def get_action(self, state, deterministic, epsilon=1e-6):
mean, log_std = self.forward(state)
normal = Normal(torch.zeros(mean.shape), torch.ones(log_std.shape))
z = normal.sample()
if self.args.stochastic_actor:
std = log_std.exp()
action_0 = mean + torch.mul(z, std)
action_1 = torch.tanh(action_0)
action = torch.mul(self.action_range, action_1) + self.action_bias
log_prob = Normal(mean, std).log_prob(action_0) - torch.log(
1. - action_1.pow(2) + epsilon) - torch.log(self.action_range)
log_prob = log_prob.sum(dim=-1, keepdim=True)
action_mean = torch.mul(self.action_range,
torch.tanh(mean)) + self.action_bias
action = action_mean.detach().cpu().numpy(
) if deterministic else action.detach().cpu().numpy()
return action, log_prob.detach().item()
else:
action_mean = torch.mul(self.action_range,
torch.tanh(mean)) + self.action_bias
action = action_mean + 0.1 * torch.mul(self.action_range, z)
action = torch.min(action, self.action_high)
action = torch.max(action, self.action_low)
action = action_mean.detach().cpu().numpy(
) if deterministic else action.detach().cpu().numpy()
return action, 0
def forward(self, ss: List, phase_use_mode: bool = False) -> Tuple:
p_pres_logits, p_where_mean, p_where_std, p_depth_mean, \
p_depth_std, p_what_mean, p_what_std = ss
if phase_use_mode:
z_pres = (p_pres_logits > 0).float()
else:
z_pres = RelaxedBernoulli(logits=p_pres_logits, temperature=self.args.train.tau_pres).rsample()
# z_where_scale, z_where_shift: (bs, dim, num_cell, num_cell)
if phase_use_mode:
z_where_scale, z_where_shift = p_where_mean.chunk(2, 1)
else:
z_where_scale, z_where_shift = \
Normal(p_where_mean, p_where_std).rsample().chunk(2, 1)
# z_where_origin: (bs, dim, num_cell, num_cell)
z_where_origin = \
torch.cat([z_where_scale.detach(), z_where_shift.detach()], dim=1)
z_where_shift = \
(2. / self.args.arch.num_cell) * \
(self.offset + 0.5 + torch.tanh(z_where_shift)) - 1.
scale, ratio = z_where_scale.chunk(2, 1)
scale = scale.sigmoid()
ratio = torch.exp(ratio)
ratio_sqrt = ratio.sqrt()
z_where_scale = torch.cat([scale / ratio_sqrt, scale * ratio_sqrt], dim=1)
# z_where: (bs, dim, num_cell, num_cell)
z_where = torch.cat([z_where_scale, z_where_shift], dim=1)
if phase_use_mode:
z_depth = p_depth_mean
z_what = p_what_mean
else:
z_depth = Normal(p_depth_mean, p_depth_std).rsample()
z_what = Normal(p_what_mean, p_what_std).rsample()
z_what_reshape = z_what.permute(0, 2, 3, 1).reshape(-1, self.args.z.z_what_dim). \
view(-1, self.args.z.z_what_dim, 1, 1)
if self.args.data.inp_channel == 1 or not self.args.arch.phase_overlap:
o = self.z_what_decoder_net(z_what_reshape)
o = o.sigmoid()
a = o.new_ones(o.size())
elif self.args.arch.phase_overlap:
o, a = self.z_what_decoder_net(z_what_reshape).split([self.args.data.inp_channel, 1], dim=1)
o, a = o.sigmoid(), a.sigmoid()
else:
raise NotImplemented
lv = [z_pres, z_where, z_depth, z_what, z_where_origin]
pa = [o, a]
return pa, lv
def optimize_model(policy, q1_net, q2_net, v_net, v_target_net, memory,
actor_optimizer, q_net_optimizer, v_net_optimizer):
if len(memory) < train_batch_size:
return 0, 0, 0 # dummy losses for consistency in presenting results
st_b, ac_b, rew_b, nst_b, dn_b = memory.sample(train_batch_size)
states_th = torch.tensor(st_b).float().to(device)
actions_th = torch.tensor(ac_b).to(device)
rewards_th = torch.tensor(rew_b).unsqueeze(1).to(device)
next_states_th = torch.tensor(nst_b).float().to(device)
dones_th = torch.tensor(dn_b).float().unsqueeze(1).to(device)
Q1_vals = q1_net(states_th, actions_th)
Q2_vals = q2_net(states_th, actions_th)
V_vals = v_net(states_th)
V_next_state_vals = v_target_net(next_states_th)
pi_action_means, pi_action_logstd = policy(states_th)
pi_action_stds = torch.exp(pi_action_logstd)
z = Normal(torch.zeros_like(pi_action_means),
torch.ones_like(pi_action_stds)).sample()
newly_sampled_actions = pi_action_means + z * pi_action_stds
newly_sampled_action_log_probs = Normal(
pi_action_means, pi_action_stds).log_prob(newly_sampled_actions)
newly_sampled_Q1_vals = q1_net(states_th, newly_sampled_actions)
newly_sampled_Q2_vals = q2_net(states_th, newly_sampled_actions)
newly_sampled_Q_minvals = torch.min(newly_sampled_Q1_vals,
newly_sampled_Q2_vals)
J_v = torch.mean(
(V_vals -
(newly_sampled_Q_minvals.detach() -
entropy_coeff * newly_sampled_action_log_probs.detach()))**2)
v_net_optimizer.zero_grad()
J_v.backward()
v_net_optimizer.step()
J_q1 = torch.mean((Q1_vals - (rewards_th + gamma * V_next_state_vals *
(1 - dones_th)))**2)
J_q2 = torch.mean((Q2_vals - (rewards_th + gamma * V_next_state_vals *
(1 - dones_th)))**2)
J_q = J_q1 + J_q2
q_net_optimizer.zero_grad()
J_q.backward()
q_net_optimizer.step()
J_pi = torch.mean(entropy_coeff * newly_sampled_action_log_probs -
newly_sampled_Q_minvals)
actor_optimizer.zero_grad()
J_pi.backward()
actor_optimizer.step()
return J_v, J_q, J_pi
(epoch, i + 1, running_loss / 20))
running_loss = 0
optimizer.step()
print('Done!')
### EVALUATE POLICY
max_steps = env.spec.timestep_limit
pi.cpu()
returns = []
for i in range(10):
print('iter', i)
obs = env.reset()
done = False
totalr = 0.
steps = 0
while not done:
a_mu, a_sigma = pi(torch.from_numpy(obs).float())
a = Normal(loc=a_mu, scale=a_sigma).sample()
obs, r, done, _ = env.step(a.detach().numpy())
if RENDER:
env.render()
totalr += r
steps += 1
if steps % 100 == 0: print("%i/%i" % (steps, max_steps))
if steps >= max_steps:
break
returns.append(totalr)
print('returns', returns)
print('mean return', np.mean(returns))
print('std of return', np.std(returns))