1
0
mirror of https://github.com/gryf/coach.git synced 2026-04-21 07:51:28 +02:00

network_imporvements branch merge

This commit is contained in:
Shadi Endrawis
2018-10-02 13:41:46 +03:00
parent 72ea933384
commit 51726a5b80
110 changed files with 1639 additions and 1161 deletions
+4 -10
View File
@@ -1,8 +1,7 @@
from rl_coach.agents.pal_agent import PALAgentParameters
from rl_coach.base_parameters import VisualizationParameters, PresetValidationParameters
from rl_coach.core_types import TrainingSteps, EnvironmentEpisodes, EnvironmentSteps, RunPhase
from rl_coach.environments.environment import SelectedPhaseOnlyDumpMethod, MaxDumpMethod
from rl_coach.environments.gym_environment import Mujoco
from rl_coach.core_types import TrainingSteps, EnvironmentEpisodes, EnvironmentSteps
from rl_coach.environments.gym_environment import GymVectorEnvironment
from rl_coach.graph_managers.basic_rl_graph_manager import BasicRLGraphManager
from rl_coach.graph_managers.graph_manager import ScheduleParameters
from rl_coach.memories.memory import MemoryGranularity
@@ -41,12 +40,7 @@ agent_params.exploration.epsilon_schedule = LinearSchedule(1.0, 0.01, 10000)
################
# Environment #
################
env_params = Mujoco()
env_params.level = 'CartPole-v0'
vis_params = VisualizationParameters()
vis_params.video_dump_methods = [SelectedPhaseOnlyDumpMethod(RunPhase.TEST), MaxDumpMethod()]
vis_params.dump_mp4 = False
env_params = GymVectorEnvironment(level='CartPole-v0')
########
# Test #
@@ -57,5 +51,5 @@ preset_validation_params.min_reward_threshold = 150
preset_validation_params.max_episodes_to_achieve_reward = 250
graph_manager = BasicRLGraphManager(agent_params=agent_params, env_params=env_params,
schedule_params=schedule_params, vis_params=vis_params,
schedule_params=schedule_params, vis_params=VisualizationParameters(),
preset_validation_params=preset_validation_params)