AI Reference Kits
In collaboration with Accenture*, Intel offers a series of downloadable AI reference kits to the open source community to help enterprises accelerate their digital transformation journey. These kits are built upon the AI application tools that Intel provides to data scientists and developers.
Solve Important Business Problems
Intel has selected AI reference kits for their impact on the problems they solve across various industries. Each kit includes an AI model developed to deliver higher accuracy, better performance in training and inference, and lower overall total cost of ownership.
Scale AI with the AI Reference Kits Video Demos
View examples of how the AI reference kits are being put to work to solve real-world problems.
Part 1: Scale AI with Optimized, Domain-Specific Reference Kits
Part 2: Scale AI with the New Releases of Optimized Reference Kits
Benefits
Innovate Faster with Open Source Machine Learning Kits
Released to the open source community, the AI models were designed, trained, and tested from among thousands of models to release the one best suited for the use case. Data scientists can further customize and fine-tune the model with data from their industries.
Designed and Optimized for the Machine Learning Pipeline
Each reference kit includes a user guide to accelerate AI deployment in the enterprise, including:
- Data ingesting
- Data preprocessing
- Machine learning modeling
- Hyperparameter tuning
- Model serving and deployment
- Benchmarking
Build More Models Using Less Compute Resources
All AI models are optimized using Intel libraries, frameworks, and tools for AI development needs, powered by oneAPI, for faster training and inferencing performance using less compute resources. The AI reference kits use components from Intel's AI software portfolio, including Intel® AI Analytics Toolkit and the Intel® Distribution of OpenVINO™ toolkit.
Example Results
Up to a 25% increase in prediction accuracy for predictive asset maintenance.
Predict Utility Asset Health to Deliver Higher Service Reliability
Up to 20% faster training and 55% faster inferencing time for visual defect detection.
Automate Visual Quality Control Inspections for Life Sciences
Up to 45% faster inferencing in batch mode for enterprise conversational chatbots.
Step Up Interactions with an Enterprise Conversational AI Chatbot
A Hub of AI Reference Kits
Over 30 AI reference kits with trained machine learning and deep learning models have been released to the open source community. Each kit includes model code, training data, instructions for the machine learning pipeline, libraries, and Intel® oneAPI components.
See all code repositories and their underlying topics in one list.
Combine Multiple Kits to Optimize Business Processes
Multiple AI reference kits can be used together to transform a business. In this example, four stages of a pharmaceutical manufacturing business were automated and optimized. By training large datasets and identifying patterns, AI can improve efficiencies, product quality, and patient experience. This combined AI Reference Kit includes the following four kits:
- Demand Forecasting
- Predictive Asset Maintenance
- Visual Anomaly Detection
- Generative AI Chatbot
Try It for Yourself
This combined AI Reference Kit comes with:
- AI and machine learning source code
- Developer guide
- Docker* container configuration files
- FastAPI* end-point script
- Instructions to deploy on Intel® Developer Cloud
- Synthetic data generation scripts (if applicable)
- Streamlit* web application front end (if applicable)
- Makefile for easier deployment (if applicable)
Featured Kits
Personally Identifiable Information (PII) Anonymization
Use to help automate the masking of personally identifiable information.
Automate Visual Quality Control Inspections for Life Sciences
Use to help pharmaceutical companies automate the visual inspection of pills.
Faster Digital Twin Insights
Use to help speed up training and inference of digital twin behavior modeling.
Image Data Generation
Use to help generate images that can be used for deep learning.
In-Depth Tutorials
Learn tips and techniques to get the most from the AI Reference Kits.
This tutorial shows how to use the Visual Quality Inspection AI Reference Kit to build computer vision solutions for the pharmaceutical industry.
This video demonstrates how to optimize the training cycles, prediction throughput, and accuracy of a machine learning workflow.
Stay Up to Date on AI Workload Optimizations
Sign up to receive hand-curated technical articles, tutorials, developer tools, training opportunities, and more to help you accelerate and optimize your end-to-end AI and data science workflows.
Take a chance and subscribe. You can change your mind at any time.