I0312 23:23:10.648330 2676 caffe.cpp:218] Using GPUs 0
I0312 23:23:10.663291 2676 caffe.cpp:223] GPU 0: GeForce GTX 1070
I0312 23:23:10.862406 2676 solver.cpp:44] Initializing solver from parameters:
test_iter: 56
test_interval: 28
base_lr: 0.001
display: 20
max_iter: 5000
lr_policy: "step"
gamma: 0.1
momentum: 0.9
weight_decay: 0.0005
stepsize: 100000
snapshot: 4999
snapshot_prefix: "models/bvlc_alexnet/caffe_alexnet_sinatrain"
solver_mode: GPU
device_id: 0
net: "examples/alexnetfinetune/train_valsina.prototxt"
train_state {
level: 0
stage: ""
}
type: "SGD"
I0312 23:23:10.863147 2676 solver.cpp:87] Creating training net from net file: examples/alexnetfinetune/train_valsina.prototxt
I0312 23:23:10.863476 2676 net.cpp:296] The NetState phase (0) differed from the phase (1) specified by a rule in layer data
I0312 23:23:10.863488 2676 net.cpp:296] The NetState phase (0) differed from the phase (1) specified by a rule in layer accuracy
I0312 23:23:10.863648 2676 net.cpp:53] Initializing net from parameters:
name: "AlexNet"
state {
phase: TRAIN
level: 0
stage: ""
}
layer {
name: "data"
type: "Data"
top: "data"
top: "label"
include {
phase: TRAIN
}
transform_param {
mirror: true
crop_size: 227
mean_file: "examples/Mydataset_train_lmdb/mean_imagetest.binaryproto"
}
data_param {
source: "examples/Mydataset_train_lmdb"
batch_size: 256
backend: LMDB
}
}
layer {
name: "conv1"
type: "Convolution"
bottom: "data"
top: "conv1"
param {
lr_mult: 1
decay_mult: 1
}
param {
lr_mult: 2
decay_mult: 0
}
convolution_param {
num_output: 96
kernel_size: 11
stride: 4
weight_filler {
type: "gaussian"
std: 0.01
}
bias_filler {
type: "constant"
value: 0
}
}
}
layer {
name: "relu1"
type: "ReLU"
bottom: "conv1"
top: "conv1"
}
layer {
name: "norm1"
type: "LRN"
bottom: "conv1"
top: "norm1"
lrn_param {
local_size: 5
alpha: 0.0001
beta: 0.75
}
}
layer {
name: "pool1"
type: "Pooling"
bottom: "norm1"
top: "pool1"
pooling_param {
pool: MAX
kernel_size: 3
stride: 2
}
}
layer {
name: "conv2"
type: "Convolution"
bottom: "pool1"
top: "conv2"
param {
lr_mult: 1
decay_mult: 1
}
param {
lr_mult: 2
decay_mult: 0
}
convolution_param {
num_output: 256
pad: 2
kernel_size: 5
group: 2
weight_filler {
type: "gaussian"
std: 0.01
}
bias_filler {
type: "constant"
value: 0.1
}
}
}
layer {
name: "relu2"
type: "ReLU"
bottom: "conv2"
top: "conv2"
}
layer {
name: "norm2"
type: "LRN"
bottom: "conv2"
top: "norm2"
lrn_param {
local_size: 5
alpha: 0.0001
beta: 0.75
}
}
layer {
name: "pool2"
type: "Pooling"
bottom: "norm2"
top: "pool2"
pooling_param {
pool: MAX
kernel_size: 3
stride: 2
}
}
layer {
name: "conv3"
type: "Convolution"
bottom: "pool2"
top: "conv3"
param {
lr_mult: 1
decay_mult: 1
}
param {
lr_mult: 2
decay_mult: 0
}
convolution_param {
num_output: 384
pad: 1
kernel_size: 3
weight_filler {
type: "gaussian"
std: 0.01
}
bias_filler {
type: "constant"
value: 0
}
}
}
layer {
name: "relu3"
type: "ReLU"
bottom: "conv3"
top: "conv3"
}
layer {
name: "conv4"
type: "Convolution"
bottom: "conv3"
top: "conv4"
param {
lr_mult: 1
decay_mult: 1
}
param {
lr_mult: 2
decay_mult: 0
}
convolution_param {
num_output: 384
pad: 1
kernel_size: 3
group: 2
weight_filler {
type: "gaussian"
std: 0.01
}
bias_filler {
type: "constant"
value: 0.1
}
}
}
layer {
name: "relu4"
type: "ReLU"
bottom: "conv4"
top: "conv4"
}
layer {
name: "conv5"
type: "Convolution"
bottom: "conv4"
top: "conv5"
param {
lr_mult: 1
decay_mult: 1
}
param {
lr_mult: 2
decay_mult: 0
}
convolution_param {
num_output: 256
pad: 1
kernel_size: 3
group: 2
weight_filler {
type: "xavier"
std: 0.01
}
bias_filler {
type: "constant"
value: 0.1
}
}
}
layer {
name: "relu5"
type: "ReLU"
bottom: "conv5"
top: "conv5"
}
layer {
name: "pool5"
type: "Pooling"
bottom: "conv5"
top: "pool5"
pooling_param {
pool: MAX
kernel_size: 3
stride: 2
}
}
layer {
name: "fc6"
type: "InnerProduct"
bottom: "pool5"
top: "fc6"
param {
lr_mult: 1
decay_mult: 1
}
param {
lr_mult: 2
decay_mult: 0
}
inner_product_param {
num_output: 4096
weight_filler {
type: "xavier"
std: 0.005
}
bias_filler {
type: "constant"
value: 0.1
}
}
}
layer {
name: "relu6"
type: "ReLU"
bottom: "fc6"
top: "fc6"
}
layer {
name: "drop6"
type: "Dropout"
bottom: "fc6"
top: "fc6"
dropout_param {
dropout_ratio: 0.5
}
}
layer {
name: "fc7"
type: "InnerProduct"
bottom: "fc6"
top: "fc7"
param {
lr_mult: 1
decay_mult: 1
}
param {
lr_mult: 2
decay_mult: 0
}
inner_product_param {
num_output: 4096
weight_filler {
type: "xavier"
std: 0.005
}
bias_filler {
type: "constant"
value: 0.1
}
}
}
layer {
name: "relu7"
type: "ReLU"
bottom: "fc7"
top: "fc7"
}
layer {
name: "drop7"
type: "Dropout"
bottom: "fc7"
top: "fc7"
dropout_param {
dropout_ratio: 0.5
}
}
layer {
name: "fc8"
type: "InnerProduct"
bottom: "fc7"
top: "fc8"
param {
lr_mult: 1
decay_mult: 1
}
param {
lr_mult: 2
decay_mult: 0
}
inner_product_param {
num_output: 4
weight_filler {
type: "xavier"
std: 0.01
}
bias_filler {
type: "constant"
value: 0
}
}
}
layer {
name: "accuracy_training"
type: "Accuracy"
bottom: "fc8"
bottom: "label"
top: "accuracy_training"
include {
phase: TRAIN
}
}
layer {
name: "loss"
type: "SoftmaxWithLoss"
bottom: "fc8"
bottom: "label"
top: "loss"
}
I0312 23:23:10.863792 2676 layer_factory.hpp:77] Creating layer data
I0312 23:23:10.990649 2676 db_lmdb.cpp:35] Opened lmdb examples/Mydataset_train_lmdb
I0312 23:23:11.008492 2676 net.cpp:86] Creating Layer data
I0312 23:23:11.008553 2676 net.cpp:382] data -> data
I0312 23:23:11.008602 2676 net.cpp:382] data -> label
I0312 23:23:11.008659 2676 data_transformer.cpp:25] Loading mean file from: examples/Mydataset_train_lmdb/mean_imagetest.binaryproto
I0312 23:23:11.035311 2676 data_layer.cpp:45] output data size: 256,3,227,227
I0312 23:23:11.238029 2676 net.cpp:124] Setting up data
I0312 23:23:11.238059 2676 net.cpp:131] Top shape: 256 3 227 227 (39574272)
I0312 23:23:11.238076 2676 net.cpp:131] Top shape: 256 (256)
I0312 23:23:11.238080 2676 net.cpp:139] Memory required for data: 158298112
I0312 23:23:11.238086 2676 layer_factory.hpp:77] Creating layer label_data_1_split
I0312 23:23:11.238109 2676 net.cpp:86] Creating Layer label_data_1_split
I0312 23:23:11.238113 2676 net.cpp:408] label_data_1_split <- label
I0312 23:23:11.238123 2676 net.cpp:382] label_data_1_split -> label_data_1_split_0
I0312 23:23:11.238131 2676 net.cpp:382] label_data_1_split -> label_data_1_split_1
I0312 23:23:11.238162 2676 net.cpp:124] Setting up label_data_1_split
I0312 23:23:11.238167 2676 net.cpp:131] Top shape: 256 (256)
I0312 23:23:11.238169 2676 net.cpp:131] Top shape: 256 (256)
I0312 23:23:11.238170 2676 net.cpp:139] Memory required for data: 158300160
I0312 23:23:11.238173 2676 layer_factory.hpp:77] Creating layer conv1
I0312 23:23:11.238183 2676 net.cpp:86] Creating Layer conv1
I0312 23:23:11.238185 2676 net.cpp:408] conv1 <- data
I0312 23:23:11.238189 2676 net.cpp:382] conv1 -> conv1
I0312 23:23:15.377827 2676 net.cpp:124] Setting up conv1
I0312 23:23:15.377876 2676 net.cpp:131] Top shape: 256 96 55 55 (74342400)
I0312 23:23:15.377888 2676 net.cpp:139] Memory required for data: 455669760
I0312 23:23:15.377938 2676 layer_factory.hpp:77] Creating layer relu1
I0312 23:23:15.377962 2676 net.cpp:86] Creating Layer relu1
I0312 23:23:15.377979 2676 net.cpp:408] relu1 <- conv1
I0312 23:23:15.377995 2676 net.cpp:369] relu1 -> conv1 (in-place)
I0312 23:23:15.378617 2676 net.cpp:124] Setting up relu1
I0312 23:23:15.378645 2676 net.cpp:131] Top shape: 256 96 55 55 (74342400)
I0312 23:23:15.378654 2676 net.cpp:139] Memory required for data: 753039360
I0312 23:23:15.378662 2676 layer_factory.hpp:77] Creating layer norm1
I0312 23:23:15.378685 2676 net.cpp:86] Creating Layer norm1
I0312 23:23:15.378693 2676 net.cpp:408] norm1 <- conv1
I0312 23:23:15.378708 2676 net.cpp:382] norm1 -> norm1
I0312 23:23:15.379875 2676 net.cpp:124] Setting up norm1
I0312 23:23:15.379904 2676 net.cpp:131] Top shape: 256 96 55 55 (74342400)
I0312 23:23:15.379911 2676 net.cpp:139] Memory required for data: 1050408960
I0312 23:23:15.379918 2676 layer_factory.hpp:77] Creating layer pool1
I0312 23:23:15.379935 2676 net.cpp:86] Creating Layer pool1
I0312 23:23:15.379943 2676 net.cpp:408] pool1 <- norm1
I0312 23:23:15.379958 2676 net.cpp:382] pool1 -> pool1
I0312 23:23:15.380061 2676 net.cpp:124] Setting up pool1
I0312 23:23:15.380084 2676 net.cpp:131] Top shape: 256 96 27 27 (17915904)
I0312 23:23:15.380090 2676 net.cpp:139] Memory required for data: 1122072576
I0312 23:23:15.380096 2676 layer_factory.hpp:77] Creating layer conv2
I0312 23:23:15.380120 2676 net.cpp:86] Creating Layer conv2
I0312 23:23:15.380129 2676 net.cpp:408] conv2 <- pool1
I0312 23:23:15.380141 2676 net.cpp:382] conv2 -> conv2
I0312 23:23:15.395615 2676 net.cpp:124] Setting up conv2
I0312 23:23:15.395655 2676 net.cpp:131] Top shape: 256 256 27 27 (47775744)
I0312 23:23:15.395663 2676 net.cpp:139] Memory required for data: 1313175552
I0312 23:23:15.395686 2676 layer_factory.hpp:77] Creating layer relu2
I0312 23:23:15.395700 2676 net.cpp:86] Creating Layer relu2
I0312 23:23:15.395706 2676 net.cpp:408] relu2 <- conv2
I0312 23:23:15.395717 2676 net.cpp:369] relu2 -> conv2 (in-place)
I0312 23:23:15.396646 2676 net.cpp:124] Setting up relu2
I0312 23:23:15.396669 2676 net.cpp:131] Top shape: 256 256 27 27 (47775744)
I0312 23:23:15.396675 2676 net.cpp:139] Memory required for data: 1504278528
I0312 23:23:15.396682 2676 layer_factory.hpp:77] Creating layer norm2
I0312 23:23:15.396697 2676 net.cpp:86] Creating Layer norm2
I0312 23:23:15.396703 2676 net.cpp:408] norm2 <- conv2
I0312 23:23:15.396713 2676 net.cpp:382] norm2 -> norm2
I0312 23:23:15.397075 2676 net.cpp:124] Setting up norm2
I0312 23:23:15.397091 2676 net.cpp:131] Top shape: 256 256 27 27 (47775744)
I0312 23:23:15.397096 2676 net.cpp:139] Memory required for data: 1695381504
I0312 23:23:15.397100 2676 layer_factory.hpp:77] Creating layer pool2
I0312 23:23:15.397111 2676 net.cpp:86] Creating Layer pool2
I0312 23:23:15.397116 2676 net.cpp:408] pool2 <- norm2
I0312 23:23:15.397125 2676 net.cpp:382] pool2 -> pool2
I0312 23:23:15.397186 2676 net.cpp:124] Setting up pool2
I0312 23:23:15.397199 2676 net.cpp:131] Top shape: 256 256 13 13 (11075584)
I0312 23:23:15.397203 2676 net.cpp:139] Memory required for data: 1739683840
I0312 23:23:15.397208 2676 layer_factory.hpp:77] Creating layer conv3
I0312 23:23:15.397222 2676 net.cpp:86] Creating Layer conv3
