In the 7th tutorial for artificial neural networks, we start looking into sequential data and time series. We're going to build a recurrent neural network for this purpose.

Sequential data can be viewed or perceived as data at multiple points in time. Think of it like having 2 axes, the x and y. On the y axis you have your data points, while the x axis is for the time points. 

In the real world, applications of sequential data are traffic analysis, weather forecast, the stock market, and so on. 

Like we did with all machine learning models in this series, we're going to start by creating our dataset. It's going to be comprised of waveforms (oscillations across time). 

For that we're defining two functions: one to generate the waves and amplitudes, and another one for visualization purposes. We are going to use these functions later for creating the dataset that's going to be fed to our recurrent neural network. 

Please see below for the full walkthrough for this tutorial:

<center><iframe width="560" height="315" src="https://www.youtube.com/embed/uwL4F-3xlqs" frameborder="0" gesture="media" allow="encrypted-media" allowfullscreen></iframe></center>
___
### <center>To stay in touch with me, follow @cristi</center>   
___

[Cristi Vlad](http://cristivlad.com) Self-Experimenter and Author