Antonio Rocha’s io#

This is my github.io’s main page, here you can find a little more about me and links to my project’s documentations. For my GitHub page, click here.

About me#

I’m a Brazilian/Portuguese Chemical Engineer from the Federal University of Rio de Janeiro (UFRJ), having done a double degree exchange at ESPCI - Paris. I’ve become interested in programming with Python and Deep Learning for engineering applications by the beginning of 2021. Since then, I’ve worked on the subject on a few different projects:

  1. A project for the Introduction to Deep Learining class at ESPCI, developing a CNN for microfluidic drop recognition and classification. Code not available.

  2. An internship on Battery Health estimation, using concepts of Transfer Learning for applying models trained for one type of battery to batteries of different chemistries and protocols. We’ve recently published an open-access article on this work and it can be found here. The complete code is also available here.

    *Graphs from the publication.*

    Some graphs from the publication.#

  3. My undergraduate thesis on Predictive Emissions Monitoring System (PEMS) at the Federal University of Rio de Janeiro (UFRJ). I developed and compared prediction accuracies for different types of neural networks (NNs, CNNs, RNNs) and hybrid models that integrate them. The final report is publicly available (though it is written in portuguese) and can be found here Some of the results were presented in the ESCAPE32 2022 conference, the proceedings of which can be found here. Code not available.

    *Schematic of the hybrid model.*

    Schematic of the hybrid model.#

  4. I did a personal project on Neural Networks for recognition of LaTeX symbols (CNN Project). Its objective was to develop skills in advanced python, get used to using GitHub, explore CNN architectures and serve as portfolio (since other projects may be private). Code available below.

*The state of the Pygame UI implementation.*

The state of the Pygame UI implementation.#

You can find more about me in my Linkedin page.

My Public Projects#

CNN Project#

This is a personal project for developing skills in python and testing different Convolutional Neural Network (CNN) architectures for different tasks. It’s also supposed to serve as a bit like a portfolio, although it is still a work in progress.

ProcessEngineering#

Project for testing the implementation of different Process Engineering concepts. This includes Equation Ordering Algorithms for Sequential Modeling and simulation. It also serves as a dive into object oriented programming and dynamic definitions of functions.

Basic Optimization Framework#

Small project for relearning FORTRAN 90 and studying optimization concepts. This project is a generic framework over which simple optimization methods can be built upon and run.

C++ Game#

Small project following the footsteps of Let’s make games’ “How to make a game” tutorial as a means of learning a little bit of C++.