Larngear Antocr Weight
Compact CRNN + CTC line recognizer for printed/typed Thai documents with Latin (English / European-accent) co-script. 33 MB, 254-character charset. Reads one text-line crop → string. Tuned for clean documents (forms,...
Source du modèle
- Type
- Other
- Taille du fichier
- 31,7 MB
- Version
- vBest
- Mis à jour
- 23 oct. 2026
- Source
- Hugging Face
Description de la source
larngear-antocr-crnn-th (v40)
Compact CRNN + CTC line recognizer for printed/typed Thai documents with Latin (English / European-accent) co-script. ~33 MB, 254-character charset. Reads one text-line crop → string. Tuned for clean documents (forms, government reports, financial statements, academic calendars/timetables).
This repo is artifacts only — no inference code. It pairs with the
recognizer in
larngear_AntOCR
(the _CRNN architecture + input normalization + CTC decode that these weights
are shape-locked to), used in production by larngear-docling.
Files
| File | What |
|---|---|
best.pt | CRNN weights — plain state_dict, load with torch.load(weights_only=True) |
classes.json | charset, {"chars": [...]} — 254 chars (Thai + Latin + accents + digits + punctuation) |
The two are a versioned pair: len(chars) + 1 (the +1 is the CTC blank) is
the model's output dimension. A mismatched pair decodes to garbage — always pull
both from the same revision.
Accuracy
| Benchmark | CER |
|---|---|
| AntOCR real-PDF line bench (3502 lines) | ~4.3% |
larngear-docling control corpus — region-aligned text CER (~20 pages, born-digital Thai PDFs) | 2.34% |
Out of scope (not trained for it): scene text, handwriting.
Usage
from huggingface_hub import hf_hub_download
from antocr.core import CRNNLineRecognizer # from larngear_AntOCR
repo = "jsaksrisuwan/larngear_antocr_weight"
weights = hf_hub_download(repo, "best.pt", revision="main")
classes = hf_hub_download(repo, "classes.json", revision="main")
rec = CRNNLineRecognizer(weights=weights, classes=classes)
texts = rec.recognize_batch([line_crop0, line_crop1, ...]) # grayscale np.uint8 crops
Line segmentation (finding the line crops) is the consumer's job —
larngear-docling does layout detection + line-crop before calling this.
Pin revision to a tag or commit SHA for reproducible deploys.
Sources
1 source
Sources
1 sourceVérifié 23 oct.
Artefacts du modèle
1 artefact
Artefacts du modèle
1 artefactExtraits de sources
2 extraits
Extraits de sources
2 extraits--- license: apache-2.0 language: - th - en tags: - ocr - thai - crnn - ctc - document-ocr pipeline_tag: image-to-text library_name: pytorch --- # larngear-antocr-crnn-th (v40) Compact **CRNN + CTC line recognizer** for printed/typed **Thai** documents with Latin (English / European-accent) co-script. ~33 MB, 254-character charset. Reads one text-line crop → string. Tuned for clean documents (forms, government reports, financial statements, academic calendars/timetables). This repo is **artifacts only** — no inference code. It pairs with the recognizer in [`larngear_AntOCR`](https://github.com/jakksaksrisuwan-tech/larngear_AntOCR) (the `_CRNN` architecture + input normalization + CTC decode that these weights are shape-locked to), used in production by `larngear-docling`. ## Files | File | What | |------|------| | `best.pt` | CRNN weights — plain `state_dict`, load with `torch.load(weights_only=True)` | | `classes.json` | charset, `{"chars": [...]}` — **254 chars** (Thai + Latin + accents + digits + punctuation) | The two are a **versioned pair**: `len(chars) + 1` (the +1 is the CTC blank) is the model's output dimension. A mismatched pair decodes to garbage — always pull both from the same revision. ## Accuracy | Benchmark | CER | |-----------|-----| | AntOCR real-PDF line bench (3502 lines) | ~4.3% | | `larngear-docling` control corpus — region-aligned text CER (~20 pages, born-digital Thai PDFs) | 2.34% | Out of scope (not trained for it): scene text, handwriting. ## Usage ```python from huggingface_hub import hf_hub_download from antocr.core import CRNNLineRecognizer # from larngear_AntOCR repo = "jsaksrisuwan/larngear_antocr_weight" weights = hf_hub_download(repo, "best.pt", revision="main") classes = hf_hub_download(repo, "classes.json", revision="main") rec = CRNNLineRecognizer(weights=weights, classes=classes) texts = rec.recognize_batch([line_crop0, line_crop1, ...]) # grayscale np.uint8 crops ``` Line segmentation (finding the line crops) is the consumer's job — `larngear-docling` does layout detection + line-crop before calling this. Pin `revision` to a tag or commit SHA for reproducible deploys.
Source context: 0 downloads · 0 likes · Pipeline image-to-text · Library pytorch · Repo jsaksrisuwan/larngear_antocr_weight
