0.276098−0.89 mAP points
3.11% relative loss · 96.9% retained
Exact loss: 0.887 pointsScores use the familiar 0–100 mAP scale; exact 0–1 report values remain visible for traceability. YOLOv8n · 320 × 320 · 5,000 COCO val2017 images. Both software evaluations were repeated with identical predictions. The INT8 result uses an ONNX Runtime QOperator proxy; complete-detector HASLAB command execution and hardware measurements remain future work.
Why the INT8 proxy is a separate result
ONNX Runtime uses QLinear sigmoid and multiply rather than HASLAB’s fused SiLU lookup, and requantizes learned-head outputs where v0 preserves an INT32 boundary. The proxy measures a calibration candidate; it does not establish whole-detector HASLAB command accuracy. See the INT8 report and calibration record.
02 / Command correctness
614,400 values. Zero mismatches.
The minimal compiler and simulator runtime execute the first two complete Conv-SiLU blocks. The first retained tensor contains 409,600 INT8 values; the second contains 204,800. Each matches the independent haslab_ref integer path exactly for the report’s deterministic input.
Command records16,245ABI 0.1 candidate
INT8 values compared614,400Both retained tensors
Integer mismatches0Golden comparison
DMA bytes1,999,320Functional schedule
The two-block report records package hash, command mix, traffic, exact integer comparisons, and separate FLOAT ONNX errors. These counts are not hardware timing or throughput. Remaining graph nodes, host tail, RTL, and FPGA execution are unfinished.
Understand command execution →03 / Independent expectations
What do the 46 fixtures check?
The candidate conformance corpus stores binary commands, initial and expected memory images, completion state, errors, hashes, and provenance. It covers successful and rejected behavior so another implementation can consume the same expected bytes. Stable ABI freeze and final corpus release await independent review.
Numerical rules
Rounding, saturation, accumulation order, layouts, conversions, and SiLU behavior have documented semantics and tests.
Command behavior
Memory bounds, DMA, lifecycle, reset, completion, and architectural faults are specified in the v0 contract.
Reproducible fixtures
The fixture format and manifest record expectations and integrity checks.
04 / Source reports
Reproduce each claim.
Reports below are linked to reviewed source revision 15a2e45. Model weights and COCO images require their own setup; the workload guide gives commands and artifact requirements.