# User Manual¶

This chapter starts with a description of the blackbox solver, provided by the solver module. If solutions of a start system for a polynomial homotopy are available, then we may better call directly the path tracking routines to solve the target system in the homotopy. The path tracking functions are described in the second section.

The third section deals with the computation of positive dimensional solution sets. In a numerical irreducible decomposition of the solution set of a polynomial system, generic points are computed on all irreducible factors of the solution sets of all dimensions. In a witness set representation, the number of generic points in the witness set equals the degree of the pure dimensional solution set. Solution sets can be computed in a top down fashion with cascade homotopies or in a bottom up manner via diagonal homotopies.

The definitions of the polynomial systems which make interesting examples and families of problems are illustrated in section four.

Pieri homotopies and Littlewood-Richardson homotopies solve problems in enumerative geometry. The fifth section of this chapter is concerned with numerical Schubert calculus.

Every polynomial in several variables has a Newton polytope, spanned by the exponents of the monomials which occur with a nonzero coefficient. The mixed volume of a tuple of Newton polytopes gives an upper bound for the number of isolated solutions. Systems with exactly two monomials in every equation can be solved fast, via unimodular coordinate transformations. Section six of this chapter ends with an illustration of the computation of power series solutions for algebraic curves.

Section seven and eight describe prototype modules for a graphical user interface and a computational server. The nineth and last section of this chapter sketches the design of the C interface, the Python interface module, and the wrappers to the C interface to PHCpack.

The last section collects the code snippets, defined for the notebook extension of Jupyter, as they pop up in the menus of the notebook.

- a blackbox solver for isolated solutions
- solving random trinomials and a particular trinomial system
- representations of solutions of polynomial systems
- reproducible runs with fixed seeds
- root counting methods
- Newton’s method and deflation
- the multiplicity of an isolated solution
- equation and variable scaling
- reduction of polynomial systems

- path trackers and sweep homotopies
- a simple example
- fixing the gamma constant
- give the next solution on a path
- solving with polyhedral homotopies
- Newton’s method at higher precision
- multitasked path tracking
- GPU accelerated path tracking
- sweep homotopies
- real versus complex sweeps
- tuning parameters, settings, and tolerances
- a polyhedral end game

- positive dimensional solution sets
- some interesting examples and families
- numerical Schubert calculus
- Newton polytopes, monomial maps, and power series
- a graphical user interface
- the module server
- the module phcpy.phcpy2c3
- code snippets