RAPIDS API Documentation#
Explore documentation for current and previous RAPIDS releases. Each project card links to its available documentation versions and related resources.
- Stable
Documentation for the current release, recommended for most users.
- Nightly
Documentation for the upcoming release, generated from development branches.
- Legacy
Documentation for the previous release, retained for reference.
RAPIDS APIs#
cuDF is a Python GPU DataFrame library (built on the Apache Arrow columnar memory format) for loading, joining, aggregating, filtering, and otherwise manipulating data.
Dask-cuDF extends Dask where necessary to allow its DataFrame partitions to be processed using cuDF GPU DataFrames instead of Pandas DataFrames. Dask-cuDF is used to leverage multiple gpus and multiple nodes for more performance at larger scales
cuML is a suite of libraries that implement machine learning algorithms and mathematical primitives functions that share compatible APIs with other RAPIDS projects.
cuGraph is a GPU accelerated graph analytics library, the core part of an ecosystem of libraries supporting graph-processing workloads like NetworkX integration (nx-cugraph), GNNs with PyG (cugraph-pyg), and more efficient memory management for large graphs (pylibwholegraph).
cuxfilter acts as a connector library, which provides the connections between different visualization libraries and a GPU dataframe without much hassle. This also allows the user to use charts from different libraries in a single dashboard, while also providing the interaction.
Java bindings for the cuDF library.
The RAPIDS cuCIM is an extensible toolkit designed to provide GPU accelerated I/O, computer vision & image processing primitives for N-Dimensional images with a focus on biomedical imaging.
cuVS is a library for GPU-accelerated vector search and clustering.
KvikIO is a Python and C++ library for high performance file IO using GPUDirect Storage (GDS).
RAFT contains fundamental widely-used algorithms and primitives for vector search, machine learning, and information retrieval.
Various utilities to improve deployment and management of Dask workers on CUDA-enabled systems.
RAPIDS Memory Manager (RMM) is a central place for all device memory allocations in cuDF (C++ and Python) and other RAPIDS libraries. In addition, it is a replacement allocator for CUDA Device Memory (and CUDA Managed Memory) and a pool allocator to make CUDA device memory allocation / deallocation faster and asynchronous.
RAPIDS Multi-Process Foundation (rapidsmpf) is a collection of multi-GPU, distributed memory algorithms written in C++ and exposed to Python.
UCXX is the Python interface for UCX, a low-level high-performance networking library. UCX and UCXX support several transport methods including InfiniBand and NVLink while still using traditional networking protocols like TCP.
nvForest is a highly-optimized and lightweight RAPIDS library that enables fast inference for decision tree models on NVIDIA GPUs and CPUs.
RAPIDS Libraries#
libcudf is a C/C++ CUDA library for implementing standard dataframe operations.
libcuml is a C/C++ CUDA library for cuML.
libkvikio is a C++ header-only library providing bindings to cuFile, which enables GPUDirect Storage (GDS).
librapidsmpf is a C++ collection of multi-GPU, distributed memory algorithms.
libucxx is an object-oriented C++ interface for UCX, with native support for Python bindings.
This is a collection of CMake modules that are useful for all CUDA RAPIDS projects. By sharing the code in a single place it makes rolling out CMake fixes easier.
Inactive Projects#
cuProj is a GPU-accelerated geographic and geodetic coordinate transformation library which supports projecting coordinates between coordinate reference systems (CRSes), compatible with PyProj.
cuSignal functionality has been moved to CuPy. Please see the CuPy documentation for more information.
cuSpatial is a GPU-accelerated vector GIS library including binary predicates (DE-9IM), point-in-polygon, spatial join, distances, and trajectory analysis.
libcuproj is a C++ header-only library for GPU-accelerated geographic and geodetic coordinate transformation library which supports projecting coordinates between coordinate reference systems (CRSes), similar to PROJ.
libcuspatial is a GPU-accelerated header-only C++ vector GIS library including binary predicates (DE-9IM), point-in-polygon, spatial join, distances, and trajectory analysis.