This project was bootstrapped with Poetry.
This is project illustrates the basic single neuron neural network demonstrated by giant_neural_network in his Beginner Introduction to Neural Networks YouTube series. giant_neural_network demonstrates the innerworkings and details of a neural network. As such, he describes the math used for the activation function and the loss (cost) function. The merit is a deeper understanding of what neural networks are and what they're doing at each step as it learns to classify new feature sets. However, giant_neural_network also notes that there are machine learning libraries that handle the math once we understand what a neural network is, what its doing, and how we can effectively use it.
As such, this project demonstrates his example using the PyTorch library.
To get started, ensure Python 3.7 and Poetry are installed. After:
This project does not have a production build. Obtain the source by cloning or downloading the repository. If you have git installed, run the following command:
git clone https://github.com/weiluntong/IntroNN.git
Otherwise, there is a green button at the top right that says "Clone or download" to obtain the source code.
If you cloned the source code, simply cd into the directory you just cloned. If you downloaded the zip, you'll have to uncompress it first and then navigate to the directory in a terminal.
Run this command:
poetry install
and that's it. You're ready to run any of the available scripts.
In the project directory, you can run:
Runs the app. You can modify the python files within the intronn directory to see different outputs at different stages.
You can learn more in the Poetry documentation.
Python documentation can be found here, and PyTorch documentation is here.