A talk that explains how neural nets work, mathematically, alongside a walkthrough of the associated code that builds them from scratch using Python. This talk was delivered at ODSC West 2017 as "Deep Learning From Scratch Using Python".
The "ODSC_Deep_Learning_2017.ipynb" notebook contains the talk. It is designed to be viewed using RISE. The instructions for viewing the talk, therefore, are:
- Clone this repo:
git clone git@github.com:SethHWeidman/ODSC_Neural_Nets_11-04-17.git - Follow the instructions in the RISE repository to download RISE and configure it correctly as a Jupyter extension.
- Open a Jupyter Notebook by typing
jupyter notebookin the terminal. - Begin slideshow mode, being sure to run all the code cells as you go along.
You may first need to install Jupyter; if so, see here for instructions.
Enjoy!