daehancon / dsa2018

Data Science Africa 2018, hands on exercise.

Introduction

Stations located in Kenya.

The TAHMO stations

Dataset

The data is a time series of weather sensor readings, consisting of different physical variables on a regular grid on the Earth, indexed by lon(gitude) and lat(itude) coordinates. The variables we have made available are:

• prec --- Percipitation.
• tair --- air temperature.
• relh --- relative humidity.
• wspd --- wind speed.
• wgst --- wind gust.
• pres --- surface pressure.
• srad --- Solar radiation.

The fields are recorded every 5 minutes for two years from 2016-2017. The dataset is observation averaged on hour scale. For this exercise a percipitation data from multiple stations is selected as primary target. gpm-tahmo-rain.csv

• station --- percipitation.
• year--- year of observation.
• datetime --- timestamp of observation.
• tahmo --- percipitation from TAHMO station.
• gpm --- percipitation from GPM satellite close to the the TAHMO station.

Station description

• from: Station id
• to: Station id
• distance (km): distance between stations.
• elevation (m) : Elevation difference between from & to station.

The prediction task

- Predict whether it rains or not.