Analytics and Machine Learning Toolkit

July 2018, 377059B-01

The LabVIEW Analytics and Machine Learning Toolkit includes VIs for training machine learning models to discover patterns in large amounts of data. You can deploy trained machine learning models to detect anomalies, recognize patterns, or make predictions in new data.

The following figure illustrates the machine learning process.

The machine learning process contains the following steps:

• Data Collection—Collects data using data acquisition devices or other sources.
• Feature Extraction—Extracts features based on your domain knowledge using available signal processing tools in LabVIEW, such as the LabVIEW Advanced Signal Processing Toolkit, Electrical Power Toolkit, NI Sound and Vibration Measurement Suite, Vision Development Module, and so on.
• Feature Reduction—Reduces the dimension of data so that you can use simplified data for model training.
• Model Creation—Trains machine learning models using the training data.
• Model Validation—Validates models using model evaluation metrics.
• Deployment—Deploys trained models on deployment data.

Note  Among these steps, the Analytics and Machine Learning Toolkit supports feature reduction, model creation, model validation, and deployment.

The Analytics and Machine Learning Toolkit uses some algorithms from the IMS Center Watchdog Agent. Refer to the IMS Center website at www.imscenter.net for more information about the center.

© 2017–2018 National Instruments. All rights reserved.