Table of Contents

• MLProject - v0.0.0 (unreleased)
  • Clone
  • Installation
    • Requirements
    • Data
  • Usage

MLProject - v0.0.0 (unreleased)

This repository contains a machine learning code to solve the Collision Avoidance Challenge from ESA released in October 2019.

Clone

• HTTPS: git clone https://github.com/freddida/MLProject.git
• SSH: git [email protected]:freddida/MLProject.git

Installation

Requirements

• Run pip install -r requirements.txt to install all dependencies

Data

• The training and testing data can be found here.
• Download the training data and add it to the folder /data/training as train_data.csv
• Download the testing data and add it to the folder /data/testing as test_data.csv

Usage

After installation and data setup, you can use the provided modules and scripts for various tasks such as data preprocessing, model training, and evaluation. Here's a brief overview of how to utilize the different components:

• Data Preprocessing: Utilize the functions in src/utils for loading, filtering, encoding labels, and calculating statistics on your dataset.
• Exploratory Data Analysis (EDA): Refer to the Jupyter notebook notebooks/exploratory/visualization.ipynb for visualizing and understanding the dataset.
• Feature Engineering: Explore and compare feature selection techniques provided in the notebooks/engineering directory, including dropping or imputing NaN values, and Principal Component Analysis (PCA).
• Modeling and Training: Use the Neural Network Notebook notebooks/modeling/neural_network_model.ipynb for building, training, and evaluating your machine learning models.