Skip to content

Instantly share code, notes, and snippets.

View iriyagupta's full-sized avatar
I might be slow to respond

Riya Gupta iriyagupta

I might be slow to respond
View GitHub Profile
iriyagupta /
Created July 14, 2018 07:45 — forked from karpathy/
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