🤖 Neural Network

This is a handmade neural network library for Python. It supports multiple layers, activation functions, optimizers... You can easily create a model and train it with backpropagation.

(I made this project to show that I can make a neural network whithout any library, if you need a neural network library, Keras or PyTorch will be much better than this)

📋 Summary

• 📋 Summary
• ✨ Features
• 🛠️ Install
• 🧪 Tests
• 🙏 Credits

✨ Features

Layers

• Input Layer: The input layer is the first layer of the network, it receives the input data

• Linear: A dense neural network layer, you can choose the number of neurons

• Activation functions: You can choose between ReLU, Tanh and Softmax

Optimizers

They are used to update the weights of the network. You can choose between SGD and Adam.

Loss functions

They are used to calculate the error of the network. You can choose between Negative log likelihood, Mean squared error and Mean absolute error.

🛠️ Install

You need Numpy to install the library:

$ pip install numpy

Then, just put the neural_network folder in the root of your project.

Example of imports:

from neural_network import Model
from neural_network.layers import Input, Linear, ReLU, Tanh, Softmax
from neural_network.optimizers import SGD, Adam
from neural_network.losses import NegativeLogLikelihood, MeanSquaredError, MeanAbsoluteError

🧪 Tests

You can find a jupyter notebook in the tests folder, it contains a simple example of how to use the library by creating a model that classify handwritten digits.

🙏 Credits

• Angel Uriot : Creator of the project.