Neural Networks

Introduction

Inspired by the human brain, neural networks represent a computational model composed of interconnected nodes organized into three layers: input, hidden, and output. In this project, I have implemented three models for digit classification, regression, and language identification.

Final result

Non-linear regression for $\sin$ function prediction

The following shows the implemented non-linear regression predicting the value of $\mathrm{sin(x)}\mathrm{over\; [-2\pi ,\; 2\pi ]}$ interval.

Digit classification

The following demonstrates the usage of the MNIST dataset, which contains 28 $\times$ 28 grayscale images, to train the implemented network for the classification of handwritten digits.

Language identification

Language identification is the task of determining the language in which a piece of text is written. For instance, your browser might be capable of detecting if you’ve accessed a webpage in a foreign language and offer to translate it. Using a dataset that includes five different languages, the following demonstrates the training of the implemented neural network for language identification, analyzing one word at a time.

_{Reference to the stub code}