Kaan Aytekin, from the TRUST AI framework engineering team, shows how to set up and run the TRUST AI framework on your local machine using Docker. Learn how to clone the repo, set up your environment, create an access token, and launch the app in minutes.
This video shows how to install a new predictive or prescriptive algorithm into the TRUST AI framework. Learn how to setup input/output files, configure parameters, build a Docker image, and run your algorithm in the TRUST interface.
Learn how to start using the Trust Framework: upload datasets, configure algorithms, and run your first training session. This video builds on the previous setup tutorial.
In this follow-up tutorial, Kaan Aytekin, from the TRUST AI framework engineering team, demonstrates how to use the explanation modules (What If and Counterfactuals) to better understand model behavior, test predictions, and analyze feature impact in the platform.
The goal of the project is to enable humans and machines to collaborate and discover new solutions. In order to achieve this, we have developed What-if Analysis and Counterfactual Analysis Interfaces for users. These interfaces provide users with powerful tools for exploring and understanding the models developed by the TRUST-AI project.
This video introduces the Explanability Assistant, a conversational AI tool that helps professionals interpret machine learning predictions by combining secure backend functions with large language models, offering transparent, interactive, and privacy-focused analysis.
This screencast illustrates the use of the TRUST AI framework for developing GP models and symbolic expressions for the energy use case.
This video illustrates the structure of a data-neutral and data-sharing environment for building applications. This work was delivered in D 1.2 of the TRUST AI project. The environment is hosted at https://ds.leiminte.com and will be used for open access to building data and for deploying them in production environments, particularly within the TRUST AI framework used for modeling purposes. More information can be found in the published paper: N. Sakkas, S. Yfanti, Open Data or Open Access? The Case of Building Data. https://www.academia.edu/55583548/Open_data_or_open_access_The_case_of_building_data, Academia Letters, October 2021.
The screencast shows how the TRUST AI API can be used to transfer data from the TRUST AI Energy data middleware within TRUST AI itself, under a specific project setup to receive data. The screencast relates to the Energy Use Case.
This screencast shows how the TRUST-AI API can be used to transfer data from an external (commercial) environment within TRUST-AI itself, under a specific project setup to receive data. The screencast relates to the Energy Use Case.