Python keras.layers.dot() Examples
The following are 14
code examples of keras.layers.dot().
You can vote up the ones you like or vote down the ones you don't like,
and go to the original project or source file by following the links above each example.
You may also want to check out all available functions/classes of the module
keras.layers
, or try the search function
.
.
Example #1
| Source File: test_keras2_numeric.py From coremltools with BSD 3-Clause "New" or "Revised" License | 6 votes |
|
def test_tiny_cos_random(self):
np.random.seed(1988)
input_dim = 10
num_channels = 6
# Define a model
input_tensor = Input(shape=(input_dim,))
x1 = Dense(num_channels)(input_tensor)
x2 = Dense(num_channels)(x1)
x3 = Dense(num_channels)(x1)
x4 = dot([x2, x3], axes=-1, normalize=True)
x5 = Dense(num_channels)(x4)
model = Model(inputs=[input_tensor], outputs=[x5])
# Set some random weights
model.set_weights([np.random.rand(*w.shape) for w in model.get_weights()])
# Get the coreml model
self._test_model(model)
Example #2
| Source File: cf.py From keras-recommender with MIT License | 6 votes |
|
def create_model(self):
user_id_input = Input(shape=[1], name='user')
item_id_input = Input(shape=[1], name='item')
user_embedding = Embedding(output_dim=EMBEDDING_SIZE, input_dim=self.max_user_id + 1,
input_length=1, name='user_embedding')(user_id_input)
item_embedding = Embedding(output_dim=EMBEDDING_SIZE, input_dim=self.max_item_id + 1,
input_length=1, name='item_embedding')(item_id_input)
# reshape from shape: (batch_size, input_length, embedding_size)
# to shape: (batch_size, input_length * embedding_size) which is
# equal to shape: (batch_size, embedding_size)
user_vecs = Flatten()(user_embedding)
item_vecs = Flatten()(item_embedding)
# y = merge([user_vecs, item_vecs], mode='dot', output_shape=(1,))
y = dot([user_vecs, item_vecs], axes=1)
model = Model(inputs=[user_id_input, item_id_input], outputs=[y])
model.compile(optimizer='adam', loss='mae')
return model
Example #3
| Source File: trimodal_attention_models.py From contextual-multimodal-fusion with MIT License | 6 votes |
|
def self_attention(x):
'''
. stands for dot product
* stands for elemwise multiplication
m = x . transpose(x)
n = softmax(m)
o = n . x
a = o * x
return a
'''
m = dot([x, x], axes=[2,2])
n = Activation('softmax')(m)
o = dot([n, x], axes=[2,1])
a = multiply([o, x])
return a
Example #4
| Source File: merge_test.py From DeepLearning_Wavelet-LSTM with MIT License | 5 votes |
|
def test_merge_dot():
i1 = layers.Input(shape=(4,))
i2 = layers.Input(shape=(4,))
o = layers.dot([i1, i2], axes=1)
assert o._keras_shape == (None, 1)
model = models.Model([i1, i2], o)
dot_layer = layers.Dot(axes=1)
o2 = dot_layer([i1, i2])
assert dot_layer.output_shape == (None, 1)
x1 = np.random.random((2, 4))
x2 = np.random.random((2, 4))
out = model.predict([x1, x2])
assert out.shape == (2, 1)
expected = np.zeros((2, 1))
expected[0, 0] = np.dot(x1[0], x2[0])
expected[1, 0] = np.dot(x1[1], x2[1])
assert_allclose(out, expected, atol=1e-4)
