Various Bandits in the Context of Warfarin Dosing

Setup

1. Create a virtualenv or conda environment
2. Run pip install -r requirements.txt
3. Prepare the dataset by running python clean_dataset.py to generate the version of the data that our models will run.

Run agents

To run an agent use python main.py --agent=AGENT_NAME to generate a set of 20 experiments. Examples of agents are fixed, linear, ucb, linucb, supervised-lin, supervised-ridge, thompson-0, ensemble-0

Generate plots

Run python plot.py --agent=AGENT_NAME1,AGENT_NAME2 to visualize that agent's performance against the other agents provided. This automatically generates the named plots in the plots/ directory.