Welcome to madupite’s documentation!

madupite is a python package for solving large-scale Markov decision processes (MDP) using distributed inexact policy iteration (iPI). The library is written in C++ and uses distributed sparse linear solvers from PETSc.

• Installation
  • Dependencies
  • Python package
  • C++
  • Euler (ETH Zurich Cluster)
• Tutorials
  • Loading and reading data with madupite
  • Generating data with madupite
  • Matrix Preallocation
  • Data format
• Examples
  • Example 1
  • Turning it into code
  • Further examples
• Options
  • Required Options
  • Optional Options
  • Useful PETSc Options
  • Usage Example
• Benchmarks
• API Reference
  • madupite.createTransitionProbabilityTensor
  • madupite.createStageCostMatrix
  • madupite.MDP
  • madupite.MatrixPreallocation
  • madupite.getCommWorld
  • madupite.mpi_rank_size
  • madupite.writePETScBinary
  • madupite.MatrixType
  • madupite.MatrixCategory
  • madupite.Matrix
  • madupite.initialize_madupite

Index

• Index