def demo_create_encoder(latent_dim, cat_dim, window_size, input_dim):
input_layer = Input(shape=(window_size, input_dim))
code = TimeDistributed(Dense(64, activation='linear'))(input_layer)
code = Bidirectional(LSTM(128, return_sequences=True))(code)
code = BatchNormalization()(code)
code = ELU()(code)
code = Bidirectional(LSTM(64))(code)
code = BatchNormalization()(code)
code = ELU()(code)
cat = Dense(64)(code)
cat = BatchNormalization()(cat)
cat = PReLU()(cat)
cat = Dense(cat_dim, activation='softmax')(cat)
latent_repr = Dense(64)(code)
latent_repr = BatchNormalization()(latent_repr)
latent_repr = PReLU()(latent_repr)
latent_repr = Dense(latent_dim, activation='linear')(latent_repr)
decode = Concatenate()([latent_repr, cat])
decode = RepeatVector(window_size)(decode)
decode = Bidirectional(LSTM(64, return_sequences=True))(decode)
decode = ELU()(decode)
decode = Bidirectional(LSTM(128, return_sequences=True))(decode)
decode = ELU()(decode)
decode = TimeDistributed(Dense(64))(decode)
decode = ELU()(decode)
decode = TimeDistributed(Dense(input_dim, activation='linear'))(decode)
error = Subtract()([input_layer, decode])
return Model(input_layer, [decode, latent_repr, cat, error])
def demo_create_discriminator(latent_dim):
input_layer = Input(shape=(latent_dim,))
disc = Dense(128)(input_layer)
disc = ELU()(disc)
disc = Dense(64)(disc)
disc = ELU()(disc)
disc = Dense(1, activation="sigmoid")(disc)
model = Model(input_layer, disc)
return model
def create_model(self):
""" DEFINE NEURAL NETWORK """
# define model as a linear stack of dense layers
self.model = Sequential()
# iteratively add hidden layers
for layer_n in range(1, self.n_layers+1):
print layer_n, "hidden layer\n",
if layer_n == 1: # input_shape needs to be specified for the first layer
self.model.add(Dense(units=self.n_hidden[layer_n], input_shape=(self.n_features,),
kernel_initializer=self.weights_init, bias_initializer=self.bias_init))
else:
self.model.add(Dense(units=self.n_hidden[layer_n], kernel_initializer=self.weights_init,
bias_initializer=self.bias_init))
if self.batch_norm:
self.model.add(BatchNormalization()) # add batch normalization before activation
# add the activation layer explicitly
if self.activation == 'LeakyReLU':
self.model.add(LeakyReLU(alpha=self.alpha)) # for x < 0, y = alpha*x -> non-zero slope in the negative region
elif self.activation == 'ReLU':
self.model.add(ReLU())
elif self.activation == 'eLU':
self.model.add(ELU())
# add output layer; no activation for the output layer
self.model.add(Dense(units=self.n_outputs, kernel_initializer=self.weights_init,
bias_initializer=self.bias_init))
def get_test_model_sequential():
"""Returns a typical (VGG-like) sequential test model."""
model = Sequential()
model.add(Conv2D(8, (3, 3), activation='relu', input_shape=(32, 32, 3)))
model.add(Conv2D(8, (3, 3), activation='relu'))
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Dropout(0.25))
model.add(Conv2D(16, (3, 3), activation='elu'))
model.add(Conv2D(16, (3, 3)))
model.add(ELU())
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Dropout(0.25))
model.add(Flatten())
model.add(Dense(64, activation='sigmoid'))
model.add(Dropout(0.5))
model.add(Dense(10, activation='softmax'))
model.compile(loss='categorical_crossentropy', optimizer='sgd')
# fit to dummy data
training_data_size = 1
data_in = [np.random.random(size=(training_data_size, 32, 32, 3))]
data_out = [np.random.random(size=(training_data_size, 10))]
model.fit(data_in, data_out, epochs=10)
return model
def discriminator(input_shape=(28, 28, 1), nb_filter=64):
model = Sequential()
model.add(Conv2D(nb_filter, (5, 5), strides=(2, 2), padding='same', input_shape=input_shape))
model.add(BatchNormalization())
model.add(ELU())
model.add(Conv2D(2*nb_filter, (5, 5), strides=(2, 2)))
model.add(BatchNormalization())
model.add(ELU())
model.add(Flatten())
model.add(Dense(4*nb_filter))
model.add(BatchNormalization())
model.add(ELU())
model.add(Dropout(0.5))
model.add(Dense(1))
model.add(Activation('sigmoid'))
print(model.summary())
return model
def discriminator(input_shape=(28, 28, 1), nb_filter=64): #kontrol edici fonksiyon, 1. parametre resmin boyutu (28,28,renkliyse 3 değilse 1), 2.parametre filtre 64 tane
model = Sequential() #ardışık model oluşturduk, resmi alıp sinir ağının sonunda değer elde ederiz.
