An example solution to the no_flood_with_ai_2020 contest

What information is contained in this repository?

• Jupyter notebook that contains the steps of implementing a predictive model and has any descriptive comments that would help explain the working of the submitted solution
• Dockerfile and forecast.py those are needed to run your machine learning code
• Abstracts describing the approach. Any formats are accepted.

The testing process

docker build -t no_flood_with_ai .
docker run --volume $(pwd)/datasets:/usr/src/app/datasets no_flood_with_ai 2020-10-11 2020-10-21

Note, you don't need to include datasets in a docker image.

As you can see, docker container runs with command line parameters that specifies the start and end of the period of data forecasts.