glue mrpc

Introduction

The GLUE-MRPC model is a fine-tuned version of bert-base-cased, specifically trained on the GLUE MRPC dataset. It is designed for text classification tasks and achieves notable performance metrics such as an accuracy of 85.54% and an F1 score of 89.74%.

Architecture

This model utilizes the BERT architecture, specifically the bert-base-cased variant. It is part of the Transformers library and is implemented in PyTorch. The model is compatible with TensorBoard for training visualization and supports inference endpoints.

Training

The GLUE-MRPC model was trained using the following hyperparameters:

• Learning Rate: 5e-05
• Training Batch Size: 16
• Evaluation Batch Size: 16
• Seed: 42
• Optimizer: Adam (betas=(0.9, 0.999), epsilon=1e-08)
• Learning Rate Scheduler Type: Linear
• Number of Epochs: 3.0

The training framework versions used include Transformers 4.13.0.dev0, PyTorch 1.10.0+cu102, Datasets 1.15.2.dev0, and Tokenizers 0.10.3.

Guide: Running Locally

To run the GLUE-MRPC model locally, follow these basic steps:

1. Install Dependencies:

   pip install transformers torch datasets tensorboard
   

2. Clone the Repository:

   git clone https://huggingface.co/sgugger/glue-mrpc
   cd glue-mrpc
   

3. Load the Model:

   from transformers import AutoModelForSequenceClassification, AutoTokenizer
   model = AutoModelForSequenceClassification.from_pretrained("sgugger/glue-mrpc")
   tokenizer = AutoTokenizer.from_pretrained("sgugger/glue-mrpc")
   

4. Inference: Prepare your input data and use the model to make predictions.

5. Optional: Utilize cloud GPUs for faster training and inference, such as those provided by AWS, Google Cloud, or Azure.

License

The GLUE-MRPC model is licensed under the Apache-2.0 License, which allows for both personal and commercial use, distribution, and modification.