cifar10训练 test 0.894
name: "CIFAR10_full"
layer {
name: "cifar"
type: "Data"
top: "data"
top: "label"
include {
phase: TRAIN
}
transform_param {
mean_file: "mean.binaryproto"
#scale:0.00390625
mirror:1
crop_size:28
}
data_param {
source: "cifar10_train_lmdb"
batch_size: 256
backend: LMDB
}
}
layer {
name: "cifar"
type: "Data"
top: "data"
top: "label"
include {
phase: TEST
}
transform_param {
mean_file: "mean.binaryproto"
#scale:0.00390625
mirror:1
crop_size: 28
}
data_param {
source: "cifar10_test_lmdb"
batch_size: 100
backend: LMDB
}
}
layer {
name: "conv1"
type: "Convolution"
bottom: "data"
top: "conv1"
param {
lr_mult: 1
}
param {
lr_mult: 2
}
convolution_param {
num_output: 32
pad:2
kernel_size: 3
stride: 1
weight_filler {
type: "gaussian"
std: 0.0001
}
bias_filler {
type: "constant"
}
}
}
layer {
name: "relu1_1"
type: "ReLU"
bottom: "conv1"
top: "conv1"
}
layer {
name: "conv1_1"
type: "Convolution"
bottom: "conv1"
top: "conv1_1"
param {
lr_mult: 1
}
param {
lr_mult: 2
}
convolution_param {
num_output: 32
pad:1
kernel_size: 3
stride: 1
weight_filler {
type: "gaussian"
std: 0.0001
}
bias_filler {
type: "constant"
}
}
}
layer {
name: "bn_conv1"
bottom: "conv1_1"
top: "conv1_1"
type: "BatchNorm"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
batch_norm_param {
use_global_stats: false
}
}
layer {
name: "scale_conv1"
bottom: "conv1_1"
top: "conv1_1"
type: "Scale"
param {
lr_mult: 0.1
decay_mult: 0
}
param {
lr_mult: 0.1
decay_mult: 0
}
scale_param {
bias_term: true
}
}
layer {
name: "pool1"
type: "Pooling"
bottom: "conv1_1"
top: "pool1"
pooling_param {
pool: MAX
kernel_size: 3
stride: 2
}
}
layer {
name: "conv2"
type: "Convolution"
bottom: "pool1"
top: "conv2"
param {
lr_mult: 1
}
param {
lr_mult: 2
}
convolution_param {
num_output: 32
pad:2
kernel_size: 3
stride: 1
weight_filler {
type: "gaussian"
std: 0.005
}
bias_filler {
type: "constant"
}
}
}
layer {
name: "relu2_2"
type: "ReLU"
bottom: "conv2"
top: "conv2"
}
layer {
name: "conv2_2"
type: "Convolution"
bottom: "conv2"
top: "conv2_2"
param {
lr_mult: 1
}
param {
lr_mult: 2
}
convolution_param {
num_output: 32
pad:1
kernel_size: 3
stride: 1
weight_filler {
type: "gaussian"
std: 0.005
}
bias_filler {
type: "constant"
}
}
}
layer {
name: "bn_conv2"
bottom: "conv2_2"
top: "conv2_2"
type: "BatchNorm"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
batch_norm_param {
use_global_stats: false
}
}
layer {
name: "scale_conv2"
bottom: "conv2_2"
top: "conv2_2"
type: "Scale"
param {
lr_mult: 0.1
decay_mult: 0
}
param {
lr_mult: 0.1
decay_mult: 0
}
scale_param {
bias_term: true
}
}
layer {
name: "pool2"
type: "Pooling"
bottom: "conv2_2"
top: "pool2"
pooling_param {
pool: MAX
kernel_size: 3
stride: 2
}
}
layer {
name: "conv3"
type: "Convolution"
bottom: "pool2"
top: "conv3"
convolution_param {
num_output: 64
pad:2
kernel_size: 3
stride: 1
weight_filler {
type: "gaussian"
std: 0.05
}
bias_filler {
type: "constant"
}
}
}
layer {
name: "relu3_3"
type: "ReLU"
bottom: "conv3"
top: "conv3"
}
layer {
name: "conv3_3"
type: "Convolution"
bottom: "conv3"
top: "conv3_3"
convolution_param {
num_output: 64
pad:1
kernel_size: 3
stride: 1
weight_filler {
type: "gaussian"
std: 0.05
}
bias_filler {
type: "constant"
}
}
}
layer {
name: "bn_conv3"
bottom: "conv3_3"
top: "conv3_3"
type: "BatchNorm"
}
layer {
name: "bn_conv3"
bottom: "conv3_3"
top: "conv3_3"
type: "BatchNorm"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
batch_norm_param {
use_global_stats: false
}
}
layer {
name: "scale_conv3"
bottom: "conv3_3"
top: "conv3_3"
type: "Scale"
param {
lr_mult: 0.1
decay_mult: 0
}
param {
lr_mult: 0.1
decay_mult: 0
}
scale_param {
bias_term: true
}
}
layer {
name: "relu3"
type: "ReLU"
bottom: "conv3_3"
top: "conv3_3"
}
layer {
name: "pool3"
type: "Pooling"
bottom: "conv3_3"
top: "pool3"
pooling_param {
pool: MAX
kernel_size: 3
stride: 2
}
}
layer {
name: "conv4"
type: "Convolution"
bottom: "pool3"
top: "conv4"
convolution_param {
num_output: 64
pad:2
kernel_size: 3
stride: 1
