Crnn Mobilenet V3 Large Gen Hw

🚀 Doctr - Optical Character Recognition

Optical Character Recognition made seamless & accessible to anyone, powered by TensorFlow 2 & PyTorch

This project enables seamless and accessible optical character recognition, leveraging the power of TensorFlow 2 and PyTorch. It provides an easy - to - use solution for converting images to text.

🚀 Quick Start

The project's GitHub repository can be found at https://github.com/mindee/doctr.

💻 Usage Examples

Basic Usage

>>> from doctr.io import DocumentFile
>>> from doctr.models import ocr_predictor, from_hub

>>> img = DocumentFile.from_images(['<image_path>'])
>>> # Load your model from the hub
>>> model = from_hub('mindee/my-model')

>>> # Pass it to the predictor
>>> # If your model is a recognition model:
>>> predictor = ocr_predictor(det_arch='db_mobilenet_v3_large',
>>>                           reco_arch=model,
>>>                           pretrained=True)

>>> # If your model is a detection model:
>>> predictor = ocr_predictor(det_arch=model,
>>>                           reco_arch='crnn_mobilenet_v3_small',
>>>                           pretrained=True)

>>> # Get your predictions
>>> res = predictor(img)

🔧 Technical Details

Run Configuration

{
  "arch": "crnn_mobilenet_v3_large",
  "train_path": "train",
  "val_path": "val",
  "train_samples": 1000,
  "val_samples": 20,
  "font": "FreeMono.ttf,FreeSans.ttf,FreeSerif.ttf",
  "min_chars": 1,
  "max_chars": 12,
  "name": "crnn_mobilenet_v3_large_gen_hw",
  "epochs": 3,
  "batch_size": 64,
  "device": null,
  "input_size": 32,
  "lr": 0.001,
  "weight_decay": 0,
  "workers": 2,
  "resume": "crnn_mobilenet_v3_large_printed.pt",
  "vocab": "tamazight",
  "test_only": false,
  "show_samples": false,
  "wb": true,
  "push_to_hub": true,
  "pretrained": false,
  "sched": "cosine",
  "amp": false,
  "find_lr": false
}