Provides a quick and easy way to get up and running with a DeepRacer training environment in AWS or Azure, using either the Azure N-Series Virtual Machines or AWS EC2 Accelerated Computing instances, or locally on your own desktop or server.
DeepRacer-For-Cloud (DRfC) started as an extension of the work done by Alex (https://github.com/alexschultz/deepracer-for-dummies), which is again a wrapper around the amazing work done by Chris (https://github.com/crr0004/deepracer). With the introduction of the second generation Deepracer Console the repository has been split up. This repository contains the scripts needed to run the training, but depends on Docker Hub to provide pre-built docker images. All the under-the-hood building capabilities have been moved to my Deepracer Build repository.
DRfC supports a wide set of features to ensure that you can focus on creating the best model:
- Time Trial
- Object Avoidance
- Multiple Robomaker instances per Sagemaker (N:1) to improve training progress.
- Multiple training sessions in parallel - each being (N:1) if hardware supports it - to test out things in parallel.
- Connect multiple nodes together (Swarm-mode only) to combine the powers of multiple computers/instances.
- Evaluate independently from training.
- Save evaluation run to MP4 file in S3.
- Training metrics and trace files are stored to S3.
- Optional integration with AWS CloudWatch.
- Optional exposure of Robomaker internal log-files.
- Supports both Docker Swarm (used for connecting multiple nodes together) and Docker Compose (used to support OpenGL)
- Initial Installation
- Upload Model to Console
- Using multiple Robomaker workers
- Running multiple parallel experiments
- GPU Accelerated OpenGL for Robomaker
- Having multiple GPUs in one Computer
- Installing on Windows
- Run a Head-to-Head Race
- For general support it is suggested to join the AWS DeepRacing Community. The Community Slack has a channel #dr-drfc-setup where the community provides active support.
- Create a GitHub issue if you find an actual code issue, or where updates to documentation would be required.