M5-B result: UniXcoder dense +4.5pp MRR over MiniLM (fair code model) 8e4942c CamQuestByte Claude Opus 4.8 commited on 2 days ago
M5: embedding-experiment infra + sub-exp A null result + UniXcoder (B) setup 4dfafb0 CamQuestByte Claude Opus 4.8 commited on 4 days ago
fix(data): use namespaced dataset id code-search-net/code_search_net 6e73712 CamQuestByte Claude Opus 4.8 commited on 5 days ago
M4: slim BM25 index (index-only save/load with id-order guard) 5132b10 CamQuestByte Claude Opus 4.8 commited on 7 days ago
M3: Hybrid RRF + MS-MARCO cross-encoder reranker b966217 CamQuestByte Claude Opus 4.7 commited on 7 days ago
Dense: minimal Qdrant payload + local query-vector cache ec8985c CamQuestByte Claude Opus 4.7 commited on 7 days ago
M2 done-criterion: qualitative BM25 vs dense comparison 5de6364 CamQuestByte Claude Sonnet 4.6 commited on May 11
M2 concept checkpoint: HNSW ef_search sweep 086a913 CamQuestByte Claude Sonnet 4.6 commited on May 11
M2 complete: dense retrieval via MiniLM + Qdrant, MRR@10=0.3891 Recall@100=0.7520 f8d9841 CamQuestByte Claude Sonnet 4.6 commited on Apr 29
M1 complete: BM25 baseline eval with correct code-token indexing d283064 CamQuestByte Claude Sonnet 4.6 commited on Apr 28
M0: fix qdrant-client API (search → query_points) e5b31ad CamQuestByte Claude Sonnet 4.6 commited on Apr 21
Initial commit: codesearch project with BM25 and dense retrieval 6a2ebe8 CamQuestByte Claude Sonnet 4.6 commited on Apr 21