nums-project / numsLinks
A library that translates Python and NumPy to optimized distributed systems code.
☆132Updated 2 years ago
Alternatives and similar repositories for nums
Users that are interested in nums are comparing it to the libraries listed below
Sorting:
- Worked example of the process from Python source to CUDA kernel execution with Numba☆41Updated 9 months ago
- Deadline-based hyperparameter tuning on RayTune.☆31Updated 5 years ago
- An Aspiring Drop-In Replacement for Pandas at Scale☆73Updated 3 years ago
- Python bindings for UCX☆135Updated last week
- Distributed XGBoost on Ray☆148Updated last year
- A suite of benchmarks for CPU and GPU performance of the most popular high-performance libraries for Python☆326Updated 8 months ago
- RAPIDS GPU-BDB☆108Updated last year
- Flow with FlorDB 🌻☆154Updated 3 weeks ago
- Serverless Scientific Computing☆87Updated 6 years ago
- A Ray-based data loader with per-epoch shuffling and configurable pipelining, for shuffling and loading training data for distributed tra…☆18Updated 2 years ago
- A benchmark to measure performance of popular Gradient boosting algorithms against popular ML datasets.☆38Updated 2 years ago
- A tensor-aware point-to-point communication primitive for machine learning☆258Updated 2 years ago
- ☆57Updated last year
- Distributed ML Optimizer☆32Updated 3 years ago
- A performant, memory-efficient checkpointing library for PyTorch applications, designed with large, complex distributed workloads in mind…☆157Updated this week
- Pygloo provides Python bindings for Gloo.☆21Updated 3 months ago
- ☆30Updated 2 years ago
- The Foundation for All Legate Libraries☆218Updated last week
- Home for OctoML PyTorch Profiler☆113Updated 2 years ago
- Open Source version of SigOpt API, performing hyperparameter optimization and visualization☆40Updated 4 months ago
- Exoshuffle-CloudSort☆26Updated 2 years ago
- A minimal shared memory object store design☆53Updated 8 years ago
- A small library for creating and manipulating custom JAX Pytree classes☆56Updated 2 years ago
- Tracking Ray Enhancement Proposals☆53Updated 2 months ago
- WIP. Veloce is a low-code Ray-based parallelization library that makes machine learning computation novel, efficient, and heterogeneous.☆18Updated 2 years ago
- Extending JAX with custom C++ and CUDA code☆396Updated 10 months ago
- Zero-copy MPI communication of JAX arrays, for turbo-charged HPC applications in Python☆481Updated this week
- Python combination of Ray and Numba providing compiled distributed arrays, remote functions, and actors.☆31Updated last year
- Benchmark framework to easily compare Bayesian optimization methods on real machine learning tasks☆148Updated 3 years ago
- Utilities for Dask and CUDA interactions☆309Updated last week