AWS DeepRacer Robomaker Image
This repository contains the extracts from the AWS DeepRacer Robomaker simapp; including
- the environment and tracks
- the robotics agent code ('markov')
- Dockerfile
Available pre-built versions
In most cases it will be sufficient to use one of our pre-built images which use optimized Tensorflow binaries. Depending on how modern processor you have available you can pick which instruction sets to use.
- AVX is available on most Intel Core processors from the second generation on - running DeepRacer on an older processor really does not make sense; hence we do not have pre-built images without this instruction set.
- AVX2 is available on more modern Intel Core processors; starting from the Haswell architecture (4th generation), and is probably the image most will use.
- AVX-512 was available on newer Xeon and Core i7-X processors. Due to convergence issues it has been removed.
- GPU uses Tensorflow-GPU instead of CPU. It is not compiled with special instruction sets for the CPU.
Available tags are:
Version 5
Version 4 is the current development version, and it is receiving both new tracks as well as functionality updates as they are released by AWS.
| Version | Comment | AVX | AVX2 | GPU |
|---|---|---|---|---|
| 5.0.0 | Initial release | 5.0.0-cpu-avx 5.0.0-cpu-gl-avx |
5.0.0-cpu-avx2 5.0.0-cpu-gl-avx2 |
5.0.0-gpu 5.0.0-gpu-gl |
| 5.0.1 | April Track | 5.0.1-cpu-avx 5.0.1-cpu-gl-avx |
5.0.1-cpu-avx2 5.0.1-cpu-gl-avx2 |
5.0.1-gpu 5.0.1-gpu-gl |
| 5.0.2 | May Track | 5.0.2-cpu-avx 5.0.2-cpu-gl-avx |
5.0.2-cpu-avx2 5.0.2-cpu-gl-avx2 |
5.0.2-gpu 5.0.2-gpu-gl |
Version 4
Version 4 is retired.
| Version | Comment | AVX | AVX2 | GPU |
|---|---|---|---|---|
| 4.0.2 | Melodic Upgrade | 4.0.2-cpu-avx 4.0.2-cpu-gl-avx |
4.0.2-cpu-avx2 4.0.2-cpu-gl-avx2 |
4.0.2-gpu 4.0.2-gpu-gl |
| 4.0.3 | May Track | 4.0.3-cpu-avx 4.0.3-cpu-gl-avx |
4.0.3-cpu-avx2 4.0.3-cpu-gl-avx2 |
4.0.3-gpu 4.0.3-gpu-gl |
| 4.0.4 | May Updates | 4.0.4-cpu-avx 4.0.4-cpu-gl-avx |
4.0.4-cpu-avx2 4.0.4-cpu-gl-avx2 |
4.0.4-gpu 4.0.4-gpu-gl |
| 4.0.5 | June Track | 4.0.5-cpu-avx 4.0.5-cpu-gl-avx |
4.0.5-cpu-avx2 4.0.5-cpu-gl-avx2 |
4.0.5-gpu 4.0.5-gpu-gl |
| 4.0.6 | July Track & Updates | 4.0.6-cpu-avx 4.0.6-cpu-gl-avx |
4.0.6-cpu-avx2 4.0.6-cpu-gl-avx2 |
4.0.6-gpu 4.0.6-gpu-gl |
| 4.0.7 | August Track | 4.0.7-cpu-avx 4.0.7-cpu-gl-avx |
4.0.7-cpu-avx2 4.0.7-cpu-gl-avx2 |
4.0.7-gpu 4.0.7-gpu-gl |
| 4.0.8 | September Track | 4.0.8-cpu-avx 4.0.8-cpu-gl-avx |
4.0.8-cpu-avx2 4.0.8-cpu-gl-avx2 |
4.0.8-gpu 4.0.8-gpu-gl |
| 4.0.9 | October Track | 4.0.9-cpu-avx 4.0.9-cpu-gl-avx |
4.0.9-cpu-avx2 4.0.9-cpu-gl-avx2 |
4.0.9-gpu 4.0.9-gpu-gl |
| 4.0.10 | ReInvent Track | 4.0.10-cpu-avx 4.0.10-cpu-gl-avx |
4.0.10-cpu-avx2 4.0.10-cpu-gl-avx2 |
4.0.10-gpu 4.0.10-gpu-gl |
| 4.0.11 | Minor fixes | 4.0.11-cpu-avx 4.0.11-cpu-gl-avx |
4.0.11-cpu-avx2 4.0.11-cpu-gl-avx |
4.0.11-gpu 4.0.11-gpu-gl |
| 4.0.12 | 2022 Summit Track | 4.0.12-cpu-avx 4.0.12-cpu-gl-avx |
4.0.12-cpu-avx2 4.0.12-cpu-gl-avx |
4.0.12-gpu 4.0.12-gpu-gl |
| 4.0.13 | 2022 March Track | 4.0.13-cpu-avx 4.0.13-cpu-gl-avx |
4.0.13-cpu-avx2 4.0.13-cpu-gl-avx |
4.0.13-gpu 4.0.13-gpu-gl |
The built image is available via docker pull awsdeepracercommunity/deepracer-robomaker:cpu. For other images replace cpu with the tag of your choice from the above table.
Building the image
To build the CPU image run docker build -t awsdeepracercommunity/deepracer-robomaker:cpu -f docker/Dockerfile.cpu . --build-arg TENSORFLOW_VER='tensorflow==1.11.0
To build the GPU image run docker build -t awsdeepracercommunity/deepracer-robomaker:gpu -f docker/Dockerfile.gpu .
To add a custom tensorflow library run docker build -t awsdeepracercommunity/deepracer-robomaker:cpu -f docker/Dockerfile.cpu . --build-arg TENSORFLOW_VER=<http-url-to-file>
Tensorflow build
Tensorflow has been built with the following parameters:
cpu-avxwith-march=x86-64 -maes -mavx -mmmx -mpclmul -mpopcnt -msse -msse2 -msse3 -msse4.1 -msse4.2 -mssse3cpu-avx2with-march=x86-64 -maes -mavx -mavx2 -mmmx -mpclmul -mpopcnt -msse -msse2 -msse3 -msse4.1 -msse4.2 -mssse3 -mf16c -mfma