def breed(self, mother, father):
"""Make two children as parts of their parents.
Args:
mother (dict): Network parameters
father (dict): Network parameters
Returns:
(list): Two network objects
"""
children = []
for _ in range(2):
child = {}
# Loop through the parameters and pick params for the kid.
for param in self.nn_param_choices:
child[param] = random.choice(
[mother.network[param], father.network[param]])
# Now create a network object.
network = Network(self.nn_param_choices)
network.create_set(child)
# Randomly mutate some of the children.
if self.mutate_chance > random.random():
network = self.mutate(network)
children.append(network)
return children
def run():
nodes = [Node('0', 0, 0), Node('1', 0, 50), Node('2', 0, 400)]
edges = [Edge(nodes[0], nodes[1]), Edge(nodes[1], nodes[2])]
newNet = Network(nodes, edges)
with open("local\\sumo.path", 'r') as f:
sumo_path = f.read()
sumo_tools = sumo_path + '\\tools'
sumo_path += '\\bin'
print(sumo_path)
os.makedirs('simple_2d', exist_ok=True)
newNet.writeNet(sumo_path, 'simple_2d\\2d')
vehTypes = []
vehTypes.append(VehicleType((0, 1, 0), 'smart'))
vehTypes.append(VehicleType((1, 0, 0), 'dumb'))
numCars = 100
starts = Demand._def_rand_departs(numCars, 2)
prob_smart = .5
vehicles = []
dumbcars = []
for x in [('%s' % i, random.random(), starts[i]) for i in range(numCars)]:
if x[1] < prob_smart:
vType = vehTypes[0]
else:
vType = vehTypes[1]
dumbcars.append(x[0])
route = Route(['0_1', '1_2'])
vehicles.append(Vehicle(x[0], x[2], vType, route))
dem = Demand(vehTypes, vehicles)
dem.genRouteFile('simple_2d\\2d.rou.xml')
sensorList = []
lane = '1_2_0'
sensorList.append(Fusion.Sensor('first', 75, lane, [.25], .95))
sensorList.append(Fusion.Sensor('second', 200, lane, [.25], .95))
fus = Fusion.FusionArchitecture(sensorList, None, None)
runSim.runSim('simple_2d\\2d.net.xml',
'simple_2d\\2d.rou.xml',
None,
fusion=fus)
def mapa(request):
idProduto = request.GET.get("idProduto", None)
markedSectors = []
if idProduto is not None:
urlSetoresProduto = 'https://scan-skip-plu-teste.herokuapp.com/setoresProduto/' + idProduto
network = Network()
response = network.request(urlSetoresProduto, 'GET')
if response != -1:
for setor in response:
markedSectors.append(setor['idSetor'])
return render(request, 'mapa.html', {
'markedSectors': markedSectors,
'sectorsRange': range(1, 53)
})
def produto(request, idProduto):
urlProduto = 'https://supermercado-carrinho.herokuapp.com/produtos/TesteProduto/' #'http://143.107.102.48:3000/produto/' + idProduto
network = Network()
response = network.request(urlProduto, 'GET')
nome = response['nome']
marca = response['marca']
preco = response['valor']
preco = ('%.2f' % (float(preco) / 100)).replace('.', ',')
categoria = response['categoria']
imagem = response['imagem']
return render(
request, 'produto-escaneado-confirmacao.html', {
'idProduto': idProduto,
'nome': nome,
'marca': marca,
'preco': preco,
'categoria': categoria,
'imagem': imagem
})
def create_population(self, count):
"""Create a population of random networks.
