Attention

This challenge has ended!

This documentation is only for the Real Robot Challenge 2020 which has ended. Following challenges have their own documentation, see the challenge website for more information.

Install Software for Simulation Phase

Installing Anaconda or Miniconda

If not already done, install conda (Miniconda is sufficient). To do so, see the official documentation.

We tested with conda version 4.8.3.

Installing the rrc_simulation_package

1. Clone this repo and cd into it:

   git clone https://github.com/rr-learning/rrc_simulation.git
   cd rrc_simulation
   

2. Set up the conda env:

   export PYTHONNOUSERSITE=True
   conda env create -f environment.yml
   

   Note that the environment.yml contains some packages (such as stable-baselines and tensorflow) which are only required for running the example we provide. If you do not wish to install them, you can safely remove them, see comments in the environment.yml file.

3. Activate the conda env (you might have to start a new terminal):

   conda activate rrc_simulation
   

4. Install the rrc_simulation package:

   python -m pip install -e .
   

Important

Note on Activating the Conda Environment

Every time you activate your rrc_simulation conda environment, it is important that you make sure that PYTHONNOUSERSITE is set to True, as is done in step 2 above. This ensures that your conda environment is isolated from the global/user site-packages, which is important for us to reproduce your env on our side. For convenience, we provide a script which does this:

source conda_activate_rrc_simulation.sh