Intel® AI Analytics Toolkit (AI Kit)
Achieve End-to-End Performance for AI Workloads
Accelerate Data Science & AI Pipelines
The AI Kit gives data scientists, AI developers, and researchers familiar Python* tools and frameworks to accelerate end-to-end data science and analytics pipelines on Intel® architectures. The components are built using oneAPI libraries for low-level compute optimizations. This toolkit maximizes performance from preprocessing through machine learning, and provides interoperability for efficient model development.
Using this toolkit, you can:
- Deliver high-performance, deep-learning training on Intel® XPUs and integrate fast inference into your AI development workflow with Intel®-optimized, deep-learning frameworks for TensorFlow* and PyTorch*, pretrained models, and low-precision tools.
- Achieve drop-in acceleration for data preprocessing and machine-learning workflows with compute-intensive Python* packages, Modin*, scikit-learn*, and XGBoost, optimized for Intel.
- Gain direct access to analytics and AI optimizations from Intel to ensure that your software works together seamlessly.
Download the Toolkit
Accelerate end-to-end machine learning and data science pipelines with optimized deep learning frameworks and high-performing Python* libraries.
Develop in the Cloud
Get what you need to build, test, and optimize your oneAPI projects for free. With an Intel® DevCloud account, you get 120 days of access to the latest Intel® hardware—CPUs, GPUs, FPGAs—and Intel oneAPI tools and frameworks. No software downloads. No configuration steps. No installations.
Data Science Workstation Powered by the AI Kit
The Data Science Workstation lineup from Intel and Intel's OEM partners provides data scientists, data analysts, and developers productive and cost-effective AI development solutions to quickly generate insights for their organizations.
- Open, optimized software tools are coupled with optimal compute and memory hardware configurations to deliver the best out-of-the-box developer experience, whether you are prototyping or developing production AI.
- High-memory systems can fit large datasets for efficient preprocessing, considerably shortening the time required to sort, filter, label, and transform your data.
- Familiar Python*‡ APIs deliver software accelerations of up to 10x to 100x for training and inference.
Together, the hardware and software bundle inside the workstations enables data scientists to easily iterate and analyze data at scale.
Central to the optimized AI software stack of the Data Science Workstation† is the AI Kit that accelerates end-to-end data science and machine-learning pipelines using tools and frameworks that are based on Python. The components of the toolkit are open and standards-based, while offering both drop-in acceleration with almost no code changes and seamless scaling to multiple nodes and architectures.
† By downloading the AI Kit for Data Science Workstation, you agree to fully comply with the terms and conditions of the License Agreement.
‡ Python support: 3.6 – 3.8. Operating system support: Centos* 7.x and 8.x, Ubuntu* 18.04 LTS and 20.04 LTS, SLES 12.x and 15.x, RHEL* 7.x and 8.x.
In the News
CERN Uses Intel® Deep Learning Boost & oneAPI to Juice Inference without Accuracy Loss
Researchers at CERN and Intel showcase promising results with low-precision optimizations that exploit heterogeneous operations on CPUs for convolutional Generative Adversarial Networks (GAN).
LAIKA Studios* & Intel Join Forces to Expand the Possibilities in Stop-Motion Film Making
See how LAIKA Studios* and the Intel Applied Machine Learning team used tools from the AI Kit to realize the limitless scope of stop-motion animation.
Accelerate PyTorch* with oneAPI Libraries
Harnessing Intel® Deep Learning Boost and oneAPI libraries, Intel and Facebook* collaboratively improved PyTorch CPU performance across multiple training and inference workloads.
MLPerf Results for Deep Learning Training and Inference
Reflecting the broad range of AI workloads, Intel submitted results for MLPerf Release v.0.7 for training and inference. Results in each use case demonstrated that Intel continues to improve standards for Intel® Xeon® Scalable processors as universal platforms for CPU-based machine learning training and inference.
An Open Road to Swift DataFrame Scaling
This podcast looks at the challenges of data preprocessing, especially time-consuming, data-wrangling tasks. It discusses how Intel and OmniSci* are collaborating to provide integrated solutions that improve DataFrame scaling.
Superior Machine-Learning Performance on the Latest Intel® Xeon® Scalable Processors
Intel® Extension for Scikit-learn* gives data scientists the performance and ease-of-use they need to run machine-learning algorithms with a simple drop-in replacement for the stock scikit-learn. This article showcases the speedups achieved on the latest Intel Xeon Scalable processors when compared to processors from NVIDIA* and AMD.
Optimize Performance of Gradient Boost Algorithms
Intel has been constantly improving training and inference performance for XGBoost algorithms. The following blogs compare the training performance of XGBoost 1.1 on a CPU with third-party GPUs, and showcase how to speed up inference with minimal code changes and no loss of quality.
Accelerate Lung Disease Diagnoses with Intel® AI
Accrad developed CheXRad, an AI-powered solution to rapidly detect COVID-19 and 14 other thoracic diseases in the clinics and hospitals of Africa. With the help of Intel, they were able to train, optimize, and deploy in less time and at a lower operational cost than available alternatives.
Features
Optimized Deep Learning
- Leverage popular, Intel-optimized frameworks—including TensorFlow and PyTorch—to use the full power of Intel architecture and yield high performance for training and inference.
