Giter Club home page Giter Club logo

asn's Introduction

Multiagent ASN

Installation instructions

This project based on SMAC and pymarl

OS: Ubuntu 14 or 16

Install StarCraftII:

chmod 755 ./install_sc2.sh
./install_sc2.sh
tar -xvf SMAC_Maps_supply.tar.gz
mv SMAC_Maps/* 3rdparty/StarCraftII/Maps/SMAC_Maps/

You must install anaconda before installing code env.

Install Code Env:

conda create -n ASN python=3.6
source activate ASN
conda install certifi==2018.8.24
pip install -r requirements.txt

Run an experiment

Activate Code Env:

source activate ASN
export PYTHONPATH=.

Then:

8m:

IQL

Vanilla
python src/main.py --config=iql_smac --env-config=sc2 with env_args.map_name=8m_8m agents_num=8 enemies_num=8 agent='dense_rnn' legal_action=False batch_size_run=1 use_tensorboard=True save_model=True runner_log_interval=2000 seed=100
Attention
python src/main.py --config=iql_smac --env-config=sc2 with env_args.map_name=8m_8m agents_num=8 enemies_num=8 agent='dense_rnn_attention' legal_action=False batch_size_run=1 use_tensorboard=True save_model=True runner_log_interval=2000 seed=100
Entity-Attention
python src/main.py --config=iql_smac --env-config=sc2 with env_args.map_name=8m_8m agents_num=8 enemies_num=8 agent='dense_rnn_entity_attention' legal_action=False batch_size_run=1 use_tensorboard=True save_model=True runner_log_interval=2000 seed=100
Dueling
python src/main.py --config=iql_smac --env-config=sc2 with env_args.map_name=8m_8m agents_num=8 enemies_num=8 agent='dense_rnn_dueling' legal_action=False batch_size_run=1 use_tensorboard=True save_model=True runner_log_interval=2000 seed=100
ASN
python src/main.py --config=iql_smac --env-config=sc2 with env_args.map_name=8m_8m agents_num=8 enemies_num=8 agent='asn_rnn' legal_action=False batch_size_run=1 use_tensorboard=True save_model=True runner_log_interval=2000 seed=100

VDN

Vanilla
python src/main.py --config=vdn_smac --env-config=sc2 with env_args.map_name=8m_8m agents_num=8 enemies_num=8 agent='dense_rnn' legal_action=False batch_size_run=1 use_tensorboard=True save_model=True runner_log_interval=2000 seed=100
Attention
python src/main.py --config=vdn_smac --env-config=sc2 with env_args.map_name=8m_8m agents_num=8 enemies_num=8 agent='dense_rnn_attention' legal_action=False batch_size_run=1 use_tensorboard=True save_model=True runner_log_interval=2000 seed=100
Entity-Attention
python src/main.py --config=vdn_smac --env-config=sc2 with env_args.map_name=8m_8m agents_num=8 enemies_num=8 agent='dense_rnn_entity_attention' legal_action=False batch_size_run=1 use_tensorboard=True save_model=True runner_log_interval=2000 seed=100
Dueling
python src/main.py --config=vdn_smac --env-config=sc2 with env_args.map_name=8m_8m agents_num=8 enemies_num=8 agent='dense_rnn_dueling' legal_action=False batch_size_run=1 use_tensorboard=True save_model=True runner_log_interval=2000 seed=100
ASN
python src/main.py --config=vdn_smac --env-config=sc2 with env_args.map_name=8m_8m agents_num=8 enemies_num=8 agent='asn_rnn' legal_action=False batch_size_run=1 use_tensorboard=True save_model=True runner_log_interval=2000 seed=100