model.add(Conv2D(nb_filter, (5, 5), strides=(2, 2), padding='same', input_shape=input_shape))
#convnet oluşturduk, 1.parametre filitre, 2. parametre boyutu, 3. parametre filitrelerin resimde nasıl kayacağı (2 pixel kayacak), 4. parametre boyutu korur resmin etrafına 0 lar ekler,5. parametre çıkış boyutu
model.add(BatchNormalization()) #normalleştirme
model.add(ELU()) #aktivasyon fonksiyonu (elu)
model.add(Conv2D(2*nb_filter, (5, 5), strides=(2, 2))) #2 katı filitre ile yeni layer. boyut 5x5x128 e düşer
model.add(BatchNormalization())
model.add(ELU())
model.add(Flatten()) #düzleştirme işlevi Dense layer eklemek için
model.add(Dense(4*nb_filter)) # nöron sayısı
model.add(BatchNormalization())
model.add(ELU())
model.add(Dropout(0.5)) # ezber yapmasını engeller.
model.add(Dense(1)) # output sinir hücresi, çıkış
model.add(Activation('sigmoid')) #aktivasyon fonksiyonu gelen değer [0,1] arasında
print(model.summary()) #model özeti
return model
def discriminator(input_shape=(WIDTH_OF_IMAGE, HEIGHT_OF_IMAGE, CHANNEL_OF_IMAGE), nb_filter=CHANNEL_COEFFICIENT):
model = Sequential()
model.add(Conv2D(nb_filter,
(CONVOLUTIONRANGE, CONVOLUTIONRANGE),
strides=(UPSAMPLINGRANGE, UPSAMPLINGRANGE),
padding='same', input_shape=input_shape))
model.add(BatchNormalization())
model.add(ELU())
model.add(Conv2D(2*nb_filter, (CONVOLUTIONRANGE, CONVOLUTIONRANGE), strides=(UPSAMPLINGRANGE, UPSAMPLINGRANGE)))
model.add(BatchNormalization())
model.add(ELU())
model.add(Flatten())
model.add(Dense(4*nb_filter))
model.add(BatchNormalization())
model.add(ELU())
model.add(Dropout(0.5))
model.add(Dense(1))
model.add(Activation('sigmoid'))
print(model.summary())
return model
def get_activation(self, x):
if self.activation_type == 'elu':
activate = ELU()(x)
elif self.activation_type == 'relu':
activate = ReLU()(x)
elif self.activation_type == 'prelu':
activate = PReLU()(x)
elif self.activation_type == 'leakyrelu':
activate = LeakyReLU()(x)
else:
raise ValueError('Undefined ACTIVATION_TYPE!')
