I0409 20:00:43.950959 15472 upgrade_proto.cpp:1082] Attempting to upgrade input file specified using deprecated 'solver_type' field (enum)': /mnt/bigdisk/DIGITS-MAN-2/digits/jobs/20210409-200041-7422/solver.prototxt
I0409 20:00:43.951164 15472 upgrade_proto.cpp:1089] Successfully upgraded file specified using deprecated 'solver_type' field (enum) to 'type' field (string).
W0409 20:00:43.951171 15472 upgrade_proto.cpp:1091] Note that future Caffe releases will only support 'type' field (string) for a solver's type.
I0409 20:00:43.951248 15472 caffe.cpp:218] Using GPUs 3
I0409 20:00:43.970679 15472 caffe.cpp:223] GPU 3: GeForce GTX 1080 Ti
I0409 20:00:44.263801 15472 solver.cpp:44] Initializing solver from parameters:
test_iter: 51
test_interval: 102
base_lr: 0.01
display: 12
max_iter: 10200
lr_policy: "exp"
gamma: 0.99980193
momentum: 0.9
weight_decay: 0.0001
snapshot: 102
snapshot_prefix: "snapshot"
solver_mode: GPU
device_id: 3
net: "train_val.prototxt"
train_state {
level: 0
stage: ""
}
type: "SGD"
I0409 20:00:44.303392 15472 solver.cpp:87] Creating training net from net file: train_val.prototxt
I0409 20:00:44.304064 15472 net.cpp:294] The NetState phase (0) differed from the phase (1) specified by a rule in layer val-data
I0409 20:00:44.304080 15472 net.cpp:294] The NetState phase (0) differed from the phase (1) specified by a rule in layer accuracy
I0409 20:00:44.304255 15472 net.cpp:51] Initializing net from parameters:
state {
phase: TRAIN
level: 0
stage: ""
}
layer {
name: "train-data"
type: "Data"
top: "data"
top: "label"
include {
phase: TRAIN
}
transform_param {
mirror: true
crop_size: 227
mean_file: "/mnt/bigdisk/DIGITS-MAN-2/digits/jobs/20210407-214532-d396/mean.binaryproto"
}
data_param {
source: "/mnt/bigdisk/DIGITS-MAN-2/digits/jobs/20210407-214532-d396/train_db"
batch_size: 128
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: "gaussian"
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: 512
weight_filler {
type: "gaussian"
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: 512
weight_filler {
type: "gaussian"
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: "fc7.5"
type: "InnerProduct"
bottom: "fc7"
top: "fc7.5"
param {
lr_mult: 1
decay_mult: 1
}
param {
lr_mult: 2
decay_mult: 0
}
inner_product_param {
num_output: 512
weight_filler {
type: "gaussian"
std: 0.005
}
bias_filler {
type: "constant"
value: 0.1
}
}
}
layer {
name: "relu7.5"
type: "ReLU"
bottom: "fc7.5"
top: "fc7.5"
}
layer {
name: "drop7.5"
type: "Dropout"
bottom: "fc7.5"
top: "fc7.5"
dropout_param {
dropout_ratio: 0.5
}
}
layer {
name: "fc8"
type: "InnerProduct"
bottom: "fc7.5"
top: "fc8"
param {
lr_mult: 1
decay_mult: 1
}
param {
lr_mult: 2
decay_mult: 0
}
inner_product_param {
num_output: 196
weight_filler {
type: "gaussian"
std: 0.01
}
bias_filler {
type: "constant"
value: 0
}
}
}
layer {
name: "loss"
type: "SoftmaxWithLoss"
bottom: "fc8"
bottom: "label"
top: "loss"
}
I0409 20:00:44.304364 15472 layer_factory.hpp:77] Creating layer train-data
I0409 20:00:44.307087 15472 db_lmdb.cpp:35] Opened lmdb /mnt/bigdisk/DIGITS-MAN-2/digits/jobs/20210407-214532-d396/train_db
I0409 20:00:44.307276 15472 net.cpp:84] Creating Layer train-data
I0409 20:00:44.307287 15472 net.cpp:380] train-data -> data
I0409 20:00:44.307307 15472 net.cpp:380] train-data -> label
I0409 20:00:44.307319 15472 data_transformer.cpp:25] Loading mean file from: /mnt/bigdisk/DIGITS-MAN-2/digits/jobs/20210407-214532-d396/mean.binaryproto
I0409 20:00:44.312110 15472 data_layer.cpp:45] output data size: 128,3,227,227
I0409 20:00:44.436342 15472 net.cpp:122] Setting up train-data
I0409 20:00:44.436367 15472 net.cpp:129] Top shape: 128 3 227 227 (19787136)
I0409 20:00:44.436372 15472 net.cpp:129] Top shape: 128 (128)
I0409 20:00:44.436376 15472 net.cpp:137] Memory required for data: 79149056
I0409 20:00:44.436388 15472 layer_factory.hpp:77] Creating layer conv1
I0409 20:00:44.436412 15472 net.cpp:84] Creating Layer conv1
I0409 20:00:44.436419 15472 net.cpp:406] conv1 <- data
I0409 20:00:44.436432 15472 net.cpp:380] conv1 -> conv1
I0409 20:00:45.003123 15472 net.cpp:122] Setting up conv1
I0409 20:00:45.003146 15472 net.cpp:129] Top shape: 128 96 55 55 (37171200)
I0409 20:00:45.003150 15472 net.cpp:137] Memory required for data: 227833856
I0409 20:00:45.003170 15472 layer_factory.hpp:77] Creating layer relu1
I0409 20:00:45.003199 15472 net.cpp:84] Creating Layer relu1
I0409 20:00:45.003204 15472 net.cpp:406] relu1 <- conv1
I0409 20:00:45.003211 15472 net.cpp:367] relu1 -> conv1 (in-place)
I0409 20:00:45.003502 15472 net.cpp:122] Setting up relu1
I0409 20:00:45.003512 15472 net.cpp:129] Top shape: 128 96 55 55 (37171200)
I0409 20:00:45.003515 15472 net.cpp:137] Memory required for data: 376518656
I0409 20:00:45.003520 15472 layer_factory.hpp:77] Creating layer norm1
I0409 20:00:45.003528 15472 net.cpp:84] Creating Layer norm1
I0409 20:00:45.003532 15472 net.cpp:406] norm1 <- conv1
I0409 20:00:45.003537 15472 net.cpp:380] norm1 -> norm1
I0409 20:00:45.003973 15472 net.cpp:122] Setting up norm1
I0409 20:00:45.003983 15472 net.cpp:129] Top shape: 128 96 55 55 (37171200)
I0409 20:00:45.003986 15472 net.cpp:137] Memory required for data: 525203456
I0409 20:00:45.003990 15472 layer_factory.hpp:77] Creating layer pool1
I0409 20:00:45.003998 15472 net.cpp:84] Creating Layer pool1
I0409 20:00:45.004002 15472 net.cpp:406] pool1 <- norm1
I0409 20:00:45.004009 15472 net.cpp:380] pool1 -> pool1
I0409 20:00:45.004043 15472 net.cpp:122] Setting up pool1
I0409 20:00:45.004050 15472 net.cpp:129] Top shape: 128 96 27 27 (8957952)
I0409 20:00:45.004053 15472 net.cpp:137] Memory required for data: 561035264
I0409 20:00:45.004056 15472 layer_factory.hpp:77] Creating layer conv2
I0409 20:00:45.004066 15472 net.cpp:84] Creating Layer conv2
I0409 20:00:45.004070 15472 net.cpp:406] conv2 <- pool1
I0409 20:00:45.004076 15472 net.cpp:380] conv2 -> conv2
I0409 20:00:45.011984 15472 net.cpp:122] Setting up conv2
I0409 20:00:45.012001 15472 net.cpp:129] Top shape: 128 256 27 27 (23887872)
I0409 20:00:45.012004 15472 net.cpp:137] Memory required for data: 656586752
I0409 20:00:45.012014 15472 layer_factory.hpp:77] Creating layer relu2
I0409 20:00:45.012022 15472 net.cpp:84] Creating Layer relu2
I0409 20:00:45.012027 15472 net.cpp:406] relu2 <- conv2
I0409 20:00:45.012032 15472 net.cpp:367] relu2 -> conv2 (in-place)
I0409 20:00:45.012444 15472 net.cpp:122] Setting up relu2
I0409 20:00:45.012452 15472 net.cpp:129] Top shape: 128 256 27 27 (23887872)
I0409 20:00:45.012456 15472 net.cpp:137] Memory required for data: 752138240
I0409 20:00:45.012460 15472 layer_factory.hpp:77] Creating layer norm2
I0409 20:00:45.012467 15472 net.cpp:84] Creating Layer norm2
I0409 20:00:45.012470 15472 net.cpp:406] norm2 <- conv2
I0409 20:00:45.012476 15472 net.cpp:380] norm2 -> norm2
I0409 20:00:45.012768 15472 net.cpp:122] Setting up norm2
I0409 20:00:45.012775 15472 net.cpp:129] Top shape: 128 256 27 27 (23887872)
I0409 20:00:45.012779 15472 net.cpp:137] Memory required for data: 847689728
I0409 20:00:45.012782 15472 layer_factory.hpp:77] Creating layer pool2
I0409 20:00:45.012790 15472 net.cpp:84] Creating Layer pool2
I0409 20:00:45.012794 15472 net.cpp:406] pool2 <- norm2
I0409 20:00:45.012799 15472 net.cpp:380] pool2 -> pool2
I0409 20:00:45.012825 15472 net.cpp:122] Setting up pool2
I0409 20:00:45.012830 15472 net.cpp:129] Top shape: 128 256 13 13 (5537792)
I0409 20:00:45.012835 15472 net.cpp:137] Memory required for data: 869840896
I0409 20:00:45.012837 15472 layer_factory.hpp:77] Creating layer conv3
I0409 20:00:45.012847 15472 net.cpp:84] Creating Layer conv3
I0409 20:00:45.012850 15472 net.cpp:406] conv3 <- pool2
I0409 20:00:45.012856 15472 net.cpp:380] conv3 -> conv3
I0409 20:00:45.022773 15472 net.cpp:122] Setting up conv3
I0409 20:00:45.022789 15472 net.cpp:129] Top shape: 128 384 13 13 (8306688)
I0409 20:00:45.022794 15472 net.cpp:137] Memory required for data: 903067648
I0409 20:00:45.022802 15472 layer_factory.hpp:77] Creating layer relu3
I0409 20:00:45.022812 15472 net.cpp:84] Creating Layer relu3
I0409 20:00:45.022816 15472 net.cpp:406] relu3 <- conv3
I0409 20:00:45.022821 15472 net.cpp:367] relu3 -> conv3 (in-place)
I0409 20:00:45.023295 15472 net.cpp:122] Setting up relu3
I0409 20:00:45.023308 15472 net.cpp:129] Top shape: 128 384 13 13 (8306688)
I0409 20:00:45.023311 15472 net.cpp:137] Memory required for data: 936294400
I0409 20:00:45.023315 15472 layer_factory.hpp:77] Creating layer conv4
I0409 20:00:45.023344 15472 net.cpp:84] Creating Layer conv4
I0409 20:00:45.023348 15472 net.cpp:406] conv4 <- conv3
I0409 20:00:45.023355 15472 net.cpp:380] conv4 -> conv4
I0409 20:00:45.033601 15472 net.cpp:122] Setting up conv4
I0409 20:00:45.033617 15472 net.cpp:129] Top shape: 128 384 13 13 (8306688)
I0409 20:00:45.033622 15472 net.cpp:137] Memory required for data: 969521152
I0409 20:00:45.033630 15472 layer_factory.hpp:77] Creating layer relu4
I0409 20:00:45.033638 15472 net.cpp:84] Creating Layer relu4
I0409 20:00:45.033643 15472 net.cpp:406] relu4 <- conv4
I0409 20:00:45.033651 15472 net.cpp:367] relu4 -> conv4 (in-place)
I0409 20:00:45.033998 15472 net.cpp:122] Setting up relu4
I0409 20:00:45.034008 15472 net.cpp:129] Top shape: 128 384 13 13 (8306688)
I0409 20:00:45.034011 15472 net.cpp:137] Memory required for data: 1002747904
I0409 20:00:45.034015 15472 layer_factory.hpp:77] Creating layer conv5
I0409 20:00:45.034026 15472 net.cpp:84] Creating Layer conv5
I0409 20:00:45.034030 15472 net.cpp:406] conv5 <- conv4
I0409 20:00:45.034036 15472 net.cpp:380] conv5 -> conv5
I0409 20:00:45.042307 15472 net.cpp:122] Setting up conv5
I0409 20:00:45.042325 15472 net.cpp:129] Top shape: 128 256 13 13 (5537792)
I0409 20:00:45.042327 15472 net.cpp:137] Memory required for data: 1024899072
I0409 20:00:45.042340 15472 layer_factory.hpp:77] Creating layer relu5
I0409 20:00:45.042347 15472 net.cpp:84] Creating Layer relu5
I0409 20:00:45.042351 15472 net.cpp:406] relu5 <- conv5
I0409 20:00:45.042358 15472 net.cpp:367] relu5 -> conv5 (in-place)
I0409 20:00:45.042834 15472 net.cpp:122] Setting up relu5
I0409 20:00:45.042845 15472 net.cpp:129] Top shape: 128 256 13 13 (5537792)
I0409 20:00:45.042848 15472 net.cpp:137] Memory required for data: 1047050240
I0409 20:00:45.042852 15472 layer_factory.hpp:77] Creating layer pool5
I0409 20:00:45.042860 15472 net.cpp:84] Creating Layer pool5
I0409 20:00:45.042863 15472 net.cpp:406] pool5 <- conv5
I0409 20:00:45.042868 15472 net.cpp:380] pool5 -> pool5
I0409 20:00:45.042906 15472 net.cpp:122] Setting up pool5
I0409 20:00:45.042912 15472 net.cpp:129] Top shape: 128 256 6 6 (1179648)
I0409 20:00:45.042915 15472 net.cpp:137] Memory required for data: 1051768832
I0409 20:00:45.042918 15472 layer_factory.hpp:77] Creating layer fc6
I0409 20:00:45.042929 15472 net.cpp:84] Creating Layer fc6
I0409 20:00:45.042932 15472 net.cpp:406] fc6 <- pool5
I0409 20:00:45.042938 15472 net.cpp:380] fc6 -> fc6
I0409 20:00:45.089130 15472 net.cpp:122] Setting up fc6
I0409 20:00:45.089148 15472 net.cpp:129] Top shape: 128 512 (65536)
I0409 20:00:45.089152 15472 net.cpp:137] Memory required for data: 1052030976
I0409 20:00:45.089161 15472 layer_factory.hpp:77] Creating layer relu6
I0409 20:00:45.089170 15472 net.cpp:84] Creating Layer relu6
I0409 20:00:45.089175 15472 net.cpp:406] relu6 <- fc6
I0409 20:00:45.089182 15472 net.cpp:367] relu6 -> fc6 (in-place)
I0409 20:00:45.090248 15472 net.cpp:122] Setting up relu6
I0409 20:00:45.090257 15472 net.cpp:129] Top shape: 128 512 (65536)
I0409 20:00:45.090260 15472 net.cpp:137] Memory required for data: 1052293120
I0409 20:00:45.090265 15472 layer_factory.hpp:77] Creating layer drop6
I0409 20:00:45.090271 15472 net.cpp:84] Creating Layer drop6
I0409 20:00:45.090274 15472 net.cpp:406] drop6 <- fc6
I0409 20:00:45.090281 15472 net.cpp:367] drop6 -> fc6 (in-place)
I0409 20:00:45.090307 15472 net.cpp:122] Setting up drop6
I0409 20:00:45.090314 15472 net.cpp:129] Top shape: 128 512 (65536)
I0409 20:00:45.090317 15472 net.cpp:137] Memory required for data: 1052555264
I0409 20:00:45.090322 15472 layer_factory.hpp:77] Creating layer fc7
I0409 20:00:45.090328 15472 net.cpp:84] Creating Layer fc7
I0409 20:00:45.090332 15472 net.cpp:406] fc7 <- fc6
I0409 20:00:45.090338 15472 net.cpp:380] fc7 -> fc7
I0409 20:00:45.092641 15472 net.cpp:122] Setting up fc7
I0409 20:00:45.092648 15472 net.cpp:129] Top shape: 128 512 (65536)
I0409 20:00:45.092650 15472 net.cpp:137] Memory required for data: 1052817408
I0409 20:00:45.092658 15472 layer_factory.hpp:77] Creating layer relu7
I0409 20:00:45.092680 15472 net.cpp:84] Creating Layer relu7
I0409 20:00:45.092684 15472 net.cpp:406] relu7 <- fc7
I0409 20:00:45.092690 15472 net.cpp:367] relu7 -> fc7 (in-place)
I0409 20:00:45.093171 15472 net.cpp:122] Setting up relu7
I0409 20:00:45.093180 15472 net.cpp:129] Top shape: 128 512 (65536)
I0409 20:00:45.093183 15472 net.cpp:137] Memory required for data: 1053079552
I0409 20:00:45.093187 15472 layer_factory.hpp:77] Creating layer drop7
I0409 20:00:45.093194 15472 net.cpp:84] Creating Layer drop7
I0409 20:00:45.093199 15472 net.cpp:406] drop7 <- fc7
I0409 20:00:45.093204 15472 net.cpp:367] drop7 -> fc7 (in-place)
I0409 20:00:45.093225 15472 net.cpp:122] Setting up drop7
I0409 20:00:45.093230 15472 net.cpp:129] Top shape: 128 512 (65536)
I0409 20:00:45.093233 15472 net.cpp:137] Memory required for data: 1053341696
I0409 20:00:45.093237 15472 layer_factory.hpp:77] Creating layer fc7.5
I0409 20:00:45.093245 15472 net.cpp:84] Creating Layer fc7.5
I0409 20:00:45.093248 15472 net.cpp:406] fc7.5 <- fc7
I0409 20:00:45.093255 15472 net.cpp:380] fc7.5 -> fc7.5
I0409 20:00:45.096127 15472 net.cpp:122] Setting up fc7.5
I0409 20:00:45.096135 15472 net.cpp:129] Top shape: 128 512 (65536)
I0409 20:00:45.096139 15472 net.cpp:137] Memory required for data: 1053603840
I0409 20:00:45.096146 15472 layer_factory.hpp:77] Creating layer relu7.5
I0409 20:00:45.096153 15472 net.cpp:84] Creating Layer relu7.5
I0409 20:00:45.096156 15472 net.cpp:406] relu7.5 <- fc7.5
I0409 20:00:45.096163 15472 net.cpp:367] relu7.5 -> fc7.5 (in-place)
I0409 20:00:45.096660 15472 net.cpp:122] Setting up relu7.5
I0409 20:00:45.096669 15472 net.cpp:129] Top shape: 128 512 (65536)
I0409 20:00:45.096673 15472 net.cpp:137] Memory required for data: 1053865984
I0409 20:00:45.096676 15472 layer_factory.hpp:77] Creating layer drop7.5
I0409 20:00:45.096681 15472 net.cpp:84] Creating Layer drop7.5
I0409 20:00:45.096685 15472 net.cpp:406] drop7.5 <- fc7.5
I0409 20:00:45.096693 15472 net.cpp:367] drop7.5 -> fc7.5 (in-place)
I0409 20:00:45.096716 15472 net.cpp:122] Setting up drop7.5
I0409 20:00:45.096721 15472 net.cpp:129] Top shape: 128 512 (65536)
I0409 20:00:45.096724 15472 net.cpp:137] Memory required for data: 1054128128
I0409 20:00:45.096727 15472 layer_factory.hpp:77] Creating layer fc8
I0409 20:00:45.096733 15472 net.cpp:84] Creating Layer fc8
I0409 20:00:45.096737 15472 net.cpp:406] fc8 <- fc7.5
I0409 20:00:45.096743 15472 net.cpp:380] fc8 -> fc8
I0409 20:00:45.097685 15472 net.cpp:122] Setting up fc8
I0409 20:00:45.097692 15472 net.cpp:129] Top shape: 128 196 (25088)
I0409 20:00:45.097695 15472 net.cpp:137] Memory required for data: 1054228480
I0409 20:00:45.097705 15472 layer_factory.hpp:77] Creating layer loss
I0409 20:00:45.097712 15472 net.cpp:84] Creating Layer loss
I0409 20:00:45.097715 15472 net.cpp:406] loss <- fc8
I0409 20:00:45.097719 15472 net.cpp:406] loss <- label
I0409 20:00:45.097726 15472 net.cpp:380] loss -> loss
I0409 20:00:45.097738 15472 layer_factory.hpp:77] Creating layer loss
I0409 20:00:45.098328 15472 net.cpp:122] Setting up loss
I0409 20:00:45.098338 15472 net.cpp:129] Top shape: (1)
I0409 20:00:45.098341 15472 net.cpp:132]     with loss weight 1
I0409 20:00:45.098359 15472 net.cpp:137] Memory required for data: 1054228484
I0409 20:00:45.098364 15472 net.cpp:198] loss needs backward computation.
I0409 20:00:45.098371 15472 net.cpp:198] fc8 needs backward computation.
I0409 20:00:45.098374 15472 net.cpp:198] drop7.5 needs backward computation.
I0409 20:00:45.098377 15472 net.cpp:198] relu7.5 needs backward computation.
I0409 20:00:45.098381 15472 net.cpp:198] fc7.5 needs backward computation.
I0409 20:00:45.098384 15472 net.cpp:198] drop7 needs backward computation.
I0409 20:00:45.098388 15472 net.cpp:198] relu7 needs backward computation.
I0409 20:00:45.098392 15472 net.cpp:198] fc7 needs backward computation.
I0409 20:00:45.098395 15472 net.cpp:198] drop6 needs backward computation.
I0409 20:00:45.098399 15472 net.cpp:198] relu6 needs backward computation.
I0409 20:00:45.098402 15472 net.cpp:198] fc6 needs backward computation.
I0409 20:00:45.098419 15472 net.cpp:198] pool5 needs backward computation.
I0409 20:00:45.098423 15472 net.cpp:198] relu5 needs backward computation.
I0409 20:00:45.098428 15472 net.cpp:198] conv5 needs backward computation.
I0409 20:00:45.098431 15472 net.cpp:198] relu4 needs backward computation.
I0409 20:00:45.098434 15472 net.cpp:198] conv4 needs backward computation.
I0409 20:00:45.098438 15472 net.cpp:198] relu3 needs backward computation.
I0409 20:00:45.098443 15472 net.cpp:198] conv3 needs backward computation.
I0409 20:00:45.098446 15472 net.cpp:198] pool2 needs backward computation.
I0409 20:00:45.098450 15472 net.cpp:198] norm2 needs backward computation.
I0409 20:00:45.098453 15472 net.cpp:198] relu2 needs backward computation.
I0409 20:00:45.098456 15472 net.cpp:198] conv2 needs backward computation.
I0409 20:00:45.098460 15472 net.cpp:198] pool1 needs backward computation.
I0409 20:00:45.098464 15472 net.cpp:198] norm1 needs backward computation.
I0409 20:00:45.098469 15472 net.cpp:198] relu1 needs backward computation.
I0409 20:00:45.098471 15472 net.cpp:198] conv1 needs backward computation.
I0409 20:00:45.098475 15472 net.cpp:200] train-data does not need backward computation.
I0409 20:00:45.098479 15472 net.cpp:242] This network produces output loss
I0409 20:00:45.098495 15472 net.cpp:255] Network initialization done.
I0409 20:00:45.126675 15472 solver.cpp:172] Creating test net (#0) specified by net file: train_val.prototxt
I0409 20:00:45.126716 15472 net.cpp:294] The NetState phase (1) differed from the phase (0) specified by a rule in layer train-data
I0409 20:00:45.126870 15472 net.cpp:51] Initializing net from parameters:
state {
phase: TEST
}
layer {
name: "val-data"
type: "Data"
top: "data"
top: "label"
include {
phase: TEST
}
transform_param {
crop_size: 227
mean_file: "/mnt/bigdisk/DIGITS-MAN-2/digits/jobs/20210407-214532-d396/mean.binaryproto"
}
data_param {
source: "/mnt/bigdisk/DIGITS-MAN-2/digits/jobs/20210407-214532-d396/val_db"
batch_size: 32
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: "gaussian"
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: 512
weight_filler {
type: "gaussian"
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: 512
weight_filler {
type: "gaussian"
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: "fc7.5"
type: "InnerProduct"
bottom: "fc7"
top: "fc7.5"
param {
lr_mult: 1
decay_mult: 1
}
param {
lr_mult: 2
decay_mult: 0
}
inner_product_param {
num_output: 512
weight_filler {
type: "gaussian"
std: 0.005
}
bias_filler {
type: "constant"
value: 0.1
}
}
}
layer {
name: "relu7.5"
type: "ReLU"
bottom: "fc7.5"
top: "fc7.5"
}
layer {
name: "drop7.5"
type: "Dropout"
bottom: "fc7.5"
top: "fc7.5"
dropout_param {
dropout_ratio: 0.5
}
}
layer {
name: "fc8"
type: "InnerProduct"
bottom: "fc7.5"
top: "fc8"
param {
lr_mult: 1
decay_mult: 1
}
param {
lr_mult: 2
decay_mult: 0
}
inner_product_param {
num_output: 196
weight_filler {
type: "gaussian"
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"
}
I0409 20:00:45.126977 15472 layer_factory.hpp:77] Creating layer val-data
I0409 20:00:45.304920 15472 db_lmdb.cpp:35] Opened lmdb /mnt/bigdisk/DIGITS-MAN-2/digits/jobs/20210407-214532-d396/val_db
I0409 20:00:45.305320 15472 net.cpp:84] Creating Layer val-data
I0409 20:00:45.305335 15472 net.cpp:380] val-data -> data
I0409 20:00:45.305347 15472 net.cpp:380] val-data -> label
I0409 20:00:45.305357 15472 data_transformer.cpp:25] Loading mean file from: /mnt/bigdisk/DIGITS-MAN-2/digits/jobs/20210407-214532-d396/mean.binaryproto
I0409 20:00:45.329939 15472 data_layer.cpp:45] output data size: 32,3,227,227
I0409 20:00:45.379732 15472 net.cpp:122] Setting up val-data
I0409 20:00:45.379757 15472 net.cpp:129] Top shape: 32 3 227 227 (4946784)
I0409 20:00:45.379763 15472 net.cpp:129] Top shape: 32 (32)
I0409 20:00:45.379767 15472 net.cpp:137] Memory required for data: 19787264
I0409 20:00:45.379797 15472 layer_factory.hpp:77] Creating layer label_val-data_1_split
I0409 20:00:45.379813 15472 net.cpp:84] Creating Layer label_val-data_1_split
I0409 20:00:45.379819 15472 net.cpp:406] label_val-data_1_split <- label
I0409 20:00:45.379828 15472 net.cpp:380] label_val-data_1_split -> label_val-data_1_split_0
I0409 20:00:45.379840 15472 net.cpp:380] label_val-data_1_split -> label_val-data_1_split_1
I0409 20:00:45.379946 15472 net.cpp:122] Setting up label_val-data_1_split
I0409 20:00:45.379956 15472 net.cpp:129] Top shape: 32 (32)
I0409 20:00:45.379961 15472 net.cpp:129] Top shape: 32 (32)
I0409 20:00:45.379964 15472 net.cpp:137] Memory required for data: 19787520
I0409 20:00:45.379969 15472 layer_factory.hpp:77] Creating layer conv1
I0409 20:00:45.379985 15472 net.cpp:84] Creating Layer conv1
I0409 20:00:45.379989 15472 net.cpp:406] conv1 <- data
I0409 20:00:45.379998 15472 net.cpp:380] conv1 -> conv1
I0409 20:00:45.382531 15472 net.cpp:122] Setting up conv1
I0409 20:00:45.382545 15472 net.cpp:129] Top shape: 32 96 55 55 (9292800)
I0409 20:00:45.382550 15472 net.cpp:137] Memory required for data: 56958720
I0409 20:00:45.382563 15472 layer_factory.hpp:77] Creating layer relu1
I0409 20:00:45.382572 15472 net.cpp:84] Creating Layer relu1
I0409 20:00:45.382577 15472 net.cpp:406] relu1 <- conv1
I0409 20:00:45.382584 15472 net.cpp:367] relu1 -> conv1 (in-place)
I0409 20:00:45.383179 15472 net.cpp:122] Setting up relu1
I0409 20:00:45.383191 15472 net.cpp:129] Top shape: 32 96 55 55 (9292800)
I0409 20:00:45.383196 15472 net.cpp:137] Memory required for data: 94129920
I0409 20:00:45.383201 15472 layer_factory.hpp:77] Creating layer norm1
I0409 20:00:45.383211 15472 net.cpp:84] Creating Layer norm1
I0409 20:00:45.383216 15472 net.cpp:406] norm1 <- conv1
I0409 20:00:45.383224 15472 net.cpp:380] norm1 -> norm1
I0409 20:00:45.384934 15472 net.cpp:122] Setting up norm1
I0409 20:00:45.384948 15472 net.cpp:129] Top shape: 32 96 55 55 (9292800)
I0409 20:00:45.384953 15472 net.cpp:137] Memory required for data: 131301120
I0409 20:00:45.384958 15472 layer_factory.hpp:77] Creating layer pool1
I0409 20:00:45.384968 15472 net.cpp:84] Creating Layer pool1
I0409 20:00:45.384972 15472 net.cpp:406] pool1 <- norm1
I0409 20:00:45.384979 15472 net.cpp:380] pool1 -> pool1
I0409 20:00:45.385021 15472 net.cpp:122] Setting up pool1
I0409 20:00:45.385030 15472 net.cpp:129] Top shape: 32 96 27 27 (2239488)
I0409 20:00:45.385033 15472 net.cpp:137] Memory required for data: 140259072
I0409 20:00:45.385037 15472 layer_factory.hpp:77] Creating layer conv2
I0409 20:00:45.385048 15472 net.cpp:84] Creating Layer conv2
I0409 20:00:45.385052 15472 net.cpp:406] conv2 <- pool1
I0409 20:00:45.385059 15472 net.cpp:380] conv2 -> conv2
I0409 20:00:45.395150 15472 net.cpp:122] Setting up conv2
I0409 20:00:45.395170 15472 net.cpp:129] Top shape: 32 256 27 27 (5971968)
I0409 20:00:45.395175 15472 net.cpp:137] Memory required for data: 164146944
I0409 20:00:45.395190 15472 layer_factory.hpp:77] Creating layer relu2
I0409 20:00:45.395200 15472 net.cpp:84] Creating Layer relu2
I0409 20:00:45.395205 15472 net.cpp:406] relu2 <- conv2
I0409 20:00:45.395212 15472 net.cpp:367] relu2 -> conv2 (in-place)
I0409 20:00:45.395864 15472 net.cpp:122] Setting up relu2
I0409 20:00:45.395875 15472 net.cpp:129] Top shape: 32 256 27 27 (5971968)
I0409 20:00:45.395879 15472 net.cpp:137] Memory required for data: 188034816
I0409 20:00:45.395884 15472 layer_factory.hpp:77] Creating layer norm2
I0409 20:00:45.395896 15472 net.cpp:84] Creating Layer norm2
I0409 20:00:45.395901 15472 net.cpp:406] norm2 <- conv2
I0409 20:00:45.395910 15472 net.cpp:380] norm2 -> norm2
I0409 20:00:45.396394 15472 net.cpp:122] Setting up norm2
I0409 20:00:45.396404 15472 net.cpp:129] Top shape: 32 256 27 27 (5971968)
I0409 20:00:45.396409 15472 net.cpp:137] Memory required for data: 211922688
I0409 20:00:45.396414 15472 layer_factory.hpp:77] Creating layer pool2
I0409 20:00:45.396422 15472 net.cpp:84] Creating Layer pool2
I0409 20:00:45.396427 15472 net.cpp:406] pool2 <- norm2
I0409 20:00:45.396453 15472 net.cpp:380] pool2 -> pool2
I0409 20:00:45.396497 15472 net.cpp:122] Setting up pool2
I0409 20:00:45.396503 15472 net.cpp:129] Top shape: 32 256 13 13 (1384448)
I0409 20:00:45.396507 15472 net.cpp:137] Memory required for data: 217460480
I0409 20:00:45.396512 15472 layer_factory.hpp:77] Creating layer conv3
I0409 20:00:45.396523 15472 net.cpp:84] Creating Layer conv3
I0409 20:00:45.396528 15472 net.cpp:406] conv3 <- pool2
I0409 20:00:45.396536 15472 net.cpp:380] conv3 -> conv3
I0409 20:00:45.411099 15472 net.cpp:122] Setting up conv3
I0409 20:00:45.411119 15472 net.cpp:129] Top shape: 32 384 13 13 (2076672)
I0409 20:00:45.411124 15472 net.cpp:137] Memory required for data: 225767168
I0409 20:00:45.411139 15472 layer_factory.hpp:77] Creating layer relu3
I0409 20:00:45.411150 15472 net.cpp:84] Creating Layer relu3
I0409 20:00:45.411155 15472 net.cpp:406] relu3 <- conv3
I0409 20:00:45.411164 15472 net.cpp:367] relu3 -> conv3 (in-place)
I0409 20:00:45.411612 15472 net.cpp:122] Setting up relu3
I0409 20:00:45.411623 15472 net.cpp:129] Top shape: 32 384 13 13 (2076672)
I0409 20:00:45.411626 15472 net.cpp:137] Memory required for data: 234073856
I0409 20:00:45.411630 15472 layer_factory.hpp:77] Creating layer conv4
I0409 20:00:45.411645 15472 net.cpp:84] Creating Layer conv4
I0409 20:00:45.411649 15472 net.cpp:406] conv4 <- conv3
I0409 20:00:45.411657 15472 net.cpp:380] conv4 -> conv4
I0409 20:00:45.425019 15472 net.cpp:122] Setting up conv4
I0409 20:00:45.425040 15472 net.cpp:129] Top shape: 32 384 13 13 (2076672)
I0409 20:00:45.425045 15472 net.cpp:137] Memory required for data: 242380544
I0409 20:00:45.425053 15472 layer_factory.hpp:77] Creating layer relu4
I0409 20:00:45.425063 15472 net.cpp:84] Creating Layer relu4
I0409 20:00:45.425068 15472 net.cpp:406] relu4 <- conv4
I0409 20:00:45.425076 15472 net.cpp:367] relu4 -> conv4 (in-place)
I0409 20:00:45.425671 15472 net.cpp:122] Setting up relu4
I0409 20:00:45.425683 15472 net.cpp:129] Top shape: 32 384 13 13 (2076672)
I0409 20:00:45.425686 15472 net.cpp:137] Memory required for data: 250687232
I0409 20:00:45.425691 15472 layer_factory.hpp:77] Creating layer conv5
I0409 20:00:45.425704 15472 net.cpp:84] Creating Layer conv5
I0409 20:00:45.425709 15472 net.cpp:406] conv5 <- conv4
I0409 20:00:45.425717 15472 net.cpp:380] conv5 -> conv5
I0409 20:00:45.437500 15472 net.cpp:122] Setting up conv5
I0409 20:00:45.437517 15472 net.cpp:129] Top shape: 32 256 13 13 (1384448)
I0409 20:00:45.437522 15472 net.cpp:137] Memory required for data: 256225024
I0409 20:00:45.437536 15472 layer_factory.hpp:77] Creating layer relu5
I0409 20:00:45.437547 15472 net.cpp:84] Creating Layer relu5
I0409 20:00:45.437553 15472 net.cpp:406] relu5 <- conv5
I0409 20:00:45.437561 15472 net.cpp:367] relu5 -> conv5 (in-place)
I0409 20:00:45.438369 15472 net.cpp:122] Setting up relu5
I0409 20:00:45.438381 15472 net.cpp:129] Top shape: 32 256 13 13 (1384448)
I0409 20:00:45.438385 15472 net.cpp:137] Memory required for data: 261762816
I0409 20:00:45.438390 15472 layer_factory.hpp:77] Creating layer pool5
I0409 20:00:45.438402 15472 net.cpp:84] Creating Layer pool5
I0409 20:00:45.438407 15472 net.cpp:406] pool5 <- conv5
I0409 20:00:45.438414 15472 net.cpp:380] pool5 -> pool5
I0409 20:00:45.438459 15472 net.cpp:122] Setting up pool5
I0409 20:00:45.438467 15472 net.cpp:129] Top shape: 32 256 6 6 (294912)
I0409 20:00:45.438470 15472 net.cpp:137] Memory required for data: 262942464
I0409 20:00:45.438474 15472 layer_factory.hpp:77] Creating layer fc6
I0409 20:00:45.438484 15472 net.cpp:84] Creating Layer fc6
I0409 20:00:45.438488 15472 net.cpp:406] fc6 <- pool5
I0409 20:00:45.438495 15472 net.cpp:380] fc6 -> fc6
I0409 20:00:45.488385 15472 net.cpp:122] Setting up fc6
I0409 20:00:45.488404 15472 net.cpp:129] Top shape: 32 512 (16384)
I0409 20:00:45.488409 15472 net.cpp:137] Memory required for data: 263008000
I0409 20:00:45.488417 15472 layer_factory.hpp:77] Creating layer relu6
I0409 20:00:45.488427 15472 net.cpp:84] Creating Layer relu6
I0409 20:00:45.488432 15472 net.cpp:406] relu6 <- fc6
I0409 20:00:45.488459 15472 net.cpp:367] relu6 -> fc6 (in-place)
I0409 20:00:45.488900 15472 net.cpp:122] Setting up relu6
I0409 20:00:45.488909 15472 net.cpp:129] Top shape: 32 512 (16384)
I0409 20:00:45.488912 15472 net.cpp:137] Memory required for data: 263073536
I0409 20:00:45.488916 15472 layer_factory.hpp:77] Creating layer drop6
I0409 20:00:45.488924 15472 net.cpp:84] Creating Layer drop6
I0409 20:00:45.488929 15472 net.cpp:406] drop6 <- fc6
I0409 20:00:45.488935 15472 net.cpp:367] drop6 -> fc6 (in-place)
I0409 20:00:45.488961 15472 net.cpp:122] Setting up drop6
I0409 20:00:45.488966 15472 net.cpp:129] Top shape: 32 512 (16384)
I0409 20:00:45.488970 15472 net.cpp:137] Memory required for data: 263139072
I0409 20:00:45.488973 15472 layer_factory.hpp:77] Creating layer fc7
I0409 20:00:45.488981 15472 net.cpp:84] Creating Layer fc7
I0409 20:00:45.488986 15472 net.cpp:406] fc7 <- fc6
I0409 20:00:45.488991 15472 net.cpp:380] fc7 -> fc7
I0409 20:00:45.492183 15472 net.cpp:122] Setting up fc7
I0409 20:00:45.492193 15472 net.cpp:129] Top shape: 32 512 (16384)
I0409 20:00:45.492197 15472 net.cpp:137] Memory required for data: 263204608
I0409 20:00:45.492203 15472 layer_factory.hpp:77] Creating layer relu7
I0409 20:00:45.492210 15472 net.cpp:84] Creating Layer relu7
I0409 20:00:45.492214 15472 net.cpp:406] relu7 <- fc7
I0409 20:00:45.492219 15472 net.cpp:367] relu7 -> fc7 (in-place)
I0409 20:00:45.492853 15472 net.cpp:122] Setting up relu7
I0409 20:00:45.492863 15472 net.cpp:129] Top shape: 32 512 (16384)
I0409 20:00:45.492867 15472 net.cpp:137] Memory required for data: 263270144
I0409 20:00:45.492871 15472 layer_factory.hpp:77] Creating layer drop7
I0409 20:00:45.492877 15472 net.cpp:84] Creating Layer drop7
I0409 20:00:45.492882 15472 net.cpp:406] drop7 <- fc7
I0409 20:00:45.492888 15472 net.cpp:367] drop7 -> fc7 (in-place)
I0409 20:00:45.492918 15472 net.cpp:122] Setting up drop7
I0409 20:00:45.492923 15472 net.cpp:129] Top shape: 32 512 (16384)
I0409 20:00:45.492925 15472 net.cpp:137] Memory required for data: 263335680
I0409 20:00:45.492929 15472 layer_factory.hpp:77] Creating layer fc7.5
I0409 20:00:45.492936 15472 net.cpp:84] Creating Layer fc7.5
I0409 20:00:45.492940 15472 net.cpp:406] fc7.5 <- fc7
I0409 20:00:45.492947 15472 net.cpp:380] fc7.5 -> fc7.5
I0409 20:00:45.495457 15472 net.cpp:122] Setting up fc7.5
I0409 20:00:45.495465 15472 net.cpp:129] Top shape: 32 512 (16384)
I0409 20:00:45.495469 15472 net.cpp:137] Memory required for data: 263401216
I0409 20:00:45.495476 15472 layer_factory.hpp:77] Creating layer relu7.5
I0409 20:00:45.495482 15472 net.cpp:84] Creating Layer relu7.5
I0409 20:00:45.495486 15472 net.cpp:406] relu7.5 <- fc7.5
I0409 20:00:45.495492 15472 net.cpp:367] relu7.5 -> fc7.5 (in-place)
I0409 20:00:45.496026 15472 net.cpp:122] Setting up relu7.5
I0409 20:00:45.496035 15472 net.cpp:129] Top shape: 32 512 (16384)
I0409 20:00:45.496039 15472 net.cpp:137] Memory required for data: 263466752
I0409 20:00:45.496043 15472 layer_factory.hpp:77] Creating layer drop7.5
I0409 20:00:45.496052 15472 net.cpp:84] Creating Layer drop7.5
I0409 20:00:45.496055 15472 net.cpp:406] drop7.5 <- fc7.5
I0409 20:00:45.496060 15472 net.cpp:367] drop7.5 -> fc7.5 (in-place)
I0409 20:00:45.496088 15472 net.cpp:122] Setting up drop7.5
I0409 20:00:45.496093 15472 net.cpp:129] Top shape: 32 512 (16384)
I0409 20:00:45.496096 15472 net.cpp:137] Memory required for data: 263532288
I0409 20:00:45.496099 15472 layer_factory.hpp:77] Creating layer fc8
I0409 20:00:45.496105 15472 net.cpp:84] Creating Layer fc8
I0409 20:00:45.496109 15472 net.cpp:406] fc8 <- fc7.5
I0409 20:00:45.496116 15472 net.cpp:380] fc8 -> fc8
I0409 20:00:45.497148 15472 net.cpp:122] Setting up fc8
I0409 20:00:45.497154 15472 net.cpp:129] Top shape: 32 196 (6272)
I0409 20:00:45.497157 15472 net.cpp:137] Memory required for data: 263557376
I0409 20:00:45.497169 15472 layer_factory.hpp:77] Creating layer fc8_fc8_0_split
I0409 20:00:45.497176 15472 net.cpp:84] Creating Layer fc8_fc8_0_split
I0409 20:00:45.497180 15472 net.cpp:406] fc8_fc8_0_split <- fc8
I0409 20:00:45.497186 15472 net.cpp:380] fc8_fc8_0_split -> fc8_fc8_0_split_0
I0409 20:00:45.497208 15472 net.cpp:380] fc8_fc8_0_split -> fc8_fc8_0_split_1
I0409 20:00:45.497243 15472 net.cpp:122] Setting up fc8_fc8_0_split
I0409 20:00:45.497248 15472 net.cpp:129] Top shape: 32 196 (6272)
I0409 20:00:45.497252 15472 net.cpp:129] Top shape: 32 196 (6272)
I0409 20:00:45.497256 15472 net.cpp:137] Memory required for data: 263607552
I0409 20:00:45.497259 15472 layer_factory.hpp:77] Creating layer accuracy
I0409 20:00:45.497267 15472 net.cpp:84] Creating Layer accuracy
I0409 20:00:45.497270 15472 net.cpp:406] accuracy <- fc8_fc8_0_split_0
I0409 20:00:45.497275 15472 net.cpp:406] accuracy <- label_val-data_1_split_0
I0409 20:00:45.497282 15472 net.cpp:380] accuracy -> accuracy
I0409 20:00:45.497289 15472 net.cpp:122] Setting up accuracy
I0409 20:00:45.497293 15472 net.cpp:129] Top shape: (1)
I0409 20:00:45.497296 15472 net.cpp:137] Memory required for data: 263607556
I0409 20:00:45.497300 15472 layer_factory.hpp:77] Creating layer loss
I0409 20:00:45.497306 15472 net.cpp:84] Creating Layer loss
I0409 20:00:45.497309 15472 net.cpp:406] loss <- fc8_fc8_0_split_1
I0409 20:00:45.497314 15472 net.cpp:406] loss <- label_val-data_1_split_1
I0409 20:00:45.497319 15472 net.cpp:380] loss -> loss
I0409 20:00:45.497328 15472 layer_factory.hpp:77] Creating layer loss
I0409 20:00:45.499023 15472 net.cpp:122] Setting up loss
I0409 20:00:45.499035 15472 net.cpp:129] Top shape: (1)
I0409 20:00:45.499039 15472 net.cpp:132]     with loss weight 1
I0409 20:00:45.499050 15472 net.cpp:137] Memory required for data: 263607560
I0409 20:00:45.499054 15472 net.cpp:198] loss needs backward computation.