QMIX

Vanilla
python src/main.py --config=qmix_smac --env-config=sc2 with env_args.map_name=8m_8m agents_num=8 enemies_num=8 legal_action=False batch_size_run=1 use_tensorboard=True save_model=True runner_log_interval=2000 env='sc2' seed=100 agent='dense_rnn'
Attention
python src/main.py --config=qmix_smac --env-config=sc2 with env_args.map_name=8m_8m agents_num=8 enemies_num=8 legal_action=False batch_size_run=1 use_tensorboard=True save_model=True runner_log_interval=2000 env='sc2' seed=100 agent='dense_rnn_attention'
Entity-Attention
python src/main.py --config=qmix_smac --env-config=sc2 with env_args.map_name=8m_8m agents_num=8 enemies_num=8 legal_action=False batch_size_run=1 use_tensorboard=True save_model=True runner_log_interval=2000 env='sc2' seed=100 agent='dense_rnn_entity_attention'
Dueling
python src/main.py --config=qmix_smac --env-config=sc2 with env_args.map_name=8m_8m agents_num=8 enemies_num=8 legal_action=False batch_size_run=1 use_tensorboard=True save_model=True runner_log_interval=2000 env='sc2' seed=100 agent='dense_rnn_dueling'
ASN
python src/main.py --config=qmix_smac --env-config=sc2 with env_args.map_name=8m_8m agents_num=8 enemies_num=8 legal_action=False batch_size_run=1 use_tensorboard=True save_model=True runner_log_interval=2000 env='sc2' seed=100 agent='asn_rnn'

-1-Paddings

QMIX

Vanilla
python src/main.py --config=qmix_smac --env-config=sc2 with env_args.map_name=8m_8m agents_num=8 enemies_num=8 legal_action=False batch_size_run=1 use_tensorboard=True save_model=True runner_log_interval=2000 seed=100 agent='dense_rnn' env='sc2_not_0'
Attention
python src/main.py --config=qmix_smac --env-config=sc2 with env_args.map_name=8m_8m agents_num=8 enemies_num=8 legal_action=False batch_size_run=1 use_tensorboard=True save_model=True runner_log_interval=2000 seed=100 agent='dense_rnn_attention' env='sc2_not_0'
Entity-Attention
python src/main.py --config=qmix_smac --env-config=sc2 with env_args.map_name=8m_8m agents_num=8 enemies_num=8 legal_action=False batch_size_run=1 use_tensorboard=True save_model=True runner_log_interval=2000 seed=100 agent='dense_rnn_entity_attention' env='sc2_not_0'
Dueling
python src/main.py --config=qmix_smac --env-config=sc2 with env_args.map_name=8m_8m agents_num=8 enemies_num=8 legal_action=False batch_size_run=1 use_tensorboard=True save_model=True runner_log_interval=2000 seed=100 agent='dense_rnn_dueling' env='sc2_not_0'
ASN
python src/main.py --config=qmix_smac --env-config=sc2 with env_args.map_name=8m_8m agents_num=8 enemies_num=8 legal_action=False batch_size_run=1 use_tensorboard=True save_model=True runner_log_interval=2000 seed=100 agent='asn_rnn' env='sc2_not_0'

1-Paddings

QMIX

Vanilla
python src/main.py --config=qmix_smac --env-config=sc2 with env_args.map_name=8m_8m agents_num=8 enemies_num=8 legal_action=False batch_size_run=1 use_tensorboard=True save_model=True runner_log_interval=2000 seed=100 agent='dense_rnn' env='sc2_set_1'
Attention
python src/main.py --config=qmix_smac --env-config=sc2 with env_args.map_name=8m_8m agents_num=8 enemies_num=8 legal_action=False batch_size_run=1 use_tensorboard=True save_model=True runner_log_interval=2000 seed=100 agent='dense_rnn_attention' env='sc2_set_1'
Entity-Attention
python src/main.py --config=qmix_smac --env-config=sc2 with env_args.map_name=8m_8m agents_num=8 enemies_num=8 legal_action=False batch_size_run=1 use_tensorboard=True save_model=True runner_log_interval=2000 seed=100 agent='dense_rnn_entity_attention' env='sc2_set_1'
Dueling
python src/main.py --config=qmix_smac --env-config=sc2 with env_args.map_name=8m_8m agents_num=8 enemies_num=8 legal_action=False batch_size_run=1 use_tensorboard=True save_model=True runner_log_interval=2000 seed=100 agent='dense_rnn_dueling' env='sc2_set_1'
ASN
python src/main.py --config=qmix_smac --env-config=sc2 with env_args.map_name=8m_8m agents_num=8 enemies_num=8 legal_action=False batch_size_run=1 use_tensorboard=True save_model=True runner_log_interval=2000 seed=100 agent='asn_rnn' env='sc2_set_1'

asn's People

Contributors

mas-anony avatar

Watchers

James Cloos avatar

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.