Step 5. Run the build

This next section will describe how to run a build.

Now that our workflow is properly configured , our model is ready to train. For that we need to "RUN" our project.

Click on the 'Run' button on the extreme left corner of the viewport. Notice the green leds switching on: they signal that the model is running properly.

From the main Dashboard, our project is also labelled as "success".

As we run our project , a series of behavior happens to some particular elements of the workflow. Some acknowledge the completion of the previous operations whereas other notify us about the status of our project.

Green leds switch on as the project runs.

Among others:

1. The evaluator for ML : sends an info about the accuracy of our model. As its name suggests, this module's role is to assess the performance of our model.

2. The Item Saver: informs about under which name our current model is saved. Here, it is saved as 'SVC_model'. It is a default name for all models.

>>> We can rename ours from the parameters. Let us change it into 'SVC_model_Iris'.

We can rename our model through the Item saver.

However, since we changed the name of our model, we need to rerun it to apply the change into the whole model. Take notice that the name has also changed in the logs.

We need to rerun after name changes.

3. The logs We can check the logs by clicking the log icon on the extreme right of the menu. We can choose to view all records, per level or per module by ticking on the corresponding radio button. .

The logs recap the information about the project.

The logs also display the name under which our model will be stored , which is ‘SVC_model’.

As we click on modules, we retrieve this latter camong the list of Trained Models.

Our model displays in the menu of trained models.