# Load environment constants
env_tmp = RecursiveListEnv(length=5, encoding_dim=conf.encoding_dim)
num_programs = env_tmp.get_num_programs()
num_non_primary_programs = env_tmp.get_num_non_primary_programs()
observation_dim = env_tmp.get_observation_dim()
programs_library = env_tmp.programs_library
# Load Alpha-NPI policy
encoder = RecursiveListEnvEncoder(env_tmp.get_observation_dim(),
conf.encoding_dim)
indices_non_primary_programs = [
p['index'] for _, p in programs_library.items() if p['level'] > 0
]
policy = Policy(encoder, conf.hidden_size, num_programs,
num_non_primary_programs, conf.program_embedding_dim,
conf.encoding_dim, indices_non_primary_programs,
conf.learning_rate)
# Load replay buffer
idx_tasks = [
prog['index'] for key, prog in env_tmp.programs_library.items()
if prog['level'] > 0
]
buffer = PrioritizedReplayBuffer(conf.buffer_max_length,
idx_tasks,
p1=conf.proba_replay_buffer)
# Prepare mcts params
max_depth_dict = {1: 5, 2: 5, 3: 5}
mcts_train_params = {
'number_of_simulations': conf.number_of_simulations,
results_file = open(results_save_path, 'w')
# Load environment constants
env_tmp = ListEnv(length=5, encoding_dim=conf.encoding_dim)
num_programs = env_tmp.get_num_programs()
num_non_primary_programs = env_tmp.get_num_non_primary_programs()
observation_dim = env_tmp.get_observation_dim()
programs_library = env_tmp.programs_library
# Load Alpha-NPI policy
encoder = ListEnvEncoder(env_tmp.get_observation_dim(), conf.encoding_dim)
indices_non_primary_programs = [
p['index'] for _, p in programs_library.items() if p['level'] > 0
]
policy = Policy(encoder, conf.hidden_size, num_programs,
num_non_primary_programs, conf.program_embedding_dim,
conf.encoding_dim, indices_non_primary_programs,
conf.learning_rate)
policy.load_state_dict(torch.load(load_path))
# Start validation
if verbose:
print('Start validation for model: {}'.format(load_path))
if save_results:
results_file.write('Validation on model: {}'.format(load_path) + ' \n')
for len in [5, 10, 20, 60, 100]:
mcts_rewards_normalized = []
mcts_rewards = []