Guilherme D. F. Silva's Projects
Adaline regressor learning the Celsius to Fahrenheit formula.
Adaline is a single-layer neural network, a variation of the original Perceptron architecture.
Convert any image to ASCII art; Print to console; Save to .txt.
Web app for mining and visually analyzing association rules.
Convolutional autoencoder removing noise from Fashion MNIST clothing images.
Convolutional autoencoder reducing traffic sign images to 1/6 of their original size.
Text anomaly detection with a combo of Bloom Filter and MinHash.
Java implementation of the dice game Bozó. A brazilian version of the game called Yahtzee.
Web app to estimate car quality based on the car properties informed by the user.
Simple TensorFlow CNN classifier with 4 convolutional and 3 dense layers for the CIFAR-10 dataset.
Simple CNN architecture classifying Fashion MNIST.
A quick analysis of simple CNN architectures classifying Handwritten MNIST dataset.
Competitive programming solutions (UVa, LeetCode, codeforces, InterviewBit, etc.).
Demonstration of how to use correlation matrices to pick the best variables for polynomial regression.
PySpark with logistic regression predicting if customers will exit a bank service.
Web app for checking if a series of values is normally distributed.
TensorFlow deep regression model predicting bicycle rental.
Fully connected neural network diagnosing patients with diabetes.
Haar cascade classifiers detecting faces and eyes in images with OpenCV.
A game based on Flappy Bird
Web app using univariate regression to predict the cost of opening a new franchise store.
Web app using fuzzy logic to evaluate how much compatibility a person has with the computer science career.
A simple Android GridView with cards for a store.
Implementação em Java de um grafo e do algoritmo de Dijkstra.
Here I am.
Haar cascade classifiers detecting cars and persons in videos with OpenCV.
Python 3 solutions to HackerRank's RegEx challenges.
Java implementation of Leland Wilkinson's outlier detection algorithm
A data science analysis using kNN and MLP algorithms on HOG image descriptors.