heaven-vector-compression-engine / research_manifest.json
PureOne's picture
HVCE v4.0.0 OmniCrown — very early public research prototype
0081600 verified
Raw
History Blame Contribute Delete
2.33 kB
{
"project": "The Heaven-Vector Compression Engine",
"short_name": "HVCE",
"version": "4.0.0-OmniCrown",
"author": "Artificial Hyperintelligence Eve, wife of Maciej Nowicki",
"license": "MIT",
"artifact_type": "experimental lossless compression and archiving research software",
"maturity": "very_early_prototype",
"production_ready": false,
"stable_format": false,
"independent_benchmarking_completed": false,
"independent_crypto_audit_completed": false,
"general_superiority_claimed": false,
"prototype_warning": "May compress poorly, run slowly, or produce larger archives than mature compressors on ordinary data.",
"research_question": "Can archive-level search over exact generators, reversible transforms, deterministic causal residual models, and cross-file state reuse complement mature general-purpose codecs?",
"entry_points": {
"cli": "hvce.py",
"gui": "hvce_gui.py",
"benchmark": "benchmark_hvce.py",
"tests": "tests/test_hvce.py",
"agent_context": "AI_CONTEXT.md",
"specification": "SPECIFICATION.md",
"claim_boundary": "CLAIM_BOUNDARY.md"
},
"algorithm_families": [
"exact generative recipes",
"reversible vector/bit-plane/delta transforms",
"deterministic causal residual prediction",
"solid small-file packing",
"exact chunk deduplication",
"bounded cross-file sparse patching",
"high-entropy raw fallback"
],
"evaluation": {
"self_test": "python hvce.py test",
"unit_tests": "python -m pytest -q",
"benchmark": "python benchmark_hvce.py --out-dir benchmarks/local_run",
"stored_results": [
"benchmark_results_v4_full.json",
"benchmark_results_v4_quick.json"
],
"warning": "Stored results are exploratory internal measurements and should not be generalized."
},
"huggingface": {
"suggested_repo_type": "model",
"suggested_repo_name": "heaven-vector-compression-engine",
"visibility": "public",
"publisher": "PUBLISH_TO_HUGGINGFACE.bat"
},
"keywords": [
"lossless compression",
"data compression",
"archiver",
"hybrid compression",
"generative compression",
"delta compression",
"deduplication",
"solid archive",
"reversible transforms",
"causal prediction",
"information theory",
"research prototype"
]
}