dlpy.layers.OutputLayer

class dlpy.layers.OutputLayer(name=None, act='softmax', fcmp_act=None, fcmp_err=None, error=None, init=None, std=None, mean=None, truncation_factor=None, init_bias=None, n=None, n_softmax_samples=None, include_bias=None, target_std=None, full_connect=None, src_layers=None, **kwargs)

Bases: dlpy.layers.Layer

Output layer

Parameters:
name : string, optional

Specifies the name of the layer.

act : string, optional

Specifies the activation function.
Valid Values: AUTO, IDENTITY, LOGISTIC, SIGMOID, TANH, RECTIFIER, RELU, SOFPLUS, ELU, LEAKY, FCMP
Default: AUTO

fcmp_act : string, optional

Specifies the FCMP activation function for the layer.

fcmp_err : string, optional

Specifies the FCMP error function for the output layer.

error : int, optional

Specifies the error function. This function is also known as the loss function.
Valid Values: AUTO, GAMMA, NORMAL, POISSON, ENTROPY, CTC, FCMPERR, CTC_ALT
Default: AUTO

init : string, optional

Specifies the initialization scheme for the layer.
Valid Values: XAVIER, UNIFORM, NORMAL, CAUCHY, XAVIER1, XAVIER2, MSRA, MSRA1, MSRA2
Default: XAVIER

std : float, optional

Specifies the standard deviation value when the init parameter is set to NORMAL.

mean : float, optional

Specifies the mean value when the init parameter is set to NORMAL.

truncation_factor : float, optional

Specifies the truncation threshold (truncationFactor x std), when the init parameter is set to NORMAL

init_bias : float, optional

Specifies the initial bias for the layer.

n : int, optional

Specifies the number of neurons in the output layer. By default, the number of neurons is automatically determined when the model training begins. The specified value cannot be smaller than the number of target variable levels.

n_softmax_samples : int, optional

Specifies the number of samples used in sampled Softmax.

include_bias : bool, optional

Includes bias neurons.
Default: True

target_std : int, optional

Specifies how to standardize the variables in the output layer.
Valid Values: MIDRANGE, NONE, STD
Default: NONE

full_connect : bool, optional

In default, the output layer is fully connected to all the previous layers. When it is false, the output layer becomes a loss function layer.
Default: True

src_layers : iter-of-Layers, optional

Specifies the layers directed to this layer.

Returns:
OutputLayer
__init__(name=None, act='softmax', fcmp_act=None, fcmp_err=None, error=None, init=None, std=None, mean=None, truncation_factor=None, init_bias=None, n=None, n_softmax_samples=None, include_bias=None, target_std=None, full_connect=None, src_layers=None, **kwargs)

Initialize self. See help(type(self)) for accurate signature.

Methods

__init__([name, act, fcmp_act, fcmp_err, …]) Initialize self.
count_instances()
format_name([block_num, local_count]) Format the name of the layer
get_number_of_instances()
to_model_params() Convert the model configuration to CAS action parameters