I0409 20:00:45.499060 15472 net.cpp:200] accuracy does not need backward computation.
I0409 20:00:45.499064 15472 net.cpp:198] fc8_fc8_0_split needs backward computation.
I0409 20:00:45.499068 15472 net.cpp:198] fc8 needs backward computation.
I0409 20:00:45.499073 15472 net.cpp:198] drop7.5 needs backward computation.
I0409 20:00:45.499075 15472 net.cpp:198] relu7.5 needs backward computation.
I0409 20:00:45.499079 15472 net.cpp:198] fc7.5 needs backward computation.
I0409 20:00:45.499083 15472 net.cpp:198] drop7 needs backward computation.
I0409 20:00:45.499086 15472 net.cpp:198] relu7 needs backward computation.
I0409 20:00:45.499090 15472 net.cpp:198] fc7 needs backward computation.
I0409 20:00:45.499094 15472 net.cpp:198] drop6 needs backward computation.
I0409 20:00:45.499099 15472 net.cpp:198] relu6 needs backward computation.
I0409 20:00:45.499101 15472 net.cpp:198] fc6 needs backward computation.
I0409 20:00:45.499105 15472 net.cpp:198] pool5 needs backward computation.
I0409 20:00:45.499110 15472 net.cpp:198] relu5 needs backward computation.
I0409 20:00:45.499114 15472 net.cpp:198] conv5 needs backward computation.
I0409 20:00:45.499119 15472 net.cpp:198] relu4 needs backward computation.
I0409 20:00:45.499123 15472 net.cpp:198] conv4 needs backward computation.
I0409 20:00:45.499127 15472 net.cpp:198] relu3 needs backward computation.
I0409 20:00:45.499131 15472 net.cpp:198] conv3 needs backward computation.
I0409 20:00:45.499135 15472 net.cpp:198] pool2 needs backward computation.
I0409 20:00:45.499140 15472 net.cpp:198] norm2 needs backward computation.
I0409 20:00:45.499143 15472 net.cpp:198] relu2 needs backward computation.
I0409 20:00:45.499147 15472 net.cpp:198] conv2 needs backward computation.
I0409 20:00:45.499150 15472 net.cpp:198] pool1 needs backward computation.
I0409 20:00:45.499155 15472 net.cpp:198] norm1 needs backward computation.
I0409 20:00:45.499158 15472 net.cpp:198] relu1 needs backward computation.
I0409 20:00:45.499161 15472 net.cpp:198] conv1 needs backward computation.
I0409 20:00:45.499166 15472 net.cpp:200] label_val-data_1_split does not need backward computation.
I0409 20:00:45.499171 15472 net.cpp:200] val-data does not need backward computation.
I0409 20:00:45.499176 15472 net.cpp:242] This network produces output accuracy
I0409 20:00:45.499181 15472 net.cpp:242] This network produces output loss
I0409 20:00:45.499202 15472 net.cpp:255] Network initialization done.
I0409 20:00:45.499292 15472 solver.cpp:56] Solver scaffolding done.
I0409 20:00:45.499845 15472 caffe.cpp:248] Starting Optimization
I0409 20:00:45.499855 15472 solver.cpp:272] Solving
I0409 20:00:45.499857 15472 solver.cpp:273] Learning Rate Policy: exp
I0409 20:00:45.500847 15472 solver.cpp:330] Iteration 0, Testing net (#0)
I0409 20:00:45.500856 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:00:45.519170 15472 blocking_queue.cpp:49] Waiting for data
I0409 20:00:49.928805 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:00:49.972939 15472 solver.cpp:397]     Test net output #0: accuracy = 0.00551471
I0409 20:00:49.972988 15472 solver.cpp:397]     Test net output #1: loss = 5.27828 (* 1 = 5.27828 loss)
I0409 20:00:50.062170 15472 solver.cpp:218] Iteration 0 (-4.56362e-18 iter/s, 4.56209s/12 iters), loss = 5.27708
I0409 20:00:50.063685 15472 solver.cpp:237]     Train net output #0: loss = 5.27708 (* 1 = 5.27708 loss)
I0409 20:00:50.063709 15472 sgd_solver.cpp:105] Iteration 0, lr = 0.01
I0409 20:00:53.855419 15472 solver.cpp:218] Iteration 12 (3.16492 iter/s, 3.79157s/12 iters), loss = 5.27845
I0409 20:00:53.855463 15472 solver.cpp:237]     Train net output #0: loss = 5.27845 (* 1 = 5.27845 loss)
I0409 20:00:53.855473 15472 sgd_solver.cpp:105] Iteration 12, lr = 0.00997626
I0409 20:00:58.675765 15472 solver.cpp:218] Iteration 24 (2.48958 iter/s, 4.82009s/12 iters), loss = 5.27819
I0409 20:00:58.675817 15472 solver.cpp:237]     Train net output #0: loss = 5.27819 (* 1 = 5.27819 loss)
I0409 20:00:58.675830 15472 sgd_solver.cpp:105] Iteration 24, lr = 0.00995257
I0409 20:01:03.498812 15472 solver.cpp:218] Iteration 36 (2.48819 iter/s, 4.82279s/12 iters), loss = 5.28011
I0409 20:01:03.498873 15472 solver.cpp:237]     Train net output #0: loss = 5.28011 (* 1 = 5.28011 loss)
I0409 20:01:03.498886 15472 sgd_solver.cpp:105] Iteration 36, lr = 0.00992894
I0409 20:01:08.324512 15472 solver.cpp:218] Iteration 48 (2.48682 iter/s, 4.82543s/12 iters), loss = 5.2848
I0409 20:01:08.324568 15472 solver.cpp:237]     Train net output #0: loss = 5.2848 (* 1 = 5.2848 loss)
I0409 20:01:08.324581 15472 sgd_solver.cpp:105] Iteration 48, lr = 0.00990537
I0409 20:01:13.106995 15472 solver.cpp:218] Iteration 60 (2.50929 iter/s, 4.78222s/12 iters), loss = 5.2762
I0409 20:01:13.107041 15472 solver.cpp:237]     Train net output #0: loss = 5.2762 (* 1 = 5.2762 loss)
I0409 20:01:13.107051 15472 sgd_solver.cpp:105] Iteration 60, lr = 0.00988185
I0409 20:01:17.925585 15472 solver.cpp:218] Iteration 72 (2.49049 iter/s, 4.81833s/12 iters), loss = 5.27892
I0409 20:01:17.927362 15472 solver.cpp:237]     Train net output #0: loss = 5.27892 (* 1 = 5.27892 loss)
I0409 20:01:17.927383 15472 sgd_solver.cpp:105] Iteration 72, lr = 0.00985839
I0409 20:01:22.800411 15472 solver.cpp:218] Iteration 84 (2.46262 iter/s, 4.87285s/12 iters), loss = 5.2822
I0409 20:01:22.800457 15472 solver.cpp:237]     Train net output #0: loss = 5.2822 (* 1 = 5.2822 loss)
I0409 20:01:22.800467 15472 sgd_solver.cpp:105] Iteration 84, lr = 0.00983498
I0409 20:01:27.745271 15472 solver.cpp:218] Iteration 96 (2.4269 iter/s, 4.94459s/12 iters), loss = 5.28335
I0409 20:01:27.745323 15472 solver.cpp:237]     Train net output #0: loss = 5.28335 (* 1 = 5.28335 loss)
I0409 20:01:27.745335 15472 sgd_solver.cpp:105] Iteration 96, lr = 0.00981163
I0409 20:01:29.410316 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:01:29.715047 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_102.caffemodel
I0409 20:01:30.193467 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_102.solverstate
I0409 20:01:30.532035 15472 solver.cpp:330] Iteration 102, Testing net (#0)
I0409 20:01:30.532065 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:01:35.047256 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:01:35.126224 15472 solver.cpp:397]     Test net output #0: accuracy = 0.00551471
I0409 20:01:35.126276 15472 solver.cpp:397]     Test net output #1: loss = 5.27935 (* 1 = 5.27935 loss)
I0409 20:01:36.833804 15472 solver.cpp:218] Iteration 108 (1.32041 iter/s, 9.08809s/12 iters), loss = 5.27825
I0409 20:01:36.833864 15472 solver.cpp:237]     Train net output #0: loss = 5.27825 (* 1 = 5.27825 loss)
I0409 20:01:36.833878 15472 sgd_solver.cpp:105] Iteration 108, lr = 0.00978834
I0409 20:01:41.677078 15472 solver.cpp:218] Iteration 120 (2.4778 iter/s, 4.843s/12 iters), loss = 5.27658
I0409 20:01:41.677130 15472 solver.cpp:237]     Train net output #0: loss = 5.27658 (* 1 = 5.27658 loss)
I0409 20:01:41.677143 15472 sgd_solver.cpp:105] Iteration 120, lr = 0.0097651
I0409 20:01:46.571100 15472 solver.cpp:218] Iteration 132 (2.4521 iter/s, 4.89376s/12 iters), loss = 5.2565
I0409 20:01:46.571139 15472 solver.cpp:237]     Train net output #0: loss = 5.2565 (* 1 = 5.2565 loss)
I0409 20:01:46.571148 15472 sgd_solver.cpp:105] Iteration 132, lr = 0.00974192
I0409 20:01:51.363373 15472 solver.cpp:218] Iteration 144 (2.50416 iter/s, 4.79202s/12 iters), loss = 5.28621
I0409 20:01:51.363507 15472 solver.cpp:237]     Train net output #0: loss = 5.28621 (* 1 = 5.28621 loss)
I0409 20:01:51.363520 15472 sgd_solver.cpp:105] Iteration 144, lr = 0.00971879
I0409 20:01:56.151988 15472 solver.cpp:218] Iteration 156 (2.50612 iter/s, 4.78827s/12 iters), loss = 5.2624
I0409 20:01:56.152042 15472 solver.cpp:237]     Train net output #0: loss = 5.2624 (* 1 = 5.2624 loss)
I0409 20:01:56.152055 15472 sgd_solver.cpp:105] Iteration 156, lr = 0.00969571
I0409 20:02:00.961881 15472 solver.cpp:218] Iteration 168 (2.495 iter/s, 4.80963s/12 iters), loss = 5.27408
I0409 20:02:00.961921 15472 solver.cpp:237]     Train net output #0: loss = 5.27408 (* 1 = 5.27408 loss)
I0409 20:02:00.961930 15472 sgd_solver.cpp:105] Iteration 168, lr = 0.00967269
I0409 20:02:05.933425 15472 solver.cpp:218] Iteration 180 (2.41387 iter/s, 4.97128s/12 iters), loss = 5.271
I0409 20:02:05.933475 15472 solver.cpp:237]     Train net output #0: loss = 5.271 (* 1 = 5.271 loss)
I0409 20:02:05.933485 15472 sgd_solver.cpp:105] Iteration 180, lr = 0.00964973
I0409 20:02:10.851018 15472 solver.cpp:218] Iteration 192 (2.44035 iter/s, 4.91732s/12 iters), loss = 5.2761
I0409 20:02:10.851068 15472 solver.cpp:237]     Train net output #0: loss = 5.2761 (* 1 = 5.2761 loss)
I0409 20:02:10.851079 15472 sgd_solver.cpp:105] Iteration 192, lr = 0.00962682
I0409 20:02:14.770146 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:02:15.467411 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_204.caffemodel
I0409 20:02:15.928716 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_204.solverstate
I0409 20:02:16.260877 15472 solver.cpp:330] Iteration 204, Testing net (#0)
I0409 20:02:16.260906 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:02:20.588850 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:02:20.710355 15472 solver.cpp:397]     Test net output #0: accuracy = 0.00551471
I0409 20:02:20.710399 15472 solver.cpp:397]     Test net output #1: loss = 5.28118 (* 1 = 5.28118 loss)
I0409 20:02:20.792577 15472 solver.cpp:218] Iteration 204 (1.20711 iter/s, 9.94108s/12 iters), loss = 5.27002
I0409 20:02:20.792626 15472 solver.cpp:237]     Train net output #0: loss = 5.27002 (* 1 = 5.27002 loss)
I0409 20:02:20.792637 15472 sgd_solver.cpp:105] Iteration 204, lr = 0.00960396
I0409 20:02:24.924990 15472 solver.cpp:218] Iteration 216 (2.90404 iter/s, 4.13217s/12 iters), loss = 5.279
I0409 20:02:24.927211 15472 solver.cpp:237]     Train net output #0: loss = 5.279 (* 1 = 5.279 loss)
I0409 20:02:24.927225 15472 sgd_solver.cpp:105] Iteration 216, lr = 0.00958116
I0409 20:02:29.735548 15472 solver.cpp:218] Iteration 228 (2.49578 iter/s, 4.80812s/12 iters), loss = 5.26319
I0409 20:02:29.735608 15472 solver.cpp:237]     Train net output #0: loss = 5.26319 (* 1 = 5.26319 loss)
I0409 20:02:29.735620 15472 sgd_solver.cpp:105] Iteration 228, lr = 0.00955841
I0409 20:02:34.581826 15472 solver.cpp:218] Iteration 240 (2.47627 iter/s, 4.84599s/12 iters), loss = 5.28796
I0409 20:02:34.581884 15472 solver.cpp:237]     Train net output #0: loss = 5.28796 (* 1 = 5.28796 loss)
I0409 20:02:34.581897 15472 sgd_solver.cpp:105] Iteration 240, lr = 0.00953572
I0409 20:02:39.429198 15472 solver.cpp:218] Iteration 252 (2.47571 iter/s, 4.84709s/12 iters), loss = 5.26745
I0409 20:02:39.429250 15472 solver.cpp:237]     Train net output #0: loss = 5.26745 (* 1 = 5.26745 loss)
I0409 20:02:39.429261 15472 sgd_solver.cpp:105] Iteration 252, lr = 0.00951308
I0409 20:02:44.355129 15472 solver.cpp:218] Iteration 264 (2.43622 iter/s, 4.92566s/12 iters), loss = 5.27574
I0409 20:02:44.355172 15472 solver.cpp:237]     Train net output #0: loss = 5.27574 (* 1 = 5.27574 loss)
I0409 20:02:44.355181 15472 sgd_solver.cpp:105] Iteration 264, lr = 0.00949049
I0409 20:02:49.284210 15472 solver.cpp:218] Iteration 276 (2.43467 iter/s, 4.92881s/12 iters), loss = 5.28886
I0409 20:02:49.284262 15472 solver.cpp:237]     Train net output #0: loss = 5.28886 (* 1 = 5.28886 loss)
I0409 20:02:49.284274 15472 sgd_solver.cpp:105] Iteration 276, lr = 0.00946796
I0409 20:02:54.129542 15472 solver.cpp:218] Iteration 288 (2.47675 iter/s, 4.84506s/12 iters), loss = 5.27773
I0409 20:02:54.129599 15472 solver.cpp:237]     Train net output #0: loss = 5.27773 (* 1 = 5.27773 loss)
I0409 20:02:54.129612 15472 sgd_solver.cpp:105] Iteration 288, lr = 0.00944548
I0409 20:02:59.006876 15472 solver.cpp:218] Iteration 300 (2.4605 iter/s, 4.87705s/12 iters), loss = 5.28189
I0409 20:02:59.007026 15472 solver.cpp:237]     Train net output #0: loss = 5.28189 (* 1 = 5.28189 loss)
I0409 20:02:59.007041 15472 sgd_solver.cpp:105] Iteration 300, lr = 0.00942305
I0409 20:02:59.967761 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:03:00.987198 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_306.caffemodel
I0409 20:03:01.480088 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_306.solverstate
I0409 20:03:01.806968 15472 solver.cpp:330] Iteration 306, Testing net (#0)
I0409 20:03:01.807000 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:03:06.026876 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:03:06.184985 15472 solver.cpp:397]     Test net output #0: accuracy = 0.00551471
I0409 20:03:06.185034 15472 solver.cpp:397]     Test net output #1: loss = 5.2827 (* 1 = 5.2827 loss)
I0409 20:03:08.054459 15472 solver.cpp:218] Iteration 312 (1.3264 iter/s, 9.04703s/12 iters), loss = 5.27794
I0409 20:03:08.054507 15472 solver.cpp:237]     Train net output #0: loss = 5.27794 (* 1 = 5.27794 loss)
I0409 20:03:08.054517 15472 sgd_solver.cpp:105] Iteration 312, lr = 0.00940068
I0409 20:03:12.859700 15472 solver.cpp:218] Iteration 324 (2.49742 iter/s, 4.80496s/12 iters), loss = 5.25025
I0409 20:03:12.859766 15472 solver.cpp:237]     Train net output #0: loss = 5.25025 (* 1 = 5.25025 loss)
I0409 20:03:12.859783 15472 sgd_solver.cpp:105] Iteration 324, lr = 0.00937836
I0409 20:03:17.685215 15472 solver.cpp:218] Iteration 336 (2.48693 iter/s, 4.82523s/12 iters), loss = 5.26433
I0409 20:03:17.685266 15472 solver.cpp:237]     Train net output #0: loss = 5.26433 (* 1 = 5.26433 loss)
I0409 20:03:17.685276 15472 sgd_solver.cpp:105] Iteration 336, lr = 0.0093561
I0409 20:03:22.455405 15472 solver.cpp:218] Iteration 348 (2.51577 iter/s, 4.76992s/12 iters), loss = 5.26783
I0409 20:03:22.455446 15472 solver.cpp:237]     Train net output #0: loss = 5.26783 (* 1 = 5.26783 loss)
I0409 20:03:22.455456 15472 sgd_solver.cpp:105] Iteration 348, lr = 0.00933388
I0409 20:03:27.252267 15472 solver.cpp:218] Iteration 360 (2.50177 iter/s, 4.7966s/12 iters), loss = 5.29048
I0409 20:03:27.252312 15472 solver.cpp:237]     Train net output #0: loss = 5.29048 (* 1 = 5.29048 loss)
I0409 20:03:27.252322 15472 sgd_solver.cpp:105] Iteration 360, lr = 0.00931172
I0409 20:03:32.057418 15472 solver.cpp:218] Iteration 372 (2.49746 iter/s, 4.80488s/12 iters), loss = 5.2719
I0409 20:03:32.057588 15472 solver.cpp:237]     Train net output #0: loss = 5.2719 (* 1 = 5.2719 loss)
I0409 20:03:32.057602 15472 sgd_solver.cpp:105] Iteration 372, lr = 0.00928961
I0409 20:03:37.017458 15472 solver.cpp:218] Iteration 384 (2.41953 iter/s, 4.95964s/12 iters), loss = 5.28033
I0409 20:03:37.017515 15472 solver.cpp:237]     Train net output #0: loss = 5.28033 (* 1 = 5.28033 loss)
I0409 20:03:37.017529 15472 sgd_solver.cpp:105] Iteration 384, lr = 0.00926756
I0409 20:03:42.177132 15472 solver.cpp:218] Iteration 396 (2.32586 iter/s, 5.15938s/12 iters), loss = 5.27512
I0409 20:03:42.177188 15472 solver.cpp:237]     Train net output #0: loss = 5.27512 (* 1 = 5.27512 loss)
I0409 20:03:42.177201 15472 sgd_solver.cpp:105] Iteration 396, lr = 0.00924556
I0409 20:03:45.168558 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:03:46.517292 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_408.caffemodel
I0409 20:03:46.995081 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_408.solverstate
I0409 20:03:47.320828 15472 solver.cpp:330] Iteration 408, Testing net (#0)
I0409 20:03:47.320847 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:03:51.606953 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:03:51.812573 15472 solver.cpp:397]     Test net output #0: accuracy = 0.00551471
I0409 20:03:51.812623 15472 solver.cpp:397]     Test net output #1: loss = 5.28465 (* 1 = 5.28465 loss)
I0409 20:03:51.895601 15472 solver.cpp:218] Iteration 408 (1.23482 iter/s, 9.71798s/12 iters), loss = 5.27689
I0409 20:03:51.895650 15472 solver.cpp:237]     Train net output #0: loss = 5.27689 (* 1 = 5.27689 loss)
I0409 20:03:51.895661 15472 sgd_solver.cpp:105] Iteration 408, lr = 0.00922361
I0409 20:03:55.897359 15472 solver.cpp:218] Iteration 420 (2.99886 iter/s, 4.00152s/12 iters), loss = 5.27286
I0409 20:03:55.897415 15472 solver.cpp:237]     Train net output #0: loss = 5.27286 (* 1 = 5.27286 loss)
I0409 20:03:55.897428 15472 sgd_solver.cpp:105] Iteration 420, lr = 0.00920171
I0409 20:04:00.729740 15472 solver.cpp:218] Iteration 432 (2.48339 iter/s, 4.8321s/12 iters), loss = 5.26895
I0409 20:04:00.729794 15472 solver.cpp:237]     Train net output #0: loss = 5.26895 (* 1 = 5.26895 loss)
I0409 20:04:00.729806 15472 sgd_solver.cpp:105] Iteration 432, lr = 0.00917986
I0409 20:04:05.512912 15472 solver.cpp:218] Iteration 444 (2.50894 iter/s, 4.7829s/12 iters), loss = 5.28673
I0409 20:04:05.513018 15472 solver.cpp:237]     Train net output #0: loss = 5.28673 (* 1 = 5.28673 loss)
I0409 20:04:05.513028 15472 sgd_solver.cpp:105] Iteration 444, lr = 0.00915807
I0409 20:04:10.330593 15472 solver.cpp:218] Iteration 456 (2.491 iter/s, 4.81735s/12 iters), loss = 5.28073
I0409 20:04:10.330651 15472 solver.cpp:237]     Train net output #0: loss = 5.28073 (* 1 = 5.28073 loss)
I0409 20:04:10.330662 15472 sgd_solver.cpp:105] Iteration 456, lr = 0.00913632
I0409 20:04:15.158551 15472 solver.cpp:218] Iteration 468 (2.48567 iter/s, 4.82768s/12 iters), loss = 5.28426
I0409 20:04:15.158601 15472 solver.cpp:237]     Train net output #0: loss = 5.28426 (* 1 = 5.28426 loss)
I0409 20:04:15.158612 15472 sgd_solver.cpp:105] Iteration 468, lr = 0.00911463
I0409 20:04:19.960803 15472 solver.cpp:218] Iteration 480 (2.49897 iter/s, 4.80198s/12 iters), loss = 5.26673
I0409 20:04:19.960847 15472 solver.cpp:237]     Train net output #0: loss = 5.26673 (* 1 = 5.26673 loss)
I0409 20:04:19.960858 15472 sgd_solver.cpp:105] Iteration 480, lr = 0.00909299
I0409 20:04:24.908260 15472 solver.cpp:218] Iteration 492 (2.42562 iter/s, 4.94718s/12 iters), loss = 5.28846
I0409 20:04:24.908313 15472 solver.cpp:237]     Train net output #0: loss = 5.28846 (* 1 = 5.28846 loss)
I0409 20:04:24.908326 15472 sgd_solver.cpp:105] Iteration 492, lr = 0.0090714
I0409 20:04:29.827833 15472 solver.cpp:218] Iteration 504 (2.43937 iter/s, 4.9193s/12 iters), loss = 5.26689
I0409 20:04:29.827877 15472 solver.cpp:237]     Train net output #0: loss = 5.26689 (* 1 = 5.26689 loss)
I0409 20:04:29.827888 15472 sgd_solver.cpp:105] Iteration 504, lr = 0.00904986
I0409 20:04:30.077127 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:04:31.784521 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_510.caffemodel
I0409 20:04:32.240392 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_510.solverstate
I0409 20:04:32.576902 15472 solver.cpp:330] Iteration 510, Testing net (#0)
I0409 20:04:32.576931 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:04:36.673548 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:04:36.910974 15472 solver.cpp:397]     Test net output #0: accuracy = 0.00612745
I0409 20:04:36.911010 15472 solver.cpp:397]     Test net output #1: loss = 5.28464 (* 1 = 5.28464 loss)
I0409 20:04:38.678064 15472 solver.cpp:218] Iteration 516 (1.35597 iter/s, 8.84978s/12 iters), loss = 5.28105
I0409 20:04:38.678110 15472 solver.cpp:237]     Train net output #0: loss = 5.28105 (* 1 = 5.28105 loss)
I0409 20:04:38.678119 15472 sgd_solver.cpp:105] Iteration 516, lr = 0.00902838
I0409 20:04:43.627677 15472 solver.cpp:218] Iteration 528 (2.42457 iter/s, 4.94933s/12 iters), loss = 5.26821
I0409 20:04:43.627725 15472 solver.cpp:237]     Train net output #0: loss = 5.26821 (* 1 = 5.26821 loss)
I0409 20:04:43.627733 15472 sgd_solver.cpp:105] Iteration 528, lr = 0.00900694
I0409 20:04:48.862504 15472 solver.cpp:218] Iteration 540 (2.29247 iter/s, 5.23453s/12 iters), loss = 5.2715
I0409 20:04:48.862545 15472 solver.cpp:237]     Train net output #0: loss = 5.2715 (* 1 = 5.2715 loss)
I0409 20:04:48.862553 15472 sgd_solver.cpp:105] Iteration 540, lr = 0.00898556
I0409 20:04:53.702951 15472 solver.cpp:218] Iteration 552 (2.47925 iter/s, 4.84018s/12 iters), loss = 5.27286
I0409 20:04:53.702998 15472 solver.cpp:237]     Train net output #0: loss = 5.27286 (* 1 = 5.27286 loss)
I0409 20:04:53.703007 15472 sgd_solver.cpp:105] Iteration 552, lr = 0.00896423
I0409 20:04:58.709131 15472 solver.cpp:218] Iteration 564 (2.39717 iter/s, 5.0059s/12 iters), loss = 5.25811
I0409 20:04:58.709177 15472 solver.cpp:237]     Train net output #0: loss = 5.25811 (* 1 = 5.25811 loss)
I0409 20:04:58.709185 15472 sgd_solver.cpp:105] Iteration 564, lr = 0.00894294
I0409 20:05:03.559561 15472 solver.cpp:218] Iteration 576 (2.47415 iter/s, 4.85016s/12 iters), loss = 5.27517
I0409 20:05:03.559602 15472 solver.cpp:237]     Train net output #0: loss = 5.27517 (* 1 = 5.27517 loss)
I0409 20:05:03.559610 15472 sgd_solver.cpp:105] Iteration 576, lr = 0.00892171
I0409 20:05:08.376350 15472 solver.cpp:218] Iteration 588 (2.49143 iter/s, 4.81652s/12 iters), loss = 5.26295
I0409 20:05:08.376482 15472 solver.cpp:237]     Train net output #0: loss = 5.26295 (* 1 = 5.26295 loss)
I0409 20:05:08.376497 15472 sgd_solver.cpp:105] Iteration 588, lr = 0.00890053
I0409 20:05:13.194715 15472 solver.cpp:218] Iteration 600 (2.49065 iter/s, 4.81802s/12 iters), loss = 5.26203
I0409 20:05:13.194763 15472 solver.cpp:237]     Train net output #0: loss = 5.26203 (* 1 = 5.26203 loss)
I0409 20:05:13.194775 15472 sgd_solver.cpp:105] Iteration 600, lr = 0.0088794
I0409 20:05:15.522891 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:05:17.589841 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_612.caffemodel
I0409 20:05:20.741294 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_612.solverstate
I0409 20:05:22.864825 15472 solver.cpp:330] Iteration 612, Testing net (#0)
I0409 20:05:22.864850 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:05:27.164515 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:05:27.451040 15472 solver.cpp:397]     Test net output #0: accuracy = 0.00551471
I0409 20:05:27.451092 15472 solver.cpp:397]     Test net output #1: loss = 5.28511 (* 1 = 5.28511 loss)
I0409 20:05:27.534157 15472 solver.cpp:218] Iteration 612 (0.836893 iter/s, 14.3388s/12 iters), loss = 5.27181
I0409 20:05:27.534209 15472 solver.cpp:237]     Train net output #0: loss = 5.27181 (* 1 = 5.27181 loss)
I0409 20:05:27.534220 15472 sgd_solver.cpp:105] Iteration 612, lr = 0.00885831
I0409 20:05:31.887256 15472 solver.cpp:218] Iteration 624 (2.75682 iter/s, 4.35285s/12 iters), loss = 5.28704
I0409 20:05:31.887300 15472 solver.cpp:237]     Train net output #0: loss = 5.28704 (* 1 = 5.28704 loss)
I0409 20:05:31.887308 15472 sgd_solver.cpp:105] Iteration 624, lr = 0.00883728
I0409 20:05:36.676990 15472 solver.cpp:218] Iteration 636 (2.5055 iter/s, 4.78947s/12 iters), loss = 5.28673
I0409 20:05:36.677029 15472 solver.cpp:237]     Train net output #0: loss = 5.28673 (* 1 = 5.28673 loss)
I0409 20:05:36.677038 15472 sgd_solver.cpp:105] Iteration 636, lr = 0.0088163
I0409 20:05:41.516324 15472 solver.cpp:218] Iteration 648 (2.47982 iter/s, 4.83907s/12 iters), loss = 5.27203
I0409 20:05:41.516489 15472 solver.cpp:237]     Train net output #0: loss = 5.27203 (* 1 = 5.27203 loss)
I0409 20:05:41.516505 15472 sgd_solver.cpp:105] Iteration 648, lr = 0.00879537
I0409 20:05:46.265080 15472 solver.cpp:218] Iteration 660 (2.52718 iter/s, 4.74837s/12 iters), loss = 5.2663
I0409 20:05:46.265133 15472 solver.cpp:237]     Train net output #0: loss = 5.2663 (* 1 = 5.2663 loss)
I0409 20:05:46.265146 15472 sgd_solver.cpp:105] Iteration 660, lr = 0.00877449
I0409 20:05:51.085824 15472 solver.cpp:218] Iteration 672 (2.48939 iter/s, 4.82047s/12 iters), loss = 5.27488
I0409 20:05:51.085865 15472 solver.cpp:237]     Train net output #0: loss = 5.27488 (* 1 = 5.27488 loss)
I0409 20:05:51.085873 15472 sgd_solver.cpp:105] Iteration 672, lr = 0.00875366
I0409 20:05:55.477345 15472 blocking_queue.cpp:49] Waiting for data
I0409 20:05:55.918025 15472 solver.cpp:218] Iteration 684 (2.48348 iter/s, 4.83193s/12 iters), loss = 5.27603
I0409 20:05:55.918084 15472 solver.cpp:237]     Train net output #0: loss = 5.27603 (* 1 = 5.27603 loss)
I0409 20:05:55.918095 15472 sgd_solver.cpp:105] Iteration 684, lr = 0.00873287
I0409 20:06:00.736968 15472 solver.cpp:218] Iteration 696 (2.49032 iter/s, 4.81866s/12 iters), loss = 5.26663
I0409 20:06:00.737023 15472 solver.cpp:237]     Train net output #0: loss = 5.26663 (* 1 = 5.26663 loss)
I0409 20:06:00.737033 15472 sgd_solver.cpp:105] Iteration 696, lr = 0.00871214
I0409 20:06:05.234395 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:06:05.604601 15472 solver.cpp:218] Iteration 708 (2.46541 iter/s, 4.86735s/12 iters), loss = 5.25814
I0409 20:06:05.604653 15472 solver.cpp:237]     Train net output #0: loss = 5.25814 (* 1 = 5.25814 loss)
I0409 20:06:05.604665 15472 sgd_solver.cpp:105] Iteration 708, lr = 0.00869145
I0409 20:06:07.564105 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_714.caffemodel
I0409 20:06:08.048394 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_714.solverstate
I0409 20:06:08.382975 15472 solver.cpp:330] Iteration 714, Testing net (#0)
I0409 20:06:08.383008 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:06:12.560356 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:06:12.884127 15472 solver.cpp:397]     Test net output #0: accuracy = 0.00551471
I0409 20:06:12.884181 15472 solver.cpp:397]     Test net output #1: loss = 5.28617 (* 1 = 5.28617 loss)
I0409 20:06:14.747648 15472 solver.cpp:218] Iteration 720 (1.31254 iter/s, 9.14259s/12 iters), loss = 5.27744
I0409 20:06:14.747686 15472 solver.cpp:237]     Train net output #0: loss = 5.27744 (* 1 = 5.27744 loss)
I0409 20:06:14.747695 15472 sgd_solver.cpp:105] Iteration 720, lr = 0.00867082
I0409 20:06:19.544118 15472 solver.cpp:218] Iteration 732 (2.50198 iter/s, 4.79621s/12 iters), loss = 5.27989
I0409 20:06:19.544167 15472 solver.cpp:237]     Train net output #0: loss = 5.27989 (* 1 = 5.27989 loss)
I0409 20:06:19.544181 15472 sgd_solver.cpp:105] Iteration 732, lr = 0.00865023
I0409 20:06:24.369392 15472 solver.cpp:218] Iteration 744 (2.48705 iter/s, 4.825s/12 iters), loss = 5.27434
I0409 20:06:24.369441 15472 solver.cpp:237]     Train net output #0: loss = 5.27434 (* 1 = 5.27434 loss)
I0409 20:06:24.369451 15472 sgd_solver.cpp:105] Iteration 744, lr = 0.0086297