I0312 23:23:15.397229 2676 net.cpp:408] conv3 <- pool2
I0312 23:23:15.397238 2676 net.cpp:382] conv3 -> conv3
I0312 23:23:15.411098 2676 net.cpp:124] Setting up conv3
I0312 23:23:15.411116 2676 net.cpp:131] Top shape: 256 384 13 13 (16613376)
I0312 23:23:15.411120 2676 net.cpp:139] Memory required for data: 1806137344
I0312 23:23:15.411131 2676 layer_factory.hpp:77] Creating layer relu3
I0312 23:23:15.411139 2676 net.cpp:86] Creating Layer relu3
I0312 23:23:15.411144 2676 net.cpp:408] relu3 <- conv3
I0312 23:23:15.411149 2676 net.cpp:369] relu3 -> conv3 (in-place)
I0312 23:23:15.411316 2676 net.cpp:124] Setting up relu3
I0312 23:23:15.411324 2676 net.cpp:131] Top shape: 256 384 13 13 (16613376)
I0312 23:23:15.411326 2676 net.cpp:139] Memory required for data: 1872590848
I0312 23:23:15.411329 2676 layer_factory.hpp:77] Creating layer conv4
I0312 23:23:15.411339 2676 net.cpp:86] Creating Layer conv4
I0312 23:23:15.411340 2676 net.cpp:408] conv4 <- conv3
I0312 23:23:15.411360 2676 net.cpp:382] conv4 -> conv4
I0312 23:23:15.419447 2676 net.cpp:124] Setting up conv4
I0312 23:23:15.419461 2676 net.cpp:131] Top shape: 256 384 13 13 (16613376)
I0312 23:23:15.419462 2676 net.cpp:139] Memory required for data: 1939044352
I0312 23:23:15.419467 2676 layer_factory.hpp:77] Creating layer relu4
I0312 23:23:15.419473 2676 net.cpp:86] Creating Layer relu4
I0312 23:23:15.419476 2676 net.cpp:408] relu4 <- conv4
I0312 23:23:15.419479 2676 net.cpp:369] relu4 -> conv4 (in-place)
I0312 23:23:15.419606 2676 net.cpp:124] Setting up relu4
I0312 23:23:15.419612 2676 net.cpp:131] Top shape: 256 384 13 13 (16613376)
I0312 23:23:15.419615 2676 net.cpp:139] Memory required for data: 2005497856
I0312 23:23:15.419615 2676 layer_factory.hpp:77] Creating layer conv5
I0312 23:23:15.419622 2676 net.cpp:86] Creating Layer conv5
I0312 23:23:15.419625 2676 net.cpp:408] conv5 <- conv4
I0312 23:23:15.419628 2676 net.cpp:382] conv5 -> conv5
I0312 23:23:15.424048 2676 net.cpp:124] Setting up conv5
I0312 23:23:15.424058 2676 net.cpp:131] Top shape: 256 256 13 13 (11075584)
I0312 23:23:15.424062 2676 net.cpp:139] Memory required for data: 2049800192
I0312 23:23:15.424083 2676 layer_factory.hpp:77] Creating layer relu5
I0312 23:23:15.424088 2676 net.cpp:86] Creating Layer relu5
I0312 23:23:15.424090 2676 net.cpp:408] relu5 <- conv5
I0312 23:23:15.424094 2676 net.cpp:369] relu5 -> conv5 (in-place)
I0312 23:23:15.424257 2676 net.cpp:124] Setting up relu5
I0312 23:23:15.424263 2676 net.cpp:131] Top shape: 256 256 13 13 (11075584)
I0312 23:23:15.424264 2676 net.cpp:139] Memory required for data: 2094102528
I0312 23:23:15.424266 2676 layer_factory.hpp:77] Creating layer pool5
I0312 23:23:15.424271 2676 net.cpp:86] Creating Layer pool5
I0312 23:23:15.424273 2676 net.cpp:408] pool5 <- conv5
I0312 23:23:15.424291 2676 net.cpp:382] pool5 -> pool5
I0312 23:23:15.424348 2676 net.cpp:124] Setting up pool5
I0312 23:23:15.424352 2676 net.cpp:131] Top shape: 256 256 6 6 (2359296)
I0312 23:23:15.424355 2676 net.cpp:139] Memory required for data: 2103539712
I0312 23:23:15.424355 2676 layer_factory.hpp:77] Creating layer fc6
I0312 23:23:15.424374 2676 net.cpp:86] Creating Layer fc6
I0312 23:23:15.424377 2676 net.cpp:408] fc6 <- pool5
I0312 23:23:15.424397 2676 net.cpp:382] fc6 -> fc6
I0312 23:23:15.601073 2676 net.cpp:124] Setting up fc6
I0312 23:23:15.601089 2676 net.cpp:131] Top shape: 256 4096 (1048576)
I0312 23:23:15.601091 2676 net.cpp:139] Memory required for data: 2107734016
I0312 23:23:15.601097 2676 layer_factory.hpp:77] Creating layer relu6
I0312 23:23:15.601117 2676 net.cpp:86] Creating Layer relu6
I0312 23:23:15.601121 2676 net.cpp:408] relu6 <- fc6
I0312 23:23:15.601125 2676 net.cpp:369] relu6 -> fc6 (in-place)
I0312 23:23:15.601327 2676 net.cpp:124] Setting up relu6
I0312 23:23:15.601332 2676 net.cpp:131] Top shape: 256 4096 (1048576)
I0312 23:23:15.601335 2676 net.cpp:139] Memory required for data: 2111928320
I0312 23:23:15.601336 2676 layer_factory.hpp:77] Creating layer drop6
I0312 23:23:15.601359 2676 net.cpp:86] Creating Layer drop6
I0312 23:23:15.601361 2676 net.cpp:408] drop6 <- fc6
I0312 23:23:15.601366 2676 net.cpp:369] drop6 -> fc6 (in-place)
I0312 23:23:15.601413 2676 net.cpp:124] Setting up drop6
I0312 23:23:15.601418 2676 net.cpp:131] Top shape: 256 4096 (1048576)
I0312 23:23:15.601419 2676 net.cpp:139] Memory required for data: 2116122624
I0312 23:23:15.601433 2676 layer_factory.hpp:77] Creating layer fc7
I0312 23:23:15.601438 2676 net.cpp:86] Creating Layer fc7
I0312 23:23:15.601439 2676 net.cpp:408] fc7 <- fc6
I0312 23:23:15.601456 2676 net.cpp:382] fc7 -> fc7
I0312 23:23:15.681016 2676 net.cpp:124] Setting up fc7
I0312 23:23:15.681032 2676 net.cpp:131] Top shape: 256 4096 (1048576)
I0312 23:23:15.681035 2676 net.cpp:139] Memory required for data: 2120316928
I0312 23:23:15.681041 2676 layer_factory.hpp:77] Creating layer relu7
I0312 23:23:15.681061 2676 net.cpp:86] Creating Layer relu7
I0312 23:23:15.681063 2676 net.cpp:408] relu7 <- fc7
I0312 23:23:15.681114 2676 net.cpp:369] relu7 -> fc7 (in-place)
I0312 23:23:15.681531 2676 net.cpp:124] Setting up relu7
I0312 23:23:15.681540 2676 net.cpp:131] Top shape: 256 4096 (1048576)
I0312 23:23:15.681541 2676 net.cpp:139] Memory required for data: 2124511232
I0312 23:23:15.681543 2676 layer_factory.hpp:77] Creating layer drop7
I0312 23:23:15.681547 2676 net.cpp:86] Creating Layer drop7
I0312 23:23:15.681550 2676 net.cpp:408] drop7 <- fc7
I0312 23:23:15.681566 2676 net.cpp:369] drop7 -> fc7 (in-place)
I0312 23:23:15.681596 2676 net.cpp:124] Setting up drop7
I0312 23:23:15.681617 2676 net.cpp:131] Top shape: 256 4096 (1048576)
I0312 23:23:15.681620 2676 net.cpp:139] Memory required for data: 2128705536
I0312 23:23:15.681622 2676 layer_factory.hpp:77] Creating layer fc8
I0312 23:23:15.681638 2676 net.cpp:86] Creating Layer fc8
I0312 23:23:15.681641 2676 net.cpp:408] fc8 <- fc7
I0312 23:23:15.681644 2676 net.cpp:382] fc8 -> fc8
I0312 23:23:15.682404 2676 net.cpp:124] Setting up fc8
I0312 23:23:15.682410 2676 net.cpp:131] Top shape: 256 4 (1024)
I0312 23:23:15.682412 2676 net.cpp:139] Memory required for data: 2128709632
I0312 23:23:15.682416 2676 layer_factory.hpp:77] Creating layer fc8_fc8_0_split
I0312 23:23:15.682420 2676 net.cpp:86] Creating Layer fc8_fc8_0_split
I0312 23:23:15.682436 2676 net.cpp:408] fc8_fc8_0_split <- fc8
I0312 23:23:15.682440 2676 net.cpp:382] fc8_fc8_0_split -> fc8_fc8_0_split_0
I0312 23:23:15.682445 2676 net.cpp:382] fc8_fc8_0_split -> fc8_fc8_0_split_1
I0312 23:23:15.682493 2676 net.cpp:124] Setting up fc8_fc8_0_split
I0312 23:23:15.682497 2676 net.cpp:131] Top shape: 256 4 (1024)
I0312 23:23:15.682499 2676 net.cpp:131] Top shape: 256 4 (1024)
I0312 23:23:15.682502 2676 net.cpp:139] Memory required for data: 2128717824
I0312 23:23:15.682503 2676 layer_factory.hpp:77] Creating layer accuracy_training
I0312 23:23:15.682507 2676 net.cpp:86] Creating Layer accuracy_training
I0312 23:23:15.682523 2676 net.cpp:408] accuracy_training <- fc8_fc8_0_split_0
I0312 23:23:15.682526 2676 net.cpp:408] accuracy_training <- label_data_1_split_0
I0312 23:23:15.682543 2676 net.cpp:382] accuracy_training -> accuracy_training
I0312 23:23:15.682548 2676 net.cpp:124] Setting up accuracy_training
I0312 23:23:15.682564 2676 net.cpp:131] Top shape: (1)
I0312 23:23:15.682567 2676 net.cpp:139] Memory required for data: 2128717828
I0312 23:23:15.682569 2676 layer_factory.hpp:77] Creating layer loss
I0312 23:23:15.682572 2676 net.cpp:86] Creating Layer loss
I0312 23:23:15.682588 2676 net.cpp:408] loss <- fc8_fc8_0_split_1
I0312 23:23:15.682592 2676 net.cpp:408] loss <- label_data_1_split_1
I0312 23:23:15.682595 2676 net.cpp:382] loss -> loss
I0312 23:23:15.682615 2676 layer_factory.hpp:77] Creating layer loss
I0312 23:23:15.682878 2676 net.cpp:124] Setting up loss
I0312 23:23:15.682898 2676 net.cpp:131] Top shape: (1)
I0312 23:23:15.682898 2676 net.cpp:134] with loss weight 1
I0312 23:23:15.682929 2676 net.cpp:139] Memory required for data: 2128717832
I0312 23:23:15.682930 2676 net.cpp:200] loss needs backward computation.
I0312 23:23:15.682936 2676 net.cpp:202] accuracy_training does not need backward computation.
I0312 23:23:15.682951 2676 net.cpp:200] fc8_fc8_0_split needs backward computation.
I0312 23:23:15.682953 2676 net.cpp:200] fc8 needs backward computation.
I0312 23:23:15.682955 2676 net.cpp:200] drop7 needs backward computation.
I0312 23:23:15.682957 2676 net.cpp:200] relu7 needs backward computation.
I0312 23:23:15.682973 2676 net.cpp:200] fc7 needs backward computation.
I0312 23:23:15.682976 2676 net.cpp:200] drop6 needs backward computation.
I0312 23:23:15.682976 2676 net.cpp:200] relu6 needs backward computation.
I0312 23:23:15.682979 2676 net.cpp:200] fc6 needs backward computation.
I0312 23:23:15.682981 2676 net.cpp:200] pool5 needs backward computation.
I0312 23:23:15.682983 2676 net.cpp:200] relu5 needs backward computation.
I0312 23:23:15.682986 2676 net.cpp:200] conv5 needs backward computation.
I0312 23:23:15.682987 2676 net.cpp:200] relu4 needs backward computation.
I0312 23:23:15.682998 2676 net.cpp:200] conv4 needs backward computation.
I0312 23:23:15.683001 2676 net.cpp:200] relu3 needs backward computation.
I0312 23:23:15.683002 2676 net.cpp:200] conv3 needs backward computation.
I0312 23:23:15.683006 2676 net.cpp:200] pool2 needs backward computation.
I0312 23:23:15.683007 2676 net.cpp:200] norm2 needs backward computation.
I0312 23:23:15.683010 2676 net.cpp:200] relu2 needs backward computation.
I0312 23:23:15.683012 2676 net.cpp:200] conv2 needs backward computation.
I0312 23:23:15.683014 2676 net.cpp:200] pool1 needs backward computation.
I0312 23:23:15.683017 2676 net.cpp:200] norm1 needs backward computation.
I0312 23:23:15.683018 2676 net.cpp:200] relu1 needs backward computation.
I0312 23:23:15.683020 2676 net.cpp:200] conv1 needs backward computation.
I0312 23:23:15.683023 2676 net.cpp:202] label_data_1_split does not need backward computation.
I0312 23:23:15.683025 2676 net.cpp:202] data does not need backward computation.
I0312 23:23:15.683027 2676 net.cpp:244] This network produces output accuracy_training
I0312 23:23:15.683029 2676 net.cpp:244] This network produces output loss
I0312 23:23:15.683043 2676 net.cpp:257] Network initialization done.
