Phase 6: Implement grouped dataset partitions #445
Labels
No labels
bug
cli
core
docs
event
experiment
figure
invariant
metrics
oracle
phase-0
phase-1
phase-2
phase-3
phase-4
phase-5
phase-6
provenance
revocation
tests
workload
No project
No assignees
1 participant
Notifications
Due date
No due date set.
Blocks
Depends on
#447 Phase 6: Fit training-only baseline preprocessing
erikinkinen/AES
#450 Phase 6: Fit interpretable temporal predictors
erikinkinen/AES
#453 Phase 6: Build discrete-time hazard rows
erikinkinen/AES
#476 Phase 6: Fit matched boosted-tree predictor pairs
erikinkinen/AES
#477 Phase 6: Add nonlinear instantaneous controls
erikinkinen/AES
#427 Phase 5: Execute phase 5 and release evidence
erikinkinen/AES
#440 Phase 6: Generate the fixed prediction corpus
erikinkinen/AES
#442 Phase 6: Implement temporal features and labels
erikinkinen/AES
Reference
erikinkinen/AES#445
Loading…
Reference in a new issue
No description provided.
Delete branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Plan ID:
P6-05Record type: Implementation leaf
Scientific protocol: 2.2 · Backlog revision: 2.2.1
Specification: sections 10.3; common contracts in sections 3–5, 11 and 17.4.
Source:
AES_PHASES_2_6_PLAN.md, issue heading at line 8134.Plan SHA-256:
8acdaeb52d6a94fcad7de64d64521cd465f608344d63560b45f9202f5af71212Protocol snapshot: Download the complete governing plan.
Goal
prevent counterfactual or near-duplicate leakage.
Prerequisites
Steps
Definition of Done
splits.csv accounts for every trace and all feature preparation
is fitted only on the appropriate training subset.
Execution and acceptance contract
Apply the governing semantic/data/statistical contracts in sections 3–5 and the relevant phase section, plus the section-11 checks applicable to this task. Section 17.4 defines the required acceptance evidence. Store the outcome and immutable prerequisite/input/output hashes at
artifacts/backlog/2.2.1/P6-05/acceptance.json.Only explicit dependency and parent-closure edges govern readiness; numbering adds no predecessor dependency. Checkboxes are work/acceptance items, not commit quotas. Keep implementation changes independently reviewable using the existing PR template. Execution, review and gate tasks close with their evidence/decision records and a PR only when tracked artifacts change.
An unsupported scientific effect is a valid result. Missing correctness validation, invalid inference or absent required review is not completion. Do not change endpoints, matrices, margins, inference rules or sample sizes outside the protocol change-control process.