# Test with negative tuple of axes.
o = layers.dot([i1, i2], axes=(-1, -1))
assert o._keras_shape == (None, 1)
model = models.Model([i1, i2], o)
out = model.predict([x1, x2])
assert out.shape == (2, 1)
assert_allclose(out, expected, atol=1e-4)
Example #5
| Source File: merge_test.py From DeepLearning_Wavelet-LSTM with MIT License | 5 votes |
|
def test_merge_dot():
i1 = layers.Input(shape=(4,))
i2 = layers.Input(shape=(4,))
o = layers.dot([i1, i2], axes=1)
assert o._keras_shape == (None, 1)
model = models.Model([i1, i2], o)
dot_layer = layers.Dot(axes=1)
o2 = dot_layer([i1, i2])
assert dot_layer.output_shape == (None, 1)
x1 = np.random.random((2, 4))
x2 = np.random.random((2, 4))
out = model.predict([x1, x2])
assert out.shape == (2, 1)
expected = np.zeros((2, 1))
expected[0, 0] = np.dot(x1[0], x2[0])
expected[1, 0] = np.dot(x1[1], x2[1])
assert_allclose(out, expected, atol=1e-4)
# Test with negative tuple of axes.
o = layers.dot([i1, i2], axes=(-1, -1))
assert o._keras_shape == (None, 1)
model = models.Model([i1, i2], o)
out = model.predict([x1, x2])
assert out.shape == (2, 1)
assert_allclose(out, expected, atol=1e-4)
Example #6
| Source File: merge_test.py From DeepLearning_Wavelet-LSTM with MIT License | 5 votes |
|
def test_merge_dot():
i1 = layers.Input(shape=(4,))
i2 = layers.Input(shape=(4,))
o = layers.dot([i1, i2], axes=1)
assert o._keras_shape == (None, 1)
model = models.Model([i1, i2], o)
dot_layer = layers.Dot(axes=1)
o2 = dot_layer([i1, i2])
assert dot_layer.output_shape == (None, 1)
x1 = np.random.random((2, 4))
x2 = np.random.random((2, 4))
out = model.predict([x1, x2])
assert out.shape == (2, 1)
expected = np.zeros((2, 1))
expected[0, 0] = np.dot(x1[0], x2[0])
expected[1, 0] = np.dot(x1[1], x2[1])
assert_allclose(out, expected, atol=1e-4)
# Test with negative tuple of axes.
o = layers.dot([i1, i2], axes=(-1, -1))
assert o._keras_shape == (None, 1)
model = models.Model([i1, i2], o)
out = model.predict([x1, x2])
assert out.shape == (2, 1)
assert_allclose(out, expected, atol=1e-4)
Example #7
| Source File: merge_test.py From DeepLearning_Wavelet-LSTM with MIT License | 5 votes |
|
def test_merge_dot():
i1 = layers.Input(shape=(4,))
i2 = layers.Input(shape=(4,))
o = layers.dot([i1, i2], axes=1)
assert o._keras_shape == (None, 1)
model = models.Model([i1, i2], o)
dot_layer = layers.Dot(axes=1)
o2 = dot_layer([i1, i2])
assert dot_layer.output_shape == (None, 1)
x1 = np.random.random((2, 4))
x2 = np.random.random((2, 4))
out = model.predict([x1, x2])
assert out.shape == (2, 1)
expected = np.zeros((2, 1))
expected[0, 0] = np.dot(x1[0], x2[0])
expected[1, 0] = np.dot(x1[1], x2[1])
assert_allclose(out, expected, atol=1e-4)
# Test with negative tuple of axes.
o = layers.dot([i1, i2], axes=(-1, -1))
assert o._keras_shape == (None, 1)
model = models.Model([i1, i2], o)
out = model.predict([x1, x2])
assert out.shape == (2, 1)
assert_allclose(out, expected, atol=1e-4)
Example #8
| Source File: merge_test.py From DeepLearning_Wavelet-LSTM with MIT License | 5 votes |
|
def test_merge_dot():
i1 = layers.Input(shape=(4,))
i2 = layers.Input(shape=(4,))
o = layers.dot([i1, i2], axes=1)
assert o._keras_shape == (None, 1)
model = models.Model([i1, i2], o)
dot_layer = layers.Dot(axes=1)
o2 = dot_layer([i1, i2])
assert dot_layer.output_shape == (None, 1)
x1 = np.random.random((2, 4))
x2 = np.random.random((2, 4))
out = model.predict([x1, x2])
assert out.shape == (2, 1)
expected = np.zeros((2, 1))
expected[0, 0] = np.dot(x1[0], x2[0])
expected[1, 0] = np.dot(x1[1], x2[1])
assert_allclose(out, expected, atol=1e-4)