I0312 23:23:15.683274 2676 solver.cpp:173] Creating test net (#0) specified by net file: examples/alexnetfinetune/train_valsina.prototxt
I0312 23:23:15.683326 2676 net.cpp:296] The NetState phase (1) differed from the phase (0) specified by a rule in layer data
I0312 23:23:15.683336 2676 net.cpp:296] The NetState phase (1) differed from the phase (0) specified by a rule in layer accuracy_training
I0312 23:23:15.683465 2676 net.cpp:53] Initializing net from parameters:
name: "AlexNet"
state {
phase: TEST
}
layer {
name: "data"
type: "Data"
top: "data"
top: "label"
include {
phase: TEST
}
transform_param {
mirror: false
crop_size: 227
mean_file: "examples/Mydataset_test_lmdb/mean_imagetest.binaryproto"
}
data_param {
source: "examples/Mydataset_test_lmdb"
batch_size: 50
backend: LMDB
}
}
layer {
name: "conv1"
type: "Convolution"
bottom: "data"
top: "conv1"
param {
lr_mult: 1
decay_mult: 1
}
param {
lr_mult: 2
decay_mult: 0
}
convolution_param {
num_output: 96
kernel_size: 11
stride: 4
weight_filler {
type: "gaussian"
std: 0.01
}
bias_filler {
type: "constant"
value: 0
}
}
}
layer {
name: "relu1"
type: "ReLU"
bottom: "conv1"
top: "conv1"
}
layer {
name: "norm1"
type: "LRN"
bottom: "conv1"
top: "norm1"
lrn_param {
local_size: 5
alpha: 0.0001
beta: 0.75
}
}
layer {
name: "pool1"
type: "Pooling"
bottom: "norm1"
top: "pool1"
pooling_param {
pool: MAX
kernel_size: 3
stride: 2
}
}
layer {
name: "conv2"
type: "Convolution"
bottom: "pool1"
top: "conv2"
param {
lr_mult: 1
decay_mult: 1
}
param {
lr_mult: 2
decay_mult: 0
}
convolution_param {
num_output: 256
pad: 2
kernel_size: 5
group: 2
weight_filler {
type: "gaussian"
std: 0.01
}
bias_filler {
type: "constant"
value: 0.1
}
}
}
layer {
name: "relu2"
type: "ReLU"
bottom: "conv2"
top: "conv2"
}
layer {
name: "norm2"
type: "LRN"
bottom: "conv2"
top: "norm2"
lrn_param {
local_size: 5
alpha: 0.0001
beta: 0.75
}
}
layer {
name: "pool2"
type: "Pooling"
bottom: "norm2"
top: "pool2"
pooling_param {
pool: MAX
kernel_size: 3
stride: 2
}
}
layer {
name: "conv3"
type: "Convolution"
bottom: "pool2"
top: "conv3"
param {
lr_mult: 1
decay_mult: 1
}
param {
lr_mult: 2
decay_mult: 0
}
convolution_param {
num_output: 384
pad: 1
kernel_size: 3
weight_filler {
type: "gaussian"
std: 0.01
}
bias_filler {
type: "constant"
value: 0
}
}
}
layer {
name: "relu3"
type: "ReLU"
bottom: "conv3"
top: "conv3"
}
layer {
name: "conv4"
type: "Convolution"
bottom: "conv3"
top: "conv4"
param {
lr_mult: 1
decay_mult: 1
}
param {
lr_mult: 2
decay_mult: 0
}
convolution_param {
num_output: 384
pad: 1
kernel_size: 3
group: 2
weight_filler {
type: "gaussian"
std: 0.01
}
bias_filler {
type: "constant"
value: 0.1
}
}
}
layer {
name: "relu4"
type: "ReLU"
bottom: "conv4"
top: "conv4"
}
layer {
name: "conv5"
type: "Convolution"
bottom: "conv4"
top: "conv5"
param {
lr_mult: 1
decay_mult: 1
}
param {
lr_mult: 2
decay_mult: 0
}
convolution_param {
num_output: 256
pad: 1
kernel_size: 3
group: 2
weight_filler {
type: "xavier"
std: 0.01
}
bias_filler {
type: "constant"
value: 0.1
}
}
}
layer {
name: "relu5"
type: "ReLU"
bottom: "conv5"
top: "conv5"
}
layer {
name: "pool5"
type: "Pooling"
bottom: "conv5"
top: "pool5"
pooling_param {
pool: MAX
kernel_size: 3
stride: 2
}
}
layer {
name: "fc6"
type: "InnerProduct"
bottom: "pool5"
top: "fc6"
param {
lr_mult: 1
decay_mult: 1
}
param {
lr_mult: 2
decay_mult: 0
}
inner_product_param {
num_output: 4096
weight_filler {
type: "xavier"
std: 0.005
}
bias_filler {
type: "constant"
value: 0.1
}
}
}
layer {
name: "relu6"
type: "ReLU"
bottom: "fc6"
top: "fc6"
}
layer {
name: "drop6"
type: "Dropout"
bottom: "fc6"
top: "fc6"
dropout_param {
dropout_ratio: 0.5
}
}
layer {
name: "fc7"
type: "InnerProduct"
bottom: "fc6"
top: "fc7"
param {
lr_mult: 1
decay_mult: 1
}
param {
lr_mult: 2
decay_mult: 0
}
inner_product_param {
num_output: 4096
weight_filler {
type: "xavier"
std: 0.005
}
bias_filler {
type: "constant"
value: 0.1
}
}
}
layer {
name: "relu7"
type: "ReLU"
bottom: "fc7"
top: "fc7"
}
layer {
name: "drop7"
type: "Dropout"
bottom: "fc7"
top: "fc7"
dropout_param {
dropout_ratio: 0.5
}
}
layer {
name: "fc8"
type: "InnerProduct"
bottom: "fc7"
top: "fc8"
param {
lr_mult: 1
decay_mult: 1
}
param {
lr_mult: 2
decay_mult: 0
}
inner_product_param {
num_output: 4
weight_filler {
type: "xavier"
std: 0.01
}
bias_filler {
type: "constant"
value: 0
}
}
}
layer {
name: "accuracy"
type: "Accuracy"
bottom: "fc8"
bottom: "label"
top: "accuracy"
include {
phase: TEST
}
}
layer {
name: "loss"
type: "SoftmaxWithLoss"
bottom: "fc8"
bottom: "label"
top: "loss"
}
I0312 23:23:15.683547 2676 layer_factory.hpp:77] Creating layer data
I0312 23:23:15.724110 2676 db_lmdb.cpp:35] Opened lmdb examples/Mydataset_test_lmdb
I0312 23:23:15.734433 2676 net.cpp:86] Creating Layer data
I0312 23:23:15.734490 2676 net.cpp:382] data -> data
I0312 23:23:15.734519 2676 net.cpp:382] data -> label
I0312 23:23:15.734539 2676 data_transformer.cpp:25] Loading mean file from: examples/Mydataset_test_lmdb/mean_imagetest.binaryproto
I0312 23:23:15.775971 2676 data_layer.cpp:45] output data size: 50,3,227,227
I0312 23:23:15.841073 2676 net.cpp:124] Setting up data
I0312 23:23:15.841104 2676 net.cpp:131] Top shape: 50 3 227 227 (7729350)
I0312 23:23:15.841121 2676 net.cpp:131] Top shape: 50 (50)
I0312 23:23:15.841122 2676 net.cpp:139] Memory required for data: 30917600
I0312 23:23:15.841127 2676 layer_factory.hpp:77] Creating layer label_data_1_split
I0312 23:23:15.841150 2676 net.cpp:86] Creating Layer label_data_1_split
I0312 23:23:15.841153 2676 net.cpp:408] label_data_1_split <- label
I0312 23:23:15.841158 2676 net.cpp:382] label_data_1_split -> label_data_1_split_0
I0312 23:23:15.841166 2676 net.cpp:382] label_data_1_split -> label_data_1_split_1
I0312 23:23:15.841298 2676 net.cpp:124] Setting up label_data_1_split
I0312 23:23:15.841322 2676 net.cpp:131] Top shape: 50 (50)
I0312 23:23:15.841325 2676 net.cpp:131] Top shape: 50 (50)
I0312 23:23:15.841352 2676 net.cpp:139] Memory required for data: 30918000
I0312 23:23:15.841354 2676 layer_factory.hpp:77] Creating layer conv1
I0312 23:23:15.841364 2676 net.cpp:86] Creating Layer conv1
I0312 23:23:15.841378 2676 net.cpp:408] conv1 <- data
I0312 23:23:15.841385 2676 net.cpp:382] conv1 -> conv1
I0312 23:23:15.845085 2676 net.cpp:124] Setting up conv1
I0312 23:23:15.845110 2676 net.cpp:131] Top shape: 50 96 55 55 (14520000)
I0312 23:23:15.845113 2676 net.cpp:139] Memory required for data: 88998000
I0312 23:23:15.845135 2676 layer_factory.hpp:77] Creating layer relu1
I0312 23:23:15.845140 2676 net.cpp:86] Creating Layer relu1
I0312 23:23:15.845142 2676 net.cpp:408] relu1 <- conv1
I0312 23:23:15.845146 2676 net.cpp:369] relu1 -> conv1 (in-place)
I0312 23:23:15.845297 2676 net.cpp:124] Setting up relu1
I0312 23:23:15.845304 2676 net.cpp:131] Top shape: 50 96 55 55 (14520000)
I0312 23:23:15.845319 2676 net.cpp:139] Memory required for data: 147078000
I0312 23:23:15.845320 2676 layer_factory.hpp:77] Creating layer norm1
I0312 23:23:15.845327 2676 net.cpp:86] Creating Layer norm1
I0312 23:23:15.845329 2676 net.cpp:408] norm1 <- conv1
I0312 23:23:15.845346 2676 net.cpp:382] norm1 -> norm1
I0312 23:23:15.845508 2676 net.cpp:124] Setting up norm1
I0312 23:23:15.845515 2676 net.cpp:131] Top shape: 50 96 55 55 (14520000)
I0312 23:23:15.845530 2676 net.cpp:139] Memory required for data: 205158000
I0312 23:23:15.845531 2676 layer_factory.hpp:77] Creating layer pool1
I0312 23:23:15.845536 2676 net.cpp:86] Creating Layer pool1
I0312 23:23:15.845538 2676 net.cpp:408] pool1 <- norm1
I0312 23:23:15.845556 2676 net.cpp:382] pool1 -> pool1
I0312 23:23:15.845613 2676 net.cpp:124] Setting up pool1
I0312 23:23:15.845618 2676 net.cpp:131] Top shape: 50 96 27 27 (3499200)
I0312 23:23:15.845619 2676 net.cpp:139] Memory required for data: 219154800
I0312 23:23:15.845621 2676 layer_factory.hpp:77] Creating layer conv2
I0312 23:23:15.845639 2676 net.cpp:86] Creating Layer conv2
I0312 23:23:15.845641 2676 net.cpp:408] conv2 <- pool1
I0312 23:23:15.845644 2676 net.cpp:382] conv2 -> conv2
I0312 23:23:15.849257 2676 net.cpp:124] Setting up conv2
I0312 23:23:15.849267 2676 net.cpp:131] Top shape: 50 256 27 27 (9331200)
I0312 23:23:15.849269 2676 net.cpp:139] Memory required for data: 256479600
I0312 23:23:15.849277 2676 layer_factory.hpp:77] Creating layer relu2
I0312 23:23:15.849294 2676 net.cpp:86] Creating Layer relu2
I0312 23:23:15.849297 2676 net.cpp:408] relu2 <- conv2
I0312 23:23:15.849299 2676 net.cpp:369] relu2 -> conv2 (in-place)
I0312 23:23:15.849706 2676 net.cpp:124] Setting up relu2
I0312 23:23:15.849714 2676 net.cpp:131] Top shape: 50 256 27 27 (9331200)
I0312 23:23:15.849716 2676 net.cpp:139] Memory required for data: 293804400
I0312 23:23:15.849719 2676 layer_factory.hpp:77] Creating layer norm2
I0312 23:23:15.849740 2676 net.cpp:86] Creating Layer norm2
I0312 23:23:15.849742 2676 net.cpp:408] norm2 <- conv2
I0312 23:23:15.849758 2676 net.cpp:382] norm2 -> norm2
I0312 23:23:15.849923 2676 net.cpp:124] Setting up norm2
I0312 23:23:15.849930 2676 net.cpp:131] Top shape: 50 256 27 27 (9331200)
I0312 23:23:15.849931 2676 net.cpp:139] Memory required for data: 331129200
I0312 23:23:15.849947 2676 layer_factory.hpp:77] Creating layer pool2
I0312 23:23:15.849951 2676 net.cpp:86] Creating Layer pool2
I0312 23:23:15.849954 2676 net.cpp:408] pool2 <- norm2
I0312 23:23:15.849972 2676 net.cpp:382] pool2 -> pool2
I0312 23:23:15.850024 2676 net.cpp:124] Setting up pool2
I0312 23:23:15.850028 2676 net.cpp:131] Top shape: 50 256 13 13 (2163200)
I0312 23:23:15.850042 2676 net.cpp:139] Memory required for data: 339782000
I0312 23:23:15.850044 2676 layer_factory.hpp:77] Creating layer conv3
I0312 23:23:15.850049 2676 net.cpp:86] Creating Layer conv3
I0312 23:23:15.850064 2676 net.cpp:408] conv3 <- pool2
I0312 23:23:15.850069 2676 net.cpp:382] conv3 -> conv3
I0312 23:23:15.858925 2676 net.cpp:124] Setting up conv3
I0312 23:23:15.858939 2676 net.cpp:131] Top shape: 50 384 13 13 (3244800)
I0312 23:23:15.858969 2676 net.cpp:139] Memory required for data: 352761200
I0312 23:23:15.858994 2676 layer_factory.hpp:77] Creating layer relu3
I0312 23:23:15.859000 2676 net.cpp:86] Creating Layer relu3
I0312 23:23:15.859017 2676 net.cpp:408] relu3 <- conv3
I0312 23:23:15.859021 2676 net.cpp:369] relu3 -> conv3 (in-place)
I0312 23:23:15.859175 2676 net.cpp:124] Setting up relu3
I0312 23:23:15.859181 2676 net.cpp:131] Top shape: 50 384 13 13 (3244800)
I0312 23:23:15.859182 2676 net.cpp:139] Memory required for data: 365740400
I0312 23:23:15.859184 2676 layer_factory.hpp:77] Creating layer conv4
I0312 23:23:15.859191 2676 net.cpp:86] Creating Layer conv4
I0312 23:23:15.859207 2676 net.cpp:408] conv4 <- conv3
I0312 23:23:15.859210 2676 net.cpp:382] conv4 -> conv4
I0312 23:23:15.866523 2676 net.cpp:124] Setting up conv4
I0312 23:23:15.866536 2676 net.cpp:131] Top shape: 50 384 13 13 (3244800)
I0312 23:23:15.866539 2676 net.cpp:139] Memory required for data: 378719600
I0312 23:23:15.866544 2676 layer_factory.hpp:77] Creating layer relu4
I0312 23:23:15.866550 2676 net.cpp:86] Creating Layer relu4
I0312 23:23:15.866554 2676 net.cpp:408] relu4 <- conv4
I0312 23:23:15.866557 2676 net.cpp:369] relu4 -> conv4 (in-place)
I0312 23:23:15.866735 2676 net.cpp:124] Setting up relu4
I0312 23:23:15.866740 2676 net.cpp:131] Top shape: 50 384 13 13 (3244800)
I0312 23:23:15.866757 2676 net.cpp:139] Memory required for data: 391698800
I0312 23:23:15.866760 2676 layer_factory.hpp:77] Creating layer conv5
I0312 23:23:15.866766 2676 net.cpp:86] Creating Layer conv5
I0312 23:23:15.866768 2676 net.cpp:408] conv5 <- conv4
I0312 23:23:15.866772 2676 net.cpp:382] conv5 -> conv5
I0312 23:23:15.871409 2676 net.cpp:124] Setting up conv5
I0312 23:23:15.871439 2676 net.cpp:131] Top shape: 50 256 13 13 (2163200)
I0312 23:23:15.871443 2676 net.cpp:139] Memory required for data: 400351600
I0312 23:23:15.871454 2676 layer_factory.hpp:77] Creating layer relu5
I0312 23:23:15.871460 2676 net.cpp:86] Creating Layer relu5
I0312 23:23:15.871464 2676 net.cpp:408] relu5 <- conv5
I0312 23:23:15.871469 2676 net.cpp:369] relu5 -> conv5 (in-place)
I0312 23:23:15.871618 2676 net.cpp:124] Setting up relu5
I0312 23:23:15.871624 2676 net.cpp:131] Top shape: 50 256 13 13 (2163200)
I0312 23:23:15.871640 2676 net.cpp:139] Memory required for data: 409004400
I0312 23:23:15.871642 2676 layer_factory.hpp:77] Creating layer pool5
I0312 23:23:15.871649 2676 net.cpp:86] Creating Layer pool5
I0312 23:23:15.871651 2676 net.cpp:408] pool5 <- conv5
I0312 23:23:15.871655 2676 net.cpp:382] pool5 -> pool5
I0312 23:23:15.871695 2676 net.cpp:124] Setting up pool5
I0312 23:23:15.871712 2676 net.cpp:131] Top shape: 50 256 6 6 (460800)
I0312 23:23:15.871714 2676 net.cpp:139] Memory required for data: 410847600
I0312 23:23:15.871716 2676 layer_factory.hpp:77] Creating layer fc6
I0312 23:23:15.871721 2676 net.cpp:86] Creating Layer fc6
I0312 23:23:15.871722 2676 net.cpp:408] fc6 <- pool5
I0312 23:23:15.871726 2676 net.cpp:382] fc6 -> fc6
I0312 23:23:16.052088 2676 net.cpp:124] Setting up fc6
I0312 23:23:16.052119 2676 net.cpp:131] Top shape: 50 4096 (204800)
I0312 23:23:16.052122 2676 net.cpp:139] Memory required for data: 411666800
I0312 23:23:16.052131 2676 layer_factory.hpp:77] Creating layer relu6
I0312 23:23:16.052139 2676 net.cpp:86] Creating Layer relu6
I0312 23:23:16.052141 2676 net.cpp:408] relu6 <- fc6
I0312 23:23:16.052148 2676 net.cpp:369] relu6 -> fc6 (in-place)
I0312 23:23:16.052384 2676 net.cpp:124] Setting up relu6
I0312 23:23:16.052388 2676 net.cpp:131] Top shape: 50 4096 (204800)
I0312 23:23:16.052390 2676 net.cpp:139] Memory required for data: 412486000
I0312 23:23:16.052392 2676 layer_factory.hpp:77] Creating layer drop6
I0312 23:23:16.052397 2676 net.cpp:86] Creating Layer drop6
I0312 23:23:16.052398 2676 net.cpp:408] drop6 <- fc6
I0312 23:23:16.052414 2676 net.cpp:369] drop6 -> fc6 (in-place)
I0312 23:23:16.052464 2676 net.cpp:124] Setting up drop6
I0312 23:23:16.052467 2676 net.cpp:131] Top shape: 50 4096 (204800)
I0312 23:23:16.052494 2676 net.cpp:139] Memory required for data: 413305200
I0312 23:23:16.052510 2676 layer_factory.hpp:77] Creating layer fc7
I0312 23:23:16.052515 2676 net.cpp:86] Creating Layer fc7
I0312 23:23:16.052516 2676 net.cpp:408] fc7 <- fc6
I0312 23:23:16.052520 2676 net.cpp:382] fc7 -> fc7
I0312 23:23:16.134677 2676 net.cpp:124] Setting up fc7
I0312 23:23:16.134694 2676 net.cpp:131] Top shape: 50 4096 (204800)
I0312 23:23:16.134696 2676 net.cpp:139] Memory required for data: 414124400
I0312 23:23:16.134716 2676 layer_factory.hpp:77] Creating layer relu7
I0312 23:23:16.134722 2676 net.cpp:86] Creating Layer relu7
I0312 23:23:16.134726 2676 net.cpp:408] relu7 <- fc7
I0312 23:23:16.134730 2676 net.cpp:369] relu7 -> fc7 (in-place)
I0312 23:23:16.135211 2676 net.cpp:124] Setting up relu7
I0312 23:23:16.135232 2676 net.cpp:131] Top shape: 50 4096 (204800)
I0312 23:23:16.135236 2676 net.cpp:139] Memory required for data: 414943600
I0312 23:23:16.135237 2676 layer_factory.hpp:77] Creating layer drop7
I0312 23:23:16.135241 2676 net.cpp:86] Creating Layer drop7
I0312 23:23:16.135243 2676 net.cpp:408] drop7 <- fc7
I0312 23:23:16.135262 2676 net.cpp:369] drop7 -> fc7 (in-place)
I0312 23:23:16.135287 2676 net.cpp:124] Setting up drop7
I0312 23:23:16.135303 2676 net.cpp:131] Top shape: 50 4096 (204800)
I0312 23:23:16.135304 2676 net.cpp:139] Memory required for data: 415762800
I0312 23:23:16.135306 2676 layer_factory.hpp:77] Creating layer fc8
I0312 23:23:16.135311 2676 net.cpp:86] Creating Layer fc8
I0312 23:23:16.135313 2676 net.cpp:408] fc8 <- fc7
I0312 23:23:16.135315 2676 net.cpp:382] fc8 -> fc8
I0312 23:23:16.135507 2676 net.cpp:124] Setting up fc8
I0312 23:23:16.135512 2676 net.cpp:131] Top shape: 50 4 (200)
I0312 23:23:16.135514 2676 net.cpp:139] Memory required for data: 415763600
I0312 23:23:16.135517 2676 layer_factory.hpp:77] Creating layer fc8_fc8_0_split
I0312 23:23:16.135560 2676 net.cpp:86] Creating Layer fc8_fc8_0_split
I0312 23:23:16.135562 2676 net.cpp:408] fc8_fc8_0_split <- fc8
I0312 23:23:16.135566 2676 net.cpp:382] fc8_fc8_0_split -> fc8_fc8_0_split_0
I0312 23:23:16.135571 2676 net.cpp:382] fc8_fc8_0_split -> fc8_fc8_0_split_1
I0312 23:23:16.135610 2676 net.cpp:124] Setting up fc8_fc8_0_split
I0312 23:23:16.135617 2676 net.cpp:131] Top shape: 50 4 (200)
I0312 23:23:16.135619 2676 net.cpp:131] Top shape: 50 4 (200)
I0312 23:23:16.135622 2676 net.cpp:139] Memory required for data: 415765200
I0312 23:23:16.135637 2676 layer_factory.hpp:77] Creating layer accuracy
I0312 23:23:16.135642 2676 net.cpp:86] Creating Layer accuracy
I0312 23:23:16.135643 2676 net.cpp:408] accuracy <- fc8_fc8_0_split_0
I0312 23:23:16.135646 2676 net.cpp:408] accuracy <- label_data_1_split_0
I0312 23:23:16.135650 2676 net.cpp:382] accuracy -> accuracy
I0312 23:23:16.135655 2676 net.cpp:124] Setting up accuracy
I0312 23:23:16.135658 2676 net.cpp:131] Top shape: (1)
I0312 23:23:16.135659 2676 net.cpp:139] Memory required for data: 415765204
I0312 23:23:16.135661 2676 layer_factory.hpp:77] Creating layer loss
I0312 23:23:16.135665 2676 net.cpp:86] Creating Layer loss
I0312 23:23:16.135668 2676 net.cpp:408] loss <- fc8_fc8_0_split_1
I0312 23:23:16.135669 2676 net.cpp:408] loss <- label_data_1_split_1
I0312 23:23:16.135673 2676 net.cpp:382] loss -> loss
I0312 23:23:16.135677 2676 layer_factory.hpp:77] Creating layer loss
I0312 23:23:16.135888 2676 net.cpp:124] Setting up loss
I0312 23:23:16.135893 2676 net.cpp:131] Top shape: (1)
I0312 23:23:16.135895 2676 net.cpp:134] with loss weight 1
I0312 23:23:16.135917 2676 net.cpp:139] Memory required for data: 415765208
I0312 23:23:16.135920 2676 net.cpp:200] loss needs backward computation.
