最近看了深度殘差收縮網路的原論文，包含（Keras和·TFlearn倆版本），想把Keras框架的拿來給Cifar10跑，最后發現過擬合非常嚴重，訓練集準確率很不錯，但是val-準確率約只有60多，Dropout和正則化都有加，實在找不到原因了，多謝各位大佬指教！附上論文原文的連接：
https://blog.csdn.net/Jordanisxu/article/details/105007339?ops_request_misc=%25257B%252522request%25255Fid%252522%25253A%252522160946653916780277046277%252522%25252C%252522scm%252522%25253A%25252220140713.130102334.pc%25255Fall.%252522%25257D&request_id=160946653916780277046277&biz_id=0&utm_medium=distribute.pc_search_result.none-task-blog-2~all~first_rank_v2~rank_v29-13-105007339.pc_search_result_no_baidu_js&utm_term=%E6%B7%B1%E5%BA%A6%E6%AE%8B%E5%B7%AE%E6%94%B6%E7%BC%A9%E7%BD%91%E7%BB%9C
以下是我的代碼：多謝各位大佬！
from __future__ import print_function
import keras
import numpy as np
from keras.datasets import cifar10
from keras.layers import Dense, Conv2D, BatchNormalization, Activation
from keras.layers import AveragePooling2D, Input, GlobalAveragePooling2D
from keras.optimizers import Adam
from keras.regularizers import l2
from keras import backend as K
from keras.models import Model
from keras.layers.core import Lambda
from keras.callbacks import LearningRateScheduler
K.set_learning_phase(1)
# Input image dimensions
img_rows, img_cols = 32, 32
# The data, split between train and test sets
(x_train, y_train), (x_test, y_test) = cifar10.load_data()
if K.image_data_format() == 'channels_first':
x_train = x_train.reshape(x_train.shape[0], 3, img_rows, img_cols)
x_test = x_test.reshape(x_test.shape[0], 3, img_rows, img_cols)
input_shape = (3, img_rows, img_cols)
else:
x_train = x_train.reshape(x_train.shape[0], img_rows, img_cols, 3)
x_test = x_test.reshape(x_test.shape[0], img_rows, img_cols, 3)
input_shape = (img_rows, img_cols, 3)
# Noised data
x_train = x_train.astype('float32') / 255. + 0.1 * np.random.random([x_train.shape[0], img_rows, img_cols, 3])
x_test = x_test.astype('float32') / 255. + 0.1 * np.random.random([x_test.shape[0], img_rows, img_cols, 3])
print('x_train shape:', x_train.shape)
print(x_train.shape[0], 'train samples')
print(x_test.shape[0], 'test samples')
# convert class vectors to binary class matrices
y_train = keras.utils.to_categorical(y_train, 10)
y_test = keras.utils.to_categorical(y_test, 10)
def scheduler(epoch):
if epoch % 50 == 0 and epoch != 0:
lr = K.get_value(model.optimizer.lr)
K.set_value(model.optimizer.lr, lr * 0.1)
print("lr changed to {}".format(lr * 0.1))
return K.get_value(model.optimizer.lr)
def abs_backend(inputs):
return K.abs(inputs)
def expand_dim_backend(inputs):
return K.expand_dims(K.expand_dims(inputs, 1), 1)
def sign_backend(inputs):
return K.sign(inputs)
def pad_backend(inputs, in_channels, out_channels):
pad_dim = (out_channels - in_channels) // 2
inputs = K.expand_dims(inputs, -1)
inputs = K.spatial_3d_padding(inputs, ((0, 0), (0, 0), (pad_dim, pad_dim)), 'channels_last')
return K.squeeze(inputs, -1)
# Residual Shrinakge Block
def residual_shrinkage_block(incoming, nb_blocks, out_channels, downsample=False,
downsample_strides=2):
residual = incoming
in_channels = incoming.get_shape().as_list()[-1]
for i in range(nb_blocks):
identity = residual
if not downsample:
downsample_strides = 1
residual = BatchNormalization()(residual)
