BUILD A PREDICTIVE MODEL IN 60 SECONDS

I have twenty-four hours to build a machine learning model to predict how NFL players are going to move while the ball is still in the air. The NFL is having another big data bowl and there are some big big prizes. So I'm partnering with AWS and in this video I'm going to show you how you can build a predictive model. First we have to gather and clean our data because as Sherlock Holmes said, it is a capital mistake to theorize before one has data. We have the location of each player before the ball is supposed to predict the location of every once the ball lands. That's a lot of people to predict. So instead of doing it all individually, we can look at it all as one big piece of information. A graph and then use a graph neural network for our model. Once we have our data we can separate it into training and test data. Train our GNN model and then test what we have. Then we can find tune accordingly or make an ensemble of models by adding in another model. Which I did by looping in an LSTM model. Now it's time to submit. We did it. Now I want to see you build a better model than mine. You can definitely do it. The link to join the competition is in my bio.