I0409 20:06:29.305442 15472 solver.cpp:218] Iteration 756 (2.43123 iter/s, 4.93577s/12 iters), loss = 5.2704
I0409 20:06:29.305487 15472 solver.cpp:237]     Train net output #0: loss = 5.2704 (* 1 = 5.2704 loss)
I0409 20:06:29.305497 15472 sgd_solver.cpp:105] Iteration 756, lr = 0.00860921
I0409 20:06:34.111500 15472 solver.cpp:218] Iteration 768 (2.49699 iter/s, 4.80579s/12 iters), loss = 5.27697
I0409 20:06:34.111553 15472 solver.cpp:237]     Train net output #0: loss = 5.27697 (* 1 = 5.27697 loss)
I0409 20:06:34.111567 15472 sgd_solver.cpp:105] Iteration 768, lr = 0.00858877
I0409 20:06:38.978013 15472 solver.cpp:218] Iteration 780 (2.46597 iter/s, 4.86623s/12 iters), loss = 5.26618
I0409 20:06:38.978062 15472 solver.cpp:237]     Train net output #0: loss = 5.26618 (* 1 = 5.26618 loss)
I0409 20:06:38.978076 15472 sgd_solver.cpp:105] Iteration 780, lr = 0.00856838
I0409 20:06:43.836125 15472 solver.cpp:218] Iteration 792 (2.47023 iter/s, 4.85784s/12 iters), loss = 5.27019
I0409 20:06:43.836239 15472 solver.cpp:237]     Train net output #0: loss = 5.27019 (* 1 = 5.27019 loss)
I0409 20:06:43.836251 15472 sgd_solver.cpp:105] Iteration 792, lr = 0.00854803
I0409 20:06:48.718921 15472 solver.cpp:218] Iteration 804 (2.45778 iter/s, 4.88245s/12 iters), loss = 5.28622
I0409 20:06:48.718962 15472 solver.cpp:237]     Train net output #0: loss = 5.28622 (* 1 = 5.28622 loss)
I0409 20:06:48.718971 15472 sgd_solver.cpp:105] Iteration 804, lr = 0.00852774
I0409 20:06:50.428591 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:06:53.149550 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_816.caffemodel
I0409 20:06:53.618355 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_816.solverstate
I0409 20:06:53.950116 15472 solver.cpp:330] Iteration 816, Testing net (#0)
I0409 20:06:53.950145 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:06:58.128584 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:06:58.481739 15472 solver.cpp:397]     Test net output #0: accuracy = 0.00551471
I0409 20:06:58.481786 15472 solver.cpp:397]     Test net output #1: loss = 5.28617 (* 1 = 5.28617 loss)
I0409 20:06:58.564797 15472 solver.cpp:218] Iteration 816 (1.21885 iter/s, 9.84539s/12 iters), loss = 5.27863
I0409 20:06:58.564848 15472 solver.cpp:237]     Train net output #0: loss = 5.27863 (* 1 = 5.27863 loss)
I0409 20:06:58.564859 15472 sgd_solver.cpp:105] Iteration 816, lr = 0.00850749
I0409 20:07:02.786047 15472 solver.cpp:218] Iteration 828 (2.84293 iter/s, 4.221s/12 iters), loss = 5.28161
I0409 20:07:02.786096 15472 solver.cpp:237]     Train net output #0: loss = 5.28161 (* 1 = 5.28161 loss)
I0409 20:07:02.786108 15472 sgd_solver.cpp:105] Iteration 828, lr = 0.00848729
I0409 20:07:07.688400 15472 solver.cpp:218] Iteration 840 (2.44794 iter/s, 4.90208s/12 iters), loss = 5.22935
I0409 20:07:07.688446 15472 solver.cpp:237]     Train net output #0: loss = 5.22935 (* 1 = 5.22935 loss)
I0409 20:07:07.688454 15472 sgd_solver.cpp:105] Iteration 840, lr = 0.00846714
I0409 20:07:12.515897 15472 solver.cpp:218] Iteration 852 (2.4859 iter/s, 4.82723s/12 iters), loss = 5.29934
I0409 20:07:12.515940 15472 solver.cpp:237]     Train net output #0: loss = 5.29934 (* 1 = 5.29934 loss)
I0409 20:07:12.515950 15472 sgd_solver.cpp:105] Iteration 852, lr = 0.00844704
I0409 20:07:17.421661 15472 solver.cpp:218] Iteration 864 (2.44624 iter/s, 4.90549s/12 iters), loss = 5.25965
I0409 20:07:17.423233 15472 solver.cpp:237]     Train net output #0: loss = 5.25965 (* 1 = 5.25965 loss)
I0409 20:07:17.423250 15472 sgd_solver.cpp:105] Iteration 864, lr = 0.00842698
I0409 20:07:22.309226 15472 solver.cpp:218] Iteration 876 (2.45611 iter/s, 4.88577s/12 iters), loss = 5.27017
I0409 20:07:22.309288 15472 solver.cpp:237]     Train net output #0: loss = 5.27017 (* 1 = 5.27017 loss)
I0409 20:07:22.309300 15472 sgd_solver.cpp:105] Iteration 876, lr = 0.00840698
I0409 20:07:27.235414 15472 solver.cpp:218] Iteration 888 (2.4361 iter/s, 4.9259s/12 iters), loss = 5.26376
I0409 20:07:27.235462 15472 solver.cpp:237]     Train net output #0: loss = 5.26376 (* 1 = 5.26376 loss)
I0409 20:07:27.235476 15472 sgd_solver.cpp:105] Iteration 888, lr = 0.00838702
I0409 20:07:32.150944 15472 solver.cpp:218] Iteration 900 (2.44138 iter/s, 4.91525s/12 iters), loss = 5.27404
I0409 20:07:32.151001 15472 solver.cpp:237]     Train net output #0: loss = 5.27404 (* 1 = 5.27404 loss)
I0409 20:07:32.151013 15472 sgd_solver.cpp:105] Iteration 900, lr = 0.0083671
I0409 20:07:35.959172 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:07:37.061993 15472 solver.cpp:218] Iteration 912 (2.44362 iter/s, 4.91075s/12 iters), loss = 5.26004
I0409 20:07:37.062052 15472 solver.cpp:237]     Train net output #0: loss = 5.26004 (* 1 = 5.26004 loss)
I0409 20:07:37.062063 15472 sgd_solver.cpp:105] Iteration 912, lr = 0.00834724
I0409 20:07:39.026834 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_918.caffemodel
I0409 20:07:39.485395 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_918.solverstate
I0409 20:07:39.801632 15472 solver.cpp:330] Iteration 918, Testing net (#0)
I0409 20:07:39.801664 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:07:43.774286 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:07:44.172905 15472 solver.cpp:397]     Test net output #0: accuracy = 0.00551471
I0409 20:07:44.172950 15472 solver.cpp:397]     Test net output #1: loss = 5.2863 (* 1 = 5.2863 loss)
I0409 20:07:46.036546 15472 solver.cpp:218] Iteration 924 (1.33718 iter/s, 8.97409s/12 iters), loss = 5.2831
I0409 20:07:46.036592 15472 solver.cpp:237]     Train net output #0: loss = 5.2831 (* 1 = 5.2831 loss)
I0409 20:07:46.036600 15472 sgd_solver.cpp:105] Iteration 924, lr = 0.00832742
I0409 20:07:50.874395 15472 solver.cpp:218] Iteration 936 (2.48058 iter/s, 4.83757s/12 iters), loss = 5.262
I0409 20:07:50.874998 15472 solver.cpp:237]     Train net output #0: loss = 5.262 (* 1 = 5.262 loss)
I0409 20:07:50.875015 15472 sgd_solver.cpp:105] Iteration 936, lr = 0.00830765
I0409 20:07:55.710525 15472 solver.cpp:218] Iteration 948 (2.48175 iter/s, 4.83531s/12 iters), loss = 5.28514
I0409 20:07:55.710568 15472 solver.cpp:237]     Train net output #0: loss = 5.28514 (* 1 = 5.28514 loss)
I0409 20:07:55.710577 15472 sgd_solver.cpp:105] Iteration 948, lr = 0.00828793
I0409 20:08:00.608448 15472 solver.cpp:218] Iteration 960 (2.45016 iter/s, 4.89765s/12 iters), loss = 5.25902
I0409 20:08:00.608503 15472 solver.cpp:237]     Train net output #0: loss = 5.25902 (* 1 = 5.25902 loss)
I0409 20:08:00.608517 15472 sgd_solver.cpp:105] Iteration 960, lr = 0.00826825
I0409 20:08:05.463820 15472 solver.cpp:218] Iteration 972 (2.47163 iter/s, 4.8551s/12 iters), loss = 5.27091
I0409 20:08:05.463865 15472 solver.cpp:237]     Train net output #0: loss = 5.27091 (* 1 = 5.27091 loss)
I0409 20:08:05.463873 15472 sgd_solver.cpp:105] Iteration 972, lr = 0.00824862
I0409 20:08:10.267259 15472 solver.cpp:218] Iteration 984 (2.49835 iter/s, 4.80317s/12 iters), loss = 5.29562
I0409 20:08:10.267313 15472 solver.cpp:237]     Train net output #0: loss = 5.29562 (* 1 = 5.29562 loss)
I0409 20:08:10.267326 15472 sgd_solver.cpp:105] Iteration 984, lr = 0.00822903
I0409 20:08:15.102007 15472 solver.cpp:218] Iteration 996 (2.48219 iter/s, 4.83443s/12 iters), loss = 5.277
I0409 20:08:15.102049 15472 solver.cpp:237]     Train net output #0: loss = 5.277 (* 1 = 5.277 loss)
I0409 20:08:15.102058 15472 sgd_solver.cpp:105] Iteration 996, lr = 0.0082095
I0409 20:08:19.926035 15472 solver.cpp:218] Iteration 1008 (2.48769 iter/s, 4.82376s/12 iters), loss = 5.29112
I0409 20:08:19.926095 15472 solver.cpp:237]     Train net output #0: loss = 5.29112 (* 1 = 5.29112 loss)
I0409 20:08:19.926106 15472 sgd_solver.cpp:105] Iteration 1008, lr = 0.00819001
I0409 20:08:20.914974 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:08:24.334477 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_1020.caffemodel
I0409 20:08:24.805406 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_1020.solverstate
I0409 20:08:25.136420 15472 solver.cpp:330] Iteration 1020, Testing net (#0)
I0409 20:08:25.136449 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:08:29.202533 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:08:29.632395 15472 solver.cpp:397]     Test net output #0: accuracy = 0.00551471
I0409 20:08:29.632445 15472 solver.cpp:397]     Test net output #1: loss = 5.28613 (* 1 = 5.28613 loss)
I0409 20:08:29.719702 15472 solver.cpp:218] Iteration 1020 (1.22534 iter/s, 9.79317s/12 iters), loss = 5.28671
I0409 20:08:29.719753 15472 solver.cpp:237]     Train net output #0: loss = 5.28671 (* 1 = 5.28671 loss)
I0409 20:08:29.719764 15472 sgd_solver.cpp:105] Iteration 1020, lr = 0.00817056
I0409 20:08:33.796355 15472 solver.cpp:218] Iteration 1032 (2.94376 iter/s, 4.07641s/12 iters), loss = 5.25184
I0409 20:08:33.796398 15472 solver.cpp:237]     Train net output #0: loss = 5.25184 (* 1 = 5.25184 loss)
I0409 20:08:33.796407 15472 sgd_solver.cpp:105] Iteration 1032, lr = 0.00815116
I0409 20:08:38.710155 15472 solver.cpp:218] Iteration 1044 (2.44224 iter/s, 4.91352s/12 iters), loss = 5.25796
I0409 20:08:38.710220 15472 solver.cpp:237]     Train net output #0: loss = 5.25796 (* 1 = 5.25796 loss)
I0409 20:08:38.710232 15472 sgd_solver.cpp:105] Iteration 1044, lr = 0.00813181
I0409 20:08:43.555891 15472 solver.cpp:218] Iteration 1056 (2.47655 iter/s, 4.84545s/12 iters), loss = 5.26152
I0409 20:08:43.555948 15472 solver.cpp:237]     Train net output #0: loss = 5.26152 (* 1 = 5.26152 loss)
I0409 20:08:43.555959 15472 sgd_solver.cpp:105] Iteration 1056, lr = 0.0081125
I0409 20:08:48.393524 15472 solver.cpp:218] Iteration 1068 (2.4807 iter/s, 4.83735s/12 iters), loss = 5.28803
I0409 20:08:48.393569 15472 solver.cpp:237]     Train net output #0: loss = 5.28803 (* 1 = 5.28803 loss)
I0409 20:08:48.393579 15472 sgd_solver.cpp:105] Iteration 1068, lr = 0.00809324
I0409 20:08:53.366647 15472 solver.cpp:218] Iteration 1080 (2.41311 iter/s, 4.97284s/12 iters), loss = 5.26863
I0409 20:08:53.366768 15472 solver.cpp:237]     Train net output #0: loss = 5.26863 (* 1 = 5.26863 loss)
I0409 20:08:53.366783 15472 sgd_solver.cpp:105] Iteration 1080, lr = 0.00807403
I0409 20:08:58.274576 15472 solver.cpp:218] Iteration 1092 (2.4452 iter/s, 4.90758s/12 iters), loss = 5.28114
I0409 20:08:58.274622 15472 solver.cpp:237]     Train net output #0: loss = 5.28114 (* 1 = 5.28114 loss)
I0409 20:08:58.274634 15472 sgd_solver.cpp:105] Iteration 1092, lr = 0.00805486
I0409 20:09:03.401372 15472 solver.cpp:218] Iteration 1104 (2.34077 iter/s, 5.12651s/12 iters), loss = 5.27222
I0409 20:09:03.401422 15472 solver.cpp:237]     Train net output #0: loss = 5.27222 (* 1 = 5.27222 loss)
I0409 20:09:03.401432 15472 sgd_solver.cpp:105] Iteration 1104, lr = 0.00803573
I0409 20:09:06.393227 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:09:08.177201 15472 solver.cpp:218] Iteration 1116 (2.5128 iter/s, 4.77555s/12 iters), loss = 5.27081
I0409 20:09:08.177256 15472 solver.cpp:237]     Train net output #0: loss = 5.27081 (* 1 = 5.27081 loss)
I0409 20:09:08.177269 15472 sgd_solver.cpp:105] Iteration 1116, lr = 0.00801666
I0409 20:09:10.154796 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_1122.caffemodel
I0409 20:09:11.534767 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_1122.solverstate
I0409 20:09:11.877540 15472 solver.cpp:330] Iteration 1122, Testing net (#0)
I0409 20:09:11.877571 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:09:16.117950 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:09:16.591939 15472 solver.cpp:397]     Test net output #0: accuracy = 0.00551471
I0409 20:09:16.591976 15472 solver.cpp:397]     Test net output #1: loss = 5.28633 (* 1 = 5.28633 loss)
I0409 20:09:18.403404 15472 solver.cpp:218] Iteration 1128 (1.17352 iter/s, 10.2257s/12 iters), loss = 5.27458
I0409 20:09:18.403460 15472 solver.cpp:237]     Train net output #0: loss = 5.27458 (* 1 = 5.27458 loss)
I0409 20:09:18.403470 15472 sgd_solver.cpp:105] Iteration 1128, lr = 0.00799762
I0409 20:09:23.304816 15472 solver.cpp:218] Iteration 1140 (2.44842 iter/s, 4.90113s/12 iters), loss = 5.2683
I0409 20:09:23.304870 15472 solver.cpp:237]     Train net output #0: loss = 5.2683 (* 1 = 5.2683 loss)
I0409 20:09:23.304883 15472 sgd_solver.cpp:105] Iteration 1140, lr = 0.00797863
I0409 20:09:28.228016 15472 solver.cpp:218] Iteration 1152 (2.43758 iter/s, 4.92291s/12 iters), loss = 5.27914
I0409 20:09:28.228142 15472 solver.cpp:237]     Train net output #0: loss = 5.27914 (* 1 = 5.27914 loss)
I0409 20:09:28.228154 15472 sgd_solver.cpp:105] Iteration 1152, lr = 0.00795969
I0409 20:09:33.220006 15472 solver.cpp:218] Iteration 1164 (2.40402 iter/s, 4.99163s/12 iters), loss = 5.27012
I0409 20:09:33.220048 15472 solver.cpp:237]     Train net output #0: loss = 5.27012 (* 1 = 5.27012 loss)
I0409 20:09:33.220057 15472 sgd_solver.cpp:105] Iteration 1164, lr = 0.00794079
I0409 20:09:38.141108 15472 solver.cpp:218] Iteration 1176 (2.43862 iter/s, 4.92082s/12 iters), loss = 5.2878
I0409 20:09:38.141162 15472 solver.cpp:237]     Train net output #0: loss = 5.2878 (* 1 = 5.2878 loss)
I0409 20:09:38.141173 15472 sgd_solver.cpp:105] Iteration 1176, lr = 0.00792194
I0409 20:09:43.040155 15472 solver.cpp:218] Iteration 1188 (2.4496 iter/s, 4.89876s/12 iters), loss = 5.2691
I0409 20:09:43.040215 15472 solver.cpp:237]     Train net output #0: loss = 5.2691 (* 1 = 5.2691 loss)
I0409 20:09:43.040228 15472 sgd_solver.cpp:105] Iteration 1188, lr = 0.00790313
I0409 20:09:47.937875 15472 solver.cpp:218] Iteration 1200 (2.45026 iter/s, 4.89743s/12 iters), loss = 5.2896
I0409 20:09:47.937920 15472 solver.cpp:237]     Train net output #0: loss = 5.2896 (* 1 = 5.2896 loss)
I0409 20:09:47.937930 15472 sgd_solver.cpp:105] Iteration 1200, lr = 0.00788437
I0409 20:09:52.857398 15472 solver.cpp:218] Iteration 1212 (2.4394 iter/s, 4.91925s/12 iters), loss = 5.26699
I0409 20:09:52.857456 15472 solver.cpp:237]     Train net output #0: loss = 5.26699 (* 1 = 5.26699 loss)
I0409 20:09:52.857470 15472 sgd_solver.cpp:105] Iteration 1212, lr = 0.00786565
I0409 20:09:53.145381 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:09:57.336344 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_1224.caffemodel
I0409 20:09:57.811426 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_1224.solverstate
I0409 20:09:58.142618 15472 solver.cpp:330] Iteration 1224, Testing net (#0)
I0409 20:09:58.142647 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:10:02.075479 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:10:02.601137 15472 solver.cpp:397]     Test net output #0: accuracy = 0.00612745
I0409 20:10:02.601184 15472 solver.cpp:397]     Test net output #1: loss = 5.28629 (* 1 = 5.28629 loss)
I0409 20:10:02.684150 15472 solver.cpp:218] Iteration 1224 (1.22122 iter/s, 9.82625s/12 iters), loss = 5.28052
I0409 20:10:02.684199 15472 solver.cpp:237]     Train net output #0: loss = 5.28052 (* 1 = 5.28052 loss)
I0409 20:10:02.684212 15472 sgd_solver.cpp:105] Iteration 1224, lr = 0.00784697
I0409 20:10:06.833372 15472 solver.cpp:218] Iteration 1236 (2.89228 iter/s, 4.14898s/12 iters), loss = 5.27052
I0409 20:10:06.833423 15472 solver.cpp:237]     Train net output #0: loss = 5.27052 (* 1 = 5.27052 loss)
I0409 20:10:06.833436 15472 sgd_solver.cpp:105] Iteration 1236, lr = 0.00782834
I0409 20:10:11.671212 15472 solver.cpp:218] Iteration 1248 (2.48059 iter/s, 4.83756s/12 iters), loss = 5.28049
I0409 20:10:11.671268 15472 solver.cpp:237]     Train net output #0: loss = 5.28049 (* 1 = 5.28049 loss)
I0409 20:10:11.671281 15472 sgd_solver.cpp:105] Iteration 1248, lr = 0.00780976
I0409 20:10:16.524601 15472 solver.cpp:218] Iteration 1260 (2.47265 iter/s, 4.8531s/12 iters), loss = 5.27019
I0409 20:10:16.524662 15472 solver.cpp:237]     Train net output #0: loss = 5.27019 (* 1 = 5.27019 loss)
I0409 20:10:16.524674 15472 sgd_solver.cpp:105] Iteration 1260, lr = 0.00779122
I0409 20:10:21.378687 15472 solver.cpp:218] Iteration 1272 (2.47229 iter/s, 4.8538s/12 iters), loss = 5.24622
I0409 20:10:21.378742 15472 solver.cpp:237]     Train net output #0: loss = 5.24622 (* 1 = 5.24622 loss)
I0409 20:10:21.378755 15472 sgd_solver.cpp:105] Iteration 1272, lr = 0.00777272
I0409 20:10:26.179325 15472 solver.cpp:218] Iteration 1284 (2.49981 iter/s, 4.80036s/12 iters), loss = 5.28506
I0409 20:10:26.179369 15472 solver.cpp:237]     Train net output #0: loss = 5.28506 (* 1 = 5.28506 loss)
I0409 20:10:26.179380 15472 sgd_solver.cpp:105] Iteration 1284, lr = 0.00775426
I0409 20:10:30.981326 15472 solver.cpp:218] Iteration 1296 (2.4991 iter/s, 4.80173s/12 iters), loss = 5.26883
I0409 20:10:30.981379 15472 solver.cpp:237]     Train net output #0: loss = 5.26883 (* 1 = 5.26883 loss)
I0409 20:10:30.981389 15472 sgd_solver.cpp:105] Iteration 1296, lr = 0.00773585
I0409 20:10:35.944638 15472 solver.cpp:218] Iteration 1308 (2.41788 iter/s, 4.96303s/12 iters), loss = 5.25219
I0409 20:10:35.944763 15472 solver.cpp:237]     Train net output #0: loss = 5.25219 (* 1 = 5.25219 loss)
I0409 20:10:35.944773 15472 sgd_solver.cpp:105] Iteration 1308, lr = 0.00771749
I0409 20:10:38.367305 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:10:40.813000 15472 solver.cpp:218] Iteration 1320 (2.46507 iter/s, 4.86801s/12 iters), loss = 5.27396
I0409 20:10:40.813052 15472 solver.cpp:237]     Train net output #0: loss = 5.27396 (* 1 = 5.27396 loss)
I0409 20:10:40.813064 15472 sgd_solver.cpp:105] Iteration 1320, lr = 0.00769916
I0409 20:10:42.886413 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_1326.caffemodel
I0409 20:10:44.428745 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_1326.solverstate
I0409 20:10:44.991611 15472 solver.cpp:330] Iteration 1326, Testing net (#0)
I0409 20:10:44.991642 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:10:48.803633 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:10:49.357733 15472 solver.cpp:397]     Test net output #0: accuracy = 0.00551471
I0409 20:10:49.357775 15472 solver.cpp:397]     Test net output #1: loss = 5.28694 (* 1 = 5.28694 loss)
I0409 20:10:51.184334 15472 solver.cpp:218] Iteration 1332 (1.15709 iter/s, 10.3708s/12 iters), loss = 5.28731
I0409 20:10:51.184381 15472 solver.cpp:237]     Train net output #0: loss = 5.28731 (* 1 = 5.28731 loss)
I0409 20:10:51.184391 15472 sgd_solver.cpp:105] Iteration 1332, lr = 0.00768088
I0409 20:10:56.037226 15472 solver.cpp:218] Iteration 1344 (2.47289 iter/s, 4.85262s/12 iters), loss = 5.28147
I0409 20:10:56.037268 15472 solver.cpp:237]     Train net output #0: loss = 5.28147 (* 1 = 5.28147 loss)
I0409 20:10:56.037277 15472 sgd_solver.cpp:105] Iteration 1344, lr = 0.00766265
I0409 20:11:00.884213 15472 solver.cpp:218] Iteration 1356 (2.4759 iter/s, 4.84672s/12 iters), loss = 5.2768
I0409 20:11:00.884258 15472 solver.cpp:237]     Train net output #0: loss = 5.2768 (* 1 = 5.2768 loss)
I0409 20:11:00.884269 15472 sgd_solver.cpp:105] Iteration 1356, lr = 0.00764446
I0409 20:11:05.740394 15472 solver.cpp:218] Iteration 1368 (2.47122 iter/s, 4.85591s/12 iters), loss = 5.26947
I0409 20:11:05.740433 15472 solver.cpp:237]     Train net output #0: loss = 5.26947 (* 1 = 5.26947 loss)
I0409 20:11:05.740442 15472 sgd_solver.cpp:105] Iteration 1368, lr = 0.00762631
I0409 20:11:05.740654 15472 blocking_queue.cpp:49] Waiting for data
I0409 20:11:10.606369 15472 solver.cpp:218] Iteration 1380 (2.46624 iter/s, 4.86571s/12 iters), loss = 5.27234
I0409 20:11:10.606465 15472 solver.cpp:237]     Train net output #0: loss = 5.27234 (* 1 = 5.27234 loss)
I0409 20:11:10.606475 15472 sgd_solver.cpp:105] Iteration 1380, lr = 0.0076082
I0409 20:11:15.527293 15472 solver.cpp:218] Iteration 1392 (2.43873 iter/s, 4.9206s/12 iters), loss = 5.27207
I0409 20:11:15.527348 15472 solver.cpp:237]     Train net output #0: loss = 5.27207 (* 1 = 5.27207 loss)
I0409 20:11:15.527362 15472 sgd_solver.cpp:105] Iteration 1392, lr = 0.00759014
I0409 20:11:20.479545 15472 solver.cpp:218] Iteration 1404 (2.42328 iter/s, 4.95196s/12 iters), loss = 5.2792
I0409 20:11:20.479598 15472 solver.cpp:237]     Train net output #0: loss = 5.2792 (* 1 = 5.2792 loss)
I0409 20:11:20.479612 15472 sgd_solver.cpp:105] Iteration 1404, lr = 0.00757212
I0409 20:11:25.021785 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:11:25.367513 15472 solver.cpp:218] Iteration 1416 (2.45515 iter/s, 4.88769s/12 iters), loss = 5.25805
I0409 20:11:25.367563 15472 solver.cpp:237]     Train net output #0: loss = 5.25805 (* 1 = 5.25805 loss)
I0409 20:11:25.367575 15472 sgd_solver.cpp:105] Iteration 1416, lr = 0.00755414
I0409 20:11:29.831876 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_1428.caffemodel
I0409 20:11:30.300218 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_1428.solverstate
I0409 20:11:30.653792 15472 solver.cpp:330] Iteration 1428, Testing net (#0)
I0409 20:11:30.653825 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:11:34.415794 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:11:35.004657 15472 solver.cpp:397]     Test net output #0: accuracy = 0.00551471
I0409 20:11:35.004693 15472 solver.cpp:397]     Test net output #1: loss = 5.2864 (* 1 = 5.2864 loss)
I0409 20:11:35.087729 15472 solver.cpp:218] Iteration 1428 (1.2346 iter/s, 9.71972s/12 iters), loss = 5.27615
I0409 20:11:35.087779 15472 solver.cpp:237]     Train net output #0: loss = 5.27615 (* 1 = 5.27615 loss)
I0409 20:11:35.087791 15472 sgd_solver.cpp:105] Iteration 1428, lr = 0.0075362
I0409 20:11:39.155995 15472 solver.cpp:218] Iteration 1440 (2.94984 iter/s, 4.06802s/12 iters), loss = 5.28608
I0409 20:11:39.156050 15472 solver.cpp:237]     Train net output #0: loss = 5.28608 (* 1 = 5.28608 loss)
I0409 20:11:39.156064 15472 sgd_solver.cpp:105] Iteration 1440, lr = 0.00751831
I0409 20:11:44.236985 15472 solver.cpp:218] Iteration 1452 (2.36188 iter/s, 5.0807s/12 iters), loss = 5.28054
I0409 20:11:44.237123 15472 solver.cpp:237]     Train net output #0: loss = 5.28054 (* 1 = 5.28054 loss)
I0409 20:11:44.237135 15472 sgd_solver.cpp:105] Iteration 1452, lr = 0.00750046
I0409 20:11:49.129758 15472 solver.cpp:218] Iteration 1464 (2.45278 iter/s, 4.89241s/12 iters), loss = 5.27604
I0409 20:11:49.129802 15472 solver.cpp:237]     Train net output #0: loss = 5.27604 (* 1 = 5.27604 loss)
I0409 20:11:49.129812 15472 sgd_solver.cpp:105] Iteration 1464, lr = 0.00748265
I0409 20:11:53.965535 15472 solver.cpp:218] Iteration 1476 (2.48165 iter/s, 4.8355s/12 iters), loss = 5.27885
I0409 20:11:53.965585 15472 solver.cpp:237]     Train net output #0: loss = 5.27885 (* 1 = 5.27885 loss)
I0409 20:11:53.965595 15472 sgd_solver.cpp:105] Iteration 1476, lr = 0.00746489
I0409 20:11:58.819299 15472 solver.cpp:218] Iteration 1488 (2.47245 iter/s, 4.85348s/12 iters), loss = 5.25153
I0409 20:11:58.819352 15472 solver.cpp:237]     Train net output #0: loss = 5.25153 (* 1 = 5.25153 loss)
I0409 20:11:58.819365 15472 sgd_solver.cpp:105] Iteration 1488, lr = 0.00744716
I0409 20:12:03.687407 15472 solver.cpp:218] Iteration 1500 (2.46517 iter/s, 4.86782s/12 iters), loss = 5.27084
I0409 20:12:03.687463 15472 solver.cpp:237]     Train net output #0: loss = 5.27084 (* 1 = 5.27084 loss)
I0409 20:12:03.687476 15472 sgd_solver.cpp:105] Iteration 1500, lr = 0.00742948
I0409 20:12:08.572594 15472 solver.cpp:218] Iteration 1512 (2.45655 iter/s, 4.8849s/12 iters), loss = 5.28332
I0409 20:12:08.572639 15472 solver.cpp:237]     Train net output #0: loss = 5.28332 (* 1 = 5.28332 loss)
I0409 20:12:08.572649 15472 sgd_solver.cpp:105] Iteration 1512, lr = 0.00741184
I0409 20:12:10.311935 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:12:13.475239 15472 solver.cpp:218] Iteration 1524 (2.4478 iter/s, 4.90237s/12 iters), loss = 5.27306
I0409 20:12:13.475284 15472 solver.cpp:237]     Train net output #0: loss = 5.27306 (* 1 = 5.27306 loss)
I0409 20:12:13.475293 15472 sgd_solver.cpp:105] Iteration 1524, lr = 0.00739425
I0409 20:12:15.455631 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_1530.caffemodel
I0409 20:12:15.916239 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_1530.solverstate
I0409 20:12:16.237113 15472 solver.cpp:330] Iteration 1530, Testing net (#0)
I0409 20:12:16.237140 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:12:20.189244 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:12:20.848454 15472 solver.cpp:397]     Test net output #0: accuracy = 0.00551471
I0409 20:12:20.848503 15472 solver.cpp:397]     Test net output #1: loss = 5.28623 (* 1 = 5.28623 loss)
I0409 20:12:22.590842 15472 solver.cpp:218] Iteration 1536 (1.31649 iter/s, 9.11514s/12 iters), loss = 5.27673
I0409 20:12:22.590884 15472 solver.cpp:237]     Train net output #0: loss = 5.27673 (* 1 = 5.27673 loss)
I0409 20:12:22.590894 15472 sgd_solver.cpp:105] Iteration 1536, lr = 0.00737669
I0409 20:12:27.449287 15472 solver.cpp:218] Iteration 1548 (2.47006 iter/s, 4.85817s/12 iters), loss = 5.23476
I0409 20:12:27.449334 15472 solver.cpp:237]     Train net output #0: loss = 5.23476 (* 1 = 5.23476 loss)
I0409 20:12:27.449345 15472 sgd_solver.cpp:105] Iteration 1548, lr = 0.00735918
I0409 20:12:32.296681 15472 solver.cpp:218] Iteration 1560 (2.4757 iter/s, 4.84712s/12 iters), loss = 5.29212
I0409 20:12:32.296732 15472 solver.cpp:237]     Train net output #0: loss = 5.29212 (* 1 = 5.29212 loss)
I0409 20:12:32.296742 15472 sgd_solver.cpp:105] Iteration 1560, lr = 0.00734171
I0409 20:12:37.168089 15472 solver.cpp:218] Iteration 1572 (2.4635 iter/s, 4.87112s/12 iters), loss = 5.25727
I0409 20:12:37.168145 15472 solver.cpp:237]     Train net output #0: loss = 5.25727 (* 1 = 5.25727 loss)
I0409 20:12:37.168157 15472 sgd_solver.cpp:105] Iteration 1572, lr = 0.00732427
I0409 20:12:42.012713 15472 solver.cpp:218] Iteration 1584 (2.47712 iter/s, 4.84434s/12 iters), loss = 5.26886
I0409 20:12:42.012761 15472 solver.cpp:237]     Train net output #0: loss = 5.26886 (* 1 = 5.26886 loss)
I0409 20:12:42.012770 15472 sgd_solver.cpp:105] Iteration 1584, lr = 0.00730688
I0409 20:12:46.871076 15472 solver.cpp:218] Iteration 1596 (2.47011 iter/s, 4.85809s/12 iters), loss = 5.26382
I0409 20:12:46.871176 15472 solver.cpp:237]     Train net output #0: loss = 5.26382 (* 1 = 5.26382 loss)
I0409 20:12:46.871186 15472 sgd_solver.cpp:105] Iteration 1596, lr = 0.00728954
I0409 20:12:51.743302 15472 solver.cpp:218] Iteration 1608 (2.4631 iter/s, 4.8719s/12 iters), loss = 5.26837
