Bisnow

5 Things To Keep In Mind To Avoid Software Buyer's Remorse

Software company Trimble highlights key points that an architecture, engineering or construction firm should consider to ...
AEC Magazine

Towards a hybrid reality

In the desktop vs. browser debate for AEC software, it’s no longer about choosing one over the other, writes Graphisoft CTO ...
IEEE

Efficient Surrogate Modeling for High-Dimensional Parametric Problems Based on Multi-Level Fidelity Simulation Combined with Incremental Learning

For high-dimensional parametric problems, data-driven surrogate models can be generated from simulation or experimental data to quickly evaluate quantities of interest, in support of applications such ...
IEEE

Proper Orthogonal Decomposition for Model Order Reduction of High Dimensional Parametric Problems with Varying Geometries Based on the Interface Adaptive Mesh

Abstract: The proper orthogonal decomposition (POD) is commonly applied for the model order reduction (MOR) of parametric problems. The method generates a reduced basis from the left singular vectors ...
GitHub

A Model Context Protocol (MCP) server implementation for Autodesk Revit 2026, enabling AI assistants like Claude to interact with Revit projects through a standardized interface.

This MCP server provides a bridge between AI assistants and Autodesk Revit 2026, allowing natural language queries and commands to interact with BIM models, extract data, modify parameters, and ...
GitHub

A PyTorch implementation of Parametric UMAP (Uniform Manifold Approximation and Projection) for learning low-dimensional parametric embeddings of high-dimensional data.

The pip install pulls the default PyTorch build from PyPI. If you need a specific CUDA version, install PyTorch first following the official instructions, then install this package.