# Test with negative tuple of axes.
o = layers.dot([i1, i2], axes=(-1, -1))
assert o._keras_shape == (None, 1)
model = models.Model([i1, i2], o)
out = model.predict([x1, x2])
assert out.shape == (2, 1)
assert_allclose(out, expected, atol=1e-4)
Example #9
| Source File: merge_test.py From DeepLearning_Wavelet-LSTM with MIT License | 5 votes |
|
def test_merge_dot():
i1 = layers.Input(shape=(4,))
i2 = layers.Input(shape=(4,))
o = layers.dot([i1, i2], axes=1)
assert o._keras_shape == (None, 1)
model = models.Model([i1, i2], o)
dot_layer = layers.Dot(axes=1)
o2 = dot_layer([i1, i2])
assert dot_layer.output_shape == (None, 1)
x1 = np.random.random((2, 4))
x2 = np.random.random((2, 4))
out = model.predict([x1, x2])
assert out.shape == (2, 1)
expected = np.zeros((2, 1))
expected[0, 0] = np.dot(x1[0], x2[0])
expected[1, 0] = np.dot(x1[1], x2[1])
assert_allclose(out, expected, atol=1e-4)
# Test with negative tuple of axes.
o = layers.dot([i1, i2], axes=(-1, -1))
assert o._keras_shape == (None, 1)
model = models.Model([i1, i2], o)
out = model.predict([x1, x2])
assert out.shape == (2, 1)
assert_allclose(out, expected, atol=1e-4)
Example #10
| Source File: merge_test.py From DeepLearning_Wavelet-LSTM with MIT License | 5 votes |
|
def test_merge_dot():
i1 = layers.Input(shape=(4,))
i2 = layers.Input(shape=(4,))
o = layers.dot([i1, i2], axes=1)
assert o._keras_shape == (None, 1)
model = models.Model([i1, i2], o)
dot_layer = layers.Dot(axes=1)
o2 = dot_layer([i1, i2])
assert dot_layer.output_shape == (None, 1)
x1 = np.random.random((2, 4))
x2 = np.random.random((2, 4))
out = model.predict([x1, x2])
assert out.shape == (2, 1)
expected = np.zeros((2, 1))
expected[0, 0] = np.dot(x1[0], x2[0])
expected[1, 0] = np.dot(x1[1], x2[1])
assert_allclose(out, expected, atol=1e-4)
# Test with negative tuple of axes.
o = layers.dot([i1, i2], axes=(-1, -1))
assert o._keras_shape == (None, 1)
model = models.Model([i1, i2], o)
out = model.predict([x1, x2])
assert out.shape == (2, 1)
assert_allclose(out, expected, atol=1e-4)
Example #11
| Source File: ast_attendgru_xtra.py From funcom with GNU General Public License v3.0 | 5 votes |
|
def create_model(self):
dat_input = Input(shape=(self.tdatlen,))
com_input = Input(shape=(self.comlen,))
sml_input = Input(shape=(self.smllen,))
ee = Embedding(output_dim=self.embdims, input_dim=self.tdatvocabsize, mask_zero=False)(dat_input)
se = Embedding(output_dim=self.smldims, input_dim=self.smlvocabsize, mask_zero=False)(sml_input)
se_enc = CuDNNGRU(self.recdims, return_state=True, return_sequences=True)
seout, state_sml = se_enc(se)
enc = CuDNNGRU(self.recdims, return_state=True, return_sequences=True)
encout, state_h = enc(ee, initial_state=state_sml)
de = Embedding(output_dim=self.embdims, input_dim=self.comvocabsize, mask_zero=False)(com_input)
dec = CuDNNGRU(self.recdims, return_sequences=True)
decout = dec(de, initial_state=state_h)
attn = dot([decout, encout], axes=[2, 2])
attn = Activation('softmax')(attn)
context = dot([attn, encout], axes=[2, 1])
ast_attn = dot([decout, seout], axes=[2, 2])
ast_attn = Activation('softmax')(ast_attn)
ast_context = dot([ast_attn, seout], axes=[2, 1])
context = concatenate([context, decout, ast_context])
