feat: OCR pipeline with NDJSON streaming and real-time progress (#226, #227, #231) #229

Merged
marcel merged 74 commits from feat/issue-226-227-ocr-pipeline-polygon into main 2026-04-13 12:39:04 +02:00
Showing only changes of commit 741979304c - Show all commits

View File

@@ -78,7 +78,7 @@ services:
dockerfile: Dockerfile
container_name: archive-ocr
restart: unless-stopped
mem_limit: 4g
mem_limit: 8g
memswap_limit: 8g
volumes:
- ocr_models:/app/models
@@ -86,8 +86,8 @@ services:
KRAKEN_MODEL_PATH: /app/models/german_kurrent.mlmodel
OCR_CONFIDENCE_THRESHOLD: "0.3"
OCR_CONFIDENCE_THRESHOLD_KURRENT: "0.5"
RECOGNITION_BATCH_SIZE: "1"
DETECTOR_BATCH_SIZE: "1"
RECOGNITION_BATCH_SIZE: "4"
DETECTOR_BATCH_SIZE: "2"
networks:
- archive-net
healthcheck: