---
title: vitstr_small-korean
canonical_url: "https://www.modelscope.ai/models/Bing-su/vitstr_small-korean"
md_url: "https://www.modelscope.ai/models/Bing-su/vitstr_small-korean.md"
repository: Bing-su/vitstr_small-korean
last_updated: 2026-05-06
license: mit
parameters: 21.4M
tensor_type:
  - F32
library_name:
  - safetensors
  - pytorch
frameworks:
  - pytorch
language:
  - en
downloads: 131
stars: 0
---

# vitstr_small-korean

> vitstr_small-korean - An open-source model by Bing-su on ModelScope. Optical Character Recognition made seamless & accessible to anyone, powered by TensorFlow 2 & PyTorch

Bing-su/vitstr_small-korean is a 21.4M-parameter machine learning model on ModelScope. licensed under mit.

- **Repository**: Bing-su/vitstr_small-korean
- **License**: mit
- **Parameters**: 21.4M
- **Downloads**: 131
- **Stars**: 0
- **Last updated**: 2026-05-06

Source: https://www.modelscope.ai/models/Bing-su/vitstr_small-korean

---

<p align="center">
<img src="https://doctr-static.mindee.com/models?id=v0.3.1/Logo_doctr.gif&src=0" width="60%">
</p>

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

## Task: recognition

https://github.com/mindee/doctr

### Example usage:

```python
>>> 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)
```
### Run Configuration

{
  "arch": "vitstr_small",
  "train_path": "C:\\Users\\smartmind\\Desktop\\workspace\\test\\train_ocr\\train",
  "val_path": "C:\\Users\\smartmind\\Desktop\\workspace\\test\\train_ocr\\validation",
  "train_samples": 1000,
  "val_samples": 20,
  "font": "FreeMono.ttf,FreeSans.ttf,FreeSerif.ttf",
  "min_chars": 1,
  "max_chars": 12,
  "name": "vitstr_small-korean",
  "epochs": 15,
  "batch_size": 64,
  "device": 0,
  "input_size": 32,
  "lr": 0.001,
  "weight_decay": 0.01,
  "workers": 6,
  "resume": null,
  "vocab": "korean",
  "test_only": false,
  "show_samples": false,
  "wb": true,
  "push_to_hub": true,
  "pretrained": true,
  "sched": "onecycle",
  "amp": true,
  "find_lr": false
}