This repository contains the script which was used for the competition. Here is the link to the competition: https://www.kaggle.com/c/competitive-data-science-predict-future-sales/leaderboard.
a. Created a robust model to predict the total sales for the next month for every product and store from a time series dataset with 22000 products in 60 stores of a Russian Software Firm. b. Used LGBM with Bayesian Optimization based hyper parameter tuning to get an RMSE of 0.858 and rank 17 amongst 8015 teams (top 0.002%) on the leaderboard.