Shortcuts

Working with habitat-lab

Setting up your environment for habitat and torchrl

Installing habitat-lab from pip

Instructions can be found on habitat github repo.

  1. Preparing conda env

    Assuming you have conda installed, let’s prepare a conda env:

    conda create -n habitat python=3.7 cmake=3.14.0
    conda activate habitat
    
  2. conda install habitat-sim To install habitat-sim with bullet physics and in headless mode (usually necessary to run habitat on a cluster)

    conda install habitat-sim withbullet headless -c conda-forge -c aihabitat-nightly -y
    pip install git+https://github.com/facebookresearch/habitat-lab.git#subdirectory=habitat-lab
    
    # This is to reduce verbosity
    export MAGNUM_LOG=quiet && export HABITAT_SIM_LOG=quiet
    

    If you don’t want to install it in headless mode, simply remove the headless package from the conda install command.

    See Habitat-Sim’s installation instructions for more details.

Installing TorchRL

Follow the instructions on the README.md.

Using Habitat

To get the list of available Habitat envs, simply run the following command:

from torchrl.envs.libs.habitat import HabitatEnv, _has_habitat
assert _has_habitat  # checks that habitat is installed
print([_env for _env in HabitatEnv.available_envs if _env.startswith("Habitat")])

Common issues

  1. OSError: Could not find/load shared object file: libllvmlite.so Solution: IOnstall the missing library and add a pointer to the file (check that the python version matches before executing the command and replace the path and env_name with yours):

conda install -c conda-forge llvmlite
conda env config vars set LD_PRELOAD=$LD_PRELOAD:/path/to/conda/envs/env_name/lib/python3.8/site-packages/llvmlite/binding/libllvmlite.so
conda deactivate && conda activate env_name
  1. ImportError: /usr/lib/x86_64-linux-gnu/libOpenGL.so.0: undefined symbol: _glapi_tls_Current Solution: as in MUJOCO debug, Link conda to the right libOpenGL.so file (replace /path/to/conda and mujoco_env with the proper paths and names):

conda install -y -c conda-forge libglvnd-glx-cos7-x86_64 --force-reinstall
conda install -y -c conda-forge xvfbwrapper --force-reinstall
conda env config vars set LD_PRELOAD=$LD_PRELOAD:/path/to/conda/envs/env_name/x86_64-conda-linux-gnu/sysroot/usr/lib64/libGLdispatch.so.0
conda deactivate && conda activate env_name

Docs

Access comprehensive developer documentation for PyTorch

View Docs

Tutorials

Get in-depth tutorials for beginners and advanced developers

View Tutorials

Resources

Find development resources and get your questions answered

View Resources