weight_filler {
type: "gaussian"
std: 0.05
}
bias_filler {
type: "constant"
}
}
}
layer {
name: "relu4_4"
type: "ReLU"
bottom: "conv4"
top: "conv4"
}
layer {
name: "conv4_4"
type: "Convolution"
bottom: "conv4"
top: "conv4_4"
convolution_param {
num_output: 64
pad:1
kernel_size: 3
stride: 1
weight_filler {
type: "gaussian"
std: 0.05
}
bias_filler {
type: "constant"
}
}
}
layer {
name: "bn_conv4"
bottom: "conv4_4"
top: "conv4_4"
type: "BatchNorm"
}
layer {
name: "bn_conv4"
bottom: "conv4_4"
top: "conv4_4"
type: "BatchNorm"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
batch_norm_param {
use_global_stats: false
}
}
layer {
name: "scale_conv4"
bottom: "conv4_4"
top: "conv4_4"
type: "Scale"
param {
lr_mult: 0.1
decay_mult: 0
}
param {
lr_mult: 0.1
decay_mult: 0
}
scale_param {
bias_term: true
}
}
layer {
name: "relu4"
type: "ReLU"
bottom: "conv4_4"
top: "conv4_4"
}
layer {
name: "pool4"
type: "Pooling"
bottom: "conv4_4"
top: "pool4"
pooling_param {
pool: MAX
kernel_size: 3
stride: 2
}
}
layer {
name: "conv5"
type: "Convolution"
bottom: "pool4"
top: "conv5"
convolution_param {
num_output: 128
pad:2
kernel_size: 3
stride: 1
weight_filler {
type: "gaussian"
std: 0.05
}
bias_filler {
type: "constant"
}
}
}
layer {
name: "relu5_5"
type: "ReLU"
bottom: "conv5"
top: "conv5"
}
layer {
name: "conv5_5"
type: "Convolution"
bottom: "conv5"
top: "conv5_5"
convolution_param {
num_output: 128
pad:1
kernel_size: 3
stride: 1
weight_filler {
type: "gaussian"
std: 0.05
}
bias_filler {
type: "constant"
}
}
}
layer {
name: "bn_conv5"
bottom: "conv5_5"
top: "conv5_5"
type: "BatchNorm"
}
layer {
name: "bn_conv5"
bottom: "conv5_5"
top: "conv5_5"
type: "BatchNorm"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
batch_norm_param {
use_global_stats: false
}
}
layer {
name: "scale_conv5"
bottom: "conv5_5"
top: "conv5_5"
type: "Scale"
param {
lr_mult: 0.1
decay_mult: 0
}
param {
lr_mult: 0.1
decay_mult: 0
}
scale_param {
bias_term: true
}
}
layer {
name: "relu5"
type: "ReLU"
bottom: "conv5_5"
top: "conv5_5"
}
layer {
name: "pool5"
type: "Pooling"
bottom: "conv5_5"
top: "pool5"
pooling_param {
pool: MAX
kernel_size: 3
stride: 2
}
}
layer {
name: "ip1"
type: "InnerProduct"
bottom: "pool5"
top: "ip1"
param {
lr_mult: 1
decay_mult: 10
}
param {
lr_mult: 2
decay_mult: 0
}
inner_product_param {
num_output: 128
weight_filler {
type: "gaussian"
std: 0.005
}
bias_filler {
type: "constant"
}
}
}
layer {
name: "ip2"
type: "InnerProduct"
bottom: "ip1"
top: "ip2"
param {
lr_mult: 1
decay_mult: 10
}
param {
lr_mult: 2
decay_mult: 0
}
inner_product_param {
num_output: 10
weight_filler {
type: "gaussian"
std: 0.005
}
bias_filler {
type: "constant"
}
}
}
layer {
name: "accuracy"
type: "Accuracy"
bottom: "ip2"
bottom: "label"
top: "accuracy"
include {
phase: TEST
}
}
layer {
name: "accuracy"
type: "Accuracy"
bottom: "ip2"
bottom: "label"
top: "accuracy"
include {
phase: TRAIN
}
}
layer {
name: "loss"
type: "SoftmaxWithLoss"
bottom: "ip2"
bottom: "label"
top: "loss"
}
solver：
# reduce learning rate after 120 epochs (60000 iters) by factor 0f 10
# then another factor of 10 after 10 more epochs (5000 iters)
# The train/test net protocol buffer definition
net: "cifar10_full_train_test.prototxt"
# test_iter specifies how many forward passes the test should carry out.
# In the case of CIFAR10, we have test batch size 100 and 100 test iterations,
# covering the full 10,000 testing images.
test_iter: 200
# Carry out testing every 1000 training iterations.
test_interval: 1000
# The base learning rate, momentum and the weight decay of the network.
base_lr: 0.0001
momentum: 0.9
weight_decay: 0.004
# The learning rate policy
#lr_policy: "multistep"
#gamma:0.1
#stepvalue:5000
lr_policy: "inv"
gamma: 0.0001
power: 0.75
# Display every 200 iterations
display: 200
# The maximum number of iterations
max_iter: 100000
# snapshot intermediate results
snapshot: 10000
#snapshot_format: LMDB
snapshot_prefix: "cifar10_full"
# solver mode: CPU or GPU
solver_mode: GPU
type:"SGD"
截图：