Predicting the Variables of Exoplanets : Machine Learning Approach
What does it do?
In this task, AutoGluon is used to predict key variables of exoplanets, including mass (MP), radius (RP), temperature (TP), atmospheric metallicity (logZ), and carbon-to-oxygen ratio (C/O). The predictions are based on the transit depth of exoplanets as measured through specific filters, such as Sloan and Bessel
How does it work?
AutoGluon works by analyzing the transit depth data, which is processed through various machine learning algorithms. It automatically handles data preprocessing and model selection, optimizing hyperparameters to determine the best predictive model for the exoplanet variables. The framework takes the filtered transit depth data as input and outputs the predicted values for the exoplanet characteristics.
What benefits does it have?
Space Exploration: Accurate predictions help identify potentially habitable exoplanets, advancing our understanding of the universe. Cost-Effective Research: AutoGluon streamlines machine learning processes, saving time and resources for more research projects. Tools & Coding Language
Development Tools: Google Colab
Programming Languages: Python
Software: Canva
Link to our model : https://huggingface.co/kxm1k4m1/variables_of_exoplanets/tree/main/model