Machine Learning Project II

This is a project for the EPFL class Machine Learning CS-433 in collaboration with the ESA lab ESTEC.

In this project, we will predict coronal mass ejections with the data from EDAC counters and magnetic sensors of several spacecrafts.

This project contains our final report to explain everything in it.

Agatha DUZAN | Sebastien CHAHOUD | Nastasia MOREL

1 - Clone the repo

  git clone https://github.com/CS-433/ml-project-2-ml4esa

2 - Add the data

Download the data files and add them to the DATA folder.

Data fetching and preprocessing: all the code we used to fetch the data online and preprocess it can be found here. It also contains the first visualizations of the data.
DATA folder: contains the original data, scraped data and cleaned up data. It allows you to run the models without having to reproduce the scraping and preprocessing.
main notebook: Training and evaluation of our final model.
Other notebooks: other models that we experimented with (different architectures, input features and hyperparameters)
Final report: contains the project report which describes in detail the project, our approach, the different models, the results, and the ethical considerations.

Name		Name	Last commit message	Last commit date
Latest commit History 82 Commits
DATA		DATA
Final Report		Final Report
__pycache__		__pycache__
data_fetching		data_fetching
data_preprocessing		data_preprocessing
.gitignore		.gitignore
Base_Models.ipynb		Base_Models.ipynb
Grid_Search.ipynb		Grid_Search.ipynb
Prediction_models_1.ipynb		Prediction_models_1.ipynb
Prediction_models_2.ipynb		Prediction_models_2.ipynb
Prediction_models_3.ipynb		Prediction_models_3.ipynb
Prediction_models_4.ipynb		Prediction_models_4.ipynb
Prediction_models_4_colab.ipynb		Prediction_models_4_colab.ipynb
README.md		README.md
final_prediction_model.ipynb		final_prediction_model.ipynb
nn_helpers.py		nn_helpers.py
project2_description.pdf		project2_description.pdf
pt_helpers.py		pt_helpers.py