from Actor_Critic_Agents.DDPG_Agent import DDPG_Agent
from Data_Structures.Config import Config
from Lunar_Lander_Continuous import Lunar_Lander_Continuous
from Open_AI_Gym_Environments.Mountain_Car_Continuous_Environment import Mountain_Car_Continuous_Environment
from PPO_Agent import PPO_Agent
from Utility_Functions import run_games_for_agents
config = Config()
config.seed = 200
config.environment = Lunar_Lander_Continuous()
config.max_episodes_to_run = 3000
config.file_to_save_data_results = "Results_Data.pkl"
config.file_to_save_data_results_graph = "Results_Graph2.png"
config.visualise_individual_results = False
config.visualise_overall_results = True
config.runs_per_agent = 10
config.hyperparameters = {
"Policy_Gradient_Agents": {
"learning_rate": 0.02,
"nn_layers": 2,
"nn_start_units": 20,
"nn_unit_decay": 1.0,
"final_layer_activation": "TANH",
"learning_iterations_per_round": 10,
"discount_rate": 0.99,
"batch_norm": False,
"clip_epsilon": 0.2,
"episodes_per_learning_round": 7,
"normalise_rewards": True,
"gradient_clipping_norm": 5,
from Actor_Critic_Agents.DDPG_Agent import DDPG_Agent
from DDPG_HER_Agent import DDPG_HER_Agent
from Data_Structures.Config import Config
from Fetch_Reach_Environment import Fetch_Reach_Environment
from Utility_Functions import run_games_for_agents
config = Config()
config.seed = 100
config.environment = Fetch_Reach_Environment()
config.max_episodes_to_run = 2000
config.file_to_save_data_results = "Results_Data.pkl"
config.file_to_save_data_results_graph = "Results_Graph.png"
config.visualise_individual_results = True
config.visualise_overall_results = True
config.runs_per_agent = 1
config.use_GPU = False
config.hyperparameters = {
"Actor_Critic_Agents": {
"Actor": {
"learning_rate": 0.001,
"nn_layers": 5,
"nn_start_units": 50,
"nn_unit_decay": 1.0,
"final_layer_activation": "TANH",
"batch_norm": False,
"tau": 0.01,
"gradient_clipping_norm": 5
},
"Critic": {
"learning_rate": 0.01,
import gym
from Actor_Critic_Agents.DDPG import DDPG
from Agents.Actor_Critic_Agents.DDPG_HER import DDPG_HER
from Data_Structures.Config import Config
from Agents.Trainer import Trainer
config = Config()
config.seed = 1
config.environment = gym.make("FetchReach-v1")
config.num_episodes_to_run = 2
config.file_to_save_data_results = None
config.file_to_save_results_graph = None
config.show_solution_score = False
config.visualise_individual_results = False
config.visualise_overall_agent_results = True
config.standard_deviation_results = 1.0
config.runs_per_agent = 3
config.use_GPU = False
config.overwrite_existing_results_file = False
config.randomise_random_seed = True
config.save_model = False
config.hyperparameters = {
"Actor_Critic_Agents": {
"Actor": {
"learning_rate": 0.001,
"linear_hidden_units": [50, 50],
"final_layer_activation": "TANH",
"batch_norm": False,
"tau": 0.01,
from PPO_Agent import PPO_Agent
from Data_Structures.Config import Config
from Agents.DQN_Agents.DDQN_Agent import DDQN_Agent
from Agents.DQN_Agents.DDQN_With_Prioritised_Experience_Replay import DDQN_With_Prioritised_Experience_Replay
from Agents.DQN_Agents.DQN_Agent import DQN_Agent
from Agents.DQN_Agents.DQN_Agent_With_Fixed_Q_Targets import DQN_Agent_With_Fixed_Q_Targets
from Environments.Open_AI_Gym_Environments.Cart_Pole_Environment import Cart_Pole_Environment
from Agents.Policy_Gradient_Agents.REINFORCE_Agent import REINFORCE_Agent
from Agents.Stochastic_Policy_Search_Agents.Genetic_Agent import Genetic_Agent
from Agents.Stochastic_Policy_Search_Agents.Hill_Climbing_Agent import Hill_Climbing_Agent
from Utilities.Utility_Functions import run_games_for_agents
config = Config()
config.seed = 100
config.environment = Cart_Pole_Environment()
config.max_episodes_to_run = 2000
config.file_to_save_data_results = "Results_Data.pkl"
config.file_to_save_data_results_graph = "Results_Graph.png"
config.visualise_individual_results = True
config.visualise_overall_results = True
config.runs_per_agent = 1
config.hyperparameters = {
"DQN_Agents": {
"learning_rate": 0.005,
"batch_size": 256,
"buffer_size": 40000,
"epsilon": 0.1,
"epsilon_decay_rate_denominator": 200,
"discount_rate": 0.99,
"tau": 0.1,