https://github.com/rizerkrof/library-simpleneuralnetwork

This is a simple fully connected neural network library for deep learning. Include example.
https://github.com/rizerkrof/library-simpleneuralnetwork

deep-learning deep-neural-networks library

Last synced: 6 days ago
JSON representation

This is a simple fully connected neural network library for deep learning. Include example.

• Host: GitHub
• URL: https://github.com/rizerkrof/library-simpleneuralnetwork
• Owner: rizerkrof
• Created: 2022-10-24T13:10:53.000Z (over 2 years ago)
• Default Branch: main
• Last Pushed: 2022-10-24T13:14:10.000Z (over 2 years ago)
• Last Synced: 2024-11-29T09:52:41.829Z (2 months ago)
• Topics: deep-learning, deep-neural-networks, library
• Language: Python
• Homepage:
• Size: 5.86 KB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.org

Awesome Lists containing this project

README

        
#+title: Readme

#+author: Edouard Lacourt

* Table of Contents :TOC_3:noexport:

- [[#description][Description]]

- [[#prerequisites][Prerequisites]]

- [[#documentation][Documentation]]

  - [[#installation][Installation]]

  - [[#create-a-neural-network][Create a neural network]]

  - [[#fit-the-model][Fit the model]]

  - [[#evaluate-the-model][Evaluate the model]]

  - [[#run-example-script][Run example script]]

  - [[#run-test][Run test]]

* Description

/!\ This repo needs to be refactored

This library allows to create and fit a custom neural network. This library was made during data predictive module course at EPF engineering school. It is a very simple library, you can only create fully connected of neural network. However, it is possible to adjust layer count and size to your needs.

* Prerequisites

- python 3.8

* Documentation

** Installation

Install dependencies:

#+begin_src sh

pip install -r requirements.txt

#+end_src

** Create a neural network

#+begin_src python

dnn = DeepNeuralNetwork(n_neurons=[784, 128, 64, 10])

#+end_src

You can define the layer count and size in n_neurons. n_neurons list correspond to the network layer list. Each element of the list correspond to the side of the layer. Please note that the first element represent the input layer and the last to the ouput layer. You network needs to have at least 2 layers.

** Fit the model

#+begin_src python

dnn.fit(X_train, Y_train, X_val, Y_val, epochs=5, batch_size=16, l_rate=1e-2)

#+end_src

The above expression fit the model called 'dnn'

** Evaluate the model

#+begin_src python

dnn.predict(X_test)

#+end_src

The above expression evaluate the model called 'dnn' with the 'X_test' data.

** Run example script

#+begin_src sh

python3 ./example/handWrittenNumbersExample.py

#+end_src

** Run test

#+begin_src sh

pytest ./tests/*

#+end_src