hongshaorou/Tianshou
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

minor typo

Browse Source
This commit is contained in:
Tongzheng Ren 2017-11-09 19:25:33 +08:00
parent d9674a3c8d
commit e0b59a3404
1 changed files with 1 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
AlphaGo/Network.py
View File

@ -60,7 +60,7 @@ v = value_heads(h, is_training)
p = policy_heads(h, is_training)
# loss = tf.reduce_mean(tf.square(z-v)) - tf.multiply(pi, tf.log(tf.clip_by_value(tf.nn.softmax(p), 1e-8, tf.reduce_max(tf.nn.softmax(p)))))
value_loss = tf.reduce_mean(tf.square(z - v))
policy_loss = - tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(labels=pi, logits=p))
policy_loss = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(labels=pi, logits=p))
reg = tf.add_n(tf.get_collection(tf.GraphKeys.REGULARIZATION_LOSSES))
total_loss = value_loss + policy_loss + reg