def initialize_custom_reset(self, alternate_route):
"""
resets the environment with a custom route for the agent
:param alternate_route: list: list of alternate route nodes for car agent
:return state: list: initial state of agent
"""
# initialize the car and light state objects
init_car_state = self.car_init_method(self.N,
self.axis,
car_id=self.agent,
alternate_route=alternate_route)
self.cars_object = Cars(init_state=init_car_state, axis=self.axis)
if self.animate:
# init animator
self.animator = Animator(fig=self.fig,
ax=self.ax,
cars_object=self.cars_object,
lights_object=self.lights_object,
num=self.num)
stateview = self.refresh_stateview()
state = stateview.determine_state()[0]
state = state.index(True)
return state
def __init__(self, n, graph, agent, dt, animate=False):
"""
initializes an environment for a car in the system
:param n: int: number of cars to simulate
:param graph: OGraph object from
:param agent: int: the ID of the car (agent)
:param animate: bool: if the environment is to be animated while learning
"""
self.N = n
self.num = None
self.graph = graph
self.fig, self.ax = self.graph.fig, self.graph.ax
self.agent = agent
self.dt = dt
self.animate = animate
self.animator = None
self.axis = self.ax.axis()
self.route_times = []
self.car_init_method = sim.init_random_node_start_location
# self.car_init_method = sim.init_culdesac_start_location
self.light_init_method = sim.init_traffic_lights
# self.car_init_method = convergent_learner.init_custom_agent
# self.light_init_method = convergent_learner.init_custom_lights
self.cars_object = Cars(self.car_init_method(self.N, self.graph),
self.graph)
self.lights_object = TrafficLights(
self.light_init_method(self.graph, prescale=40), self.graph)
self.high = 10
self.low = 2
self.shortest_route_thresh = 5
def __init__(self, n, fig, ax, agent, dt, animate=False):
"""
initializes an environment for a car in the system
:param n: int: number of cars to simulate
:param fig: figure: from matplotlib
:param ax: axis: from matplotlib
:param agent: int: the ID of the car (agent)
:param animate: bool: if the environment is to be animated while learning
"""
self.N = n
self.num = None
self.fig = fig
self.ax = ax
self.agent = agent
self.dt = dt
self.animate = animate
self.animator = None
self.axis = self.ax.axis()
self.route_times = []
# self.car_init_method = sim.init_culdesac_start_location
# self.light_init_method = sim.init_traffic_lights
self.car_init_method = convergent_learner.init_custom_agent
self.light_init_method = convergent_learner.init_custom_lights
self.cars_object = Cars(self.car_init_method(self.N, self.axis),
self.axis)
self.lights_object = TrafficLights(
self.light_init_method(self.axis, prescale=40), self.axis)
self.high = 10
self.low = 2
self.shortest_route_thresh = 5
def __init__(self, drones_positions, cars_inits):
self.steps_left = TIME_LENGTH
# création des voitures
self.cars = [Cars(car_init) for car_init in cars_inits]
# création des drones
self.drones = [Drones(drone_z, 0) for drone_z in drones_positions]
# la visibilité au niveau env est le cumul des visibiités des drones
self.visibilities = [np.zeros((SIZE_X, SIZE_Y)) < 0]
# pour se lier avec la librairie Acme
self.spec = Environment_spec(
number_of_drones=len(drones_positions),
number_of_cars=len(cars_inits)
)
def reset(self, num):
"""
resets the environment
:param num: tuple: int, int
:return state: int
"""
# initialize cars every reset
init_cars = self.car_init_method(self.N, self.axis)
self.cars_object = Cars(init_state=init_cars, graph=self.graph)
stateview = self.refresh_stateview()
state = stateview.determine_state()[0]
state = state.index(True)
if self.animate:
# init animator
self.num = num
self.animator = Animator(fig=self.fig,
ax=self.ax,
cars_object=self.cars_object,
lights_object=self.lights_object,
num=self.num)
return state
def addCustomer(customer):
car_list = Cars()
car_list.get_available_cars()
car_number = input("Enter number of car, which you want to take: ")
hours = input("And enter hours for rent: ")
while car_number is not '' and hours is not '':
if car_list.is_available(car_number):
car = car_list.set_unavailable_cars(str(car_number))
customer.add_customer_cars(car, hours)
car_list.get_available_cars()
# print(car_number + " " + json.dumps(car, indent=4))
else:
print("This cart is unavailable, please choose another car!")
