def do_square(self, arg):
"""Draw Square: square 100"""
self.results.writeToFile("Drawing a square")
self.parse(arg)
turtle_drawer = TurtleDrawer()
directions = [0, 90, 180, 270]
for i in directions:
turtle_drawer.draw_line(i, self.data)
def do_S(self, arg):
"""Draw line 120 degrees : S 100"""
self.results.writeToFile("Draw line 120 degrees : S", arg)
self.parse(arg)
turtle_drawer = TurtleDrawer()
turtle_drawer.draw_line(180, self.data)
def do_W(self, arg):
"""Draw line 270 degrees : W 100"""
self.results.writeToFile("Draw line 270 degrees : W", arg)
self.parse(arg)
turtle_drawer = TurtleDrawer()
turtle_drawer.draw_line(270, self.data)
def do_E(self, arg):
"""Draw line 90 degrees : E 100"""
self.results.writeToFile("Draw line 90 degrees : E", arg)
self.parse(arg)
turtle_drawer = TurtleDrawer()
turtle_drawer.draw_line(90, self.data)
def do_P(self, arg):
"""Select Pen: P 10"""
self.results.writeToFile("Selected pen", arg)
self.parse(arg)
turtle_drawer = TurtleDrawer()
turtle_drawer.select_pen(self.data)
def setUp(self):
self.Reader = ArgumentSourceReader(ArgumentParser(TurtleDrawer()))
self.source = ['-e', '-t', '-k', '-g']
def setUp(self):
self.tkinterParser = ArgumentParser(TurtleDrawer())
self.source = [
'-c', '-e', '-t', '-k', '-g', '1', '2', '3', '4', '5', '6', '7',
'8', '9', '0'
]
def main():
# Generate random population of chromosomes
population = [] # Chromosomes are appended in this
for i in range(POPULATION_SIZE):
# Every chromosome is a list of numbers
chromosome = list(range(array_size))
# ...randomized/shuffled.
random.shuffle(chromosome)
# Add to population
population.append(chromosome)
turtle_drawer = TurtleDrawer()
plot_drawer = PlotDrawer()
# Result arrays for plotting
generation_indices = []
average_results = []
min_results = []
max_results = []
# Keep track of the last max_chromosome to avoid redundancy
previous_max_chromosome = []
for gen_count in range(NUMBER_OF_GENERATIONS):
# Calculate fitness from distance and flow using imported function
fitness_scores = get_fitness_scores(population, distance_array,
flow_array)
# Normalize fitness score by inverting and dividing by the total
fitness_scores_normalized = normalise_fitness_scores(fitness_scores)
# While it's not normalized yet, max is the worst, therefore min is the most suitable
# Use numpy to get min, max, and average
max_fitness = np.min(fitness_scores)
min_fitness = np.max(fitness_scores)
average_fitness = np.mean(fitness_scores)
max_chromosome = population[np.argmin(fitness_scores)]
max_chromosome = list(map(lambda value: value + 1, max_chromosome))
# Use the imported strategy functions to select chromosomes from
# population based on the normalized scores
selected_chromosomes = selection_strategy.start_tournament(
population, fitness_scores_normalized)
# Input the selected chromosomes to the imported crossover function
crossed_chromosomes = crossover_strategy.start_crossover(
selected_chromosomes)
# Apply mutation over the crossever chromosomes
mutated_chromosomes = mutation_strategy.start_mutation(
crossed_chromosomes)
# Print current values to the terminal
print(
"Generation_Count: \t\t\t\t{}\nMean fitness.: \t\t\t{}\nMax score: \t\t\t{}\nMax chromosome.: \t\t{}\n\n"
.format(gen_count, average_fitness, max_fitness, max_chromosome))
# DRAWING FRAME
# Implementing the turtle functions and iteratively drawing
def draw_visual_frame():
# The function takes flow and distance arrays, the best chromosome,
# and other data and draws points and lines representing flow and distance
turtle_drawer.draw_main_frame(max_chromosome, gen_count,
max_fitness, max_chromosome,
flow_array, distance_array)
# Delay between each iteration of drawing
time.sleep(SLEEP_TIME)
return
if previous_max_chromosome != max_chromosome:
draw_visual_frame()
# Update the max chromosome
previous_max_chromosome = max_chromosome
# Update the population to the one generated by the mutation strategy
population = crossed_chromosomes
# For plot drawing
# ----------------
# Add fitness scores to the results arrays, to be used in plot drawing
max_results.append(max_fitness)
min_results.append(min_fitness)
average_results.append(average_fitness)
# Add an increment to the generation indices, for the x-axis of the plot
generation_indices.append(gen_count)
# wait for 2 seconds between consective output generation
# time.sleep(2)
# Draw and save a plot at the first generation, and every 1/5ths of the total generations after that,
# and then at the very last generation.
if gen_count == 0 or gen_count % (
NUMBER_OF_GENERATIONS /
5) == 0 or gen_count == NUMBER_OF_GENERATIONS - 1:
plot_drawer.drawPlot("Plot_For_" + 'Gen_' + str(gen_count),
generation_indices,
average_results,
max_results,
min_results,
path='Graphs')