Aims and Objectives of this project: Finding the aggregate transactions each month during years 2015- 2019 and analyse the trends. Finding the top 10 Smart Contracts (Addresses that made the largest transactions) that took place during these years. Analysing if gas prices has changed over time, or contracts have become more complicated (the amount of gas consumed per transaction has increased or not). Finding the most lucrative form of scam taking place in Ethereum community?

In this project, we cover some of the time series methods used in past competitions [7][8][9] to investigate the dataset provided by Walmart. First, we build a special kind of RNN model called LSTM which attempts to predict the unit sales of individual stores. This is done by considering a univariate model and then using a more complex multivariate model, which we hope increases the test accuracy. Next, from the scikit-learn library in Python, we build an RF model which analyzes the item HOUSEHOLD_1_272_CA_3_validation, whilst selecting the most important features which help predict the sales of this particular product.