gryf/coach
1
0
Fork 0
mirror of https://github.com/gryf/coach.git synced 2026-02-28 05:15:46 +01:00
Code Issues Projects Releases Wiki Activity

SAC algorithm (#282)

Browse Source 
* SAC algorithm

* SAC - updates to agent (learn_from_batch), sac_head and sac_q_head to fix problem in gradient calculation. Now SAC agents is able to train.
gym_environment - fixing an error in access to gym.spaces

* Soft Actor Critic - code cleanup

* code cleanup

* V-head initialization fix

* SAC benchmarks

* SAC Documentation

* typo fix

* documentation fixes

* documentation and version update

* README typo
This commit is contained in:
guyk1971
2019-05-01 18:37:49 +03:00
committed by shadiendrawis
parent 33dc29ee99
commit 74db141d5e
92 changed files with 2812 additions and 402 deletions
Download Patch File Download Diff File Expand all files Collapse all files

11
rl_coach/architectures/tensorflow_components/heads/v_head.py
View File

@@ -26,7 +26,7 @@ from rl_coach.spaces import SpacesDefinition
class VHead(Head):
def __init__(self, agent_parameters: AgentParameters, spaces: SpacesDefinition, network_name: str,
head_idx: int = 0, loss_weight: float = 1., is_local: bool = True, activation_function: str='relu',
dense_layer=Dense):
dense_layer=Dense, initializer='normalized_columns'):
super().__init__(agent_parameters, spaces, network_name, head_idx, loss_weight, is_local, activation_function,
dense_layer=dense_layer)
self.name = 'v_values_head'
@@ -37,10 +37,15 @@ class VHead(Head):
else:
self.loss_type = tf.losses.mean_squared_error
self.initializer = initializer
def _build_module(self, input_layer):
# Standard V Network
self.output = self.dense_layer(1)(input_layer, name='output',
kernel_initializer=normalized_columns_initializer(1.0))
if self.initializer == 'normalized_columns':
self.output = self.dense_layer(1)(input_layer, name='output',
kernel_initializer=normalized_columns_initializer(1.0))
elif self.initializer == 'xavier' or self.initializer is None:
self.output = self.dense_layer(1)(input_layer, name='output')
def __str__(self):
result = [