andre/PyTorch-Tutorial
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

update for torch 0.2

Browse Source
This commit is contained in:
Morvan Zhou
2017-09-05 12:57:25 +10:00
committed by Morvan Zhou
parent f31c88b81e
commit 3b23f6bced
1 changed files with 2 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
tutorial-contents/405_DQN_Reinforcement_learning.py
View File

@ -58,7 +58,7 @@ class DQN(object):
# input only one sample
if np.random.uniform() < EPSILON: # greedy
actions_value = self.eval_net.forward(x)
action = torch.max(actions_value, 1)[1].data.numpy()[0, 0] # return the argmax
action = torch.max(actions_value, 1)[1].data.numpy()[0] # return the argmax
else: # random
action = np.random.randint(0, N_ACTIONS)
return action
@ -87,7 +87,7 @@ class DQN(object):
# q_eval w.r.t the action in experience
q_eval = self.eval_net(b_s).gather(1, b_a) # shape (batch, 1)
q_next = self.target_net(b_s_).detach() # detach from graph, don't backpropagate
q_target = b_r + GAMMA * q_next.max(1)[0] # shape (batch, 1)
q_target = b_r + GAMMA * q_next.max(1)[0].view(BATCH_SIZE, 1) # shape (batch, 1)
loss = self.loss_func(q_eval, q_target)
self.optimizer.zero_grad()