Project Setup

• Clone the project
• Go to root directory using comman below cd rl_project
• Create a virualenv using command: virtalenv venv
• Activate virtualenv using command: source venv/bin/activate
• Install project dependencies using command: pip install -r requirements.txt

To train the models

In this project, we have used DQN and DDQN agents to train our recommendation system

• To train DQN use command below: python src/train.py DQN
• To train DDQN use command below: python src/train.py DDQN

To evaluate the models

In this project, we have used DQN and DDQN agents to train our recommendation system

• To evaluate DQN use command below: python src/evaluate.py DQN
• To evaluate DDQN use command below: python src/evaluate.py DDQN

To generate recommendation for a single user

• To generate recommendation using DQN use command below: python src/recommend.py DQN [user_id]
• To generate recommendation using DDQN use command below: python src/recommend.py DDQN [user_id]