Claude/amazing cori kw a2 v#101
Merged
Merged
Conversation
…ring train_models.jl (bag-of-words MVP), train_advanced_models.jl (GPU-assumed), train_complete_final.jl (logistic-regression copy-paste), train_final_models.jl (broken stubs), and train_and_evaluate.jl (synthetic-metrics only) are moved to src/julia/archive/ with a one-line deprecation header. run_training.jl is now the single canonical entry point. https://claude.ai/code/session_01YPqu7gti4azBach6ZvpRFJ
…/ artefacts Both run_training.jl and run_training_cpu.jl now produce the full set of canonical outputs after every training run: - models/neural/premise_selector.bson (NeuralSolver weights) - models/neural/tactic_predictor.bson (same weights, tactic side alias) - models/neural/vocab.json (token→id map + vocab_size + tactic_classes) - models/neural/gnn_ranker/ (best_model → gnn_ranker deploy alias) - training_data/metrics_baseline.jsonl (appended JSONL row with MRR/top-k/metadata) - models/model_metadata.txt (rewritten with real vocab size, classes, MRR) Adds `using Dates` to both scripts so Dates.now() is available at top level. https://claude.ai/code/session_01YPqu7gti4azBach6ZvpRFJ
…fallback load_gnn_model() now tries three candidate directories in order: gnn_ranker/ → best_model/ → final_model/ (all under models/neural/, written by run_training*.jl) The cosine similarity path (rank_with_cosine) is kept intact and reachable but only runs when all candidate directories are absent or fail to deserialise — it is the genuine missing-model fallback for CI smoke runs, not the default. The warning message now explains how to produce real weights (just train-cpu). https://claude.ai/code/session_01YPqu7gti4azBach6ZvpRFJ
train — full corpus run via run_training.jl (GPU auto-selected) train-cpu — ECHIDNA_MAX_PROOF_STATES=2000 ECHIDNA_NUM_EPOCHS=2 CPU smoke pass Both produce the canonical artefact set (premise_selector.bson, tactic_predictor.bson, vocab.json, model_metadata.txt, metrics_baseline.jsonl). https://claude.ai/code/session_01YPqu7gti4azBach6ZvpRFJ
hyperpolymath
enabled auto-merge (squash)
May 24, 2026 17:02
🔍 Hypatia Security ScanFindings: 261 issues detected
View findings[
{
"reason": "Issue in quality.yml",
"type": "missing_workflow",
"file": "quality.yml",
"action": "create",
"rule_module": "workflow_audit",
"severity": "high"
},
{
"reason": "Issue in security-policy.yml",
"type": "missing_workflow",
"file": "security-policy.yml",
"action": "create",
"rule_module": "workflow_audit",
"severity": "medium"
},
{
"reason": "Workflow executes remote script directly (curl/wget piped to shell). Download, verify checksum/signature, then execute.",
"type": "download_then_run",
"file": "mirror.yml",
"action": "verify_download_integrity",
"rule_module": "workflow_audit",
"severity": "high"
},
{
"reason": "Agda postulate assumes without proof -- potential soundness hole (1 occurrences, CWE-704)",
"type": "agda_postulate",
"file": "/home/runner/work/echidna/echidna/proofs/agda/Basic.agda",
"action": "flag",
"rule_module": "code_safety",
"severity": "critical"
},
{
"reason": "Agda postulate assumes without proof -- potential soundness hole (2 occurrences, CWE-704)",
"type": "agda_postulate",
"file": "/home/runner/work/echidna/echidna/proofs/agda/SoundnessPreservation.agda",
"action": "flag",
"rule_module": "code_safety",
"severity": "critical"
},
{
"reason": "Agda postulate assumes without proof -- potential soundness hole (1 occurrences, CWE-704)",
"type": "agda_postulate",
"file": "/home/runner/work/echidna/echidna/meta-checker/src/Echidna/AxiomSafety.agda",
"action": "flag",
"rule_module": "code_safety",
"severity": "critical"
},
{
"reason": "UNSOUND: --type-in-type collapses type hierarchy (Girard's paradox) (1 occurrences, CWE-704)",
"type": "agda_type_in_type",
"file": "/home/runner/work/echidna/echidna/meta-checker/src/Echidna/AxiomSafety.agda",
"action": "flag",
"rule_module": "code_safety",
"severity": "critical"
},
{
"reason": "User-defined axiom -- not verified by Lean kernel (1 occurrences, CWE-704)",
"type": "lean_axiom",
"file": "/home/runner/work/echidna/echidna/verification/proofs/lean4/IntegrityVerification.lean",
"action": "flag",
"rule_module": "code_safety",
"severity": "medium"
},
{
"reason": "User-defined axiom -- not verified by Lean kernel (1 occurrences, CWE-704)",
"type": "lean_axiom",
"file": "/home/runner/work/echidna/echidna/verification/proofs/lean4/ConfidenceLattice.lean",
"action": "flag",
"rule_module": "code_safety",
"severity": "medium"
},
{
"reason": "User-defined axiom -- not verified by Lean kernel (1 occurrences, CWE-704)",
"type": "lean_axiom",
"file": "/home/runner/work/echidna/echidna/proofs/lean/mvp_basic.lean",
"action": "flag",
"rule_module": "code_safety",
"severity": "medium"
}
]Powered by Hypatia Neurosymbolic CI/CD Intelligence |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
No description provided.