138 lines
14 KiB
JSON
138 lines
14 KiB
JSON
{
|
|
"release": "graphrag-v7-frozen",
|
|
"evaluation_kind": "fresh_query_group_holdout_plus_historical_regression",
|
|
"graph_freeze_sha256": "9e97794200520d0f2288e452c9e186c0291a54c63d2c1382f693a9753e4081cf",
|
|
"runtime_files": {
|
|
"app/__init__.py": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855",
|
|
"app/main.py": "ff1adeff8b770047139fe3f632ad837517c95ec04649849c4315d6f892f9eef3",
|
|
"app/routes.py": "dfb82171fea33d33b6e48cb238d2c19a29254b47a9a87308f4864cdbac53fa7b",
|
|
"app/security.py": "24f437aff6642477049f02fdccfb3429c15e610d044bbb66128903771510055f",
|
|
"requirements.txt": "26a6d097727eba03c1b6c71cd02b06373ef01ce03e567dd2bc4357ae330e5a1a",
|
|
"tools/__init__.py": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855",
|
|
"tools/answer_evaluation/__init__.py": "795be75bf60e4aa9209e5b820207e52e8420c41a4dc2e37d0b6018cd2934a51a",
|
|
"tools/answer_evaluation/answer_format.py": "86a27b4aebc6a5077e13794b81957a1cf063a1f3e070c583120ba53c1788d2f8",
|
|
"tools/answer_evaluation/check_context_scenarios.py": "99dbae2747183622e0823cd8f6536f4ff38ef56d0e41dd7855096d47af3b1c37",
|
|
"tools/answer_evaluation/check_deployment.py": "14618b102c0ade954dc73cbc5036a628166f0eb703d435b679d1a000d045ab5a",
|
|
"tools/answer_evaluation/check_rag_runtime.py": "1e1b8bd48ac4d668ec057fee162ef30d34ae20ae026361bc8cc296bc455d4067",
|
|
"tools/answer_evaluation/checked_generation.py": "cb821a65327424c19cbd996190b14e6aa30d9b909493f11716552f71249a0529",
|
|
"tools/answer_evaluation/context_policy.py": "0dc2a056ca0d886d7cc8248ed4632c910c5e35bebd03ef94573df41802297f47",
|
|
"tools/answer_evaluation/evaluate_graph_answers.py": "d81504e8fe51598c6bd1b0d482f61b9b6f51f87ba3f38b03ae1bc584675a3b76",
|
|
"tools/answer_evaluation/graph_answer_prompt.txt": "fc9f2f8d52e70a8d67b9f4814b3f15e6be64bf88afe53ce4db4f037bde00e064",
|
|
"tools/answer_evaluation/rag_delivery.py": "b83f93d446f44fdd7d41c12eeeaae69c0b02908076624f4eb3f4f72b9b88dbc8",
|
|
"tools/answer_evaluation/regrade_citations.py": "330f437aec12927fd8c99b0aa601373f6ab26cb111971a6fb06098931d1e4aab",
|
|
"tools/answer_evaluation/render_answer.py": "609c534a37899b3febcea32d983da4b806c287de8f2ddf9bc0675e3280859002",
|
|
"tools/graphrag/__init__.py": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855",
|
|
"tools/graphrag/build_graphrag.py": "256292101d6d7d1946e926e2717ac45465b169f4091c51ecfed4f1a779dfd700",
|
|
"tools/graphrag/evaluation/__init__.py": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855",
|
|
"tools/graphrag/evaluation/audit_benchmark.py": "6619c818a52fc284e87ed6c48a503a54a3c9dd95447463ed78f075b7c58bad80",
|
|
"tools/graphrag/evaluation/benchmark_v2.py": "62a0b6af852d5ae8060c06db3f983ef4b797d74443f9df25a6b8bd77bf133330",
|
|
"tools/graphrag/evaluation/benchmark_v3.py": "30bae6c605ed26f4b49973acec799087c4e2b97a883ccedcafb965dd909c4c4d",
