quq99

""" Trains an agent with (stochastic) Policy Gradients on Pong. Uses OpenAI Gym. """
import numpy as np
import cPickle as pickle
import gym
import copy

# hyperparameters
H = 200 # number of hidden layer neurons
batch_size = 10 # every how many episodes to do a param update?
learning_rate = 1e-4

quq99

import numpy as np
import gym
from gym.spaces import Discrete, Box
from gym import wrappers
#===================================
#Polices
#==================================

class DeterministicDiscreteActionLinearPolicy(object):
    def __init__(self, theta, ob_space, ac_space):

quq99

import numpy as np
import gym
from gym.spaces import Discrete, Box
from gym import wrappers
#===================================
#Polices
#==================================

class DeterministicDiscreteActionLinearPolicy(object):
    def __init__(self, theta, ob_space, ac_space):

quq99

import numpy as np, os
os.environ["THEANO_FLAGS"]="device=cpu,floatX=float64"
import theano, theano.tensor as T
import gym
from gym import wrappers

def discount(x, gamma):
    """
    Given vector x, computes a vector y such that
    y[i] = x[i] + gamma * x[i+1] + gamma^2 * x[i+2] ...

quq99

import gym
from gym import wrappers
import tensorflow as tf 
import numpy as np 
import random
from collections import deque

# Hyper Parameters for DQN
GAMMA = 0.9 # discount factor for target Q 
INITIAL_EPSILON = 0.5 # starting value of epsilon