Accelerating Model Development and Generating Ensembles with Diviner
October 28, 2025
Eigenvector Research, Inc. is pleased to offer Accelerating Model Development and Generating Ensembles with Diviner, a live, online short course demonstrating the use of the Diviner tool in PLS_Toolbox and Solo. This course is free for our software users with up-to-date maintenance agreements.
Complete information about the course can be found by following the links below.
- Target Audience
- Course Description
- About the Instructors
- Registration and Fees
- Schedule
- Course Outline
Target Audience
Accelerating Model Development and Generating Ensembles with Diviner is aimed at spectroscopists and engineers who need to develop quality regression calibration models efficiently. It is recommended that participants be familiar with Principal Components Analysis (PCA) and multivariate regression methods such as Partial Least Squares (PLS). Some knowledge of Preprocessing methods and Variable Selection would also be useful. Courses in these topics can be found on the EigenU Recorded Courses page.
Course Description
Diviner, our semi-automated machine learning tool (semi-autoML) for developing calibration models, was introduced with PLS_Toolbox and Solo 9.5 in 2024. (Divine—to discover or locate something by intuition, insight or supernatural means.) Unlike conventional autoML methods that yield a single optimal model, our approach creates a diverse family of models and ranks them by their cross-validation performance, degree of overfitting, and prediction error on validation sets. These models may be further refined to produce a single “optimal” model. Alternately, a group of candidate models can be used to create an ensemble model that harnesses the diversity of the candidate models. These models are combined using voting regression. This leverages the complementary strengths and reduces individual model weaknesses, enhancing overall predictive accuracy and robustness. In this hands-on class participants will learn to use Diviner to create single models and ensembles.
In order to take advantage of the hands-on examples participants should equip their computers with Solo 9.5 or PLS_Toolbox 9.5 (and MATLAB 2021a through 2024b) installed. Demo copies will work just fine. Users with Eigenvector accounts can download free demos. If you don’t have an account, start by creating one.
About the Instructors
The course will be led by Eigenvector’s Principal Consultant and Data Scientist Manuel A. Palacios. Manny has delivered many chemometrics courses at scientific conferences, on-site for companies, on-line and at our popular Eigenvector University each year in Seattle. Manny will be assisted by PLS_Toolbox creator and Eigenvector President Barry M. Wise.
Registration and Fees
Please register through your Eigenvector account. The course is free for PLS_Toolbox and Solo users with current maintenance agreements as of October 1, 2025. $795 for all others. Current users please to orders@eigenvector.com for a coupon code that will waive the course fee. Registration will open September 1, 2025.
Schedule
Schedule, Eastern Daylight Time (EDT)
09:45 – 10:00 Zoom available for log in
10:00 – 10:50 Instruction
10:50 – 11:00 Break
11:00 – 11:50 Instruction
11:50 – 12:00 Break
12:00 – 12:30 Instruction
12:30 – 12:50 Wrap-up Discussion and Questions
Course Outline
Accelerating Model Development and Generating Ensembles with Diviner will cover the follow topics:
- The Problem of Calibration Model Development
- The Promise and Problems with Automated Machine Learning (AutoML)
- Diviner Approach – Semi-AutoML
- A perfect model?
- Diviner Workflow
- Hands-on Examples
- Improving Prediction Error and Robustness with Ensemble Models
- Hands-on Examples
- Conclusions
- Wrap-up and Discussion