I0312 23:23:16.135922 2676 net.cpp:202] accuracy does not need backward computation.
I0312 23:23:16.135937 2676 net.cpp:200] fc8_fc8_0_split needs backward computation.
I0312 23:23:16.135939 2676 net.cpp:200] fc8 needs backward computation.
I0312 23:23:16.135941 2676 net.cpp:200] drop7 needs backward computation.
I0312 23:23:16.135967 2676 net.cpp:200] relu7 needs backward computation.
I0312 23:23:16.135984 2676 net.cpp:200] fc7 needs backward computation.
I0312 23:23:16.135987 2676 net.cpp:200] drop6 needs backward computation.
I0312 23:23:16.135988 2676 net.cpp:200] relu6 needs backward computation.
I0312 23:23:16.135990 2676 net.cpp:200] fc6 needs backward computation.
I0312 23:23:16.135993 2676 net.cpp:200] pool5 needs backward computation.
I0312 23:23:16.136008 2676 net.cpp:200] relu5 needs backward computation.
I0312 23:23:16.136008 2676 net.cpp:200] conv5 needs backward computation.
I0312 23:23:16.136010 2676 net.cpp:200] relu4 needs backward computation.
I0312 23:23:16.136013 2676 net.cpp:200] conv4 needs backward computation.
I0312 23:23:16.136014 2676 net.cpp:200] relu3 needs backward computation.
I0312 23:23:16.136016 2676 net.cpp:200] conv3 needs backward computation.
I0312 23:23:16.136018 2676 net.cpp:200] pool2 needs backward computation.
I0312 23:23:16.136021 2676 net.cpp:200] norm2 needs backward computation.
I0312 23:23:16.136023 2676 net.cpp:200] relu2 needs backward computation.
I0312 23:23:16.136024 2676 net.cpp:200] conv2 needs backward computation.
I0312 23:23:16.136026 2676 net.cpp:200] pool1 needs backward computation.
I0312 23:23:16.136041 2676 net.cpp:200] norm1 needs backward computation.
I0312 23:23:16.136044 2676 net.cpp:200] relu1 needs backward computation.
I0312 23:23:16.136045 2676 net.cpp:200] conv1 needs backward computation.
I0312 23:23:16.136047 2676 net.cpp:202] label_data_1_split does not need backward computation.
I0312 23:23:16.136065 2676 net.cpp:202] data does not need backward computation.
I0312 23:23:16.136065 2676 net.cpp:244] This network produces output accuracy
I0312 23:23:16.136067 2676 net.cpp:244] This network produces output loss
I0312 23:23:16.136093 2676 net.cpp:257] Network initialization done.
I0312 23:23:16.136191 2676 solver.cpp:56] Solver scaffolding done.
I0312 23:23:16.136680 2676 caffe.cpp:248] Starting Optimization
I0312 23:23:16.136683 2676 solver.cpp:273] Solving AlexNet
I0312 23:23:16.136685 2676 solver.cpp:274] Learning Rate Policy: step
I0312 23:23:16.138639 2676 solver.cpp:331] Iteration 0, Testing net (#0)
I0312 23:23:16.161350 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:23:16.401897 2676 blocking_queue.cpp:49] Waiting for data
I0312 23:23:22.186173 2676 solver.cpp:398] Test net output #0: accuracy = 0.545357
I0312 23:23:22.186219 2676 solver.cpp:398] Test net output #1: loss = 1.26114 (* 1 = 1.26114 loss)
I0312 23:23:22.457931 2676 solver.cpp:219] Iteration 0 (-7.07474e-05 iter/s, 6.31881s/20 iters), loss = 1.38639
I0312 23:23:22.460357 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.335938
I0312 23:23:22.460372 2676 solver.cpp:238] Train net output #1: loss = 1.38639 (* 1 = 1.38639 loss)
I0312 23:23:22.460417 2676 sgd_solver.cpp:105] Iteration 0, lr = 0.001
I0312 23:23:33.165230 2676 solver.cpp:219] Iteration 20 (1.86903 iter/s, 10.7008s/20 iters), loss = 1.13092
I0312 23:23:33.165272 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.605469
I0312 23:23:33.165280 2676 solver.cpp:238] Train net output #1: loss = 1.13092 (* 1 = 1.13092 loss)
I0312 23:23:33.165285 2676 sgd_solver.cpp:105] Iteration 20, lr = 0.001
I0312 23:23:36.300024 2676 solver.cpp:331] Iteration 28, Testing net (#0)
I0312 23:23:36.300057 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:23:38.279549 2676 solver.cpp:398] Test net output #0: accuracy = 0.545357
I0312 23:23:38.279577 2676 solver.cpp:398] Test net output #1: loss = 1.16134 (* 1 = 1.16134 loss)
I0312 23:23:41.725931 2676 solver.cpp:219] Iteration 40 (2.33717 iter/s, 8.55736s/20 iters), loss = 1.20381
I0312 23:23:41.738235 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.515625
I0312 23:23:41.738255 2676 solver.cpp:238] Train net output #1: loss = 1.20381 (* 1 = 1.20381 loss)
I0312 23:23:41.738261 2676 sgd_solver.cpp:105] Iteration 40, lr = 0.001
I0312 23:23:45.567281 2676 solver.cpp:331] Iteration 56, Testing net (#0)
I0312 23:23:45.567314 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:23:47.537325 2676 solver.cpp:398] Test net output #0: accuracy = 0.545714
I0312 23:23:47.537364 2676 solver.cpp:398] Test net output #1: loss = 1.14157 (* 1 = 1.14157 loss)
I0312 23:23:48.830137 2676 solver.cpp:219] Iteration 60 (2.7988 iter/s, 7.14593s/20 iters), loss = 1.18281
I0312 23:23:48.842432 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.523438
I0312 23:23:48.842471 2676 solver.cpp:238] Train net output #1: loss = 1.18281 (* 1 = 1.18281 loss)
I0312 23:23:48.842490 2676 sgd_solver.cpp:105] Iteration 60, lr = 0.001
I0312 23:23:54.134619 2676 solver.cpp:219] Iteration 80 (3.7488 iter/s, 5.33504s/20 iters), loss = 1.20185
I0312 23:23:54.146620 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.539062
I0312 23:23:54.146651 2676 solver.cpp:238] Train net output #1: loss = 1.20185 (* 1 = 1.20185 loss)
I0312 23:23:54.146656 2676 sgd_solver.cpp:105] Iteration 80, lr = 0.001
I0312 23:23:54.778365 2676 solver.cpp:331] Iteration 84, Testing net (#0)
I0312 23:23:54.778399 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:23:56.768332 2676 solver.cpp:398] Test net output #0: accuracy = 0.545357
I0312 23:23:56.768355 2676 solver.cpp:398] Test net output #1: loss = 1.10908 (* 1 = 1.10908 loss)
I0312 23:24:01.276773 2676 solver.cpp:219] Iteration 100 (2.79421 iter/s, 7.15766s/20 iters), loss = 1.10751
I0312 23:24:01.288919 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.550781
I0312 23:24:01.288950 2676 solver.cpp:238] Train net output #1: loss = 1.10751 (* 1 = 1.10751 loss)
I0312 23:24:01.288956 2676 sgd_solver.cpp:105] Iteration 100, lr = 0.001
I0312 23:24:04.057387 2676 solver.cpp:331] Iteration 112, Testing net (#0)
I0312 23:24:04.057422 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:24:06.030336 2676 solver.cpp:398] Test net output #0: accuracy = 0.548214
I0312 23:24:06.030366 2676 solver.cpp:398] Test net output #1: loss = 1.05868 (* 1 = 1.05868 loss)
I0312 23:24:08.410758 2676 solver.cpp:219] Iteration 120 (2.8032 iter/s, 7.1347s/20 iters), loss = 1.07986
I0312 23:24:08.422947 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.542969
I0312 23:24:08.422967 2676 solver.cpp:238] Train net output #1: loss = 1.07986 (* 1 = 1.07986 loss)
I0312 23:24:08.422972 2676 sgd_solver.cpp:105] Iteration 120, lr = 0.001
I0312 23:24:13.330821 2676 solver.cpp:331] Iteration 140, Testing net (#0)
I0312 23:24:13.330871 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:24:15.309830 2676 solver.cpp:398] Test net output #0: accuracy = 0.571786
I0312 23:24:15.309869 2676 solver.cpp:398] Test net output #1: loss = 1.00978 (* 1 = 1.00978 loss)
I0312 23:24:15.570574 2676 solver.cpp:219] Iteration 140 (2.7953 iter/s, 7.15487s/20 iters), loss = 1.1149
I0312 23:24:15.573030 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.503906
I0312 23:24:15.573060 2676 solver.cpp:238] Train net output #1: loss = 1.1149 (* 1 = 1.1149 loss)
I0312 23:24:15.573066 2676 sgd_solver.cpp:105] Iteration 140, lr = 0.001
I0312 23:24:20.904796 2676 solver.cpp:219] Iteration 160 (3.74837 iter/s, 5.33566s/20 iters), loss = 1.15146
I0312 23:24:20.916966 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.488281
I0312 23:24:20.916983 2676 solver.cpp:238] Train net output #1: loss = 1.15146 (* 1 = 1.15146 loss)
I0312 23:24:20.916988 2676 sgd_solver.cpp:105] Iteration 160, lr = 0.001
I0312 23:24:22.625000 2676 solver.cpp:331] Iteration 168, Testing net (#0)
I0312 23:24:22.625020 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:24:24.634449 2676 solver.cpp:398] Test net output #0: accuracy = 0.5775
I0312 23:24:24.634475 2676 solver.cpp:398] Test net output #1: loss = 0.970126 (* 1 = 0.970126 loss)
I0312 23:24:28.095824 2676 solver.cpp:219] Iteration 180 (2.78425 iter/s, 7.18325s/20 iters), loss = 1.03344
I0312 23:24:28.108048 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.5625
I0312 23:24:28.108081 2676 solver.cpp:238] Train net output #1: loss = 1.03344 (* 1 = 1.03344 loss)
I0312 23:24:28.108086 2676 sgd_solver.cpp:105] Iteration 180, lr = 0.001
I0312 23:24:31.963044 2676 solver.cpp:331] Iteration 196, Testing net (#0)
I0312 23:24:31.963080 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:24:33.957167 2676 solver.cpp:398] Test net output #0: accuracy = 0.576429
I0312 23:24:33.957206 2676 solver.cpp:398] Test net output #1: loss = 0.958613 (* 1 = 0.958613 loss)
I0312 23:24:35.273638 2676 solver.cpp:219] Iteration 200 (2.78957 iter/s, 7.16956s/20 iters), loss = 0.960444
I0312 23:24:35.285887 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.5625
I0312 23:24:35.285918 2676 solver.cpp:238] Train net output #1: loss = 0.960444 (* 1 = 0.960444 loss)
I0312 23:24:35.285924 2676 sgd_solver.cpp:105] Iteration 200, lr = 0.001
I0312 23:24:40.657598 2676 solver.cpp:219] Iteration 220 (3.72122 iter/s, 5.37458s/20 iters), loss = 0.904809
I0312 23:24:40.669728 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.578125
I0312 23:24:40.669744 2676 solver.cpp:238] Train net output #1: loss = 0.904809 (* 1 = 0.904809 loss)
I0312 23:24:40.669764 2676 sgd_solver.cpp:105] Iteration 220, lr = 0.001
I0312 23:24:41.304330 2676 solver.cpp:331] Iteration 224, Testing net (#0)
I0312 23:24:41.304366 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:24:43.297623 2676 solver.cpp:398] Test net output #0: accuracy = 0.595
I0312 23:24:43.297662 2676 solver.cpp:398] Test net output #1: loss = 0.935005 (* 1 = 0.935005 loss)
I0312 23:24:47.847645 2676 solver.cpp:219] Iteration 240 (2.78486 iter/s, 7.18169s/20 iters), loss = 0.964164
I0312 23:24:47.859764 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.582031
I0312 23:24:47.859781 2676 solver.cpp:238] Train net output #1: loss = 0.964164 (* 1 = 0.964164 loss)
I0312 23:24:47.859800 2676 sgd_solver.cpp:105] Iteration 240, lr = 0.001
I0312 23:24:50.647053 2676 solver.cpp:331] Iteration 252, Testing net (#0)
I0312 23:24:50.647089 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:24:52.633581 2676 solver.cpp:398] Test net output #0: accuracy = 0.599286
I0312 23:24:52.633618 2676 solver.cpp:398] Test net output #1: loss = 0.980615 (* 1 = 0.980615 loss)
I0312 23:24:55.028817 2676 solver.cpp:219] Iteration 260 (2.78831 iter/s, 7.17279s/20 iters), loss = 0.988547
I0312 23:24:55.041051 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.59375
I0312 23:24:55.041085 2676 solver.cpp:238] Train net output #1: loss = 0.988547 (* 1 = 0.988547 loss)
I0312 23:24:55.041091 2676 sgd_solver.cpp:105] Iteration 260, lr = 0.001
I0312 23:24:59.975488 2676 solver.cpp:331] Iteration 280, Testing net (#0)
I0312 23:24:59.975523 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:25:01.941673 2676 solver.cpp:398] Test net output #0: accuracy = 0.6
I0312 23:25:01.941711 2676 solver.cpp:398] Test net output #1: loss = 0.919044 (* 1 = 0.919044 loss)
I0312 23:25:02.204524 2676 solver.cpp:219] Iteration 280 (2.79049 iter/s, 7.1672s/20 iters), loss = 1.0459
I0312 23:25:02.206957 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.585938
I0312 23:25:02.206985 2676 solver.cpp:238] Train net output #1: loss = 1.0459 (* 1 = 1.0459 loss)
I0312 23:25:02.207006 2676 sgd_solver.cpp:105] Iteration 280, lr = 0.001
I0312 23:25:07.572079 2676 solver.cpp:219] Iteration 300 (3.72585 iter/s, 5.36791s/20 iters), loss = 0.892142
I0312 23:25:07.584277 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.605469
I0312 23:25:07.584297 2676 solver.cpp:238] Train net output #1: loss = 0.892142 (* 1 = 0.892142 loss)
I0312 23:25:07.584317 2676 sgd_solver.cpp:105] Iteration 300, lr = 0.001
I0312 23:25:09.297149 2676 solver.cpp:331] Iteration 308, Testing net (#0)
I0312 23:25:09.297185 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:25:11.270056 2676 solver.cpp:398] Test net output #0: accuracy = 0.617857
I0312 23:25:11.270094 2676 solver.cpp:398] Test net output #1: loss = 0.876603 (* 1 = 0.876603 loss)
I0312 23:25:14.751806 2676 solver.cpp:219] Iteration 320 (2.78891 iter/s, 7.17125s/20 iters), loss = 0.935945
I0312 23:25:14.763928 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.601562
I0312 23:25:14.763962 2676 solver.cpp:238] Train net output #1: loss = 0.935945 (* 1 = 0.935945 loss)
I0312 23:25:14.763967 2676 sgd_solver.cpp:105] Iteration 320, lr = 0.001
I0312 23:25:18.636039 2676 solver.cpp:331] Iteration 336, Testing net (#0)
I0312 23:25:18.636075 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:25:20.618952 2676 solver.cpp:398] Test net output #0: accuracy = 0.634643
