J8ahmed notes on recurrent neural networks 1

Day 90: Recurrent Neural Network Basics

I’m finally back to what I want to focus on. How to build a great “Rock Paper Scissors” Bot for my freeCodeCamp project. This has been the longest I have ever spent on a freeCodeCamp certification project. I have failed many times but I think I am finally on the right track. I guess I am failing forward and making progress.

Now, about the Recurrent Neural Networks (RNNs). I am only vaguely familiar with RNNs. I know that they are a better fit for sequential data. I was previously trying to make a Neural Network fit for complex classification predict the next move of an opponent in ‘Rock Paper Scissors’. That didn’t go so well. lol

From what I understood from my googling, is that, an RNN is always carrying along previous information from each past step along with new information introduced at that moment.

I still don’t really know how it works under the hood but we will slowly peel this RNN onion back layer by layer (Ha, get it – Neural Networks have layers! XD). On that horrible joke I will see myself out. Until next time.