gym-socioTEQ

OpenAI Gym Style SocioTEQ Environment.

Each observation includes four dimensions that represent:

• The percentage of positive interaction within the member of all the social sub-groups [range 0-100]
  The overall members interaction is the sum of each participant positive interactions within the Action sub-group, multiplied by its reverse popularity
• The Action Evaluation rate on behalf of all the participants [range 0-100]
• The Action Evaluation rate on behalf of the Group Facilitator [range 0-100]
• The Facilitaror evaluation on behalf of the Group Members [range 0-100]

observation [7, 15, 86, 28]
action 3
new observation [7.7, 16.5, 94.6, 30.8]
reward 37.4
done 0

Requirement

• Python 3.5+
• OpenAI Gym
• NumPy

Install

git clone https://github.com/efotopoulou/gym-socioTEQ.git
cd gym-socioTEQ/
pip install -e .

Try example agents

cd examples/
python random_agent_SocioTEQEnv.py