return activate
def get_activation(self):
activation_type = self.activation_type
if activation_type ==1:
activation = LeakyReLU(alpha=0.1)
elif activation_type == 2:
activation = ReLU()
elif activation_type == 3:
activation = PReLU(alpha_initializer='zeros', alpha_regularizer=None, alpha_constraint=None, shared_axes=None)
elif activation_type == 4:
activation=ELU(alpha=1.0)
else:
raise Exception('Not a valid activation type')
return activation
def test_delete_channels_advanced_activations(channel_index, data_format):
layer_test_helper_flatten_2d(LeakyReLU(), channel_index, data_format)
layer_test_helper_flatten_2d(ELU(), channel_index, data_format)
layer_test_helper_flatten_2d(ThresholdedReLU(), channel_index, data_format)
def init_model(self,
input_shape,
num_classes,
**kwargs):
layers = 5
filters_size = [64, 128, 256, 512, 512]
kernel_size = (3, 3)
pool_size = [(2, 2), (2, 2), (2, 2), (4, 1), (4, 1)]
freq_axis = 2
channel_axis = 3
channel_size = 128
min_size = min(input_shape[:2])
melgram_input = Input(shape=input_shape)
# x = ZeroPadding2D(padding=(0, 37))(melgram_input)
x = Reshape((input_shape[0], input_shape[1], 1))(melgram_input)
x = BatchNormalization(axis=freq_axis, name='bn_0_freq')(x)
# Conv block 1
x = Convolution2D(
filters=filters_size[0],
kernel_size=kernel_size,
padding='same',
name='conv1')(x)
x = ELU()(x)
x = BatchNormalization(axis=channel_axis, name='bn1')(x)
x = MaxPooling2D(
pool_size=pool_size[0],
strides=pool_size[0],
name='pool1')(x)
x = Dropout(0.1, name='dropout1')(x)
min_size = min_size // pool_size[0][0]
for layer in range(1, layers):
min_size = min_size // pool_size[layer][0]
if min_size < 1:
break
x = Convolution2D(
filters=filters_size[layer],
kernel_size=kernel_size,
padding='same',
name='conv' + str(layer + 1))(x)
x = ELU()(x)
x = BatchNormalization(axis=channel_axis, name='bn'+str(layer + 1)+'')(x)
x = MaxPooling2D(
pool_size=pool_size[layer],
strides=pool_size[layer],
name='pool'+str(layer + 1)+'')(x)
x = Dropout(0.1, name='dropout'+str(layer + 1)+'')(x)
x = Reshape((-1, channel_size))(x)
gru_units = 32
if num_classes > 32:
gru_units = int(num_classes * 1.5)
# GRU block 1, 2, output
x = CuDNNGRU(gru_units, return_sequences=True, name='gru1')(x)
x = CuDNNGRU(gru_units, return_sequences=False, name='gru2')(x)
x = Dropout(0.3)(x)
outputs = Dense(num_classes, activation='softmax', name='output')(x)
model = TFModel(inputs=melgram_input, outputs=outputs)
optimizer = optimizers.Adam(
# learning_rate=1e-3,
lr=1e-3,
beta_1=0.9,
beta_2=0.999,
epsilon=1e-08,
decay=1e-4,
amsgrad=True)
model.compile(
optimizer=optimizer,
loss="sparse_categorical_crossentropy",
metrics=['accuracy'])
model.summary()
self._model = model
self.is_init = True
def init_model(self, input_shape, num_classes, **kwargs):
freq_axis = 2
channel_axis = 3
channel_size = 128
min_size = min(input_shape[:2])
melgram_input = Input(shape=input_shape)
# x = ZeroPadding2D(padding=(0, 37))(melgram_input)
# x = BatchNormalization(axis=freq_axis, name='bn_0_freq')(x)
x = Reshape((input_shape[0], input_shape[1], 1))(melgram_input)
# Conv block 1
x = Convolution2D(64, 3, 1, padding='same', name='conv1')(x)
x = BatchNormalization(axis=channel_axis, name='bn1')(x)
x = ELU()(x)
x = MaxPooling2D(pool_size=(2, 2), strides=(2, 2), name='pool1')(x)
x = Dropout(0.1, name='dropout1')(x)
