TensorTrace provides a drag-and-drop interface for building tensor networks:
Completed networks can be exported from TensorTrace into code via copy-paste into a text editor:
Flexible: code can be exported in the users choice of Python, Julia and MATLAB languages (where the Python code is also directly compatible with Google's TensorNetwork library).
Efficient: TensorTrace contains built in solvers that automatically determine the optimal contraction order of each network.
Networks diagrams can be reconstructed by copy-pasting code back into TensorTrace:
reconstruction of diagrams allows users to easily visualize and modify the networks associated to a computer code.
TensorTrace also provides a host of advanced features such as cost-scaling analysis and network auto-differentiaition:
single tensor derivatives