FINAL FIX: Enable Face Verification weights and fix 0-byte image serving 67ac72b hoda.fakhar commited on about 7 hours ago
docs: update README with live URL and accurate tech details 26086fb hoda.fakhar commited on about 7 hours ago
docs: add direct live URL and free-tier performance note 77417bf hoda.fakhar commited on about 7 hours ago
fix: use tensorflow/tensorflow:2.15.0 base image to avoid 300MB pip download 5e0e5c7 hoda.fakhar commited on about 7 hours ago
debug: split pip steps with echo markers so build logs show progress 368c8bf hoda.fakhar commited on about 7 hours ago
UN-BRICK SPACE: Remove LFS trackers for images to fix hanging build ad27889 hoda.fakhar commited on about 7 hours ago
Fix permission error: switch from /app to /home/user in build script a01d457 hoda.fakhar commited on about 7 hours ago
Optimize Hugging Face build context and permissions 5ff59cc hoda.fakhar commited on about 7 hours ago
debug: add progress-bar to tensorflow install to show download progress in logs 79134ab hoda.fakhar commited on about 7 hours ago
fix: use tensorflow-cpu==2.15.1 (2.13.0 has ABI incompatibility with Debian Bookworm) 53c0ef5 hoda.fakhar commited on about 8 hours ago
fix: split pip install into steps for visibility, isolate tensorflow install aae2413 hoda.fakhar commited on about 8 hours ago
fix: downgrade to tensorflow-cpu==2.13.0, add pip timeout/retry flags to prevent build hang 271c6f2 hoda.fakhar commited on about 8 hours ago
fix: pin tensorflow-cpu==2.15.1, add 5min download timeout to prevent build hang 7189f3f hoda.fakhar commited on about 8 hours ago
fix: remove aggressive sharpening, fix gender confidence mismatch in live view 84b1835 hoda.fakhar commited on about 11 hours ago
feat: enable retinaface detector, download retinaface.h5 from HF Hub during build bb776bc hoda.fakhar commited on about 11 hours ago
fix: replace libgl1-mesa-glx with libgl1 for Debian Trixie 779b029 hoda.fakhar commited on about 12 hours ago
fix: remove retinaface from live view, clean unused import cdfeda4 hoda.fakhar commited on about 12 hours ago
fix: stop 500 crash + reliable model download from HF Hub ff31e2d hoda.fakhar commited on about 14 hours ago
fix: OOM fix - models baked into Docker image, mobile responsive, 1 worker ab8cebe hoda.fakhar commited on about 14 hours ago