def __init__(self, num_inputs, use_gru):
super(CNNBase, self).__init__()
init_ = lambda m: init(m, nn.init.orthogonal_, lambda x: nn.init.
constant_(x, 0), nn.init.calculate_gain('relu'))
self.main = nn.Sequential(
init_(nn.Conv2d(num_inputs, 32, 8, stride=4)), nn.ReLU(),
init_(nn.Conv2d(32, 64, 4, stride=2)), nn.ReLU(),
init_(nn.Conv2d(64, 32, 3, stride=1)), nn.ReLU(), Flatten(),
init_(nn.Linear(32 * 7 * 7, 512)), nn.ReLU())
if use_gru:
self.gru = nn.GRUCell(512, 512)
nn.init.orthogonal_(self.gru.weight_ih.data)
nn.init.orthogonal_(self.gru.weight_hh.data)
self.gru.bias_ih.data.fill_(0)
self.gru.bias_hh.data.fill_(0)
init_ = lambda m: init(m, nn.init.orthogonal_, lambda x: nn.init.
constant_(x, 0))
self.critic_linear = init_(nn.Linear(512, 1))
self.train()
def __init__(self, num_inputs, num_outputs):
super(DiagGaussian, self).__init__()
init_ = lambda m: init(m,
init_normc_,
lambda x: nn.init.constant_(x, 0))
self.fc_mean = init_(nn.Linear(num_inputs, num_outputs))
self.logstd = AddBias(torch.zeros(num_outputs))
def __init__(self, num_inputs, num_outputs):
super(Categorical, self).__init__()
init_ = lambda m: init(m,
nn.init.orthogonal_,
lambda x: nn.init.constant_(x, 0),
gain=0.01)
self.linear = init_(nn.Linear(num_inputs, num_outputs))
def __init__(self, num_inputs):
super(MLPBase, self).__init__()
init_ = lambda m: init(m, init_normc_, lambda x: nn.init.constant_(
x, 0))
self.actor = nn.Sequential(init_(nn.Linear(num_inputs, 64)), nn.Tanh(),
init_(nn.Linear(64, 64)), nn.Tanh())
self.critic = nn.Sequential(init_(nn.Linear(num_inputs, 64)),
nn.Tanh(), init_(nn.Linear(64, 64)),
nn.Tanh())
self.critic_linear = init_(nn.Linear(64, 1))
self.train()