example.py

import sys
import math
import glob
import os
import matplotlib.pyplot as plt
import TrafficModel as tfdm
from tqdm import tqdm
import matplotlib
matplotlib.use('TkAgg')

'''
Example script to show how to use the TrafficModel package
The script can be run with the following arguments:
  - constant:     constant traffic flow
  - peaked:       peaked traffic flow
  - periodic:     periodic traffic flow
  - traveltime:   traveltime histogram
These analyses are described in the report.
'''


# utility
def clear():
    '''
    Clear the temp_data folder
    '''
    files = glob.glob("./temp_data/*")
    for f in files:
        os.remove(f)


# utility
def plot():
    '''
    Plot the k distribution with respect to time
    '''
    file = open("./temp_data/k-t.dat", "r")
    lines = file.readlines()
    file.close()
    x = []
    y = []
    for line in lines:
        x.append(float(line.split()[0]))
        y.append(float(line.split()[1]))
    plt.plot(x, y)
    plt.show()


# utility
def hist(filename):
    '''
    Plot a histogram (used in the traveltime analysis)
    '''
    file = open(filename, "r")
    lines = file.readlines()
    file.close()
    x = []
    y = []
    for line in lines:
        x.append(float(line.split()[0]))
        y.append(float(line.split()[1]))
    plt.bar(x, y)
    plt.show()


# Constant analysis
def constant():
    '''
    Run the constant analysis. For more information, see the report.
    Many vehicles are added to the network constantly.
    The k distribution is plotted every 25 time steps.
    '''
    clear()
    graph = tfdm.Graph("./data/matrix.dat")
    tfdm.Vehicle.addVehicleType("./data/vehicletype.dat")
    graph.setTemperature(300)
    graph.updateTransMatrix()
    for t in tqdm(range(15001)):
        if t % 25 == 0:
            graph.fprintTimeDistribution("./temp_data/", "k", 0.)
        if (t < 12e3) and (t % 60 == 0) and (t != 0):
            graph.addVehiclesUniformly(250)
        graph.evolve(False)
    plot()


# Peaked analysis
def peaked():
    '''
    Run the peaked analysis. For more information, see the report.
    Many vehicles are added to the network at the beginning, in a short time.
    The k distribution is plotted every 100 time steps.
    '''
    clear()
    graph = tfdm.Graph("./data/matrix.dat")
    tfdm.Vehicle.addVehicleType("./data/vehicletype.dat")
    graph.setTemperature(300)
    graph.updateTransMatrix()
    for t in tqdm(range(12000)):
        if t % 100 == 0:
            graph.fprintTimeDistribution("./temp_data/", "k", 0.)
        if (t < 500) and (t % 50 == 0) and (t != 0):
            graph.addVehiclesUniformly(2500)
        if t < 0.4e4:
            graph.evolve()
        else:
            graph.evolve(False)
    plot()


# Periodic analysis
def periodic():
    '''
    Run the periodic analysis. For more information, see the report.
    Vehicles are added to the network periodically, with three peaks per day.
    The k distribution is plotted every 900 time steps.
    '''
    clear()
    graph = tfdm.Graph("./data/matrix.dat")
    tfdm.Vehicle.addVehicleType("./data/vehicletype.dat")
    graph.setTemperature(300)
    graph.updateTransMatrix()
    for t in tqdm(range(55000)):
        if t % 900 == 0:
            graph.fprintTimeDistribution("./temp_data/", "k", 0.)
        if (t % 300 == 0) and (t < 32400 * 1.5):
            vehicles = abs(2200 * math.sin(2 * math.pi * t / 32400))
            if (t > 1.62e4) and (t < 2.43e4):
                vehicles = vehicles / 1.125
            graph.addVehiclesUniformly(int(vehicles))
        graph.evolve(False)
    plot()


# Traveltime analysis
def traveltime():
    '''
    Run the traveltime analysis. For more information, see the report.
    Vehicles are added constantly to the network.
    The traveltime histogram is plotted every 3000 time steps.
    '''
    clear()
    graph = tfdm.Graph("./data/matrix.dat")
    tfdm.Vehicle.addVehicleType("./data/vehicletype.dat")
    graph.setTemperature(300)
    graph.updateTransMatrix()
    for t in tqdm(range(15001)):
        if t % 100 == 0:
            graph.fprintHistogram("./temp_data/", "traveltime", 100, "latex")
        if (t % 60 == 0) and (t != 0):
            graph.addRndmVehicles(200)
            graph.addVehiclesUniformly(int(200 / 1.125))
        graph.evolve(False)
    hist("./temp_data/3000_t.dat")
    hist("./temp_data/6000_t.dat")
    hist("./temp_data/9000_t.dat")
    hist("./temp_data/12000_t.dat")
    hist("./temp_data/15000_t.dat")


if __name__ == "__main__":
    if len(sys.argv) > 1:
        todo = sys.argv[1].lower()
        if todo == "constant":
            constant()
        elif todo == "peaked":
            peaked()
        elif todo == "periodic":
            periodic()
        elif todo == "traveltime":
            traveltime()