vitstr_small_DoctrOcrEngine

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

Task: recognition

Example 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_resnet50',
>>>                           reco_arch=model,
>>>                           pretrained=True)

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

Training configuration and logs: https://wandb.ai/xbankov/text-recognition

Run Configuration

{ "hf_dataset_name": "fimu-docproc-research/born_digital_recognition", "name": "vitstr_small_10_512_32_0.07510091396452692_0.08993428426580115_cosine_da2de2d1_f3c04964", "epochs": 10, "lr": 0.07510091396452692, "weight_decay": 0.08993428426580115, "batch_size": 512, "input_size": 32, "sched": "cosine", "sample": null, "workers": 16, "wb": true, "push_to_hub": "fimu-docproc-research/vitstr_small", "test_only": false, "arch": "vitstr_small" }