# Conv block 2
x = Convolution2D(channel_size, 3, 1, padding='same', name='conv2')(x)
x = BatchNormalization(axis=channel_axis, name='bn2')(x)
x = ELU()(x)
x = MaxPooling2D(pool_size=(3, 3), strides=(3, 3), name='pool2')(x)
x = Dropout(0.1, name='dropout2')(x)
# Conv block 3
x = Convolution2D(channel_size, 3, 1, padding='same', name='conv3')(x)
x = BatchNormalization(axis=channel_axis, name='bn3')(x)
x = ELU()(x)
x = MaxPooling2D(pool_size=(3, min_size / 6),
strides=(3, min_size / 6),
name='pool3')(x)
x = Dropout(0.1, name='dropout3')(x)
# if min_size // 24 >= 4:
# # Conv block 4
# x = Convolution2D(
# channel_size,
# 3,
# 1,
# padding='same',
# name='conv4')(x)
# x = BatchNormalization(axis=channel_axis, name='bn4')(x)
# x = ELU()(x)
# x = MaxPooling2D(pool_size=(4, 4), strides=(4, 4), name='pool4')(x)
# x = Dropout(0.1, name='dropout4')(x)
x = Reshape((-1, channel_size))(x)
avg = GlobalAvgPool1D()(x)
max = GlobalMaxPool1D()(x)
x = concatenate([avg, max], axis=-1)
# x = Dense(max(int(num_classes*1.5), 128), activation='relu', name='dense1')(x)
x = Dropout(0.3)(x)
outputs = Dense(num_classes, activation='softmax', name='output')(x)
model = TFModel(inputs=melgram_input, outputs=outputs)
optimizer = optimizers.Nadam(lr=0.002,
beta_1=0.9,
beta_2=0.999,
epsilon=None,
schedule_decay=0.004)
model.compile(optimizer=optimizer,
loss='sparse_categorical_crossentropy',
metrics=['accuracy'])
model.summary()
self._model = model
self.is_init = True
def init_model(self, input_shape, num_classes, **kwargs):
freq_axis = 2
channel_axis = 3
channel_size = 128
min_size = min(input_shape[:2])
melgram_input = Input(shape=input_shape)
# x = ZeroPadding2D(padding=(0, 37))(melgram_input)
# x = BatchNormalization(axis=freq_axis, name='bn_0_freq')(x)
x = Reshape((input_shape[0], input_shape[1], 1))(melgram_input)
# Conv block 1
x = Convolution2D(64, 3, 1, padding='same', name='conv1')(x)
x = BatchNormalization(axis=channel_axis, name='bn1')(x)
x = ELU()(x)
x = MaxPooling2D(pool_size=(2, 2), strides=(2, 2), name='pool1')(x)
x = Dropout(0.1, name='dropout1')(x)
# Conv block 2
x = Convolution2D(channel_size, 3, 1, padding='same', name='conv2')(x)
x = BatchNormalization(axis=channel_axis, name='bn2')(x)
x = ELU()(x)
x = MaxPooling2D(pool_size=(3, 3), strides=(3, 3), name='pool2')(x)
x = Dropout(0.1, name='dropout2')(x)
# Conv block 3
x = Convolution2D(channel_size, 3, 1, padding='same', name='conv3')(x)
x = BatchNormalization(axis=channel_axis, name='bn3')(x)
x = ELU()(x)
x = MaxPooling2D(pool_size=(4, 4), strides=(4, 4), name='pool3')(x)
x = Dropout(0.1, name='dropout3')(x)
if min_size // 24 >= 4:
# Conv block 4
x = Convolution2D(channel_size, 3, 1, padding='same',
name='conv4')(x)
x = BatchNormalization(axis=channel_axis, name='bn4')(x)
x = ELU()(x)
x = MaxPooling2D(pool_size=(4, 4), strides=(4, 4), name='pool4')(x)
x = Dropout(0.1, name='dropout4')(x)
x = Reshape((-1, channel_size))(x)
gru_units = 128
if num_classes > gru_units:
gru_units = int(num_classes * 1.5)
# GRU block 1, 2, output
x = CuDNNGRU(gru_units, return_sequences=True, name='gru1')(x)
x = CuDNNGRU(gru_units, return_sequences=False, name='gru2')(x)
# x = Dense(max(int(num_classes*1.5), 128), activation='relu', name='dense1')(x)
x = Dropout(0.3)(x)
outputs = Dense(num_classes, activation='softmax', name='output')(x)
model = TFModel(inputs=melgram_input, outputs=outputs)
