LocallyConnected1D
classtf.keras.layers.LocallyConnected1D(
filters,
kernel_size,
strides=1,
padding="valid",
data_format=None,
activation=None,
use_bias=True,
kernel_initializer="glorot_uniform",
bias_initializer="zeros",
kernel_regularizer=None,
bias_regularizer=None,
activity_regularizer=None,
kernel_constraint=None,
bias_constraint=None,
implementation=1,
**kwargs
)
Locally-connected layer for 1D inputs.
The LocallyConnected1D
layer works similarly to
the Conv1D
layer, except that weights are unshared,
that is, a different set of filters is applied at each different patch
of the input.
Note: layer attributes cannot be modified after the layer has been called
once (except the trainable
attribute).
Example
# apply a unshared weight convolution 1d of length 3 to a sequence with
# 10 timesteps, with 64 output filters
model = Sequential()
model.add(LocallyConnected1D(64, 3, input_shape=(10, 32)))
# now model.output_shape == (None, 8, 64)
# add a new conv1d on top
model.add(LocallyConnected1D(32, 3))
# now model.output_shape == (None, 6, 32)
Arguments
dilation_rate
value != 1."valid"
(case-insensitive).
"same"
may be supported in the future.
"valid"
means no padding.channels_last
(default) or channels_first
.
The ordering of the dimensions in the inputs.
channels_last
corresponds to inputs with shape
(batch, length, channels)
while channels_first
corresponds to inputs with shape
(batch, channels, length)
.
It defaults to the image_data_format
value found in your
Keras config file at ~/.keras/keras.json
.
If you never set it, then it will be "channels_last".a(x) = x
).kernel
weights matrix.kernel
weights matrix.1
, 2
, or 3
.
1
loops over input spatial locations to perform the forward pass.
It is memory-efficient but performs a lot of (small) ops.2
stores layer weights in a dense but sparsely-populated 2D matrix
and implements the forward pass as a single matrix-multiply. It uses
a lot of RAM but performs few (large) ops.
3
stores layer weights in a sparse tensor and implements the forward
pass as a single sparse matrix-multiply.
-
How to choose:
-
1
: large, dense models,
2
: small models,
3
: large, sparse models,
where "large" stands for large input/output activations
(i.e. many filters
, input_filters
, large input_size
,
output_size
), and "sparse" stands for few connections between inputs
and outputs, i.e. small ratio
filters * input_filters * kernel_size / (input_size * strides)
,
where inputs to and outputs of the layer are assumed to have shapes
(input_size, input_filters)
, (output_size, filters)
respectively.
It is recommended to benchmark each in the setting of interest to pick the most efficient one (in terms of speed and memory usage). Correct choice of implementation can lead to dramatic speed improvements (e.g. 50X), potentially at the expense of RAM.
Also, only padding="valid"
is supported by implementation=1
.
Input shape
3D tensor with shape: (batch_size, steps, input_dim)
Output shape
3D tensor with shape: (batch_size, new_steps, filters)
steps
value might have changed due to padding or strides.