Args:
count (int): Number of networks to generate, aka the
size of the population
Returns:
(list): Population of network objects
"""
pop = []
for _ in range(0, count):
# Create a random network.
network = Network(self.nn_param_choices)
network.create_random()
# Add the network to our population.
pop.append(network)
return pop
def produto(request, idProduto):
urlProduto = 'https://scan-skip-plu-teste.herokuapp.com/produto/' + idProduto
network = Network()
response = network.request(urlProduto, 'GET')
if response != -1:
quantidade = 1
nome = response['nome']
marca = response['marca']
preco = response['valor']
preco = ('%.2f' % (float(preco)/100)).replace('.', ',')
categoria = response['categoria']
imagem = response['imagem']
# Salvando ultimo produto escaneado na sessão
request.session['lastScannedProduct'] = idProduto
return render(request, 'produto-escaneado-confirmacao.html', {'idProduto': idProduto, 'nome': nome, 'marca': marca, 'preco': preco, 'categoria': categoria, 'imagem': imagem, 'quantidade': quantidade, 'registrado': True})
else:
nome = 'Produto não registrado'
marca = '-------'
preco = '-------'
categoria = '-------'
return render(request, 'produto-escaneado-confirmacao.html', {'idProduto': idProduto, 'nome': nome, 'marca': marca, 'preco': preco, 'categoria': categoria, 'registrado': False})
def main(arguments):
print("Loading client...")
try:
if len(arguments) != globalsettings.ARGUMENTS_LENGTH:
raise Exception(f"required 3 arguments received {len(arguments)}.")
client_frame = window_frame(
client_app,
"uSpeak - client",
GetSystemMetrics(0) / 2 - globalsettings.CLIENT_SCREEN_WIDTH / 2,
GetSystemMetrics(1) / 2 - globalsettings.CLIENT_SCREEN_HEIGHT / 2,
globalsettings.CLIENT_SCREEN_WIDTH,
globalsettings.CLIENT_SCREEN_HEIGHT,
stylesheet='',
resizable=False)
chat_box = text_field(client_frame, 10, 32, '')
chat_box.setReadOnly(True)
chat_box.resize(470, 350)
userlist = QListWidget(client_frame)
userlist.move(520, 32)
userlist.resize(100, 120)
channellist = QListWidget(client_frame)
channellist.itemClicked.connect(
lambda item: event.set_selected_channel(item))
channellist.move(520, 160)
channellist.resize(100, 200)
chat_input = input_field(client_frame, 10, 387,
'Message #selected channel')
chat_input.resize(470, 30)
message_button = button(
client_frame,
"Send message", 520, 387, lambda: client.send_message(
globalsettings.USER_MESSAGE, event.selected_channel,
event.get_input_text()))
chat_input.returnPressed.connect(message_button.click)
exitAct = QAction(QIcon(''), '&Disconnect', client_frame)
exitAct.triggered.connect(os._exit)
menubar = client_frame.menuBar()
server_menu = menubar.addMenu('Server')
action = server_menu.addAction(exitAct)
event = Event(chat_box, chat_input, userlist, channellist)
client = Network(arguments[0], int(arguments[1]), arguments[2], event)
client.establish_connection()
time.sleep(0.3)
event.set_selected_channel(channellist.item(0))
client_frame.gui_show()
except Exception as exception:
print(f"{exception}")
print('test')
exit()
except IndexError:
print(f"Required 3 arguments received {len(arguments)}.")
exit()
except:
pass
Node('6', 0, 225),
Node('7', 0, 250)
]
edges = [
Edge(nodes[0], nodes[1]),
Edge(nodes[1], nodes[2]),
Edge(nodes[1], nodes[3]),
Edge(nodes[2], nodes[4], speed=5),
Edge(nodes[3], nodes[5], speed=2),
Edge(nodes[4], nodes[6]),
Edge(nodes[5], nodes[6]),
Edge(nodes[6], nodes[7], numLanes=2)
]
newNet = Network(nodes, edges)
sumo_path = 'C:\\dev\\Traffic\\Sumo\\bin'
os.makedirs('oneway_data', exist_ok=True)
newNet.writeNet(sumo_path, 'oneway_data\\oneway')
vehTypes = []
vehTypes.append(VehicleType((0, 1, 0), 'smart'))
vehTypes.append(VehicleType((1, 0, 0), 'dumb'))
numCars = 100
starts = Demand._def_rand_departs(numCars, 2)
prob_smart = .5
vehicles = []
ap.add_argument("-APRILTAG_16h5", "--DICT_APRILTAG_16h5", required=False, action="store_true", default=False, help="Include DICT_APRILTAG_16h5 into dataset.")