|
|
"tools/graphrag/evaluation/benchmark_v4.py": "7a731fe649d3d5170045b8bbcabf31ad527e5d26a934d434a4be3a4c94dc5c38",
|
|
"tools/graphrag/evaluation/benchmark_v6.py": "f955e600d17358fafcb33bab772aef09d787452f6b33352e299fdd42669aaf34",
|
|
"tools/graphrag/evaluation/benchmark_v7.py": "9477cd6e9a6dd0558837cf044535de75a8f101b17115e28e57a82d96d1aa01cb",
|
|
"tools/graphrag/evaluation/check_openwebui.py": "863801048231d474ba82930cf6bf6853b4410f8295b3f97fd3d5934852431fbe",
|
|
"tools/graphrag/evaluation/evaluate_graphrag.py": "22f48e9ac6dfae92d79ad648ca173547c6a37f98aa8199f7c37595f1b0b186f9",
|
|
"tools/graphrag/evaluation/evidence_contract.py": "78e24264c176e163e1782fef1634b6c45c7f31bd4692726d02a30ab45a06648e",
|
|
"tools/graphrag/evaluation/freeze_graphrag.py": "d8764d1aeb72208de55604785acff880002da1f818d8a8ad9477cf39e1f29efb",
|
|
"tools/graphrag/evaluation/graphrag_evaluator.py": "7f99bf0f97974058650b619ea49a8ee32284cb4b324f1ffab9baf54f5c4495f4",
|
|
"tools/graphrag/evaluation/reproducibility.py": "0e2dab6ffb17691f05f9d81dbec12ffc0f2e4e0b1a2114a75b66a5165cf7b4d2",
|
|
"tools/graphrag/evaluation/source_oracle.py": "aed0309435fd21bc274fde7905a0720f68f359b3c30bb0b912e3371ebf0d8072",
|
|
"tools/graphrag/evaluation/source_validation.py": "9671b8978623888dbed8b8b2d1b6b01c1ce73cb0a512b03919e48873990f15ed",
|
|
"tools/graphrag/graph_db.py": "c4b1a2edb765856b807398e4c243a8d17753529a5d2dfdb51791e28bc8de17f7",
|
|
"tools/graphrag/graph_search.py": "c41cbfbac0d5daa1d0b51f5f4b798a15688bbd2e0498d2cbe93c6584a99f6b7d",
|
|
"tools/graphrag/graphrag.py": "dbc75403aa038e4021b5a70d1f17c62784b478252a94bac8db682bf38dcb39b3",
|
|
"tools/graphrag/neo4j_smoke.py": "fb5244c72c29ef2acf96d1e7f54335ff168eed4119db1692025eb1b99194b7bb",
|
|
"tools/graphrag/publication.py": "fba1fa0d0bed325fd87f8d3593baae9271de1766b772d7214c7f63fd3f1c37b6",
|
|
"tools/graphrag/tests/__init__.py": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855",
|
|
"tools/graphrag/tests/test_evaluation_pipeline_v2.py": "63d1b3e80ae3c7b563374fd7ab4aa5ffad3c614eafbd15142c577abd7f756bf9",
|
|
"tools/graphrag/tests/test_evaluation_v5.py": "9387e9d1418ebab87530986b0e29dac31cf33d762777a29a31cff24cc72e552a",
|
|
"tools/graphrag/tests/test_evaluation_wide.py": "1b28bb74fd6b9457c869186b88606f1f7863d573aa3ab5226162e878f236d99e",
|
|
"tools/graphrag/tests/test_evaluator_connected_v2.py": "29d7c9838ed1ff37ea3276df6feb1b6142158c43a0525712f490b648181882de",
|
|
"tools/graphrag/tests/test_graph_publication.py": "96e53c1ae28e36c7a4290e1e60c8b398bf095ceaf883e819e1a477e2672b11c0",
|
|
"tools/graphrag/tests/test_graph_rebuild_atomic.py": "fe37d4fff0b1f0fb1d565f0a96152b9b4e336837bc78b5977cf3b4879d7fbaf3",
|
|
"tools/graphrag/tests/test_graph_search.py": "eb58661aef9dafe6bf7188fcdc2d79cef6652d57c64d1b81ae81a52e8e0204d7",
|
|
