This repository contains my work for the capstone project of the IBM Data Science course sequence offered on Coursera. All of the work is based on notebook and code templates provided by IBM, cleaned up and edited by myself. The code has been edited in order to complete the assignments as well as to ensure it runs properly on my local machine.
In this project, we pose as a data scientist working for a new rocket company that would like to compete with SpaceX. SpaceX offers relatively inexpensive rocket launches, largely due to the fact that they can usually recover and reuse the "first stage" of the launch, which does most of the work of sending a payload into orbit. Thus, if we can determine whether we can recover and reuse the first stage, we can largely determine the cost of a launch. Our goal is to predict whether we can reuse the first stage based on several factors such as payload mass, customer, and orbit.
/dashboard- contains the dashboard app created with dash and plotly/data- contains the cleaned copies of the data used for our analyses/notebooks- contains the assignment notebooks comprising data collection, cleaning, and analysis/plots- .png files used for the final report/report- contains the final report, summarizing our findings in a ~50 slide powerpoint presentation
- Lab 1 - Collecting Falcon 9 rocket launch data from the API provided by SpaceX.
- Lab 2 - Collecting Falcon 9 rocket launch data by scraping data from Wikipedia.
- Lab 3 - Data cleaning including encoding categorical variables.
- Lab 4 - EDA using seaborn and matplotlib for data visualization.
- Lab 5 - Analytical EDA using SQL queries within python.
- Lab 6 - Creating interactive maps of the Falcon 9 launch sites using Folium.
- Lab 7 - Predicting launch success using scikit-learn.