ap.add_argument("-APRILTAG_25h9", "--DICT_APRILTAG_25h9", required=False, action="store_true", default=False, help="Include DICT_APRILTAG_25h9 into dataset.")
ap.add_argument("-APRILTAG_36h10", "--DICT_APRILTAG_36h10", required=False, action="store_true", default=False, help="Include DICT_APRILTAG_36h10 into dataset.")
ap.add_argument("-APRILTAG_36h11", "--DICT_APRILTAG_36h11", required=False, action="store_true", default=False, help="Include DICT_APRILTAG_36h11 into dataset.")
args = vars(ap.parse_args())
# Verify the passed parameters
if not isinstance(args["shape"], tuple) or len(args["shape"]) != 3:
raise Exception("Shape parameter is invalid. Should be something like '(256,256,3)'.")
if not isinstance(args["learning_rate"], float) or args["learning_rate"] <= 0.0:
raise Exception("Learning rate parameter is invalid. Should be a float bigger than '0.0'.")
if not isinstance(args["iterations"], int) or args["iterations"] < 1:
raise Exception("Iterations has an invalid value.")
if not isinstance(args["steps"], int) or args["steps"] < 1:
raise Exception("Steps argument has an invalid value.")
if not isinstance(args["batch_size"], int) or args["batch_size"] < 1:
raise Exception("Batch size has an invalid value.")
if not os.path.isdir(os.path.dirname(args["weights"])):
raise Exception("Path to store weights is invalid.")
if not isinstance(args["saving_iterations"], int) or args["saving_iterations"] < 1:
raise Exception("Saving iterations has an invalid value.")
if not (args["DICT_4X4_50"] or args["DICT_4X4_100"] or args["DICT_4X4_250"] or args["DICT_4X4_1000"] or args["DICT_5X5_50"] or args["DICT_5X5_100"] or args["DICT_5X5_250"] or args["DICT_5X5_1000"] or args["DICT_6X6_50"] or args["DICT_6X6_100"] or args["DICT_6X6_250"] or args["DICT_6X6_1000"] or args["DICT_7X7_50"] or args["DICT_7X7_100"] or args["DICT_7X7_250"] or args["DICT_7X7_1000"] or args["DICT_ARUCO_ORIGINAL"] or args["DICT_APRILTAG_16h5"] or args["DICT_APRILTAG_25h9"] or args["DICT_APRILTAG_36h10"] or args["DICT_APRILTAG_36h11"]):
raise Exception("At least one marker family need to be enabled.")
# Initalize the segnet network
network = Network(shape=args["shape"], learning_rate=args["learning_rate"])
# Load checkpoint if is setted
if args["checkpoint"] != '':
network.load_weights(weights_path=args["weights"])
# Start training
network.train_synthetic_data(iterations=args["iterations"], steps=args["steps"], batch_size=args["batch_size"], weights_path=args["weights"], saving_iterations=args["saving_iterations"], DICT_4X4_50=args["DICT_4X4_50"], DICT_4X4_100=args["DICT_4X4_100"], DICT_4X4_250=args["DICT_4X4_250"], DICT_4X4_1000=args["DICT_4X4_1000"], DICT_5X5_50=args["DICT_5X5_50"], DICT_5X5_100=args["DICT_5X5_100"], DICT_5X5_250=args["DICT_5X5_250"], DICT_5X5_1000=args["DICT_5X5_1000"], DICT_6X6_50=args["DICT_6X6_50"], DICT_6X6_100=args["DICT_6X6_100"], DICT_6X6_250=args["DICT_6X6_250"], DICT_6X6_1000=args["DICT_6X6_1000"], DICT_7X7_50=args["DICT_7X7_50"], DICT_7X7_100=args["DICT_7X7_100"], DICT_7X7_250=args["DICT_7X7_250"], DICT_7X7_1000=args["DICT_7X7_1000"], DICT_ARUCO_ORIGINAL=args["DICT_ARUCO_ORIGINAL"], DICT_APRILTAG_16h5=args["DICT_APRILTAG_16h5"], DICT_APRILTAG_25h9=args["DICT_APRILTAG_25h9"], DICT_APRILTAG_36h10=args["DICT_APRILTAG_36h10"], DICT_APRILTAG_36h11=args["DICT_APRILTAG_36h11"])
help="Define the shape of a tile (default=(256,256,3))).")
