Image-to-Text
PaddleOCR
Safetensors
English
Chinese
pp_ocrv5_mobile_rec
OCR
PaddlePaddle
textline_recognition
Instructions to use tomsanbear/pp-ocrv5-mobile-rec with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PaddleOCR
How to use tomsanbear/pp-ocrv5-mobile-rec with PaddleOCR:
# 1. See https://www.paddlepaddle.org.cn/en/install to install paddlepaddle # 2. pip install paddleocr from paddleocr import TextRecognition model = TextRecognition(model_name="pp-ocrv5-mobile-rec") output = model.predict(input="path/to/image.png", batch_size=1) for res in output: res.print() res.save_to_img(save_path="./output/") res.save_to_json(save_path="./output/res.json") - Notebooks
- Google Colab
- Kaggle
| { | |
| "conversion_tool": "tools/models/candle/convert_ppocr_recognition.py@schema-v1", | |
| "dtype_device_results": [ | |
| { | |
| "device": "cpu", | |
| "dtype": "f32", | |
| "note": "native safetensors metadata and CTC probability shape contract smoke; Candle forward parity lands in impl-candle-recognition", | |
| "status": "passed" | |
| }, | |
| { | |
| "device": "cuda", | |
| "dtype": "f16", | |
| "note": "accelerator validation deferred until native recognition forward exists", | |
| "status": "unvalidated" | |
| }, | |
| { | |
| "device": "metal", | |
| "dtype": "f16", | |
| "note": "accelerator validation deferred until native recognition forward exists", | |
| "status": "unvalidated" | |
| } | |
| ], | |
| "generated_files": [ | |
| { | |
| "path": "model.safetensors", | |
| "sha256": "0bce74901be5dc8aef6d62a09eef40e46d9687fe7b087b37639e80f5d80219a3", | |
| "size_bytes": 31152668 | |
| }, | |
| { | |
| "path": "config.json", | |
| "sha256": "e33ccc6f68b0e710bbc5116b38995d8430f4ead7f6e15e745325e844cfe463f4", | |
| "size_bytes": 1134 | |
| }, | |
| { | |
| "path": "preprocessor_config.json", | |
| "sha256": "66c17f46fe813495d27e2c246de2256b5d7388780b8c87861f0fb47c5043427f", | |
| "size_bytes": 203006 | |
| }, | |
| { | |
| "path": "inference.yml", | |
| "sha256": "5dfeb2777f6d0db8177d8128a8acfcf6e6276dc4ac73ea3bf0dc06d6a5e85d8e", | |
| "size_bytes": 148345 | |
| }, | |
| { | |
| "path": "README.md", | |
| "sha256": "679f46bc3b144c11c91720aa4b3d4f68b46b79af6045fe761874e14c59b1aa20", | |
| "size_bytes": 1950 | |
| }, | |
| { | |
| "path": "ocr_pipeline.py", | |
| "sha256": "50cc6b790afc3c2073d45a94dd57843fa971d1dfa45bece8ccbf77830d5c27e9", | |
| "size_bytes": 5477 | |
| }, | |
| { | |
| "path": "tensor_map.json", | |
| "sha256": "23df8ce4046792700afc49382e4d5f097607b4160ecb73d7b461bfa0b450d8a3", | |
| "size_bytes": 336548 | |
| }, | |
| { | |
| "path": "bundle.json", | |
| "sha256": "357d82e2fed5d5c22181fa99be701ed048564f2af99ffc98f5f5b3de24af6fdd", | |
| "size_bytes": 5794 | |
| } | |
| ], | |
| "parity_fixture_ids": [ | |
| "axera-ppocrv5-rec-synthetic-line-ctc-probabilities", | |
| "axera-ppocrv5-rec-omnidocbench-line-text" | |
| ], | |
| "schema_version": 1, | |
| "source_files": [ | |
| { | |
| "path": "README.md", | |
| "sha256": "679f46bc3b144c11c91720aa4b3d4f68b46b79af6045fe761874e14c59b1aa20", | |
| "size_bytes": 1950 | |
| }, | |
| { | |
| "path": "config.json", | |
| "sha256": "e33ccc6f68b0e710bbc5116b38995d8430f4ead7f6e15e745325e844cfe463f4", | |
| "size_bytes": 1134 | |
| }, | |
| { | |
| "path": "inference.yml", | |
| "sha256": "5dfeb2777f6d0db8177d8128a8acfcf6e6276dc4ac73ea3bf0dc06d6a5e85d8e", | |
| "size_bytes": 148345 | |
| }, | |
| { | |
| "path": "model.safetensors", | |
| "sha256": "0bce74901be5dc8aef6d62a09eef40e46d9687fe7b087b37639e80f5d80219a3", | |
| "size_bytes": 31152668 | |
| }, | |
| { | |
| "path": "ocr_pipeline.py", | |
| "sha256": "50cc6b790afc3c2073d45a94dd57843fa971d1dfa45bece8ccbf77830d5c27e9", | |
| "size_bytes": 5477 | |
| }, | |
| { | |
| "path": "preprocessor_config.json", | |
| "sha256": "66c17f46fe813495d27e2c246de2256b5d7388780b8c87861f0fb47c5043427f", | |
| "size_bytes": 203006 | |
| } | |
| ], | |
| "tested_io": [ | |
| { | |
| "id": "ppocr-recognition-native-ctc-probability-shape-smoke", | |
| "inputs": [ | |
| { | |
| "dtype": "f32", | |
| "name": "pixel_values", | |
| "shape": [ | |
| 1, | |
| 3, | |
| 48, | |
| 320 | |
| ] | |
| } | |
| ], | |
| "outputs": [ | |
| { | |
| "dtype": "f32", | |
| "name": "ctc_probabilities", | |
| "shape": [ | |
| 1, | |
| 40, | |
| 18385 | |
| ] | |
| } | |
| ] | |
| }, | |
| { | |
| "id": "ppocr-recognition-source-dynamic-width-contract", | |
| "inputs": [ | |
| { | |
| "dtype": "f32", | |
| "name": "pixel_values", | |
| "shape": [ | |
| "batch", | |
| 3, | |
| 48, | |
| "width" | |
| ] | |
| } | |
| ], | |
| "outputs": [ | |
| { | |
| "dtype": "f32", | |
| "name": "ctc_probabilities", | |
| "shape": [ | |
| "batch", | |
| "width_div_8", | |
| 18385 | |
| ] | |
| } | |
| ] | |
| } | |
| ], | |
| "vocab_source": { | |
| "blank_index": 0, | |
| "body_size": 18383, | |
| "json_pointer": "/character_list", | |
| "path": "preprocessor_config.json", | |
| "policy": "embedded_config", | |
| "space_token_index": 18384, | |
| "vocab_size": 18385 | |
| } | |
| } | |