Skip to main content

No project description provided

Project description

CuProphet || Prophet built to use CUDA

Build


prophet

Compile instructions:

  • install python 3.9.8

  • Run these:

      cd [insert Dir]/CuProphet-Master
      python -m cmdstanpy.install_cmdstan(version="2.32.1", dir=[put dir here], verbose=True, cores=-1)
    
  • Add to PATH:

      ~.../Users/[user]/.cmdstan/RTools40/mingw64/bin
      ~.../Users/[user]/.cmdstan/RTools40/usr/bin
    
  • Remove any other compilers from user path and system path

  • Merge these folders from the replace folder: \

      ~.../Users/{user}/.cmdstan/\
      ~.../Python39/Lib/site-packages/prophet/\
      ~.../Python39/Lib/site-packages/cmdstanpy/\
    
  • Run these commands:

      cd ".\CuProphet-Master"
      python setup.py install
      cd ".\CuProphet-Master\hppfiles"
      nvcc -o cuda_func_help.dll --shared -v -arch=sm_86 --fmad=false -g cuda_func_help.cu
    

Install instructions

  • Run this:

      pip install cuprophet
    

Donate with PayPal

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

cuprophet-0.2.1.tar.gz (4.8 MB view hashes)

Uploaded Source

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page