pranz24

#!/bin/bash

# Emoji
TICK='\U2714'
THUMBS_UP='\U1F44D'
DOWN_ARROW='\U2B07'

# Colors
BRED='\033[1;31m'
BGREEN='\033[1;32m'

pranz24

"""
small neural network trained using actor-critic in Pytorch
References:
David Silver's Lecture 7-Policy Gradient Methods: https://www.youtube.com/watch?v=KHZVXao4qXs&t=46s
Actor-critic example in Pytorch: https://github.com/pytorch/examples/blob/master/reinforcement_learning/actor_critic.py
I think a higher score can be achieved by running the algorithm for more number of episodes(>80000)
"""
import argparse
import gym
import numpy as np

pranz24

# A small neural network is trained using Covariance Matrix Adaptation Evolution Strategies
# The idea is from this paper: https://arxiv.org/abs/1604.00772
# This gist is using pybrain (http://pybrain.org/)


import logging
import os
import numpy as np

import gym