def test_lambda():
layer_test(layers.Lambda,
kwargs={'function': lambda x: x + 1},
input_shape=(3, 2))
layer_test(layers.Lambda,
kwargs={
'function': lambda x, a, b: x * a + b,
'arguments': {
'a': 0.6,
'b': 0.4
}
},
input_shape=(3, 2))
def antirectifier(x):
x -= K.mean(x, axis=1, keepdims=True)
x = K.l2_normalize(x, axis=1)
pos = K.relu(x)
neg = K.relu(-x)
return K.concatenate([pos, neg], axis=1)
def antirectifier_output_shape(input_shape):
shape = list(input_shape)
assert len(shape) == 2 # only valid for 2D tensors
shape[-1] *= 2
return tuple(shape)
layer_test(layers.Lambda,
kwargs={
'function': antirectifier,
'output_shape': antirectifier_output_shape
},
input_shape=(3, 2))
# test serialization with function
def f(x):
return x + 1
ld = layers.Lambda(f)
config = ld.get_config()
ld = deserialize_layer({'class_name': 'Lambda', 'config': config})
# test with lambda
ld = layers.Lambda(
lambda x: K.concatenate([K.square(x), x]),
output_shape=lambda s: tuple(list(s)[:-1] + [2 * s[-1]]))
config = ld.get_config()
ld = layers.Lambda.from_config(config)
# test serialization with output_shape function
def f(x):
return K.concatenate([K.square(x), x])
def f_shape(s):
return tuple(list(s)[:-1] + [2 * s[-1]])
ld = layers.Lambda(f, output_shape=f_shape)
config = ld.get_config()
ld = deserialize_layer({'class_name': 'Lambda', 'config': config})
def test_lambda():
layer_test(layers.Lambda,
kwargs={'function': lambda x: x + 1},
input_shape=(3, 2))
layer_test(layers.Lambda,
kwargs={'function': lambda x, a, b: x * a + b,
'arguments': {'a': 0.6, 'b': 0.4}},
input_shape=(3, 2))
# test serialization with function
def f(x):
return x + 1
ld = layers.Lambda(f)
config = ld.get_config()
ld = deserialize_layer({'class_name': 'Lambda', 'config': config})
# test with lambda
ld = layers.Lambda(
lambda x: K.concatenate([K.square(x), x]),
output_shape=lambda s: tuple(list(s)[:-1] + [2 * s[-1]]))
config = ld.get_config()
ld = layers.Lambda.from_config(config)
# test serialization with output_shape function
def f(x):
return K.concatenate([K.square(x), x])
def f_shape(s):
return tuple(list(s)[:-1] + [2 * s[-1]])
ld = layers.Lambda(f, output_shape=f_shape)
config = ld.get_config()
ld = deserialize_layer({'class_name': 'Lambda', 'config': config})
def from_config(cls, config, custom_objects=None):
if custom_objects is None:
custom_objects = {'LSTMPeephole': LSTMPeephole}
from keras.layers import deserialize as deserialize_layer
recurrent_layer = deserialize_layer(config.pop('recurrent_layer'),
custom_objects=custom_objects)
dense_layer = deserialize_layer(config.pop('dense_layer'),
custom_objects=custom_objects)
return cls(recurrent_layer, dense_layer, **config)
def test_lambda():
layer_test(layers.Lambda,
kwargs={'function': lambda x: x + 1},
input_shape=(3, 2))
layer_test(layers.Lambda,
kwargs={'function': lambda x, a, b: x * a + b,
'arguments': {'a': 0.6, 'b': 0.4}},
input_shape=(3, 2))
def antirectifier(x):
x -= K.mean(x, axis=1, keepdims=True)
x = K.l2_normalize(x, axis=1)
pos = K.relu(x)
neg = K.relu(-x)
return K.concatenate([pos, neg], axis=1)
def antirectifier_output_shape(input_shape):
shape = list(input_shape)
assert len(shape) == 2 # only valid for 2D tensors
shape[-1] *= 2
return tuple(shape)
layer_test(layers.Lambda,
kwargs={'function': antirectifier,
'output_shape': antirectifier_output_shape},
input_shape=(3, 2))
# test serialization with function
def f(x):
return x + 1
ld = layers.Lambda(f)
config = ld.get_config()
ld = deserialize_layer({'class_name': 'Lambda', 'config': config})
# test with lambda
ld = layers.Lambda(
lambda x: K.concatenate([K.square(x), x]),
output_shape=lambda s: tuple(list(s)[:-1] + [2 * s[-1]]))
config = ld.get_config()
ld = layers.Lambda.from_config(config)
# test serialization with output_shape function
def f(x):
return K.concatenate([K.square(x), x])
def f_shape(s):
return tuple(list(s)[:-1] + [2 * s[-1]])
ld = layers.Lambda(f, output_shape=f_shape)
config = ld.get_config()
ld = deserialize_layer({'class_name': 'Lambda', 'config': config})
def from_config(cls, config, custom_objects=None):
from keras.layers import deserialize as deserialize_layer
dna_layer = deserialize_layer(config.pop('layer'),
custom_objects=custom_objects)
layer = cls(dna_layer, **config)
return layer
def from_config(cls, config, custom_objects=None):
cell = deserialize_layer(config.pop('grnn_cell'),
custom_objects=custom_objects)
num_constants = config.pop('num_constants', None)
layer = cls(cell, **config)
layer._num_constants = num_constants
return layer
def from_config(cls, config, custom_objects=None):
from keras.layers import deserialize as deserialize_layer # pylint: disable=g-import-not-at-top
cells = []
for cell_config in config.pop('cells'):
cells.append(
deserialize_layer(cell_config, custom_objects=custom_objects))
return cls(cells, **config)
def from_config(cls, config, custom_objects=None):