car_number = input(
"If you want to get another car enter its number, or just press Enter: "
)
if car_number is '':
break
hours = input("And enter hours for rent: ")
return customer.get_customer_cars()
# G = ox.load_graphml('data/sanfrancisco.graphml')
# G = ox.project_graph(G)
# fig, ax = ox.plot_graph(G, fig_height=12, fig_width=10, node_size=0, edge_linewidth=0.5)
# ax.set_title('San Francisco, California')
"""Piedmont, California"""
G = ox.load_graphml('piedmont.graphml')
G = ox.project_graph(G)
fig, ax = ox.plot_graph(G, node_size=0, edge_linewidth=0.5, show=False)
ax.set_title('Piedmont, California')
# grab the dimensions of the figure
axis = ax.axis()
""" initialize the car and light state objects """
N = 33 # cars
# cars = Cars(sim.init_culdesac_start_location(N, axis), axis)
cars = Cars(sim.init_random_node_start_location(N, axis), axis)
lights = TrafficLights(sim.init_traffic_lights(axis, prescale=40), axis)
""" for an example of learning using a single, convergent learner, initialize the sim using these cars and lights: """
# cars = Cars(cl.init_custom_agent(n=1, fig_axis=axis), axis=axis)
# lights = TrafficLights(cl.init_custom_lights(fig_axis=axis, prescale=None), axis)
# initialize the Animator
animator = Animator(fig=fig,
ax=ax,
cars_object=cars,
lights_object=lights,
num=(1, 10),
n=N)
init = animator.reset
animate = animator.animate
from cars import Cars
if __name__ == "__main__":
Cars(10)
else:
connection.close()
elif command == '3':
result = red_light.get_state()
print(result)
connection.sendall(str.encode(str(result)))
else:
result = 'Not authorised'
connection.sendall(bytes(result, 'utf-8'))
connection.close()
if __name__ == "__main__":
global red_light
global user_id
red_light = AccessPoint()
user_id = Cars()
while True:
with socket.socket(socket.AF_INET,
socket.SOCK_STREAM) as SystemSideSocket:
try:
SystemSideSocket.bind((HOST, PORT))
except socket.error as e:
print(e)
print('System Socket is listenning..')
SystemSideSocket.listen(1)
while True:
client, address = SystemSideSocket.accept()
print('Connected to: ' + address[0] + ':' + str(address[1]))
_thread.start_new_thread(con_handler, (client, ))
SystemSideSocket.close()
from turtle import Screen
from player import Player
from scoreboard import Scoreboard
from cars import Cars
from time import sleep
screen = Screen()
screen.colormode(255)
screen.setup(width=600, height=600)
screen.listen()
screen.tracer(0)
player = Player()
scoreboard = Scoreboard()
cars = Cars()
screen.onkey(player.moveup, 'w')
screen.onkey(player.moveup, 'Up')
screen.onkey(player.movedown, 's')
screen.onkey(player.movedown, 'Down')
screen.onkey(player.moveright, 'Right')
screen.onkey(player.moveright, 'd')
screen.onkey(player.moveleft, 'a')
screen.onkey(player.moveleft, 'Left')
level = 1
game_on = True
while game_on:
screen.update()
scoreboard.update_scoreboard(level)
cars.move_car(level)
if player.ycor() >= 285:
from turtle import Turtle, Screen
from cars import Cars, car_list
from racer import Racer
from finishline import RaceLine
from levels import Level
# a turtle crossing game. Get the turtle to the finish line without hitting a car. Move forward with spacebar
turtle.listen()
# generate screen
screen = Screen()
screen.setup(height=600, width=600)
screen.tracer(0)
# generate game objects
car = Cars()
racer_turtle = Racer()
finish_line = RaceLine(250)
start_line = RaceLine(-250)
level_start = Level()
# instructions = level_start.instructions()
# start game
game_on = True
while game_on:
time.sleep(level_start.time_set)
screen.update()
for car in car_list:
car.move()
# generate one car after previous reaches a given coordinate
if car_list[-1].xcor() in range(200, 250):
car = Cars()
from osm_request import OGraph
from tqdm import tqdm
import sys
# ask the user for a geo-codable location
query = input(
'Please input a geo-codable place, like "Harlem, NY" or "Kigali, Rwanda": '
)
# get OGraph object
graph = OGraph(query, save=True)
# initialize the car and light state objects
N = int(input('Number of cars to simulate: '))
# cars = Cars(sim.init_culdesac_start_location(N, graph), graph)
cars = Cars(sim.init_random_node_start_location(N, graph), graph)
lights = TrafficLights(sim.init_traffic_lights(graph, prescale=15), graph)
""" for an example of learning using a single, convergent learner, initialize the sim using these cars and lights: """
# cars = Cars(cl.init_custom_agent(n=1, fig_axis=axis), axis=axis)
# lights = TrafficLights(cl.init_custom_lights(fig_axis=axis, prescale=None), axis)
# time of simulation (in seconds)
duration = int(input('Duration of time to simulate (in seconds): '))
