def test_noise(self):
x = K.placeholder((2, 3))
f1 = N.Noise(level=0.5, noise_dims=0, noise_type='gaussian')
y = f1(x)
f = K.function(x, y, defaults={K.is_training(): True})
z = f(np.ones((2, 3)))
z = z.tolist()
self.assertTrue(all(i == z[0] for i in z))
f1 = N.Noise(level=0.5, noise_dims=1, noise_type='gaussian')
y = f1(x)
f = K.function(x, y, defaults={K.is_training(): True})
z = f(np.ones((2, 3)))
z = z.T.tolist()
self.assertTrue(all(i == z[0] for i in z))
def test_seq(self):
X = K.placeholder((None, 28, 28, 1))
f = N.Sequence([
N.Conv(8, (3, 3), strides=1, pad='same'),
N.Dimshuffle(pattern=(0, 3, 1, 2)),
N.Flatten(outdim=2),
N.Noise(level=0.3, noise_dims=None, noise_type='gaussian'),
N.Dense(128, activation=tf.nn.relu),
N.Dropout(level=0.3, noise_dims=None),
N.Dense(10, activation=tf.nn.softmax)
])
y = f(X)
yT = f.T(y)
f1 = K.function(X, y, defaults={K.is_training(): True})
f2 = K.function(X, yT, defaults={K.is_training(): False})
f = cPickle.loads(cPickle.dumps(f))
y = f(X)
yT = f.T(y)
f3 = K.function(X, y, defaults={K.is_training(): True})
f4 = K.function(X, yT, defaults={K.is_training(): False})
x = np.random.rand(12, 28, 28, 1)
self.assertEquals(f1(x).shape, (2688, 10))
self.assertEquals(f3(x).shape, (2688, 10))
self.assertEqual(np.round(f1(x).sum(), 4), np.round(f3(x).sum(), 4))
self.assertEquals(y.shape.as_list(), (None, 10))
self.assertEquals(f2(x).shape, (12, 28, 28, 1))
self.assertEquals(f4(x).shape, (12, 28, 28, 1))
self.assertEqual(str(f2(x).sum())[:4], str(f4(x).sum())[:4])
self.assertEquals(yT.shape.as_list(), (None, 28, 28, 1))
def create():
f = N.Sequence([
N.Conv(8, (3, 3), strides=1, pad='same'),
N.Dimshuffle(pattern=(0, 3, 1, 2)),
N.FlattenLeft(outdim=2),
N.Noise(level=0.3, noise_dims=None, noise_type='gaussian'),
N.Dense(128, activation=K.relu),
N.Dropout(level=0.3, noise_dims=None),
N.Dense(10, activation=K.softmax)
],
debug=True)
y = f(X)
yT = f.T(y)
f1 = K.function(X, y)
f2 = K.function(X, yT)
cPickle.dump(f, open(U.get_modelpath('dummy.ai', override=True), 'w'))
_ = f1(x)
print(_.shape, _.sum())
_ = f2(x)
print(_.shape, _.sum())
N.BatchNorm(),
N.Conv(num_filters=64, filter_size=3, pad='same'),
N.BatchNorm(),
N.Conv(num_filters=64, filter_size=3, pad='valid'),
N.BatchNorm(activation=tf.nn.elu),
N.Pool(),
N.Flatten(outdim=2),
N.Dense(num_units=args.dim)
], debug=True, name='EncoderNetwork')
f_decoder = N.Sequence([
N.Dropout(level=LATENT_DROPOUT, noise_type='uniform'),
N.Noise(level=1.0, noise_type='gaussian'),
N.Dimshuffle((0, 'x', 'x', 1)),
N.TransposeConv(num_filters=64, filter_size=3, pad='valid'),
N.Upsample(size=2, axes=(1, 2)),
N.BatchNorm(),
N.TransposeConv(num_filters=64, filter_size=3, pad='same'),
N.BatchNorm(),
N.TransposeConv(num_filters=32, filter_size=3, pad='valid'),
N.Upsample(size=2, axes=(1, 2),
desire_shape=None if is_cifar10 else (None, 14, 14, None)),
N.BatchNorm(),
N.TransposeConv(num_filters=3 if is_cifar10 else 1,