residual = Activation('relu')(residual)
residual = Conv2D(out_channels, 3, strides=(downsample_strides, downsample_strides),
padding='same', kernel_initializer='he_normal',
kernel_regularizer=l2(1e-4))(residual)
residual = BatchNormalization()(residual)
residual = Activation('relu')(residual)
residual = Conv2D(out_channels, 3, padding='same', kernel_initializer='he_normal',
kernel_regularizer=l2(1e-4))(residual)
# Calculate global means
residual_abs = Lambda(abs_backend)(residual)
abs_mean = GlobalAveragePooling2D()(residual_abs)
# Calculate scaling coefficients
scales = Dense(out_channels, activation=None, kernel_initializer='he_normal',
kernel_regularizer=l2(1e-4))(abs_mean)
scales = BatchNormalization()(scales)
scales = Activation('relu')(scales)
scales = Dense(out_channels, activation='sigmoid', kernel_regularizer='L2')(scales)
scales = Lambda(expand_dim_backend)(scales)
# Calculate thresholds
thres = keras.layers.multiply([abs_mean, scales])
# Soft thresholding
sub = keras.layers.subtract([residual_abs, thres])
zeros = keras.layers.subtract([sub, sub])
n_sub = keras.layers.maximum([sub, zeros])
residual = keras.layers.multiply([Lambda(sign_backend)(residual), n_sub])
# Downsampling (it is important to use the pooL-size of (1, 1))
if downsample_strides > 1:
identity = AveragePooling2D(pool_size=(1, 1), strides=(2, 2))(identity)
# Zero_padding to match channels (it is important to use zero padding rather than 1by1 convolution)
if in_channels != out_channels:
identity = Lambda(pad_backend, arguments={'in_channels': in_channels, 'out_channels': out_channels})(
identity)
residual = keras.layers.add([residual, identity])
return residual
# define and train a model
inputs = Input(shape=input_shape)
net = Conv2D(16, 3, padding='same', kernel_initializer='he_normal', kernel_regularizer=l2(1e-4))(inputs)
net = residual_shrinkage_block(net, 1, 16)
net = residual_shrinkage_block(net, 1, 32, downsample=True)
net = residual_shrinkage_block(net, 1, 32, downsample=True)
net = residual_shrinkage_block(net, 1, 64, downsample=True)
net = BatchNormalization()(net)
net = Activation('relu')(net)
net = GlobalAveragePooling2D()(net)
outputs = Dense(10, activation='softmax', kernel_initializer='he_normal', kernel_regularizer=l2(1e-4))(net)
model = Model(inputs=inputs, outputs=outputs)
model.compile(loss='categorical_crossentropy', optimizer=Adam(), metrics=['accuracy'])
reduce_lr = LearningRateScheduler(scheduler)
model.fit(x_train, y_train, batch_size=100, epochs=500, verbose=1, validation_data=https://bbs.csdn.net/topics/(x_test, y_test),shuffle=True, callbacks=[reduce_lr])
# get results
K.set_learning_phase(0)
DRSN_train_score = model.evaluate(x_train, y_train, batch_size=100, verbose=0)
print('Train loss:', DRSN_train_score[0])
print('Train accuracy:', DRSN_train_score[1])
DRSN_test_score = model.evaluate(x_test, y_test, batch_size=100, verbose=0)
print('Test loss:', DRSN_test_score[0])
print('Test accuracy:', DRSN_test_score[1])
uj5u.com熱心網友回復：
小白畢設求助！uj5u.com熱心網友回復：
把kernel_regularizer調大一點試試呢。這樣正則化效果更強一些，例如l2(2e-4)、l2(5e-4)轉載請註明出處，本文鏈接：https://www.uj5u.com/qita/244414.html
標籤：人工智能技術
上一篇：MongoDB遷移之后，啟動報錯 WT_PANIC: WiredTiger library panic
下一篇：求教Python