I0409 20:12:51.743342 15472 solver.cpp:237]     Train net output #0: loss = 5.26837 (* 1 = 5.26837 loss)
I0409 20:12:51.743352 15472 sgd_solver.cpp:105] Iteration 1608, lr = 0.00727223
I0409 20:12:55.554728 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:12:56.610783 15472 solver.cpp:218] Iteration 1620 (2.46548 iter/s, 4.86721s/12 iters), loss = 5.25718
I0409 20:12:56.610841 15472 solver.cpp:237]     Train net output #0: loss = 5.25718 (* 1 = 5.25718 loss)
I0409 20:12:56.610854 15472 sgd_solver.cpp:105] Iteration 1620, lr = 0.00725496
I0409 20:13:01.021155 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_1632.caffemodel
I0409 20:13:01.464869 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_1632.solverstate
I0409 20:13:02.188323 15472 solver.cpp:330] Iteration 1632, Testing net (#0)
I0409 20:13:02.188342 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:13:06.154745 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:13:06.823195 15472 solver.cpp:397]     Test net output #0: accuracy = 0.00551471
I0409 20:13:06.823246 15472 solver.cpp:397]     Test net output #1: loss = 5.28638 (* 1 = 5.28638 loss)
I0409 20:13:06.906411 15472 solver.cpp:218] Iteration 1632 (1.1656 iter/s, 10.2951s/12 iters), loss = 5.28689
I0409 20:13:06.906464 15472 solver.cpp:237]     Train net output #0: loss = 5.28689 (* 1 = 5.28689 loss)
I0409 20:13:06.906476 15472 sgd_solver.cpp:105] Iteration 1632, lr = 0.00723774
I0409 20:13:11.026548 15472 solver.cpp:218] Iteration 1644 (2.91271 iter/s, 4.11988s/12 iters), loss = 5.25254
I0409 20:13:11.026607 15472 solver.cpp:237]     Train net output #0: loss = 5.25254 (* 1 = 5.25254 loss)
I0409 20:13:11.026619 15472 sgd_solver.cpp:105] Iteration 1644, lr = 0.00722056
I0409 20:13:15.844914 15472 solver.cpp:218] Iteration 1656 (2.49062 iter/s, 4.81808s/12 iters), loss = 5.29212
I0409 20:13:15.844956 15472 solver.cpp:237]     Train net output #0: loss = 5.29212 (* 1 = 5.29212 loss)
I0409 20:13:15.844965 15472 sgd_solver.cpp:105] Iteration 1656, lr = 0.00720341
I0409 20:13:20.680691 15472 solver.cpp:218] Iteration 1668 (2.48164 iter/s, 4.83551s/12 iters), loss = 5.26114
I0409 20:13:20.680819 15472 solver.cpp:237]     Train net output #0: loss = 5.26114 (* 1 = 5.26114 loss)
I0409 20:13:20.680833 15472 sgd_solver.cpp:105] Iteration 1668, lr = 0.00718631
I0409 20:13:25.503384 15472 solver.cpp:218] Iteration 1680 (2.48842 iter/s, 4.82234s/12 iters), loss = 5.27963
I0409 20:13:25.503432 15472 solver.cpp:237]     Train net output #0: loss = 5.27963 (* 1 = 5.27963 loss)
I0409 20:13:25.503443 15472 sgd_solver.cpp:105] Iteration 1680, lr = 0.00716925
I0409 20:13:30.438037 15472 solver.cpp:218] Iteration 1692 (2.43192 iter/s, 4.93437s/12 iters), loss = 5.2855
I0409 20:13:30.438089 15472 solver.cpp:237]     Train net output #0: loss = 5.2855 (* 1 = 5.2855 loss)
I0409 20:13:30.438102 15472 sgd_solver.cpp:105] Iteration 1692, lr = 0.00715223
I0409 20:13:35.365780 15472 solver.cpp:218] Iteration 1704 (2.43533 iter/s, 4.92746s/12 iters), loss = 5.26918
I0409 20:13:35.365835 15472 solver.cpp:237]     Train net output #0: loss = 5.26918 (* 1 = 5.26918 loss)
I0409 20:13:35.365847 15472 sgd_solver.cpp:105] Iteration 1704, lr = 0.00713525
I0409 20:13:40.312299 15472 solver.cpp:218] Iteration 1716 (2.42609 iter/s, 4.94623s/12 iters), loss = 5.27949
I0409 20:13:40.312355 15472 solver.cpp:237]     Train net output #0: loss = 5.27949 (* 1 = 5.27949 loss)
I0409 20:13:40.312366 15472 sgd_solver.cpp:105] Iteration 1716, lr = 0.00711831
I0409 20:13:41.341017 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:13:45.224236 15472 solver.cpp:218] Iteration 1728 (2.44317 iter/s, 4.91166s/12 iters), loss = 5.28056
I0409 20:13:45.224273 15472 solver.cpp:237]     Train net output #0: loss = 5.28056 (* 1 = 5.28056 loss)
I0409 20:13:45.224283 15472 sgd_solver.cpp:105] Iteration 1728, lr = 0.00710141
I0409 20:13:47.195483 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_1734.caffemodel
I0409 20:13:47.630650 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_1734.solverstate
I0409 20:13:47.949038 15472 solver.cpp:330] Iteration 1734, Testing net (#0)
I0409 20:13:47.949066 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:13:51.606644 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:13:52.309813 15472 solver.cpp:397]     Test net output #0: accuracy = 0.00551471
I0409 20:13:52.309860 15472 solver.cpp:397]     Test net output #1: loss = 5.28662 (* 1 = 5.28662 loss)
I0409 20:13:54.213290 15472 solver.cpp:218] Iteration 1740 (1.33502 iter/s, 8.9886s/12 iters), loss = 5.25577
I0409 20:13:54.213346 15472 solver.cpp:237]     Train net output #0: loss = 5.25577 (* 1 = 5.25577 loss)
I0409 20:13:54.213361 15472 sgd_solver.cpp:105] Iteration 1740, lr = 0.00708455
I0409 20:13:59.026257 15472 solver.cpp:218] Iteration 1752 (2.49341 iter/s, 4.81269s/12 iters), loss = 5.26747
I0409 20:13:59.026304 15472 solver.cpp:237]     Train net output #0: loss = 5.26747 (* 1 = 5.26747 loss)
I0409 20:13:59.026315 15472 sgd_solver.cpp:105] Iteration 1752, lr = 0.00706773
I0409 20:14:03.914559 15472 solver.cpp:218] Iteration 1764 (2.45498 iter/s, 4.88803s/12 iters), loss = 5.26579
I0409 20:14:03.914615 15472 solver.cpp:237]     Train net output #0: loss = 5.26579 (* 1 = 5.26579 loss)
I0409 20:14:03.914630 15472 sgd_solver.cpp:105] Iteration 1764, lr = 0.00705094
I0409 20:14:08.882899 15472 solver.cpp:218] Iteration 1776 (2.41543 iter/s, 4.96806s/12 iters), loss = 5.28018
I0409 20:14:08.882938 15472 solver.cpp:237]     Train net output #0: loss = 5.28018 (* 1 = 5.28018 loss)
I0409 20:14:08.882949 15472 sgd_solver.cpp:105] Iteration 1776, lr = 0.0070342
I0409 20:14:13.757859 15472 solver.cpp:218] Iteration 1788 (2.4617 iter/s, 4.87469s/12 iters), loss = 5.26547
I0409 20:14:13.757915 15472 solver.cpp:237]     Train net output #0: loss = 5.26547 (* 1 = 5.26547 loss)
I0409 20:14:13.757930 15472 sgd_solver.cpp:105] Iteration 1788, lr = 0.0070175
I0409 20:14:18.581303 15472 solver.cpp:218] Iteration 1800 (2.488 iter/s, 4.82316s/12 iters), loss = 5.27995
I0409 20:14:18.581354 15472 solver.cpp:237]     Train net output #0: loss = 5.27995 (* 1 = 5.27995 loss)
I0409 20:14:18.581367 15472 sgd_solver.cpp:105] Iteration 1800, lr = 0.00700084
I0409 20:14:23.465718 15472 solver.cpp:218] Iteration 1812 (2.45694 iter/s, 4.88413s/12 iters), loss = 5.26877
I0409 20:14:23.465858 15472 solver.cpp:237]     Train net output #0: loss = 5.26877 (* 1 = 5.26877 loss)
I0409 20:14:23.465871 15472 sgd_solver.cpp:105] Iteration 1812, lr = 0.00698422
I0409 20:14:26.614059 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:14:28.324944 15472 solver.cpp:218] Iteration 1824 (2.46972 iter/s, 4.85885s/12 iters), loss = 5.27732
I0409 20:14:28.324995 15472 solver.cpp:237]     Train net output #0: loss = 5.27732 (* 1 = 5.27732 loss)
I0409 20:14:28.325006 15472 sgd_solver.cpp:105] Iteration 1824, lr = 0.00696764
I0409 20:14:32.757345 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_1836.caffemodel
I0409 20:14:33.226263 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_1836.solverstate
I0409 20:14:33.546732 15472 solver.cpp:330] Iteration 1836, Testing net (#0)
I0409 20:14:33.546749 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:14:37.237195 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:14:37.984149 15472 solver.cpp:397]     Test net output #0: accuracy = 0.00551471
I0409 20:14:37.984200 15472 solver.cpp:397]     Test net output #1: loss = 5.28594 (* 1 = 5.28594 loss)
I0409 20:14:38.067309 15472 solver.cpp:218] Iteration 1836 (1.2318 iter/s, 9.74187s/12 iters), loss = 5.27492
I0409 20:14:38.067365 15472 solver.cpp:237]     Train net output #0: loss = 5.27492 (* 1 = 5.27492 loss)
I0409 20:14:38.067378 15472 sgd_solver.cpp:105] Iteration 1836, lr = 0.0069511
I0409 20:14:42.286913 15472 solver.cpp:218] Iteration 1848 (2.84404 iter/s, 4.21935s/12 iters), loss = 5.27357
I0409 20:14:42.286965 15472 solver.cpp:237]     Train net output #0: loss = 5.27357 (* 1 = 5.27357 loss)
I0409 20:14:42.286975 15472 sgd_solver.cpp:105] Iteration 1848, lr = 0.00693459
I0409 20:14:47.063385 15472 solver.cpp:218] Iteration 1860 (2.51246 iter/s, 4.77619s/12 iters), loss = 5.28083
I0409 20:14:47.063427 15472 solver.cpp:237]     Train net output #0: loss = 5.28083 (* 1 = 5.28083 loss)
I0409 20:14:47.063436 15472 sgd_solver.cpp:105] Iteration 1860, lr = 0.00691813
I0409 20:14:51.914646 15472 solver.cpp:218] Iteration 1872 (2.47373 iter/s, 4.85098s/12 iters), loss = 5.27144
I0409 20:14:51.914707 15472 solver.cpp:237]     Train net output #0: loss = 5.27144 (* 1 = 5.27144 loss)
I0409 20:14:51.914721 15472 sgd_solver.cpp:105] Iteration 1872, lr = 0.0069017
I0409 20:14:56.723018 15472 solver.cpp:218] Iteration 1884 (2.4958 iter/s, 4.80808s/12 iters), loss = 5.28772
I0409 20:14:56.724915 15472 solver.cpp:237]     Train net output #0: loss = 5.28772 (* 1 = 5.28772 loss)
I0409 20:14:56.724927 15472 sgd_solver.cpp:105] Iteration 1884, lr = 0.00688532
I0409 20:15:01.601356 15472 solver.cpp:218] Iteration 1896 (2.46093 iter/s, 4.87621s/12 iters), loss = 5.26633
I0409 20:15:01.601402 15472 solver.cpp:237]     Train net output #0: loss = 5.26633 (* 1 = 5.26633 loss)
I0409 20:15:01.601411 15472 sgd_solver.cpp:105] Iteration 1896, lr = 0.00686897
I0409 20:15:06.579813 15472 solver.cpp:218] Iteration 1908 (2.41052 iter/s, 4.97817s/12 iters), loss = 5.28548
I0409 20:15:06.579864 15472 solver.cpp:237]     Train net output #0: loss = 5.28548 (* 1 = 5.28548 loss)
I0409 20:15:06.579876 15472 sgd_solver.cpp:105] Iteration 1908, lr = 0.00685266
I0409 20:15:11.482203 15472 solver.cpp:218] Iteration 1920 (2.44793 iter/s, 4.90211s/12 iters), loss = 5.27606
I0409 20:15:11.482246 15472 solver.cpp:237]     Train net output #0: loss = 5.27606 (* 1 = 5.27606 loss)
I0409 20:15:11.482256 15472 sgd_solver.cpp:105] Iteration 1920, lr = 0.00683639
I0409 20:15:11.802548 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:15:16.335137 15472 solver.cpp:218] Iteration 1932 (2.47287 iter/s, 4.85265s/12 iters), loss = 5.2772
I0409 20:15:16.335191 15472 solver.cpp:237]     Train net output #0: loss = 5.2772 (* 1 = 5.2772 loss)
I0409 20:15:16.335202 15472 sgd_solver.cpp:105] Iteration 1932, lr = 0.00682016
I0409 20:15:18.307161 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_1938.caffemodel
I0409 20:15:19.206739 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_1938.solverstate
I0409 20:15:19.569387 15472 solver.cpp:330] Iteration 1938, Testing net (#0)
I0409 20:15:19.569409 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:15:23.534409 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:15:24.316587 15472 solver.cpp:397]     Test net output #0: accuracy = 0.00551471
I0409 20:15:24.316637 15472 solver.cpp:397]     Test net output #1: loss = 5.28631 (* 1 = 5.28631 loss)
I0409 20:15:26.074754 15472 solver.cpp:218] Iteration 1944 (1.23214 iter/s, 9.73912s/12 iters), loss = 5.27321
I0409 20:15:26.074797 15472 solver.cpp:237]     Train net output #0: loss = 5.27321 (* 1 = 5.27321 loss)
I0409 20:15:26.074805 15472 sgd_solver.cpp:105] Iteration 1944, lr = 0.00680397
I0409 20:15:31.096158 15472 solver.cpp:218] Iteration 1956 (2.3899 iter/s, 5.02112s/12 iters), loss = 5.28157
I0409 20:15:31.096259 15472 solver.cpp:237]     Train net output #0: loss = 5.28157 (* 1 = 5.28157 loss)
I0409 20:15:31.096269 15472 sgd_solver.cpp:105] Iteration 1956, lr = 0.00678782
I0409 20:15:35.883796 15472 solver.cpp:218] Iteration 1968 (2.50663 iter/s, 4.7873s/12 iters), loss = 5.27407
I0409 20:15:35.883857 15472 solver.cpp:237]     Train net output #0: loss = 5.27407 (* 1 = 5.27407 loss)
I0409 20:15:35.883872 15472 sgd_solver.cpp:105] Iteration 1968, lr = 0.0067717
I0409 20:15:40.763298 15472 solver.cpp:218] Iteration 1980 (2.45941 iter/s, 4.87921s/12 iters), loss = 5.25671
I0409 20:15:40.763345 15472 solver.cpp:237]     Train net output #0: loss = 5.25671 (* 1 = 5.25671 loss)
I0409 20:15:40.763353 15472 sgd_solver.cpp:105] Iteration 1980, lr = 0.00675562
I0409 20:15:45.615546 15472 solver.cpp:218] Iteration 1992 (2.47322 iter/s, 4.85197s/12 iters), loss = 5.28148
I0409 20:15:45.615592 15472 solver.cpp:237]     Train net output #0: loss = 5.28148 (* 1 = 5.28148 loss)
I0409 20:15:45.615602 15472 sgd_solver.cpp:105] Iteration 1992, lr = 0.00673958
I0409 20:15:50.407946 15472 solver.cpp:218] Iteration 2004 (2.50411 iter/s, 4.79212s/12 iters), loss = 5.27376
I0409 20:15:50.408004 15472 solver.cpp:237]     Train net output #0: loss = 5.27376 (* 1 = 5.27376 loss)
I0409 20:15:50.408017 15472 sgd_solver.cpp:105] Iteration 2004, lr = 0.00672358
I0409 20:15:55.243914 15472 solver.cpp:218] Iteration 2016 (2.48155 iter/s, 4.83568s/12 iters), loss = 5.25378
I0409 20:15:55.243971 15472 solver.cpp:237]     Train net output #0: loss = 5.25378 (* 1 = 5.25378 loss)
I0409 20:15:55.243983 15472 sgd_solver.cpp:105] Iteration 2016, lr = 0.00670762
I0409 20:15:57.696018 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:16:00.058043 15472 solver.cpp:218] Iteration 2028 (2.49281 iter/s, 4.81384s/12 iters), loss = 5.27575
I0409 20:16:00.058104 15472 solver.cpp:237]     Train net output #0: loss = 5.27575 (* 1 = 5.27575 loss)
I0409 20:16:00.058122 15472 sgd_solver.cpp:105] Iteration 2028, lr = 0.00669169
I0409 20:16:04.455539 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_2040.caffemodel
I0409 20:16:05.667407 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_2040.solverstate
I0409 20:16:06.661684 15472 solver.cpp:330] Iteration 2040, Testing net (#0)
I0409 20:16:06.661715 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:16:10.285892 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:16:11.112426 15472 solver.cpp:397]     Test net output #0: accuracy = 0.00612745
I0409 20:16:11.112462 15472 solver.cpp:397]     Test net output #1: loss = 5.28593 (* 1 = 5.28593 loss)
I0409 20:16:11.195791 15472 solver.cpp:218] Iteration 2040 (1.07747 iter/s, 11.1372s/12 iters), loss = 5.28006
I0409 20:16:11.195839 15472 solver.cpp:237]     Train net output #0: loss = 5.28006 (* 1 = 5.28006 loss)
I0409 20:16:11.195852 15472 sgd_solver.cpp:105] Iteration 2040, lr = 0.00667581
I0409 20:16:15.236308 15472 solver.cpp:218] Iteration 2052 (2.97009 iter/s, 4.04028s/12 iters), loss = 5.28315
I0409 20:16:15.236358 15472 solver.cpp:237]     Train net output #0: loss = 5.28315 (* 1 = 5.28315 loss)
I0409 20:16:15.236371 15472 sgd_solver.cpp:105] Iteration 2052, lr = 0.00665996
I0409 20:16:15.589004 15472 blocking_queue.cpp:49] Waiting for data
I0409 20:16:20.111240 15472 solver.cpp:218] Iteration 2064 (2.46172 iter/s, 4.87464s/12 iters), loss = 5.2716
I0409 20:16:20.111299 15472 solver.cpp:237]     Train net output #0: loss = 5.2716 (* 1 = 5.2716 loss)
I0409 20:16:20.111310 15472 sgd_solver.cpp:105] Iteration 2064, lr = 0.00664414
I0409 20:16:24.983209 15472 solver.cpp:218] Iteration 2076 (2.46321 iter/s, 4.87168s/12 iters), loss = 5.28117
I0409 20:16:24.983261 15472 solver.cpp:237]     Train net output #0: loss = 5.28117 (* 1 = 5.28117 loss)
I0409 20:16:24.983273 15472 sgd_solver.cpp:105] Iteration 2076, lr = 0.00662837
I0409 20:16:29.831260 15472 solver.cpp:218] Iteration 2088 (2.47537 iter/s, 4.84777s/12 iters), loss = 5.27219
I0409 20:16:29.831305 15472 solver.cpp:237]     Train net output #0: loss = 5.27219 (* 1 = 5.27219 loss)
I0409 20:16:29.831315 15472 sgd_solver.cpp:105] Iteration 2088, lr = 0.00661263
I0409 20:16:34.758422 15472 solver.cpp:218] Iteration 2100 (2.43562 iter/s, 4.92688s/12 iters), loss = 5.27076
I0409 20:16:34.758510 15472 solver.cpp:237]     Train net output #0: loss = 5.27076 (* 1 = 5.27076 loss)
I0409 20:16:34.758520 15472 sgd_solver.cpp:105] Iteration 2100, lr = 0.00659693
I0409 20:16:39.759462 15472 solver.cpp:218] Iteration 2112 (2.39966 iter/s, 5.00072s/12 iters), loss = 5.27849
I0409 20:16:39.759508 15472 solver.cpp:237]     Train net output #0: loss = 5.27849 (* 1 = 5.27849 loss)
I0409 20:16:39.759519 15472 sgd_solver.cpp:105] Iteration 2112, lr = 0.00658127
I0409 20:16:44.235108 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:16:44.543748 15472 solver.cpp:218] Iteration 2124 (2.50835 iter/s, 4.78401s/12 iters), loss = 5.25696
I0409 20:16:44.543802 15472 solver.cpp:237]     Train net output #0: loss = 5.25696 (* 1 = 5.25696 loss)
I0409 20:16:44.543813 15472 sgd_solver.cpp:105] Iteration 2124, lr = 0.00656564
I0409 20:16:49.372511 15472 solver.cpp:218] Iteration 2136 (2.48525 iter/s, 4.82849s/12 iters), loss = 5.27179
I0409 20:16:49.372553 15472 solver.cpp:237]     Train net output #0: loss = 5.27179 (* 1 = 5.27179 loss)
I0409 20:16:49.372562 15472 sgd_solver.cpp:105] Iteration 2136, lr = 0.00655006
I0409 20:16:51.356227 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_2142.caffemodel
I0409 20:16:51.839267 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_2142.solverstate
I0409 20:16:52.170783 15472 solver.cpp:330] Iteration 2142, Testing net (#0)
I0409 20:16:52.170811 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:16:55.729537 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:16:56.588425 15472 solver.cpp:397]     Test net output #0: accuracy = 0.00551471
I0409 20:16:56.588469 15472 solver.cpp:397]     Test net output #1: loss = 5.28579 (* 1 = 5.28579 loss)
I0409 20:16:58.485877 15472 solver.cpp:218] Iteration 2148 (1.31681 iter/s, 9.1129s/12 iters), loss = 5.27749
I0409 20:16:58.485929 15472 solver.cpp:237]     Train net output #0: loss = 5.27749 (* 1 = 5.27749 loss)
I0409 20:16:58.485942 15472 sgd_solver.cpp:105] Iteration 2148, lr = 0.00653451
I0409 20:17:03.391083 15472 solver.cpp:218] Iteration 2160 (2.44652 iter/s, 4.90492s/12 iters), loss = 5.28403
I0409 20:17:03.391139 15472 solver.cpp:237]     Train net output #0: loss = 5.28403 (* 1 = 5.28403 loss)
I0409 20:17:03.391150 15472 sgd_solver.cpp:105] Iteration 2160, lr = 0.00651899
I0409 20:17:08.325515 15472 solver.cpp:218] Iteration 2172 (2.43203 iter/s, 4.93415s/12 iters), loss = 5.27672
I0409 20:17:08.325664 15472 solver.cpp:237]     Train net output #0: loss = 5.27672 (* 1 = 5.27672 loss)
I0409 20:17:08.325678 15472 sgd_solver.cpp:105] Iteration 2172, lr = 0.00650351
I0409 20:17:13.186833 15472 solver.cpp:218] Iteration 2184 (2.46866 iter/s, 4.86094s/12 iters), loss = 5.27365
I0409 20:17:13.186882 15472 solver.cpp:237]     Train net output #0: loss = 5.27365 (* 1 = 5.27365 loss)
I0409 20:17:13.186892 15472 sgd_solver.cpp:105] Iteration 2184, lr = 0.00648807
I0409 20:17:18.135254 15472 solver.cpp:218] Iteration 2196 (2.42515 iter/s, 4.94814s/12 iters), loss = 5.25432
I0409 20:17:18.135298 15472 solver.cpp:237]     Train net output #0: loss = 5.25432 (* 1 = 5.25432 loss)
I0409 20:17:18.135306 15472 sgd_solver.cpp:105] Iteration 2196, lr = 0.00647267
I0409 20:17:23.047812 15472 solver.cpp:218] Iteration 2208 (2.44285 iter/s, 4.91229s/12 iters), loss = 5.26852
I0409 20:17:23.047853 15472 solver.cpp:237]     Train net output #0: loss = 5.26852 (* 1 = 5.26852 loss)
I0409 20:17:23.047863 15472 sgd_solver.cpp:105] Iteration 2208, lr = 0.0064573
I0409 20:17:28.015115 15472 solver.cpp:218] Iteration 2220 (2.41593 iter/s, 4.96702s/12 iters), loss = 5.28029
I0409 20:17:28.015159 15472 solver.cpp:237]     Train net output #0: loss = 5.28029 (* 1 = 5.28029 loss)
I0409 20:17:28.015170 15472 sgd_solver.cpp:105] Iteration 2220, lr = 0.00644197
I0409 20:17:29.993392 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:17:33.110824 15472 solver.cpp:218] Iteration 2232 (2.35505 iter/s, 5.09543s/12 iters), loss = 5.28331
I0409 20:17:33.110864 15472 solver.cpp:237]     Train net output #0: loss = 5.28331 (* 1 = 5.28331 loss)
I0409 20:17:33.110872 15472 sgd_solver.cpp:105] Iteration 2232, lr = 0.00642668
I0409 20:17:37.539119 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_2244.caffemodel
I0409 20:17:38.622092 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_2244.solverstate
I0409 20:17:39.487845 15472 solver.cpp:330] Iteration 2244, Testing net (#0)
I0409 20:17:39.487865 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:17:43.031507 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:17:43.936414 15472 solver.cpp:397]     Test net output #0: accuracy = 0.00551471
I0409 20:17:43.936458 15472 solver.cpp:397]     Test net output #1: loss = 5.285 (* 1 = 5.285 loss)
I0409 20:17:44.019305 15472 solver.cpp:218] Iteration 2244 (1.10012 iter/s, 10.9079s/12 iters), loss = 5.27679
I0409 20:17:44.019356 15472 solver.cpp:237]     Train net output #0: loss = 5.27679 (* 1 = 5.27679 loss)
I0409 20:17:44.019366 15472 sgd_solver.cpp:105] Iteration 2244, lr = 0.00641142
I0409 20:17:48.008023 15472 solver.cpp:218] Iteration 2256 (3.00867 iter/s, 3.98848s/12 iters), loss = 5.24523
I0409 20:17:48.008072 15472 solver.cpp:237]     Train net output #0: loss = 5.24523 (* 1 = 5.24523 loss)
I0409 20:17:48.008085 15472 sgd_solver.cpp:105] Iteration 2256, lr = 0.0063962
I0409 20:17:52.814275 15472 solver.cpp:218] Iteration 2268 (2.49689 iter/s, 4.80597s/12 iters), loss = 5.28104
I0409 20:17:52.814319 15472 solver.cpp:237]     Train net output #0: loss = 5.28104 (* 1 = 5.28104 loss)
I0409 20:17:52.814327 15472 sgd_solver.cpp:105] Iteration 2268, lr = 0.00638101
I0409 20:17:57.719476 15472 solver.cpp:218] Iteration 2280 (2.44652 iter/s, 4.90493s/12 iters), loss = 5.25654
I0409 20:17:57.719523 15472 solver.cpp:237]     Train net output #0: loss = 5.25654 (* 1 = 5.25654 loss)
I0409 20:17:57.719534 15472 sgd_solver.cpp:105] Iteration 2280, lr = 0.00636586
I0409 20:18:02.616050 15472 solver.cpp:218] Iteration 2292 (2.45083 iter/s, 4.8963s/12 iters), loss = 5.2696
I0409 20:18:02.616099 15472 solver.cpp:237]     Train net output #0: loss = 5.2696 (* 1 = 5.2696 loss)
I0409 20:18:02.616111 15472 sgd_solver.cpp:105] Iteration 2292, lr = 0.00635075
I0409 20:18:07.601426 15472 solver.cpp:218] Iteration 2304 (2.40718 iter/s, 4.98509s/12 iters), loss = 5.27025
I0409 20:18:07.601477 15472 solver.cpp:237]     Train net output #0: loss = 5.27025 (* 1 = 5.27025 loss)
I0409 20:18:07.601488 15472 sgd_solver.cpp:105] Iteration 2304, lr = 0.00633567
I0409 20:18:12.424243 15472 solver.cpp:218] Iteration 2316 (2.48832 iter/s, 4.82254s/12 iters), loss = 5.2615
I0409 20:18:12.424399 15472 solver.cpp:237]     Train net output #0: loss = 5.2615 (* 1 = 5.2615 loss)
I0409 20:18:12.424414 15472 sgd_solver.cpp:105] Iteration 2316, lr = 0.00632063
I0409 20:18:16.309211 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:18:17.370090 15472 solver.cpp:218] Iteration 2328 (2.42647 iter/s, 4.94546s/12 iters), loss = 5.25838
I0409 20:18:17.370138 15472 solver.cpp:237]     Train net output #0: loss = 5.25838 (* 1 = 5.25838 loss)
I0409 20:18:17.370148 15472 sgd_solver.cpp:105] Iteration 2328, lr = 0.00630562
I0409 20:18:22.198702 15472 solver.cpp:218] Iteration 2340 (2.48533 iter/s, 4.82833s/12 iters), loss = 5.28491
I0409 20:18:22.198757 15472 solver.cpp:237]     Train net output #0: loss = 5.28491 (* 1 = 5.28491 loss)
I0409 20:18:22.198770 15472 sgd_solver.cpp:105] Iteration 2340, lr = 0.00629065
I0409 20:18:24.286814 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_2346.caffemodel
I0409 20:18:24.783869 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_2346.solverstate
I0409 20:18:25.113664 15472 solver.cpp:330] Iteration 2346, Testing net (#0)
I0409 20:18:25.113693 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:18:28.498858 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:18:29.475301 15472 solver.cpp:397]     Test net output #0: accuracy = 0.00551471
I0409 20:18:29.475351 15472 solver.cpp:397]     Test net output #1: loss = 5.28376 (* 1 = 5.28376 loss)
I0409 20:18:31.290779 15472 solver.cpp:218] Iteration 2352 (1.3199 iter/s, 9.0916s/12 iters), loss = 5.25539
I0409 20:18:31.290839 15472 solver.cpp:237]     Train net output #0: loss = 5.25539 (* 1 = 5.25539 loss)
I0409 20:18:31.290853 15472 sgd_solver.cpp:105] Iteration 2352, lr = 0.00627571
I0409 20:18:36.367442 15472 solver.cpp:218] Iteration 2364 (2.36389 iter/s, 5.07637s/12 iters), loss = 5.2969
I0409 20:18:36.367489 15472 solver.cpp:237]     Train net output #0: loss = 5.2969 (* 1 = 5.2969 loss)
I0409 20:18:36.367501 15472 sgd_solver.cpp:105] Iteration 2364, lr = 0.00626081
I0409 20:18:41.251008 15472 solver.cpp:218] Iteration 2376 (2.45736 iter/s, 4.88329s/12 iters), loss = 5.25816
I0409 20:18:41.251056 15472 solver.cpp:237]     Train net output #0: loss = 5.25816 (* 1 = 5.25816 loss)
I0409 20:18:41.251067 15472 sgd_solver.cpp:105] Iteration 2376, lr = 0.00624595
I0409 20:18:46.098706 15472 solver.cpp:218] Iteration 2388 (2.47554 iter/s, 4.84742s/12 iters), loss = 5.27301
I0409 20:18:46.098798 15472 solver.cpp:237]     Train net output #0: loss = 5.27301 (* 1 = 5.27301 loss)
I0409 20:18:46.098809 15472 sgd_solver.cpp:105] Iteration 2388, lr = 0.00623112
I0409 20:18:50.915825 15472 solver.cpp:218] Iteration 2400 (2.49128 iter/s, 4.8168s/12 iters), loss = 5.27958
I0409 20:18:50.915877 15472 solver.cpp:237]     Train net output #0: loss = 5.27958 (* 1 = 5.27958 loss)
I0409 20:18:50.915890 15472 sgd_solver.cpp:105] Iteration 2400, lr = 0.00621633
I0409 20:18:55.728134 15472 solver.cpp:218] Iteration 2412 (2.49375 iter/s, 4.81202s/12 iters), loss = 5.26843
I0409 20:18:55.728188 15472 solver.cpp:237]     Train net output #0: loss = 5.26843 (* 1 = 5.26843 loss)
I0409 20:18:55.728200 15472 sgd_solver.cpp:105] Iteration 2412, lr = 0.00620157
I0409 20:19:00.599253 15472 solver.cpp:218] Iteration 2424 (2.46364 iter/s, 4.87083s/12 iters), loss = 5.27248
I0409 20:19:00.599308 15472 solver.cpp:237]     Train net output #0: loss = 5.27248 (* 1 = 5.27248 loss)
I0409 20:19:00.599321 15472 sgd_solver.cpp:105] Iteration 2424, lr = 0.00618684
I0409 20:19:01.658469 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:19:05.434181 15472 solver.cpp:218] Iteration 2436 (2.48208 iter/s, 4.83465s/12 iters), loss = 5.27311
I0409 20:19:05.434235 15472 solver.cpp:237]     Train net output #0: loss = 5.27311 (* 1 = 5.27311 loss)
I0409 20:19:05.434247 15472 sgd_solver.cpp:105] Iteration 2436, lr = 0.00617215
I0409 20:19:09.801431 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_2448.caffemodel
I0409 20:19:10.505946 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_2448.solverstate
I0409 20:19:10.986194 15472 solver.cpp:330] Iteration 2448, Testing net (#0)
