#! /usr/bin/python
# -*- coding: utf-8 -*-
import numbers
import tensorlayerx as tlx
from tensorlayerx import logging
from tensorlayerx.nn.core import Module
__all__ = [
'DropconnectLinear',
]
[docs]class DropconnectLinear(Module):
"""
The :class:`DropconnectLinear` class is :class:`Dense` with DropConnect
behaviour which randomly removes connections between this layer and the previous
layer according to a keeping probability.
Parameters
----------
keep : float
The keeping probability.
The lower the probability it is, the more activations are set to zero.
out_features : int
The number of units of this layer.
act : activation function
The activation function of this layer.
W_init : weights initializer or str
The initializer for the weight matrix.
b_init : biases initializer or str
The initializer for the bias vector.
in_features: int
The number of channels of the previous layer.
If None, it will be automatically detected when the layer is forwarded for the first time.
name : str
A unique layer name.
Examples
--------
>>> net = tlx.nn.Input([10, 784], name='input')
>>> net = tlx.nn.DropconnectLinear(keep=0.8, out_features=800, act=tlx.ReLU, name='DropconnectLinear1')(net)
>>> output shape :(10, 800)
>>> net = tlx.nn.DropconnectLinear(keep=0.5, out_features=800, act=tlx.ReLU, name='DropconnectLinear2')(net)
>>> output shape :(10, 800)
>>> net = tlx.nn.DropconnectLinear(keep=0.5, out_features=10, name='DropconnectLinear3')(net)
>>> output shape :(10, 10)
References
----------
- `Wan, L. (2013). Regularization of neural networks using dropconnect <http://machinelearning.wustl.edu/mlpapers/papers/icml2013_wan13>`__
"""
def __init__(
self,
keep=0.5,
out_features=100,
act=None,
W_init='truncated_normal',
b_init='constant',
in_features=None,
name=None, # 'dropconnect',
):
super().__init__(name, act=act)
if isinstance(keep, numbers.Real) and not (keep > 0 and keep <= 1):
raise ValueError("keep must be a scalar tensor or a float in the " "range (0, 1], got %g" % keep)
self.keep = keep
self.out_features = out_features
self.W_init = self.str_to_init(W_init)
self.b_init = self.str_to_init(b_init)
self.in_features = in_features
if self.in_features is not None:
self.build((None, self.in_features))
self._built = True
logging.info(
"DropconnectLinear %s: %d %s" %
(self.name, out_features, self.act.__class__.__name__ if self.act is not None else 'No Activation')
)
def __repr__(self):
actstr = self.act.__name__ if self.act is not None else 'No Activation'
s = ('{classname}(out_features={out_features}, ' + actstr)
s += ', keep={keep}'
if self.in_features is not None:
s += ', in_features=\'{in_features}\''
if self.name is not None:
s += ', name=\'{name}\''
s += ')'
return s.format(classname=self.__class__.__name__, **self.__dict__)
def build(self, inputs_shape):
if len(inputs_shape) != 2:
raise Exception("The input dimension must be rank 2")
if self.in_features is None:
self.in_features = inputs_shape[1]
n_in = inputs_shape[-1]
self.weights = self._get_weights("weights", shape=(n_in, self.out_features), init=self.W_init)
if self.b_init:
self.biases = self._get_weights("biases", shape=(self.out_features), init=self.b_init)
self.dropout = tlx.ops.Dropout(p=self.keep)
self.matmul = tlx.ops.MatMul()
self.bias_add = tlx.ops.BiasAdd()
def forward(self, inputs):
if self._forward_state == False:
if self._built == False:
self.build(tlx.get_tensor_shape(inputs))
self._built = True
self._forward_state = True
W_dropcon = self.dropout(self.weights)
outputs = self.matmul(inputs, W_dropcon)
if self.b_init:
outputs = self.bias_add(outputs, self.biases)
if self.act:
outputs = self.act(outputs)
if not self._nodes_fixed and self._build_graph:
self._add_node(inputs, outputs)
self._nodes_fixed = True
return outputs