frames_per_second = 60
n_frames = duration * frames_per_second
# initialize the Animator
animator = Animator(fig=graph.fig,
ax=graph.ax,
cars_object=cars,
lights_object=lights,
def setUp(self):
print('setUp')
self.toyota = Cars('Toyota', 'Supra', 236000000000)
from cars import Cars
from electric_car_m import ElectricCar
my_car = ElectricCar('tesla', 'model s', 2017)
print(my_car.summary())
my_car.battery.read_battery()
your_car = Cars('toyoya', 'prius', 2000)
print(your_car.summary())
your_car.odometer_reading()
from turtle import Screen
from time import sleep
from player import Player
from cars import Cars
from random import randint
from score import Levels
screen = Screen()
screen.setup(width=600, height=600)
screen.tracer(0)
screen.listen()
player1 = Player()
cars = []
for _ in range(40):
cars.append(Cars(randint(-240, 240)))
score_levels = Levels()
screen.onkey(player1.go_up, "Up")
game_on = True
while game_on:
if player1.ycor() > 300:
player1.go_to_start()
score_levels.level += 1
score_levels.show_level()
for i in cars:
i.more_speed()
from turtle import Screen
from playerturtle import PlayerTurtle
import time
from cars import Cars
from scoreboard import Scoreboard
screen = Screen()
screen.title("Cross Road Game")
screen.setup(width=600, height=600)
screen.bgcolor("white")
screen.tracer(0)
car = Cars()
player = PlayerTurtle()
scoreboard = Scoreboard()
screen.listen()
screen.onkey(player.go_up, "Up")
game_is_on = True
while game_is_on:
time.sleep(0.1)
screen.update()
# move cars
car.create_cars()
car.move()
from cars import Cars
#figured it out in the coolcars statement - need to fix the other ones though
cool_car = Cars('Red', '2')
lame_car = Cars('Brown', '4')
great_car = Cars('Green', '4')
great_car.honk()
print('Information for cool car: color: {}, number of doors: {}\nInformation for lame car: color: {}, number of doors: {}\nInformation for great car: color: {}, number of doors: {}'.format(cool_car.color, cool_car.doorsnum, lame_car.color, lame_car.doorsnum, great_car.color, great_car.doorsnum))
"""Piedmont, California"""
G = ox.load_graphml('piedmont.graphml')
G = ox.project_graph(G)
fig, ax = ox.plot_graph(G, node_size=0, edge_linewidth=0.5)
ax.set_title('Piedmont, California')
# grab the dimensions of the figure
axis = ax.axis()
# initialize the car and light state objects
# cars = Cars(sim.init_culdesac_start_location(N, axis), axis)
# cars = Cars(sim.init_random_node_start_location(N, axis), axis)
# lights = TrafficLights(sim.init_traffic_lights(axis, prescale=40), axis)
cars = Cars(cl.init_custom_agent(n=1, fig_axis=axis), axis=axis)
lights = TrafficLights(cl.init_custom_lights(fig_axis=axis, prescale=None), axis)
# initialize the Animator
animator = Animator(fig=fig, ax=ax, cars_object=cars, lights_object=lights, num=(1, 10))
init = animator.reset
animate = animator.animate
# for creating HTML frame-movies
# ani = animation.FuncAnimation(fig, animate, init_func=init, frames=1200, interval=30, blit=True)
# ani.save('traffic.html', fps=300, extra_args=['-vcodec', 'libx264'])
# for creating mp4 movies
ani = animation.FuncAnimation(fig, animate, init_func=init, frames=20000)
mywriter = animation.FFMpegWriter(fps=300)
class Env:
def __init__(self, n, graph, agent, dt, animate=False):
"""
initializes an environment for a car in the system
:param n: int: number of cars to simulate
:param graph: OGraph object from
:param agent: int: the ID of the car (agent)
:param animate: bool: if the environment is to be animated while learning
"""
self.N = n
self.num = None
self.graph = graph
self.fig, self.ax = self.graph.fig, self.graph.ax
self.agent = agent
self.dt = dt
self.animate = animate
self.animator = None
self.axis = self.ax.axis()
self.route_times = []
self.car_init_method = sim.init_random_node_start_location
# self.car_init_method = sim.init_culdesac_start_location
self.light_init_method = sim.init_traffic_lights
# self.car_init_method = convergent_learner.init_custom_agent
# self.light_init_method = convergent_learner.init_custom_lights
self.cars_object = Cars(self.car_init_method(self.N, self.graph),
self.graph)
self.lights_object = TrafficLights(
self.light_init_method(self.graph, prescale=40), self.graph)
self.high = 10
self.low = 2
self.shortest_route_thresh = 5
def reset(self, num):
"""
resets the environment
:param num: tuple: int, int
:return state: int
"""
# initialize cars every reset
init_cars = self.car_init_method(self.N, self.axis)
