Building Information Models (BIM) to Structural Analysis Models (SAM) Using Artificial Intelligence
1. Build the SAMBuilder and TclBuilder excutables
2. Install Tensorflow if you haven't.
pip install tensorflow
Or do it in a virtual environment (virtualenv is recommended.) Tensorflow does not work with Python 3.7 or above at the time this code is made, so Python 3.6 is recommended.
Training the neural net
cd Python/NeuralNets python BIM2FeaturesNN_V1.py
Use the neural net in a BIM->SAM->tcl pipeline. This will generate a tcl file that can be run in OpenSees:
cd Python python main.py
How to cite
Charles Wang, Caigui Jiang, Stella X. Yu, Frank McKenna, & Kincho H. Law. (2019, October 18). NHERI-SimCenter/BIM2SAM.AI: Release v1.0 (Version 1.0). Zenodo. http://doi.org/10.5281/zenodo.3509957
This material is based upon work supported by the National Science Foundation under Grant No. 1612843.