maiya11 / nnets

Neural network tutorials and implementations.

This repository will help you master neural networks. It contains writeups of how different neural networks work, along with full implementations of different network types.

• The explanations folder has writeups of each algorithm.
• The nnets folder has clean implementations that are best for someone who understands the high level concept.
• The exploration folder has implementations that are more exploratory and easier to understand for beginners.

I recommend reading the writeup, then looking at the exploratory implementation, then looking at the clean implementation.

1. Gradient Descent

Gradient descent is an important building block for neural networks.

• Gradient descent with linear regression tutorial
• Clean implementation - linear regression is equivalent to a dense network with no activation function.
• Notebook implementation

2. Dense networks

Dense networks are networks where every input is connected to an output.

• Dense network tutorial coming soon
• Clean implementation
• Notebook implementation

3. Convolutional networks

Convolutional neural networks are used for working with images and time series.

• Convolutional network tutorial coming soon
• Clean implementation
• Notebook implementation

Installation

• pip install -r requirements.txt