ANN Visualizer: A python library for visualizing Artificial Neural Networks (ANN)
The process of being able to Visualize data lies at the core of the skill set of any data scientist who is good at what he does. As doing so results in the provision of a holistic view of what all could be hidden in the data. And, today, with the help of advancement in the field of deep learning, an individual can now visualize the entire deep learning process or just even the Convolutional Neural Network (CNN or ConvNet) that has been built.
Also, adding on to it and taking it a step further, you can as well visualize an Artificial Neural Network(ANN) by just making the use of only a single line of code!
In order to help us do so, we have ANN Visualizer, a visualization library that is used in order to work with Keras. The visualization library makes the use of the ‘graphviz’ library of Python in order to create a graph that is neat and presentable of the neural network that is being built.
How is it useful?
This library can be useful in a lot of ways. For, let's say, the library can be made use for the purpose of teaching when an individual, without having to run a lot of code, want to explain how the NN looks like. There have been a number of previous efforts that have taken place in this area but with the ease of effort that it presents itself with and the beautifully optimized output, this is a library that is definitely worth.
The library, as of now, currently only visualizes dense layers but the developers have indicated that very soon there might be the addition of convolution as well as LSTM layers.
For all those who wish to go ahead and explore this library, do keep in mind that this is still an unstable release and there can be chances of bugs, so don't let them bother you. Presently, it has been tested with python3.5, but it should run just fine on any python3 as well!
So, Go ahead and start Visualizing your ANN with just a line of code, making the use of the ‘ANN Visualizer’ python library!
For more information regarding the same, go through the link mentioned below:
Source and Information: GitHub