Average Ratings 0 Ratings
Average Ratings 0 Ratings
Description
Amazon SageMaker Model Monitor enables users to choose which data to observe and assess without any coding requirements. It provides a selection of data types, including prediction outputs, while also capturing relevant metadata such as timestamps, model identifiers, and endpoints, allowing for comprehensive analysis of model predictions in relation to this metadata. Users can adjust the data capture sampling rate as a percentage of total traffic, particularly beneficial for high-volume real-time predictions, with all captured data securely stored in their designated Amazon S3 bucket. Additionally, the data can be encrypted, and users have the ability to set up fine-grained security measures, establish data retention guidelines, and implement access control protocols to ensure secure data handling. Amazon SageMaker Model Monitor also includes built-in analytical capabilities, utilizing statistical rules to identify shifts in data and variations in model performance. Moreover, users have the flexibility to create custom rules and define specific thresholds for each of those rules, enhancing the monitoring process further. This level of customization allows for a tailored monitoring experience that can adapt to varying project requirements and objectives.
Description
A closed-loop universal multivariable optimizer is designed to enhance both the performance and quality of Model Predictive Control (MPC) systems. This optimizer utilizes data from Excel files sourced from Dynamic Matrix Control (DMC) by Aspen Tech, Robust Model Predictive Control Technology (RMPCT) from Honeywell, or Predict Pro from Emerson to develop and refine accurate models for various multivariable-controller variable (MV-CV) pairs. This innovative optimization technology eliminates the need for step tests typically required by Aspen Tech and Honeywell, operating entirely within the time domain while remaining user-friendly, compact, and efficient. Given that Model Predictive Controls (MPC) can encompass tens or even hundreds of dynamic models, the possibility of incorrect models is a significant concern. The presence of inaccurate dynamic models in MPCs leads to bias, which is identified as model prediction error, manifesting as discrepancies between predicted signals and actual measurements from sensors. COLUMBO serves as a powerful tool to enhance the accuracy of Model Predictive Control (MPC) models, effectively utilizing either open-loop or fully closed-loop data to ensure optimal performance. By addressing the potential for errors in dynamic models, COLUMBO aims to significantly improve overall control system effectiveness.
API Access
Has API
No
API Access
Has API
No
Integrations
Amazon SageMaker
Yes
Amazon Web Services (AWS)
Yes
Aspen DMC3
No
Microsoft Excel
No
Integrations
Amazon SageMaker
No
Amazon Web Services (AWS)
No
Aspen DMC3
Yes
Microsoft Excel
Yes
Pricing Details
No price information available.
Free Trial
No
Free Version
No
Pricing Details
No price information available.
Free Trial
Yes
Free Version
No
Deployment
Web-Based
Yes
On-Premises
No
iPhone App
No
iPad App
No
Android App
No
Windows
No
Mac
No
Linux
No
Chromebook
No
Deployment
Web-Based
Yes
On-Premises
No
iPhone App
No
iPad App
No
Android App
No
Windows
No
Mac
No
Linux
No
Chromebook
No
Customer Support
Business Hours
No
Live Rep (24/7)
No
Online Support
Yes
Customer Support
Business Hours
Yes
Live Rep (24/7)
No
Online Support
Yes
Types of Training
Training Docs
Yes
Webinars
No
Live Training (Online)
No
In Person
No
Types of Training
Training Docs
Yes
Webinars
No
Live Training (Online)
Yes
In Person
Yes
Vendor Details
Company Name
Amazon
Founded
2006
Country
United States
Website
aws.amazon.com/sagemaker/model-monitor/
Vendor Details
Company Name
PiControl Solutions
Country
United States
Website
www.picontrolsolutions.com/products/columbo/
Product Features
Machine Learning
Deep Learning
No
ML Algorithm Library
No
Model Training
No
Natural Language Processing (NLP)
No
Predictive Modeling
No
Statistical / Mathematical Tools
No
Templates
No
Visualization
No