Nov 14 6 days ago 17:35 Adapt FireViewer presentations and sourced graphics to this repository Task-specific presentation from pinned evidence: metrics, annotation views, conversion contracts, geographic catalogues or audit status as appropriate. Keep original artifacts, source rights, visibility and qualification boundaries. No new training, inference or dataset modification. Oct 10 10:22 docs: separate Cadryl catalogues from the native PyTorch cascade 10:13 docs: English quick cards, variant guides and recorded comparisons Oct 7 00:45 Add four verified v2.5 LiteRT cascade models, Cadryl contracts and bilingual 680-image benchmark Sep 20 23:05 docs: add bilingual LiteRT evidence, figures and integration guides unicorn who dev
00:35 Complete FireViewer Android learning SDK, source and verification records unicorn who dev
00:33 Add on-device learning and optimizer resume: fireviewer_dfine_m_strict_v1 unicorn who dev
00:18 Add on-device learning and optimizer resume: fireviewer_yolo11m_strict_v1 unicorn who dev
00:16 Add on-device learning and optimizer resume: fireviewer_dinov3_pointing_pilot_v1 unicorn who dev
00:15 Add on-device learning and optimizer resume: fireviewer_rtdetr_v2_r50 unicorn who dev
00:15 Add on-device learning and optimizer resume: fireviewer_rfdetr_medium_v110 unicorn who dev
Sep 19 23:58 Add on-device learning and optimizer resume: fireviewer_rtdetr_v2_r50 unicorn who dev
23:58 Add on-device learning and optimizer resume: fireviewer_dinov3_pointing_pilot_v1 unicorn who dev
23:30 Add on-device learning and optimizer resume: fireviewer_rfdetr_medium_v110 unicorn who dev
23:30 Add on-device learning and optimizer resume: fireviewer_yolo11m_strict_v1 unicorn who dev
21:36 Add LiteRT conversion: FireViewer YOLO11-M strict v1 unicorn who dev