"tools/graphrag/tests/test_graph_search_entity_v2.py": "0ef1a8ebd513c5a4507eb38fcf5ae6137c69e9901b5a2975ba513be5902ac693",
|
|
"tools/graphrag/tests/test_graph_search_neo4j.py": "fec2bf72de8f752f17c9164fee5754bdb2e153828855e47efbf0000dbe82cf3b",
|
|
"tools/graphrag/tests/test_graph_search_regressions.py": "9db7779aeb8ed9e8c0f92c1657011a81080aff3c928649e811358e3009a18c48",
|
|
"tools/graphrag/tests/test_graphrag.py": "9ee8629d6b528dd78c2ddfc25cee95e17613f45b454fc0a38d4c591862637a9f",
|
|
"tools/graphrag/tests/test_graphrag_evaluator.py": "1c6d08aeec9ae56ac4123239a5aed12a0081d66cc3f34b2047d95764579d7298",
|
|
"tools/graphrag/tests/test_graphrag_freeze.py": "c4b760baf4c7ebf13b16bd4cd3c51874f5512eef3b4c61470dc4d428b0303d9f",
|
|
"tools/graphrag/tests/test_graphrag_research_v2.py": "711008a91ab8a58ae0854e7f90329bcd42e5189fde2e2001589e97b0c22c5198",
|
|
"tools/graphrag/tests/test_graphrag_root_title.py": "085651fe6e6081521a5dde9554abfee93456250108cf15978df25378d886a46c",
|
|
"tools/graphrag/tests/test_source_oracle_v2.py": "0afdbe235e185ffe9f38aa667f90f5a6c20504fe29e9ac572642b39a8eb7faca",
|
|
"tools/graphrag/tests/test_title_labels.py": "2983ead68ad6e2e346d24af0fc96dc6771c163e9c2a351af426a1e0383998e6f",
|
|
"tools/index_state.py": "c33920a63d04009f11ef23ca9e9eee9bf86453b7d0586d027bf3b39156afea0d",
|
|
"tools/publication.py": "f99d445073390a733afc8d849b9fcd4b4b0890bb54d136db67d4015b78436e1a",
|
|
"tools/rag/__init__.py": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855",
|
|
"tools/rag/build_chunks.py": "fe060e25b138abcb689a6eb1b1e993d7017a4b3c5dd63e7cd275e1279da6053b",
|
|
"tools/rag/build_sqlite_index.py": "107c0987e9ec4086ff41f9e7ae30195a67be34616875635e705dc79236e79416",
|
|
"tools/rag/common.py": "8497c46d5c6c27a7c76df1568360ad5e9d00d5e5a7ef186e758960cc79d81d1c",
|
|
"tools/rag/embedding_utils.py": "a2ea0b5be1a1782efff1c85c0eabfedba043e0e8bf7eb8b6617eab60eb4559c3",
|
|
"tools/rag/evaluation/__init__.py": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855",
|
|
"tools/rag/evaluation/evaluate_rag_answers.py": "945c2da655537fb37fe82a08e10dc3ed66a2d90980921f61f61efb171a4a26a2",
|
|
"tools/rag/evaluation/evaluate_retrieval.py": "aba865751ab74d9c2b962548b1e476508dbe6cf7bf811bda79e25f853061cbd7",
|
|
"tools/rag/evaluation/metrics.py": "40b7d16f7729599c1d2fa2370f7cb40dfa04a968cdadef3c4fae83a168202c28",
|
|
"tools/rag/evaluation/rag_answer_data.py": "9a369e9ced254cf723a5d3006877959c6f144f4c770e382835af776f7f569762",
|
|
"tools/rag/evaluation/rag_answer_state.py": "da4ab19f22378899d5c13e93e803a978518f8edcdd62e4990a06aae12d8eb469",
|
|
"tools/rag/evaluation/rag_metrics.py": "3e95922df90dd1b757a550040a2d9bd4f857e17d179f60170d7cc881a08f94b6",
|
|
"tools/rag/evaluation/rag_runner.py": "031abf7d5b1a026eb199aa2a579bfb8baaa01175279d0018a364bfe50b76c3c1",
|
|
"tools/rag/evaluation/retrieval_runner.py": "fe3dd05459f825d72252b7a0841f28004fc81537096a7de487807e4d5f0c4c45",
