Bert Sdg French

🚀 ilovebots/bert-sdg-french

This model enables the classification of texts according to the United Nations' Sustainable Development Goals (SDGs).

Source: https://www.un.org/development/desa/disabilities/about-us/sustainable-development-goals-sdgs-and-disability.html

🚀 Quick Start

Use the following code to start using the model.

from transformers import AutoTokenizer, AutoModelForSequenceClassification
tokenizer = AutoTokenizer.from_pretrained("ilovebots/bert-sdg-french")
model = AutoModelForSequenceClassification.from_pretrained("ilovebots/bert-sdg-french")

✨ Features

This text classification model can classify texts based on the United Nations' Sustainable Development Goals (SDGs).

📦 Installation

No specific installation steps other than the code in the quick start section are provided in the original document.

💻 Usage Examples

Basic Usage

from transformers import AutoTokenizer, AutoModelForSequenceClassification
tokenizer = AutoTokenizer.from_pretrained("ilovebots/bert-sdg-french")
model = AutoModelForSequenceClassification.from_pretrained("ilovebots/bert-sdg-french")

📚 Documentation

Model Details

Description of the Model

This text classification model was developed by fine - tuning the pre - trained model dbmdz/bert - base - french - europeana - cased. The training data for this fine - tuned model comes from the publicly accessible OSDG community dataset (OSDG - CD) at https://zenodo.org/record/5550238#.ZBulfcJByF4. This model was created as part of an academic research project at [Université Laval](https://www.ulaval.ca/developpement - durable/objectifs - de - developpement - durable - de - lonu). The goal was to create a French - based SDG text classification model using transformers. The main details of the model are highlighted below:

Model Sources

• Repository: https://huggingface.co/ilovebots/bert - sdg - french

Training Data

The dataset available at https://zenodo.org/record/5550238#.ZBulfcJByF4 has been enriched with the United Nations' Sustainable Development Goals and translated into French.

Training Hyperparameters

• Num_epoch = 4
• Learning rate = 2e - 5
• Epsilon = 1e - 8
• Optimizer = AdamW
• Batch size = 32
• Seed random = 42

Evaluation

Metrics

• Accuracy = 0.84


🔧 Technical Details

The model is fine - tuned from the pre - trained model dbmdz/bert - base - french - europeana - cased. The training data is from the OSDG community dataset. The hyperparameters and evaluation metrics are also provided, which shows the technical process of model training and evaluation.

📄 License

The model is released under the mit license.

Citation

Martinez, D.F. (2023). SDG classification with BERT. https://huggingface.co/ilovebots/bert - sdg - french