Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Feature: added Contexts and CUDA integration + JaccardWeights preprocessing + 1D Array formats #81

Merged
merged 25 commits into from
Mar 29, 2022

Conversation

AmroAlJundi
Copy link
Contributor

Added support for different architectures using Contexts. For details, check #80.

Changes:

  • Expanded library to accept differnt architectures (GPU, IPU, etc.) through Contexts.
  • Added a new preprocessing: JaccardWeights.
  • Added a new Formats: the 1D Array.
  • Added CUDA support.
    • CUDAContext class.
    • Two new formats, CUDACSR and CUDAArray.
    • Added a Jaccard Weights CUDACSR function.
    • Compilation and installation modification to include CUDA in compilation and header-only modes.
    • Changed file structure.
    • Updated readme and getting started page with CUDA compilation details.

A note about installation:

  • if the library is in compile-mode, then regardless of whether the user compiles and installs it with CUDA on or off, they will be able to use it with C++ compilers and nvcc .
  • On the other hand, in header-only mode, if the library is installed with CUDA on, then the user must use nvcc to compile it. With CUDA off, they can use either. However, they won't have access to the CUDA functionality.

@AmroAlJundi AmroAlJundi added state: review needed type: feature Brand new functionality, features, workflows, endpoints, etc labels Mar 19, 2022
@AmroAlJundi AmroAlJundi merged commit 0af7b7c into develop Mar 29, 2022
@AmroAlJundi AmroAlJundi deleted the feature/cuda_integration branch March 29, 2022 08:20
SinanEkm pushed a commit that referenced this pull request Aug 19, 2024
Feature: added Contexts and CUDA integration + JaccardWeights preprocessing + 1D Array formats
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
state: review needed type: feature Brand new functionality, features, workflows, endpoints, etc
Projects
None yet
Development

Successfully merging this pull request may close these issues.

Integrating different architectures (GPUs, IPUs, etc.) into SparseBase
3 participants