Icon

10.1 Integrated Deployment - solution

Integrated Deployment - Solution

Solution to the exercise 6.1 for KNIME User Training
- Train a Decision Tree model
- Capture preprocessing and trained model
- Deploy workflow with Integrated Deployment

URL: KNIME Integrated Deployment https://www.knime.com/integrated-deployment
URL: Integrated Deployment - How to Move Data Science into Production https://www.youtube.com/watch?v=MJq154i4O-0

Step 1: Capture Preprocessing- use the Capture Workflow Start node with data port types to start capturing your preprocessing part of the workflow. - use the Number to String node to convert the SentimentRating and WebActivity columns- use the Capture Workflow End node with data port types to end your preprocessing part of the workflow
Step 2: Capture Prediction Model- use the Capture Workflow Start node with data port types to start capturing your prediction model- use the Decision Tree Predictor node for your model predictions- use the Capture Workflow End node with data port types to end your prediction model
Step 3: Combine captured Workflows and Deploy - use the Workflow Combiner node to combine the preprocessing workflow and the model prediction workflow (HINT: the first part needs to be connected to the upper input port)- use the Workflow Writer node to deploy your workflow
Activity I: Create Model for Deployment - capture the preprocessing of your data and the model prediction using the Capture Workflow Start and Capture Workflow End nodes - combine your captured workflows using the Workflow Combiner - write/deploy your workflow using the Workflow Writer node
Number to String
Partitioning
Decision Tree Learner
Decision Tree Predictor
Read CurrentDetailData.table
Table Reader
Scorer

Nodes

Extensions

Links