I cover the full process of building a beginner machine learning project. This includes creating a hypothesis, setting up the model, and measuring error.
I use data from historical Olympic games and try to predict how many medals a country will win based on historical and current data.
Most machine learning projects follow a similar outline, which I also follow here.
Project Steps
- Form a hypothesis.
- Find and explore the data.
- (If necessary) Reshape the data to predict your target.
- Clean the data for ML.
- Pick an error metric.
- Split your data.
- Train a model.
I am using data from the Olympics, which was originally on Kaggle.
- teams.csv - the team-level data that we use in this project.
- athlete_events.csv - this is the original athlete-level data