Hackathon participation files.
Aim of the hackathon was to forcast and optimize warehouse Pick and Pay, i.e. using current information of fictious customers, predict how a new customer would require items from their warehouse.
Method used: Data from 4 customers was used to pre-train an LSTM model and then tested on the last customer. Data aggrigated as per both, weeks and monthly basis. Monthly data was presented and a third place was achieved.