Scoring Model – Machine Learning Fundamental Concepts

Scoring Model

1. Search the Score Model component and drag it to the canvas.
2. Connect the Score Model’s left input to the Train Model’s output and the right input to the Split Data output as shown in Figure 3-29.

Figure 3-29Adding and configuring the scoring model

Evaluation

1. Find and drag the Evaluate Model component in the canvas. Connect the Score Model’s output to Evaluate Model as shown in Figure 3-30.

Figure 3-30Adding and configuring the evaluation model

Submission

1. Click the Submit button to run our model as shown in Figure 3-31.

Figure 3-31Running our machine learning model

2. Click Create new.
3. Name our experiment.
4. Hit Submit as shown in Figure 3-32.

Figure 3-32Naming the experiment
We see “Completed” marks in some tabs and “Running” in others as shown in Figure 3-33.

Figure 3-33We can see the completed modules and running modules as the workflow progresses.

5. Click the Job Detail as shown in Figure 3-34.

Figure 3-34Click the job detail to check the status of the model

Scored Labels

1. Check the Score Model’s scored labels. Select Scored dataset under “Preview Data” by right-clicking the Score Model component as shown in Figure 3-35.

Figure 3-35Checking the scored labels

2. The Scored Labels column is now available. Our regression model predicts these prices as you can see in Figure 3-36.

Figure 3-36The Scored Labels column has been added

Evaluation Result

1. Select “Evaluation Result” under “Preview Data” by right-clicking the “Evaluate Model” component as shown in Figure 3-37.

Figure 3-37Examining evaluation results

2. The mean absolute error, coefficient of determination, root mean squared error, and others are shown here as shown in Figure 3-38.

Figure 3-38Evaluation results
This includes data preparation, model training, testing, scoring, and evaluation.