Giulia Ventagli (data generation/pipeline, deterministic neural network & analysis)
Ippocratis Saltas (Bayesian neural network & analysis)
The data used for the machine learning analysis were generated on Julia (e.g. solution of ODEs, construction of mass-radius-tidal data, etc.). The deep learning models and analysis were done in Python using Tensorflow/Keras. The version of libraries used are as follows:
julia 1.10.2
DifferentialEquations v7.13.0
DelimitedFiles v1.9.1
Interpolations v0.15.1
Symbolics v5.22.1
Plots v1.40.1
Roots v2.1.2
Distributions v0.25.107
CSV v0.10.13
LaTeXStrings v.1.3.1
Images v0.26.0
python 3
tensorflow version 2.14.0
numpy version 1.24.3
pandas version 2.0.3
sklearn version 1.3.0
matplotlib version 3.5.1
statmodels version 0.14.1
IPython : 8.14.0
ipykernel : 6.25.0
ipywidgets : 6.0.0
jupyter_client : 8.3.0
jupyter_core : 5.3.1
jupyter_server : 2.7.0
jupyterlab : 4.0.3
nbclient : 0.8.0
nbconvert : 7.7.3
nbformat : 5.9.1
notebook : 6.1.5
qtconsole : 5.5.1
traitlets : 5.9.0
Data: This folder contains the various versions of the training/testing data we used.
DataGenerator: This folder containts the code and necessary data files needed to generate the training/testing data.
Datagenerator.ipynb: This is the notebook (Julia) which solves the equations and generates the training/testing data set we used in the deep-learning analysis.
Deterministic_Neural_Network.ipynb: This is the notebook (Python) for the analysis of the equation of state based on the deterministic neural network.
Baysian_Neural_Network.ipynb: This is the notebook (Python) for the analysis of the equation of state based on the Bayesian/probabilistic neural network.
model_det.py: The Python file containing the definition of the deep-learning model and the required functionality needed for the notebook "Deterministic_Neural_Network.ipynb".
model_Bayesian.py: The Python file containing the definition of the deep-learning model and the required functionality for the notebook "Bayesian_Neural_Network.ipynb".
If you use any part of this reproduction package for independent work we ask that you cite both this Zenodo package and the science paper.