self.cars_object = Cars(init_state=init_cars, graph=self.graph)
stateview = self.refresh_stateview()
state = stateview.determine_state()[0]
state = state.index(True)
if self.animate:
# init animator
self.num = num
self.animator = Animator(fig=self.fig,
ax=self.ax,
cars_object=self.cars_object,
lights_object=self.lights_object,
num=self.num)
return state
def refresh_stateview(self):
"""
this function prepares a fresh depiction of what state the car is in
:return stateview: object
"""
stateview = nav.StateView(graph=self.graph,
car_index=self.agent,
cars=self.cars_object.state,
lights=self.lights_object.state)
return stateview
def initialize_custom_reset(self, alternate_route):
"""
resets the environment with a custom route for the agent
:param alternate_route: list: list of alternate route nodes for car agent
:return state: list: initial state of agent
"""
# initialize the car and light state objects
init_car_state = self.car_init_method(self.N,
self.axis,
car_id=self.agent,
alternate_route=alternate_route)
self.cars_object = Cars(init_state=init_car_state, axis=self.axis)
if self.animate:
# init animator
self.animator = Animator(fig=self.fig,
ax=self.ax,
cars_object=self.cars_object,
lights_object=self.lights_object,
num=self.num)
stateview = self.refresh_stateview()
state = stateview.determine_state()[0]
state = state.index(True)
return state
def step(self, action, num):
"""
This function runs a full simulation of a car from origin to destination
(if action, then use the alternate route)
:param action: int: 0 or 1
:param num: tuple: the simulation number out of the total number of simulations
:return new_state, reward, done, _: list: the end of the return is free to contain debugging info
"""
debug_report = []
if self.animate:
self.animator.reset(self.num)
stateview = self.refresh_stateview()
state, new_route, new_xpath, new_ypath = stateview.determine_state()
if action:
new_state = self.initialize_custom_reset(
alternate_route=(new_route, new_xpath, new_ypath))
else:
new_state = state.index(True)
arrived = False
i = 0
while not arrived:
arrived = self.simulation_step(i)
i += 1
route_time = self.cars_object.state.loc[self.agent]['route-time']
self.route_times.append(route_time)
# TODO: need new way of identifying shortest route time.
if len(self.route_times) < self.shortest_route_thresh:
shortest_route_found_reward = 0
done = False
elif np.isclose(0,
self.route_times[-1] - np.min(self.route_times),
atol=5 * self.dt).all():
"""
If the route time achieved after the simulation is within 5 x dt second of the minimum time achieved.
Define this environment condition as having found the shortest route (locally).
"""
shortest_route_found_reward = self.high
done = True
else:
shortest_route_found_reward = 0
done = False
if num[0] < 1:
reward = 0
else:
time_delta = self.route_times[num[0] - 1] - self.route_times[
num[0]] + shortest_route_found_reward
if time_delta > 0:
reward = time_delta
else:
reward = 0
return new_state, reward, done, debug_report
def simulation_step(self, i):
"""
make one step in the simulation
:param i: simulation step
:return arrived: bool
"""
frontview = nav.FrontView(self.cars_object.state.loc[self.agent],
self.graph)
end_of_route = frontview.end_of_route()
if not end_of_route:
if self.animate:
self.animator.animate(i)
else:
self.lights_object.update(self.dt)
self.cars_object.update(self.dt, self.lights_object.state)
arrived = False
else:
arrived = True
return arrived
from turtle import Screen, Turtle
from player import Player
from cars import Cars
from score import Score
import time
#Setting up the screen
sc = Screen()
sc.tracer(0)
sc.setup(width=600, height=600)
difficulty_increase = 1.25
#Instantiating the Cars, Player, and Scoreboard
player = Player()
score = Score()
cars = Cars()
sc.listen()
sc.onkey(player.move_up, 'Up')
playing = True
while playing:
time.sleep(0.1)
sc.update()
cars.move_cars()
#Checking for collision
for car in cars.cars:
if player.distance(car) < 20:
playing = False
score.goto(0, 0)
score.write("YOU LOSE",
screen.tracer(0)
screen.setup(width=600, height=600)
scoreboard = Scoreboard()
player = Player()
car_list = []
loop_count = 0
screen.listen()
screen.onkey(player.move, "Up")
is_game_on = True
while is_game_on:
loop_count += 1
if loop_count % 6 == 0:
new_car = Cars()
new_car.increase_speed(scoreboard.score)
car_list.append(new_car)
for car in car_list:
car.move()
if car.distance(player) < 25:
scoreboard.game_over()
is_game_on = False
time.sleep(0.1)
screen.update()
if player.level_up():
new_car = []
screen.update()