Skip to content
A MLP Neural Network that trains on an exoplanet dataset.
Python
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.

Files

Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
data
.gitignore
LICENSE
README.md
data_utils.py
main.py
models.py
requirements.txt

README.md

MLP Exoplanet Hunting

Requirements

  • keras
  • numpy
  • requests
  • scikit
  • tensorflow
  • tqdm

Usage

git clone https://github.com/drewszurko/keras-exoplanet-analysis.git
cd keras-exoplanet-analysis/ 
pip install -r requirements.txt
python main.py

Datasets

The datasets were created and open-sourced by Winter Delta (WΔ) and can be found on Kaggle and GitHub. They have also been pre-split into training and testing files, exo_train.csv.gz and exo_test.csv.gz.

Dataset Descriptions

exo_train.csv.gz

  • 5087 observations
  • 3198 features
  • 37 confirmed exoplanet stars
  • 5050 non-exoplanet stars
  • Column 1 is the target column (0 = non-exoplanet star, 1 = confirmed exoplanet star)
  • Columns 2-3198 are feature columns that display the star's flux values over time

exo_test.csv.gz

  • 570 observations
  • 3198 features
  • 5 confirmed exoplanet stars
  • 565 non-exoplanet stars
  • Column 1 is the target column (0 = non-exoplanet star, 1 = confirmed exoplanet star)
  • Columns 2-3198 are feature columns that display the star's flux values over time
You can’t perform that action at this time.