https://github.com/nnamanx/olympic-games

I use data from historical Olympic games and try to predict how many medals a country will win based on historical and current data
https://github.com/nnamanx/olympic-games

linear-regression machine-learning mean-absolute-error

Last synced: 16 days ago
JSON representation

I use data from historical Olympic games and try to predict how many medals a country will win based on historical and current data

• Host: GitHub
• URL: https://github.com/nnamanx/olympic-games
• Owner: nnamanx
• Created: 2024-03-25T17:10:46.000Z (10 months ago)
• Default Branch: main
• Last Pushed: 2024-03-26T05:51:35.000Z (10 months ago)
• Last Synced: 2024-11-06T01:50:06.440Z (2 months ago)
• Topics: linear-regression, machine-learning, mean-absolute-error
• Language: Jupyter Notebook
• Homepage:
• Size: 199 KB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# Project

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.

# Machine learning project steps

Most machine learning projects follow a similar outline, which I also follow here. 

**Project Steps**

1. Form a hypothesis.

2. Find and explore the data.

3. (If necessary) Reshape the data to predict your target.

4. Clean the data for ML.

5. Pick an error metric.

6. Split your data.

7. Train a model.

## Data

I am using data from the Olympics, which was originally on [Kaggle](https://www.kaggle.com/datasets/heesoo37/120-years-of-olympic-history-athletes-and-results).

* [teams.csv](https://drive.google.com/uc?export=download&id=1L3YAlts8tijccIndVPB-mOsRpEpVawk7) - the team-level data that we use in this project.

* [athlete_events.csv](https://drive.google.com/uc?export=download&id=1Ah4wOyNFMGREq8Yw_Jbv7u2CeI_6tpn5) - this is the original athlete-level data