The Exoplanet Predictor Version 1.0 - Predict the number of planets of a system based on the stellar characteristics. The usage of this tool is through the command line (CMD).
To use this tool you have two options, the first one is to predict the number of planets that a system with only ONE star has. For this, enter the directory of the model_predictor.py file and open the CMD. And type:
model_predictor.py -h
This will display the help menu of the tool, in which you can find the 8 required inputs with their description.
For example, you will need to enter the following in the command line: model_predictor.py -t VALUE -r VALUE -m VALUE -mt VALUE -a VALUE -d VALUE -v VALUE -l VALUE
After you enter the required fields, the algorithm will do its job ;).
So, what happens if you need to predict the number of planets for multiple systems at once? Here is when the second option of usage comes to the game. For this option you will need to provide the path of the CSV (dataframe file) in which you store all the hundreds of systems. Please note that for the headers of your database it must contain st_teff st_rad st_mass st_met st_age st_dens st_radv st_logg, separated with a comma ",". If you need an example of what a dataframe file should look like please refer to Data/PSCompPars_2021.04.20_19.50.36.csv
This means you should enter the following in the command line: model_predictor.py -data PATH TO YOUR FILE
After you enter the required fields, it will promt you the following: The Predictions were saved to "predictions.txt" file. Each line of the TXT file corresponds to each prediction. PD: The datafile was cleaned, which means that every NaN value was removed.
You can find the predictions.txt file in the current directory.
The database used to train the algorithm was produced by the NASA Exoplanet Archive http://exoplanetarchive.ipac.caltech.edu