Skip to content

release: verification 20.16% per record / 34.01% per product - #159

Merged
Seungpyo1007 merged 5 commits into
mainfrom
develop
Aug 11, 2026
Merged

release: verification 20.16% per record / 34.01% per product#159
Seungpyo1007 merged 5 commits into
mainfrom
develop

Conversation

@Seungpyo1007

@Seungpyo1007 Seungpyo1007 commented Aug 11, 2026

Copy link
Copy Markdown
Member

Release of 5 commits. Merging deploys the site and notifies TechEngine.

records  105,742   verified 21,314  = 20.16%   (was 20.11%)
models    62,148   verified 21,136  = 34.01%   (was 33.92%)

CPU moved 65.25% → 66.61%.

What is in it

54 CPU promotions. passmark_multi_ge_single asserted passmark_cpu_mark >= passmark_single, but those are two separately normalised PassMark scales, so the comparison is not meaningful. All 51 single-core parts failed it, and failure tracked absolute weakness rather than parallelism (failing CPU Mark median 641 against 19,657 for passing), with near-ties at the boundary — Core 2 Duo E8600 at 1378 vs 1388. Dropped; Cinebench and Geekbench keep their multi ≥ single checks because those report one scale (GetTechAPI/TechEngine#58).

The submodule pointer is moving again. bump-engine had failed 9 consecutive times since 2026-08-03 — TechEngine carries TechAPI back as a submodule, and git submodule update --recursive walked that cycle, re-fetching the million-file data tree at each level until the runner ran out of disk. Dropping --recursive fixed it; this release carries the first successful bump since (d84bcec), which is what finally brings the merged engine fixes into the pointer TechAPI publishes.

Closes #1

TechEngineBot and others added 5 commits August 10, 2026 07:00
Auto-promotions from the verification layer (green+live-source or crossref-confirm).
Each flip is verified:false->true only; see data/_verify/ledger.jsonl. Refs #1
TechEngine carries TechAPI back as a submodule, so `git submodule
update --recursive` follows that cycle — TechEngine/TechAPI/TechEngine/
TechAPI/… — re-fetching the million-file data tree at every level until
the runner dies with "No space left on device":

    fatal: Unable to checkout '679e487…' in submodule path
    'TechEngine/TechAPI/TechEngine/TechAPI/TechEngine/TechAPI/…'

Every bump-engine run has failed this way since 2026-08-03, which is
when the job started switching branches before touching the submodule
and so began re-resolving it from scratch. The submodule pointer has
not moved automatically since.

The gitlink being bumped is at this repo's top level; nothing below it
needs updating, so drop --recursive from both submodule commands.

Refs #1
@TechEngineBot

Copy link
Copy Markdown
Member

🔎 Data verification — Tier 0 (offline existence/trust)

Scored by app.verify; posted by TechEngineBot. Informational only — the structural gate (app.validate) is separate and authoritative for merge.

Changed records in this PR

54 record(s) scored.

%%{init: {"theme":"base","themeVariables":{"pie1":"#3fb950","pie2":"#d29922","pie3":"#f85149","pieStrokeWidth":"0px","pieOpacity":"1"}}}%%
pie showData
    title Verification bands — all records
    "Green" : 54
    "Yellow" : 0
    "Red" : 0
Loading
Category Distribution Total 🟢 🟡 🔴 🟢 %
cpu 🟩🟩🟩🟩🟩🟩🟩🟩🟩🟩🟩🟩 54 54 0 0 100.0%
All 🟩🟩🟩🟩🟩🟩🟩🟩🟩🟩🟩🟩 54 54 0 0 100.0%

Full-dataset baseline

105742 record(s) scored.

%%{init: {"theme":"base","themeVariables":{"pie1":"#3fb950","pie2":"#d29922","pie3":"#f85149","pieStrokeWidth":"0px","pieOpacity":"1"}}}%%
pie showData
    title Verification bands — all records
    "Green" : 21313
    "Yellow" : 83228
    "Red" : 1201
Loading
Category Distribution Total 🟢 🟡 🔴 🟢 %
brand 🟩🟨🟨🟨🟨🟨🟨🟨🟨🟨🟨🟨 207 24 183 0 11.6%
soc 🟩🟨🟨🟨🟨🟨🟥🟥🟥🟥🟥🟥 2104 146 844 1114 6.9%
smartphone 🟩🟩🟨🟨🟨🟨🟨🟨🟨🟨🟨🟥 93396 17382 75934 80 18.6%
tablet 🟩🟨🟨🟨🟨🟨🟨🟨🟨🟨🟨🟥 3455 218 3236 1 6.3%
watch 🟩🟨🟨🟨🟨🟨🟨🟨🟨🟨🟨🟥 433 18 410 5 4.2%
pda 🟩🟩🟩🟩🟩🟩🟨🟨🟨🟨🟨🟨 140 66 74 0 47.1%
gpu 🟩🟩🟩🟩🟩🟨🟨🟨🟨🟨🟨🟨 2030 810 1220 0 39.9%
cpu 🟩🟩🟩🟩🟩🟩🟩🟨🟨🟨🟨🟥 3977 2649 1327 1 66.6%
All 🟩🟩🟨🟨🟨🟨🟨🟨🟨🟨🟨🟥 105742 21313 83228 1201 20.2%

Hard violations (forced red):

Count Check
1 !boost_ge_base

green = authoritative source + complete + consistent · yellow = plausible, needs confirmation · red = sparse/weak source or a hard contradiction. Promotion to verified runs in the scheduled verify-network workflow.

