Views
No views yet
ketos train --arch ppocrv6) and usable
with oxygraphos-ocr / Ppocrv6-OCR
(ppocrv6-ocr) or stock kraken ≥ 7.0.| File | Variant | Val char accuracy* | Codec | Size (approx.) | Notes |
|---|---|---|---|---|---|
ppocrv6_sam_heb_small.safetensors | small | 0.9872 | 28 Hebrew letters + space | ~12 MB | From scratch |
ppocrv6_sam_heb_medium.safetensors | medium | 0.9877 | 28 Hebrew letters + space | ~60 MB | From scratch |
ppocrv6_sam_heb_medium_ft.safetensors | medium | 0.9886 | ~1622 chars (multilingual head retained) | ~61 MB | Fine-tuned from a multilingual PP-OCRv6 checkpoint |
best_*.safetensors filename (validation character accuracy).ppocrv6_sam_heb_medium_ft.safetensors for best accuracy on this domain. Use small when latency / footprint matters. Prefer the Hebrew-only small / medium codecs if you want a locked Samaritan alphabet (no Latin / punctuation emissions).| Architecture | PPOCRv6Model (kraken), task recognition |
| Input | RGB line images, target height 96, baseline / polygon-dewarped lines (seg_type: baselines) |
| Normalization | NFD + whitespace normalize; BiDi reordering enabled at train time |
| Data | Samaritan Hebrew PAGE-XML line corpus (sam_44_mss_pango / additional split), compiled to binary Arrow |
| Min kraken | 7.0.0 (_kraken_min_version in metadata) |
.safetensors file under kraken_meta — no sidecar alphabet file is required.hf download johnlockejrr/ppocrv6-sam-heb ppocrv6_sam_heb_medium_ft.safetensors --local-dir ./models1ppocrv6-ocr -d cuda:0 infer \
2 -m ./models/ppocrv6_sam_heb_medium_ft.safetensors \
3 -f page --output-dir ./out page.xmlkraken -i line.png text.txt ocr -m ./models/ppocrv6_sam_heb_medium_ft.safetensorsmedium_ft retains a large multilingual output vocabulary; greedy decoding can occasionally emit non-Hebrew symbols on ambiguous linessmall / medium cannot emit characters outside their 28-letter (+ space) codecketos / PP-OCRv6 (AdamW+Muon, cosine schedule, bf16-mixed)--base-dir R / RTL-aware setup for Hebrewmedium / small: trained from scratch on the Samaritan binary datasetmedium_ft: fine-tuned from a multilingual medium checkpoint onto the same data (lower LR); final weights keep the large pretrained codecPpocrv6-OCR)