Code Monkey home page Code Monkey logo

metarl-offloading's Introduction

System environment

Unbuntu 16.04

Preinstall Package

sudo apt-get update && sudo apt-get install cmake libopenmpi-dev python3-dev zlib1g-dev

Create Conda Environment

conda create --name tf-1.15 anaconda python=3.6
conda activate tf-1.15

Install Tensorflow-1.15 GPU or CPU

pip install tensorflow-gpu==1.15

or

pip install tensorflow==1.15

Install Third-party Python Pakage

pip install gym
pip install graphviz
pip install pydotplus
pip install pyprind
pip install mpi4py

Start Meta Training:

python meta_trainer.py

All the hyperparameters are defined in meta_trainer.py including the log file and save path of the trained model.

Start Meta Evaluation:

After training, you will get the meta model. In order to fast adapt the meta model for new learning tasks in MEC, we need to conduct fine-tuning steps for the trained meta moodel.

python meta_evaluator.py

The training might take long time because of the large training set. All the training results and evaluation results can be found in the log file.

Related paper: Fast Adaptive Task Offloading in Edge Computing based on Meta Reinforcement Learning

If you like this research, please cite this paper:

@article{wang2020fast,
  title={Fast Adaptive Task Offloading in Edge Computing Based on Meta Reinforcement Learning},
  author={Wang, Jin and Hu, Jia and Min, Geyong and Zomaya, Albert Y and Georgalas, Nektarios},
  journal={IEEE Transactions on Parallel and Distributed Systems},
  volume={32},
  number={1},
  pages={242--253},
  year={2020},
  publisher={IEEE}
}

metarl-offloading's People

Stargazers

 avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar

Watchers

 avatar  avatar  avatar  avatar  avatar  avatar

metarl-offloading's Issues

InvalidSymbolNameError

tensorflow.python.util.tf_export.InvalidSymbolNameError: Can only export symbols under package name of component. e.g. tensorflow_estimator must export all symbols under tf.estimator
您好,运行代码是遇到这个问题,该怎么处理呢

DAG生成问题

你好 看见/env 中有DAG图的数据 但是想知道DAG图的生成方式 是否可以开源DAG的生成器 非常感谢

A question about MDPs in this paper

Hello, I would like to ask a question. The difference between different MDPs in meta-reinforcement learning in the previous papers is that the state transition probability(P) and reward function(R) are different, but in this article, according to the definition of the state, it is obvious that the states(S) of different MDPs are different, here I do not understand. Hope to answer that, thank you.

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.