Mission 10 - Train Classification Model
Estimated time for completing this mission: 15 mins
Learning Objective
Learning how training the classification model works using LOGIBLOX tools.
Scenario
Now, you have spotted that in the dataset you have been using there are some cells in "STATUS" that are empty. You report it, and get a response saying that you should perform classification and fill the empty records using trained model. First step is to train on the data.
Know-How Refresh
Classification is used in order to predict a certain category using different feature values. To have a working model, you need to first train it so that it can learn the underlying pattern in the data set. Once the model is trained it can later be used for classification (see next section).
BLOX used in this mission:
- Basics/Start
- MyData/FinalData
- AI/Classifier
Data
The same data set will be used as for the previous task FinalData.
Steps
Please refer to the Navigation Guide to perform the steps below
- In the Module 4 folder click the "Add Item" button to create new logic named Train Classifier
- Drag-and-drop all the necessary BLOX listed above including the dataset FinalData
- Connect the "Start" BLOX to the "MyData" BLOX
- Connect the output from "MyData" BLOX to "Classifier" BLOX. In the "Classifier" BLOX settings, specify "Target Variable" as "STATUS", "Feature Variables" as "PRODUCTLINE", "COUNTRY", and "DEALSIZE", "Training Intensity" as "Medium", and "Model Name" as "Classification Model"
- Press the play button on "Start" BLOX to execute the logic
- After the logic has been executed, double click on the "Classifier" BLOX to see training results
Results and Summary