@TechEngineBot

Copy link
Copy Markdown
Member

TechEngine change review: PASS

Check Result
python -m app.validate PASS
python integrity_check.py TechAPI/data --strict PASS

Changed data

Category Added Modified Deleted Added verified Added unverified Added Kaggle-sourced
brand 0 0 0 0 0 0
soc 0 0 0 0 0 0
smartphone 0 0 0 0 0 0
tablet 0 0 0 0 0 0
watch 0 0 0 0 0 0
pda 0 0 0 0 0 0
gpu 0 0 0 0 0 0
cpu 0 54 0 0 0 0

Changed record examples

cpu modified

  • cpu/amd/2008/consumer/amd-athlon-64-x2-dual-core-5000-plus.json - AMD Athlon 64 X2 Dual Core 5000+
  • cpu/amd/2008/consumer/amd-athlon-64-x2-dual-core-5400-plus.json - AMD Athlon 64 X2 Dual Core 5400+
  • cpu/amd/2008/consumer/amd-athlon-64-x2-dual-core-5600-plus.json - AMD Athlon 64 X2 Dual Core 5600+
  • cpu/amd/2008/consumer/amd-athlon-64-x2-dual-core-6000-plus.json - AMD Athlon 64 X2 Dual Core 6000+
  • cpu/amd/2008/consumer/amd-athlon-64-x2-dual-core-6400-plus.json - AMD Athlon 64 X2 Dual Core 6400+
  • cpu/amd/2008/consumer/amd-sempron-2400-plus.json - AMD Sempron 2400+
  • cpu/amd/2008/consumer/amd-sempron-3300-plus.json - AMD Sempron 3300+
  • cpu/amd/2008/consumer/amd-turion-64-mobile-ml-32.json - AMD Turion 64 Mobile ML-32
  • cpu/amd/2008/consumer/amd-turion-x2-dual-core-mobile-rm-72.json - AMD Turion X2 Dual-Core Mobile RM-72
  • cpu/amd/2008/enterprise/amd-opteron-144.json - AMD Opteron 144
  • cpu/amd/2008/enterprise/amd-opteron-175.json - AMD Opteron 175
  • cpu/amd/2009/consumer/amd-athlon-1640b.json - AMD Athlon 1640B
  • cpu/amd/2009/consumer/amd-athlon-2650e.json - AMD Athlon 2650e
  • cpu/amd/2009/consumer/amd-athlon-2850e.json - AMD Athlon 2850e
  • cpu/amd/2009/consumer/amd-athlon-5000-dual-core.json - AMD Athlon 5000 Dual-Core
  • ... 39 more

Heuristic review

  • Heuristic warnings: none found.

@TechEngineBot

Copy link
Copy Markdown
Member

TechEngine validation stats: PASS

Data summary

Category Total Verified Unverified Missing verified Tracked Verified % of tracked
brand 207 24 183 0 207 11.6%
soc 2104 146 1958 0 2104 6.9%
smartphone 93396 17384 76012 0 93396 18.6%
tablet 3455 218 3237 0 3455 6.3%
watch 433 18 415 0 433 4.2%
pda 140 66 74 0 140 47.1%
gpu 2030 809 1221 0 2030 39.9%
cpu 3977 2649 1328 0 3977 66.6%
all 105742 21314 84428 0 105742 20.2%

Warning

Tracked verified coverage is below 50% for watch 4.2% (18/433), tablet 6.3% (218/3455), soc 6.9% (146/2104), brand 11.6% (24/207), smartphone 18.6% (17384/93396), all 20.2% (21314/105742), gpu 39.9% (809/2030), pda 47.1% (66/140).
Tracked coverage excludes records missing the verified field; see the Missing verified column for those records.
This does not fail validation. Keep imported records verified: false until manual audit, but treat this as follow-up verification work before relying on the affected categories as curated data.

Validation notes

  • Full advisory outlier listings are suppressed on successful runs because they are dataset-wide and mostly stable between PRs.
  • Failure runs still include a detailed log excerpt for debugging.

Key output:

## app.validate
## integrity_check.py --strict
loaded CPU=3977 GPU=2030
✅ integrity gate: no hard anomalies.
Integrity section Flagged lines
structural 0
CPU name/tier consistency (desktop mainstream only) 0
CPU single>multi (cinebench/geekbench — should be multi>=single) 0
CPU era-vs-score outliers 8
CPU cross-source ratio outliers (possible wrong-variant) 152
GPU cross-source ratio outliers + sanity 18

@Seungpyo1007
Seungpyo1007 merged commit dcfd64f into main Aug 11, 2026
10 checks passed
@github-project-automation github-project-automation Bot moved this from In Progress to Done in TechAPI-Project Aug 11, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

ci CI and workflow changes data Dataset changes enhancement New feature or request

Projects

Status: Done

Development

Successfully merging this pull request may close these issues.

2 participants