This is a light simulator for autonomous driving researchers.

Download simulator for windows in baidu cloud, the extract code is `mjtr`. The linux version will be released later.

The inference server can be run within a docker container:

first docker pull floydhub/tensorflow:1.12-py3_aws.40, then start it
docker cp /src into the container
docker cp model.h5(you can find a pre-trained one in HydraMini) too.
python3 predict_server.py to start the service

Doc

Please refer to HydraMini for more Information.