-
Book Overview & Buying
-
Table Of Contents
-
Feedback & Rating

Computer Vision on AWS
By :

Once the model training finishes, you will see the performance of the trained model in the Evaluation tab. As you can see in the screenshot here, our trained model has really good precision and recall and a good F1 score for the packt
label (with just eight training images):
Figure 3.13: Evaluating model performance
If you’d like to review how the model performed on your test dataset, select on View test results.
The next step is to start the model so that we can start using it to detect Packt’s logo in images. You can go to the Use model tab and select on Start:
Figure 3.14: Starting the model using the Rekognition Custom Labels console
You will also need to select the number of inference units when you start your model— you will be charged for each inference unit.
Important note
A higher number of inference units will increase...