"batch_norm": False,
"tau": 0.01,
"gradient_clipping_norm": 5
},
"Critic": {
"learning_rate": 0.01,
"linear_hidden_units": [400, 300],
"final_layer_activation": "None",
"batch_norm": False,
"buffer_size": 100000,
"tau": 0.01,
"gradient_clipping_norm": 5
},
"batch_size": 64,
"discount_rate": 0.99,
"mu": 0.0, # for O-H noise
"theta": 0.15, # for O-H noise
"sigma": 0.2, # for O-H noise
"action_noise_std": 0.2, # for TD3
"action_noise_clipping_range": 0.5, # for TD3
"update_every_n_steps": 1,
"learning_updates_per_learning_session": 1,
"clip_rewards": False
}
}
if __name__ == "__main__":
AGENTS = [DDPG, HIRO]
trainer = Trainer(config, AGENTS)
trainer.run_games_for_agents()
def predictionToFloat(prediction: Variable) -> float:
return prediction.data.numpy()[0]
# steeringData = SteeringTrainingData(pandas.read_csv('training-data/train_data/alpine-1.csv'))
# brakingData = BrakingTrainingData(pandas.read_csv('training-data/train_data/alpine-1.csv'))
steeringData = ExtendedSteeringData()
observations = []
driver = MyDriver()
for i in range(len(steeringData)):
state = dataToStateExtended(steeringData[i][0])
sample = Trainer.stateToSample(state)
command = driver.drive(state)
#
# observation = {
# 'target': (brakingData[i][1]).numpy()[0],
# 'actual': command.brake
# }
observation = {
'target': (steeringData[i][1]).numpy()[0],
'actual': command.steering
}
observations.append(observation)