wassname/keras-contrib
1
0
Fork 0
mirror of https://github.com/wassname/keras-contrib.git synced 2026-08-14 12:30:09 +08:00
Code Issues Packages Projects Releases Wiki Activity

update pelu to K2

Browse Source
This commit is contained in:
farizrahman4u
2017-03-19 08:40:45 +05:30
parent 69e1a5d487
commit 4295b8e91f
1 changed files with 12 additions and 15 deletions
Download Patch File Download Diff File Expand all files Collapse all files
keras_contrib/layers/advanced_activations.py
+12 -15
View File
@@ -1,6 +1,5 @@
from .. import initializers
from keras.engine import Layer
from keras.utils.generic_utils import get_custom_objects
from .. import backend as K
import numpy as np
@@ -18,8 +17,8 @@ class PELU(Layer):
# Output shape
Same shape as the input.
# Arguments
alphas_init: initialization function for the alpha variable weights.
betas_init: initialization function for the beta variable weights.
alphas_initializer: initialization function for the alpha variable weights.
betas_initializer: initialization function for the beta variable weights.
weights: initial weights, as a list of a single Numpy array.
shared_axes: the axes along which to share learnable
parameters for the activation function.
@@ -33,16 +32,16 @@ class PELU(Layer):
- [PARAMETRIC EXPONENTIAL LINEAR UNIT FOR DEEP CONVOLUTIONAL NEURAL NETWORKS](https://arxiv.org/abs/1605.09332v3)
"""
def __init__(self, alphas_init='one', betas_init='one', weights=None, shared_axes=None, **kwargs):
def __init__(self, alphas_initializer='one', betas_initializer='one', weights=None, shared_axes=None, **kwargs):
self.supports_masking = True
self.alphas_init = initializations.get(alphas_init)
self.betas_init = initializations.get(betas_init)
self.alphas_initializer = initializers.get(alphas_initializer)
self.betas_initializer = initializers.get(betas_initializer)
self.initial_weights = weights
if not isinstance(shared_axes, (list, tuple)):
self.shared_axes = [shared_axes]
else:
self.shared_axes = list(shared_axes)
super(PELU, self).__init__(**kwargs)
super(PELU, self).__initializer__(**kwargs)
def build(self, input_shape):
param_shape = list(input_shape[1:])
@@ -53,10 +52,10 @@ class PELU(Layer):
self.param_broadcast[i - 1] = True
# Initialised as ones to emulate the default ELU
self.alphas = self.alphas_init(param_shape,
name='{}_alphas'.format(self.name))
self.betas = self.betas_init(param_shape,
name='{}_betas'.format(self.name))
self.alphas = self.add_weight(param_shape,
name='alpha',
initializer=self.alphas_initializerializer)
self.betas = self.add_weight(param_shape, name='betas', initializer=self.betas_initializerializer)
self.trainable_weights = [self.alphas, self.betas]
@@ -76,9 +75,7 @@ class PELU(Layer):
return neg + pos
def get_config(self):
config = {'alphas_init': self.alphas_init.__name__,
'betas_init': self.betas_init.__name__}
config = {'alphas_initializer': initializers.serialize(self.alphas_initializer),
'betas_initializer': initializers.serialize(betas_initializer)}
base_config = super(PELU, self).get_config()
return dict(list(base_config.items()) + list(config.items()))
get_custom_objects().update({"PELU": PELU})