Jupyter Notebooks

OMLT provides Jupyter notebooks to demonstrate its core capabilities. All notebooks can be found on the OMLT github page. The notebooks are summarized as follows:

• build_network.ipynb shows how to manually create a NetworkDefinition object. This notebook is helpful for understanding the details of the internal layer structure that OMLT uses to represent neural networks.

• import_network.ipynb shows how to import neural networks from Keras and PyTorch using ONNX interoperability. The notebook also shows how to import variable bounds from data.

• neural_network_formulations.ipynb showcases the different neural network formulations available in OMLT.

• mnist_example_dense.ipynb trains a fully dense neural network on MNIST and uses OMLT to find adversarial examples.

• mnist_example_convolutional.ipynb trains a convolutional neural network on MNIST and uses OMLT to find adversarial examples.

• auto-thermal-reformer.ipynb develops a neural network surrogate (using sigmoid activations) with data from a process model built using IDAES-PSE.

• auto-thermal-reformer-relu.ipynb develops a neural network surrogate (using ReLU activations) with data from a process model built using IDAES-PSE.

• bo_with_trees.ipynb incorporates gradient-boosted-trees into a Bayesian optimization loop to optimize the Rosenbrock function.

• linear_tree_formulations.ipynb showcases the different linear model decision tree formulations available in OMLT.