# Avoid mutating the input dict.
config = copy.deepcopy(config)
model = deserialize_layer(config.pop("model"),
custom_objects=custom_objects)
config["model"] = model
return super().from_config(config, custom_objects)
def from_config(cls, config, custom_objects=None):
from keras.layers import deserialize as deserialize_layer # pylint: disable=g-import-not-at-top
# Avoid mutating the input dict
config = copy.deepcopy(config)
layer = deserialize_layer(
config.pop('layer'), custom_objects=custom_objects)
return cls(layer, **config)
def from_config(cls, config, custom_objects=None):
from keras.layers import deserialize as deserialize_layer
cell = deserialize_layer(config.pop('cell'),
custom_objects=custom_objects)
num_constants = config.pop('num_constants', None)
layer = cls(cell, **config)
layer._num_constants = num_constants
return layer
def from_config(cls, config, custom_objects=None):
from keras.layers import deserialize as deserialize_layer # pylint: disable=g-import-not-at-top
cell = deserialize_layer(config.pop('cell'),
custom_objects=custom_objects)
num_constants = config.pop('num_constants', 0)
layer = cls(cell, **config)
layer._num_constants = num_constants # pylint: disable=protected-access
return layer
def from_config(cls, config, custom_objects=None):
from keras.layers import deserialize as deserialize_layer
# Avoid mutating the input dict
config = copy.deepcopy(config)
layer = deserialize_layer(config.pop("layer"),
custom_objects=custom_objects)
return cls(layer, **config)
def from_config(cls, config, custom_objects=None):
from keras.layers import deserialize as deserialize_layer
cells = []
for cell_config in config.pop("cells"):
cells.append(
deserialize_layer(cell_config, custom_objects=custom_objects)
)
return cls(cells, **config)
def from_config(cls, config, custom_objects=None):
# Instead of updating the input, create a copy and use that.
config = copy.deepcopy(config)
num_constants = config.pop('num_constants', 0)
# Handle forward layer instantiation (as would parent class).
from keras.layers import deserialize as deserialize_layer # pylint: disable=g-import-not-at-top
config['layer'] = deserialize_layer(
config['layer'], custom_objects=custom_objects)
# Handle (optional) backward layer instantiation.
backward_layer_config = config.pop('backward_layer', None)
if backward_layer_config is not None:
backward_layer = deserialize_layer(
backward_layer_config, custom_objects=custom_objects)
config['backward_layer'] = backward_layer
# Instantiate the wrapper, adjust it and return it.
layer = cls(**config)
layer._num_constants = num_constants
return layer
def from_config(cls, config):
from keras.layers import deserialize as deserialize_layer
layers_config = config.pop('layers')
layers = [
deserialize_layer(layers_config[i])
for i in range(len(layers_config))
]
return cls(layers, **config)
def test_lambda():
layer_test(layers.Lambda,
kwargs={'function': lambda x: x + 1},
input_shape=(3, 2))
layer_test(layers.Lambda,
kwargs={'function': lambda x, a, b: x * a + b,
'arguments': {'a': 0.6, 'b': 0.4}},
input_shape=(3, 2))
def antirectifier(x):
x -= K.mean(x, axis=1, keepdims=True)
x = K.l2_normalize(x, axis=1)
pos = K.relu(x)
neg = K.relu(-x)
return K.concatenate([pos, neg], axis=1)
def antirectifier_output_shape(input_shape):
shape = list(input_shape)
assert len(shape) == 2 # only valid for 2D tensors
shape[-1] *= 2
return tuple(shape)
layer_test(layers.Lambda,
kwargs={'function': antirectifier,
'output_shape': antirectifier_output_shape},
input_shape=(3, 2))
# test layer with multiple outputs
def test_multiple_outputs():
def func(x):
return [x * 0.2, x * 0.3]
def output_shape(input_shape):
return [input_shape, input_shape]
def mask(inputs, mask=None):
return [None, None]
i = layers.Input(shape=(64, 64, 3))
o = layers.Lambda(function=func,
output_shape=output_shape,
mask=mask)(i)
o1, o2 = o
assert o1._keras_shape == (None, 64, 64, 3)
assert o2._keras_shape == (None, 64, 64, 3)
model = Model(i, o)
x = np.random.random((4, 64, 64, 3))
out1, out2 = model.predict(x)
assert out1.shape == (4, 64, 64, 3)
assert out2.shape == (4, 64, 64, 3)
assert_allclose(out1, x * 0.2, atol=1e-4)
assert_allclose(out2, x * 0.3, atol=1e-4)
test_multiple_outputs()
# test serialization with function
def f(x):
return x + 1
ld = layers.Lambda(f)
config = ld.get_config()
ld = deserialize_layer({'class_name': 'Lambda', 'config': config})
# test with lambda
ld = layers.Lambda(
lambda x: K.concatenate([K.square(x), x]),
output_shape=lambda s: tuple(list(s)[:-1] + [2 * s[-1]]))
config = ld.get_config()
ld = layers.Lambda.from_config(config)
# test serialization with output_shape function
def f(x):
return K.concatenate([K.square(x), x])
def f_shape(s):
return tuple(list(s)[:-1] + [2 * s[-1]])
ld = layers.Lambda(f, output_shape=f_shape)
config = ld.get_config()
ld = deserialize_layer({'class_name': 'Lambda', 'config': config})
def from_config(cls, config, custom_objects=None):
cells = []
for cell_config in config.pop('cells'):
cells.append(
deserialize_layer(cell_config, custom_objects=custom_objects))
return cls(cells, **config)
