To train the causal forest model, execute the following script:
python experiments/wtt_experiment.py
and the following for training the predictive model:
python experiments/pred_experiment.py
After training the predictive model, load the results to the Jupyter notebook below and execute the cells.
conformal_prediction.ipynb
The scripts for running the policy learning experiments are the following:
python ppo_temp_cost_reward_noPred.py
python ppo_temp_cost_reward_conformal.py