I0312 23:25:20.618978 2676 solver.cpp:398] Test net output #1: loss = 0.85954 (* 1 = 0.85954 loss)
I0312 23:25:21.936133 2676 solver.cpp:219] Iteration 340 (2.7871 iter/s, 7.17593s/20 iters), loss = 0.869285
I0312 23:25:21.948242 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.636719
I0312 23:25:21.948276 2676 solver.cpp:238] Train net output #1: loss = 0.869285 (* 1 = 0.869285 loss)
I0312 23:25:21.948282 2676 sgd_solver.cpp:105] Iteration 340, lr = 0.001
I0312 23:25:27.332120 2676 solver.cpp:219] Iteration 360 (3.71286 iter/s, 5.38669s/20 iters), loss = 0.899806
I0312 23:25:27.344280 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.613281
I0312 23:25:27.344297 2676 solver.cpp:238] Train net output #1: loss = 0.899806 (* 1 = 0.899806 loss)
I0312 23:25:27.344303 2676 sgd_solver.cpp:105] Iteration 360, lr = 0.001
I0312 23:25:27.983243 2676 solver.cpp:331] Iteration 364, Testing net (#0)
I0312 23:25:27.983280 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:25:30.007056 2676 solver.cpp:398] Test net output #0: accuracy = 0.645357
I0312 23:25:30.007093 2676 solver.cpp:398] Test net output #1: loss = 0.844139 (* 1 = 0.844139 loss)
I0312 23:25:32.783954 2676 blocking_queue.cpp:49] Waiting for data
I0312 23:25:34.568060 2676 solver.cpp:219] Iteration 380 (2.7672 iter/s, 7.22753s/20 iters), loss = 0.840824
I0312 23:25:34.580109 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.664062
I0312 23:25:34.580127 2676 solver.cpp:238] Train net output #1: loss = 0.840824 (* 1 = 0.840824 loss)
I0312 23:25:34.580132 2676 sgd_solver.cpp:105] Iteration 380, lr = 0.001
I0312 23:25:37.373685 2676 solver.cpp:331] Iteration 392, Testing net (#0)
I0312 23:25:37.373728 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:25:39.393951 2676 solver.cpp:398] Test net output #0: accuracy = 0.652143
I0312 23:25:39.393977 2676 solver.cpp:398] Test net output #1: loss = 0.85232 (* 1 = 0.85232 loss)
I0312 23:25:41.798810 2676 solver.cpp:219] Iteration 400 (2.76914 iter/s, 7.22245s/20 iters), loss = 0.829997
I0312 23:25:41.811102 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.65625
I0312 23:25:41.811122 2676 solver.cpp:238] Train net output #1: loss = 0.829997 (* 1 = 0.829997 loss)
I0312 23:25:41.811127 2676 sgd_solver.cpp:105] Iteration 400, lr = 0.001
I0312 23:25:46.767246 2676 solver.cpp:331] Iteration 420, Testing net (#0)
I0312 23:25:46.767431 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:25:48.779335 2676 solver.cpp:398] Test net output #0: accuracy = 0.664286
I0312 23:25:48.779371 2676 solver.cpp:398] Test net output #1: loss = 0.804686 (* 1 = 0.804686 loss)
I0312 23:25:49.040144 2676 solver.cpp:219] Iteration 420 (2.76518 iter/s, 7.2328s/20 iters), loss = 0.937281
I0312 23:25:49.042621 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.597656
I0312 23:25:49.042652 2676 solver.cpp:238] Train net output #1: loss = 0.937281 (* 1 = 0.937281 loss)
I0312 23:25:49.042657 2676 sgd_solver.cpp:105] Iteration 420, lr = 0.001
I0312 23:25:54.423571 2676 solver.cpp:219] Iteration 440 (3.71489 iter/s, 5.38374s/20 iters), loss = 0.970864
I0312 23:25:54.435729 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.597656
I0312 23:25:54.435745 2676 solver.cpp:238] Train net output #1: loss = 0.970864 (* 1 = 0.970864 loss)
I0312 23:25:54.435765 2676 sgd_solver.cpp:105] Iteration 440, lr = 0.001
I0312 23:25:56.153745 2676 solver.cpp:331] Iteration 448, Testing net (#0)
I0312 23:25:56.153781 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:25:58.141585 2676 solver.cpp:398] Test net output #0: accuracy = 0.673214
I0312 23:25:58.141623 2676 solver.cpp:398] Test net output #1: loss = 0.799573 (* 1 = 0.799573 loss)
I0312 23:26:01.631712 2676 solver.cpp:219] Iteration 460 (2.77788 iter/s, 7.19972s/20 iters), loss = 0.886954
I0312 23:26:01.643872 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.613281
I0312 23:26:01.643903 2676 solver.cpp:238] Train net output #1: loss = 0.886954 (* 1 = 0.886954 loss)
I0312 23:26:01.643908 2676 sgd_solver.cpp:105] Iteration 460, lr = 0.001
I0312 23:26:05.524873 2676 solver.cpp:331] Iteration 476, Testing net (#0)
I0312 23:26:05.524893 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:26:07.525322 2676 solver.cpp:398] Test net output #0: accuracy = 0.674643
I0312 23:26:07.525360 2676 solver.cpp:398] Test net output #1: loss = 0.781635 (* 1 = 0.781635 loss)
I0312 23:26:08.875036 2676 solver.cpp:219] Iteration 480 (2.773 iter/s, 7.21241s/20 iters), loss = 0.790123
I0312 23:26:08.887622 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.644531
I0312 23:26:08.887660 2676 solver.cpp:238] Train net output #1: loss = 0.790123 (* 1 = 0.790123 loss)
I0312 23:26:08.887665 2676 sgd_solver.cpp:105] Iteration 480, lr = 0.001
I0312 23:26:14.376476 2676 solver.cpp:219] Iteration 500 (3.70628 iter/s, 5.39625s/20 iters), loss = 0.727805
I0312 23:26:14.388772 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.660156
I0312 23:26:14.388804 2676 solver.cpp:238] Train net output #1: loss = 0.727805 (* 1 = 0.727805 loss)
I0312 23:26:14.388809 2676 sgd_solver.cpp:105] Iteration 500, lr = 0.001
I0312 23:26:15.037009 2676 solver.cpp:331] Iteration 504, Testing net (#0)
I0312 23:26:15.037044 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:26:17.052551 2676 solver.cpp:398] Test net output #0: accuracy = 0.678572
I0312 23:26:17.052588 2676 solver.cpp:398] Test net output #1: loss = 0.757533 (* 1 = 0.757533 loss)
I0312 23:26:21.647141 2676 solver.cpp:219] Iteration 520 (2.77673 iter/s, 7.20271s/20 iters), loss = 0.833957
I0312 23:26:21.659446 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.648438
I0312 23:26:21.659463 2676 solver.cpp:238] Train net output #1: loss = 0.833957 (* 1 = 0.833957 loss)
I0312 23:26:21.659482 2676 sgd_solver.cpp:105] Iteration 520, lr = 0.001
I0312 23:26:24.474599 2676 solver.cpp:331] Iteration 532, Testing net (#0)
I0312 23:26:24.474622 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:26:26.479739 2676 solver.cpp:398] Test net output #0: accuracy = 0.706072
I0312 23:26:26.479763 2676 solver.cpp:398] Test net output #1: loss = 0.735266 (* 1 = 0.735266 loss)
I0312 23:26:28.900995 2676 solver.cpp:219] Iteration 540 (2.77077 iter/s, 7.21822s/20 iters), loss = 0.796514
I0312 23:26:28.913269 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.660156
I0312 23:26:28.913302 2676 solver.cpp:238] Train net output #1: loss = 0.796514 (* 1 = 0.796514 loss)
I0312 23:26:28.913308 2676 sgd_solver.cpp:105] Iteration 540, lr = 0.001
I0312 23:26:33.882889 2676 solver.cpp:331] Iteration 560, Testing net (#0)
I0312 23:26:33.882967 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:26:35.880746 2676 solver.cpp:398] Test net output #0: accuracy = 0.701429
I0312 23:26:35.880770 2676 solver.cpp:398] Test net output #1: loss = 0.724797 (* 1 = 0.724797 loss)
I0312 23:26:36.143487 2676 solver.cpp:219] Iteration 560 (2.77038 iter/s, 7.21922s/20 iters), loss = 0.798369
I0312 23:26:36.145952 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.6875
I0312 23:26:36.145998 2676 solver.cpp:238] Train net output #1: loss = 0.798369 (* 1 = 0.798369 loss)
I0312 23:26:36.146019 2676 sgd_solver.cpp:105] Iteration 560, lr = 0.001
I0312 23:26:41.539515 2676 solver.cpp:219] Iteration 580 (3.71042 iter/s, 5.39022s/20 iters), loss = 0.653225
I0312 23:26:41.551695 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.742188
I0312 23:26:41.551728 2676 solver.cpp:238] Train net output #1: loss = 0.653225 (* 1 = 0.653225 loss)
I0312 23:26:41.551734 2676 sgd_solver.cpp:105] Iteration 580, lr = 0.001
I0312 23:26:43.272722 2676 solver.cpp:331] Iteration 588, Testing net (#0)
I0312 23:26:43.272761 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:26:45.273249 2676 solver.cpp:398] Test net output #0: accuracy = 0.720357
I0312 23:26:45.273273 2676 solver.cpp:398] Test net output #1: loss = 0.69917 (* 1 = 0.69917 loss)
I0312 23:26:48.760844 2676 solver.cpp:219] Iteration 600 (2.77404 iter/s, 7.20969s/20 iters), loss = 0.766637
I0312 23:26:48.773182 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.671875
I0312 23:26:48.773214 2676 solver.cpp:238] Train net output #1: loss = 0.766637 (* 1 = 0.766637 loss)
I0312 23:26:48.773221 2676 sgd_solver.cpp:105] Iteration 600, lr = 0.001
I0312 23:26:52.661382 2676 solver.cpp:331] Iteration 616, Testing net (#0)
I0312 23:26:52.661420 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:26:54.657719 2676 solver.cpp:398] Test net output #0: accuracy = 0.683929
I0312 23:26:54.657742 2676 solver.cpp:398] Test net output #1: loss = 0.751855 (* 1 = 0.751855 loss)
I0312 23:26:55.990639 2676 solver.cpp:219] Iteration 620 (2.77134 iter/s, 7.21672s/20 iters), loss = 0.716449
I0312 23:26:56.002892 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.683594
I0312 23:26:56.002925 2676 solver.cpp:238] Train net output #1: loss = 0.716449 (* 1 = 0.716449 loss)
I0312 23:26:56.002931 2676 sgd_solver.cpp:105] Iteration 620, lr = 0.001
I0312 23:27:01.403511 2676 solver.cpp:219] Iteration 640 (3.70404 iter/s, 5.39951s/20 iters), loss = 0.686625
I0312 23:27:01.415530 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.742188
I0312 23:27:01.415561 2676 solver.cpp:238] Train net output #1: loss = 0.686625 (* 1 = 0.686625 loss)
I0312 23:27:01.415566 2676 sgd_solver.cpp:105] Iteration 640, lr = 0.001
I0312 23:27:02.059046 2676 solver.cpp:331] Iteration 644, Testing net (#0)
I0312 23:27:02.059067 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:27:04.075769 2676 solver.cpp:398] Test net output #0: accuracy = 0.728214
I0312 23:27:04.075808 2676 solver.cpp:398] Test net output #1: loss = 0.672072 (* 1 = 0.672072 loss)
I0312 23:27:08.655361 2676 solver.cpp:219] Iteration 660 (2.76324 iter/s, 7.23788s/20 iters), loss = 0.61903
I0312 23:27:08.667495 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.753906
I0312 23:27:08.667526 2676 solver.cpp:238] Train net output #1: loss = 0.61903 (* 1 = 0.61903 loss)
I0312 23:27:08.667531 2676 sgd_solver.cpp:105] Iteration 660, lr = 0.001
I0312 23:27:11.481791 2676 solver.cpp:331] Iteration 672, Testing net (#0)
I0312 23:27:11.481827 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:27:13.467962 2676 solver.cpp:398] Test net output #0: accuracy = 0.738214
I0312 23:27:13.467988 2676 solver.cpp:398] Test net output #1: loss = 0.633847 (* 1 = 0.633847 loss)
I0312 23:27:15.883911 2676 solver.cpp:219] Iteration 680 (2.77234 iter/s, 7.21412s/20 iters), loss = 0.621791
I0312 23:27:15.896040 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.742188
I0312 23:27:15.896075 2676 solver.cpp:238] Train net output #1: loss = 0.621791 (* 1 = 0.621791 loss)
I0312 23:27:15.896080 2676 sgd_solver.cpp:105] Iteration 680, lr = 0.001
I0312 23:27:20.870307 2676 solver.cpp:331] Iteration 700, Testing net (#0)
I0312 23:27:20.870440 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:27:22.878713 2676 solver.cpp:398] Test net output #0: accuracy = 0.744286
I0312 23:27:22.878738 2676 solver.cpp:398] Test net output #1: loss = 0.618683 (* 1 = 0.618683 loss)
I0312 23:27:23.141259 2676 solver.cpp:219] Iteration 700 (2.7614 iter/s, 7.24269s/20 iters), loss = 0.676554
I0312 23:27:23.143703 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.707031
I0312 23:27:23.143733 2676 solver.cpp:238] Train net output #1: loss = 0.676554 (* 1 = 0.676554 loss)
I0312 23:27:23.143739 2676 sgd_solver.cpp:105] Iteration 700, lr = 0.001
I0312 23:27:28.554875 2676 solver.cpp:219] Iteration 720 (3.69742 iter/s, 5.40918s/20 iters), loss = 0.570761
I0312 23:27:28.566990 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.757812
I0312 23:27:28.567008 2676 solver.cpp:238] Train net output #1: loss = 0.570761 (* 1 = 0.570761 loss)
I0312 23:27:28.567013 2676 sgd_solver.cpp:105] Iteration 720, lr = 0.001
I0312 23:27:30.292814 2676 solver.cpp:331] Iteration 728, Testing net (#0)
I0312 23:27:30.292850 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:27:32.317675 2676 solver.cpp:398] Test net output #0: accuracy = 0.766429
I0312 23:27:32.317713 2676 solver.cpp:398] Test net output #1: loss = 0.59132 (* 1 = 0.59132 loss)
I0312 23:27:35.824728 2676 solver.cpp:219] Iteration 740 (2.75672 iter/s, 7.25499s/20 iters), loss = 0.562286
I0312 23:27:35.836900 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.796875
I0312 23:27:35.837059 2676 solver.cpp:238] Train net output #1: loss = 0.562286 (* 1 = 0.562286 loss)
I0312 23:27:35.837067 2676 sgd_solver.cpp:105] Iteration 740, lr = 0.001
I0312 23:27:39.736904 2676 solver.cpp:331] Iteration 756, Testing net (#0)
I0312 23:27:39.736940 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:27:41.320724 2676 blocking_queue.cpp:49] Waiting for data
I0312 23:27:41.724932 2676 solver.cpp:398] Test net output #0: accuracy = 0.778929
I0312 23:27:41.724970 2676 solver.cpp:398] Test net output #1: loss = 0.561568 (* 1 = 0.561568 loss)
