目录
摘要
Diganta Misra的一篇题为“Mish: A Self Regularized Non-Monotonic Neural Activation Function”的新论文介绍了一个新的深度学习激活函数，该函数在最终准确度上比Swish(+.494%)和ReLU(+ 1.671%)都有提高
公式如下：
-
如何在Pytorch使用Mish函数
定义Mish函数。
class Mish(torch.nn.Module):
def __init__(self):
super().__init__()
def forward(self, x):
x = x * (torch.tanh(torch.nn.functional.softplus(x)))
return x
调用函数：
class Path1_64(nn.Module):
def __init__(self):
super().__init__()
self.conv1 = ConvBN(32, 64, 3)
self.conv2 = ConvBN(64, 64, [1, 9])
self.conv3 = ConvBN(64, 64, [9, 1])
self.conv4 = ConvBN(64, 64, 1)
self.resBlock = ResBlock(ch=64, nblocks=2)
self.conv5 = ConvBN(64, 64, [1, 7])
self.conv6 = ConvBN(64, 64, [7, 1])
self.conv7 = ConvBN(64, 64, 1)
self.relu = Mish()
def forward(self, input):
x1 = self.conv1(input)
x2 = self.conv2(x1)
x3 = self.conv3(x2)
x4 = self.conv4(x3)
r1 = self.resBlock(x4)
x5 = self.conv5(r1)
x6 = self.conv6(x5)
x7 = self.conv7(x6)
x7 = self.relu(x7 + x4)
return x7
调用Mish激活函数和调用其他的激活函数一样，直接调用即可。
-
如何在Keras中使用Mish激活函数。
定义Mish激活函数
import tensorflow as tf
from tensorflow.python.keras.layers import *
from tensorflow.keras.layers import Activation
from tensorflow.keras.utils import get_custom_objects
class Mish(Activation):
def __init__(self, activation, **kwargs):
super(Mish, self).__init__(activation, **kwargs)
self.__name__ = 'Mish'
def mish(inputs):
return inputs * tf.math.tanh(tf.math.softplus(inputs))
get_custom_objects().update({'Mish': Mish(mish)})
调用激活函数：
def bn_prelu(x):
x = BatchNormalization(epsilon=1e-5)(x)
x = Activation('Mish')(x)
return x
def build_model(out_dims, input_shape=(100, 100, 3)):
inputs_dim = Input(input_shape)
x = Conv2D(32, (3, 3), strides=(2, 2), padding='same')(inputs_dim)
x = bn_prelu(x)
x = Conv2D(32, (3, 3), strides=(1, 1), padding='same')(x)
x = bn_prelu(x)
x = MaxPooling2D(pool_size=(2, 2))(x)
x = Conv2D(64, (3, 3), strides=(1, 1), padding='same')(x)
x = bn_prelu(x)
x = Conv2D(64, (3, 3), strides=(1, 1), padding='same')(x)
x = bn_prelu(x)
x = MaxPooling2D(pool_size=(2, 2))(x)
x = Conv2D(128, (3, 3), strides=(1, 1), padding='same')(x)
x = bn_prelu(x)
x = Conv2D(128, (3, 3), strides=(1, 1), padding='same')(x)
x = bn_prelu(x)
x = MaxPooling2D(pool_size=(2, 2))(x)
x = Conv2D(256, (3, 3), strides=(1, 1), padding='same')(x)
x = bn_prelu(x)
x = Conv2D(256, (3, 3), strides=(1, 1), padding='same')(x)
x = bn_prelu(x)
x = GlobalAveragePooling2D()(x)
dp_1 = Dropout(0.5)(x)
fc2 = Dense(out_dims)(dp_1)
fc2 = Activation('softmax')(fc2) # 此处注意，为sigmoid函数
model = Model(inputs=inputs_dim, outputs=fc2)
return model
model = build_model(2) # 生成模型
optimizer = Adam(lr=1e-3) # 加入优化器，设置优化器的学习率。
model.compile(optimizer=optimizer, loss='sparse_categorical_crossentropy', metrics=['accuracy'])
