RAFT

E890465

RAFT is a core NVIDIA RAPIDS library that provides reusable, GPU-accelerated primitives and algorithms for building high-performance data science and machine learning workflows.

Try in SPARQL Jump to: Surface forms Statements Referenced by

All labels observed (1)

Label Occurrences
RAFT canonical 1

Statements (48)

Predicate Object
instanceOf GPU-accelerated library
NVIDIA RAPIDS library
open-source software library
developer NVIDIA NERFINISHED
domain data science
high-performance computing
machine learning
ecosystem RAPIDS AI NERFINISHED
feature distributed computing support
high-level algorithmic building blocks
low-level GPU primitives
multi-GPU support
performance-optimized implementations
focusesOn composability of algorithms
reusability of GPU components
goal enable reusable GPU-accelerated components
simplify development of GPU-accelerated ML algorithms
integration C++ APIs
Python bindings via RAPIDS
license Apache License 2.0
optimizedFor high throughput
low latency
partOf RAPIDS NERFINISHED
programmingLanguage C++
CUDA NERFINISHED
Python
provides GPU-accelerated algorithms
GPU-accelerated primitives
building blocks for data science workflows
building blocks for machine learning workflows
repository https://github.com/rapidsai/raft
supports clustering primitives
dense linear algebra operations
distance and similarity computations
graph analytics primitives
multi-node workloads
random number generation
single-GPU workloads
sparse linear algebra operations
statistics primitives
supportsHardware NVIDIA GPUs NERFINISHED
supportsPlatform CUDA-enabled systems
targetUser HPC developers
data scientists
machine learning engineers
usedBy cuGraph NERFINISHED
cuML NERFINISHED
other RAPIDS libraries

Referenced by (1)

Full triples — surface form annotated when it differs from this entity's canonical label.

NVIDIA RAPIDS component RAFT