Welcome to Parfun’s documentation!

Parfun is a lightweight and user-friendly Python library designed to help users run Python functions in parallel with ease.

Even with minimal knowledge of parallelism or distributed systems, users can significantly speed up their Python code using Parfun.

Parfun supports multiple execution backends, including Python’s built-in multiprocessing, as well as Scaler and Dask.

Content

• Quick start
  • When to use it
  • Setup and backend selection
  • Your first parallel function
  • Partitioning functions
  • Combining functions
  • Custom partitioning and combining generators
  • Partition size estimate
  • Nested parallel function calls
  • Profiling
• Examples
  • Parallely count bigrams in a text
  • Parallely train a random tree regressor
• Implementation details
  • Design goals
  • Implementation
  • Partition size estimation
• API
  • parallel()
  • Backend setup
  • Partitioning
  • Python lists
  • Pandas dataframes

Indices and tables

• Index

• Module Index

• Search Page