Deploying Machine Learning models for SpaceY who wants to bid against SpaceX
- 0 Collaborators
In this, we deployed different ML models for a company SpaceY that wants to bid against SpaceX. We have collected data regarding successful landings of SpaceX from SpaceX's public API and Wikipedia page. Then we have produces 4 ML models, which produced similar results of accuracy of about 83.33%. ...learn more
Project status: Concept
Intel Technologies
Optane,
10th Gen Intel® Core™ Processors,
Intel powered desktop PC,
Intel CPU
Overview / Usage
The age of Commercial Space travel is here. Space X has the best pricing which is $62millions. Nasa spends $165 million USD on average compared to $62 million from SpaceX.
SpaceX makes this happen Largely due to the ability to recover part of the rocket (Stage 1).
Space Y wants to compete with SpaceX, that's why Space Y tasks us to train a machine learning model to predict successful Stage1 recovery.
Methodology / Approach
- Data collection methodology: Combined data from SpaceX public API and SpaceX Wikipedia page.
- Performed data wrangling: Classifying true landings as successful and unsuccessful otherwise.
- Performed exploratory data analysis (EDA) using visualization and SQL.
- Performed interactive visual analytics using FoliumandPlotly Dash.
- Performed predictive analysis using classification models: Tuned models using GridSearchCV
Technologies Used
- Python
- SQL
- Folium
- Plotly Dash
- GridSearchCV
- Logistic Regression, Support Vector Machine, Decision Tree Classifier, and K Nearest Neighbors.
