![]() The container automatically starts TensorBoard and the HTTP server, so this does not need to be done manually. This starts the Docker container in the background, publishing port 6006 for TensorBoard and port 8000 for the You can use it as follows: docker run -d -p 6006:6006 -p 8000:8000 -rm -shm-size=3gb -name deepcomp deepcomp If you want to use the deepcomp Docker container and pulled the corresponding image from Docker Hub, To save them, copy them from the Docker container or use a Docker volume. Note: By default, results within the Docker container are not stored persistently. You can choose a custom location with -result-dir. If you installed via PyPi, this is in your virtualenv's site packages. If you cloned the repo from GitHub, this is where the Readme is. To run DeepCoMP, use -alg ppo -agent central.įor DD-CoMP, use -alg ppo -agent multi, and for D3-CoMP, use -alg ppo -agent multi -separate-agent-nns.īy default, training logs, results, videos, and trained agents are saved in /results, Usage # get an overview of all optionsįor example: deepcomp -env medium -slow-ues 3 -agent central -workers 2 -train-steps 50000 -seed 42 -video both Use the -no-cache option is to force a rebuild of the image, pulling the latest deepcomp version from PyPI. alternatively, write out "stefanbschneider/deepcomp" in all following commands.ĭocker tag stefanbschneider/deepcomp:latest deepcompĪlternatively, to build the Docker image manually from the Dockerfile, clone this repository and run docker build -t deepcomp. To use the Docker image, simply pull the latest version from Docker Hub: docker pull stefanbschneider/deepcomp There is a Docker image that comes with deepcomp preinstalled. On Ubuntu: sudo apt install ffmpeg imagemagick Tested on Ubuntu 20.04 and Windows 10 with Python 3.8.įor saving videos and gifs, you also need to install ffmpeg (not on Windows) and ImageMagick. # for development install (when changing code): pip install -e. # "python setup.py install" does not work for some reason: ![]() Then install everything, following these steps: # only on ubuntu Sudo apt install cmake build-essential zlib1g-dev python3-devįor adjusting or further developing DeepCoMP, it's better to install manually rather than from PyPi.Ĭlone the repository. Simple Installation via PyPi sudo apt update You can install deepcomp either directly from PyPi or manually after cloning this repository. Visualized cell selection policy of DeepCoMP after 2M training steps.īase station icon by Clea Doltz from the Noun Project Setup Three variants: DeepCoMP (central agent), DD-CoMP (distributed agents using central policy), D3-CoMP (distributed agents with separate policies).Īll three approaches self-learn and adapt to various scenarios in mobile networks without expert knowledge, human intervention, or detailed assumptions about the underlying system.Ĭompared to other approaches, they are more flexible and achieve higher Quality of Experience. DeepCoMP: Self-Learning Dynamic Multi-Cell Selection for Coordinated Multipoint (CoMP)ĭeep reinforcement learning for dynamic multi-cell selection in CoMP scenarios.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |