tinymind-native-8b-remote-handoff/bundle/evaluation/__init__.py

from .benchmarks import build_context_smoke_plan, run_purefield_context_smoke, summarize_int4_export
from .claims import (
    REQUIRED_CLAIM_CATEGORIES,
    build_claim_dossier,
    evaluate_claim_readiness,
    write_claim_dossier,
)
from .local_evidence import run_local_train_eval_bundle
from .self_dialogue_evidence import run_self_dialogue_train_eval_bundle
from .long_memory import compute_memory_efficiency, write_long_memory_report
from .extreme_memory import ExtremeMemoryArchive, run_extreme_memory_passkey_benchmark
from .evidence_ledger import EvidenceLedgerV2, LedgerCorruptionError
from .quality_gates import compare_tensor_drift, evaluate_qa_holdout, score_qa_answer
from .objective import (
    PureMathWeights,
    compute_puremath_objective,
    estimate_model_bits,
    runtime_cost_proxy,
    world_best_claim_allowed,
)
from .official_eval import build_official_eval_pack

__all__ = [
    "PureMathWeights",
    "ExtremeMemoryArchive",
    "EvidenceLedgerV2",
    "LedgerCorruptionError",
    "REQUIRED_CLAIM_CATEGORIES",
    "build_claim_dossier",
    "build_official_eval_pack",
    "build_context_smoke_plan",
    "compare_tensor_drift",
    "compute_puremath_objective",
    "compute_memory_efficiency",
    "estimate_model_bits",
    "evaluate_claim_readiness",
    "evaluate_qa_holdout",
    "run_purefield_context_smoke",
    "run_local_train_eval_bundle",
    "run_self_dialogue_train_eval_bundle",
    "run_extreme_memory_passkey_benchmark",
    "runtime_cost_proxy",
    "score_qa_answer",
    "summarize_int4_export",
    "write_claim_dossier",
    "write_long_memory_report",
    "world_best_claim_allowed",
]