FrontalLobe
A library to create and train Artificial Nueral Network classifier and analyse it's performance.
Currently, it only includes a class to implement Artificial Nueral Networks and a confusion matrix class for its analysis and validation named 'NueralNetwork' and 'confusionMatrix' respectively. It also includes three other suppport classes namely 'NueronLayer', 'rolledVector' and its subclass 'skeletonRolledVector' for smooth and fast execution of the 'NueralNetwork' class.
This library also contains a special feature of real time display of change in activation values of the component nuerons in the Nueral Network object while training, so that the user can understand the source of any numerical errors and also because it looks cool!
Although there are much more advanced and suitable frameworks developed by experts in field of machine learning and deep learning, I created this library to test and show my capabilities in python and my understanding of inner workings of ANN. In near future I aim to include more ML techniques and make this library much more advanced.
Requirements
python >= 3.6
Installation
pip install FrontalLobe
Version
0.0.1
URLs
The url address of detailed explaination and features of component classes of this library are given below: