Usmanbhat/fiqh-ai-api

chore: remove search traceback debug code

fix: gracefully handle missing score column in _row_to_result

debug: capture and return traceback in /search endpoint

fix: implement true hybrid search by lifting FTS restriction on vector candidates

trigger rebuild to pull new database from dataset

fix: use torch.cuda.amp.autocast() for mixed-precision encoding

perf: use FP16 precision on CUDA and SQLite executemany

chore: host database on Hugging Face Datasets and keep git repo lightweight

feat: auto-detect GPU/MPS in generate_embeddings.py

feat: model auto-detection, startup warmup, RRF k=40, CAMeL-BERT default, switch_model helper

feat: add /debug/semantic endpoint + semantic status in /health

Deploy fully-indexed AI vector database and hybrid search

Implement backward-compatible Hybrid RRF Search support

Deploy Fiqh.ai FastAPI Backend with Database and Git LFS

initial commit