Stars
This is the repository of the journal paper: Neural networks for fatigue crack propagation predictions in real-time under uncertainty - Giannella et al.
Convert any music library into a music production sample-library with ML
Tensorflow implementation of Levenberg-Marquardt training algorithm
Cross X-AI: Explainable semantic segmentation of laparoscopic images in relation with depth estimations.
This is the repository of the journal paper: StaSiS-Net: a stacked and siamese stereo network for depth reconstruction in modern 3D laparoscopy. Bardozzo et al.
Slight modification to one of the Facebook Salina examples, to test the A2C algorithm on financial series.
a Lightweight library for sequential learning agents, including reinforcement learning
This is the repository of the journal paper: Sugeno integral generalization applied to improve adaptive image binarization. Bardozzo et al.
The script below shows how to modify, with a customized layer, the value of the tensors of a network being trained. This behaviour is modelled considering static and dynamic operation assignments. …
This repository contains the supplementary materials of the manuscript mentioned in the title, and associated source code, the whole data set of signals and some functions to obtain fast results wi…
Specularity detection and removal for endoscopic images/videos.
Metabolic Networks Based Adjacency for Multi Omic Patterns Analyses
Some documented examples of bio semantic-web technologies in R, XSLT and SPRQL
Java Lexer, Parser, Seman and CGEN for the COOL language.
Didattic Material for the LSTM workshop done the 23-25 May 2018 @ unisa.
Some relevant examples of bio medical semantic-web leveraging R, SPARQL querying PubMed and MeSH
A study on multi-omic oscillations in Escherichia coli metabolic networks. Journal paper.
This is a simple example where the 3D reconstruction through ZED sdk is combined with the aruco markers detection.
Here we introduce the problem of 3D solids classification with a CNN (spheres and octahedra). We implemented a 3D GradCam model, in order to underline the most relevant 3D volumes useful for classi…