parser.add_argument('--learning-rate', type=float, default=0.0001)
parser.add_argument('--gpu-num', type=int, default=0)
args = parser.parse_args()
build_directory_structure(
'.', {
args.run_dir: {
args.name: {
'autoencoder': {},
'correlation': {},
'policies': {}
}
}
})
LOG.setup(os.path.join('.', args.run_dir, args.name))
env = SimpleGridworld()
dummy_env = SimpleGridworld()
net = IndepFeatureLearner(lmbda=args.lmbda,
learning_rate=args.learning_rate,
gpu_num=args.gpu_num)
buffer = replay_buffer.ReplayBuffer(10000)
visualization_freq = 10000
batch_size = args.batch_size
def run_training_step(buffer: replay_buffer.ReplayBuffer,
net: IndepFeatureLearner):
positions, _, _, _, _ = buffer.sample(batch_size)
