PyOptimizer

Python implementation of general nonlinear optimizer, this library is simple and just reveals the essential of the non-linear optimizer.

• Project backbone are implemented in backend
• Example use of this repository is implemented in tests/testMonoBA.py run

python tests/testMonoBA.py

• Generally for a nonlinear problem, we can define the problem with parameters (unknown variables) and constraints (equations).
• Parameter are abstracted as Vertex, constraints are abstracted as edge.
• In a problem, vertices must form edges and these edges must generate a residual. and this code will adjust the values of vertices to get minimum residual.
  • A basic optimization problem of finding a,b,c to get minimum y in y = a * x * x * x + b * x * x + c * x + noise given a set of y and x is implemented in tests.py. Feel free to adjust codes there to solve other simple problem.

• BAL dataset test code might be implemented later but it is simple and nearly implemented in tests/testBA.py.