• Python
  • Machine Learning
  • Time Series Data
  • Sales Forecasting
  • Retail Analytics
  • Gradient Boosting


I worked on this project as part of my summer internship at Invent Analytics.

I worked on implementing a machine learning model for sales forecasting. I used the dataset provided by the M5 Forecasting – Accuracy Competition hosted on Kaggle. The dataset contains sales information collected from Walmart over several years. The data was hierarchical, representing different levels, i.e. state level, store level, …etc. The goal of the competition was to predict the unit sales of different Walmart products for the next 28 days. My role was to study the top solutions in this competitions and come up with a model that combines the best ideas from these competitions.