I0409 20:19:10.986210 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:19:14.380414 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:19:15.368923 15472 solver.cpp:397]     Test net output #0: accuracy = 0.00551471
I0409 20:19:15.368975 15472 solver.cpp:397]     Test net output #1: loss = 5.2787 (* 1 = 5.2787 loss)
I0409 20:19:15.450217 15472 solver.cpp:218] Iteration 2448 (1.19814 iter/s, 10.0155s/12 iters), loss = 5.2499
I0409 20:19:15.450263 15472 solver.cpp:237]     Train net output #0: loss = 5.2499 (* 1 = 5.2499 loss)
I0409 20:19:15.450274 15472 sgd_solver.cpp:105] Iteration 2448, lr = 0.0061575
I0409 20:19:19.887900 15472 solver.cpp:218] Iteration 2460 (2.70427 iter/s, 4.43743s/12 iters), loss = 5.26027
I0409 20:19:19.888186 15472 solver.cpp:237]     Train net output #0: loss = 5.26027 (* 1 = 5.26027 loss)
I0409 20:19:19.888196 15472 sgd_solver.cpp:105] Iteration 2460, lr = 0.00614288
I0409 20:19:24.674249 15472 solver.cpp:218] Iteration 2472 (2.5074 iter/s, 4.78584s/12 iters), loss = 5.25937
I0409 20:19:24.674299 15472 solver.cpp:237]     Train net output #0: loss = 5.25937 (* 1 = 5.25937 loss)
I0409 20:19:24.674310 15472 sgd_solver.cpp:105] Iteration 2472, lr = 0.0061283
I0409 20:19:29.657138 15472 solver.cpp:218] Iteration 2484 (2.40838 iter/s, 4.9826s/12 iters), loss = 5.26675
I0409 20:19:29.657189 15472 solver.cpp:237]     Train net output #0: loss = 5.26675 (* 1 = 5.26675 loss)
I0409 20:19:29.657202 15472 sgd_solver.cpp:105] Iteration 2484, lr = 0.00611375
I0409 20:19:34.658531 15472 solver.cpp:218] Iteration 2496 (2.39947 iter/s, 5.00111s/12 iters), loss = 5.26224
I0409 20:19:34.658586 15472 solver.cpp:237]     Train net output #0: loss = 5.26224 (* 1 = 5.26224 loss)
I0409 20:19:34.658598 15472 sgd_solver.cpp:105] Iteration 2496, lr = 0.00609923
I0409 20:19:39.499557 15472 solver.cpp:218] Iteration 2508 (2.47896 iter/s, 4.84074s/12 iters), loss = 5.27485
I0409 20:19:39.499612 15472 solver.cpp:237]     Train net output #0: loss = 5.27485 (* 1 = 5.27485 loss)
I0409 20:19:39.499624 15472 sgd_solver.cpp:105] Iteration 2508, lr = 0.00608475
I0409 20:19:44.297547 15472 solver.cpp:218] Iteration 2520 (2.5012 iter/s, 4.7977s/12 iters), loss = 5.24371
I0409 20:19:44.297607 15472 solver.cpp:237]     Train net output #0: loss = 5.24371 (* 1 = 5.24371 loss)
I0409 20:19:44.297621 15472 sgd_solver.cpp:105] Iteration 2520, lr = 0.0060703
I0409 20:19:47.366624 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:19:49.073592 15472 solver.cpp:218] Iteration 2532 (2.51269 iter/s, 4.77575s/12 iters), loss = 5.24219
I0409 20:19:49.073650 15472 solver.cpp:237]     Train net output #0: loss = 5.24219 (* 1 = 5.24219 loss)
I0409 20:19:49.073662 15472 sgd_solver.cpp:105] Iteration 2532, lr = 0.00605589
I0409 20:19:53.937892 15472 solver.cpp:218] Iteration 2544 (2.4671 iter/s, 4.86401s/12 iters), loss = 5.26913
I0409 20:19:53.938045 15472 solver.cpp:237]     Train net output #0: loss = 5.26913 (* 1 = 5.26913 loss)
I0409 20:19:53.938057 15472 sgd_solver.cpp:105] Iteration 2544, lr = 0.00604151
I0409 20:19:55.922399 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_2550.caffemodel
I0409 20:19:57.128976 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_2550.solverstate
I0409 20:19:57.785471 15472 solver.cpp:330] Iteration 2550, Testing net (#0)
I0409 20:19:57.785495 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:20:01.106882 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:20:02.127885 15472 solver.cpp:397]     Test net output #0: accuracy = 0.0067402
I0409 20:20:02.127935 15472 solver.cpp:397]     Test net output #1: loss = 5.22468 (* 1 = 5.22468 loss)
I0409 20:20:03.790119 15472 solver.cpp:218] Iteration 2556 (1.21807 iter/s, 9.85162s/12 iters), loss = 5.24101
I0409 20:20:03.790169 15472 solver.cpp:237]     Train net output #0: loss = 5.24101 (* 1 = 5.24101 loss)
I0409 20:20:03.790181 15472 sgd_solver.cpp:105] Iteration 2556, lr = 0.00602717
I0409 20:20:08.634582 15472 solver.cpp:218] Iteration 2568 (2.4772 iter/s, 4.84418s/12 iters), loss = 5.19633
I0409 20:20:08.634630 15472 solver.cpp:237]     Train net output #0: loss = 5.19633 (* 1 = 5.19633 loss)
I0409 20:20:08.634639 15472 sgd_solver.cpp:105] Iteration 2568, lr = 0.00601286
I0409 20:20:13.549685 15472 solver.cpp:218] Iteration 2580 (2.4416 iter/s, 4.91481s/12 iters), loss = 5.18197
I0409 20:20:13.549754 15472 solver.cpp:237]     Train net output #0: loss = 5.18197 (* 1 = 5.18197 loss)
I0409 20:20:13.549774 15472 sgd_solver.cpp:105] Iteration 2580, lr = 0.00599858
I0409 20:20:18.460158 15472 solver.cpp:218] Iteration 2592 (2.4439 iter/s, 4.91018s/12 iters), loss = 5.19295
I0409 20:20:18.460212 15472 solver.cpp:237]     Train net output #0: loss = 5.19295 (* 1 = 5.19295 loss)
I0409 20:20:18.460223 15472 sgd_solver.cpp:105] Iteration 2592, lr = 0.00598434
I0409 20:20:23.294687 15472 solver.cpp:218] Iteration 2604 (2.48229 iter/s, 4.83424s/12 iters), loss = 5.1731
I0409 20:20:23.294736 15472 solver.cpp:237]     Train net output #0: loss = 5.1731 (* 1 = 5.1731 loss)
I0409 20:20:23.294749 15472 sgd_solver.cpp:105] Iteration 2604, lr = 0.00597013
I0409 20:20:28.144439 15472 solver.cpp:218] Iteration 2616 (2.4745 iter/s, 4.84947s/12 iters), loss = 5.17172
I0409 20:20:28.144526 15472 solver.cpp:237]     Train net output #0: loss = 5.17172 (* 1 = 5.17172 loss)
I0409 20:20:28.144536 15472 sgd_solver.cpp:105] Iteration 2616, lr = 0.00595596
I0409 20:20:33.046700 15472 solver.cpp:218] Iteration 2628 (2.44801 iter/s, 4.90193s/12 iters), loss = 5.22175
I0409 20:20:33.046756 15472 solver.cpp:237]     Train net output #0: loss = 5.22175 (* 1 = 5.22175 loss)
I0409 20:20:33.046768 15472 sgd_solver.cpp:105] Iteration 2628, lr = 0.00594182
I0409 20:20:33.477557 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:20:37.926093 15472 solver.cpp:218] Iteration 2640 (2.45947 iter/s, 4.87911s/12 iters), loss = 5.16477
I0409 20:20:37.926147 15472 solver.cpp:237]     Train net output #0: loss = 5.16477 (* 1 = 5.16477 loss)
I0409 20:20:37.926160 15472 sgd_solver.cpp:105] Iteration 2640, lr = 0.00592771
I0409 20:20:42.347005 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_2652.caffemodel
I0409 20:20:43.216053 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_2652.solverstate
I0409 20:20:43.766336 15472 solver.cpp:330] Iteration 2652, Testing net (#0)
I0409 20:20:43.766357 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:20:47.315145 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:20:48.568928 15472 solver.cpp:397]     Test net output #0: accuracy = 0.00796569
I0409 20:20:48.568964 15472 solver.cpp:397]     Test net output #1: loss = 5.16959 (* 1 = 5.16959 loss)
I0409 20:20:48.653337 15472 solver.cpp:218] Iteration 2652 (1.1187 iter/s, 10.7267s/12 iters), loss = 5.18308
I0409 20:20:48.653383 15472 solver.cpp:237]     Train net output #0: loss = 5.18308 (* 1 = 5.18308 loss)
I0409 20:20:48.653393 15472 sgd_solver.cpp:105] Iteration 2652, lr = 0.00591364
I0409 20:20:52.833094 15472 solver.cpp:218] Iteration 2664 (2.87115 iter/s, 4.17951s/12 iters), loss = 5.13648
I0409 20:20:52.833153 15472 solver.cpp:237]     Train net output #0: loss = 5.13648 (* 1 = 5.13648 loss)
I0409 20:20:52.833164 15472 sgd_solver.cpp:105] Iteration 2664, lr = 0.0058996
I0409 20:20:57.753093 15472 solver.cpp:218] Iteration 2676 (2.43917 iter/s, 4.9197s/12 iters), loss = 5.09224
I0409 20:20:57.753149 15472 solver.cpp:237]     Train net output #0: loss = 5.09224 (* 1 = 5.09224 loss)
I0409 20:20:57.753161 15472 sgd_solver.cpp:105] Iteration 2676, lr = 0.00588559
I0409 20:21:02.659554 15472 solver.cpp:218] Iteration 2688 (2.4459 iter/s, 4.90617s/12 iters), loss = 5.15869
I0409 20:21:02.670068 15472 solver.cpp:237]     Train net output #0: loss = 5.15869 (* 1 = 5.15869 loss)
I0409 20:21:02.670087 15472 sgd_solver.cpp:105] Iteration 2688, lr = 0.00587162
I0409 20:21:07.496639 15472 solver.cpp:218] Iteration 2700 (2.48635 iter/s, 4.82636s/12 iters), loss = 5.12084
I0409 20:21:07.496692 15472 solver.cpp:237]     Train net output #0: loss = 5.12084 (* 1 = 5.12084 loss)
I0409 20:21:07.496704 15472 sgd_solver.cpp:105] Iteration 2700, lr = 0.00585768
I0409 20:21:12.336892 15472 solver.cpp:218] Iteration 2712 (2.47935 iter/s, 4.83997s/12 iters), loss = 5.15612
I0409 20:21:12.336941 15472 solver.cpp:237]     Train net output #0: loss = 5.15612 (* 1 = 5.15612 loss)
I0409 20:21:12.336951 15472 sgd_solver.cpp:105] Iteration 2712, lr = 0.00584377
I0409 20:21:17.186394 15472 solver.cpp:218] Iteration 2724 (2.47462 iter/s, 4.84922s/12 iters), loss = 5.12642
I0409 20:21:17.186444 15472 solver.cpp:237]     Train net output #0: loss = 5.12642 (* 1 = 5.12642 loss)
I0409 20:21:17.186453 15472 sgd_solver.cpp:105] Iteration 2724, lr = 0.0058299
I0409 20:21:19.645524 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:21:21.968151 15472 solver.cpp:218] Iteration 2736 (2.50968 iter/s, 4.78148s/12 iters), loss = 5.20597
I0409 20:21:21.968206 15472 solver.cpp:237]     Train net output #0: loss = 5.20597 (* 1 = 5.20597 loss)
I0409 20:21:21.968219 15472 sgd_solver.cpp:105] Iteration 2736, lr = 0.00581605
I0409 20:21:26.808202 15472 solver.cpp:218] Iteration 2748 (2.47946 iter/s, 4.83977s/12 iters), loss = 5.1388
I0409 20:21:26.808251 15472 solver.cpp:237]     Train net output #0: loss = 5.1388 (* 1 = 5.1388 loss)
I0409 20:21:26.808261 15472 sgd_solver.cpp:105] Iteration 2748, lr = 0.00580225
I0409 20:21:28.784869 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_2754.caffemodel
I0409 20:21:30.150599 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_2754.solverstate
I0409 20:21:30.493407 15472 solver.cpp:330] Iteration 2754, Testing net (#0)
I0409 20:21:30.493433 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:21:33.322993 15472 blocking_queue.cpp:49] Waiting for data
I0409 20:21:33.829394 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:21:34.932857 15472 solver.cpp:397]     Test net output #0: accuracy = 0.00796569
I0409 20:21:34.932904 15472 solver.cpp:397]     Test net output #1: loss = 5.1455 (* 1 = 5.1455 loss)
I0409 20:21:36.738835 15472 solver.cpp:218] Iteration 2760 (1.20844 iter/s, 9.93013s/12 iters), loss = 5.05103
I0409 20:21:36.738893 15472 solver.cpp:237]     Train net output #0: loss = 5.05103 (* 1 = 5.05103 loss)
I0409 20:21:36.738905 15472 sgd_solver.cpp:105] Iteration 2760, lr = 0.00578847
I0409 20:21:41.515339 15472 solver.cpp:218] Iteration 2772 (2.51245 iter/s, 4.77622s/12 iters), loss = 5.11901
I0409 20:21:41.515379 15472 solver.cpp:237]     Train net output #0: loss = 5.11901 (* 1 = 5.11901 loss)
I0409 20:21:41.515389 15472 sgd_solver.cpp:105] Iteration 2772, lr = 0.00577473
I0409 20:21:46.335045 15472 solver.cpp:218] Iteration 2784 (2.48992 iter/s, 4.81943s/12 iters), loss = 5.13529
I0409 20:21:46.335098 15472 solver.cpp:237]     Train net output #0: loss = 5.13529 (* 1 = 5.13529 loss)
I0409 20:21:46.335109 15472 sgd_solver.cpp:105] Iteration 2784, lr = 0.00576102
I0409 20:21:51.179664 15472 solver.cpp:218] Iteration 2796 (2.47712 iter/s, 4.84434s/12 iters), loss = 5.07925
I0409 20:21:51.179706 15472 solver.cpp:237]     Train net output #0: loss = 5.07925 (* 1 = 5.07925 loss)
I0409 20:21:51.179714 15472 sgd_solver.cpp:105] Iteration 2796, lr = 0.00574734
I0409 20:21:55.977506 15472 solver.cpp:218] Iteration 2808 (2.50127 iter/s, 4.79757s/12 iters), loss = 4.98551
I0409 20:21:55.977560 15472 solver.cpp:237]     Train net output #0: loss = 4.98551 (* 1 = 4.98551 loss)
I0409 20:21:55.977571 15472 sgd_solver.cpp:105] Iteration 2808, lr = 0.00573369
I0409 20:22:00.854744 15472 solver.cpp:218] Iteration 2820 (2.46056 iter/s, 4.87695s/12 iters), loss = 5.10966
I0409 20:22:00.854801 15472 solver.cpp:237]     Train net output #0: loss = 5.10966 (* 1 = 5.10966 loss)
I0409 20:22:00.854813 15472 sgd_solver.cpp:105] Iteration 2820, lr = 0.00572008
I0409 20:22:05.627472 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:22:05.925220 15472 solver.cpp:218] Iteration 2832 (2.36678 iter/s, 5.07018s/12 iters), loss = 5.2026
I0409 20:22:05.925262 15472 solver.cpp:237]     Train net output #0: loss = 5.2026 (* 1 = 5.2026 loss)
I0409 20:22:05.925271 15472 sgd_solver.cpp:105] Iteration 2832, lr = 0.0057065
I0409 20:22:10.828784 15472 solver.cpp:218] Iteration 2844 (2.44734 iter/s, 4.90329s/12 iters), loss = 5.18574
I0409 20:22:10.828842 15472 solver.cpp:237]     Train net output #0: loss = 5.18574 (* 1 = 5.18574 loss)
I0409 20:22:10.828855 15472 sgd_solver.cpp:105] Iteration 2844, lr = 0.00569295
I0409 20:22:15.197391 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_2856.caffemodel
I0409 20:22:15.644291 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_2856.solverstate
I0409 20:22:15.959692 15472 solver.cpp:330] Iteration 2856, Testing net (#0)
I0409 20:22:15.959724 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:22:19.522002 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:22:20.668550 15472 solver.cpp:397]     Test net output #0: accuracy = 0.00919118
I0409 20:22:20.668577 15472 solver.cpp:397]     Test net output #1: loss = 5.11171 (* 1 = 5.11171 loss)
I0409 20:22:20.751664 15472 solver.cpp:218] Iteration 2856 (1.20939 iter/s, 9.92237s/12 iters), loss = 5.04406
I0409 20:22:20.751704 15472 solver.cpp:237]     Train net output #0: loss = 5.04406 (* 1 = 5.04406 loss)
I0409 20:22:20.751713 15472 sgd_solver.cpp:105] Iteration 2856, lr = 0.00567944
I0409 20:22:24.790452 15472 solver.cpp:218] Iteration 2868 (2.97136 iter/s, 4.03855s/12 iters), loss = 5.08123
I0409 20:22:24.790493 15472 solver.cpp:237]     Train net output #0: loss = 5.08123 (* 1 = 5.08123 loss)
I0409 20:22:24.790500 15472 sgd_solver.cpp:105] Iteration 2868, lr = 0.00566595
I0409 20:22:29.651178 15472 solver.cpp:218] Iteration 2880 (2.46891 iter/s, 4.86045s/12 iters), loss = 5.14922
I0409 20:22:29.651234 15472 solver.cpp:237]     Train net output #0: loss = 5.14922 (* 1 = 5.14922 loss)
I0409 20:22:29.651247 15472 sgd_solver.cpp:105] Iteration 2880, lr = 0.0056525
I0409 20:22:34.524749 15472 solver.cpp:218] Iteration 2892 (2.4624 iter/s, 4.87329s/12 iters), loss = 5.05504
I0409 20:22:34.524787 15472 solver.cpp:237]     Train net output #0: loss = 5.05504 (* 1 = 5.05504 loss)
I0409 20:22:34.524797 15472 sgd_solver.cpp:105] Iteration 2892, lr = 0.00563908
I0409 20:22:39.423763 15472 solver.cpp:218] Iteration 2904 (2.44961 iter/s, 4.89874s/12 iters), loss = 5.13536
I0409 20:22:39.425446 15472 solver.cpp:237]     Train net output #0: loss = 5.13536 (* 1 = 5.13536 loss)
I0409 20:22:39.425457 15472 sgd_solver.cpp:105] Iteration 2904, lr = 0.00562569
I0409 20:22:44.255681 15472 solver.cpp:218] Iteration 2916 (2.48447 iter/s, 4.83001s/12 iters), loss = 5.08167
I0409 20:22:44.255735 15472 solver.cpp:237]     Train net output #0: loss = 5.08167 (* 1 = 5.08167 loss)
I0409 20:22:44.255748 15472 sgd_solver.cpp:105] Iteration 2916, lr = 0.00561233
I0409 20:22:49.149821 15472 solver.cpp:218] Iteration 2928 (2.45206 iter/s, 4.89385s/12 iters), loss = 5.123
I0409 20:22:49.149878 15472 solver.cpp:237]     Train net output #0: loss = 5.123 (* 1 = 5.123 loss)
I0409 20:22:49.149889 15472 sgd_solver.cpp:105] Iteration 2928, lr = 0.00559901
I0409 20:22:50.954532 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:22:54.043524 15472 solver.cpp:218] Iteration 2940 (2.45228 iter/s, 4.89341s/12 iters), loss = 5.14361
I0409 20:22:54.043587 15472 solver.cpp:237]     Train net output #0: loss = 5.14361 (* 1 = 5.14361 loss)
I0409 20:22:54.043599 15472 sgd_solver.cpp:105] Iteration 2940, lr = 0.00558572
I0409 20:22:58.969498 15472 solver.cpp:218] Iteration 2952 (2.43621 iter/s, 4.92568s/12 iters), loss = 5.14696
I0409 20:22:58.969542 15472 solver.cpp:237]     Train net output #0: loss = 5.14696 (* 1 = 5.14696 loss)
I0409 20:22:58.969552 15472 sgd_solver.cpp:105] Iteration 2952, lr = 0.00557245
I0409 20:23:00.951105 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_2958.caffemodel
I0409 20:23:01.439949 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_2958.solverstate
I0409 20:23:01.770372 15472 solver.cpp:330] Iteration 2958, Testing net (#0)
I0409 20:23:01.770403 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:23:05.018923 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:23:06.197124 15472 solver.cpp:397]     Test net output #0: accuracy = 0.0116422
I0409 20:23:06.197158 15472 solver.cpp:397]     Test net output #1: loss = 5.08547 (* 1 = 5.08547 loss)
I0409 20:23:08.048943 15472 solver.cpp:218] Iteration 2964 (1.32173 iter/s, 9.07898s/12 iters), loss = 5.07453
I0409 20:23:08.048995 15472 solver.cpp:237]     Train net output #0: loss = 5.07453 (* 1 = 5.07453 loss)
I0409 20:23:08.049006 15472 sgd_solver.cpp:105] Iteration 2964, lr = 0.00555922
I0409 20:23:12.934805 15472 solver.cpp:218] Iteration 2976 (2.45621 iter/s, 4.88558s/12 iters), loss = 5.03333
I0409 20:23:12.934929 15472 solver.cpp:237]     Train net output #0: loss = 5.03333 (* 1 = 5.03333 loss)
I0409 20:23:12.934944 15472 sgd_solver.cpp:105] Iteration 2976, lr = 0.00554603
I0409 20:23:17.863230 15472 solver.cpp:218] Iteration 2988 (2.43503 iter/s, 4.92807s/12 iters), loss = 5.08339
I0409 20:23:17.863286 15472 solver.cpp:237]     Train net output #0: loss = 5.08339 (* 1 = 5.08339 loss)
I0409 20:23:17.863298 15472 sgd_solver.cpp:105] Iteration 2988, lr = 0.00553286
I0409 20:23:22.748955 15472 solver.cpp:218] Iteration 3000 (2.45628 iter/s, 4.88544s/12 iters), loss = 5.13679
I0409 20:23:22.749001 15472 solver.cpp:237]     Train net output #0: loss = 5.13679 (* 1 = 5.13679 loss)
I0409 20:23:22.749011 15472 sgd_solver.cpp:105] Iteration 3000, lr = 0.00551972
I0409 20:23:27.649478 15472 solver.cpp:218] Iteration 3012 (2.44886 iter/s, 4.90024s/12 iters), loss = 5.1428
I0409 20:23:27.649536 15472 solver.cpp:237]     Train net output #0: loss = 5.1428 (* 1 = 5.1428 loss)
I0409 20:23:27.649549 15472 sgd_solver.cpp:105] Iteration 3012, lr = 0.00550662
I0409 20:23:32.460618 15472 solver.cpp:218] Iteration 3024 (2.49436 iter/s, 4.81085s/12 iters), loss = 5.1702
I0409 20:23:32.460675 15472 solver.cpp:237]     Train net output #0: loss = 5.1702 (* 1 = 5.1702 loss)
I0409 20:23:32.460688 15472 sgd_solver.cpp:105] Iteration 3024, lr = 0.00549354
I0409 20:23:36.276688 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:23:37.274116 15472 solver.cpp:218] Iteration 3036 (2.49314 iter/s, 4.81321s/12 iters), loss = 5.03197
I0409 20:23:37.274168 15472 solver.cpp:237]     Train net output #0: loss = 5.03197 (* 1 = 5.03197 loss)
I0409 20:23:37.274180 15472 sgd_solver.cpp:105] Iteration 3036, lr = 0.0054805
I0409 20:23:42.101354 15472 solver.cpp:218] Iteration 3048 (2.48604 iter/s, 4.82696s/12 iters), loss = 5.15482
I0409 20:23:42.101408 15472 solver.cpp:237]     Train net output #0: loss = 5.15482 (* 1 = 5.15482 loss)
I0409 20:23:42.101420 15472 sgd_solver.cpp:105] Iteration 3048, lr = 0.00546749
I0409 20:23:46.594851 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_3060.caffemodel
I0409 20:23:47.634677 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_3060.solverstate
I0409 20:23:48.183167 15472 solver.cpp:330] Iteration 3060, Testing net (#0)
I0409 20:23:48.183198 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:23:51.924355 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:23:53.136273 15472 solver.cpp:397]     Test net output #0: accuracy = 0.0183824
I0409 20:23:53.136312 15472 solver.cpp:397]     Test net output #1: loss = 5.07607 (* 1 = 5.07607 loss)
I0409 20:23:53.219358 15472 solver.cpp:218] Iteration 3060 (1.07939 iter/s, 11.1174s/12 iters), loss = 5.15479
I0409 20:23:53.219413 15472 solver.cpp:237]     Train net output #0: loss = 5.15479 (* 1 = 5.15479 loss)
I0409 20:23:53.219424 15472 sgd_solver.cpp:105] Iteration 3060, lr = 0.00545451
I0409 20:23:57.332110 15472 solver.cpp:218] Iteration 3072 (2.91793 iter/s, 4.11251s/12 iters), loss = 5.07056
I0409 20:23:57.332139 15472 solver.cpp:237]     Train net output #0: loss = 5.07056 (* 1 = 5.07056 loss)
I0409 20:23:57.332147 15472 sgd_solver.cpp:105] Iteration 3072, lr = 0.00544156
I0409 20:24:02.455816 15472 solver.cpp:218] Iteration 3084 (2.34218 iter/s, 5.12343s/12 iters), loss = 5.01811
I0409 20:24:02.455873 15472 solver.cpp:237]     Train net output #0: loss = 5.01811 (* 1 = 5.01811 loss)
I0409 20:24:02.455886 15472 sgd_solver.cpp:105] Iteration 3084, lr = 0.00542864
I0409 20:24:07.354601 15472 solver.cpp:218] Iteration 3096 (2.44973 iter/s, 4.8985s/12 iters), loss = 5.12432
I0409 20:24:07.354653 15472 solver.cpp:237]     Train net output #0: loss = 5.12432 (* 1 = 5.12432 loss)
I0409 20:24:07.354665 15472 sgd_solver.cpp:105] Iteration 3096, lr = 0.00541575
I0409 20:24:12.187091 15472 solver.cpp:218] Iteration 3108 (2.48334 iter/s, 4.83221s/12 iters), loss = 5.12327
I0409 20:24:12.187148 15472 solver.cpp:237]     Train net output #0: loss = 5.12327 (* 1 = 5.12327 loss)
I0409 20:24:12.187161 15472 sgd_solver.cpp:105] Iteration 3108, lr = 0.00540289
I0409 20:24:16.992236 15472 solver.cpp:218] Iteration 3120 (2.49747 iter/s, 4.80486s/12 iters), loss = 5.01595
I0409 20:24:16.992369 15472 solver.cpp:237]     Train net output #0: loss = 5.01595 (* 1 = 5.01595 loss)
I0409 20:24:16.992384 15472 sgd_solver.cpp:105] Iteration 3120, lr = 0.00539006
I0409 20:24:21.839869 15472 solver.cpp:218] Iteration 3132 (2.47562 iter/s, 4.84727s/12 iters), loss = 5.10265
I0409 20:24:21.839915 15472 solver.cpp:237]     Train net output #0: loss = 5.10265 (* 1 = 5.10265 loss)
I0409 20:24:21.839924 15472 sgd_solver.cpp:105] Iteration 3132, lr = 0.00537727
I0409 20:24:22.948635 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:24:26.695338 15472 solver.cpp:218] Iteration 3144 (2.47158 iter/s, 4.85519s/12 iters), loss = 5.06887
I0409 20:24:26.695386 15472 solver.cpp:237]     Train net output #0: loss = 5.06887 (* 1 = 5.06887 loss)
I0409 20:24:26.695399 15472 sgd_solver.cpp:105] Iteration 3144, lr = 0.0053645
I0409 20:24:31.517305 15472 solver.cpp:218] Iteration 3156 (2.48875 iter/s, 4.82169s/12 iters), loss = 5.061
I0409 20:24:31.517361 15472 solver.cpp:237]     Train net output #0: loss = 5.061 (* 1 = 5.061 loss)
I0409 20:24:31.517374 15472 sgd_solver.cpp:105] Iteration 3156, lr = 0.00535176
I0409 20:24:33.511556 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_3162.caffemodel
I0409 20:24:34.245474 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_3162.solverstate
I0409 20:24:35.284356 15472 solver.cpp:330] Iteration 3162, Testing net (#0)
I0409 20:24:35.284384 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:24:38.409373 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:24:39.669992 15472 solver.cpp:397]     Test net output #0: accuracy = 0.0147059
I0409 20:24:39.670043 15472 solver.cpp:397]     Test net output #1: loss = 5.05053 (* 1 = 5.05053 loss)
I0409 20:24:41.411748 15472 solver.cpp:218] Iteration 3168 (1.21286 iter/s, 9.89393s/12 iters), loss = 5.17955
I0409 20:24:41.411796 15472 solver.cpp:237]     Train net output #0: loss = 5.17955 (* 1 = 5.17955 loss)
I0409 20:24:41.411808 15472 sgd_solver.cpp:105] Iteration 3168, lr = 0.00533906
I0409 20:24:46.232280 15472 solver.cpp:218] Iteration 3180 (2.4895 iter/s, 4.82025s/12 iters), loss = 5.09617
I0409 20:24:46.232327 15472 solver.cpp:237]     Train net output #0: loss = 5.09617 (* 1 = 5.09617 loss)
I0409 20:24:46.232337 15472 sgd_solver.cpp:105] Iteration 3180, lr = 0.00532638
I0409 20:24:51.037393 15472 solver.cpp:218] Iteration 3192 (2.49748 iter/s, 4.80484s/12 iters), loss = 5.10989
I0409 20:24:51.037554 15472 solver.cpp:237]     Train net output #0: loss = 5.10989 (* 1 = 5.10989 loss)
I0409 20:24:51.037566 15472 sgd_solver.cpp:105] Iteration 3192, lr = 0.00531374
I0409 20:24:55.872772 15472 solver.cpp:218] Iteration 3204 (2.48191 iter/s, 4.83499s/12 iters), loss = 5.14439
I0409 20:24:55.872826 15472 solver.cpp:237]     Train net output #0: loss = 5.14439 (* 1 = 5.14439 loss)
I0409 20:24:55.872838 15472 sgd_solver.cpp:105] Iteration 3204, lr = 0.00530112
I0409 20:25:00.703050 15472 solver.cpp:218] Iteration 3216 (2.48448 iter/s, 4.82999s/12 iters), loss = 5.09261
I0409 20:25:00.703114 15472 solver.cpp:237]     Train net output #0: loss = 5.09261 (* 1 = 5.09261 loss)
I0409 20:25:00.703130 15472 sgd_solver.cpp:105] Iteration 3216, lr = 0.00528853
I0409 20:25:05.563323 15472 solver.cpp:218] Iteration 3228 (2.46914 iter/s, 4.85999s/12 iters), loss = 5.06528
I0409 20:25:05.563372 15472 solver.cpp:237]     Train net output #0: loss = 5.06528 (* 1 = 5.06528 loss)
I0409 20:25:05.563385 15472 sgd_solver.cpp:105] Iteration 3228, lr = 0.00527598
I0409 20:25:08.702147 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:25:10.380816 15472 solver.cpp:218] Iteration 3240 (2.49107 iter/s, 4.81721s/12 iters), loss = 5.08999
I0409 20:25:10.380869 15472 solver.cpp:237]     Train net output #0: loss = 5.08999 (* 1 = 5.08999 loss)
I0409 20:25:10.380882 15472 sgd_solver.cpp:105] Iteration 3240, lr = 0.00526345
I0409 20:25:15.210228 15472 solver.cpp:218] Iteration 3252 (2.48492 iter/s, 4.82913s/12 iters), loss = 5.13831
I0409 20:25:15.210274 15472 solver.cpp:237]     Train net output #0: loss = 5.13831 (* 1 = 5.13831 loss)
I0409 20:25:15.210283 15472 sgd_solver.cpp:105] Iteration 3252, lr = 0.00525095
I0409 20:25:19.592986 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_3264.caffemodel
I0409 20:25:20.086094 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_3264.solverstate
I0409 20:25:20.412498 15472 solver.cpp:330] Iteration 3264, Testing net (#0)
I0409 20:25:20.412520 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:25:23.621817 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:25:25.016374 15472 solver.cpp:397]     Test net output #0: accuracy = 0.0147059
I0409 20:25:25.016405 15472 solver.cpp:397]     Test net output #1: loss = 5.05178 (* 1 = 5.05178 loss)
I0409 20:25:25.097559 15472 solver.cpp:218] Iteration 3264 (1.21374 iter/s, 9.88683s/12 iters), loss = 5.19304
I0409 20:25:25.097599 15472 solver.cpp:237]     Train net output #0: loss = 5.19304 (* 1 = 5.19304 loss)
I0409 20:25:25.097607 15472 sgd_solver.cpp:105] Iteration 3264, lr = 0.00523849
I0409 20:25:29.167856 15472 solver.cpp:218] Iteration 3276 (2.94836 iter/s, 4.07006s/12 iters), loss = 5.01241
I0409 20:25:29.167899 15472 solver.cpp:237]     Train net output #0: loss = 5.01241 (* 1 = 5.01241 loss)
I0409 20:25:29.167909 15472 sgd_solver.cpp:105] Iteration 3276, lr = 0.00522605
I0409 20:25:34.099678 15472 solver.cpp:218] Iteration 3288 (2.43332 iter/s, 4.93154s/12 iters), loss = 4.94854
