def run_autoencoder():
imfiles = glob.glob('cropped/*.pgm')
X = np.vstack([np.array(Image.open(fname).convert('L')).flatten()
for fname in imfiles])
# digits = datasets.load_digits()
# X = np.asarray(digits.data, 'float32')
rbm = train_rbm(X, n_components=100, n_iter=5)
images = [comp.reshape(192, 168) for comp in rbm.components_]
images = images[:10]
hlp.smartshow(images)
def run_autoencoder():
imfiles = glob.glob('cropped/*.pgm')
X = np.vstack([
np.array(Image.open(fname).convert('L')).flatten() for fname in imfiles
])
# digits = datasets.load_digits()
# X = np.asarray(digits.data, 'float32')
rbm = train_rbm(X, n_components=100, n_iter=5)
images = [comp.reshape(192, 168) for comp in rbm.components_]
images = images[:10]
hlp.smartshow(images)
def run_auto():
X = load_data('gender/male')
X = X.astype(np.float32) / 256
rbm = BernoulliRBM(random_state=0, verbose=True)
rbm.learning_rate = 0.06
rbm.n_iter = 20
rbm.n_components = 2000
rbm.fit(X)
cimgs = [comp.reshape(100, 100) for comp in rbm.components_]
smartshow(cimgs[:12])
return rbm
def run_deep():
X = load_data('gender/male')
X = X.astype(np.float32) / 256
rbm1 = BernoulliRBM(n_components=10000, n_iter=20, learning_rate=0.05,
random_state=0, verbose=True)
rbm2 = BernoulliRBM(n_components=200, n_iter=20, learning_rate=0.05,
random_state=0, verbose=True)
model = Pipeline(steps=[('rbm1', rbm1), ('rbm2', rbm)])
model.fit(X)
cimgs = [comp.reshape(100, 100) for comp in model.steps[1][1].components_]
smartshow(cimgs[:12])
return model