fmqbl / dengai-predicting-disease-spread

Practice solution of Driven data competition 'DengAI Predicting disease spread'. (Preprocessing, Visualization & Hypothesis Testing)

DengAI: Predicting Disease Spread (Practice Exploration with Hypothesis Testing) Competition by Driven Data

Dengue fever is a mosquito-borne disease that occurs in tropical and sub-tropical parts of the world. In mild cases, symptoms are similar to the flu: fever, rash, and muscle and joint pain. In severe cases, dengue fever can cause severe bleeding, low blood pressure, and even death.

Problem statement and description of data is given in link

Goal is to predict the total_cases label for each (city, year, weekofyear) in the test set. There are two cities, San Juan and Iquitos, with test data for each city spanning 5 and 3 years respectively. You will make one submission that contains predictions for both cities. The data for each city have been concatenated along with a city column indicating the source: sj for San Juan and iq for Iquitos. The test set is a pure future hold-out, meaning the test data are sequential and non-overlapping with any of the training data. Throughout, missing values have been filled as NaNs.