I0409 20:25:34.099725 15472 solver.cpp:237]     Train net output #0: loss = 4.94854 (* 1 = 4.94854 loss)
I0409 20:25:34.099735 15472 sgd_solver.cpp:105] Iteration 3288, lr = 0.00521364
I0409 20:25:38.998306 15472 solver.cpp:218] Iteration 3300 (2.44981 iter/s, 4.89835s/12 iters), loss = 4.98795
I0409 20:25:38.998351 15472 solver.cpp:237]     Train net output #0: loss = 4.98795 (* 1 = 4.98795 loss)
I0409 20:25:38.998361 15472 sgd_solver.cpp:105] Iteration 3300, lr = 0.00520126
I0409 20:25:43.930068 15472 solver.cpp:218] Iteration 3312 (2.43335 iter/s, 4.93148s/12 iters), loss = 5.03514
I0409 20:25:43.930119 15472 solver.cpp:237]     Train net output #0: loss = 5.03514 (* 1 = 5.03514 loss)
I0409 20:25:43.930132 15472 sgd_solver.cpp:105] Iteration 3312, lr = 0.00518892
I0409 20:25:48.851660 15472 solver.cpp:218] Iteration 3324 (2.43838 iter/s, 4.92131s/12 iters), loss = 5.0855
I0409 20:25:48.851719 15472 solver.cpp:237]     Train net output #0: loss = 5.0855 (* 1 = 5.0855 loss)
I0409 20:25:48.851733 15472 sgd_solver.cpp:105] Iteration 3324, lr = 0.0051766
I0409 20:25:53.807008 15472 solver.cpp:218] Iteration 3336 (2.42177 iter/s, 4.95505s/12 iters), loss = 5.08792
I0409 20:25:53.807166 15472 solver.cpp:237]     Train net output #0: loss = 5.08792 (* 1 = 5.08792 loss)
I0409 20:25:53.807181 15472 sgd_solver.cpp:105] Iteration 3336, lr = 0.00516431
I0409 20:25:54.269796 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:25:58.749403 15472 solver.cpp:218] Iteration 3348 (2.42816 iter/s, 4.942s/12 iters), loss = 5.02384
I0409 20:25:58.749457 15472 solver.cpp:237]     Train net output #0: loss = 5.02384 (* 1 = 5.02384 loss)
I0409 20:25:58.749469 15472 sgd_solver.cpp:105] Iteration 3348, lr = 0.00515204
I0409 20:26:03.654039 15472 solver.cpp:218] Iteration 3360 (2.44681 iter/s, 4.90434s/12 iters), loss = 5.05318
I0409 20:26:03.654101 15472 solver.cpp:237]     Train net output #0: loss = 5.05318 (* 1 = 5.05318 loss)
I0409 20:26:03.654114 15472 sgd_solver.cpp:105] Iteration 3360, lr = 0.00513981
I0409 20:26:05.661384 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_3366.caffemodel
I0409 20:26:06.919708 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_3366.solverstate
I0409 20:26:07.923043 15472 solver.cpp:330] Iteration 3366, Testing net (#0)
I0409 20:26:07.923072 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:26:11.027334 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:26:12.358258 15472 solver.cpp:397]     Test net output #0: accuracy = 0.0153186
I0409 20:26:12.358294 15472 solver.cpp:397]     Test net output #1: loss = 5.01636 (* 1 = 5.01636 loss)
I0409 20:26:14.152009 15472 solver.cpp:218] Iteration 3372 (1.14314 iter/s, 10.4974s/12 iters), loss = 4.93576
I0409 20:26:14.152070 15472 solver.cpp:237]     Train net output #0: loss = 4.93576 (* 1 = 4.93576 loss)
I0409 20:26:14.152081 15472 sgd_solver.cpp:105] Iteration 3372, lr = 0.00512761
I0409 20:26:18.951331 15472 solver.cpp:218] Iteration 3384 (2.5005 iter/s, 4.79903s/12 iters), loss = 4.94995
I0409 20:26:18.951390 15472 solver.cpp:237]     Train net output #0: loss = 4.94995 (* 1 = 4.94995 loss)
I0409 20:26:18.951402 15472 sgd_solver.cpp:105] Iteration 3384, lr = 0.00511544
I0409 20:26:23.761415 15472 solver.cpp:218] Iteration 3396 (2.49491 iter/s, 4.8098s/12 iters), loss = 5.05106
I0409 20:26:23.761471 15472 solver.cpp:237]     Train net output #0: loss = 5.05106 (* 1 = 5.05106 loss)
I0409 20:26:23.761483 15472 sgd_solver.cpp:105] Iteration 3396, lr = 0.00510329
I0409 20:26:28.555215 15472 solver.cpp:218] Iteration 3408 (2.50338 iter/s, 4.79351s/12 iters), loss = 4.88615
I0409 20:26:28.555328 15472 solver.cpp:237]     Train net output #0: loss = 4.88615 (* 1 = 4.88615 loss)
I0409 20:26:28.555341 15472 sgd_solver.cpp:105] Iteration 3408, lr = 0.00509117
I0409 20:26:33.349151 15472 solver.cpp:218] Iteration 3420 (2.50334 iter/s, 4.79359s/12 iters), loss = 4.94937
I0409 20:26:33.349213 15472 solver.cpp:237]     Train net output #0: loss = 4.94937 (* 1 = 4.94937 loss)
I0409 20:26:33.349225 15472 sgd_solver.cpp:105] Iteration 3420, lr = 0.00507909
I0409 20:26:38.223192 15472 solver.cpp:218] Iteration 3432 (2.46217 iter/s, 4.87375s/12 iters), loss = 4.979
I0409 20:26:38.223240 15472 solver.cpp:237]     Train net output #0: loss = 4.979 (* 1 = 4.979 loss)
I0409 20:26:38.223249 15472 sgd_solver.cpp:105] Iteration 3432, lr = 0.00506703
I0409 20:26:40.748322 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:26:43.090986 15472 solver.cpp:218] Iteration 3444 (2.46533 iter/s, 4.86751s/12 iters), loss = 4.93619
I0409 20:26:43.091040 15472 solver.cpp:237]     Train net output #0: loss = 4.93619 (* 1 = 4.93619 loss)
I0409 20:26:43.091051 15472 sgd_solver.cpp:105] Iteration 3444, lr = 0.005055
I0409 20:26:47.892853 15472 solver.cpp:218] Iteration 3456 (2.49917 iter/s, 4.80159s/12 iters), loss = 4.97306
I0409 20:26:47.892900 15472 solver.cpp:237]     Train net output #0: loss = 4.97306 (* 1 = 4.97306 loss)
I0409 20:26:47.892910 15472 sgd_solver.cpp:105] Iteration 3456, lr = 0.005043
I0409 20:26:52.226897 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_3468.caffemodel
I0409 20:26:52.683333 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_3468.solverstate
I0409 20:26:52.994180 15472 solver.cpp:330] Iteration 3468, Testing net (#0)
I0409 20:26:52.994199 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:26:53.109759 15472 blocking_queue.cpp:49] Waiting for data
I0409 20:26:55.922654 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:26:57.301327 15472 solver.cpp:397]     Test net output #0: accuracy = 0.0196078
I0409 20:26:57.301374 15472 solver.cpp:397]     Test net output #1: loss = 4.932 (* 1 = 4.932 loss)
I0409 20:26:57.384886 15472 solver.cpp:218] Iteration 3468 (1.26428 iter/s, 9.49155s/12 iters), loss = 4.87869
I0409 20:26:57.384935 15472 solver.cpp:237]     Train net output #0: loss = 4.87869 (* 1 = 4.87869 loss)
I0409 20:26:57.384946 15472 sgd_solver.cpp:105] Iteration 3468, lr = 0.00503102
I0409 20:27:01.456987 15472 solver.cpp:218] Iteration 3480 (2.94706 iter/s, 4.07185s/12 iters), loss = 4.95256
I0409 20:27:01.457118 15472 solver.cpp:237]     Train net output #0: loss = 4.95256 (* 1 = 4.95256 loss)
I0409 20:27:01.457129 15472 sgd_solver.cpp:105] Iteration 3480, lr = 0.00501908
I0409 20:27:06.371345 15472 solver.cpp:218] Iteration 3492 (2.44201 iter/s, 4.91399s/12 iters), loss = 4.9911
I0409 20:27:06.371392 15472 solver.cpp:237]     Train net output #0: loss = 4.9911 (* 1 = 4.9911 loss)
I0409 20:27:06.371407 15472 sgd_solver.cpp:105] Iteration 3492, lr = 0.00500716
I0409 20:27:11.350872 15472 solver.cpp:218] Iteration 3504 (2.41 iter/s, 4.97925s/12 iters), loss = 4.90612
I0409 20:27:11.350919 15472 solver.cpp:237]     Train net output #0: loss = 4.90612 (* 1 = 4.90612 loss)
I0409 20:27:11.350927 15472 sgd_solver.cpp:105] Iteration 3504, lr = 0.00499527
I0409 20:27:16.184401 15472 solver.cpp:218] Iteration 3516 (2.48281 iter/s, 4.83324s/12 iters), loss = 4.84117
I0409 20:27:16.184474 15472 solver.cpp:237]     Train net output #0: loss = 4.84117 (* 1 = 4.84117 loss)
I0409 20:27:16.184490 15472 sgd_solver.cpp:105] Iteration 3516, lr = 0.00498341
I0409 20:27:21.020938 15472 solver.cpp:218] Iteration 3528 (2.48127 iter/s, 4.83624s/12 iters), loss = 4.88834
I0409 20:27:21.020982 15472 solver.cpp:237]     Train net output #0: loss = 4.88834 (* 1 = 4.88834 loss)
I0409 20:27:21.020989 15472 sgd_solver.cpp:105] Iteration 3528, lr = 0.00497158
I0409 20:27:25.608845 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:27:25.868371 15472 solver.cpp:218] Iteration 3540 (2.47568 iter/s, 4.84715s/12 iters), loss = 4.92886
I0409 20:27:25.868432 15472 solver.cpp:237]     Train net output #0: loss = 4.92886 (* 1 = 4.92886 loss)
I0409 20:27:25.868448 15472 sgd_solver.cpp:105] Iteration 3540, lr = 0.00495978
I0409 20:27:30.759052 15472 solver.cpp:218] Iteration 3552 (2.45379 iter/s, 4.89039s/12 iters), loss = 4.98001
I0409 20:27:30.759094 15472 solver.cpp:237]     Train net output #0: loss = 4.98001 (* 1 = 4.98001 loss)
I0409 20:27:30.759104 15472 sgd_solver.cpp:105] Iteration 3552, lr = 0.004948
I0409 20:27:35.637192 15472 solver.cpp:218] Iteration 3564 (2.46009 iter/s, 4.87786s/12 iters), loss = 4.88886
I0409 20:27:35.637372 15472 solver.cpp:237]     Train net output #0: loss = 4.88886 (* 1 = 4.88886 loss)
I0409 20:27:35.637383 15472 sgd_solver.cpp:105] Iteration 3564, lr = 0.00493626
I0409 20:27:37.636075 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_3570.caffemodel
I0409 20:27:38.121562 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_3570.solverstate
I0409 20:27:38.446568 15472 solver.cpp:330] Iteration 3570, Testing net (#0)
I0409 20:27:38.446588 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:27:42.033605 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:27:43.442786 15472 solver.cpp:397]     Test net output #0: accuracy = 0.0110294
I0409 20:27:43.442826 15472 solver.cpp:397]     Test net output #1: loss = 4.92814 (* 1 = 4.92814 loss)
I0409 20:27:45.246421 15472 solver.cpp:218] Iteration 3576 (1.24888 iter/s, 9.60861s/12 iters), loss = 4.93116
I0409 20:27:45.246469 15472 solver.cpp:237]     Train net output #0: loss = 4.93116 (* 1 = 4.93116 loss)
I0409 20:27:45.246477 15472 sgd_solver.cpp:105] Iteration 3576, lr = 0.00492454
I0409 20:27:50.086553 15472 solver.cpp:218] Iteration 3588 (2.47941 iter/s, 4.83985s/12 iters), loss = 4.97491
I0409 20:27:50.086599 15472 solver.cpp:237]     Train net output #0: loss = 4.97491 (* 1 = 4.97491 loss)
I0409 20:27:50.086609 15472 sgd_solver.cpp:105] Iteration 3588, lr = 0.00491284
I0409 20:27:54.909525 15472 solver.cpp:218] Iteration 3600 (2.48823 iter/s, 4.8227s/12 iters), loss = 4.92112
I0409 20:27:54.909569 15472 solver.cpp:237]     Train net output #0: loss = 4.92112 (* 1 = 4.92112 loss)
I0409 20:27:54.909577 15472 sgd_solver.cpp:105] Iteration 3600, lr = 0.00490118
I0409 20:27:59.948120 15472 solver.cpp:218] Iteration 3612 (2.38175 iter/s, 5.03831s/12 iters), loss = 4.88366
I0409 20:27:59.948168 15472 solver.cpp:237]     Train net output #0: loss = 4.88366 (* 1 = 4.88366 loss)
I0409 20:27:59.948180 15472 sgd_solver.cpp:105] Iteration 3612, lr = 0.00488954
I0409 20:28:04.997871 15472 solver.cpp:218] Iteration 3624 (2.37649 iter/s, 5.04946s/12 iters), loss = 4.90454
I0409 20:28:04.997927 15472 solver.cpp:237]     Train net output #0: loss = 4.90454 (* 1 = 4.90454 loss)
I0409 20:28:04.997939 15472 sgd_solver.cpp:105] Iteration 3624, lr = 0.00487793
I0409 20:28:09.897622 15472 solver.cpp:218] Iteration 3636 (2.44925 iter/s, 4.89946s/12 iters), loss = 5.0622
I0409 20:28:09.897759 15472 solver.cpp:237]     Train net output #0: loss = 5.0622 (* 1 = 5.0622 loss)
I0409 20:28:09.897773 15472 sgd_solver.cpp:105] Iteration 3636, lr = 0.00486635
I0409 20:28:11.743878 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:28:14.823814 15472 solver.cpp:218] Iteration 3648 (2.43614 iter/s, 4.92583s/12 iters), loss = 4.90291
I0409 20:28:14.823859 15472 solver.cpp:237]     Train net output #0: loss = 4.90291 (* 1 = 4.90291 loss)
I0409 20:28:14.823868 15472 sgd_solver.cpp:105] Iteration 3648, lr = 0.0048548
I0409 20:28:19.718307 15472 solver.cpp:218] Iteration 3660 (2.45187 iter/s, 4.89422s/12 iters), loss = 5.00038
I0409 20:28:19.718359 15472 solver.cpp:237]     Train net output #0: loss = 5.00038 (* 1 = 5.00038 loss)
I0409 20:28:19.718372 15472 sgd_solver.cpp:105] Iteration 3660, lr = 0.00484327
I0409 20:28:24.163822 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_3672.caffemodel
I0409 20:28:24.626127 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_3672.solverstate
I0409 20:28:24.982844 15472 solver.cpp:330] Iteration 3672, Testing net (#0)
I0409 20:28:24.982864 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:28:27.913796 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:28:29.370781 15472 solver.cpp:397]     Test net output #0: accuracy = 0.0189951
I0409 20:28:29.370829 15472 solver.cpp:397]     Test net output #1: loss = 4.85291 (* 1 = 4.85291 loss)
I0409 20:28:29.454067 15472 solver.cpp:218] Iteration 3672 (1.23263 iter/s, 9.73526s/12 iters), loss = 4.7539
I0409 20:28:29.454142 15472 solver.cpp:237]     Train net output #0: loss = 4.7539 (* 1 = 4.7539 loss)
I0409 20:28:29.454159 15472 sgd_solver.cpp:105] Iteration 3672, lr = 0.00483177
I0409 20:28:33.518947 15472 solver.cpp:218] Iteration 3684 (2.95231 iter/s, 4.06461s/12 iters), loss = 4.90469
I0409 20:28:33.518996 15472 solver.cpp:237]     Train net output #0: loss = 4.90469 (* 1 = 4.90469 loss)
I0409 20:28:33.519008 15472 sgd_solver.cpp:105] Iteration 3684, lr = 0.0048203
I0409 20:28:38.493098 15472 solver.cpp:218] Iteration 3696 (2.41261 iter/s, 4.97387s/12 iters), loss = 4.79205
I0409 20:28:38.493144 15472 solver.cpp:237]     Train net output #0: loss = 4.79205 (* 1 = 4.79205 loss)
I0409 20:28:38.493155 15472 sgd_solver.cpp:105] Iteration 3696, lr = 0.00480886
I0409 20:28:43.481647 15472 solver.cpp:218] Iteration 3708 (2.40565 iter/s, 4.98826s/12 iters), loss = 4.83694
I0409 20:28:43.481791 15472 solver.cpp:237]     Train net output #0: loss = 4.83694 (* 1 = 4.83694 loss)
I0409 20:28:43.481804 15472 sgd_solver.cpp:105] Iteration 3708, lr = 0.00479744
I0409 20:28:48.320134 15472 solver.cpp:218] Iteration 3720 (2.4803 iter/s, 4.83812s/12 iters), loss = 4.87388
I0409 20:28:48.320185 15472 solver.cpp:237]     Train net output #0: loss = 4.87388 (* 1 = 4.87388 loss)
I0409 20:28:48.320195 15472 sgd_solver.cpp:105] Iteration 3720, lr = 0.00478605
I0409 20:28:53.151937 15472 solver.cpp:218] Iteration 3732 (2.48369 iter/s, 4.83153s/12 iters), loss = 4.85563
I0409 20:28:53.151984 15472 solver.cpp:237]     Train net output #0: loss = 4.85563 (* 1 = 4.85563 loss)
I0409 20:28:53.151994 15472 sgd_solver.cpp:105] Iteration 3732, lr = 0.00477469
I0409 20:28:57.006768 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:28:57.983268 15472 solver.cpp:218] Iteration 3744 (2.48393 iter/s, 4.83106s/12 iters), loss = 4.84678
I0409 20:28:57.983314 15472 solver.cpp:237]     Train net output #0: loss = 4.84678 (* 1 = 4.84678 loss)
I0409 20:28:57.983325 15472 sgd_solver.cpp:105] Iteration 3744, lr = 0.00476335
I0409 20:29:02.804024 15472 solver.cpp:218] Iteration 3756 (2.48938 iter/s, 4.82048s/12 iters), loss = 4.90946
I0409 20:29:02.804075 15472 solver.cpp:237]     Train net output #0: loss = 4.90946 (* 1 = 4.90946 loss)
I0409 20:29:02.804087 15472 sgd_solver.cpp:105] Iteration 3756, lr = 0.00475204
I0409 20:29:07.643687 15472 solver.cpp:218] Iteration 3768 (2.47966 iter/s, 4.83938s/12 iters), loss = 4.8037
I0409 20:29:07.643740 15472 solver.cpp:237]     Train net output #0: loss = 4.8037 (* 1 = 4.8037 loss)
I0409 20:29:07.643754 15472 sgd_solver.cpp:105] Iteration 3768, lr = 0.00474076
I0409 20:29:09.583133 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_3774.caffemodel
I0409 20:29:10.049083 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_3774.solverstate
I0409 20:29:10.366328 15472 solver.cpp:330] Iteration 3774, Testing net (#0)
I0409 20:29:10.366348 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:29:13.362452 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:29:14.891919 15472 solver.cpp:397]     Test net output #0: accuracy = 0.0177696
I0409 20:29:14.892042 15472 solver.cpp:397]     Test net output #1: loss = 4.81419 (* 1 = 4.81419 loss)
I0409 20:29:16.626562 15472 solver.cpp:218] Iteration 3780 (1.33594 iter/s, 8.98241s/12 iters), loss = 4.83997
I0409 20:29:16.626617 15472 solver.cpp:237]     Train net output #0: loss = 4.83997 (* 1 = 4.83997 loss)
I0409 20:29:16.626631 15472 sgd_solver.cpp:105] Iteration 3780, lr = 0.00472951
I0409 20:29:21.497360 15472 solver.cpp:218] Iteration 3792 (2.46381 iter/s, 4.87051s/12 iters), loss = 4.8391
I0409 20:29:21.497402 15472 solver.cpp:237]     Train net output #0: loss = 4.8391 (* 1 = 4.8391 loss)
I0409 20:29:21.497411 15472 sgd_solver.cpp:105] Iteration 3792, lr = 0.00471828
I0409 20:29:26.296427 15472 solver.cpp:218] Iteration 3804 (2.50063 iter/s, 4.79879s/12 iters), loss = 4.9267
I0409 20:29:26.296471 15472 solver.cpp:237]     Train net output #0: loss = 4.9267 (* 1 = 4.9267 loss)
I0409 20:29:26.296481 15472 sgd_solver.cpp:105] Iteration 3804, lr = 0.00470707
I0409 20:29:31.203454 15472 solver.cpp:218] Iteration 3816 (2.44561 iter/s, 4.90675s/12 iters), loss = 4.88522
I0409 20:29:31.203500 15472 solver.cpp:237]     Train net output #0: loss = 4.88522 (* 1 = 4.88522 loss)
I0409 20:29:31.203511 15472 sgd_solver.cpp:105] Iteration 3816, lr = 0.0046959
I0409 20:29:36.029477 15472 solver.cpp:218] Iteration 3828 (2.48666 iter/s, 4.82575s/12 iters), loss = 4.76183
I0409 20:29:36.029518 15472 solver.cpp:237]     Train net output #0: loss = 4.76183 (* 1 = 4.76183 loss)
I0409 20:29:36.029527 15472 sgd_solver.cpp:105] Iteration 3828, lr = 0.00468475
I0409 20:29:40.851550 15472 solver.cpp:218] Iteration 3840 (2.4887 iter/s, 4.8218s/12 iters), loss = 4.94007
I0409 20:29:40.851608 15472 solver.cpp:237]     Train net output #0: loss = 4.94007 (* 1 = 4.94007 loss)
I0409 20:29:40.851620 15472 sgd_solver.cpp:105] Iteration 3840, lr = 0.00467363
I0409 20:29:41.979341 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:29:45.912441 15472 solver.cpp:218] Iteration 3852 (2.37126 iter/s, 5.06059s/12 iters), loss = 4.64241
I0409 20:29:45.912586 15472 solver.cpp:237]     Train net output #0: loss = 4.64241 (* 1 = 4.64241 loss)
I0409 20:29:45.912600 15472 sgd_solver.cpp:105] Iteration 3852, lr = 0.00466253
I0409 20:29:50.739837 15472 solver.cpp:218] Iteration 3864 (2.486 iter/s, 4.82702s/12 iters), loss = 4.70866
I0409 20:29:50.739887 15472 solver.cpp:237]     Train net output #0: loss = 4.70866 (* 1 = 4.70866 loss)
I0409 20:29:50.739900 15472 sgd_solver.cpp:105] Iteration 3864, lr = 0.00465146
I0409 20:29:55.164978 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_3876.caffemodel
I0409 20:29:55.623895 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_3876.solverstate
I0409 20:29:55.947700 15472 solver.cpp:330] Iteration 3876, Testing net (#0)
I0409 20:29:55.947729 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:29:58.827289 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:30:00.365209 15472 solver.cpp:397]     Test net output #0: accuracy = 0.0220588
I0409 20:30:00.365254 15472 solver.cpp:397]     Test net output #1: loss = 4.74453 (* 1 = 4.74453 loss)
I0409 20:30:00.448597 15472 solver.cpp:218] Iteration 3876 (1.23606 iter/s, 9.70826s/12 iters), loss = 4.83701
I0409 20:30:00.448644 15472 solver.cpp:237]     Train net output #0: loss = 4.83701 (* 1 = 4.83701 loss)
I0409 20:30:00.448655 15472 sgd_solver.cpp:105] Iteration 3876, lr = 0.00464042
I0409 20:30:04.679267 15472 solver.cpp:218] Iteration 3888 (2.8366 iter/s, 4.23042s/12 iters), loss = 4.77264
I0409 20:30:04.679318 15472 solver.cpp:237]     Train net output #0: loss = 4.77264 (* 1 = 4.77264 loss)
I0409 20:30:04.679330 15472 sgd_solver.cpp:105] Iteration 3888, lr = 0.0046294
I0409 20:30:09.508175 15472 solver.cpp:218] Iteration 3900 (2.48518 iter/s, 4.82863s/12 iters), loss = 4.85974
I0409 20:30:09.508231 15472 solver.cpp:237]     Train net output #0: loss = 4.85974 (* 1 = 4.85974 loss)
I0409 20:30:09.508244 15472 sgd_solver.cpp:105] Iteration 3900, lr = 0.00461841
I0409 20:30:14.377540 15472 solver.cpp:218] Iteration 3912 (2.46453 iter/s, 4.86908s/12 iters), loss = 4.89907
I0409 20:30:14.377581 15472 solver.cpp:237]     Train net output #0: loss = 4.89907 (* 1 = 4.89907 loss)
I0409 20:30:14.377590 15472 sgd_solver.cpp:105] Iteration 3912, lr = 0.00460744
I0409 20:30:19.216392 15472 solver.cpp:218] Iteration 3924 (2.48007 iter/s, 4.83858s/12 iters), loss = 4.80373
I0409 20:30:19.216481 15472 solver.cpp:237]     Train net output #0: loss = 4.80373 (* 1 = 4.80373 loss)
I0409 20:30:19.216496 15472 sgd_solver.cpp:105] Iteration 3924, lr = 0.0045965
I0409 20:30:24.093642 15472 solver.cpp:218] Iteration 3936 (2.46056 iter/s, 4.87694s/12 iters), loss = 4.73638
I0409 20:30:24.093695 15472 solver.cpp:237]     Train net output #0: loss = 4.73638 (* 1 = 4.73638 loss)
I0409 20:30:24.093708 15472 sgd_solver.cpp:105] Iteration 3936, lr = 0.00458559
I0409 20:30:27.362056 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:30:28.943481 15472 solver.cpp:218] Iteration 3948 (2.47445 iter/s, 4.84956s/12 iters), loss = 4.87618
I0409 20:30:28.943534 15472 solver.cpp:237]     Train net output #0: loss = 4.87618 (* 1 = 4.87618 loss)
I0409 20:30:28.943547 15472 sgd_solver.cpp:105] Iteration 3948, lr = 0.0045747
I0409 20:30:33.745836 15472 solver.cpp:218] Iteration 3960 (2.49892 iter/s, 4.80207s/12 iters), loss = 4.89762
I0409 20:30:33.745887 15472 solver.cpp:237]     Train net output #0: loss = 4.89762 (* 1 = 4.89762 loss)
I0409 20:30:33.745895 15472 sgd_solver.cpp:105] Iteration 3960, lr = 0.00456384
I0409 20:30:38.621052 15472 solver.cpp:218] Iteration 3972 (2.46157 iter/s, 4.87494s/12 iters), loss = 4.92153
I0409 20:30:38.621091 15472 solver.cpp:237]     Train net output #0: loss = 4.92153 (* 1 = 4.92153 loss)
I0409 20:30:38.621099 15472 sgd_solver.cpp:105] Iteration 3972, lr = 0.00455301
I0409 20:30:40.583112 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_3978.caffemodel
I0409 20:30:41.381841 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_3978.solverstate
I0409 20:30:42.109138 15472 solver.cpp:330] Iteration 3978, Testing net (#0)
I0409 20:30:42.109165 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:30:44.853457 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:30:46.428215 15472 solver.cpp:397]     Test net output #0: accuracy = 0.0226716
I0409 20:30:46.428262 15472 solver.cpp:397]     Test net output #1: loss = 4.70279 (* 1 = 4.70279 loss)
I0409 20:30:48.218349 15472 solver.cpp:218] Iteration 3984 (1.25042 iter/s, 9.59681s/12 iters), loss = 4.79154
I0409 20:30:48.218405 15472 solver.cpp:237]     Train net output #0: loss = 4.79154 (* 1 = 4.79154 loss)
I0409 20:30:48.218416 15472 sgd_solver.cpp:105] Iteration 3984, lr = 0.0045422
I0409 20:30:53.017980 15472 solver.cpp:218] Iteration 3996 (2.50035 iter/s, 4.79933s/12 iters), loss = 4.78865
I0409 20:30:53.018113 15472 solver.cpp:237]     Train net output #0: loss = 4.78865 (* 1 = 4.78865 loss)
I0409 20:30:53.018123 15472 sgd_solver.cpp:105] Iteration 3996, lr = 0.00453141
I0409 20:30:57.973673 15472 solver.cpp:218] Iteration 4008 (2.42163 iter/s, 4.95533s/12 iters), loss = 4.72532
I0409 20:30:57.973714 15472 solver.cpp:237]     Train net output #0: loss = 4.72532 (* 1 = 4.72532 loss)
I0409 20:30:57.973724 15472 sgd_solver.cpp:105] Iteration 4008, lr = 0.00452066
I0409 20:31:02.829823 15472 solver.cpp:218] Iteration 4020 (2.47123 iter/s, 4.85587s/12 iters), loss = 4.74143
I0409 20:31:02.829880 15472 solver.cpp:237]     Train net output #0: loss = 4.74143 (* 1 = 4.74143 loss)
I0409 20:31:02.829892 15472 sgd_solver.cpp:105] Iteration 4020, lr = 0.00450992
I0409 20:31:07.757241 15472 solver.cpp:218] Iteration 4032 (2.4355 iter/s, 4.92713s/12 iters), loss = 4.85256
I0409 20:31:07.757284 15472 solver.cpp:237]     Train net output #0: loss = 4.85256 (* 1 = 4.85256 loss)
I0409 20:31:07.757294 15472 sgd_solver.cpp:105] Iteration 4032, lr = 0.00449921
I0409 20:31:12.855787 15472 solver.cpp:218] Iteration 4044 (2.35375 iter/s, 5.09826s/12 iters), loss = 4.79927
I0409 20:31:12.855844 15472 solver.cpp:237]     Train net output #0: loss = 4.79927 (* 1 = 4.79927 loss)
I0409 20:31:12.855857 15472 sgd_solver.cpp:105] Iteration 4044, lr = 0.00448853
I0409 20:31:13.363425 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:31:17.764071 15472 solver.cpp:218] Iteration 4056 (2.44499 iter/s, 4.908s/12 iters), loss = 4.77344
I0409 20:31:17.764125 15472 solver.cpp:237]     Train net output #0: loss = 4.77344 (* 1 = 4.77344 loss)
I0409 20:31:17.764138 15472 sgd_solver.cpp:105] Iteration 4056, lr = 0.00447788
I0409 20:31:22.642532 15472 solver.cpp:218] Iteration 4068 (2.45994 iter/s, 4.87817s/12 iters), loss = 4.8226
I0409 20:31:22.642592 15472 solver.cpp:237]     Train net output #0: loss = 4.8226 (* 1 = 4.8226 loss)
I0409 20:31:22.642606 15472 sgd_solver.cpp:105] Iteration 4068, lr = 0.00446724
I0409 20:31:27.008128 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_4080.caffemodel
I0409 20:31:27.486290 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_4080.solverstate
I0409 20:31:27.815604 15472 solver.cpp:330] Iteration 4080, Testing net (#0)
I0409 20:31:27.815629 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:31:30.747817 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:31:32.359920 15472 solver.cpp:397]     Test net output #0: accuracy = 0.0245098
I0409 20:31:32.359966 15472 solver.cpp:397]     Test net output #1: loss = 4.69797 (* 1 = 4.69797 loss)
I0409 20:31:32.443346 15472 solver.cpp:218] Iteration 4080 (1.22445 iter/s, 9.8003s/12 iters), loss = 4.7349
I0409 20:31:32.443405 15472 solver.cpp:237]     Train net output #0: loss = 4.7349 (* 1 = 4.7349 loss)
I0409 20:31:32.443419 15472 sgd_solver.cpp:105] Iteration 4080, lr = 0.00445664
I0409 20:31:36.578964 15472 solver.cpp:218] Iteration 4092 (2.9018 iter/s, 4.13536s/12 iters), loss = 4.59259
I0409 20:31:36.579010 15472 solver.cpp:237]     Train net output #0: loss = 4.59259 (* 1 = 4.59259 loss)
I0409 20:31:36.579020 15472 sgd_solver.cpp:105] Iteration 4092, lr = 0.00444606
I0409 20:31:41.460749 15472 solver.cpp:218] Iteration 4104 (2.45826 iter/s, 4.88151s/12 iters), loss = 4.7417
I0409 20:31:41.460796 15472 solver.cpp:237]     Train net output #0: loss = 4.7417 (* 1 = 4.7417 loss)
I0409 20:31:41.460808 15472 sgd_solver.cpp:105] Iteration 4104, lr = 0.0044355
I0409 20:31:46.277000 15472 solver.cpp:218] Iteration 4116 (2.49171 iter/s, 4.81598s/12 iters), loss = 4.62049
I0409 20:31:46.277050 15472 solver.cpp:237]     Train net output #0: loss = 4.62049 (* 1 = 4.62049 loss)
I0409 20:31:46.277062 15472 sgd_solver.cpp:105] Iteration 4116, lr = 0.00442497
I0409 20:31:51.133020 15472 solver.cpp:218] Iteration 4128 (2.4713 iter/s, 4.85574s/12 iters), loss = 4.77436
I0409 20:31:51.133078 15472 solver.cpp:237]     Train net output #0: loss = 4.77436 (* 1 = 4.77436 loss)
I0409 20:31:51.133091 15472 sgd_solver.cpp:105] Iteration 4128, lr = 0.00441447
I0409 20:31:55.949833 15472 solver.cpp:218] Iteration 4140 (2.49142 iter/s, 4.81653s/12 iters), loss = 4.73368