optimizer = optimizers.Adam(
# learning_rate=1e-3,
lr=1e-3,
beta_1=0.9,
beta_2=0.999,
epsilon=1e-08,
decay=1e-4,
amsgrad=True)
model.compile(optimizer=optimizer,
loss="sparse_categorical_crossentropy",
metrics=['accuracy'])
model.summary()
self._model = model
self.is_init = True
def add(self, alpha=1.0, **kwargs):
return self._add_layer(ELU(alpha, **kwargs))
num_classes = len(np.unique(y_train))
# Normalize data
mean = np.mean(x_train,axis=(0,1,2,3))
std = np.std(x_train,axis=(0,1,2,3))
x_train = (x_train-mean)/(std+1e-7)
x_test = (x_test-mean)/(std+1e-7)
if args.model_type == 'softmax':
y_train = utils.to_categorical(y_train,num_classes)
y_test = utils.to_categorical(y_test,num_classes)
model = Sequential()
model.add(Conv2D(baseMapNum, (3,3), padding='same', kernel_regularizer=regularizers.l2(weight_decay), input_shape=x_train.shape[1:]))
model.add(ELU())
model.add(BatchNormalization())
model.add(Conv2D(baseMapNum, (3,3), padding='same', kernel_regularizer=regularizers.l2(weight_decay)))
model.add(ELU())
model.add(BatchNormalization())
model.add(MaxPooling2D(pool_size=(2,2)))
model.add(Dropout(0.2))
model.add(Conv2D(2*baseMapNum, (3,3), padding='same', kernel_regularizer=regularizers.l2(weight_decay)))
model.add(ELU())
model.add(BatchNormalization())
model.add(Conv2D(2*baseMapNum, (3,3), padding='same', kernel_regularizer=regularizers.l2(weight_decay)))
model.add(ELU())
model.add(BatchNormalization())
model.add(MaxPooling2D(pool_size=(2,2)))
model.add(Dropout(0.3))
def get_test_model_full():
"""Returns a maximally complex test model,
using all supported layer types with different parameter combination.
"""
input_shapes = [
(26, 28, 3),
(4, 4, 3),
(4, 4, 3),
(4, ),
(2, 3),
(27, 29, 1),
(17, 1),
(17, 4),
]
inputs = [Input(shape=s) for s in input_shapes]
outputs = []
for inp in inputs[6:8]:
for padding in ['valid', 'same']:
for s in range(1, 6):
for out_channels in [1, 2]:
for d in range(1, 4):
outputs.append(
Conv1D(out_channels,
s,
padding=padding,
dilation_rate=d)(inp))
for padding_size in range(0, 5):
outputs.append(ZeroPadding1D(padding_size)(inp))
for crop_left in range(0, 2):
for crop_right in range(0, 2):
outputs.append(Cropping1D((crop_left, crop_right))(inp))
for upsampling_factor in range(1, 5):
outputs.append(UpSampling1D(upsampling_factor)(inp))
for padding in ['valid', 'same']:
for pool_factor in range(1, 6):
for s in range(1, 4):
outputs.append(
MaxPooling1D(pool_factor, strides=s,
padding=padding)(inp))
outputs.append(
AveragePooling1D(pool_factor,
strides=s,
padding=padding)(inp))
outputs.append(GlobalMaxPooling1D()(inp))
outputs.append(GlobalAveragePooling1D()(inp))
for inp in [inputs[0], inputs[5]]:
for padding in ['valid', 'same']:
for h in range(1, 6):
for out_channels in [1, 2]:
for d in range(1, 4):
outputs.append(
Conv2D(out_channels, (h, 1),
padding=padding,
dilation_rate=(d, 1))(inp))
outputs.append(
SeparableConv2D(out_channels, (h, 1),
padding=padding,
dilation_rate=(d, 1))(inp))
for sy in range(1, 4):
outputs.append(
Conv2D(out_channels, (h, 1),
strides=(1, sy),
padding=padding)(inp))
outputs.append(
SeparableConv2D(out_channels, (h, 1),
strides=(sy, sy),
padding=padding)(inp))
for sy in range(1, 4):
outputs.append(
MaxPooling2D((h, 1), strides=(1, sy),
padding=padding)(inp))
for w in range(1, 6):
for out_channels in [1, 2]:
for d in range(1, 4) if sy == 1 else [1]:
outputs.append(
Conv2D(out_channels, (1, w),
padding=padding,
dilation_rate=(1, d))(inp))
outputs.append(
SeparableConv2D(out_channels, (1, w),
padding=padding,
dilation_rate=(1, d))(inp))
for sx in range(1, 4):
outputs.append(
Conv2D(out_channels, (1, w),
strides=(sx, 1),
padding=padding)(inp))
outputs.append(
SeparableConv2D(out_channels, (1, w),
strides=(sx, sx),
padding=padding)(inp))
for sx in range(1, 4):
outputs.append(
MaxPooling2D((1, w), strides=(1, sx),
padding=padding)(inp))
outputs.append(ZeroPadding2D(2)(inputs[0]))
outputs.append(ZeroPadding2D((2, 3))(inputs[0]))
outputs.append(ZeroPadding2D(((1, 2), (3, 4)))(inputs[0]))
outputs.append(Cropping2D(2)(inputs[0]))
outputs.append(Cropping2D((2, 3))(inputs[0]))
outputs.append(Cropping2D(((1, 2), (3, 4)))(inputs[0]))
for y in range(1, 3):
for x in range(1, 3):
outputs.append(UpSampling2D(size=(y, x))(inputs[0]))
outputs.append(GlobalAveragePooling2D()(inputs[0]))
outputs.append(GlobalMaxPooling2D()(inputs[0]))
outputs.append(AveragePooling2D((2, 2))(inputs[0]))
outputs.append(MaxPooling2D((2, 2))(inputs[0]))
outputs.append(UpSampling2D((2, 2))(inputs[0]))
outputs.append(keras.layers.concatenate([inputs[0], inputs[0]]))
outputs.append(Dropout(0.5)(inputs[0]))
outputs.append(BatchNormalization()(inputs[0]))
outputs.append(BatchNormalization(center=False)(inputs[0]))
outputs.append(BatchNormalization(scale=False)(inputs[0]))
outputs.append(Conv2D(2, (3, 3), use_bias=True)(inputs[0]))
outputs.append(Conv2D(2, (3, 3), use_bias=False)(inputs[0]))
outputs.append(SeparableConv2D(2, (3, 3), use_bias=True)(inputs[0]))
outputs.append(SeparableConv2D(2, (3, 3), use_bias=False)(inputs[0]))
outputs.append(Dense(2, use_bias=True)(inputs[3]))
outputs.append(Dense(2, use_bias=False)(inputs[3]))
shared_conv = Conv2D(1, (1, 1),
padding='valid',
name='shared_conv',
activation='relu')
up_scale_2 = UpSampling2D((2, 2))
x1 = shared_conv(up_scale_2(inputs[1])) # (1, 8, 8)
x2 = shared_conv(up_scale_2(inputs[2])) # (1, 8, 8)
x3 = Conv2D(1, (1, 1), padding='valid')(up_scale_2(inputs[2])) # (1, 8, 8)
x = keras.layers.concatenate([x1, x2, x3]) # (3, 8, 8)
outputs.append(x)
x = Conv2D(3, (1, 1), padding='same', use_bias=False)(x) # (3, 8, 8)
outputs.append(x)
x = Dropout(0.5)(x)
outputs.append(x)
x = keras.layers.concatenate(
[MaxPooling2D((2, 2))(x),
AveragePooling2D((2, 2))(x)]) # (6, 4, 4)
outputs.append(x)
x = Flatten()(x) # (1, 1, 96)
x = Dense(4, use_bias=False)(x)
outputs.append(x)
x = Dense(3)(x) # (1, 1, 3)
outputs.append(x)
intermediate_input_shape = (3, )
intermediate_in = Input(intermediate_input_shape)
intermediate_x = intermediate_in
intermediate_x = Dense(8)(intermediate_x)
intermediate_x = Dense(5)(intermediate_x)
intermediate_model = Model(inputs=[intermediate_in],
outputs=[intermediate_x],
name='intermediate_model')
intermediate_model.compile(loss='mse', optimizer='nadam')
x = intermediate_model(x) # (1, 1, 5)
intermediate_model_2 = Sequential()
intermediate_model_2.add(Dense(7, input_shape=(5, )))
intermediate_model_2.add(Dense(5))
intermediate_model_2.compile(optimizer='rmsprop',
loss='categorical_crossentropy')
x = intermediate_model_2(x) # (1, 1, 5)
x = Dense(3)(x) # (1, 1, 3)
shared_activation = Activation('tanh')
outputs = outputs + [
Activation('tanh')(inputs[3]),
