How to use CNN, LSTM and GAN to conduct Sales prediction in Designer?

[wilburwen838]

I learn there is an in-depth learning neural network framework for sales prediction as the below diagram:

1. The CNN model is used to calculate the feature weight

2. Adopt LSTM RNN to carry on the automatic selection to the features, output the future N replenishment cycle sales prediction data;

3. Using the antagonistic generation network (GAN) to continue to improve the accuracy of the prediction.

I want to know how to find the CNN, LSTM and GAN tools in designers,  is there similar use case for sales prediction using this approach? any sample?

[danilang · Accepted solution]

hi @wilburwen838 

Out-of-the-box Alteryx(if you've got the Alteryx Predictive tools installed) comes with the Neural Network Tool which corresponds to the CNN in your model.  This article is a good introduction to the topic.  The "Tool Mastery | Neural Network" contains an example and a good explanation on how to interpret the results. 

As far as examples go, I haven't been able to find any.

Good Luck

Dan