To boost interpretability with concept vectors, a reverse engineering approach automates concept identification by analyzing the latent space of deep neural networks using Singular Value Decomposition. This framework combines factorization, latent space c... read more of Unique Concept Vectors through Latent Space Decomposition

Jupyter notebook demonstrating usage of multivariate anomaly detection based on machine learning to process monitoring and detection of a suspicious state of a process. read more of DIH4AI: I-PRAG-4 Demo for experiment - Process monitoring using one-class support vector machine

This asset includes code forthe prediction of fuel consumption using AI regression models and time series forecasting techniques. The asset was developed in the context of H2020 VesselAI project read more of VesselAI Fuel Consumption Regression and Time Series Forecasting

Using a unique dataset, comprising historical maritime data for different vessel voyages, including vessel type, voyage information and environmental factors, we employ decision trees, a number of regression, and artificial neural network algorithms for p... read more of VesselAI CO2 Emissions Prediction using ML algorithms

The use of GPT3.5 to annotate frames (perspectives) in headlines on the anti-vaccine movement. read more of Frame analysis with GPT3.5

A novel progressive training algorithm for learning Graph Neural Networks with Differential Privacy guarantees read more of ProGAP: Progressive Graph Neural Networks with Differential Privacy Guarantees

Jupyter Notebook providing the execution of a simulation of a given Grid2Op agent using the PowSyBl backend. read more of AIRGo - I-NERGY - Simulation Jupyter Notebook

A toolkit for processing geospatial data, featuring an array of functions and workflows crucial for managing and analyzing information related to geospatial and terrain features. Developed by AMIGO s.r.l. for the ARIA project, part of the I-NERGY 2nd Ope... read more of Geospatial Data Downloading and Processing Toolkit

A collection of Python methods intended as a practical tool for fetching and preprocessing data related to climate and weather conditions, useful in climate science studies. Developed by AMIGO s.r.l. for the ARIA project, part of the I-NERGY 2nd Open Cal... read more of Toolkit for the retrieval and preprocessing of climate and meteorological data

AI Model based on LSTM forecasting the Utilization and Energy Consumption of Charging Stations. We forecast the future activity/consumption/utilization/Customer-experience on EV Charging stations using LSTM timeseries forecasting yielding highly accurate ... read more of CNROpt Charging Stations Forecasting