willtanoe/thesis-prototype-agregation
1
0 Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
27 Commits 2 Branches 0 Tags
ad203eb2ee6e41427b7bd698a3e9df0234141ef6
Commit Graph
27 Commits
This Branch
This Branch
All Branches
Author SHA1 Message Date
willtanoe ad203eb2ee docs: record 1-thread seed-0 sweep, noise checks and grid comparison 2026-10-29 01:40:05 +01:00
willtanoe 91b0e371f7 chore: declare matplotlib and apply three behavior-preserving nits 
- pyproject.toml: add a figures extra with matplotlib (used by the two figure
  scripts) so the environment is reproducible from pyproject alone.
- partition.py: correct the partition_eval_split docstring (simulated per-client
  eval sets from the global split, not client-owned held-out data).
- protean_cnn.py: extract_embeddings now moves batches to the model device.
- posthoc_statistical_audit.py: refuse to write an empty table.
 2026-10-17 04:08:38 +07:00
willtanoe 783127296a docs: refresh README and session handoff for current state 
- README: add deep-comparison, post-hoc audit, and defense-preparation phases;
  correct the not-computed list (exploratory post-hoc tests now exist, with
  explicit small-n limits); add kisikisithesis/ and POSTHOC_STATISTICAL_AUDIT.md
  to the structure, artifact index, key documents, and reproduction commands.
- SESSION_HANDOFF.md: reflect HEAD 34bab45 and the post-16 sessions, document
  the prototype-snapshot tracking rule, refresh the next-session prompt.
 2026-10-17 04:08:38 +07:00
willtanoe 111adb272e research: add post-hoc statistical audit and defense preparation aids 
- scripts/posthoc_statistical_audit.py + docs/research/POSTHOC_STATISTICAL_AUDIT.md:
  read-only paired audit of the completed grid (exact sign-flip / Wilcoxon,
  selection stability, per-class/per-client summaries); writes only posthoc__*.
- kisikisithesis/: robustness audit and examiner-preparation document with
  deterministic read-only checks (robustness_checks.py).
- artifacts/analysis/posthoc__*.csv: audit outputs (6 tables).
- artifacts/analysis: track the 4 remaining beta0-snapshot effect/weight CSVs
  (xiiotid a0.5 seeds 1-2), completing the tracked beta0 set.
- .gitignore: codify that non-beta0 prototype snapshots stay untracked.
 2026-10-17 04:06:04 +07:00
willtanoe 34bab45865 docs: improve readability of PROTEAN comparison figures 2026-10-09 14:19:33 +07:00
willtanoe ad8b2d7671 research: add paired original vs proposed PROTEAN analysis 2026-10-09 13:40:20 +07:00
willtanoe 87f9b14ada docs: redesign final research figures and update README 2026-10-09 00:47:01 +07:00
willtanoe 8a9bdcf63c docs: document completed PROTEAN experiments and findings 2026-10-09 00:20:24 +07:00
willtanoe 8255393da9 docs: refresh README for finalized research state 2026-10-02 14:35:14 +07:00
willtanoe eaa9217552 docs: refresh project handoff after proposal finalization 2026-10-02 14:21:50 +07:00
willtanoe e27c5a83ab docs: finalize thesis proposal markdown 2026-10-02 14:13:59 +07:00
willtanoe b74a81f6c4 docs: strengthen thesis proposal literature and positioning 2026-10-02 13:28:49 +07:00
willtanoe 9b25d1b5e4 docs: add thesis proposal draft and source map 2026-10-02 13:02:46 +07:00
willtanoe 2ab6d74433 docs: update readme for completed research study 2026-10-01 01:07:23 +07:00
willtanoe 8a571a2387 research: finalize experimental synthesis 2026-10-01 01:00:32 +07:00
willtanoe d30e9831e6 experiment: complete two-dataset beta study 2026-09-30 22:03:46 +07:00
willtanoe bbd3508506 experiment: validate 5gnidd beta selection at alpha 0.75 2026-09-29 23:56:12 +07:00
willtanoe a7f2b0049e experiment: validate 5gnidd beta selection at alpha 0.50 2026-09-28 21:58:30 +07:00
willtanoe 2d27abc350 experiment: complete xiiotid three-alpha beta study 2026-09-27 15:12:58 +07:00
willtanoe ff68d380e2 experiment: validate beta selection at alpha 0.25 2026-09-26 11:23:38 +07:00
willtanoe 6b97a0d38b experiment: validate beta selection across three seeds 2026-09-25 13:14:28 +00:00
willtanoe 50f2da7216 experiment: validate beta selection on one paired full condition 2026-09-25 04:22:55 +00:00
willtanoe ad9aa53ba7 feat: implement support-weighted prototype aggregation 2026-09-24 23:37:26 +00:00
willtanoe 16b3add2cf docs: validate baseline and add deterministic regression coverage 
Session 3 baseline validation (no beta/support-weighted code added).

- Two measured full-data original_protean conditions (X-IIoTID and 5G-NIDD,
  alpha=0.5, seed=0); frozen manifests reused and reload-verified.
- Partition consistency, per-class/per-client inspection, round-0, loss audit,
  inference validation, reproducibility and a deterministic end-to-end test.
- S_j participant rule remains an explicitly labelled REPRODUCTION ASSUMPTION;
  |S_j| < M quantified in 53/162 class-condition cells (32.7%), 15/18 conditions.
- Adds docs/baseline/BASELINE_VALIDATION.md, tests/test_baseline_validation.py,
  scripts/validate_sj_participant_rule.py and small analysis artifacts.
- Test suite: 46 passed.

planningawal/ and raw datasets unchanged.
 2026-09-24 20:42:56 +00:00
willtanoe 374651a0fe docs: add project agent instructions 2026-09-24 17:41:09 +00:00
willtanoe 0aa8e2ba79 feat: implement reproducible PROTEAN baseline pipeline 2026-09-24 17:39:00 +00:00
willtanoe 2cc6a9859e chore: checkpoint PROTEAN research planning and architecture 2026-09-24 17:08:56 +00:00