def main():
##########################
# CONFIG
##########################
seeds = [11,7,13,19,5189,600,1000,1500,1500,1500, 1000]
size = [50,100,250,500,1000,1500,2000,2500,3000,3500,7000]
epcohs = []
##########################
##########################
##########################
# RUN THE EXPERIMENTS
##########################
# Throw the experiments
for exp in range(5):
print(f">>>> EXPERIMENTO: {exp + 1} <<<<")
results = []
for i,seed in enumerate(seeds):
np.random.seed(seed)
vetor = gera_seq_aleatoria(size[i])
cron = Cronometro()
cron.Iniciar()
conta_somas(vetor)
cron.Parar()
results.append(cron.Exibir())
print(f"Tempo gasto com {size[i]} elementos foi {cron} segundos")
del vetor
del cron
epcohs.append(results)
print()
##########################
##########################
##########################
# PLOT THE RESULTS
##########################
# Generate the scatter X and Y
scatter_x = size * len(epcohs)
scatter_y = np.concatenate(epcohs)
# Create a polinomial function of 1 dimension
fit_fn = np.poly1d(np.polyfit(scatter_x,scatter_y,1))
plt.title("Resultados")
# PLOT: Linear Regression
line, = plt.plot(scatter_x, fit_fn(scatter_x), '--', c="#0c29d0")
line.set_dashes([10,5,10,5])
# PLOT: Poly Regression
plt.plot(size, np.mean(epcohs, axis=0), c="#ff5900", label="Regressão Poly")
# PLOT: STM
plt.fill_between(size, np.min(epcohs, axis=0), np.max(epcohs, axis=0), alpha=.3, facecolor="#ff5900", label="Desvio")
# PLOT: Scatter
plt.scatter(scatter_x, scatter_y, c="#0c29d0", label="Tempos")
plt.legend(['Tendência', 'Regressão Polinomial', 'Desvio', 'Tempos'])
plt.show()