I0409 20:31:55.949874 15472 solver.cpp:237]     Train net output #0: loss = 4.73368 (* 1 = 4.73368 loss)
I0409 20:31:55.949883 15472 sgd_solver.cpp:105] Iteration 4140, lr = 0.00440398
I0409 20:31:58.482065 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:32:00.729579 15472 solver.cpp:218] Iteration 4152 (2.51074 iter/s, 4.77946s/12 iters), loss = 4.58737
I0409 20:32:00.729650 15472 solver.cpp:237]     Train net output #0: loss = 4.58737 (* 1 = 4.58737 loss)
I0409 20:32:00.729665 15472 sgd_solver.cpp:105] Iteration 4152, lr = 0.00439353
I0409 20:32:01.095403 15472 blocking_queue.cpp:49] Waiting for data
I0409 20:32:05.605376 15472 solver.cpp:218] Iteration 4164 (2.46129 iter/s, 4.8755s/12 iters), loss = 4.72057
I0409 20:32:05.605432 15472 solver.cpp:237]     Train net output #0: loss = 4.72057 (* 1 = 4.72057 loss)
I0409 20:32:05.605443 15472 sgd_solver.cpp:105] Iteration 4164, lr = 0.0043831
I0409 20:32:10.419876 15472 solver.cpp:218] Iteration 4176 (2.49262 iter/s, 4.81421s/12 iters), loss = 4.53993
I0409 20:32:10.419935 15472 solver.cpp:237]     Train net output #0: loss = 4.53993 (* 1 = 4.53993 loss)
I0409 20:32:10.419948 15472 sgd_solver.cpp:105] Iteration 4176, lr = 0.00437269
I0409 20:32:12.380262 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_4182.caffemodel
I0409 20:32:12.834887 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_4182.solverstate
I0409 20:32:13.166348 15472 solver.cpp:330] Iteration 4182, Testing net (#0)
I0409 20:32:13.166376 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:32:15.930532 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:32:17.620255 15472 solver.cpp:397]     Test net output #0: accuracy = 0.0208333
I0409 20:32:17.620294 15472 solver.cpp:397]     Test net output #1: loss = 4.63206 (* 1 = 4.63206 loss)
I0409 20:32:19.503798 15472 solver.cpp:218] Iteration 4188 (1.32109 iter/s, 9.08344s/12 iters), loss = 4.58799
I0409 20:32:19.503854 15472 solver.cpp:237]     Train net output #0: loss = 4.58799 (* 1 = 4.58799 loss)
I0409 20:32:19.503867 15472 sgd_solver.cpp:105] Iteration 4188, lr = 0.00436231
I0409 20:32:24.349325 15472 solver.cpp:218] Iteration 4200 (2.47666 iter/s, 4.84524s/12 iters), loss = 4.64997
I0409 20:32:24.349391 15472 solver.cpp:237]     Train net output #0: loss = 4.64997 (* 1 = 4.64997 loss)
I0409 20:32:24.349408 15472 sgd_solver.cpp:105] Iteration 4200, lr = 0.00435195
I0409 20:32:29.179525 15472 solver.cpp:218] Iteration 4212 (2.48452 iter/s, 4.82991s/12 iters), loss = 4.79727
I0409 20:32:29.179682 15472 solver.cpp:237]     Train net output #0: loss = 4.79727 (* 1 = 4.79727 loss)
I0409 20:32:29.179695 15472 sgd_solver.cpp:105] Iteration 4212, lr = 0.00434162
I0409 20:32:34.027505 15472 solver.cpp:218] Iteration 4224 (2.47545 iter/s, 4.8476s/12 iters), loss = 4.59717
I0409 20:32:34.027554 15472 solver.cpp:237]     Train net output #0: loss = 4.59717 (* 1 = 4.59717 loss)
I0409 20:32:34.027565 15472 sgd_solver.cpp:105] Iteration 4224, lr = 0.00433131
I0409 20:32:38.876423 15472 solver.cpp:218] Iteration 4236 (2.47492 iter/s, 4.84864s/12 iters), loss = 4.69331
I0409 20:32:38.876463 15472 solver.cpp:237]     Train net output #0: loss = 4.69331 (* 1 = 4.69331 loss)
I0409 20:32:38.876471 15472 sgd_solver.cpp:105] Iteration 4236, lr = 0.00432103
I0409 20:32:43.479230 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:32:43.700394 15472 solver.cpp:218] Iteration 4248 (2.48772 iter/s, 4.8237s/12 iters), loss = 4.64874
I0409 20:32:43.700439 15472 solver.cpp:237]     Train net output #0: loss = 4.64874 (* 1 = 4.64874 loss)
I0409 20:32:43.700449 15472 sgd_solver.cpp:105] Iteration 4248, lr = 0.00431077
I0409 20:32:48.514034 15472 solver.cpp:218] Iteration 4260 (2.49306 iter/s, 4.81336s/12 iters), loss = 4.59082
I0409 20:32:48.514077 15472 solver.cpp:237]     Train net output #0: loss = 4.59082 (* 1 = 4.59082 loss)
I0409 20:32:48.514086 15472 sgd_solver.cpp:105] Iteration 4260, lr = 0.00430053
I0409 20:32:53.316454 15472 solver.cpp:218] Iteration 4272 (2.49888 iter/s, 4.80215s/12 iters), loss = 4.67175
I0409 20:32:53.316504 15472 solver.cpp:237]     Train net output #0: loss = 4.67175 (* 1 = 4.67175 loss)
I0409 20:32:53.316516 15472 sgd_solver.cpp:105] Iteration 4272, lr = 0.00429032
I0409 20:32:57.661083 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_4284.caffemodel
I0409 20:32:58.144688 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_4284.solverstate
I0409 20:32:58.477645 15472 solver.cpp:330] Iteration 4284, Testing net (#0)
I0409 20:32:58.477674 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:33:01.252331 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:33:02.942047 15472 solver.cpp:397]     Test net output #0: accuracy = 0.0251225
I0409 20:33:02.942095 15472 solver.cpp:397]     Test net output #1: loss = 4.71001 (* 1 = 4.71001 loss)
I0409 20:33:03.025245 15472 solver.cpp:218] Iteration 4284 (1.23606 iter/s, 9.70829s/12 iters), loss = 4.76529
I0409 20:33:03.025295 15472 solver.cpp:237]     Train net output #0: loss = 4.76529 (* 1 = 4.76529 loss)
I0409 20:33:03.025305 15472 sgd_solver.cpp:105] Iteration 4284, lr = 0.00428014
I0409 20:33:07.116515 15472 solver.cpp:218] Iteration 4296 (2.93325 iter/s, 4.09103s/12 iters), loss = 4.773
I0409 20:33:07.116560 15472 solver.cpp:237]     Train net output #0: loss = 4.773 (* 1 = 4.773 loss)
I0409 20:33:07.116569 15472 sgd_solver.cpp:105] Iteration 4296, lr = 0.00426998
I0409 20:33:12.021375 15472 solver.cpp:218] Iteration 4308 (2.4467 iter/s, 4.90458s/12 iters), loss = 4.60561
I0409 20:33:12.021431 15472 solver.cpp:237]     Train net output #0: loss = 4.60561 (* 1 = 4.60561 loss)
I0409 20:33:12.021443 15472 sgd_solver.cpp:105] Iteration 4308, lr = 0.00425984
I0409 20:33:16.901286 15472 solver.cpp:218] Iteration 4320 (2.45921 iter/s, 4.87962s/12 iters), loss = 4.66593
I0409 20:33:16.901337 15472 solver.cpp:237]     Train net output #0: loss = 4.66593 (* 1 = 4.66593 loss)
I0409 20:33:16.901350 15472 sgd_solver.cpp:105] Iteration 4320, lr = 0.00424972
I0409 20:33:21.806725 15472 solver.cpp:218] Iteration 4332 (2.44641 iter/s, 4.90516s/12 iters), loss = 4.54971
I0409 20:33:21.806777 15472 solver.cpp:237]     Train net output #0: loss = 4.54971 (* 1 = 4.54971 loss)
I0409 20:33:21.806789 15472 sgd_solver.cpp:105] Iteration 4332, lr = 0.00423964
I0409 20:33:26.641449 15472 solver.cpp:218] Iteration 4344 (2.48219 iter/s, 4.83444s/12 iters), loss = 4.72805
I0409 20:33:26.641501 15472 solver.cpp:237]     Train net output #0: loss = 4.72805 (* 1 = 4.72805 loss)
I0409 20:33:26.641512 15472 sgd_solver.cpp:105] Iteration 4344, lr = 0.00422957
I0409 20:33:28.514711 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:33:31.554920 15472 solver.cpp:218] Iteration 4356 (2.44241 iter/s, 4.91318s/12 iters), loss = 4.68755
I0409 20:33:31.555081 15472 solver.cpp:237]     Train net output #0: loss = 4.68755 (* 1 = 4.68755 loss)
I0409 20:33:31.555096 15472 sgd_solver.cpp:105] Iteration 4356, lr = 0.00421953
I0409 20:33:36.494865 15472 solver.cpp:218] Iteration 4368 (2.42937 iter/s, 4.93955s/12 iters), loss = 4.70232
I0409 20:33:36.494921 15472 solver.cpp:237]     Train net output #0: loss = 4.70232 (* 1 = 4.70232 loss)
I0409 20:33:36.494935 15472 sgd_solver.cpp:105] Iteration 4368, lr = 0.00420951
I0409 20:33:41.394331 15472 solver.cpp:218] Iteration 4380 (2.44939 iter/s, 4.89918s/12 iters), loss = 4.48195
I0409 20:33:41.394387 15472 solver.cpp:237]     Train net output #0: loss = 4.48195 (* 1 = 4.48195 loss)
I0409 20:33:41.394400 15472 sgd_solver.cpp:105] Iteration 4380, lr = 0.00419952
I0409 20:33:43.388417 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_4386.caffemodel
I0409 20:33:44.450700 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_4386.solverstate
I0409 20:33:44.784621 15472 solver.cpp:330] Iteration 4386, Testing net (#0)
I0409 20:33:44.784651 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:33:47.481428 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:33:49.218089 15472 solver.cpp:397]     Test net output #0: accuracy = 0.0214461
I0409 20:33:49.218133 15472 solver.cpp:397]     Test net output #1: loss = 4.58565 (* 1 = 4.58565 loss)
I0409 20:33:51.045684 15472 solver.cpp:218] Iteration 4392 (1.24341 iter/s, 9.65086s/12 iters), loss = 4.65161
I0409 20:33:51.045724 15472 solver.cpp:237]     Train net output #0: loss = 4.65161 (* 1 = 4.65161 loss)
I0409 20:33:51.045734 15472 sgd_solver.cpp:105] Iteration 4392, lr = 0.00418954
I0409 20:33:55.932476 15472 solver.cpp:218] Iteration 4404 (2.45574 iter/s, 4.88651s/12 iters), loss = 4.62091
I0409 20:33:55.932526 15472 solver.cpp:237]     Train net output #0: loss = 4.62091 (* 1 = 4.62091 loss)
I0409 20:33:55.932538 15472 sgd_solver.cpp:105] Iteration 4404, lr = 0.0041796
I0409 20:34:00.803701 15472 solver.cpp:218] Iteration 4416 (2.46359 iter/s, 4.87094s/12 iters), loss = 4.64989
I0409 20:34:00.803755 15472 solver.cpp:237]     Train net output #0: loss = 4.64989 (* 1 = 4.64989 loss)
I0409 20:34:00.803766 15472 sgd_solver.cpp:105] Iteration 4416, lr = 0.00416967
I0409 20:34:05.693431 15472 solver.cpp:218] Iteration 4428 (2.45427 iter/s, 4.88945s/12 iters), loss = 4.59436
I0409 20:34:05.693521 15472 solver.cpp:237]     Train net output #0: loss = 4.59436 (* 1 = 4.59436 loss)
I0409 20:34:05.693532 15472 sgd_solver.cpp:105] Iteration 4428, lr = 0.00415977
I0409 20:34:10.515265 15472 solver.cpp:218] Iteration 4440 (2.48885 iter/s, 4.82151s/12 iters), loss = 4.62904
I0409 20:34:10.515311 15472 solver.cpp:237]     Train net output #0: loss = 4.62904 (* 1 = 4.62904 loss)
I0409 20:34:10.515321 15472 sgd_solver.cpp:105] Iteration 4440, lr = 0.0041499
I0409 20:34:14.579743 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:34:15.502547 15472 solver.cpp:218] Iteration 4452 (2.40626 iter/s, 4.987s/12 iters), loss = 4.53577
I0409 20:34:15.502593 15472 solver.cpp:237]     Train net output #0: loss = 4.53577 (* 1 = 4.53577 loss)
I0409 20:34:15.502604 15472 sgd_solver.cpp:105] Iteration 4452, lr = 0.00414005
I0409 20:34:20.348968 15472 solver.cpp:218] Iteration 4464 (2.47619 iter/s, 4.84615s/12 iters), loss = 4.67238
I0409 20:34:20.349011 15472 solver.cpp:237]     Train net output #0: loss = 4.67238 (* 1 = 4.67238 loss)
I0409 20:34:20.349020 15472 sgd_solver.cpp:105] Iteration 4464, lr = 0.00413022
I0409 20:34:25.191090 15472 solver.cpp:218] Iteration 4476 (2.47839 iter/s, 4.84185s/12 iters), loss = 4.57699
I0409 20:34:25.191140 15472 solver.cpp:237]     Train net output #0: loss = 4.57699 (* 1 = 4.57699 loss)
I0409 20:34:25.191150 15472 sgd_solver.cpp:105] Iteration 4476, lr = 0.00412041
I0409 20:34:29.604809 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_4488.caffemodel
I0409 20:34:30.095798 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_4488.solverstate
I0409 20:34:30.645458 15472 solver.cpp:330] Iteration 4488, Testing net (#0)
I0409 20:34:30.645486 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:34:33.291774 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:34:35.065409 15472 solver.cpp:397]     Test net output #0: accuracy = 0.0294118
I0409 20:34:35.065446 15472 solver.cpp:397]     Test net output #1: loss = 4.52007 (* 1 = 4.52007 loss)
I0409 20:34:35.148293 15472 solver.cpp:218] Iteration 4488 (1.20522 iter/s, 9.95669s/12 iters), loss = 4.50746
I0409 20:34:35.148340 15472 solver.cpp:237]     Train net output #0: loss = 4.50746 (* 1 = 4.50746 loss)
I0409 20:34:35.148351 15472 sgd_solver.cpp:105] Iteration 4488, lr = 0.00411063
I0409 20:34:39.524658 15472 solver.cpp:218] Iteration 4500 (2.74216 iter/s, 4.37611s/12 iters), loss = 4.55243
I0409 20:34:39.524821 15472 solver.cpp:237]     Train net output #0: loss = 4.55243 (* 1 = 4.55243 loss)
I0409 20:34:39.524834 15472 sgd_solver.cpp:105] Iteration 4500, lr = 0.00410087
I0409 20:34:44.346789 15472 solver.cpp:218] Iteration 4512 (2.48872 iter/s, 4.82175s/12 iters), loss = 4.60697
I0409 20:34:44.346834 15472 solver.cpp:237]     Train net output #0: loss = 4.60697 (* 1 = 4.60697 loss)
I0409 20:34:44.346843 15472 sgd_solver.cpp:105] Iteration 4512, lr = 0.00409113
I0409 20:34:49.186939 15472 solver.cpp:218] Iteration 4524 (2.4794 iter/s, 4.83987s/12 iters), loss = 4.74062
I0409 20:34:49.187005 15472 solver.cpp:237]     Train net output #0: loss = 4.74062 (* 1 = 4.74062 loss)
I0409 20:34:49.187023 15472 sgd_solver.cpp:105] Iteration 4524, lr = 0.00408142
I0409 20:34:54.017251 15472 solver.cpp:218] Iteration 4536 (2.48446 iter/s, 4.83002s/12 iters), loss = 4.56787
I0409 20:34:54.017308 15472 solver.cpp:237]     Train net output #0: loss = 4.56787 (* 1 = 4.56787 loss)
I0409 20:34:54.017321 15472 sgd_solver.cpp:105] Iteration 4536, lr = 0.00407173
I0409 20:34:58.803884 15472 solver.cpp:218] Iteration 4548 (2.50713 iter/s, 4.78635s/12 iters), loss = 4.61398
I0409 20:34:58.803927 15472 solver.cpp:237]     Train net output #0: loss = 4.61398 (* 1 = 4.61398 loss)
I0409 20:34:58.803938 15472 sgd_solver.cpp:105] Iteration 4548, lr = 0.00406206
I0409 20:35:00.021443 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:35:03.639514 15472 solver.cpp:218] Iteration 4560 (2.48172 iter/s, 4.83536s/12 iters), loss = 4.4173
I0409 20:35:03.639556 15472 solver.cpp:237]     Train net output #0: loss = 4.4173 (* 1 = 4.4173 loss)
I0409 20:35:03.639565 15472 sgd_solver.cpp:105] Iteration 4560, lr = 0.00405242
I0409 20:35:08.527963 15472 solver.cpp:218] Iteration 4572 (2.45491 iter/s, 4.88817s/12 iters), loss = 4.45263
I0409 20:35:08.528018 15472 solver.cpp:237]     Train net output #0: loss = 4.45263 (* 1 = 4.45263 loss)
I0409 20:35:08.528031 15472 sgd_solver.cpp:105] Iteration 4572, lr = 0.0040428
I0409 20:35:13.351780 15472 solver.cpp:218] Iteration 4584 (2.4878 iter/s, 4.82353s/12 iters), loss = 4.6604
I0409 20:35:13.351869 15472 solver.cpp:237]     Train net output #0: loss = 4.6604 (* 1 = 4.6604 loss)
I0409 20:35:13.351882 15472 sgd_solver.cpp:105] Iteration 4584, lr = 0.0040332
I0409 20:35:15.304455 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_4590.caffemodel
I0409 20:35:16.410018 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_4590.solverstate
I0409 20:35:17.028132 15472 solver.cpp:330] Iteration 4590, Testing net (#0)
I0409 20:35:17.028165 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:35:19.789237 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:35:21.600312 15472 solver.cpp:397]     Test net output #0: accuracy = 0.033701
I0409 20:35:21.600361 15472 solver.cpp:397]     Test net output #1: loss = 4.50422 (* 1 = 4.50422 loss)
I0409 20:35:23.543586 15472 solver.cpp:218] Iteration 4596 (1.17748 iter/s, 10.1912s/12 iters), loss = 4.475
I0409 20:35:23.543638 15472 solver.cpp:237]     Train net output #0: loss = 4.475 (* 1 = 4.475 loss)
I0409 20:35:23.543650 15472 sgd_solver.cpp:105] Iteration 4596, lr = 0.00402362
I0409 20:35:28.440950 15472 solver.cpp:218] Iteration 4608 (2.45044 iter/s, 4.89707s/12 iters), loss = 4.49234
I0409 20:35:28.441011 15472 solver.cpp:237]     Train net output #0: loss = 4.49234 (* 1 = 4.49234 loss)
I0409 20:35:28.441025 15472 sgd_solver.cpp:105] Iteration 4608, lr = 0.00401407
I0409 20:35:33.237171 15472 solver.cpp:218] Iteration 4620 (2.50212 iter/s, 4.79594s/12 iters), loss = 4.64744
I0409 20:35:33.237218 15472 solver.cpp:237]     Train net output #0: loss = 4.64744 (* 1 = 4.64744 loss)
I0409 20:35:33.237229 15472 sgd_solver.cpp:105] Iteration 4620, lr = 0.00400454
I0409 20:35:38.053408 15472 solver.cpp:218] Iteration 4632 (2.49171 iter/s, 4.81596s/12 iters), loss = 4.41941
I0409 20:35:38.053457 15472 solver.cpp:237]     Train net output #0: loss = 4.41941 (* 1 = 4.41941 loss)
I0409 20:35:38.053467 15472 sgd_solver.cpp:105] Iteration 4632, lr = 0.00399503
I0409 20:35:42.899371 15472 solver.cpp:218] Iteration 4644 (2.47643 iter/s, 4.84568s/12 iters), loss = 4.50904
I0409 20:35:42.899421 15472 solver.cpp:237]     Train net output #0: loss = 4.50904 (* 1 = 4.50904 loss)
I0409 20:35:42.899435 15472 sgd_solver.cpp:105] Iteration 4644, lr = 0.00398555
I0409 20:35:46.153611 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:35:47.704869 15472 solver.cpp:218] Iteration 4656 (2.49728 iter/s, 4.80522s/12 iters), loss = 4.76207
I0409 20:35:47.704913 15472 solver.cpp:237]     Train net output #0: loss = 4.76207 (* 1 = 4.76207 loss)
I0409 20:35:47.704923 15472 sgd_solver.cpp:105] Iteration 4656, lr = 0.00397608
I0409 20:35:52.799605 15472 solver.cpp:218] Iteration 4668 (2.3555 iter/s, 5.09445s/12 iters), loss = 4.54964
I0409 20:35:52.799651 15472 solver.cpp:237]     Train net output #0: loss = 4.54964 (* 1 = 4.54964 loss)
I0409 20:35:52.799661 15472 sgd_solver.cpp:105] Iteration 4668, lr = 0.00396664
I0409 20:35:57.697203 15472 solver.cpp:218] Iteration 4680 (2.45032 iter/s, 4.89732s/12 iters), loss = 4.55845
I0409 20:35:57.697253 15472 solver.cpp:237]     Train net output #0: loss = 4.55845 (* 1 = 4.55845 loss)
I0409 20:35:57.697263 15472 sgd_solver.cpp:105] Iteration 4680, lr = 0.00395723
I0409 20:36:02.088030 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_4692.caffemodel
I0409 20:36:02.951740 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_4692.solverstate
I0409 20:36:03.466934 15472 solver.cpp:330] Iteration 4692, Testing net (#0)
I0409 20:36:03.466956 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:36:06.150851 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:36:08.019588 15472 solver.cpp:397]     Test net output #0: accuracy = 0.0484069
I0409 20:36:08.019639 15472 solver.cpp:397]     Test net output #1: loss = 4.41442 (* 1 = 4.41442 loss)
I0409 20:36:08.102041 15472 solver.cpp:218] Iteration 4692 (1.15337 iter/s, 10.4043s/12 iters), loss = 4.5635
I0409 20:36:08.102099 15472 solver.cpp:237]     Train net output #0: loss = 4.5635 (* 1 = 4.5635 loss)
I0409 20:36:08.102113 15472 sgd_solver.cpp:105] Iteration 4692, lr = 0.00394783
I0409 20:36:12.333793 15472 solver.cpp:218] Iteration 4704 (2.83588 iter/s, 4.2315s/12 iters), loss = 4.38169
I0409 20:36:12.333842 15472 solver.cpp:237]     Train net output #0: loss = 4.38169 (* 1 = 4.38169 loss)
I0409 20:36:12.333853 15472 sgd_solver.cpp:105] Iteration 4704, lr = 0.00393846
I0409 20:36:17.136292 15472 solver.cpp:218] Iteration 4716 (2.49884 iter/s, 4.80223s/12 iters), loss = 4.58653
I0409 20:36:17.136390 15472 solver.cpp:237]     Train net output #0: loss = 4.58653 (* 1 = 4.58653 loss)
I0409 20:36:17.136399 15472 sgd_solver.cpp:105] Iteration 4716, lr = 0.00392911
I0409 20:36:22.117460 15472 solver.cpp:218] Iteration 4728 (2.40923 iter/s, 4.98084s/12 iters), loss = 4.49347
I0409 20:36:22.117512 15472 solver.cpp:237]     Train net output #0: loss = 4.49347 (* 1 = 4.49347 loss)
I0409 20:36:22.117524 15472 sgd_solver.cpp:105] Iteration 4728, lr = 0.00391978
I0409 20:36:26.915472 15472 solver.cpp:218] Iteration 4740 (2.50119 iter/s, 4.79773s/12 iters), loss = 4.55179
I0409 20:36:26.915529 15472 solver.cpp:237]     Train net output #0: loss = 4.55179 (* 1 = 4.55179 loss)
I0409 20:36:26.915544 15472 sgd_solver.cpp:105] Iteration 4740, lr = 0.00391047
I0409 20:36:31.755532 15472 solver.cpp:218] Iteration 4752 (2.47946 iter/s, 4.83977s/12 iters), loss = 4.47694
I0409 20:36:31.755589 15472 solver.cpp:237]     Train net output #0: loss = 4.47694 (* 1 = 4.47694 loss)
I0409 20:36:31.755601 15472 sgd_solver.cpp:105] Iteration 4752, lr = 0.00390119
I0409 20:36:32.272430 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:36:36.603461 15472 solver.cpp:218] Iteration 4764 (2.47543 iter/s, 4.84764s/12 iters), loss = 4.49109
I0409 20:36:36.603513 15472 solver.cpp:237]     Train net output #0: loss = 4.49109 (* 1 = 4.49109 loss)
I0409 20:36:36.603523 15472 sgd_solver.cpp:105] Iteration 4764, lr = 0.00389193
I0409 20:36:41.432864 15472 solver.cpp:218] Iteration 4776 (2.48492 iter/s, 4.82912s/12 iters), loss = 4.39919
I0409 20:36:41.432917 15472 solver.cpp:237]     Train net output #0: loss = 4.39919 (* 1 = 4.39919 loss)
I0409 20:36:41.432929 15472 sgd_solver.cpp:105] Iteration 4776, lr = 0.00388269
I0409 20:36:46.254618 15472 solver.cpp:218] Iteration 4788 (2.48886 iter/s, 4.82148s/12 iters), loss = 4.41023
I0409 20:36:46.254658 15472 solver.cpp:237]     Train net output #0: loss = 4.41023 (* 1 = 4.41023 loss)
I0409 20:36:46.254667 15472 sgd_solver.cpp:105] Iteration 4788, lr = 0.00387347
I0409 20:36:48.211764 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_4794.caffemodel
I0409 20:36:49.419498 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_4794.solverstate
I0409 20:36:50.245862 15472 solver.cpp:330] Iteration 4794, Testing net (#0)
I0409 20:36:50.245880 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:36:52.667727 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:36:54.707973 15472 solver.cpp:397]     Test net output #0: accuracy = 0.0508578
I0409 20:36:54.708007 15472 solver.cpp:397]     Test net output #1: loss = 4.40551 (* 1 = 4.40551 loss)
I0409 20:36:56.441712 15472 solver.cpp:218] Iteration 4800 (1.17802 iter/s, 10.1866s/12 iters), loss = 4.30047
I0409 20:36:56.441771 15472 solver.cpp:237]     Train net output #0: loss = 4.30047 (* 1 = 4.30047 loss)
I0409 20:36:56.441784 15472 sgd_solver.cpp:105] Iteration 4800, lr = 0.00386427
I0409 20:37:01.244073 15472 solver.cpp:218] Iteration 4812 (2.49892 iter/s, 4.80207s/12 iters), loss = 4.35253
I0409 20:37:01.244138 15472 solver.cpp:237]     Train net output #0: loss = 4.35253 (* 1 = 4.35253 loss)
I0409 20:37:01.244153 15472 sgd_solver.cpp:105] Iteration 4812, lr = 0.0038551
I0409 20:37:06.051059 15472 solver.cpp:218] Iteration 4824 (2.49652 iter/s, 4.80669s/12 iters), loss = 4.32754
I0409 20:37:06.051113 15472 solver.cpp:237]     Train net output #0: loss = 4.32754 (* 1 = 4.32754 loss)
I0409 20:37:06.051127 15472 sgd_solver.cpp:105] Iteration 4824, lr = 0.00384594
I0409 20:37:10.957898 15472 solver.cpp:218] Iteration 4836 (2.44571 iter/s, 4.90655s/12 iters), loss = 4.3561
I0409 20:37:10.957978 15472 solver.cpp:237]     Train net output #0: loss = 4.3561 (* 1 = 4.3561 loss)
I0409 20:37:10.957990 15472 sgd_solver.cpp:105] Iteration 4836, lr = 0.00383681
I0409 20:37:11.724088 15472 blocking_queue.cpp:49] Waiting for data
I0409 20:37:15.814740 15472 solver.cpp:218] Iteration 4848 (2.47089 iter/s, 4.85655s/12 iters), loss = 4.44201
I0409 20:37:15.814795 15472 solver.cpp:237]     Train net output #0: loss = 4.44201 (* 1 = 4.44201 loss)
I0409 20:37:15.814805 15472 sgd_solver.cpp:105] Iteration 4848, lr = 0.0038277
I0409 20:37:18.392724 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:37:20.627434 15472 solver.cpp:218] Iteration 4860 (2.49355 iter/s, 4.81241s/12 iters), loss = 4.21141
I0409 20:37:20.627477 15472 solver.cpp:237]     Train net output #0: loss = 4.21141 (* 1 = 4.21141 loss)
I0409 20:37:20.627486 15472 sgd_solver.cpp:105] Iteration 4860, lr = 0.00381862
I0409 20:37:25.485371 15472 solver.cpp:218] Iteration 4872 (2.47033 iter/s, 4.85765s/12 iters), loss = 4.3787
I0409 20:37:25.485424 15472 solver.cpp:237]     Train net output #0: loss = 4.3787 (* 1 = 4.3787 loss)
I0409 20:37:25.485435 15472 sgd_solver.cpp:105] Iteration 4872, lr = 0.00380955
I0409 20:37:30.278865 15472 solver.cpp:218] Iteration 4884 (2.50354 iter/s, 4.79321s/12 iters), loss = 4.3167
I0409 20:37:30.278910 15472 solver.cpp:237]     Train net output #0: loss = 4.3167 (* 1 = 4.3167 loss)
I0409 20:37:30.278920 15472 sgd_solver.cpp:105] Iteration 4884, lr = 0.0038005
I0409 20:37:34.688539 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_4896.caffemodel
I0409 20:37:35.118551 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_4896.solverstate
I0409 20:37:35.431419 15472 solver.cpp:330] Iteration 4896, Testing net (#0)
I0409 20:37:35.431442 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:37:37.912187 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:37:39.878957 15472 solver.cpp:397]     Test net output #0: accuracy = 0.0514706
I0409 20:37:39.879006 15472 solver.cpp:397]     Test net output #1: loss = 4.35668 (* 1 = 4.35668 loss)
I0409 20:37:39.964819 15472 solver.cpp:218] Iteration 4896 (1.23897 iter/s, 9.68546s/12 iters), loss = 4.48475
I0409 20:37:39.964867 15472 solver.cpp:237]     Train net output #0: loss = 4.48475 (* 1 = 4.48475 loss)
I0409 20:37:39.964879 15472 sgd_solver.cpp:105] Iteration 4896, lr = 0.00379148
I0409 20:37:44.062886 15472 solver.cpp:218] Iteration 4908 (2.92839 iter/s, 4.09782s/12 iters), loss = 4.30728
I0409 20:37:44.062942 15472 solver.cpp:237]     Train net output #0: loss = 4.30728 (* 1 = 4.30728 loss)
I0409 20:37:44.062952 15472 sgd_solver.cpp:105] Iteration 4908, lr = 0.00378248
I0409 20:37:48.865990 15472 solver.cpp:218] Iteration 4920 (2.49854 iter/s, 4.80281s/12 iters), loss = 4.28012
I0409 20:37:48.866087 15472 solver.cpp:237]     Train net output #0: loss = 4.28012 (* 1 = 4.28012 loss)
I0409 20:37:48.866097 15472 sgd_solver.cpp:105] Iteration 4920, lr = 0.0037735
I0409 20:37:53.705530 15472 solver.cpp:218] Iteration 4932 (2.47974 iter/s, 4.83921s/12 iters), loss = 4.35292
I0409 20:37:53.705582 15472 solver.cpp:237]     Train net output #0: loss = 4.35292 (* 1 = 4.35292 loss)
I0409 20:37:53.705595 15472 sgd_solver.cpp:105] Iteration 4932, lr = 0.00376454
I0409 20:37:58.514575 15472 solver.cpp:218] Iteration 4944 (2.49544 iter/s, 4.80876s/12 iters), loss = 4.40599
I0409 20:37:58.514627 15472 solver.cpp:237]     Train net output #0: loss = 4.40599 (* 1 = 4.40599 loss)
I0409 20:37:58.514638 15472 sgd_solver.cpp:105] Iteration 4944, lr = 0.0037556
I0409 20:38:03.310047 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:38:03.497262 15472 solver.cpp:218] Iteration 4956 (2.40848 iter/s, 4.9824s/12 iters), loss = 4.26605
I0409 20:38:03.497313 15472 solver.cpp:237]     Train net output #0: loss = 4.26605 (* 1 = 4.26605 loss)
I0409 20:38:03.497324 15472 sgd_solver.cpp:105] Iteration 4956, lr = 0.00374669
I0409 20:38:08.348142 15472 solver.cpp:218] Iteration 4968 (2.47392 iter/s, 4.8506s/12 iters), loss = 4.29269
I0409 20:38:08.348201 15472 solver.cpp:237]     Train net output #0: loss = 4.29269 (* 1 = 4.29269 loss)
I0409 20:38:08.348214 15472 sgd_solver.cpp:105] Iteration 4968, lr = 0.00373779
I0409 20:38:13.333235 15472 solver.cpp:218] Iteration 4980 (2.40732 iter/s, 4.9848s/12 iters), loss = 4.23948
I0409 20:38:13.333290 15472 solver.cpp:237]     Train net output #0: loss = 4.23948 (* 1 = 4.23948 loss)
I0409 20:38:13.333302 15472 sgd_solver.cpp:105] Iteration 4980, lr = 0.00372892
