2 questions · 2 experiments · 6 criteria · 9 runs, 1 retracted · 4 resources
2 of 6 criteria were declared after their evidence existed · 2 of 8 runs record their command (as given)
All criteria · 6: 4 criteria met · 2 criteria not met
4 calls wait on you.
accept c3: the ONNX export holds onnx_top1
note: one run is enough for an export check · not confirmed
rb decide c3 accept --why 'one run is enough for an export check'
Confirm it in your own terminal:
rb confirm d7k2pqa
park e2, as notes.md:3 reads: "we park the 128-image cell for now"
Confirm it in your own terminal:
rb confirm d4m8xqe
Everything on this page was computed from the record this bundle carries. Rerun that computation, and the runs behind it, on your machine:
Rerun this researchResearch
2 results to review
fp32 top1 matches the published 76.13 c23 runs · 3 seeds per arm · 1 run not counted
rule: change.top1 ≥ -0.01 (n ≥ 3)
Recorded interpretation
the ONNX export keeps the INT8 reading · not confirmed
criterion met · loss
read in imagenet-val
declared 2026-10-12 · experiment e1 · metric top1 · what the words mean
not reproduced · measured
criterion met · measured
read through a1
criterion not met · point only
rests on runsql7
Experiments
design confirmed, then changed: rb confirm e1 lists it
ready: code, environment and inputs recorded
Rerun this research with Rabbit Brain (rb). Its record: https://example.org/int8/page.html 1. Install rb: pip install rabbit-brain 2. Clone the record into a new folder: rb clone https://example.org/int8/bundle.zip int8-resnet50 Where the record names the code the runs need, rb clone says its repository and commit. If it names none, ask me where the code is. 3. In that folder, run rb context and read it before anything else. It names each resource, such as a model or a dataset, with how to get it. 4. Fetch that repository and check out its commit in the same folder, beside .rb/, leaving out its own .rb/ and the results its runs wrote (its results folders and databases). Get those resources. Rerun experiment e1, "int8 against fp32", the way the record describes it, with rb journal run -- <command>, and record each run on its experiment: rb run add <experiment> --from <the file it wrote> --command "<that command>" (rb says any per-run value it needs). 5. Then run rb status, and tell me which of its rules (c1, c2) reproduced and which did not, with their numbers. Do not change a rule. If something cannot run here, say what and why.
git checkout 3f2a9c1 python eval.py --precision int8 --seed 1
| run | arms | top1 |
|---|---|---|
| runs1 | fp32 and int8 · seed 1 | 0.7598 · 0.7569 |
| runs3 | fp32 and int8 · seed 3 | 0.7599 · 0.7572 |
design not confirmed: rb confirm e2 lists it
Record
fp32: top1=0.7598 onnx_top1=0.7569 · int8: top1=0.7569 onnx_top1=0.7566
counts toward c1, c2, c3 · made ckpt-int8-s1 · read imagenet-val
python eval.py --precision int8 --seed 1
replaced by runs2b
fp32: latency_ms=6.1 · int8: latency_ms=3.9
4 calls wait on you.
Not reproduced · estimate 0.7597 outside 0.7613 ± 0.001
It rests on 3 runs · 1 run not counted · the whole finding
Its review will read:
rb decide c2 accept --why '‹why›'
rb decide c2 reject --why '‹why›'
rb decide c2 investigate --why '‹why›'
Its review will read:
rb decide e2 continue --why '‹why›'
rb decide e2 park --why '‹why›'
rb decide e2 stop --why '‹why›'