discount=0.99,
step_size=0.01,
use_finite_diff_hvp=True,
symmetric_finite_diff=False,
# Uncomment both lines (this and the plot parameter below) to enable plotting
# plot=True,
)
algo.train()
# if __name__ == '__main__':
# run_task()
variants = VG().variants()
for v in variants:
run_experiment_lite(
run_task,
exp_prefix='trpo_cartpole_comparison',
# Number of parallel workers for sampling
n_parallel=1,
# Only keep the snapshot parameters for the last iteration
snapshot_mode="last",
# Specifies the seed for the experiment. If this is not provided, a random seed
# will be used
seed=v['seed'],
variant=v,
# dry=True,
# plot=True,
)
# plot=True,
)
algo.train()
# if __name__ == '__main__':
# run_task()
variants = VG().variants()
for v in variants:
run_experiment_lite(
run_task,
exp_prefix="softq_multigoal",
# Number of parallel workers for sampling
n_parallel=1,
# Only keep the snapshot parameters for the last iteration
snapshot_mode="last",
# Specifies the seed for the experiment. If this is not provided, a random seed
# will be used
seed=v["seed"],
variant=v,
# plot=True,
# terminate_machine=False,
)
batch_size=50000,
max_path_length=500,
discount=0.99,
step_size=0.01,
use_finite_diff_hvp=True,
# plot=True,
)
algo.train()
variants = VG().variants()
for v in variants:
run_experiment_lite(
run_task,
exp_prefix='trpo_half_cheetah',
# Number of parallel workers for sampling
n_parallel=1,
# Only keep the snapshot parameters for the last iteration
snapshot_mode='last',
# Specifies the seed for the experiment. If this is not provided, a random seed
# will be used
seed=v['seed'],
variant=v,
mode='local',
# dry=True,
# plot=True,
# terminate_machine=False,
)