Sample Efficiency Evaluation

This project aims to measure the sample efficiency of different language model architectures using the BEAR dataset.

Getting Started

Prerequisites

• Python 3.9 or higher
• pip package manager

Installation

1. Clone the repository:

   git clone https://github.com/Jabbawukis/sample_efficiency_evaluation.git
   cd sample_efficiency_evaluation

2. Install the required dependencies:

   make install

3. Download the BEAR dataset from the following link: BEAR Dataset

4. Place the downloaded dataset in the appropriate directory as specified in your configuration.

Experiment Results

See sample-efficiency-evaluation-results