Gabriel's Projects
A really simple clock system using a 16x2 lcd and STM32F446RE with an alarm set
Some ipython notebooks implementing AI algorithms
Simple synopsis based recommender built with FastAPI. Embeddings are projected with Bigbird-ROBERTA, kNN and SVM supported for recommendations.
Some scripts written in linux x86 assembly
commands and files in my main machine
Official implementation of Character Region Awareness for Text Detection (CRAFT)
A clean Pytorch re-implementation of the paper Deep One-Class Classification extending it to 128x128 resolution.
A collection of deep learning implementations in Pytorch among different domains
A deep learning template project for re-using parts or the whole of it.
PyTorch Lightning implementation of Facebook DETR on Pascal VOC 2012 Dataset
A lightweight Neural Net to recognize text from documents
https://gabrieldornelles.github.io/
A post-modern modal text editor.
Dataset format for AI. Easily build and manage datasets for machine and deep learning. Stream data real-time & version-control it. https://activeloop.ai
2D BGR Convolution with Sobel operator on C for educational purposes.
end-to-end image captioning with vit + transformer
A PyTorch implementation of Y. LeCun, L. Bottou, Y. Bengio, and P. Haffner. Gradient-based learning applied to document recognition
Mathemagics implementations
Julia implementations learned through MIT 18.S191 Fall 2020
A Convolutional Neural Net for Steering control at Mario Kart 8
implementing ml architectures and training algorithms
My highly visual course to introduce the mathematics behind the modern computer vision. From linear classifiers to convnets.
A Pytorch implementation of the paper Image-to-Image Translation with Conditional Adversarial Networks.
Simple Pytorch framework to train OCRs. Supports CRNNs, Attention, CTC and Cross Entropy Loss.
A Transfer Learning model based on ResNet-18 to classify cats and dogs beetween 38 different breeds with 78.7% accuracy
An API for IP camera video streaming and object detection with DETR or Faster-RCNN
Basic server to N clients in C sockets.