|
|
"tools/rag/rag_document_expansion.py": "d565b9794e30bc0f387c791a0557d5d03f2c1fa2fe5db53e49a9b19b1158f48e",
|
|
"tools/rag/rag_query_evidence.py": "90fced0fecd8ef976a1537506d2e7e8e9d3ba22a6371131f978fb91df5cc0b19",
|
|
"tools/rag/rag_supplemental_expansion.py": "a593579e543149633e3146c3886536a9196aff43f4df6f3d7ade94ca16dd243a",
|
|
"tools/rag/rag_utils.py": "4f898cf0facc2ae9f2ea28a36f48d4ed37bd7fc716e097b319bcf4d36a4a4f03",
|
|
"tools/rag/rebuild_index.py": "6a4d86a551d1141c403a0f6570c568896af37faeaa8c67cac170d2e6a393ca06",
|
|
"tools/rag/scan_zpwiki.py": "ef770ce4edb6352ff0690af5783766c95062a0e56f638cca6e3d929973728273",
|
|
"tools/rag/search_core.py": "624c8ddfe3edf78d06014579b747a59b2e33c0191a7cf90cdfd624b52ed5e96b",
|
|
"tools/rag/search_db.py": "cc63aeaf2854e87b7448c3978959569b097074cdb16a9b2c16b2c37a4677fdbb",
|
|
"tools/rag/search_utils.py": "69597bf5785f7852b1131a0648d9d2c65b6babba9ad279581f08b40d185ee646",
|
|
"tools/rag/tests/__init__.py": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855",
|
|
"tools/rag/tests/conftest.py": "0c94402d5ca971233d9802ded81cdb39ae7dfddda5eb9126e2035f7599a8f1c7",
|
|
"tools/rag/tests/test_chunking.py": "94904a20cfed0512d508017eaadbb43ef45ca74a5e96631241f6ca481222c498",
|
|
"tools/rag/tests/test_common.py": "86f0aecb79b7a3764112c833ea3b732073c073a5d570ef8c99900fb2a3f04430",
|
|
"tools/rag/tests/test_database.py": "31fc3553d76127c2d7e2c113b9053e537a72bb92f57a9c24cad6a579a1ffc589",
|
|
"tools/rag/tests/test_e5_1_remaining.py": "a8a52d5204e7376f33bb1aef96aad20990126a0b92266596b7cf8dd0dcbea4bc",
|
|
"tools/rag/tests/test_evaluate_rag_answers.py": "2143f3c9efbacf7b9119eeb9ae69e6cd7ea3f428ff5076d8e9973d0943675296",
|
|
"tools/rag/tests/test_live_data.py": "3c5e04d4a29ec8360d89881be980dacacccf0e12594fe51a22ec23c32c98b3e3",
|
|
"tools/rag/tests/test_rag.py": "13e97c88bf4afe45a8d633b269d110890f0f646887cee861bbf55ada0be7ad24",
|
|
"tools/rag/tests/test_rag_document_expansion.py": "9adf39c7a9b683b1d40c36fce576272e71869b7c1236c371c1b077820874eb7e",
|
|
"tools/rag/tests/test_rag_metrics_morphology.py": "070ee314d326e4c7218e7c19959a94076a2b241c4f19b9ecd9ca1fb3cc2639d4",
|
|
"tools/rag/tests/test_rag_metrics_semantic.py": "3642394d2636edd7420589348fb75bf1d214b539983b7e620008d626deffabf0",
|
|
"tools/rag/tests/test_rag_query_evidence.py": "c3fb5115306171010759b0cfd0cd05d1439d6a57de049efea02330ae7c553bfc",
|
|
"tools/rag/tests/test_rag_query_evidence_topn.py": "ebf01ad57a52f030b908b7fb0312d8d32cf9f60ea2953c5a3fb8849c1f2dceaf",
|
|
"tools/rag/tests/test_rag_runner.py": "1813086999d1e9826b42c2ac7978ea77d29e1c4b12f5e606f1fba572829e24ea",
|
|
"tools/rag/tests/test_rag_runner_routing.py": "c9454e5dddfdb7afc013567811efec15545daf20733b7167f78f2dd9ff0380d6",
|
|
"tools/rag/tests/test_rag_runner_tool_prompt.py": "073403907717d8d388c28ea92d1d1f9426edab2e8bd320c8f6329bcbff082e6c",
