To join a AWS DeepRacer community race - AWS DeepRacer

To join a AWS DeepRacer community race

Note

To join an AWS DeepRacer community race, you first need to receive a link to the race from the race organizer.

When you receive an invitation to join an AWS DeepRacer race, find out whether it's a LIVE or classic race.

Classic race

Classic races are asynchronous events that do not require real-time interaction. Your invitation link gives you access to submit a model to the race and view the leaderboard. You can submit unlimited models at any time within the race's opening and closing dates to achieve your best standing on the leaderboard. Results and videos for classic races are viewable for submitted models on the Leaderboard page as soon as the race is initiated. All classic races are private events.

LIVE race

LIVE races are real-time racing events where you gather virtually with other racers taking turns to compete for the fastest time on the leaderboard. You can enter multiple models, but only the last model you submit before the submission window closes will be used. During your race, you have the option to try interactive speed controls, which temporarily override your model’s speed parameters giving you the opportunity to make strategic, real-time adjustments. LIVE races can be broadcast privately among invited racers or publicly for anyone to see.

If the competition format is not specified in your invitation, check your race card. LIVE races say, "LIVE," and tell you the date and time of the synchronous event. Classic races give you the date range for the asynchronous competition.


                
                    Your race card indicates your competition format.

Join an AWS DeepRacer community race as a race participant

If you're new to AWS and receive an invitation to join an AWS DeepRacer community race, follow the steps in To join as a new user. If you're invited to an active community race and you've entered an AWS DeepRacer race before, follow the steps below in To join a Classic race or To join a LIVE race as is appropriate for your competition format.

If you're new to AWS and receive an invitation link to join an AWS DeepRacer community race, choose the link to go to the AWS DeepRacer console and then sign up for an AWS account before proceeding to join the race.

As a new AWS DeepRacer user or a first-time participant to any AWS DeepRacer race, follow the steps to join a community race in the AWS DeepRacer console.

To join a race as a new user
  1. Create an AWS account in the AWS DeepRacer console.

  2. Once you are set up and signed in, choose the link shared with you by the race organizer to open the Race page.

  3. When prompted to create an AWS DeepRacer racer name, enter a name that you will use as identification across all the AWS DeepRacer leaderboards. Once you choose a racer name, you cannot change it.

  4. On the Race details page, expand Get started racing.

  5. Choose Get started with RL to get a quick introduction to training an AWS DeepRacer model for autonomous driving.

  6. Train and evaluate your model for the race in the AWS DeepRacer console.

    For more information on training your model, see Train your first AWS DeepRacer model .

  7. Navigate to Community races.

  8. Find the race you're invited to. Choose Enter race on the race card.

    
                                    
                                        Image: Enter a AWS DeepRacer community race from the console.
  9. Follow the steps in To join a Classic race or To join a LIVE race as is appropriate for your race's competition format.

  1. Select the link you received from the race organizer. If you're not already signed in to your account in the AWS DeepRacer console, you'll be prompted to sign in.

  2. Once signed in to the AWS DeepRacer console, the link will take you to the Race page. The Race page displays the race details, leaderboard, and your racer info. Choose Enter race.

    
                                    
                                        Image: Enter a Classic community race from the console.
  3. On the Enter race page, under Choose model, choose a trained model and then choose Enter race.

    
                                    
                                        Image: Submit a model to a AWS DeepRacer community race from the console.
  4. If your model is evaluated successfully against the racing criteria, watch the event's leaderboard to see how your model ranks against other participants.

  5. Optionally, choose Watch to view a video of your vehicle's performance or choose Download evaluation logs to review a detailed look at the outputs produced.

    
                                    
                                        Image: Submit a model to a AWS DeepRacer community race from the console.
  6. Choose Race again to enter another model. You can submit unlimited models at any time within the race's opening and closing dates to achieve your best standing on the leaderboard.

  1. Select the link you received from the race organizer. If you're not already signed in to your account in the AWS DeepRacer console, you'll be prompted to sign in.

  2. Once signed in to the AWS DeepRacer console, the link will take you to the Race page. The Race page displays the race details and leaderboard. Choose Enter race.

    
                                    Image: Enter a LIVE community race from the console.
  3. On the Enter race page, under Choose model, choose a trained model and then choose Enter race.

    
                                    
                                        Image: Submit a model to a AWS DeepRacer community race from the console.
  4. If your model is evaluated successfully against the racing criteria, watch the event's leaderboard to see how your model ranks against other participants.

  5. Optionally, for LIVE races, select Calendar to add the LIVE racing event to your calendar.

  6. Choose Race again to enter another model. You can enter multiple models, but only the last model you submit before the submission window closes will be used.