Activation('hard_sigmoid')(inputs[3]),
Activation('selu')(inputs[3]),
Activation('sigmoid')(inputs[3]),
Activation('softplus')(inputs[3]),
Activation('softmax')(inputs[3]),
Activation('relu')(inputs[3]),
LeakyReLU()(inputs[3]),
ELU()(inputs[3]),
shared_activation(inputs[3]),
inputs[4],
inputs[1],
x,
shared_activation(x),
]
print('Model has {} outputs.'.format(len(outputs)))
model = Model(inputs=inputs, outputs=outputs, name='test_model_full')
model.compile(loss='mse', optimizer='nadam')
# fit to dummy data
training_data_size = 1
batch_size = 1
epochs = 10
data_in = generate_input_data(training_data_size, input_shapes)
data_out = generate_output_data(training_data_size, outputs)
model.fit(data_in, data_out, epochs=epochs, batch_size=batch_size)
return model
def init_model(self, input_shape, num_classes, **kwargs):
freq_axis = 2
channel_axis = 3
channel_size = 128
min_size = min(input_shape[:2])
inputs = Input(shape=input_shape)
# x = ZeroPadding2D(padding=(0, 37))(melgram_input)
# x = BatchNormalization(axis=freq_axis, name='bn_0_freq')(x)
x = Reshape((input_shape[0], input_shape[1], 1))(inputs)
# Conv block 1
x = Convolution2D(64, 3, 1, padding='same', name='conv1')(x)
x = BatchNormalization(axis=channel_axis, name='bn1')(x)
x = ELU()(x)
x = MaxPooling2D(pool_size=(2, 2), strides=(2, 2), name='pool1')(x)
x = Dropout(0.1, name='dropout1')(x)
# Conv block 2
x = Convolution2D(channel_size, 3, 1, padding='same', name='conv2')(x)
x = BatchNormalization(axis=channel_axis, name='bn2')(x)
x = ELU()(x)
x = MaxPooling2D(pool_size=(3, 3), strides=(3, 3), name='pool2')(x)
x = Dropout(0.1, name='dropout2')(x)
# Conv block 3
x = Convolution2D(channel_size, 3, 1, padding='same', name='conv3')(x)
x = BatchNormalization(axis=channel_axis, name='bn3')(x)
x = ELU()(x)
x = MaxPooling2D(pool_size=(4, 4), strides=(4, 4), name='pool3')(x)
x = Dropout(0.1, name='dropout3')(x)
if min_size // 24 >= 4:
# Conv block 4
x = Convolution2D(channel_size, 3, 1, padding='same',
name='conv4')(x)
x = BatchNormalization(axis=channel_axis, name='bn4')(x)
x = ELU()(x)
x = MaxPooling2D(pool_size=(4, 4), strides=(4, 4), name='pool4')(x)
x = Dropout(0.1, name='dropout4')(x)
x = Reshape((-1, channel_size))(x)
avg = GlobalAvgPool1D()(x)
max = GlobalMaxPool1D()(x)
x = concatenate([avg, max], axis=-1)
# x = Dense(max(int(num_classes*1.5), 128), activation='relu', name='dense1')(x)
x = Dropout(0.3)(x)
outputs1 = Dense(num_classes, activation='softmax', name='output')(x)
# bnorm_1 = BatchNormalization(axis=2)(inputs)
lstm_1 = Bidirectional(CuDNNLSTM(64,
name='blstm_1',
return_sequences=True),
merge_mode='concat')(inputs)
activation_1 = Activation('tanh')(lstm_1)
dropout1 = SpatialDropout1D(0.5)(activation_1)
attention_1 = Attention(8, 16)([dropout1, dropout1, dropout1])
pool_1 = GlobalMaxPool1D()(attention_1)
dropout2 = Dropout(rate=0.5)(pool_1)
dense_1 = Dense(units=256, activation='relu')(dropout2)
outputs2 = Dense(units=num_classes, activation='softmax')(dense_1)
outputs = Average()([outputs1, outputs2])
model = TFModel(inputs=inputs, outputs=outputs)
optimizer = optimizers.Adam(
# learning_rate=1e-3,
lr=1e-3,
beta_1=0.9,
beta_2=0.999,
epsilon=1e-08,
decay=0.0002,
amsgrad=True)
model.compile(optimizer=optimizer,
loss='sparse_categorical_crossentropy',
metrics=['accuracy'])
model.summary()
self._model = model
self.is_init = True
