Trocr Base Handwritten

🚀 Trocr-base-handwritten Model for Transformers.js

This project adapts the microsoft/trocr-base-handwritten model with ONNX weights to be compatible with the Transformers.js library. It enables optical character recognition for handwritten text in a web environment.

🚀 Quick Start

Prerequisites

Ensure you have Node.js and npm installed on your system.

Installation

You can install the Transformers.js JavaScript library from NPM using the following command:

npm i @xenova/transformers

✨ Features

• Optical Character Recognition: Perform OCR on handwritten text using the Xenova/trocr-base-handwritten model.
• Web Compatibility: Utilize the model in a web environment with the Transformers.js library.

💻 Usage Examples

Basic Usage

Here is an example of performing optical character recognition with the Xenova/trocr-base-handwritten model:

import { pipeline } from '@xenova/transformers';

// Create image-to-text pipeline
const captioner = await pipeline('image-to-text', 'Xenova/trocr-base-handwritten');

// Perform optical character recognition
const image = 'https://huggingface.co/datasets/Xenova/transformers.js-docs/resolve/main/handwriting.jpg';
const output = await captioner(image);
// [{ generated_text: 'Mr. Brown commented icily.' }]

Visual Example

📚 Documentation

Temporary Solution Note

⚠️ Important Note

Having a separate repo for ONNX weights is intended to be a temporary solution until WebML gains more traction.

Model Conversion Recommendation

💡 Usage Tip

If you would like to make your models web-ready, we recommend converting to ONNX using 🤗 Optimum and structuring your repo like this one (with ONNX weights located in a subfolder named onnx).

📄 Model Information