Python v0.3.0a0+fdf26ef

Dense

class dragon.vm.tensorlayer.layers.Dense(
  n_units,
  act=None,
  W_init='glorot_uniform',
  b_init='zeros',
  in_channels=None,
  name=None
)[source]

The fully connected layer.

Examples:

# Define and call a dense layer with the input
input = tl.layers.Input([100, 50])
output = tl.layers.Dense(n_units=800)(input)

__init__

Dense.__init__(
  n_units,
  act=None,
  W_init='glorot_uniform',
  b_init='zeros',
  in_channels=None,
  name=None
)[source]

Create a Dense layer.

Parameters:
  • n_units (int, required) – The number of output units.
  • act (callable, optional) – The optional activation function.
  • W_init (Union[callable, str], optional) – The initializer for weight matrix.
  • b_init (Union[callable, str], optional) – The initializer for bias vector.
  • in_channels (int, optional) – The number of input units.
  • name (str, optional) – The optional layer name.