RAPIDS AI

Our GPU-optimized pipeline enabled us to quickly iterate and run many experiments in a short time, giving us a competitive advantage to…

As part of its RAPIDS initiative, NVIDIA is partnering with the community to build an ecosystem for distributed data science on GPUs.

Learn how to implement a fast TargetEncoder on GPUs with built-in optimizations and advanced features

See how to build end-to-end NLP pipelines in a fast and scalable way on GPUs — from feature engineering to inference.

Learn how metadata and reference counting can be used to determine when a checkpoint should be created to achieve a zero data-loss…

See the latest work from the GPU data community and a summary of new functionality in RAPIDS core libraries

For developers who use python, anyone can easily use the power of the GPU to accelerate their code as we’ve shown. Try it out today!

Learn how to use RAPIDS cuML with scikit-learn’s ensemble model APIs to achieve more than 100x faster boosting, bagging, and stacking.

This video tutorial walks through an example of accelerated hyperparameter optimization (HPO) jobs using RAPIDS on Microsoft AzureML.

See the latest performance improvements for string processing across RAPIDS dataframe and machine learning libraries.

One can view cuStreamz as a bridge that connects Python-Streaming and GPUs — with sophisticated and reliable streaming features.

Accelerating data preparation is critical to address the volume and scale that large-scale recommender systems require.

cuSpatial accelerates spatial distance and similarity functions, indexing, and trajectory smoothing to provide immediate results.

In this blog post, we discuss how to use the feature_extraction.text package in RAPIDS cuML to perform fast text vectorizing on GPUs.

Learn how to use Unified Memory and Oversubscription to scale cuGraph to utilize datasets 4x larger than what fits into GPU memory

Combining Python/CUDA JIT for Flexible Acceleration in RAPIDS

Frigate is a simple tool which will read in the files from a Helm Chart and automatically generate a documentation page.

A tutorial on how to use the Hyperopt HPO package with RAPIDS on Databricks Cloud to optimize the accuracy of a random forest classifier.

A simple walk through how to integrate a RAPIDS cuML model with MLFlow to make our lives easier

Supporting the Community by Accelerating Learning and Insight on NVIDIA GPUs

RAPIDS on DGX A100: Even Faster with Lower TCO

The Numba compilation pipeline starts with Python source code and takes it through the following stages to generate PTX code for CUDA GPUs

This tokenizer, applied as a pre-processing step before input into a BERT language model, runs up to 270x faster than CPU implementations

Full of new features like deterministic training on GPUs, tighter ecosystem integrations, large reductions in memory usage, and more