These Jupyter Notebooks are made as I am learning Neural Networks. The order in which the notebooks should be read is given below. Anyone is welcome to make suggestions and edits to notebooks. I believe that they can be helpful to anyone who is getting started in this field.
Each notebook consists of implementation of some learning algorithm for neural networks or representation tools like matplotlib, performance surfaces, etc.
I'm putting these notebooks online so anyone can look them up as reference to Neural Network implementations. Again, just to mention, these cover material from very basic level to some advanced algorithms. (Please refer to the list given below.)
You can view these notebooks in nbviewer here: http://nbviewer.jupyter.org/github/KrnTneja/NeuralNetworkNotebooks/tree/master/
All the imports used in a notebook are given at the top of the notebook. Any errors coming up on your machine can generally be due missing libraries. Please look up at top of notebook in that case and install required libraries/modules.
For all the functions and classes defined in notebook, they have a test example that I've kept as a standard example to test whenever I make a change to function or class. You can look them up as examples or tests for checking the working on you machine.
Most of the concepts have been explained in the notebook before they are implemented. Latex has been exploited for the best experience for the reader.
Comments after 'def' line of a function explains what the function is made to do. It also explains the format of input function expects and the output one must expect from function.
Anyone can use these notebooks for educational purposes (but not for any commercial purposes or similar reasons whatsoever).