# Please add a `--gen-data` flag the first time using any particular environment


#ENV：Push

# SAC
python scripts/train.py --algo sac --env push --n-demos 500 --num-eval-episodes 5 --alpha 0.1

# SAC + SIG
python scripts/train-sig.py --algo sac --env push --n-demos 500 --num-eval-episodes 5 --alpha 0.1

# SM
python scripts/train.py --algo sac --env push --n-demos 500 --num-eval-episodes 5 --alpha 0.1 --do-mcac-bonus

# SM + SIG
python scripts/train-sig.py --algo sac --env push --n-demos 500 --num-eval-episodes 5 --alpha 0.1 --do-mcac-bonus

# SM (Less Interact)
python scripts/train-nosig.py --algo sac --env push --n-demos 500 --num-eval-episodes 5 --alpha 0.1 --do-mcac-bonus

# SM + SIG (Fixed Switch)
python scripts/train-sig-fixed.py --algo sac --env push --n-demos 500 --num-eval-episodes 5 --alpha 0.1 --do-mcac-bonus

# SM + SIG (w/o AVM)
python scripts/train-sig-noavm.py --algo sac --env push --n-demos 500 --num-eval-episodes 5 --alpha 0.1 --do-mcac-bonus

# TD3
python scripts/train.py --algo td3 --env push --n-demos 500 --num-eval-episodes 5

# TD3 + SIG
python scripts/train-sig.py --algo td3 --env push --n-demos 500 --num-eval-episodes 5

# TM
python scripts/train.py --algo td3 --env push --n-demos 500 --num-eval-episodes 5 --do-mcac-bonus

# TM + SIG
python scripts/train-sig.py --algo td3 --env push --n-demos 500 --num-eval-episodes 5 --do-mcac-bonus

# TM (Less Interact)
python scripts/train-nosig.py --algo td3 --env push --n-demos 500 --num-eval-episodes 5 --do-mcac-bonus

# TM + SIG (Fixed Switch)
python scripts/train-sig-fixed.py --algo td3 --env push --n-demos 500 --num-eval-episodes 5 --do-mcac-bonus

# TM + SIG (w/o AVM)
python scripts/train-sig-noavm.py --algo td3 --env push --n-demos 500 --num-eval-episodes 5 --do-mcac-bonus

# GQE
python scripts/train.py --algo gqe --env push --n-demos 500 --alpha 0.1 --num-eval-episodes 5 --total-timesteps 250000 --gqe --gqe-n 16 --gqe-lambda 0.95

# GQE + SIG
python scripts/train-sig.py --algo gqe --env push --n-demos 500 --alpha 0.1 --num-eval-episodes 5 --total-timesteps 250000 --gqe --gqe-n 16 --gqe-lambda 0.95

# GM
python scripts/train.py --algo gqe --env push --n-demos 500 --alpha 0.1 --num-eval-episodes 5 --total-timesteps 250000 --do-mcac-bonus --gqe --gqe-n 16 --gqe-lambda 0.95

# GM + SIG
python scripts/train-sig.py --algo gqe --env push --n-demos 500 --alpha 0.1 --num-eval-episodes 5 --total-timesteps 250000 --do-mcac-bonus --gqe --gqe-n 16 --gqe-lambda 0.95

# GM (Less Interact)
python scripts/train-nosig.py --algo gqe --env push --n-demos 500 --alpha 0.1 --num-eval-episodes 5 --total-timesteps 250000 --do-mcac-bonus --gqe --gqe-n 16 --gqe-lambda 0.95

# GM + SIG (Fixed Switch)
python scripts/train-sig-fixed.py --algo gqe --env push --n-demos 500 --alpha 0.1 --num-eval-episodes 5 --total-timesteps 250000 --do-mcac-bonus --gqe --gqe-n 16 --gqe-lambda 0.95

# GM + SIG (w/o AVM)
python scripts/train-sig-noavm.py --algo gqe --env push --n-demos 500 --alpha 0.1 --num-eval-episodes 5 --total-timesteps 250000 --do-mcac-bonus --gqe --gqe-n 16 --gqe-lambda 0.95

# OEFD
python scripts/train.py --algo td3 --env push --n-demos 500 --num-eval-episodes 5 --do-bc-loss --do-q-filter

# OEFD + SIG
python scripts/train-sig.py --algo td3 --env push --n-demos 500 --num-eval-episodes 5 --do-bc-loss --do-q-filter

# OM
python scripts/train.py --algo td3 --env push --n-demos 500 --num-eval-episodes 5 --do-bc-loss --do-q-filter --do-mcac-bonus

# OM + SIG
python scripts/train-sig.py --algo td3 --env push --n-demos 500 --num-eval-episodes 5 --do-bc-loss --do-q-filter --do-mcac-bonus

# OM (Less Interact)
python scripts/train-nosig.py --algo td3 --env push --n-demos 500 --num-eval-episodes 5 --do-bc-loss --do-q-filter --do-mcac-bonus

# OM + SIG (Fixed Switch)
python scripts/train-sig-fixed.py --algo td3 --env push --n-demos 500 --num-eval-episodes 5 --do-bc-loss --do-q-filter --do-mcac-bonus

# OM + SIG (w/o AVM)
python scripts/train-sig-noavm.py --algo td3 --env push --n-demos 500 --num-eval-episodes 5 --do-bc-loss --do-q-filter --do-mcac-bonus

# CQL
python scripts/train.py --algo cql --env push --n-demos 500 --init-iters 10000 --num-eval-episodes 5 

# CQL + SIG
python scripts/train-sig.py --algo cql --env push --n-demos 500 --init-iters 10000 --num-eval-episodes 5 

# CM
python scripts/train.py --algo cql --env push --n-demos 500 --init-iters 10000 --num-eval-episodes 5 --do-mcac-bonus

# CM + SIG
python scripts/train-sig.py --algo cql --env push --n-demos 500 --init-iters 10000 --num-eval-episodes 5 --do-mcac-bonus

# CM (Less Interact)
python scripts/train-nosig.py --algo cql --env push --n-demos 500 --init-iters 10000 --num-eval-episodes 5 --do-mcac-bonus

# CM + SIG (Fixed Switch) 
python scripts/train-sig-fixed.py --algo cql --env push --n-demos 500 --init-iters 10000 --num-eval-episodes 5 --do-mcac-bonus

# CM + SIG (w/o AVM)
python scripts/train-sig-noavm.py --algo cql --env push --n-demos 500 --init-iters 10000 --num-eval-episodes 5 --do-mcac-bonus