|
|
"tools/rag/tests/test_rag_supplemental_expansion.py": "65449054e2d94396def0dc8d505a9573306adbf184820ddd6c9b5ea92e165f8b",
|
|
"tools/rag/tests/test_rebuild_index.py": "aa0688b81e4b699ce99e97a4b8c1a6d4a49c5c153caa02301ebd5a5d8be764a4",
|
|
"tools/rag/tests/test_scan_zpwiki.py": "324510ca504dff11ecf39b28aff26b44a358e14084b3b015c0c410690d8aa129",
|
|
"tools/rag/tests/test_search.py": "82c96d35719d2233e4636b6f25d89b22f004848d8bae46a9705ee384d4abe9a8",
|
|
"tools/retrieval_contract.py": "b8fa272c3babf924a3ce8b937fe7ae437cd959ba9f7667193ac67c18d2c5b58d",
|
|
"tools/sync_indexes.py": "5b909b4394c24e5ce22848594b16d9aea45fb06418f5cd92bb6aa6381a0ccf10",
|
|
"tools/tool_health.py": "8ad37345889ad58a737edac5960b5b6c5002af2397d60248f434e03fb4eddd89"
|
|
},
|
|
"deployment_configuration": {
|
|
"Dockerfile": "fc7aefe4a99222daedc321acf8359773dc24eff82be04e688507e135956a92a6",
|
|
"docker-compose.yml": "e33b65c58f155ac057b01be272bc323fab3ff7eb4f51c9e4ca1b53904cc17af4"
|
|
},
|
|
"measured_artifacts": {
|
|
"tools/graphrag/evaluation/graphrag_questions_v7.json": "0f78e9749de369d752b6e1079f174cf3f6b390f4718a672119dfd43fa5d9dfac",
|
|
"tools/graphrag/evaluation/results/final_v7_2026-10-26/graphrag_dev.json": "d1a5d7b6d5aff3c0002f36a60cebc07627a79ec6d57de119dd587e8ab70143b0",
|
|
"tools/graphrag/evaluation/results/final_v7_2026-10-26/graphrag_test.json": "07d3d73c1b7ab0805549d2d3ebea0c02d1c00eacf3d6092911e34e9787cd9612",
|
|
"tools/graphrag/evaluation/results/final_v7_2026-10-26/historical_graph_summary.json": "f0386fde85ed1779d1262264614d4d93faf90cca9a3671cd7591ff6b57759a9b",
|
|
"tools/graphrag/evaluation/results/final_v7_2026-10-26/summary.json": "cdc7809eb1033bdf27569ea80deb53b3e82cc2beb0141e64d6251756737a396c",
|
|
"tools/graphrag/evaluation/results/v7_preparation/source_audit.json": "7825d4f3e4e535eb5f49e975dfef363b9b813f1de5c97e2387e5c2f3e578e6a5",
|
|
"tools/graphrag/evaluation/results/v7_preparation/protocol.json": "7ae343bba36fe25e8603f00325fe2dec1f5753616db9ef9c54f74ccffc1a7d83",
|
|
"tools/graphrag/evaluation/results/v7_preparation/source_review.csv": "7ee21230e5dda7be9dc847b8ada6111d14d38f7b075a9c64d38f3ff69950b5c0",
|
|
"tools/graphrag/evaluation/results/answers_v7_dev_2026-10-26/answers_dev.json": "4af9b52c250c837707c0690d892ec833347b3d3ee71d338ec631f6450eef8045",
|
|
"tools/graphrag/evaluation/results/answers_v7_test_2026-10-26/answers_test.json": "19d5e6ef484495a0acc1402b2d8f6ba91b3b63b8229d1118cd75d763082c148a",
|
|
"tools/graphrag/evaluation/results/answers_v7_regression_2026-10-26/answers_regression.json": "4bad6d54955b7645d861ea1ae2b412625b1ea032297caa57984f3b0cf21e3a25"
|
|
},
|
|
"policy": "Runtime code, prompt and gold remained unchanged throughout the final TEST and historical regression. All failures retained.",
|
|
"secrets_excluded": [
|
|
".env",
|
|
"credentials",
|
|
"private keys"
|
|
]
|
|
}
|