gryf/coach
1
0
Fork 0
mirror of https://github.com/gryf/coach.git synced 2025-12-18 11:40:18 +01:00
Code Issues Projects Releases Wiki Activity

Batch RL (#238)

Browse Source
This commit is contained in:
Gal Leibovich
2019-03-19 18:07:09 +02:00
committed by GitHub
parent 4a8451ff02
commit e3c7e526c7
38 changed files with 1003 additions and 87 deletions
Download Patch File Download Diff File Expand all files Collapse all files

5
rl_coach/architectures/tensorflow_components/heads/q_head.py
View File

@@ -50,6 +50,11 @@ class QHead(Head):
# Standard Q Network
self.output = self.dense_layer(self.num_actions)(input_layer, name='output')
# TODO add this to other Q heads. e.g. dueling.
temperature = self.ap.network_wrappers[self.network_name].softmax_temperature
temperature_scaled_outputs = self.output / temperature
self.softmax = tf.nn.softmax(temperature_scaled_outputs, name="softmax")
def __str__(self):
result = [
"Dense (num outputs = {})".format(self.num_actions)

2
rl_coach/architectures/tensorflow_components/savers.py
View File

@@ -42,7 +42,7 @@ class GlobalVariableSaver(Saver):
self._variable_placeholders.append(variable_placeholder)
self._variable_update_ops.append(v.assign(variable_placeholder))
self._saver = tf.train.Saver(self._variables)
self._saver = tf.train.Saver(self._variables, max_to_keep=None)
@property
def path(self):