Legal T5 Small Trans Fr En

🚀 legal_t5_small_trans_fr_en model

A model for translating legal text from French to English, trained on multiple parallel corpora.

🚀 Quick Start

The legal_t5_small_trans_fr_en model is designed for translating legal text from French to English. It was first released in this repository and trained on three parallel corpora from JRC-ACQUIS, Europarl, and DCEP.

✨ Features

• Based on the t5-small model, trained on a large parallel text corpus.
• A smaller model with about 60 million parameters, achieved by scaling down the baseline t5 model.
• Suitable for translating legal texts from French to English.

📦 Installation

This section is skipped as no installation steps are provided in the original document.

💻 Usage Examples

Basic Usage

Here is how to use this model to translate legal text from French to English in PyTorch:

from transformers import AutoTokenizer, AutoModelWithLMHead, TranslationPipeline

pipeline = TranslationPipeline(
model=AutoModelWithLMHead.from_pretrained("SEBIS/legal_t5_small_trans_fr_en"),
tokenizer=AutoTokenizer.from_pretrained(pretrained_model_name_or_path = "SEBIS/legal_t5_small_trans_fr_en", do_lower_case=False, 
                                            skip_special_tokens=True),
    device=0
)

fr_text = "quels montants ont été attribués et quelles sommes ont été effectivement utilisées dans chaque État membre? 4."

pipeline([fr_text], max_length=512)

📚 Documentation

Model description

The legal_t5_small_trans_fr_en model is based on the t5-small model and was trained on a large corpus of parallel text. It is a smaller model, scaling the baseline t5 model down by using dmodel = 512, dff = 2,048, 8 - headed attention, and only 6 layers each in the encoder and decoder. This variant has about 60 million parameters.

Intended uses & limitations

The model can be used for the translation of legal texts from French to English.

Training data

The legal_t5_small_trans_fr_en model was trained on [JRC - ACQUIS](https://wt - public.emm4u.eu/Acquis/index_2.2.html), EUROPARL, and [DCEP](https://ec.europa.eu/jrc/en/language - technologies/dcep) datasets, which consist of 5 million parallel texts.

Training procedure

The model was trained on a single TPU Pod V3 - 8 for a total of 250K steps, using a sequence length of 512 (batch size 4096). It has a total of approximately 220M parameters and was trained using the encoder - decoder architecture. The optimizer used is AdaFactor with an inverse square root learning rate schedule for pre - training.

Preprocessing

An unigram model was trained with 88M lines of text from the parallel corpus (of all possible language pairs) to get the vocabulary (with byte pair encoding), which is used with this model.

Pretraining

No specific details about pretraining are provided in the original document.

Evaluation results

When the model is used for the translation test dataset, it achieves the following results:

BibTeX entry and citation info

Created by Ahmed Elnaggar/@Elnaggar_AI | [LinkedIn](https://www.linkedin.com/in/prof - ahmed - elnaggar/)

🔧 Technical Details

The model is based on the t5 - small architecture. It uses specific hyperparameters (dmodel = 512, dff = 2,048, 8 - headed attention, 6 layers in encoder and decoder) to scale down the baseline t5 model. The training was carried out on a single TPU Pod V3 - 8 for 250K steps with a sequence length of 512 and a batch size of 4096. The AdaFactor optimizer with an inverse square root learning rate schedule was used for pre - training.

📄 License

This section is skipped as no license information is provided in the original document.