def loadModule(self, name, filename):
if lua is None:
return None
lua.globals().package.loaded[name] = None
CurrentDir = os.getcwd()
os.chdir('quirks')
try:
return lua.require(name)
except Error as e:
print(e)
return None
finally:
os.chdir(CurrentDir)
def loadModule(self, name, filename):
if lua is None:
return None
lua.globals().package.loaded[name] = None
CurrentDir = os.getcwd()
os.chdir('quirks')
try:
return lua.require(name)
except Error as e:
logging.error(e)
return None
finally:
os.chdir(CurrentDir)
print "----- import lua -----"
import lua
print "----- lg = lua.globals() -----"
lg = lua.globals()
print "lg:", lg
print "lg._G:", lg._G
print "lg['_G']:", lg['_G']
print "----- lg.foo = \"bar\" -----"
lg.foo = 'bar'
print "----- lg.tmp = [] -----"
lg.tmp = []
print "----- print lg.tmp -----"
print lg.tmp
print "----- lua.execute(\"xxx = {1,2,3,foo={4,5}}\") -----"
lua.execute("xxx = {1,2,3,foo={4,5}}")
print "----- print lg.xxx[1] -----"
print lg.xxx[1]
print "----- print lg.xxx[2] -----"
print lg.xxx[2]
print "----- print lg.xxx[3] -----"
print lg.xxx[3]
print "----- print lg.xxx['foo'][1] -----"
print lg.xxx['foo'][1]
print "lua.require =", lua.require
try:
lua.require("foo")
except:
print "lua.require('foo') raised an exception"
import lua
from vizdoom import *
from random import choice
from time import sleep
from time import time
import itertools as it
import cv2
import numpy as np
import argparse
import sys
from opencv.cv import *
from opencv.highgui import *
torch = lua.require('torch')
lua.require('trepl')
lua.require('cunn') # We run the network on GPU
dqn = lua.eval("dofile('dqn/NeuralQLearner.lua')")
tt = lua.eval("dofile('dqn/TransitionTable.lua')")
#lua.execute("dofile('dqn/Scale.lua')") # for the preproc
spectator_action_mapping = {
1 : [0,0,0,0,0,0,0,0,0] ,
2 : [1,0,0,0,0,0,0,0,0] ,
3 : [0,0,0,0,1,0,0,0,0] ,
4 : [0,0,0,0,0,1,0,0,0] ,
5 : [0,0,0,0,0,0,1,0,0] ,
6 : [0,0,0,0,0,0,0,1,0] ,
7 : [0,0,0,0,0,0,0,0,1] ,
8 : [0,0,0,0,0,1,0,1,0] ,
def style(rule):
colour = lua.eval("%s.Colour" % rule)
boldQ = lua.eval("%s.Bold" % rule)
bold = r"\bfseries" if boldQ else ""
return color(colour) + bold
if __name__ == "__main__":
print consts.common_style
for theme in os.listdir(consts.THEMES_DIR):
lua.require(theme)
name = consts.theme_name(theme)
styles = {
"name": name,
"backgroundcolor": style("Canvas"),
"basicstyle": style("Default"),
"identifierstyle": style("Keywords[2]"),
"commentstyle": style("BlockComment"),
"stringstyle": style("String"),
"keywordstyle": style("Keywords[1]"),
"procnamestyle": style("Keywords[4]"),
}
print consts.listings_style % styles
import numpy as np
import random
import time
import math
from tqdm import tqdm
import sys, DLFCN
sys.setdlopenflags(DLFCN.RTLD_NOW | DLFCN.RTLD_GLOBAL)
import lua
lua.require('xlua')
Brain = lua.require("deepql_model/deepqlearn")
from game256 import *
####### LUA & PYTHON for learning Game 2048 with Deep Reinforcement Learning #######
__author__ = 'Wonjun'
# params #
num_of_train = 100000
num_actions = 4
num_states = 192
Brain.init(num_states, num_actions)
Brain.load("inf_model_256.net")
#print "temp_window ", Brain.temporal_window
# Manipulating params of Nets #
Brain.start_learn_threshold = 2000
Brain.learning_steps_total = 100000
Brain.learning_steps_burnin = 2000
Brain.gamma = 0.95
actions = ['H','P','K','M']
#Brain.age = 1
import gym
import lua
import numpy as np
torch = lua.require('torch')
lua.require('trepl')
lua.require('cunn') # We run the network on GPU
dqn = lua.eval("dofile('dqn/NeuralQLearner.lua')")
tt = lua.eval("dofile('dqn/TransitionTable.lua')")
lua.execute("dofile('dqn/Scale.lua')") # for the preproc
env = gym.make('Breakout-v0')
possible_actions = lua.toTable({
1: 0,
2: 2,
3: 3,
4: 1
})
agent = torch.load("out/net-10000.t7")
agent.bestq = 0
observation = env.reset()
action_index = 4
done = False
t=1
while True:
t += 1
env.render()
observation, reward, done, info = env.step(possible_actions[action_index])
# This is a simple interface to the prosody XMPP server and presently takes care of simple use cases like
# creation, deletion of accounts. It should be used in further development to change the server if required!
# Note that this script should be run with privellages that can allow prosody databases to be accessed.
# If you are not sure enough, use sudo/root privellages( is that a stiable idea ? )
# Import the Lunatic Python, Lua-Python, interpolability module
import lua
# Import the Lua side of API
lua.require('prosody_api')
# Import types module for assertion reasoning
import types
class XMPPApi:
def isUser(self, uname, host):
# Check if we have got valid strings
assert ( type(uname) == types.StringType ) and ( type(host) == types.StringType )
# Evaluate the function from the prosody API
pro_isUser = '******'+uname+'","'+host+'")'
return lua.eval(pro_isUser)
def createUser(self, uname, passwd, host):
def loadModule(self, name, filename):
if lua is None:
return None
fullname = os.path.join('quirks', name)
lua.globals().package.loaded[fullname] = None
return lua.require(fullname)
from random import choice
from time import sleep
from time import time
import itertools as it
#changes made
import lua
import numpy as np
torch = lua.require('torch')
lua.require('trepl')
dqn = lua.eval("dofile('dqn/NeuralQLearner.lua')")
tt = lua.eval("dofile('dqn/TransitionTable.lua')")
# Create DoomGame instance. It will run the game and communicate with you.
game = DoomGame()
screen_width = 320
screen_height = 240
color_palette = 24
# Now it's time for configuration!
# load_config could be used to load configuration instead of doing it here with code.
# If load_config is used in-code configuration will work. Note that the most recent changes will add to previous ones.