Cut the current onnx file at the given node to create two networks. The output of the first network is the output of the given node.

Create output file and write header

Start scanning a new suite

Update the score of the constraint with the given score in the costTracker. */

Print usage info and exit

Construct program usage string

Write network data into an .nnet file Args: file_name (str): Path to the file to which the network will be written

Write the world definitions

Function to generate equations corresponding to all types of add/subtraction operations Args: op: (tf.op) representing addition o

Function to generate equations corresponding to addition Args: node (node): ONNX node representing the Add operation

Function to add a new Leaky Relu constraint Args: v1 (int): Variable representing input of Leaky Relu v2 (int): Varia

Append the line to the current CXXTEST_CODE() block

Identify all relus and their associated variable numbers and add them to Marabou network :meta private:

Add a test function to the current suite

Function to generate equations for a BatchNormalization Args: node (node): ONNX node representing the BatchNormalization operatio

Function that searches the graph recursively and builds equations necessary to calculate op Args: op: (tf.op) representing operat

Construct the Marabou equations :meta private:

Function representing cast Args: node (node): ONNX node representing cast operation :meta private:

Return true if stack is currently compliant, false otherwise If there is no stored solution, return false --- incompliant.

Remove unused shapes After constructing equations, remove shapes from self.shapeMap that are part of the graph but not relevant for t