This guide walks you through building, training, and evaluating your first reinforcement learning model via the AWS DeepRacer console.

To get started building your first model, follow the steps below.

1.    In your AWS account, go to the Management Console.

2.    Choose  us-east-1 region at the top right corner of the Regions dropdown menu.

3.    From the top left of the console,  click Services, type DeepRacer in the search  box, and select AWS DeepRacer.  That will open  the AWS DeepRacer console.

4.    On the landing page, click Create model.

5.    Create a model  name and description.

6.    Click Create resource.

7.    Under Environment simulation, select a track from the list. It’s recommended to start with the “re:Invent 2018 Track” and then explore more  tracks from there.

8.    Under Action space,  familiarize yourself with these settings and accept the defaults.

9.    Read through both basic and advanced reward functions to familiarize yourself with the code. For the  purposes of this exercise, choose to insert  code for the  basic reward  function and leave the code unchanged.

10.  Expand Algorithm settings and review the  different hyperparameters available  to set. For this exercise, accept the  default hyperparameter settings.

11.  For the Stop conditions go ahead and choose a max time of 60 minutes and click Start training.

12.  Select the name of the model  to watch  the  live stream in the simulator. Notice how your car is moving and become familiar with the general look and feel of the simulation. The new training will initiate in about 6 minutes. You then must  wait for the training job to complete before proceeding. If you choose a maximum time of 60 minutes, it will take up to 60 minutes for this training job to complete. It is complete when  the status reads “Ready.”

Note: Since the training job may take more  than an hour.

13.  Now, select the name of the model  you just trained and click Start new  evaluation.

14.  Select the same  track you used for training.

15.  Click Start evaluation. The evaluation results will update in approximately 5 minutes.

16.  Before the evaluation job completes, take some  time to watch the simulator to see how your car’s performing in real time. Look for things that you might want to tweak in the  future and note them down. You’ll have a chance  to apply some of those ideas in a later exercise.

17.  Watch the evaluation results update and write them down. You will use your results from this model  as a benchmark that you can compare to as you tweak your model in later exercises.

Additional Resources

•    For more  information about AWS DeepRacer, see https://aws.amazon.com/deepracer/.

•    To troubleshoot and collaborate about the AWS DeepRacer, see https://forums.aws.amazon.com/forum.jspa?forumID=318.