Merge branch 'master' of https://github.com/p-christ/Deep_RL_Implementations

Author: p-christ

.gitignore

@@ -22,4 +22,8 @@ Results/Notebook.ipynb
 *.ipynb_checkpoints
 *.drive_access_key.json
 drive_access_key.json
-drive_access_key
+drive_access_key
+settings.json
+launch.json
+results/data_and_graphs/Cart_Pole_Results_Data.pkl
+results/data_and_graphs/Cart_Pole_Results_Graph.png

agents/Base_Agent.py

@@ -21,7 +21,7 @@ def __init__(self, config):
         self.set_random_seeds(config.seed)
         self.environment = config.environment
         self.environment_title = self.get_environment_title()
-        self.action_types = "DISCRETE" if self.environment.action_space.dtype == int else "CONTINUOUS"
+        self.action_types = "DISCRETE" if self.environment.action_space.dtype == np.int64 else "CONTINUOUS"
         self.action_size = int(self.get_action_size())
         self.config.action_size = self.action_size
 
@@ -106,14 +106,18 @@ def get_score_required_to_win(self):
 
     def get_trials(self):
         """Gets the number of trials to average a score over"""
-        if self.environment_title in ["AntMaze", "FetchReach", "Hopper", "Walker2d"]: return 100
+        if self.environment_title in ["AntMaze", "FetchReach", "Hopper", "Walker2d", "CartPole"]: return 100
         try: return self.environment.unwrapped.trials
         except AttributeError: return self.environment.spec.trials
 
     def setup_logger(self):
         """Sets up the logger"""
         filename = "Training.log"
-        if os.path.isfile(filename): os.remove(filename)
+        try: 
+            if os.path.isfile(filename): 
+                os.remove(filename)
+        except: pass
+
         logger = logging.getLogger(__name__)
         logger.setLevel(logging.INFO)
         # create a file handler

results/Cart_Pole.py

@@ -1,8 +1,13 @@
+import os
+import sys
+from os.path import dirname, abspath
+sys.path.append(dirname(dirname(abspath(__file__))))
+
 import gym
 
-from A2C import A2C
-from Dueling_DDQN import Dueling_DDQN
-from SAC_Discrete import SAC_Discrete
+from agents.actor_critic_agents.A2C import A2C
+from agents.DQN_agents.Dueling_DDQN import Dueling_DDQN
+from agents.actor_critic_agents.SAC_Discrete import SAC_Discrete
 from agents.actor_critic_agents.A3C import A3C
 from agents.policy_gradient_agents.PPO import PPO
 from agents.Trainer import Trainer
@@ -16,8 +21,8 @@
 config.seed = 1
 config.environment = gym.make("CartPole-v0")
 config.num_episodes_to_run = 450
-config.file_to_save_data_results = "data_and_graphs/Cart_Pole_Results_Data.pkl"
-config.file_to_save_results_graph = "data_and_graphs/Cart_Pole_Results_Graph.png"
+config.file_to_save_data_results = "results/data_and_graphs/Cart_Pole_Results_Data.pkl"
+config.file_to_save_results_graph = "results/data_and_graphs/Cart_Pole_Results_Graph.png"
 config.show_solution_score = False
 config.visualise_individual_results = False
 config.visualise_overall_agent_results = True