RobBERT Local Example

Table of Contents

• 📍 Overview
• 🗂️ Repository Structure
• 🚀 Getting Started
  • ⚙️ Installation
  • 🤖 Usage
• 🎗 License

---

📍 Overview

The RobBERT_Local_Example repository provides a comprehensive exploration of the Dutch-based BERT model, RobBERT. This project is designed to guide users from basic to advanced usage, starting with the high-level pipeline API from Hugging Face and progressively diving into the low-level APIs. The goal is to offer a deep understanding of the model’s execution process and its application to various natural language processing tasks.

---

🗂️ Repository Structure

└── RobBERT_local_example/
    ├── LICENSE
    ├── convert_tf_models.sh
    ├── convert_tf_models_quantized.sh
    ├── main.py
    ├── requirements.txt
    ├── tf-main.py
    ├── tflite-main.py
    └── utils.py

---

🚀 Getting Started

System Requirements:

• Python: version 3.11.5

⚙️ Installation

From source

1. Clone the RobBERT_local_example repository:

$ git clone https://github.com/Chmpy/RobBERT_local_example

2. Change to the project directory:

$ cd RobBERT_local_example

3. Install the dependencies:

$ pip install -r requirements.txt

---

🤖 Usage

From source

Run RobBERT_local_example using the command below:

$ python main.py
$ python tf-main.py
$ python tflite-main.py

Run the model conversion scripts before executing tflite-main.py:

$ ./convert_tf_models.sh
$ ./convert_tf_models_quantized.sh

🎗 License

This project is protected under the MIT License. For more details, refer to the LICENSE file.