| Using RAPIDS in Azure Machine Learning
| AI model training time can be reduced by up to 20 times
| Azure Machine Learning SDK and template scripts provided
On the 21st, NVIDIA announced that NVIDIA CUDA-X AI, a data science acceleration library, is now available on Microsoft Azure. This enables data scientists to perform machine learning projects up to 20 times faster.

Azure Machine Learning service has become the first major cloud platform to integrate RAPIDS, a core component of NVIDIA CUDA-X AI. Data scientists can access the RAPIDS open source library set and leverage NVIDIA GPUs in Azure Machine Learning service to perform predictions and analysis at high speed.

RAPIDS delivers impressive performance improvements to various enterprises across multiple industries that use machine learning to generate predictive AI models from vast amounts of data. Examples include retailers looking to improve inventory management, financial institutions seeking more sophisticated financial forecasting, and healthcare organizations aiming to detect diseases more rapidly while reducing operational costs.

Enterprises using RAPIDS can reduce the time required to train AI models by up to 20 times, cutting work that would take days depending on dataset size down to hours or even minutes.
This is not the first time RAPIDS has been integrated by default into a cloud data science platform. Walmart adopted RAPIDS early to improve prediction accuracy.

Srini Venkatesan, Senior Vice President of Walmart Supply Chain Technology and Cloud, stated, "RAPIDS has the potential to significantly expand our feature engineering process," and added, "Using it, we have executed the most complex machine learning models and further improved prediction accuracy."

RAPIDS on Azure Machine Learning service is provided in the form of Jupyter Notebook.

Using Azure Machine Learning service SDK, resource groups, workstations, clusters, and environments are created with configurations and libraries suitable for using RAPIDS code. Additionally, template scripts are provided to allow users to experiment with various data sizes and different types of GPUs and set CPU-based configurations.

Eric Boyd, Corporate Vice President of Microsoft Azure AI, stated, "Our vision is to provide the best technology to support our customers' innovation work," and added, "Azure Machine Learning service is a leading platform for building and deploying machine learning models."

He continued, "We are pleased that data scientists are experiencing significant performance improvements by leveraging Azure combined with NVIDIA GPU acceleration."