This project was my final Bachelor's degree thesis. In it I decided to mix my passion, music, and the syllabus that I liked the most in my degree, deep learning.

Classification of different landcover classes using Hyperspectral data.

Different kinds of deep neural networks (DNNs) implemented from scratch using Python and NumPy, with a TensorFlow-like object-oriented API.

Digit recognition (MNIST dataset) using a fully connected neural network (97+ on test)

Train a Tic-Tac-Toe and a Connect-4 player using the TD-lambda algorithm

This is an implementation of a fully connected feedforward Neural Network (multi-layer perceptron) from scratch to classify MNIST hand-written digits

This repository holds one of my first Deep Learning projects. The project implements an MNIST classifying fully-connected neural network from scratch (in python) using only NumPy for numeric computations. For further information, please see README.

A fully connected neural net implemented by Numpy

Deep Learning in Computer Vision algorithms. Implementations of relevant codes, some from scratch others re-used and modified.

Code for Point Cloud Generation

keras implemention of fcn

CIFAR-10-Image-Classifcation done with Pytorch

A project for my Advanced Artificial Intelligence class to apply AI methods to a real-world problem.

LeNet5 from Scratch

Python illustration of Neural net from scratch

Yet another basic neural network implementation heavily inspired by, and based on, micrograd

Python implementation of couple machine learning algorithms.

Minimal, limited in features, deep learning library, created with the goal of understanding more of the field.

A CNN is a kind of network architecture for deep learning algorithms and is specifically used for image recognition and tasks that involve the processing of pixel data. This repository contains the code for CNN with a categorical classification dataset.

An implementation for an FCNN from scratch, for educational purposes