size_batch = 1000
competition = AdversarialCompetition(
size_batch=size_batch,
true_model=GenerativeNormalModel(1, 2),
discriminative=pipeline.make_pipeline(
preprocessing.PolynomialFeatures(4),
linear_model.LogisticRegression()),
generative=GenerativeNormalModel(
0, 1, updates=["mu", "sigma"]),
gradient_descent=GradientDescent(
0.03, inertia=0.0, annealing=100),
)
print(competition)
for i in range(200):
if i % 50 == 0:
competition.plot()
plt.show()
pass
competition.iteration()
print("final model %s" % competition.generatives[-1])
competition.plot_params()
plt.show()
competition.plot_auc()
plt.show()
np.random.seed(0)
size_batch = 1000
competition = AdversarialCompetition(
size_batch=size_batch,
true_model=GenerativeNormalMixtureModel([-3, 3], [1, 1]),
discriminative=pipeline.make_pipeline(preprocessing.PolynomialFeatures(4),
linear_model.LogisticRegression()),
generative=GenerativeNormalMixtureModel([-1, 1], [1, 1],
updates=["mu", "sigma"]),
gradient_descent=GradientDescent(0.01, 0.5),
)
print(competition)
for i in range(1000):
if i % 200 == 0:
competition.plot()
plt.show()
pass
competition.iteration()
print("final model %s" % competition.generatives[-1])
competition.plot_params()
plt.show()
competition.plot_auc()
plt.show()
discriminative=pipeline.make_pipeline(
preprocessing.PolynomialFeatures(4),
linear_model.LogisticRegression()),
generative=GenerativeNormalQuasiMixtureModel(
2.5,2.1,updates=["mu", "sigma"]),
gradient_descent=GradientDescent(
0.03, inertia=0.0, annealing=100),
x_dataset = true_data
)
print(competition)
for i in range(1001):
if i % 50 == 0:
plt.figure()
competition.plot()
pyplot.savefig('Pooling.png')
pyplot.close()
plt.figure()
competition1.plot()
competition2.plot()
pyplot.savefig('Separating.png')
pyplot.close()
pass
competition.iteration()
competition1.iteration()
competition2.iteration()
print("final model pooling %s" % competition.generatives[-1])
print("final model separating %s" % competition1.generatives[-1], competition2.generatives[-1])