I0312 23:27:43.058217 2676 solver.cpp:219] Iteration 760 (2.77065 iter/s, 7.21853s/20 iters), loss = 0.60303
I0312 23:27:43.070521 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.722656
I0312 23:27:43.070555 2676 solver.cpp:238] Train net output #1: loss = 0.60303 (* 1 = 0.60303 loss)
I0312 23:27:43.070560 2676 sgd_solver.cpp:105] Iteration 760, lr = 0.001
I0312 23:27:48.485563 2676 solver.cpp:219] Iteration 780 (3.69265 iter/s, 5.41616s/20 iters), loss = 0.535647
I0312 23:27:48.497586 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.78125
I0312 23:27:48.497617 2676 solver.cpp:238] Train net output #1: loss = 0.535647 (* 1 = 0.535647 loss)
I0312 23:27:48.497622 2676 sgd_solver.cpp:105] Iteration 780, lr = 0.001
I0312 23:27:49.140836 2676 solver.cpp:331] Iteration 784, Testing net (#0)
I0312 23:27:49.140872 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:27:51.124541 2676 solver.cpp:398] Test net output #0: accuracy = 0.7775
I0312 23:27:51.124577 2676 solver.cpp:398] Test net output #1: loss = 0.565756 (* 1 = 0.565756 loss)
I0312 23:27:55.714097 2676 solver.cpp:219] Iteration 800 (2.77081 iter/s, 7.21811s/20 iters), loss = 0.538049
I0312 23:27:55.726258 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.785156
I0312 23:27:55.726303 2676 solver.cpp:238] Train net output #1: loss = 0.538049 (* 1 = 0.538049 loss)
I0312 23:27:55.726310 2676 sgd_solver.cpp:105] Iteration 800, lr = 0.001
I0312 23:27:58.538839 2676 solver.cpp:331] Iteration 812, Testing net (#0)
I0312 23:27:58.538875 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:28:00.541971 2676 solver.cpp:398] Test net output #0: accuracy = 0.798929
I0312 23:28:00.542008 2676 solver.cpp:398] Test net output #1: loss = 0.513049 (* 1 = 0.513049 loss)
I0312 23:28:02.957697 2676 solver.cpp:219] Iteration 820 (2.76539 iter/s, 7.23224s/20 iters), loss = 0.516776
I0312 23:28:02.969964 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.789062
I0312 23:28:02.970010 2676 solver.cpp:238] Train net output #1: loss = 0.516776 (* 1 = 0.516776 loss)
I0312 23:28:02.970016 2676 sgd_solver.cpp:105] Iteration 820, lr = 0.001
I0312 23:28:07.946287 2676 solver.cpp:331] Iteration 840, Testing net (#0)
I0312 23:28:07.946333 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:28:09.939030 2676 solver.cpp:398] Test net output #0: accuracy = 0.794643
I0312 23:28:09.939067 2676 solver.cpp:398] Test net output #1: loss = 0.525727 (* 1 = 0.525727 loss)
I0312 23:28:10.201859 2676 solver.cpp:219] Iteration 840 (2.76546 iter/s, 7.23208s/20 iters), loss = 0.550795
I0312 23:28:10.204303 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.769531
I0312 23:28:10.204332 2676 solver.cpp:238] Train net output #1: loss = 0.550795 (* 1 = 0.550795 loss)
I0312 23:28:10.204339 2676 sgd_solver.cpp:105] Iteration 840, lr = 0.001
I0312 23:28:15.605336 2676 solver.cpp:219] Iteration 860 (3.70313 iter/s, 5.40083s/20 iters), loss = 0.433974
I0312 23:28:15.617573 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.828125
I0312 23:28:15.617605 2676 solver.cpp:238] Train net output #1: loss = 0.433974 (* 1 = 0.433974 loss)
I0312 23:28:15.617610 2676 sgd_solver.cpp:105] Iteration 860, lr = 0.001
I0312 23:28:17.345782 2676 solver.cpp:331] Iteration 868, Testing net (#0)
I0312 23:28:17.345818 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:28:19.325099 2676 solver.cpp:398] Test net output #0: accuracy = 0.803571
I0312 23:28:19.325136 2676 solver.cpp:398] Test net output #1: loss = 0.47534 (* 1 = 0.47534 loss)
I0312 23:28:22.825062 2676 solver.cpp:219] Iteration 880 (2.77514 iter/s, 7.20685s/20 iters), loss = 0.478455
I0312 23:28:22.837374 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.808594
I0312 23:28:22.837409 2676 solver.cpp:238] Train net output #1: loss = 0.478455 (* 1 = 0.478455 loss)
I0312 23:28:22.837415 2676 sgd_solver.cpp:105] Iteration 880, lr = 0.001
I0312 23:28:26.740339 2676 solver.cpp:331] Iteration 896, Testing net (#0)
I0312 23:28:26.740376 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:28:28.738651 2676 solver.cpp:398] Test net output #0: accuracy = 0.776071
I0312 23:28:28.738687 2676 solver.cpp:398] Test net output #1: loss = 0.570051 (* 1 = 0.570051 loss)
I0312 23:28:30.070143 2676 solver.cpp:219] Iteration 900 (2.76557 iter/s, 7.23179s/20 iters), loss = 0.382397
I0312 23:28:30.082314 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.863281
I0312 23:28:30.082353 2676 solver.cpp:238] Train net output #1: loss = 0.382397 (* 1 = 0.382397 loss)
I0312 23:28:30.082358 2676 sgd_solver.cpp:105] Iteration 900, lr = 0.001
I0312 23:28:35.504490 2676 solver.cpp:219] Iteration 920 (3.68917 iter/s, 5.42127s/20 iters), loss = 0.394418
I0312 23:28:35.516649 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.847656
I0312 23:28:35.516667 2676 solver.cpp:238] Train net output #1: loss = 0.394418 (* 1 = 0.394418 loss)
I0312 23:28:35.516688 2676 sgd_solver.cpp:105] Iteration 920, lr = 0.001
I0312 23:28:36.158797 2676 solver.cpp:331] Iteration 924, Testing net (#0)
I0312 23:28:36.158833 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:28:38.178876 2676 solver.cpp:398] Test net output #0: accuracy = 0.803214
I0312 23:28:38.178915 2676 solver.cpp:398] Test net output #1: loss = 0.493671 (* 1 = 0.493671 loss)
I0312 23:28:42.767585 2676 solver.cpp:219] Iteration 940 (2.7588 iter/s, 7.24952s/20 iters), loss = 0.368876
I0312 23:28:42.779748 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.875
I0312 23:28:42.779781 2676 solver.cpp:238] Train net output #1: loss = 0.368876 (* 1 = 0.368876 loss)
I0312 23:28:42.779786 2676 sgd_solver.cpp:105] Iteration 940, lr = 0.001
I0312 23:28:45.597579 2676 solver.cpp:331] Iteration 952, Testing net (#0)
I0312 23:28:45.597618 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:28:47.598242 2676 solver.cpp:398] Test net output #0: accuracy = 0.817143
I0312 23:28:47.598280 2676 solver.cpp:398] Test net output #1: loss = 0.505942 (* 1 = 0.505942 loss)
I0312 23:28:50.013950 2676 solver.cpp:219] Iteration 960 (2.76525 iter/s, 7.23261s/20 iters), loss = 0.39874
I0312 23:28:50.026216 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.859375
I0312 23:28:50.026249 2676 solver.cpp:238] Train net output #1: loss = 0.39874 (* 1 = 0.39874 loss)
I0312 23:28:50.026254 2676 sgd_solver.cpp:105] Iteration 960, lr = 0.001
I0312 23:28:55.011466 2676 solver.cpp:331] Iteration 980, Testing net (#0)
I0312 23:28:55.011608 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:28:56.987771 2676 solver.cpp:398] Test net output #0: accuracy = 0.858571
I0312 23:28:56.987810 2676 solver.cpp:398] Test net output #1: loss = 0.39668 (* 1 = 0.39668 loss)
I0312 23:28:57.250288 2676 solver.cpp:219] Iteration 980 (2.76918 iter/s, 7.22234s/20 iters), loss = 0.406421
I0312 23:28:57.252743 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.8125
I0312 23:28:57.252770 2676 solver.cpp:238] Train net output #1: loss = 0.406421 (* 1 = 0.406421 loss)
I0312 23:28:57.252789 2676 sgd_solver.cpp:105] Iteration 980, lr = 0.001
I0312 23:29:02.665797 2676 solver.cpp:219] Iteration 1000 (3.6957 iter/s, 5.4117s/20 iters), loss = 0.294406
I0312 23:29:02.677963 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.894531
I0312 23:29:02.677981 2676 solver.cpp:238] Train net output #1: loss = 0.294406 (* 1 = 0.294406 loss)
I0312 23:29:02.677986 2676 sgd_solver.cpp:105] Iteration 1000, lr = 0.001
I0312 23:29:04.405819 2676 solver.cpp:331] Iteration 1008, Testing net (#0)
I0312 23:29:04.405855 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:29:06.397883 2676 solver.cpp:398] Test net output #0: accuracy = 0.843929
I0312 23:29:06.397920 2676 solver.cpp:398] Test net output #1: loss = 0.438704 (* 1 = 0.438704 loss)
I0312 23:29:09.909572 2676 solver.cpp:219] Iteration 1020 (2.76637 iter/s, 7.2297s/20 iters), loss = 0.363993
I0312 23:29:09.921815 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.871094
I0312 23:29:09.921834 2676 solver.cpp:238] Train net output #1: loss = 0.363993 (* 1 = 0.363993 loss)
I0312 23:29:09.921840 2676 sgd_solver.cpp:105] Iteration 1020, lr = 0.001
I0312 23:29:13.822978 2676 solver.cpp:331] Iteration 1036, Testing net (#0)
I0312 23:29:13.823012 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:29:15.791075 2676 solver.cpp:398] Test net output #0: accuracy = 0.863571
I0312 23:29:15.791116 2676 solver.cpp:398] Test net output #1: loss = 0.378179 (* 1 = 0.378179 loss)
I0312 23:29:17.124424 2676 solver.cpp:219] Iteration 1040 (2.77753 iter/s, 7.20064s/20 iters), loss = 0.31136
I0312 23:29:17.136637 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.894531
I0312 23:29:17.136672 2676 solver.cpp:238] Train net output #1: loss = 0.31136 (* 1 = 0.31136 loss)
I0312 23:29:17.136677 2676 sgd_solver.cpp:105] Iteration 1040, lr = 0.001
I0312 23:29:22.559360 2676 solver.cpp:219] Iteration 1060 (3.68923 iter/s, 5.42119s/20 iters), loss = 0.317804
I0312 23:29:22.571579 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.867188
I0312 23:29:22.571599 2676 solver.cpp:238] Train net output #1: loss = 0.317804 (* 1 = 0.317804 loss)
I0312 23:29:22.571604 2676 sgd_solver.cpp:105] Iteration 1060, lr = 0.001
I0312 23:29:23.216394 2676 solver.cpp:331] Iteration 1064, Testing net (#0)
I0312 23:29:23.216416 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:29:25.206918 2676 solver.cpp:398] Test net output #0: accuracy = 0.872143
I0312 23:29:25.206955 2676 solver.cpp:398] Test net output #1: loss = 0.345706 (* 1 = 0.345706 loss)
I0312 23:29:29.798655 2676 solver.cpp:219] Iteration 1080 (2.76816 iter/s, 7.225s/20 iters), loss = 0.234061
I0312 23:29:29.810819 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.9375
I0312 23:29:29.810850 2676 solver.cpp:238] Train net output #1: loss = 0.234061 (* 1 = 0.234061 loss)
I0312 23:29:29.810855 2676 sgd_solver.cpp:105] Iteration 1080, lr = 0.001
I0312 23:29:32.624083 2676 solver.cpp:331] Iteration 1092, Testing net (#0)
I0312 23:29:32.624104 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:29:34.633342 2676 solver.cpp:398] Test net output #0: accuracy = 0.86
I0312 23:29:34.633378 2676 solver.cpp:398] Test net output #1: loss = 0.361714 (* 1 = 0.361714 loss)
I0312 23:29:37.051342 2676 solver.cpp:219] Iteration 1100 (2.76304 iter/s, 7.23839s/20 iters), loss = 0.233078
I0312 23:29:37.063549 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.910156
I0312 23:29:37.063581 2676 solver.cpp:238] Train net output #1: loss = 0.233078 (* 1 = 0.233078 loss)
I0312 23:29:37.063587 2676 sgd_solver.cpp:105] Iteration 1100, lr = 0.001
I0312 23:29:42.050027 2676 solver.cpp:331] Iteration 1120, Testing net (#0)
I0312 23:29:42.050062 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:29:44.046098 2676 solver.cpp:398] Test net output #0: accuracy = 0.897143
I0312 23:29:44.046136 2676 solver.cpp:398] Test net output #1: loss = 0.308057 (* 1 = 0.308057 loss)
I0312 23:29:44.310664 2676 solver.cpp:219] Iteration 1120 (2.76054 iter/s, 7.24495s/20 iters), loss = 0.241999
I0312 23:29:44.313117 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.910156
I0312 23:29:44.313146 2676 solver.cpp:238] Train net output #1: loss = 0.241999 (* 1 = 0.241999 loss)
I0312 23:29:44.313153 2676 sgd_solver.cpp:105] Iteration 1120, lr = 0.001
I0312 23:29:49.729884 2676 solver.cpp:219] Iteration 1140 (3.69336 iter/s, 5.41513s/20 iters), loss = 0.248057
I0312 23:29:49.742183 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.898438
I0312 23:29:49.742218 2676 solver.cpp:238] Train net output #1: loss = 0.248057 (* 1 = 0.248057 loss)
I0312 23:29:49.742223 2676 sgd_solver.cpp:105] Iteration 1140, lr = 0.001
I0312 23:29:51.471709 2676 solver.cpp:331] Iteration 1148, Testing net (#0)
I0312 23:29:51.471745 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:29:52.632112 2676 blocking_queue.cpp:49] Waiting for data
I0312 23:29:53.471439 2676 solver.cpp:398] Test net output #0: accuracy = 0.857857
I0312 23:29:53.471462 2676 solver.cpp:398] Test net output #1: loss = 0.41419 (* 1 = 0.41419 loss)
I0312 23:29:56.975296 2676 solver.cpp:219] Iteration 1160 (2.76473 iter/s, 7.23398s/20 iters), loss = 0.205069
I0312 23:29:56.987603 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.929688
I0312 23:29:56.987637 2676 solver.cpp:238] Train net output #1: loss = 0.205069 (* 1 = 0.205069 loss)
I0312 23:29:56.987643 2676 sgd_solver.cpp:105] Iteration 1160, lr = 0.001
I0312 23:30:00.894788 2676 solver.cpp:331] Iteration 1176, Testing net (#0)
I0312 23:30:00.894824 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:30:02.892915 2676 solver.cpp:398] Test net output #0: accuracy = 0.896071
I0312 23:30:02.892940 2676 solver.cpp:398] Test net output #1: loss = 0.304438 (* 1 = 0.304438 loss)
I0312 23:30:04.234632 2676 solver.cpp:219] Iteration 1180 (2.75867 iter/s, 7.24988s/20 iters), loss = 0.237971
I0312 23:30:04.246827 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.910156
