Spaces:
Sleeping
Sleeping
Commit History
-
Switch to transformers5 SIE image; LightOnOCR as default recognition
Filip Makraduli- Base image: latest-cpu-default → latest-cpu-transformers5 (where the LightOnOCR adapter actually lives, fixes the ImportError on Space boot) - Recognition default: Florence-2-base-ft → LightOnOCR-2-1B (Florence-2 isn't loadable on transformers5 due to tokenizer API change; tracked in sie-internal#828) - Drop Florence-2 entries from the dropdown so users don't click into an error - Expand alternates: 3 Donut variants, 4 NER models (GLiNER multi/large/ PII + NuNER Zero) - Swap sample images: replace dense table/handwritten/multi-column with event-poster/slide/letter to better suit the new recognition model - Server: /api/health now reports cuda availability; UI auto-disables gpuRequired models on the CPU image4e0f10e
-
Dockerfile: multi-stage Node install (HF apt sandbox workaround)
d4ff4a4 -
Initial: document-ocr demo for HF Spaces
Single-container build of the document-ocr demo from superlinked/brave-new-demos. Extends ghcr.io/superlinked/sie-server:latest-cpu-default with Node 22, runs SIE on 127.0.0.1:8080 and the UI on 0.0.0.0:7860. Preloads three small models from the default bundle (Florence-2-base, Donut-CORD-v2, GLiNER-multi). Alternates are dropdown-selectable and lazy-load on first click. HF persistent /data is used as HF_HOME so model weights survive Space restarts.ffe59ba