ap.add_argument("-w",
"--weights",
required=False,
default='./weights/weights.ckpt',
help="Path to the weights (default='./weights/weights.ckpt').")
ap.add_argument("-o",
"--output",
required=True,
help="Path to save prediction.")
args = vars(ap.parse_args())
# Verify the passed parameters
if not os.path.isfile(args["image"]):
raise Exception("Path to image is invalid.")
if not isinstance(args["shape"], tuple) or len(args["shape"]) != 3:
raise Exception(
"Shape parameter is invalid. Should be something like '(256,256,3)'.")
if not os.path.isdir(os.path.dirname(args["weights"])):
raise Exception("Path to weights is invalid.")
# Load the image
image = cv2.cvtColor(cv2.imread(args["image"], 3), cv2.COLOR_BGR2RGB)
# Initalize the segnet network
network = Network(shape=args["shape"])
# Load the weights
network.load_weights(weights_path=args["weights"])
# Start prediction and save the results
prediction = network.predict(image)
cv2.imwrite(args["output"], cv2.cvtColor(prediction, cv2.COLOR_RGB2BGR))
dataset = []
index_count = 0
current_sample = args["dataset_path"] + "/" + str(index_count) + ".jpg"
current_sample_gt = args["dataset_path"] + "/" + str(index_count) + "_gt.jpg"
while os.path.isfile(current_sample) and os.path.isfile(current_sample_gt):
dataset.append([current_sample, current_sample_gt])
index_count += 1
current_sample = args["dataset_path"] + "/" + str(index_count) + ".jpg"
current_sample_gt = args["dataset_path"] + "/" + str(
index_count) + "_gt.jpg"
if len(dataset) <= 0:
raise Exception(
"Cannot finde training samples in the directory " +
args["dataset_path"] +
".\r\n To get more information take a look at the original github repo."
)
# Initalize the segnet network
network = Network(shape=args["shape"], learning_rate=args["learning_rate"])
# Load checkpoint if is setted
if args["checkpoint"] != '':
network.load_weights(weights_path=args["weights"])
# Start training
network.train_real_data(iterations=args["iterations"],
steps=args["steps"],
batch_size=args["batch_size"],
weights_path=args["weights"],
saving_iterations=args["saving_iterations"],
dataset=dataset)
def main(): server_settings = config.load_config(globalsettings.CONFIG_FILE_LOCATION) server = Network(server_settings['hostname'], server_settings['port'], server_settings['max_users'], server_settings['rcon'], server_settings['channels']) server.start()
# Acquisition of Data: FOV, Cropped Original Image and Whole Image
print('Data Generation')
print(time.strftime("%Y-%m-%d %H:%M:%S", time.localtime()))
Preprocessing = preprocessing.Data(options.train_image, options.train_label,
options.train_probability_map, C)
FOV_Path, LABEL_Path, IMAGE_Path, FOV_size, Fov_num = Preprocessing.create_data(
options.path)
print('Data Generation finished')
print(time.strftime("%Y-%m-%d %H:%M:%S", time.localtime()))
# Input of 2 models
FOV_input = Input(shape=FOV_size)
Stack_input = Input(shape=FOV_size)
# Network of two CNNs
Net = Network(options.network, options.structure)
First_Network_Output = Net.get_first_network(FOV_input)
Second_Network_Output = Net.get_second_network(Stack_input)
# Build the model
First_Model = Model(FOV_input, First_Network_Output)
Second_Model = Model(Stack_input, Second_Network_Output)
# print(First_Model.summary())
# load the weight
if options.input_weight_path:
First_Model.load_weights(options.input_weight_path[0], by_name=True)
Second_Model.load_weights(options.input_weight_path[1], by_name=True)
# Complie the model
optimizer = RMSprop(lr=1e-5)