out = TimeDistributed(Dense(self.recdims, activation="relu"))(context)
out = Flatten()(out)
out = Dense(self.comvocabsize, activation="softmax")(out)
model = Model(inputs=[dat_input, com_input, sml_input], outputs=out)
if self.config['multigpu']:
model = keras.utils.multi_gpu_model(model, gpus=2)
model.compile(loss='categorical_crossentropy', optimizer='adam', metrics=['accuracy'])
return self.config, model
Example #12
| Source File: trimodal_attention_models.py From contextual-multimodal-fusion with MIT License | 5 votes |
|
def bi_modal_attention(x, y):
'''
. stands for dot product
* stands for elemwise multiplication
{} stands for concatenation
m1 = x . transpose(y) || m2 = y . transpose(x)
n1 = softmax(m1) || n2 = softmax(m2)
o1 = n1 . y || o2 = m2 . x
a1 = o1 * x || a2 = o2 * y
return {a1, a2}
'''
m1 = dot([x, y], axes=[2, 2])
n1 = Activation('softmax')(m1)
o1 = dot([n1, y], axes=[2, 1])
a1 = multiply([o1, x])
m2 = dot([y, x], axes=[2, 2])
n2 = Activation('softmax')(m2)
o2 = dot([n2, x], axes=[2, 1])
a2 = multiply([o2, y])
return concatenate([a1, a2])
Example #13
| Source File: layers_export.py From Fabrik with GNU General Public License v3.0 | 5 votes |
|
def eltwise(layer, layer_in, layerId):
out = {}
if (layer['params']['layer_type'] == 'Multiply'):
# This input reverse is to handle visualization
out[layerId] = multiply(layer_in[::-1])
elif (layer['params']['layer_type'] == 'Sum'):
out[layerId] = add(layer_in[::-1])
elif (layer['params']['layer_type'] == 'Average'):
out[layerId] = average(layer_in[::-1])
elif (layer['params']['layer_type'] == 'Dot'):
out[layerId] = dot(layer_in[::-1], -1)
else:
out[layerId] = maximum(layer_in[::-1])
return out
Example #14
| Source File: keras_item2vec.py From MovieTaster-Open with MIT License | 4 votes |
|
def skipgram_model(vocab_size, embedding_dim=100, paradigm='Functional'):
# Sequential paradigm
if paradigm == 'Sequential':
target = Sequential()
target.add(Embedding(vocab_size, embedding_dim, input_length=1))
context = Sequential()
context.add(Embedding(vocab_size, embedding_dim, input_length=1))
# merge the pivot and context models
model = Sequential()
model.add(Merge([target, context], mode='dot'))
model.add(Reshape((1,), input_shape=(1,1)))
model.add(Activation('sigmoid'))
model.compile(optimizer='adam', loss='binary_crossentropy')
return model
# Functional paradigm
elif paradigm == 'Functional':
target = Input(shape=(1,), name='target')
context = Input(shape=(1,), name='context')
#print target.shape, context.shape
shared_embedding = Embedding(vocab_size, embedding_dim, input_length=1, name='shared_embedding')
embedding_target = shared_embedding(target)
embedding_context = shared_embedding(context)
#print embedding_target.shape, embedding_context.shape
merged_vector = dot([embedding_target, embedding_context], axes=-1)
reshaped_vector = Reshape((1,), input_shape=(1,1))(merged_vector)
#print merged_vector.shape
prediction = Dense(1, input_shape=(1,), activation='sigmoid')(reshaped_vector)
#print prediction.shape
model = Model(inputs=[target, context], outputs=prediction)
model.compile(optimizer='adam', loss='binary_crossentropy')
return model
else:
print('paradigm error')
return None
