Kaggle COVID-19 Prediction

View Notebook

Topics Covered

• Week 1
  • XGBoost algorithm for regression
  • Feature Engineering - add new features using the existing data to boost the training data
  • Encoding Categorical Data
  • Handling time series data
  • Plotly maps
• Week 4
  • Time Series Analysis
  • ARIMA model for prediction
  • Linear regression with lasso and ridge

How to run

1. Build docker image

   docker build -t kagglecovid-19 .

2. Run docker container

   docker run -it -v $(pwd):/workdir -p 8888:8888 -t kagglecovid-19 bash

3. From inside the container workdir run to start jupyter notebooks

   jupyter-notebook --ip='0.0.0.0' --port=8888 --no-browser --allow-root

Resources

• Dataset