Welcome to crow documentation!

Crow is a multi-GPU and multi-CPU implementation of non-negative matrix tri-factorization.

Contents:

• Setup guide
• Docker containers
  • Connect to a running container
  • Docker volumes
  • Non-sudo users
  • Test run
  • Supported platforms
• Host requirements
  • Quick setup
  • CUDA
  • Docker engine
  • Nvidia-docker
  • Add user to docker group
• Legacy nvidia-docker
  • nvidia-docker-compose
• Manual setup
  • Python
  • CUDA
  • openMPI
  • openBLAS
  • NumPy
  • Other system dependencies
  • Install crow and python dependencies
  • GPU module dependencies
  • Install crow module
  • Test the configuration
• Tutorial
  • Examples
  • Command line arguments
• Docker volumes
• Data manipulation
  • Input data
  • Convert between different csv and coo formats
  • Convert between npz and csv
  • Pickle to csv
• Download data
• Benchmark
  • Quick start
  • Benchmark list
  • Visualize results
  • Manual setup
• Co-clustering example
  • Setup instructions
  • Run Experiments
• Reference

Indices and tables

• Index
• Module Index
• Search Page