I0312 23:30:04.246846 2676 solver.cpp:238] Train net output #1: loss = 0.237971 (* 1 = 0.237971 loss)
I0312 23:30:04.246865 2676 sgd_solver.cpp:105] Iteration 1180, lr = 0.001
I0312 23:30:09.661489 2676 solver.cpp:219] Iteration 1200 (3.69245 iter/s, 5.41646s/20 iters), loss = 0.230362
I0312 23:30:09.673662 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.90625
I0312 23:30:09.673696 2676 solver.cpp:238] Train net output #1: loss = 0.230362 (* 1 = 0.230362 loss)
I0312 23:30:09.673702 2676 sgd_solver.cpp:105] Iteration 1200, lr = 0.001
I0312 23:30:10.316509 2676 solver.cpp:331] Iteration 1204, Testing net (#0)
I0312 23:30:10.316542 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:30:12.316997 2676 solver.cpp:398] Test net output #0: accuracy = 0.888214
I0312 23:30:12.317036 2676 solver.cpp:398] Test net output #1: loss = 0.310524 (* 1 = 0.310524 loss)
I0312 23:30:16.907405 2676 solver.cpp:219] Iteration 1220 (2.76405 iter/s, 7.23577s/20 iters), loss = 0.201871
I0312 23:30:16.919622 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.917969
I0312 23:30:16.919641 2676 solver.cpp:238] Train net output #1: loss = 0.201871 (* 1 = 0.201871 loss)
I0312 23:30:16.919661 2676 sgd_solver.cpp:105] Iteration 1220, lr = 0.001
I0312 23:30:19.735288 2676 solver.cpp:331] Iteration 1232, Testing net (#0)
I0312 23:30:19.735324 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:30:21.744154 2676 solver.cpp:398] Test net output #0: accuracy = 0.878572
I0312 23:30:21.744191 2676 solver.cpp:398] Test net output #1: loss = 0.376121 (* 1 = 0.376121 loss)
I0312 23:30:24.159935 2676 solver.cpp:219] Iteration 1240 (2.76169 iter/s, 7.24195s/20 iters), loss = 0.265296
I0312 23:30:24.172222 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.921875
I0312 23:30:24.172255 2676 solver.cpp:238] Train net output #1: loss = 0.265296 (* 1 = 0.265296 loss)
I0312 23:30:24.172260 2676 sgd_solver.cpp:105] Iteration 1240, lr = 0.001
I0312 23:30:29.161650 2676 solver.cpp:331] Iteration 1260, Testing net (#0)
I0312 23:30:29.161741 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:30:31.143518 2676 solver.cpp:398] Test net output #0: accuracy = 0.91
I0312 23:30:31.143555 2676 solver.cpp:398] Test net output #1: loss = 0.277132 (* 1 = 0.277132 loss)
I0312 23:30:31.408638 2676 solver.cpp:219] Iteration 1260 (2.76331 iter/s, 7.23769s/20 iters), loss = 0.180875
I0312 23:30:31.411087 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.933594
I0312 23:30:31.411118 2676 solver.cpp:238] Train net output #1: loss = 0.180875 (* 1 = 0.180875 loss)
I0312 23:30:31.411123 2676 sgd_solver.cpp:105] Iteration 1260, lr = 0.001
I0312 23:30:36.829504 2676 solver.cpp:219] Iteration 1280 (3.69061 iter/s, 5.41916s/20 iters), loss = 0.146877
I0312 23:30:36.841655 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.964844
I0312 23:30:36.841683 2676 solver.cpp:238] Train net output #1: loss = 0.146877 (* 1 = 0.146877 loss)
I0312 23:30:36.841689 2676 sgd_solver.cpp:105] Iteration 1280, lr = 0.001
I0312 23:30:38.570597 2676 solver.cpp:331] Iteration 1288, Testing net (#0)
I0312 23:30:38.570631 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:30:40.577872 2676 solver.cpp:398] Test net output #0: accuracy = 0.909643
I0312 23:30:40.577893 2676 solver.cpp:398] Test net output #1: loss = 0.277754 (* 1 = 0.277754 loss)
I0312 23:30:44.083091 2676 solver.cpp:219] Iteration 1300 (2.7616 iter/s, 7.24218s/20 iters), loss = 0.256302
I0312 23:30:44.095170 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.917969
I0312 23:30:44.095202 2676 solver.cpp:238] Train net output #1: loss = 0.256302 (* 1 = 0.256302 loss)
I0312 23:30:44.095208 2676 sgd_solver.cpp:105] Iteration 1300, lr = 0.001
I0312 23:30:47.997551 2676 solver.cpp:331] Iteration 1316, Testing net (#0)
I0312 23:30:47.997588 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:30:50.009754 2676 solver.cpp:398] Test net output #0: accuracy = 0.8925
I0312 23:30:50.009794 2676 solver.cpp:398] Test net output #1: loss = 0.319198 (* 1 = 0.319198 loss)
I0312 23:30:51.349707 2676 solver.cpp:219] Iteration 1320 (2.75671 iter/s, 7.25503s/20 iters), loss = 0.193846
I0312 23:30:51.361913 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.921875
I0312 23:30:51.361948 2676 solver.cpp:238] Train net output #1: loss = 0.193846 (* 1 = 0.193846 loss)
I0312 23:30:51.361953 2676 sgd_solver.cpp:105] Iteration 1320, lr = 0.001
I0312 23:30:56.793694 2676 solver.cpp:219] Iteration 1340 (3.68189 iter/s, 5.43199s/20 iters), loss = 0.181212
I0312 23:30:56.805820 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.949219
I0312 23:30:56.805851 2676 solver.cpp:238] Train net output #1: loss = 0.181212 (* 1 = 0.181212 loss)
I0312 23:30:56.805855 2676 sgd_solver.cpp:105] Iteration 1340, lr = 0.001
I0312 23:30:57.449738 2676 solver.cpp:331] Iteration 1344, Testing net (#0)
I0312 23:30:57.449772 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:30:59.448271 2676 solver.cpp:398] Test net output #0: accuracy = 0.911428
I0312 23:30:59.448293 2676 solver.cpp:398] Test net output #1: loss = 0.27244 (* 1 = 0.27244 loss)
I0312 23:31:04.041152 2676 solver.cpp:219] Iteration 1360 (2.76417 iter/s, 7.23543s/20 iters), loss = 0.101827
I0312 23:31:04.053272 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.972656
I0312 23:31:04.053288 2676 solver.cpp:238] Train net output #1: loss = 0.101827 (* 1 = 0.101827 loss)
I0312 23:31:04.053308 2676 sgd_solver.cpp:105] Iteration 1360, lr = 0.001
I0312 23:31:06.868304 2676 solver.cpp:331] Iteration 1372, Testing net (#0)
I0312 23:31:06.868325 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:31:08.883689 2676 solver.cpp:398] Test net output #0: accuracy = 0.910714
I0312 23:31:08.883728 2676 solver.cpp:398] Test net output #1: loss = 0.287065 (* 1 = 0.287065 loss)
I0312 23:31:11.306362 2676 solver.cpp:219] Iteration 1380 (2.75748 iter/s, 7.25301s/20 iters), loss = 0.0753419
I0312 23:31:11.318665 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.984375
I0312 23:31:11.318699 2676 solver.cpp:238] Train net output #1: loss = 0.0753419 (* 1 = 0.0753419 loss)
I0312 23:31:11.318704 2676 sgd_solver.cpp:105] Iteration 1380, lr = 0.001
I0312 23:31:16.307917 2676 solver.cpp:331] Iteration 1400, Testing net (#0)
I0312 23:31:16.307952 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:31:18.326223 2676 solver.cpp:398] Test net output #0: accuracy = 0.925357
I0312 23:31:18.326246 2676 solver.cpp:398] Test net output #1: loss = 0.239625 (* 1 = 0.239625 loss)
I0312 23:31:18.592233 2676 solver.cpp:219] Iteration 1400 (2.74978 iter/s, 7.27331s/20 iters), loss = 0.0991599
I0312 23:31:18.594676 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.972656
I0312 23:31:18.594702 2676 solver.cpp:238] Train net output #1: loss = 0.0991599 (* 1 = 0.0991599 loss)
I0312 23:31:18.594722 2676 sgd_solver.cpp:105] Iteration 1400, lr = 0.001
I0312 23:31:24.013437 2676 solver.cpp:219] Iteration 1420 (3.69108 iter/s, 5.41847s/20 iters), loss = 0.107285
I0312 23:31:24.025619 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.953125
I0312 23:31:24.025655 2676 solver.cpp:238] Train net output #1: loss = 0.107285 (* 1 = 0.107285 loss)
I0312 23:31:24.025660 2676 sgd_solver.cpp:105] Iteration 1420, lr = 0.001
I0312 23:31:25.758661 2676 solver.cpp:331] Iteration 1428, Testing net (#0)
I0312 23:31:25.758697 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:31:27.764217 2676 solver.cpp:398] Test net output #0: accuracy = 0.925714
I0312 23:31:27.764256 2676 solver.cpp:398] Test net output #1: loss = 0.253493 (* 1 = 0.253493 loss)
I0312 23:31:31.272397 2676 solver.cpp:219] Iteration 1440 (2.76004 iter/s, 7.24627s/20 iters), loss = 0.102318
I0312 23:31:31.284703 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.972656
I0312 23:31:31.284739 2676 solver.cpp:238] Train net output #1: loss = 0.102318 (* 1 = 0.102318 loss)
I0312 23:31:31.284744 2676 sgd_solver.cpp:105] Iteration 1440, lr = 0.001
I0312 23:31:35.189929 2676 solver.cpp:331] Iteration 1456, Testing net (#0)
I0312 23:31:35.189951 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:31:37.209007 2676 solver.cpp:398] Test net output #0: accuracy = 0.933571
I0312 23:31:37.209045 2676 solver.cpp:398] Test net output #1: loss = 0.225641 (* 1 = 0.225641 loss)
I0312 23:31:38.545217 2676 solver.cpp:219] Iteration 1460 (2.75487 iter/s, 7.25988s/20 iters), loss = 0.120464
I0312 23:31:38.557488 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.953125
I0312 23:31:38.557521 2676 solver.cpp:238] Train net output #1: loss = 0.120464 (* 1 = 0.120464 loss)
I0312 23:31:38.557526 2676 sgd_solver.cpp:105] Iteration 1460, lr = 0.001
I0312 23:31:43.982724 2676 solver.cpp:219] Iteration 1480 (3.68685 iter/s, 5.42469s/20 iters), loss = 0.0925529
I0312 23:31:43.994879 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.96875
I0312 23:31:43.994912 2676 solver.cpp:238] Train net output #1: loss = 0.0925529 (* 1 = 0.0925529 loss)
I0312 23:31:43.994917 2676 sgd_solver.cpp:105] Iteration 1480, lr = 0.001
I0312 23:31:44.639364 2676 solver.cpp:331] Iteration 1484, Testing net (#0)
I0312 23:31:44.639400 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:31:46.645072 2676 solver.cpp:398] Test net output #0: accuracy = 0.930357
I0312 23:31:46.645110 2676 solver.cpp:398] Test net output #1: loss = 0.258542 (* 1 = 0.258542 loss)
I0312 23:31:51.239519 2676 solver.cpp:219] Iteration 1500 (2.76097 iter/s, 7.24383s/20 iters), loss = 0.0513216
I0312 23:31:51.251762 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.980469
I0312 23:31:51.251782 2676 solver.cpp:238] Train net output #1: loss = 0.0513216 (* 1 = 0.0513216 loss)
I0312 23:31:51.251788 2676 sgd_solver.cpp:105] Iteration 1500, lr = 0.001
I0312 23:31:54.071509 2676 solver.cpp:331] Iteration 1512, Testing net (#0)
I0312 23:31:54.071529 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:31:56.074208 2676 solver.cpp:398] Test net output #0: accuracy = 0.932857
I0312 23:31:56.074246 2676 solver.cpp:398] Test net output #1: loss = 0.253556 (* 1 = 0.253556 loss)
I0312 23:31:58.501843 2676 solver.cpp:219] Iteration 1520 (2.75893 iter/s, 7.24918s/20 iters), loss = 0.0631661
I0312 23:31:58.513955 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.984375
I0312 23:31:58.513990 2676 solver.cpp:238] Train net output #1: loss = 0.0631661 (* 1 = 0.0631661 loss)
I0312 23:31:58.513995 2676 sgd_solver.cpp:105] Iteration 1520, lr = 0.001
I0312 23:32:03.509240 2676 solver.cpp:331] Iteration 1540, Testing net (#0)
I0312 23:32:03.509333 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:32:04.005400 2676 blocking_queue.cpp:49] Waiting for data
I0312 23:32:05.520295 2676 solver.cpp:398] Test net output #0: accuracy = 0.937857
I0312 23:32:05.520334 2676 solver.cpp:398] Test net output #1: loss = 0.24086 (* 1 = 0.24086 loss)
I0312 23:32:05.784499 2676 solver.cpp:219] Iteration 1540 (2.7512 iter/s, 7.26955s/20 iters), loss = 0.103841
I0312 23:32:05.786963 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.964844
I0312 23:32:05.786994 2676 solver.cpp:238] Train net output #1: loss = 0.103841 (* 1 = 0.103841 loss)
I0312 23:32:05.786999 2676 sgd_solver.cpp:105] Iteration 1540, lr = 0.001
I0312 23:32:11.207936 2676 solver.cpp:219] Iteration 1560 (3.68991 iter/s, 5.42019s/20 iters), loss = 0.0426333
I0312 23:32:11.220110 2676 solver.cpp:238] Train net output #0: accuracy_training = 1
I0312 23:32:11.220142 2676 solver.cpp:238] Train net output #1: loss = 0.0426333 (* 1 = 0.0426333 loss)
I0312 23:32:11.220161 2676 sgd_solver.cpp:105] Iteration 1560, lr = 0.001
I0312 23:32:12.950600 2676 solver.cpp:331] Iteration 1568, Testing net (#0)
I0312 23:32:12.950635 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:32:14.952039 2676 solver.cpp:398] Test net output #0: accuracy = 0.943571
I0312 23:32:14.952078 2676 solver.cpp:398] Test net output #1: loss = 0.227019 (* 1 = 0.227019 loss)
I0312 23:32:18.464334 2676 solver.cpp:219] Iteration 1580 (2.76124 iter/s, 7.24312s/20 iters), loss = 0.0886056
I0312 23:32:18.476521 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.976562
I0312 23:32:18.476539 2676 solver.cpp:238] Train net output #1: loss = 0.0886056 (* 1 = 0.0886056 loss)
I0312 23:32:18.476546 2676 sgd_solver.cpp:105] Iteration 1580, lr = 0.001
I0312 23:32:22.385947 2676 solver.cpp:331] Iteration 1596, Testing net (#0)
I0312 23:32:22.385984 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:32:24.370674 2676 solver.cpp:398] Test net output #0: accuracy = 0.926785
I0312 23:32:24.370699 2676 solver.cpp:398] Test net output #1: loss = 0.288649 (* 1 = 0.288649 loss)
I0312 23:32:25.715095 2676 solver.cpp:219] Iteration 1600 (2.76342 iter/s, 7.23741s/20 iters), loss = 0.0739714
I0312 23:32:25.727288 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.984375
I0312 23:32:25.727321 2676 solver.cpp:238] Train net output #1: loss = 0.0739714 (* 1 = 0.0739714 loss)
