208e5d91d7f88cc1ab10ab1a4bfdd856b2691671,tensorforce/tests/test_trpo_agent.py,TestTRPOAgent,test_discrete,#TestTRPOAgent#,30
Before Change
def test_discrete(self):
environment = MinimalTest(continuous=False)
config = {
"batch_size": 16,
"override_line_search": False,
"cg_iterations": 20,
"use_gae": False,
"normalize_advantage": False,
"gae_lambda": 0.97,
"cg_damping": 0.001,
"line_search_steps": 20,
"max_kl_divergence": 0.05,
"max_episode_length": 4,
"continuous": False,
"state_shape": (2,),
"actions": 2,
"gamma": 0.99
}
config = create_config(config)
tf.reset_default_graph()
network_builder = layered_network_builder([{"type": "dense",
"num_outputs": 8}])
After Change
def test_discrete(self):
environment = MinimalTest(continuous=False)
config = Configuration(
batch_size=8,
learning_rate=0.0001,
cg_iterations=20,
cg_damping=0.001,
line_search_steps=20,
max_kl_divergence=0.05,
states=environment.states,
actions=environment.actions
)
network_builder = layered_network_builder(layers_config=[{"type": "dense", "size": 32}])
agent = TRPOAgent(config=config, network_builder=network_builder)
runner = Runner(agent=agent, environment=environment)
In pattern: SUPERPATTERN
Frequency: 3
Non-data size: 4
Instances
Project Name: reinforceio/tensorforce
Commit Name: 208e5d91d7f88cc1ab10ab1a4bfdd856b2691671
Time: 2017-05-22
Author: aok25@cl.cam.ac.uk
File Name: tensorforce/tests/test_trpo_agent.py
Class Name: TestTRPOAgent
Method Name: test_discrete
Project Name: reinforceio/tensorforce
Commit Name: ad1a625cd2b2dd42701435e6174a98c323be5a3e
Time: 2017-10-16
Author: mi.schaarschmidt@gmail.com
File Name: tensorforce/tests/test_reward_estimation.py
Class Name: TestRewardEstimation
Method Name: test_gae
Project Name: reinforceio/tensorforce
Commit Name: 93aafb5b8aaae27ca6542db0b896af64f96a2eb6
Time: 2017-05-21
Author: mi.schaarschmidt@gmail.com
File Name: tensorforce/tests/test_dqfd_agent.py
Class Name: TestDQFDAgent
Method Name: test_dqfd_agent