Sloan Digital Sky Survey data exploration Final project for PC120: Topics in Physics: Cosmology, Antigravity, and the Runaway Universe, a class I took at Colorado College during high school. Idea was to train a machine learning model to identify star systems containing exoplanets using spectral data from the Sloan Digital Sky Survey for training data and data from the Keppler space telescope to verify the data. Ended up having trouble with .h5 files so I ended up using stellar images instead of spectral data. I successfully training some models using Keras (which I believe has since been incorporated directly into Tensorflow), although the amount of data was insufficient to get reliable results. Also, a lot of the images were false color so I would have been surprised if it worked.
- Exoplanet_Classification.ipynb: Jupyter Notebook containing my data exploration, including accessing the data using the SDSS API.
- A Classification of Exoplanet Systems.pdf: Presentation about the project.