def downsample(inputs,
pool_type: str = 'max',
sorted: bool = True,
stage: int = 1): # noqa: A002
layers_pool = []
if pool_type == 'max':
layers_pool.append(
L.MaxPooling1D(pool_size=3,
strides=2,
padding='same',
name=f'pool_{stage}'))
elif pool_type == 'k_max':
k = int(inputs.shape[1].value / 2)
layers_pool.append(
KMaxPoolingLayer(k=k, sorted=sorted, name=f'pool_{stage}'))
elif pool_type == 'conv':
layers_pool.append(
L.Conv1D(filters=inputs.shape[-1].value,
kernel_size=3,
strides=2,
padding='same',
name=f'pool_{stage}'))
layers_pool.append(L.BatchNormalization())
elif pool_type is None:
layers_pool = []
else:
raise ValueError(f'unsupported pooling type `{pool_type}`!')
tensor_out = inputs
for layer in layers_pool:
tensor_out = layer(tensor_out)
return tensor_out
def build_model_arc(self):
output_dim = len(self.pre_processor.label2idx)
config = self.hyper_parameters
embed_model = self.embedding.embed_model
layers_region = [
L.Conv1D(**config['region_embedding']),
L.BatchNormalization(),
L.PReLU(),
L.Dropout(**config['region_dropout'])
]
layers_main = [
L.GlobalMaxPooling1D(),
L.Dense(**config['dense']),
L.BatchNormalization(),
L.PReLU(),
L.Dropout(**config['dropout']),
L.Dense(output_dim, **config['activation'])
]
tensor_out = embed_model.output
# build region tensors
for layer in layers_region:
tensor_out = layer(tensor_out)
# build the base pyramid layer
tensor_out = conv_block(tensor_out, **config['conv_block'])
# build the above pyramid layers while `steps > 2`
seq_len = tensor_out.shape[1].value
if seq_len is None:
raise ValueError(
'`sequence_length` should be explicitly assigned, but it is `None`.'
)
for i in range(floor(log2(seq_len)) - 2):
tensor_out = resnet_block(tensor_out,
stage=i + 1,
**config['resnet_block'])
for layer in layers_main:
tensor_out = layer(tensor_out)
self.tf_model = tf.keras.Model(embed_model.inputs, tensor_out)
def conv_block(inputs,
filters: int,
kernel_size: int = 3,
activation: str = 'linear',
shortcut: bool = True):
layers_conv_unit = []
layers_conv_unit.append(L.BatchNormalization())
layers_conv_unit.append(L.PReLU())
layers_conv_unit.append(
L.Conv1D(filters=filters,
kernel_size=kernel_size,
strides=1,
padding='same',
activation=activation))
layers_conv_block = layers_conv_unit * 2
tensor_out = inputs
for layer in layers_conv_block:
tensor_out = layer(tensor_out)
if shortcut:
tensor_out = L.Add()([inputs, tensor_out])
return tensor_out