I0312 23:32:25.727327 2676 sgd_solver.cpp:105] Iteration 1600, lr = 0.001
I0312 23:32:31.149646 2676 solver.cpp:219] Iteration 1620 (3.68905 iter/s, 5.42145s/20 iters), loss = 0.127903
I0312 23:32:31.161710 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.964844
I0312 23:32:31.161742 2676 solver.cpp:238] Train net output #1: loss = 0.127903 (* 1 = 0.127903 loss)
I0312 23:32:31.161747 2676 sgd_solver.cpp:105] Iteration 1620, lr = 0.001
I0312 23:32:31.809602 2676 solver.cpp:331] Iteration 1624, Testing net (#0)
I0312 23:32:31.809638 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:32:33.796406 2676 solver.cpp:398] Test net output #0: accuracy = 0.922857
I0312 23:32:33.796447 2676 solver.cpp:398] Test net output #1: loss = 0.277072 (* 1 = 0.277072 loss)
I0312 23:32:38.398217 2676 solver.cpp:219] Iteration 1640 (2.76424 iter/s, 7.23526s/20 iters), loss = 0.0360228
I0312 23:32:38.410373 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.992188
I0312 23:32:38.410404 2676 solver.cpp:238] Train net output #1: loss = 0.0360228 (* 1 = 0.0360228 loss)
I0312 23:32:38.410409 2676 sgd_solver.cpp:105] Iteration 1640, lr = 0.001
I0312 23:32:41.233564 2676 solver.cpp:331] Iteration 1652, Testing net (#0)
I0312 23:32:41.233599 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:32:43.231456 2676 solver.cpp:398] Test net output #0: accuracy = 0.934643
I0312 23:32:43.231490 2676 solver.cpp:398] Test net output #1: loss = 0.253163 (* 1 = 0.253163 loss)
I0312 23:32:45.651796 2676 solver.cpp:219] Iteration 1660 (2.76238 iter/s, 7.24013s/20 iters), loss = 0.0863931
I0312 23:32:45.663851 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.972656
I0312 23:32:45.663884 2676 solver.cpp:238] Train net output #1: loss = 0.0863931 (* 1 = 0.0863931 loss)
I0312 23:32:45.663889 2676 sgd_solver.cpp:105] Iteration 1660, lr = 0.001
I0312 23:32:50.653645 2676 solver.cpp:331] Iteration 1680, Testing net (#0)
I0312 23:32:50.653681 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:32:52.647557 2676 solver.cpp:398] Test net output #0: accuracy = 0.945714
I0312 23:32:52.647593 2676 solver.cpp:398] Test net output #1: loss = 0.239182 (* 1 = 0.239182 loss)
I0312 23:32:52.910799 2676 solver.cpp:219] Iteration 1680 (2.76029 iter/s, 7.24561s/20 iters), loss = 0.0613022
I0312 23:32:52.913255 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.976562
I0312 23:32:52.913285 2676 solver.cpp:238] Train net output #1: loss = 0.0613022 (* 1 = 0.0613022 loss)
I0312 23:32:52.913293 2676 sgd_solver.cpp:105] Iteration 1680, lr = 0.001
I0312 23:32:58.327962 2676 solver.cpp:219] Iteration 1700 (3.69434 iter/s, 5.41369s/20 iters), loss = 0.033861
I0312 23:32:58.340250 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.988281
I0312 23:32:58.340270 2676 solver.cpp:238] Train net output #1: loss = 0.033861 (* 1 = 0.033861 loss)
I0312 23:32:58.340276 2676 sgd_solver.cpp:105] Iteration 1700, lr = 0.001
I0312 23:33:00.073436 2676 solver.cpp:331] Iteration 1708, Testing net (#0)
I0312 23:33:00.073472 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:33:02.110150 2676 solver.cpp:398] Test net output #0: accuracy = 0.938928
I0312 23:33:02.110188 2676 solver.cpp:398] Test net output #1: loss = 0.250893 (* 1 = 0.250893 loss)
I0312 23:33:05.622922 2676 solver.cpp:219] Iteration 1720 (2.74765 iter/s, 7.27896s/20 iters), loss = 0.0755461
I0312 23:33:05.632652 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.96875
I0312 23:33:05.632681 2676 solver.cpp:238] Train net output #1: loss = 0.075546 (* 1 = 0.075546 loss)
I0312 23:33:05.632686 2676 sgd_solver.cpp:105] Iteration 1720, lr = 0.001
I0312 23:33:09.548780 2676 solver.cpp:331] Iteration 1736, Testing net (#0)
I0312 23:33:09.548804 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:33:11.557363 2676 solver.cpp:398] Test net output #0: accuracy = 0.942142
I0312 23:33:11.557402 2676 solver.cpp:398] Test net output #1: loss = 0.232101 (* 1 = 0.232101 loss)
I0312 23:33:12.893023 2676 solver.cpp:219] Iteration 1740 (2.75522 iter/s, 7.25894s/20 iters), loss = 0.0302835
I0312 23:33:12.905261 2676 solver.cpp:238] Train net output #0: accuracy_training = 1
I0312 23:33:12.905295 2676 solver.cpp:238] Train net output #1: loss = 0.0302835 (* 1 = 0.0302835 loss)
I0312 23:33:12.905302 2676 sgd_solver.cpp:105] Iteration 1740, lr = 0.001
I0312 23:33:18.338508 2676 solver.cpp:219] Iteration 1760 (3.68177 iter/s, 5.43217s/20 iters), loss = 0.06192
I0312 23:33:18.350666 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.984375
I0312 23:33:18.350698 2676 solver.cpp:238] Train net output #1: loss = 0.06192 (* 1 = 0.06192 loss)
I0312 23:33:18.350703 2676 sgd_solver.cpp:105] Iteration 1760, lr = 0.001
I0312 23:33:18.994849 2676 solver.cpp:331] Iteration 1764, Testing net (#0)
I0312 23:33:18.994885 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:33:21.028154 2676 solver.cpp:398] Test net output #0: accuracy = 0.927857
I0312 23:33:21.028192 2676 solver.cpp:398] Test net output #1: loss = 0.283865 (* 1 = 0.283865 loss)
I0312 23:33:25.627306 2676 solver.cpp:219] Iteration 1780 (2.74907 iter/s, 7.27518s/20 iters), loss = 0.0653128
I0312 23:33:25.639580 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.988281
I0312 23:33:25.639614 2676 solver.cpp:238] Train net output #1: loss = 0.0653128 (* 1 = 0.0653128 loss)
I0312 23:33:25.639619 2676 sgd_solver.cpp:105] Iteration 1780, lr = 0.001
I0312 23:33:28.458091 2676 solver.cpp:331] Iteration 1792, Testing net (#0)
I0312 23:33:28.458127 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:33:30.467082 2676 solver.cpp:398] Test net output #0: accuracy = 0.943214
I0312 23:33:30.467120 2676 solver.cpp:398] Test net output #1: loss = 0.22913 (* 1 = 0.22913 loss)
I0312 23:33:32.888509 2676 solver.cpp:219] Iteration 1800 (2.75959 iter/s, 7.24745s/20 iters), loss = 0.0399365
I0312 23:33:32.900758 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.996094
I0312 23:33:32.900791 2676 solver.cpp:238] Train net output #1: loss = 0.0399365 (* 1 = 0.0399365 loss)
I0312 23:33:32.900796 2676 sgd_solver.cpp:105] Iteration 1800, lr = 0.001
I0312 23:33:37.896623 2676 solver.cpp:331] Iteration 1820, Testing net (#0)
I0312 23:33:37.896724 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:33:39.885713 2676 solver.cpp:398] Test net output #0: accuracy = 0.944285
I0312 23:33:39.885751 2676 solver.cpp:398] Test net output #1: loss = 0.232312 (* 1 = 0.232312 loss)
I0312 23:33:40.148910 2676 solver.cpp:219] Iteration 1820 (2.75989 iter/s, 7.24666s/20 iters), loss = 0.0454593
I0312 23:33:40.151473 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.984375
I0312 23:33:40.151504 2676 solver.cpp:238] Train net output #1: loss = 0.0454593 (* 1 = 0.0454593 loss)
I0312 23:33:40.151510 2676 sgd_solver.cpp:105] Iteration 1820, lr = 0.001
I0312 23:33:45.579064 2676 solver.cpp:219] Iteration 1840 (3.68565 iter/s, 5.42645s/20 iters), loss = 0.028868
I0312 23:33:45.591228 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.992188
I0312 23:33:45.591262 2676 solver.cpp:238] Train net output #1: loss = 0.028868 (* 1 = 0.028868 loss)
I0312 23:33:45.591267 2676 sgd_solver.cpp:105] Iteration 1840, lr = 0.001
I0312 23:33:47.322468 2676 solver.cpp:331] Iteration 1848, Testing net (#0)
I0312 23:33:47.322490 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:33:49.315287 2676 solver.cpp:398] Test net output #0: accuracy = 0.946428
I0312 23:33:49.315312 2676 solver.cpp:398] Test net output #1: loss = 0.232146 (* 1 = 0.232146 loss)
I0312 23:33:52.824077 2676 solver.cpp:219] Iteration 1860 (2.76574 iter/s, 7.23133s/20 iters), loss = 0.0465076
I0312 23:33:52.836303 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.988281
I0312 23:33:52.836321 2676 solver.cpp:238] Train net output #1: loss = 0.0465076 (* 1 = 0.0465076 loss)
I0312 23:33:52.836325 2676 sgd_solver.cpp:105] Iteration 1860, lr = 0.001
I0312 23:33:56.744417 2676 solver.cpp:331] Iteration 1876, Testing net (#0)
I0312 23:33:56.744454 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:33:58.743549 2676 solver.cpp:398] Test net output #0: accuracy = 0.947142
I0312 23:33:58.743587 2676 solver.cpp:398] Test net output #1: loss = 0.236239 (* 1 = 0.236239 loss)
I0312 23:34:00.080644 2676 solver.cpp:219] Iteration 1880 (2.76136 iter/s, 7.2428s/20 iters), loss = 0.0632973
I0312 23:34:00.093142 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.984375
I0312 23:34:00.093181 2676 solver.cpp:238] Train net output #1: loss = 0.0632973 (* 1 = 0.0632973 loss)
I0312 23:34:00.093188 2676 sgd_solver.cpp:105] Iteration 1880, lr = 0.001
I0312 23:34:05.528113 2676 solver.cpp:219] Iteration 1900 (3.68066 iter/s, 5.43381s/20 iters), loss = 0.141142
I0312 23:34:05.540240 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.964844
I0312 23:34:05.540271 2676 solver.cpp:238] Train net output #1: loss = 0.141142 (* 1 = 0.141142 loss)
I0312 23:34:05.540277 2676 sgd_solver.cpp:105] Iteration 1900, lr = 0.001
I0312 23:34:06.184927 2676 solver.cpp:331] Iteration 1904, Testing net (#0)
I0312 23:34:06.184963 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:34:08.188815 2676 solver.cpp:398] Test net output #0: accuracy = 0.941071
I0312 23:34:08.188855 2676 solver.cpp:398] Test net output #1: loss = 0.250776 (* 1 = 0.250776 loss)
I0312 23:34:11.808614 2676 blocking_queue.cpp:49] Waiting for data
I0312 23:34:12.791224 2676 solver.cpp:219] Iteration 1920 (2.75755 iter/s, 7.25282s/20 iters), loss = 0.035214
I0312 23:34:12.803364 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.988281
I0312 23:34:12.803396 2676 solver.cpp:238] Train net output #1: loss = 0.035214 (* 1 = 0.035214 loss)
I0312 23:34:12.803401 2676 sgd_solver.cpp:105] Iteration 1920, lr = 0.001
I0312 23:34:15.624873 2676 solver.cpp:331] Iteration 1932, Testing net (#0)
I0312 23:34:15.624908 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:34:17.618657 2676 solver.cpp:398] Test net output #0: accuracy = 0.951785
I0312 23:34:17.618696 2676 solver.cpp:398] Test net output #1: loss = 0.226289 (* 1 = 0.226289 loss)
I0312 23:34:20.037569 2676 solver.cpp:219] Iteration 1940 (2.76299 iter/s, 7.23853s/20 iters), loss = 0.0472051
I0312 23:34:20.049713 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.996094
I0312 23:34:20.049732 2676 solver.cpp:238] Train net output #1: loss = 0.0472051 (* 1 = 0.0472051 loss)
I0312 23:34:20.049751 2676 sgd_solver.cpp:105] Iteration 1940, lr = 0.001
I0312 23:34:25.038463 2676 solver.cpp:331] Iteration 1960, Testing net (#0)
I0312 23:34:25.038498 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:34:27.059917 2676 solver.cpp:398] Test net output #0: accuracy = 0.946785
I0312 23:34:27.059942 2676 solver.cpp:398] Test net output #1: loss = 0.235603 (* 1 = 0.235603 loss)
I0312 23:34:27.325182 2676 solver.cpp:219] Iteration 1960 (2.74754 iter/s, 7.27925s/20 iters), loss = 0.0221539
I0312 23:34:27.327626 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.992188
I0312 23:34:27.327656 2676 solver.cpp:238] Train net output #1: loss = 0.0221539 (* 1 = 0.0221539 loss)
I0312 23:34:27.327661 2676 sgd_solver.cpp:105] Iteration 1960, lr = 0.001
I0312 23:34:32.754545 2676 solver.cpp:219] Iteration 1980 (3.68365 iter/s, 5.4294s/20 iters), loss = 0.0225337
I0312 23:34:32.766762 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.992188
I0312 23:34:32.766795 2676 solver.cpp:238] Train net output #1: loss = 0.0225337 (* 1 = 0.0225337 loss)
I0312 23:34:32.766801 2676 sgd_solver.cpp:105] Iteration 1980, lr = 0.001
I0312 23:34:34.501754 2676 solver.cpp:331] Iteration 1988, Testing net (#0)
I0312 23:34:34.501788 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:34:36.490181 2676 solver.cpp:398] Test net output #0: accuracy = 0.944642
I0312 23:34:36.490207 2676 solver.cpp:398] Test net output #1: loss = 0.232204 (* 1 = 0.232204 loss)
I0312 23:34:39.991500 2676 solver.cpp:219] Iteration 2000 (2.76715 iter/s, 7.22765s/20 iters), loss = 0.0507981
I0312 23:34:40.003767 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.988281
I0312 23:34:40.003788 2676 solver.cpp:238] Train net output #1: loss = 0.0507981 (* 1 = 0.0507981 loss)
I0312 23:34:40.003806 2676 sgd_solver.cpp:105] Iteration 2000, lr = 0.001
I0312 23:34:43.910578 2676 solver.cpp:331] Iteration 2016, Testing net (#0)
I0312 23:34:43.910599 2676 net.cpp:678] Ignoring source layer accuracy_training
I0312 23:34:45.930817 2676 solver.cpp:398] Test net output #0: accuracy = 0.948928
I0312 23:34:45.930855 2676 solver.cpp:398] Test net output #1: loss = 0.241619 (* 1 = 0.241619 loss)
I0312 23:34:47.266039 2676 solver.cpp:219] Iteration 2020 (2.75301 iter/s, 7.26478s/20 iters), loss = 0.0206446
I0312 23:34:47.278142 2676 solver.cpp:238] Train net output #0: accuracy_training = 0.996094
I0312 23:34:47.278175 2676 solver.cpp:238] Train net output #1: loss = 0.0206446 (* 1 = 0.0206446 loss)
I0312 23:34:47.278180 2676 sgd_solver.cpp:105] Iteration 2020, lr = 0.001