After downloading the code, follow the following steps to reproduce the results of Twins and Jobs datases. 

Twins Dataset:

Go to Code/Twins/Nesy and run the following command to reproduce the results of Twins dataset. One can tune the hyperparameters as given in the command below. For more hyperparameter settings see 'argparse' section in train.py.

python train.py --algorithm astar-near --exp_name twins --trial 1 --train_data ./ --test_data ./ --train_labels ./ --test_labels ./ --input_type "atom" --output_type "atom" --input_size 31 --output_size 1 --num_labels=0 --lossfxn "mseloss" --ite_beta 1  --batch_size 128 --symbolic_epochs 7 --max_depth 2 --neural_epochs 7


Jobs Dataset:

Go to Code/Jobs/Nesy and run the following command to reproduce the results of Twins dataset. One can tune the hyperparameters as given in the command below. For more hyperparameter settings see 'argparse' section in train.py.

python train.py --algorithm astar-near --exp_name jobs --trial 1 --train_data ./ --test_data ./ --train_labels ./ --test_labels ./ --input_type "atom" --output_type "atom" --input_size 18 --output_size 1 --num_labels=0 --lossfxn "mseloss" --ite_beta 1  --batch_size 64 --symbolic_epochs 10 --max_depth 5 --neural_epochs 10
