We are provided with daily historical sales data by one of the largest Russian software firms - 1C Company. The task is to forecast the total amount of products sold in every shop for the test set. Note that the list of shops and products slightly changes every month. This is part of a Kaggle competition
After extensive EDA and feature engineering built CATBOOST,LGBM and XGBOOST models and averaged the predictions from these models as the final prediction.
Please view the jupyter notebook at : https://nbviewer.ipython.org/github/Bhujith10/Predict-Future-Sales/blob/main/predict-future-sales-vbm-updated.ipynb
You can download the data at : https://www.kaggle.com/competitions/competitive-data-science-predict-future-sales/data