bartpho finetuned qa

Introduction

The BARTPHO-FINETUNED-QA model is a text generation model, part of the Transformers library, designed for question-answering tasks. It has been fine-tuned using specific hyperparameters to optimize its performance for such tasks.

Architecture

The model utilizes the mbart architecture, which is a variation of the BART model, adapted for multilingual tasks. It is compatible with inference endpoints, allowing for deployment in various applications.

Training

The model was trained from scratch, achieving a training loss of 0.0573. The training procedure employed a learning rate of 5e-05, a batch size of 8, and the Adam optimizer with specific beta values and epsilon. The training spanned 3 epochs using a linear learning rate scheduler. However, detailed information about the dataset used is not provided.

Training Hyperparameters

• Learning Rate: 5e-05
• Train Batch Size: 8
• Eval Batch Size: 8
• Seed: 42
• Optimizer: Adam (betas=(0.9, 0.999), epsilon=1e-08)
• LR Scheduler Type: Linear
• Num Epochs: 3

Framework Versions

• Transformers: 4.44.2
• PyTorch: 2.4.1
• Datasets: 3.1.0
• Tokenizers: 0.19.1

Guide: Running Locally

To run the BARTPHO-FINETUNED-QA model locally, follow these steps:

1. Install Dependencies: Ensure you have the required versions of the Transformers library, PyTorch, Datasets, and Tokenizers.
2. Clone the Repository: Use the download link or clone the repository to your local machine.
3. Load the Model: Use the Transformers library to load the model and tokenizer.
4. Run Inference: Pass your data to the model for inference.

For optimal performance, especially if working with large datasets or requiring fast processing times, consider using cloud GPUs such as those provided by AWS, GCP, or Azure.

License

Information about the license for this model has not been provided. It is important to verify the licensing terms before using the model for commercial purposes.