- Expedite development by using the open-source, pretrained, machine-learning models that are optimized by Intel for best performance.
- Take advantage of automatic accuracy-driven tuning strategies along with additional objectives like performance, model size, or memory footprint using low-precision optimizations.
Data Analytics and Machine-Learning Acceleration
- Increase machine-learning model accuracy and performance with algorithms in scikit-learn and XGBoost, optimized for Intel architectures.
- Scale out efficiently to clusters and perform distributed machine learning by using Intel Extension for Scikit-learn.
High-Performance Python*
- Take advantage of the most popular and fastest growing programming language for AI and data analytics with underlying instruction sets optimized for Intel® architectures.
- Process larger scientific data sets more quickly using drop-in performance enhancements to existing Python code.
- Achieve highly efficient multithreading, vectorization, and memory management, and scale scientific computations efficiently across a cluster.
Simplified Scaling across Multi-node DataFrames
- Seamlessly scale and accelerate pandas workflows to multicores and multi-nodes with only one line of code change using the Intel® Distribution of Modin*, an extremely lightweight parallel DataFrame.
- Accelerate data analytics with high-performance back ends, such as OmniSci.
What’s Included
Intel® Optimization for TensorFlow*
In collaboration with Google*, TensorFlow has been directly optimized for Intel architecture using the primitives of Intel® oneAPI Deep Neural Network Library (oneDNN) to maximize performance. This package provides the latest TensorFlow binary version compiled with CPU-enabled settings (--config=mkl).
Intel® Optimization for PyTorch*
In collaboration with Facebook*, this popular deep-learning framework is now directly combined with many optimizations from Intel to provide superior performance on Intel® architecture. This package provides the binary version of latest PyTorch release for CPUs, and further adds extensions and bindings from Intel with Intel® oneAPI Collective Communications Library (oneCCL) for efficient distributed training.
Model Zoo for Intel® Architecture
Access pretrained models, sample scripts, best practices, and step-by-step tutorials for many popular open-source, machine-learning models optimized by Intel to run on Intel Xeon Scalable processors.
Provide a unified, low-precision inference interface across multiple deep-learning frameworks optimized by Intel with this open-source Python library.
Intel® Extension for Scikit-learn*
Seamlessly speed up your scikit-learn applications on Intel® CPUs and GPUs across single-nodes and multi-nodes. This extension package dynamically patches scikit-learn estimators to use Intel® oneAPI Data Analytics Library (oneDAL) as the underlying solver, while achieving the speed up for your machine-learning algorithms. The toolkit also includes stock scikit-learn to provide a comprehensive Python environment installed with all required packages. The extension supports up to the last four versions of scikit-learn, and provides flexibility to use with your existing packages.
XGBoost Optimized for Intel Architecture
In collaboration with the XGBoost community, Intel has been directly upstreaming many optimizations to provide superior performance on Intel CPUs. This well-known, machine-learning package for gradient-boosted decision trees now includes seamless, drop-in acceleration for Intel architectures to significantly speed up model training and improve accuracy for better predictions.
Accelerate your pandas workflows and scale data preprocessing across multi-nodes using this intelligent, distributed DataFrame library with an identical API to pandas. The library integrates with OmniSci in the back end for accelerated analytics. This component is available only via the Anaconda* distribution of the toolkit. To download and install it, refer to the Installation Guide.
Intel® Distribution for Python*
Achieve greater performance through acceleration of core Python numerical and scientific packages that are built using Intel® Performance Libraries. This package includes Numba Compiler*, a just-in-time compiler for decorated Python code that allows the latest Single Instruction Multiple Data (SIMD) features and multicore execution to fully use modern CPUs. You can program multiple devices using the same programming model, DPPy (Data Parallel Python) without rewriting CPU code to device code.
Documentation & Code Samples
Documentation
- Installation Guides: Intel | Anaconda | Docker*
- Package Managers: YUM | APT | Zypper
- Get Started Guides: Linux* | Windows* | Containers
- Release Notes
- Maximize TensorFlow Performance on CPUs: Considerations and Recommendations for Inference Workloads
- Scikit-learn Get Started
Training
Specifications
Processors:
- Intel® Xeon® processors
- Intel Xeon Scalable processors
- Intel® Core™ processors
Language:
- Python
Operating systems:
- Linux*
Development environments:
- Compatible with Intel® compilers and others that follow established language standards
- Linux: Eclipse* IDE
Distributed environments:
- MPI (MPICH-based, Open MPI)
Support varies by tool. For details, see the system requirements.
Get Help
Your success is our success. Access these support resources when you need assistance.
- AI Kit Support Forum
- Deep-Learning Frameworks Support Forum
- Machine-Learning and Data Analytics Support Forum
For additional help, see our general oneAPI Support.
Stay in the Know with All Things CODE
Sign up to receive the latest trends, tutorials, tools, training, and more to
help you write better code optimized for CPUs, GPUs, FPGAs, and other
accelerators—stand-alone or in any combination.
Product and Performance Information
Performance varies by use, configuration and other factors. Learn more at www.Intel.com/PerformanceIndex.