Skip to content

Instantly share code, notes, and snippets.

View LUKELIEM's full-sized avatar

Luke Liem LUKELIEM

View GitHub Profile
@LUKELIEM
LUKELIEM / pg-pong.py
Created August 30, 2017 10:10 — forked from karpathy/pg-pong.py
Training a Neural Network ATARI Pong agent with Policy Gradients from raw pixels
""" Trains an agent with (stochastic) Policy Gradients on Pong. Uses OpenAI Gym. """
import numpy as np
import cPickle as pickle
import gym
# hyperparameters
H = 200 # number of hidden layer neurons
batch_size = 10 # every how many episodes to do a param update?
learning_rate = 1e-4
gamma = 0.99 # discount factor for reward