Do you know which inputs your neural network likes most? :: Päpper's Coding Blog -- Have fun coding.

Recent advances in training deep neural networks have led to a whole bunch of impressive machine learning models which are able to tackle a very diverse range of tasks. When you are developing such a model, one of the notable downsides is that it is considered a "black-box" approach in the sense that your model learns from data you feed it, but you don't really know what is going on inside the model. To make it clearer: you don't really know what your model actually learned and if you have a flaw in your training / data approach it might work well according to your metrics while having learnt the wrong thing. As a self-respecting developer you want to do better than that, so today I will show you a method you can use to get some better introspection into your model by using visualization techniques. So what is a visualization techniqe when we talk about deep neural networks?