Welcome to uarray’s documentation!¶

python-moa is an ambitious project to provide a high level array operation interface that can compile down into low level machine code. In addition moa can provide predictions on the runtime and performance of a computation.

Contents:

  • Introduction
    • Frontend
    • Shape Calculation
    • Denotational Normal Form (DNF)
    • Operational Normal Form (ONF)
    • Performance
    • Conclusion
  • Theory
    • Arrays
    • Shape \(\rho\) rho
    • Dimension \(delta\) delta
    • Pi \(\pi\)
    • Total \(\tau\) tau
  • Implementation
    • Symbol Table
    • Abstract Syntax Tree
    • Symbol Table
    • Shape Calculation
    • Reduction
  • Development Roadmap
    • Reduce (binary operation)
    • MOA Compiler
    • Numpy Broadcasting
    • gufuncs
    • Slice MOA Operation
    • Conditional Indexing
    • Reshape
    • PSI Reduction Implementation
  • Comparisons
  • Machine Independent Optimization
  • Machine Dependent Optimization
  • Publications

Indices and tables¶

  • Index

  • Module Index

  • Search Page

python-moa

Navigation

Contents:

  • Introduction
  • Theory
  • Implementation
  • Development Roadmap
  • Comparisons
  • Machine Independent Optimization
  • Machine Dependent Optimization
  • Publications

Related Topics

  • Documentation overview
    • Next: Introduction

Quick search

©2019, Quansight. | Powered by Sphinx 3.5.1 & Alabaster 0.7.12 | Page source