articles | Antonin L.

2024

Hybrid Quantum-Classical Framework for Clustering: a Comprehensive Approach with QMeans

Antonin L.

Mar 2024

Abstract Preprint PDF

This project aims to introduce the concept of quantum computing and its applications in machine learning. We will implement a quantum version of the K-Means clustering algorithm. The goal is to compare the performance of the quantum algorithm with the classical version.

2023

Design of a Connected and Intelligent Weather Station with ESP32 and MQTT

Antonin L.

Jun 2023

Abstract Medium PDF

This project aims to develop a connected and intelligent weather station. It will be equipped with various sensors: the BMP180 for measuring temperature and atmospheric pressure, the HTU21 for humidity, and a Grove sensor for assessing air quality. The data collected by these sensors will be transmitted via an ESP32 to a local MQTT server on a Mac. This server, in addition to displaying the received information using a Python script, will be capable of storing this data for visualization through a Streamlit dashboard. The weather station will provide real-time information on environmental conditions, contributing to a better understanding and prediction of local climate.
Monitoring Structural Deformation with Accelerometer and Gyroscope

Antonin L.

Jun 2023

Abstract Medium PDF

This project aims to develop a real-time structural deformation monitoring system using a combination of acceleration sensors and gyroscopes. Three MPU6050 sensors are used to detect variations in acceleration and rotation (gyroscope) at three specific points on the structure. These sensors are connected to ESP32 modules that transmit the collected data to a local MQTT server. A Python script retrieves this data, processes it to estimate the deformation of the structure, and stores it in a MongoDB database for later analysis. The deformations of the structure are visualized in real-time using a 3D Bezier curve, allowing for the observation of the structure’s tilt and movement. This system offers a real-time structural monitoring solution.
Denoising Autoencoder for Vibratory Anomaly Detection

Antonin L.

May 2023

Abstract Medium PDF

This project aims to detect shocks in real-time using a denoising autoencoder model and an acceleration sensor. The goal is to identify shocks among the sensor data by eliminating noise and normal variations in acceleration. We will use the MPU-6050 (GY-521) sensor and an Arduino Uno to collect acceleration data. The interface for visualizing the data in real-time is developed with Streamlit.
Variational Autoencoder for the Generation of Electrocardiograms

Antonin L.

Apr 2023

Abstract Medium PDF

This article presents an innovative Streamlit application capable of generating synthetic electrocardiograms (ECGs) using a variational autoencoder. The tool offers a new way to augment ECG datasets for medical research and diagnostic algorithm development. By combining the advantages of Streamlit’s intuitive user interface with the power of generative models, this application serves as a bridge between the fields of medicine and machine learning.

2022

Dynamic Graph Neural Network for signal clustering

Antonin L.

Jun 2022

Abstract PDF

This project aims to develop a dynamic graph neural network for signal clustering. The goal is to cluster signals such as birds song, ECG and random signal. We will use the Dynamic Graph Neural Network (DGNN) model proposed by the authors of the paper “A Dynamic Neural Network for Continual”.

2021

Epidemic Simulation with Python

Antonin L.

Sep 2021

Abstract PDF

This project aims to simulate the spread of a virus within a population by using the SIR and SIDR models. The objective is to study how various parameters impact the rate and extent of the virus’s spread.