Python Config.file_to_save_data_results Examples

Example #1

File: Test_HIRO.py Project: hyzcn/Deep-Reinforcement-Learning-Algorithms-with-PyTorch

import random
import numpy as np
import torch

from Hierarchical_Agents.HIRO import HIRO
from Utilities.Data_Structures.Config import Config

random.seed(1)
np.random.seed(1)
torch.manual_seed(1)

config = Config()
config.seed = 1
config.environment = gym.make("Pendulum-v0")
config.num_episodes_to_run = 1500
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 = 1
config.use_GPU = False
config.overwrite_existing_results_file = False
config.randomise_random_seed = True
config.save_model = False

config.hyperparameters = {
    "LOWER_LEVEL": {
        "max_lower_level_timesteps": 3,
        "Actor": {

Example #2

File: Long_Corridor.py Project: hyzcn/Deep-Reinforcement-Learning-Algorithms-with-PyTorch

from Hierarchical_Agents.SNN_HRL import SNN_HRL
from Agents.Trainer import Trainer
from Utilities.Data_Structures.Config import Config
from Agents.DQN_Agents.DQN import DQN
from Agents.Hierarchical_Agents.h_DQN import h_DQN
from Environments.Long_Corridor_Environment import Long_Corridor_Environment

config = Config()
config.seed = 1
config.env_parameters = {"stochasticity_of_action_right": 0.5}
config.environment = Long_Corridor_Environment(
    stochasticity_of_action_right=config.
    env_parameters["stochasticity_of_action_right"])
config.num_episodes_to_run = 10000
config.file_to_save_data_results = "Data_and_Graphs/Long_Corridor_Results_Data.pkl"
config.file_to_save_results_graph = "Data_and_Graphs/Long_Corridor_Results_Graph.png"
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 = {
    "h_DQN": {
        "CONTROLLER": {
            "batch_size":
            256,

Example #3

from Hierarchical_Agents.SNN_HRL import SNN_HRL
from Agents.Trainer import Trainer
from Utilities.Data_Structures.Config import Config
from Agents.DQN_Agents.DQN import DQN

config = Config()
config.seed = 1
config.environment = Four_Rooms_Environment(
    15,
    15,
    stochastic_actions_probability=0.25,
    random_start_user_place=True,
    random_goal_place=False)

config.num_episodes_to_run = 1000
config.file_to_save_data_results = "Data_and_Graphs/Four_Rooms.pkl"
config.file_to_save_results_graph = "Data_and_Graphs/Four_Rooms.png"
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 = {
    "DQN_Agents": {
        "linear_hidden_units": [10, 5],
        "learning_rate":

Example #4

from Agents.DQN_Agents.DQN_HER_Agent import DQN_HER_Agent
from Environments.Other_Enrivonments.Bit_Flipping_Environment import Bit_Flipping_Environment
from Utilities.Data_Structures.Config import Config
from Agents.DQN_Agents.DQN_Agent import DQN_Agent
from Utilities.Utility_Functions import run_games_for_agents

config = Config()
config.seed = 100
config.environment = Bit_Flipping_Environment(14)
config.max_episodes_to_run = 6000
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 = 3
config.use_GPU = False

config.hyperparameters = {
    "DQN_Agents": {
        "learning_rate": 0.001,
        "batch_size": 128,
        "buffer_size": 100000,
        "epsilon": 0.1,
        "epsilon_decay_rate_denominator": 500,
        "discount_rate": 0.98,
        "tau": 0.1,
        "alpha_prioritised_replay": 0.6,
        "beta_prioritised_replay": 0.4,
        "incremental_td_error": 1e-8,
        "update_every_n_steps": 1,
        "nn_layers": 2,

Example #5

from gym.wrappers import FlattenDictWrapper
from Agents.DQN_Agents.DQN_HER import DQN_HER
from Bit_Flipping_Environment import Bit_Flipping_Environment
from Agents.Trainer import Trainer
from Utilities.Data_Structures.Config import Config
from Agents.DQN_Agents.DQN import DQN

config = Config()
config.seed = 1
config.environment = Bit_Flipping_Environment(14)
config.num_episodes_to_run = 4500
config.file_to_save_data_results = None #"Data_and_Graphs/Bit_Flipping_Results_Data.pkl"
config.file_to_save_results_graph = None #"Data_and_Graphs/Bit_Flipping_Results_Graph.png"
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 = {
    "DQN_Agents": {
        "learning_rate": 0.001,
        "batch_size": 128,
        "buffer_size": 100000,
        "epsilon_decay_rate_denominator": 150,
        "discount_rate": 0.999,