Hello, everyone, and welcome to the final video of the modeling phase. Before we start with our testing, I want to congratulate each and every one of you for coming this far in the course, if there are some parts of the code that you haven't understood, don't be discouraged, ask in the comment section, I'll be very happy to come and help you. Or you can simply go back to that part and code again, or just watch a video. Let's get back to the testing. For this video, I've already written the code, but don't worry, the same exact code will be used in the flask application down the road. And I'll explain each part of the code in that lesson as well.

To test our pipeline, we have defined our model and the new session, execute the cell. Now we have a newly created session with a completely random initialized model. So the next step is to load our checkpoint and restore the same model. Then let's load our training set. That these will help us load and visualize the most similar images to our query image. Here you can see the first time pass as an example.

And before our testing, there is only one more thing to load training set vectors. Now that we are done with that we have everything ready, we can choose a test image from our test set. Let's choose this airplane. execute a simple inference function with all our inputs. And now we have our resulting images. Let's visualize our query image just to see what we are working with.

Now, are you ready to see our results? execute the cell and congratulations, you have successfully implemented a very good image to Image Search system. We have created a system that recommends us all similar images to the one that we have choose as a reference. How cool is that? Once again, great Java guys, in the next tutorial, we are going to prepare the code for our web application and finally, start creating our flask back end. For now, if you have any questions or comments, please post them in the comment section.

Otherwise, I'll see you in the next tutorial.