Hub
  • Software
  • Blog
  • Forum
  • Events
  • Documentation
  • About KNIME
  • KNIME Hub
  • k10shetty1
  • Spaces
  • Parameter Optimization for Classification
  • 01_Parameter_Optimization_with_Nodes
  • 03_Parameter_Optimization_Loop_on_Logit_with_CV
WorkflowWorkflow

Parameter Optimization Loop with Cross Validation

Parameter optimization Optimization Machine learning Onboarding Data Science
+3

Last edited: 

Drag & drop
Like
Download workflow
Copy short link
Workflow preview
This workflow shows an example of parameter optimization in a logistic regression model. In the logistic regression we optimize step size and variance. The different parameters are output as flow variables by the Parameter Optimization Loop Start node. The parameter settings of the logistic regression algorithm are overwritten by the flow variables and trees with different settings are trained. Since this is a binary classification we can use the ROC Curve node and create a flow variable with the AUC in each iteration. This is then fed into the Parameter Optimization Loop End node. The end node compares the accuracies and supplies the best value in the first output. We use "Hill Climbing Strategy". Please notice that in each iteration cross validation is adopted in order to test the combination of parameters over different folds and get an average performance. This should avoid finding a set of parameter values which performs well only on the training data, but which does not work for new data. That is we are avoiding to overfitting parameters on a training partition.

External resources

  • ML Algorithms and the Art of Parameter Selection - KNIME Blog
  • Parameter Optimization Video - KNIME TV - YouTube

Used extensions & nodes

Created with KNIME Analytics Platform version 4.5.2
  • Go to item
    KNIME Base nodes Trusted extension

    KNIME AG, Zurich, Switzerland

    Version 4.5.2

  • Go to item
    KNIME JavaScript Views Trusted extension

    KNIME AG, Zurich, Switzerland

    Version 4.5.2

  • Go to item
    KNIME Optimization extension Trusted extension

    KNIME AG, Zurich, Switzerland

    Version 4.5.0

  1. Go to item
  2. Go to item
  3. Go to item
  4. Go to item
  5. Go to item
  6. Go to item

Legal

By using or downloading the workflow, you agree to our terms and conditions.

Discussion
Discussions are currently not available, please try again later.

KNIME
Open for Innovation

KNIME AG
Hardturmstrasse 66
8005 Zurich, Switzerland
  • Software
  • Getting started
  • Documentation
  • E-Learning course
  • Solutions
  • KNIME Hub
  • KNIME Forum
  • Blog
  • Events
  • Partner
  • Developers
  • KNIME Home
  • KNIME Open Source Story
  • Careers
  • Contact us
Download KNIME Analytics Platform Read more on KNIME Server
© 2022 KNIME AG. All rights reserved.
  • Trademarks
  • Imprint
  • Privacy
  • Terms & Conditions
  • Credits