Weidmueller Industrial AutoML

Автор: Weidmüller Interface GmbH & Co. KG

Microsoft preferred solution

Easily create, operate and optimize ML models - No ML expertise needed

Weidmüller Industrial AutoML empowers machine and application experts to create, validate, deploy and operate machine learning models for manufacturing use cases – no data science skills required. To generate a ML model, a domain expert is guided through four simple steps: upload and enrich machine or process data, contextualize the data based on domain knowledge (e.g., labeling normal and abnormal behavior or creating own features), trigger the automatic model creation process for a specific model type (e.g., anomaly detection, classification) and finally select the best performing ML model. This solution is designed for application and machine experts, so that machine learning models can be created with no help from a data scientist. The solution includes the following features:
  • Machine and process data import
  • Data exploration and automatic quality check
  • Custom feature engineering
  • Enrichment of data using domain knowledge through data tagging 
  • Projects and datasets management 
  • Automated model creation, including feature engineering, pre-processing and post-processing
  • Model results explanation and validation

Быстрый обзор