Skip to main content

SINNER - Simplest Implementation of Neural Networks for Effortless Runs

Project description

SINNER - Simplest Implementation of Neural Networks for Effortless Runs

I worked with Neural Networks more years ago than I'd like to remember. I'm returning to study "black box models", and I felt that there is no simple way of creating a neural network, and that was a mistake. That's why I've created this Python library.

Creating a new neural network is (and always will be) as simple as NeuralNetwork(list), where list is a list of integers with the number of neurons on every layers (the first one being the input, the last one the output, and the rest the hidden ones). Of course there are and will be optional parameters, but it will always work with a standard view for starters.

Of course "simplest" is not the same as "simplistic", and every aspect of a neural network that makes this implementation more robust is welcome.


Public methods

This list needs to be as short as possible, always. Nowadays we have two methods only:

eval(inputs): eval an array of inputs with current configuration of the neural network.

train(trainingInputs, trainingOutputs): train the network from a set of inputs and outputs


To Do List

  • import/export the neural network
  • add a log system for training outputs
  • create a packaging for PIP
  • add usage examples on Git
  • make transfer functions selectable on creating

The original version of this implementation was loosely based on Jason Brownlee's "How to Code a Neural Network with Backpropagation In Python (from scratch)"


Comments and suggestions, feel free to contact me!

--Friar Hob

Project details


Release history Release notifications

This version

0.0.1

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Files for sinner-friarhob, version 0.0.1
Filename, size File type Python version Upload date Hashes
Filename, size sinner_friarhob-0.0.1-py3-none-any.whl (4.4 kB) File type Wheel Python version py3 Upload date Hashes View
Filename, size sinner-friarhob-0.0.1.tar.gz (3.2 kB) File type Source Python version None Upload date Hashes View

Supported by

Pingdom Pingdom Monitoring Google Google Object Storage and Download Analytics Sentry Sentry Error logging AWS AWS Cloud computing DataDog DataDog Monitoring Fastly Fastly CDN DigiCert DigiCert EV certificate StatusPage StatusPage Status page