gaspersic/baselines
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Merge pull request #268 from ei-grad/master

Browse Source 
Fix fc call in AcerLstmPolicy
This commit is contained in:
John Schulman
2018-01-27 18:42:31 -08:00
committed by GitHub
parent ebb8afff2e 634e37c5b8
commit 2e29b41592
1 changed files with 2 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
baselines/acer/policies.py
View File

@@ -59,9 +59,9 @@ class AcerLstmPolicy(object):
h5, snew = lstm(xs, ms, S, 'lstm1', nh=nlstm) h5, snew = lstm(xs, ms, S, 'lstm1', nh=nlstm)
h5 = seq_to_batch(h5) h5 = seq_to_batch(h5)
pi_logits = fc(h5, 'pi', nact, act=lambda x: x, init_scale=0.01) pi_logits = fc(h5, 'pi', nact, init_scale=0.01)
pi = tf.nn.softmax(pi_logits) pi = tf.nn.softmax(pi_logits)
q = fc(h5, 'q', nact, act=lambda x: x) q = fc(h5, 'q', nact)
a = sample(pi_logits) # could change this to use self.pi instead a = sample(pi_logits) # could change this to use self.pi instead
self.initial_state = np.zeros((nenv, nlstm*2), dtype=np.float32) self.initial_state = np.zeros((nenv, nlstm*2), dtype=np.float32)