Image-to-Text
Transformers
Safetensors
qwen3_5
image-text-to-text
ocr
table-recognition
document-understanding
Instructions to use lschiemanowski/tablereader-lightonocr-v0 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use lschiemanowski/tablereader-lightonocr-v0 with Transformers:
# Use a pipeline as a high-level helper # Warning: Pipeline type "image-to-text" is no longer supported in transformers v5. # You must load the model directly (see below) or downgrade to v4.x with: # pip install "transformers<5.0.0" from transformers import pipeline pipe = pipeline("image-to-text", model="lschiemanowski/tablereader-lightonocr-v0")# pip install -U transformers accelerate # Load model directly from transformers import AutoProcessor, AutoModelForMultimodalLM processor = AutoProcessor.from_pretrained("lschiemanowski/tablereader-lightonocr-v0") model = AutoModelForMultimodalLM.from_pretrained("lschiemanowski/tablereader-lightonocr-v0", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Download release.json from lschiemanowski/tablereader-lightonocr-v0: direct link, hf CLI and curl.
- Browser
- Download file 1.84 kB
-
https://huggingface.co/lschiemanowski/tablereader-lightonocr-v0/resolve/main/release.json
- Command line
-
hf download hf://lschiemanowski/tablereader-lightonocr-v0/release.json
-
curl -L -o release.json https://huggingface.co/lschiemanowski/tablereader-lightonocr-v0/resolve/main/release.json
1.84 kB
| { | |
| "release_id": "tablereader-lightonocr-v0", | |
| "contract_revision": 1, | |
| "preparation": { | |
| "dpi": 300, | |
| "reference_dpi": 300, | |
| "pad_px": 20, | |
| "text_gap_px": 60, | |
| "text_max_px": 450, | |
| "foot_max_px": 900, | |
| "foot_start_px": 40, | |
| "foot_line_gap_px": 30, | |
| "foot_first_margin": 0.04, | |
| "foot_margin": 0.12, | |
| "ink_min_px": 3, | |
| "word_gap_px": 40, | |
| "layout_model": "docling-project/docling-layout-heron@8f39ad3c0b4c58e9c2d2c84a38465abf757272d8", | |
| "layout_dpi": 72, | |
| "layout_threshold": 0.5, | |
| "stop_above": [ | |
| "table", | |
| "text", | |
| "list_item", | |
| "key_value_region", | |
| "picture", | |
| "page_header", | |
| "page_footer", | |
| "formula", | |
| "code", | |
| "document_index", | |
| "form", | |
| "checkbox_selected", | |
| "checkbox_unselected" | |
| ], | |
| "prose_labels": [ | |
| "text", | |
| "list_item", | |
| "key_value_region" | |
| ], | |
| "stop_min_lines": 3, | |
| "stop_below": [ | |
| "table", | |
| "picture", | |
| "page_header", | |
| "page_footer", | |
| "section_header", | |
| "title" | |
| ], | |
| "stop_overlap_px": 10, | |
| "stop_margin_px": 8 | |
| }, | |
| "renderer": "pypdfium2 5.15.1 / pdfium 159.0.8149.0@sourcebuild-toolchained", | |
| "layout_model": "docling-project/docling-layout-heron@8f39ad3c0b4c58e9c2d2c84a38465abf757272d8", | |
| "table_threshold": 0.5, | |
| "max_edge": 2048, | |
| "max_tokens": 16000, | |
| "temperature": 0.0, | |
| "enable_thinking": false, | |
| "base_model": "lightonai/LightOnOCR-3-4B", | |
| "training": { | |
| "run": "r1", | |
| "stage": "teacher", | |
| "examples": 995, | |
| "epochs": 3, | |
| "lora_rank": 16, | |
| "lr": 0.0001, | |
| "labels": "GPT-6.1 Terra transcriptions of 1,004 training crops, structurally accepted", | |
| "validation_records_1": { | |
| "all": 0.7967, | |
| "born_digital": 0.835, | |
| "scanned": 0.5112 | |
| }, | |
| "validation_parts": 1006 | |
| } | |
| } | |