Mwandamena/yango-hackathon

# 2024 Yango Hackthon In my third year second semester of my education at Cavendish University Zambia. I was privaleged to have participated in a machine learning competition hosted by Yango Zambia. Been an economic student, by then, we only learn't how to model using STATA and solve complex statistical problems on paper. I was the odd one out. ## Background I participated in the competition a day before the actual event. I only saw their advertisement a day before the event and decided to sell myself to the opportunity of revealing myself to the competition. ## The Challenge ### Can you predict taxi ride times in Lusaka using weather and trip data? Yango uses a wide range of data in order to predict travel and arrival times of its partners’ drivers and vehicles. In this Mobility Prediction Challenge, you are tasked to predict ride times based on the provided trip and weather data from Yango's data science team, exploring correlations between various factors such as distance, location, time of day, and weather conditions. We have prepared a sampled dataset of trips via the Yango platform from January 2024 in Lusaka. January was chosen as it is the rainiest month, allowing us to explore the relationship between travel time and precipitation. The models you build could be useful for improving ride time predictions and increasing efficiency of mobility in Lusaka. More about Yango ## My Approach Funny thing is I did not know that there was a starter notebook. And I thought it was somewhat unfair and a disadvantage to some people. But, I begain the competition by creating my own notebook with the workflow of machine learning that am comfortable with. 1. Loading data and packages: I start by loading the necessary packages for the solving the problem. This includes models, data wrangling tools, and visualisation tools. 2. Config: a include all the constants in a class for easy access. This can include things such as paths to the data, number of folds, seeds, …