I0409 20:38:18.198297 15472 solver.cpp:218] Iteration 4992 (2.46671 iter/s, 4.86478s/12 iters), loss = 4.35887
I0409 20:38:18.198345 15472 solver.cpp:237]     Train net output #0: loss = 4.35887 (* 1 = 4.35887 loss)
I0409 20:38:18.198354 15472 sgd_solver.cpp:105] Iteration 4992, lr = 0.00372006
I0409 20:38:20.184048 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_4998.caffemodel
I0409 20:38:21.503803 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_4998.solverstate
I0409 20:38:22.218441 15472 solver.cpp:330] Iteration 4998, Testing net (#0)
I0409 20:38:22.218464 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:38:24.657218 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:38:26.624022 15472 solver.cpp:397]     Test net output #0: accuracy = 0.060049
I0409 20:38:26.624071 15472 solver.cpp:397]     Test net output #1: loss = 4.33911 (* 1 = 4.33911 loss)
I0409 20:38:28.533808 15472 solver.cpp:218] Iteration 5004 (1.1611 iter/s, 10.335s/12 iters), loss = 4.25198
I0409 20:38:28.533865 15472 solver.cpp:237]     Train net output #0: loss = 4.25198 (* 1 = 4.25198 loss)
I0409 20:38:28.533877 15472 sgd_solver.cpp:105] Iteration 5004, lr = 0.00371123
I0409 20:38:33.359331 15472 solver.cpp:218] Iteration 5016 (2.48693 iter/s, 4.82523s/12 iters), loss = 4.41452
I0409 20:38:33.359390 15472 solver.cpp:237]     Train net output #0: loss = 4.41452 (* 1 = 4.41452 loss)
I0409 20:38:33.359402 15472 sgd_solver.cpp:105] Iteration 5016, lr = 0.00370242
I0409 20:38:38.161761 15472 solver.cpp:218] Iteration 5028 (2.49888 iter/s, 4.80214s/12 iters), loss = 4.34795
I0409 20:38:38.161813 15472 solver.cpp:237]     Train net output #0: loss = 4.34795 (* 1 = 4.34795 loss)
I0409 20:38:38.161824 15472 sgd_solver.cpp:105] Iteration 5028, lr = 0.00369363
I0409 20:38:42.970245 15472 solver.cpp:218] Iteration 5040 (2.49574 iter/s, 4.8082s/12 iters), loss = 4.32396
I0409 20:38:42.970300 15472 solver.cpp:237]     Train net output #0: loss = 4.32396 (* 1 = 4.32396 loss)
I0409 20:38:42.970310 15472 sgd_solver.cpp:105] Iteration 5040, lr = 0.00368486
I0409 20:38:47.787693 15472 solver.cpp:218] Iteration 5052 (2.49109 iter/s, 4.81717s/12 iters), loss = 4.48173
I0409 20:38:47.787744 15472 solver.cpp:237]     Train net output #0: loss = 4.48173 (* 1 = 4.48173 loss)
I0409 20:38:47.787755 15472 sgd_solver.cpp:105] Iteration 5052, lr = 0.00367611
I0409 20:38:49.658438 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:38:52.616190 15472 solver.cpp:218] Iteration 5064 (2.48539 iter/s, 4.82822s/12 iters), loss = 4.28953
I0409 20:38:52.616271 15472 solver.cpp:237]     Train net output #0: loss = 4.28953 (* 1 = 4.28953 loss)
I0409 20:38:52.616283 15472 sgd_solver.cpp:105] Iteration 5064, lr = 0.00366738
I0409 20:38:57.530056 15472 solver.cpp:218] Iteration 5076 (2.44222 iter/s, 4.91355s/12 iters), loss = 4.31526
I0409 20:38:57.530105 15472 solver.cpp:237]     Train net output #0: loss = 4.31526 (* 1 = 4.31526 loss)
I0409 20:38:57.530117 15472 sgd_solver.cpp:105] Iteration 5076, lr = 0.00365868
I0409 20:39:02.397210 15472 solver.cpp:218] Iteration 5088 (2.46565 iter/s, 4.86688s/12 iters), loss = 4.19761
I0409 20:39:02.397250 15472 solver.cpp:237]     Train net output #0: loss = 4.19761 (* 1 = 4.19761 loss)
I0409 20:39:02.397258 15472 sgd_solver.cpp:105] Iteration 5088, lr = 0.00364999
I0409 20:39:06.798420 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_5100.caffemodel
I0409 20:39:07.265319 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_5100.solverstate
I0409 20:39:07.652882 15472 solver.cpp:330] Iteration 5100, Testing net (#0)
I0409 20:39:07.652912 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:39:10.022723 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:39:12.059587 15472 solver.cpp:397]     Test net output #0: accuracy = 0.0631127
I0409 20:39:12.059635 15472 solver.cpp:397]     Test net output #1: loss = 4.20866 (* 1 = 4.20866 loss)
I0409 20:39:12.141024 15472 solver.cpp:218] Iteration 5100 (1.23161 iter/s, 9.74332s/12 iters), loss = 4.34682
I0409 20:39:12.141081 15472 solver.cpp:237]     Train net output #0: loss = 4.34682 (* 1 = 4.34682 loss)
I0409 20:39:12.141093 15472 sgd_solver.cpp:105] Iteration 5100, lr = 0.00364132
I0409 20:39:16.236014 15472 solver.cpp:218] Iteration 5112 (2.93059 iter/s, 4.09474s/12 iters), loss = 4.38548
I0409 20:39:16.236060 15472 solver.cpp:237]     Train net output #0: loss = 4.38548 (* 1 = 4.38548 loss)
I0409 20:39:16.236071 15472 sgd_solver.cpp:105] Iteration 5112, lr = 0.00363268
I0409 20:39:21.119220 15472 solver.cpp:218] Iteration 5124 (2.45754 iter/s, 4.88292s/12 iters), loss = 4.07102
I0409 20:39:21.119277 15472 solver.cpp:237]     Train net output #0: loss = 4.07102 (* 1 = 4.07102 loss)
I0409 20:39:21.119288 15472 sgd_solver.cpp:105] Iteration 5124, lr = 0.00362405
I0409 20:39:26.046129 15472 solver.cpp:218] Iteration 5136 (2.43575 iter/s, 4.92662s/12 iters), loss = 4.19882
I0409 20:39:26.047569 15472 solver.cpp:237]     Train net output #0: loss = 4.19882 (* 1 = 4.19882 loss)
I0409 20:39:26.047580 15472 sgd_solver.cpp:105] Iteration 5136, lr = 0.00361545
I0409 20:39:30.918608 15472 solver.cpp:218] Iteration 5148 (2.46366 iter/s, 4.87081s/12 iters), loss = 4.20451
I0409 20:39:30.918663 15472 solver.cpp:237]     Train net output #0: loss = 4.20451 (* 1 = 4.20451 loss)
I0409 20:39:30.918675 15472 sgd_solver.cpp:105] Iteration 5148, lr = 0.00360687
I0409 20:39:34.892683 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:39:35.805737 15472 solver.cpp:218] Iteration 5160 (2.45557 iter/s, 4.88684s/12 iters), loss = 4.06778
I0409 20:39:35.805792 15472 solver.cpp:237]     Train net output #0: loss = 4.06778 (* 1 = 4.06778 loss)
I0409 20:39:35.805804 15472 sgd_solver.cpp:105] Iteration 5160, lr = 0.0035983
I0409 20:39:40.668437 15472 solver.cpp:218] Iteration 5172 (2.46791 iter/s, 4.86242s/12 iters), loss = 4.17782
I0409 20:39:40.668478 15472 solver.cpp:237]     Train net output #0: loss = 4.17782 (* 1 = 4.17782 loss)
I0409 20:39:40.668486 15472 sgd_solver.cpp:105] Iteration 5172, lr = 0.00358976
I0409 20:39:45.534054 15472 solver.cpp:218] Iteration 5184 (2.46643 iter/s, 4.86534s/12 iters), loss = 4.32356
I0409 20:39:45.534111 15472 solver.cpp:237]     Train net output #0: loss = 4.32356 (* 1 = 4.32356 loss)
I0409 20:39:45.534122 15472 sgd_solver.cpp:105] Iteration 5184, lr = 0.00358124
I0409 20:39:50.336882 15472 solver.cpp:218] Iteration 5196 (2.49867 iter/s, 4.80255s/12 iters), loss = 4.08896
I0409 20:39:50.336930 15472 solver.cpp:237]     Train net output #0: loss = 4.08896 (* 1 = 4.08896 loss)
I0409 20:39:50.336939 15472 sgd_solver.cpp:105] Iteration 5196, lr = 0.00357273
I0409 20:39:52.339818 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_5202.caffemodel
I0409 20:39:52.988703 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_5202.solverstate
I0409 20:39:53.318398 15472 solver.cpp:330] Iteration 5202, Testing net (#0)
I0409 20:39:53.318418 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:39:55.855208 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:39:58.034833 15472 solver.cpp:397]     Test net output #0: accuracy = 0.0686275
I0409 20:39:58.034952 15472 solver.cpp:397]     Test net output #1: loss = 4.14914 (* 1 = 4.14914 loss)
I0409 20:39:59.923151 15472 solver.cpp:218] Iteration 5208 (1.25185 iter/s, 9.58578s/12 iters), loss = 4.04211
I0409 20:39:59.923209 15472 solver.cpp:237]     Train net output #0: loss = 4.04211 (* 1 = 4.04211 loss)
I0409 20:39:59.923223 15472 sgd_solver.cpp:105] Iteration 5208, lr = 0.00356425
I0409 20:40:04.729629 15472 solver.cpp:218] Iteration 5220 (2.49678 iter/s, 4.80619s/12 iters), loss = 4.24408
I0409 20:40:04.729673 15472 solver.cpp:237]     Train net output #0: loss = 4.24408 (* 1 = 4.24408 loss)
I0409 20:40:04.729686 15472 sgd_solver.cpp:105] Iteration 5220, lr = 0.00355579
I0409 20:40:09.562069 15472 solver.cpp:218] Iteration 5232 (2.48336 iter/s, 4.83217s/12 iters), loss = 4.20937
I0409 20:40:09.562117 15472 solver.cpp:237]     Train net output #0: loss = 4.20937 (* 1 = 4.20937 loss)
I0409 20:40:09.562129 15472 sgd_solver.cpp:105] Iteration 5232, lr = 0.00354735
I0409 20:40:14.392231 15472 solver.cpp:218] Iteration 5244 (2.48453 iter/s, 4.82989s/12 iters), loss = 4.07473
I0409 20:40:14.392280 15472 solver.cpp:237]     Train net output #0: loss = 4.07473 (* 1 = 4.07473 loss)
I0409 20:40:14.392292 15472 sgd_solver.cpp:105] Iteration 5244, lr = 0.00353892
I0409 20:40:19.214874 15472 solver.cpp:218] Iteration 5256 (2.48841 iter/s, 4.82236s/12 iters), loss = 4.14202
I0409 20:40:19.214933 15472 solver.cpp:237]     Train net output #0: loss = 4.14202 (* 1 = 4.14202 loss)
I0409 20:40:19.214946 15472 sgd_solver.cpp:105] Iteration 5256, lr = 0.00353052
I0409 20:40:20.451126 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:40:24.004629 15472 solver.cpp:218] Iteration 5268 (2.5055 iter/s, 4.78947s/12 iters), loss = 4.06166
I0409 20:40:24.004685 15472 solver.cpp:237]     Train net output #0: loss = 4.06166 (* 1 = 4.06166 loss)
I0409 20:40:24.004696 15472 sgd_solver.cpp:105] Iteration 5268, lr = 0.00352214
I0409 20:40:28.966042 15472 solver.cpp:218] Iteration 5280 (2.41881 iter/s, 4.96113s/12 iters), loss = 3.99387
I0409 20:40:28.966184 15472 solver.cpp:237]     Train net output #0: loss = 3.99387 (* 1 = 3.99387 loss)
I0409 20:40:28.966198 15472 sgd_solver.cpp:105] Iteration 5280, lr = 0.00351378
I0409 20:40:33.808619 15472 solver.cpp:218] Iteration 5292 (2.47821 iter/s, 4.8422s/12 iters), loss = 4.28227
I0409 20:40:33.808674 15472 solver.cpp:237]     Train net output #0: loss = 4.28227 (* 1 = 4.28227 loss)
I0409 20:40:33.808686 15472 sgd_solver.cpp:105] Iteration 5292, lr = 0.00350544
I0409 20:40:38.205369 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_5304.caffemodel
I0409 20:40:39.619105 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_5304.solverstate
I0409 20:40:40.711522 15472 solver.cpp:330] Iteration 5304, Testing net (#0)
I0409 20:40:40.711547 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:40:43.050107 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:40:45.139050 15472 solver.cpp:397]     Test net output #0: accuracy = 0.0631127
I0409 20:40:45.139099 15472 solver.cpp:397]     Test net output #1: loss = 4.11082 (* 1 = 4.11082 loss)
I0409 20:40:45.222098 15472 solver.cpp:218] Iteration 5304 (1.05144 iter/s, 11.4129s/12 iters), loss = 3.93702
I0409 20:40:45.222148 15472 solver.cpp:237]     Train net output #0: loss = 3.93702 (* 1 = 3.93702 loss)
I0409 20:40:45.222160 15472 sgd_solver.cpp:105] Iteration 5304, lr = 0.00349711
I0409 20:40:49.424278 15472 solver.cpp:218] Iteration 5316 (2.85583 iter/s, 4.20192s/12 iters), loss = 4.04446
I0409 20:40:49.424325 15472 solver.cpp:237]     Train net output #0: loss = 4.04446 (* 1 = 4.04446 loss)
I0409 20:40:49.424335 15472 sgd_solver.cpp:105] Iteration 5316, lr = 0.00348881
I0409 20:40:54.243991 15472 solver.cpp:218] Iteration 5328 (2.48991 iter/s, 4.81944s/12 iters), loss = 4.21351
I0409 20:40:54.244033 15472 solver.cpp:237]     Train net output #0: loss = 4.21351 (* 1 = 4.21351 loss)
I0409 20:40:54.244041 15472 sgd_solver.cpp:105] Iteration 5328, lr = 0.00348053
I0409 20:40:59.083570 15472 solver.cpp:218] Iteration 5340 (2.47969 iter/s, 4.83931s/12 iters), loss = 3.99951
I0409 20:40:59.083650 15472 solver.cpp:237]     Train net output #0: loss = 3.99951 (* 1 = 3.99951 loss)
I0409 20:40:59.083663 15472 sgd_solver.cpp:105] Iteration 5340, lr = 0.00347226
I0409 20:41:03.898221 15472 solver.cpp:218] Iteration 5352 (2.49255 iter/s, 4.81434s/12 iters), loss = 4.09764
I0409 20:41:03.898267 15472 solver.cpp:237]     Train net output #0: loss = 4.09764 (* 1 = 4.09764 loss)
I0409 20:41:03.898277 15472 sgd_solver.cpp:105] Iteration 5352, lr = 0.00346402
I0409 20:41:07.184118 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:41:08.688947 15472 solver.cpp:218] Iteration 5364 (2.50499 iter/s, 4.79045s/12 iters), loss = 4.2451
I0409 20:41:08.689007 15472 solver.cpp:237]     Train net output #0: loss = 4.2451 (* 1 = 4.2451 loss)
I0409 20:41:08.689018 15472 sgd_solver.cpp:105] Iteration 5364, lr = 0.0034558
I0409 20:41:13.661918 15472 solver.cpp:218] Iteration 5376 (2.41319 iter/s, 4.97268s/12 iters), loss = 4.1462
I0409 20:41:13.661993 15472 solver.cpp:237]     Train net output #0: loss = 4.1462 (* 1 = 4.1462 loss)
I0409 20:41:13.662005 15472 sgd_solver.cpp:105] Iteration 5376, lr = 0.00344759
I0409 20:41:18.490860 15472 solver.cpp:218] Iteration 5388 (2.48517 iter/s, 4.82864s/12 iters), loss = 4.21206
I0409 20:41:18.490909 15472 solver.cpp:237]     Train net output #0: loss = 4.21206 (* 1 = 4.21206 loss)
I0409 20:41:18.490921 15472 sgd_solver.cpp:105] Iteration 5388, lr = 0.00343941
I0409 20:41:23.247483 15472 solver.cpp:218] Iteration 5400 (2.52294 iter/s, 4.75635s/12 iters), loss = 4.21238
I0409 20:41:23.247532 15472 solver.cpp:237]     Train net output #0: loss = 4.21238 (* 1 = 4.21238 loss)
I0409 20:41:23.247545 15472 sgd_solver.cpp:105] Iteration 5400, lr = 0.00343124
I0409 20:41:25.275828 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_5406.caffemodel
I0409 20:41:25.736424 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_5406.solverstate
I0409 20:41:26.072479 15472 solver.cpp:330] Iteration 5406, Testing net (#0)
I0409 20:41:26.072500 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:41:28.373426 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:41:30.551764 15472 solver.cpp:397]     Test net output #0: accuracy = 0.0741422
I0409 20:41:30.551937 15472 solver.cpp:397]     Test net output #1: loss = 4.06264 (* 1 = 4.06264 loss)
I0409 20:41:32.325367 15472 solver.cpp:218] Iteration 5412 (1.32196 iter/s, 9.07741s/12 iters), loss = 4.15505
I0409 20:41:32.325428 15472 solver.cpp:237]     Train net output #0: loss = 4.15505 (* 1 = 4.15505 loss)
I0409 20:41:32.325439 15472 sgd_solver.cpp:105] Iteration 5412, lr = 0.00342309
I0409 20:41:37.111416 15472 solver.cpp:218] Iteration 5424 (2.50744 iter/s, 4.78576s/12 iters), loss = 4.13578
I0409 20:41:37.111480 15472 solver.cpp:237]     Train net output #0: loss = 4.13578 (* 1 = 4.13578 loss)
I0409 20:41:37.111493 15472 sgd_solver.cpp:105] Iteration 5424, lr = 0.00341497
I0409 20:41:41.887758 15472 solver.cpp:218] Iteration 5436 (2.51254 iter/s, 4.77605s/12 iters), loss = 4.15404
I0409 20:41:41.887810 15472 solver.cpp:237]     Train net output #0: loss = 4.15404 (* 1 = 4.15404 loss)
I0409 20:41:41.887821 15472 sgd_solver.cpp:105] Iteration 5436, lr = 0.00340686
I0409 20:41:46.919363 15472 solver.cpp:218] Iteration 5448 (2.38506 iter/s, 5.03132s/12 iters), loss = 3.92762
I0409 20:41:46.919407 15472 solver.cpp:237]     Train net output #0: loss = 3.92762 (* 1 = 3.92762 loss)
I0409 20:41:46.919416 15472 sgd_solver.cpp:105] Iteration 5448, lr = 0.00339877
I0409 20:41:51.966851 15472 solver.cpp:218] Iteration 5460 (2.37756 iter/s, 5.0472s/12 iters), loss = 4.34824
I0409 20:41:51.966917 15472 solver.cpp:237]     Train net output #0: loss = 4.34824 (* 1 = 4.34824 loss)
I0409 20:41:51.966933 15472 sgd_solver.cpp:105] Iteration 5460, lr = 0.0033907
I0409 20:41:52.529670 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:41:56.902164 15472 solver.cpp:218] Iteration 5472 (2.4316 iter/s, 4.93502s/12 iters), loss = 4.07815
I0409 20:41:56.902212 15472 solver.cpp:237]     Train net output #0: loss = 4.07815 (* 1 = 4.07815 loss)
I0409 20:41:56.902221 15472 sgd_solver.cpp:105] Iteration 5472, lr = 0.00338265
I0409 20:42:01.797526 15472 solver.cpp:218] Iteration 5484 (2.45144 iter/s, 4.89508s/12 iters), loss = 3.97104
I0409 20:42:01.797694 15472 solver.cpp:237]     Train net output #0: loss = 3.97104 (* 1 = 3.97104 loss)
I0409 20:42:01.797708 15472 sgd_solver.cpp:105] Iteration 5484, lr = 0.00337462
I0409 20:42:06.805539 15472 solver.cpp:218] Iteration 5496 (2.39635 iter/s, 5.00761s/12 iters), loss = 3.9249
I0409 20:42:06.805595 15472 solver.cpp:237]     Train net output #0: loss = 3.9249 (* 1 = 3.9249 loss)
I0409 20:42:06.805608 15472 sgd_solver.cpp:105] Iteration 5496, lr = 0.00336661
I0409 20:42:11.179121 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_5508.caffemodel
I0409 20:42:11.599700 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_5508.solverstate
I0409 20:42:11.915760 15472 solver.cpp:330] Iteration 5508, Testing net (#0)
I0409 20:42:11.915791 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:42:14.248703 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:42:16.419155 15472 solver.cpp:397]     Test net output #0: accuracy = 0.0851716
I0409 20:42:16.419203 15472 solver.cpp:397]     Test net output #1: loss = 4.01158 (* 1 = 4.01158 loss)
I0409 20:42:16.502617 15472 solver.cpp:218] Iteration 5508 (1.23755 iter/s, 9.69658s/12 iters), loss = 3.79501
I0409 20:42:16.502666 15472 solver.cpp:237]     Train net output #0: loss = 3.79501 (* 1 = 3.79501 loss)
I0409 20:42:16.502677 15472 sgd_solver.cpp:105] Iteration 5508, lr = 0.00335861
I0409 20:42:20.610642 15472 solver.cpp:218] Iteration 5520 (2.92129 iter/s, 4.10777s/12 iters), loss = 3.94548
I0409 20:42:20.610699 15472 solver.cpp:237]     Train net output #0: loss = 3.94548 (* 1 = 3.94548 loss)
I0409 20:42:20.610711 15472 sgd_solver.cpp:105] Iteration 5520, lr = 0.00335064
I0409 20:42:21.745281 15472 blocking_queue.cpp:49] Waiting for data
I0409 20:42:25.422948 15472 solver.cpp:218] Iteration 5532 (2.49376 iter/s, 4.81202s/12 iters), loss = 3.91834
I0409 20:42:25.423004 15472 solver.cpp:237]     Train net output #0: loss = 3.91834 (* 1 = 3.91834 loss)
I0409 20:42:25.423017 15472 sgd_solver.cpp:105] Iteration 5532, lr = 0.00334268
I0409 20:42:30.237390 15472 solver.cpp:218] Iteration 5544 (2.49265 iter/s, 4.81415s/12 iters), loss = 4.07292
I0409 20:42:30.237449 15472 solver.cpp:237]     Train net output #0: loss = 4.07292 (* 1 = 4.07292 loss)
I0409 20:42:30.237460 15472 sgd_solver.cpp:105] Iteration 5544, lr = 0.00333475
I0409 20:42:35.238135 15472 solver.cpp:218] Iteration 5556 (2.39979 iter/s, 5.00044s/12 iters), loss = 4.01601
I0409 20:42:35.238528 15472 solver.cpp:237]     Train net output #0: loss = 4.01601 (* 1 = 4.01601 loss)
I0409 20:42:35.238544 15472 sgd_solver.cpp:105] Iteration 5556, lr = 0.00332683
I0409 20:42:37.958392 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:42:40.285077 15472 solver.cpp:218] Iteration 5568 (2.37797 iter/s, 5.04631s/12 iters), loss = 3.89118
I0409 20:42:40.285133 15472 solver.cpp:237]     Train net output #0: loss = 3.89118 (* 1 = 3.89118 loss)
I0409 20:42:40.285149 15472 sgd_solver.cpp:105] Iteration 5568, lr = 0.00331893
I0409 20:42:45.261075 15472 solver.cpp:218] Iteration 5580 (2.41172 iter/s, 4.97571s/12 iters), loss = 4.09479
I0409 20:42:45.261121 15472 solver.cpp:237]     Train net output #0: loss = 4.09479 (* 1 = 4.09479 loss)
I0409 20:42:45.261133 15472 sgd_solver.cpp:105] Iteration 5580, lr = 0.00331105
I0409 20:42:50.180554 15472 solver.cpp:218] Iteration 5592 (2.43942 iter/s, 4.9192s/12 iters), loss = 3.99706
I0409 20:42:50.180608 15472 solver.cpp:237]     Train net output #0: loss = 3.99706 (* 1 = 3.99706 loss)
I0409 20:42:50.180620 15472 sgd_solver.cpp:105] Iteration 5592, lr = 0.00330319
I0409 20:42:55.026513 15472 solver.cpp:218] Iteration 5604 (2.47644 iter/s, 4.84567s/12 iters), loss = 4.05955
I0409 20:42:55.026568 15472 solver.cpp:237]     Train net output #0: loss = 4.05955 (* 1 = 4.05955 loss)
I0409 20:42:55.026580 15472 sgd_solver.cpp:105] Iteration 5604, lr = 0.00329535
I0409 20:42:56.969554 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_5610.caffemodel
I0409 20:42:58.068814 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_5610.solverstate
I0409 20:43:00.400189 15472 solver.cpp:330] Iteration 5610, Testing net (#0)
I0409 20:43:00.400215 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:43:02.639057 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:43:04.844306 15472 solver.cpp:397]     Test net output #0: accuracy = 0.0955882
I0409 20:43:04.844357 15472 solver.cpp:397]     Test net output #1: loss = 3.91316 (* 1 = 3.91316 loss)
I0409 20:43:06.655020 15472 solver.cpp:218] Iteration 5616 (1.032 iter/s, 11.6279s/12 iters), loss = 3.85208
I0409 20:43:06.655165 15472 solver.cpp:237]     Train net output #0: loss = 3.85208 (* 1 = 3.85208 loss)
I0409 20:43:06.655176 15472 sgd_solver.cpp:105] Iteration 5616, lr = 0.00328752
I0409 20:43:11.459684 15472 solver.cpp:218] Iteration 5628 (2.49776 iter/s, 4.8043s/12 iters), loss = 4.13116
I0409 20:43:11.459726 15472 solver.cpp:237]     Train net output #0: loss = 4.13116 (* 1 = 4.13116 loss)
I0409 20:43:11.459735 15472 sgd_solver.cpp:105] Iteration 5628, lr = 0.00327972
I0409 20:43:16.298271 15472 solver.cpp:218] Iteration 5640 (2.4802 iter/s, 4.83831s/12 iters), loss = 4.00118
I0409 20:43:16.298324 15472 solver.cpp:237]     Train net output #0: loss = 4.00118 (* 1 = 4.00118 loss)
I0409 20:43:16.298336 15472 sgd_solver.cpp:105] Iteration 5640, lr = 0.00327193
I0409 20:43:21.078459 15472 solver.cpp:218] Iteration 5652 (2.51051 iter/s, 4.77991s/12 iters), loss = 3.95088
I0409 20:43:21.078512 15472 solver.cpp:237]     Train net output #0: loss = 3.95088 (* 1 = 3.95088 loss)
I0409 20:43:21.078524 15472 sgd_solver.cpp:105] Iteration 5652, lr = 0.00326416
I0409 20:43:25.806316 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:43:25.963140 15472 solver.cpp:218] Iteration 5664 (2.4568 iter/s, 4.8844s/12 iters), loss = 4.03966
I0409 20:43:25.963193 15472 solver.cpp:237]     Train net output #0: loss = 4.03966 (* 1 = 4.03966 loss)
I0409 20:43:25.963204 15472 sgd_solver.cpp:105] Iteration 5664, lr = 0.00325641
I0409 20:43:30.901517 15472 solver.cpp:218] Iteration 5676 (2.4301 iter/s, 4.93808s/12 iters), loss = 3.83565
I0409 20:43:30.901579 15472 solver.cpp:237]     Train net output #0: loss = 3.83565 (* 1 = 3.83565 loss)
I0409 20:43:30.901593 15472 sgd_solver.cpp:105] Iteration 5676, lr = 0.00324868
I0409 20:43:35.699710 15472 solver.cpp:218] Iteration 5688 (2.50109 iter/s, 4.7979s/12 iters), loss = 3.88169
I0409 20:43:35.699766 15472 solver.cpp:237]     Train net output #0: loss = 3.88169 (* 1 = 3.88169 loss)
I0409 20:43:35.699779 15472 sgd_solver.cpp:105] Iteration 5688, lr = 0.00324097
I0409 20:43:40.507025 15472 solver.cpp:218] Iteration 5700 (2.49634 iter/s, 4.80703s/12 iters), loss = 4.13429
I0409 20:43:40.507136 15472 solver.cpp:237]     Train net output #0: loss = 4.13429 (* 1 = 4.13429 loss)
I0409 20:43:40.507150 15472 sgd_solver.cpp:105] Iteration 5700, lr = 0.00323328
I0409 20:43:44.889143 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_5712.caffemodel
I0409 20:43:45.337814 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_5712.solverstate
I0409 20:43:45.660303 15472 solver.cpp:330] Iteration 5712, Testing net (#0)
I0409 20:43:45.660322 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:43:47.847506 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:43:50.258364 15472 solver.cpp:397]     Test net output #0: accuracy = 0.09375
I0409 20:43:50.258410 15472 solver.cpp:397]     Test net output #1: loss = 3.90155 (* 1 = 3.90155 loss)
I0409 20:43:50.341512 15472 solver.cpp:218] Iteration 5712 (1.22027 iter/s, 9.83393s/12 iters), loss = 3.82047
I0409 20:43:50.341563 15472 solver.cpp:237]     Train net output #0: loss = 3.82047 (* 1 = 3.82047 loss)
I0409 20:43:50.341573 15472 sgd_solver.cpp:105] Iteration 5712, lr = 0.0032256
I0409 20:43:54.394838 15472 solver.cpp:218] Iteration 5724 (2.96071 iter/s, 4.05308s/12 iters), loss = 3.98602
I0409 20:43:54.394896 15472 solver.cpp:237]     Train net output #0: loss = 3.98602 (* 1 = 3.98602 loss)
I0409 20:43:54.394908 15472 sgd_solver.cpp:105] Iteration 5724, lr = 0.00321794
I0409 20:43:59.271452 15472 solver.cpp:218] Iteration 5736 (2.46087 iter/s, 4.87633s/12 iters), loss = 4.07332
I0409 20:43:59.271494 15472 solver.cpp:237]     Train net output #0: loss = 4.07332 (* 1 = 4.07332 loss)
I0409 20:43:59.271503 15472 sgd_solver.cpp:105] Iteration 5736, lr = 0.0032103
I0409 20:44:04.057062 15472 solver.cpp:218] Iteration 5748 (2.50766 iter/s, 4.78534s/12 iters), loss = 3.87005
I0409 20:44:04.057117 15472 solver.cpp:237]     Train net output #0: loss = 3.87005 (* 1 = 3.87005 loss)
I0409 20:44:04.057129 15472 sgd_solver.cpp:105] Iteration 5748, lr = 0.00320268
I0409 20:44:08.884191 15472 solver.cpp:218] Iteration 5760 (2.4861 iter/s, 4.82684s/12 iters), loss = 3.9299
I0409 20:44:08.884246 15472 solver.cpp:237]     Train net output #0: loss = 3.9299 (* 1 = 3.9299 loss)
I0409 20:44:08.884259 15472 sgd_solver.cpp:105] Iteration 5760, lr = 0.00319508
I0409 20:44:10.764015 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:44:13.696538 15472 solver.cpp:218] Iteration 5772 (2.49373 iter/s, 4.81206s/12 iters), loss = 3.87428
I0409 20:44:13.696589 15472 solver.cpp:237]     Train net output #0: loss = 3.87428 (* 1 = 3.87428 loss)
I0409 20:44:13.696600 15472 sgd_solver.cpp:105] Iteration 5772, lr = 0.00318749
I0409 20:44:18.529779 15472 solver.cpp:218] Iteration 5784 (2.48295 iter/s, 4.83296s/12 iters), loss = 3.79113
I0409 20:44:18.529824 15472 solver.cpp:237]     Train net output #0: loss = 3.79113 (* 1 = 3.79113 loss)
I0409 20:44:18.529835 15472 sgd_solver.cpp:105] Iteration 5784, lr = 0.00317992
I0409 20:44:23.363005 15472 solver.cpp:218] Iteration 5796 (2.48296 iter/s, 4.83294s/12 iters), loss = 3.7546
I0409 20:44:23.363061 15472 solver.cpp:237]     Train net output #0: loss = 3.7546 (* 1 = 3.7546 loss)
I0409 20:44:23.363075 15472 sgd_solver.cpp:105] Iteration 5796, lr = 0.00317237
I0409 20:44:28.180538 15472 solver.cpp:218] Iteration 5808 (2.49105 iter/s, 4.81725s/12 iters), loss = 3.89739
I0409 20:44:28.180593 15472 solver.cpp:237]     Train net output #0: loss = 3.89739 (* 1 = 3.89739 loss)
I0409 20:44:28.180606 15472 sgd_solver.cpp:105] Iteration 5808, lr = 0.00316484
I0409 20:44:30.161609 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_5814.caffemodel
I0409 20:44:31.041319 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_5814.solverstate
I0409 20:44:31.755007 15472 solver.cpp:330] Iteration 5814, Testing net (#0)
I0409 20:44:31.755038 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:44:33.879554 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:44:36.164103 15472 solver.cpp:397]     Test net output #0: accuracy = 0.0919118
I0409 20:44:36.164152 15472 solver.cpp:397]     Test net output #1: loss = 3.85491 (* 1 = 3.85491 loss)
I0409 20:44:37.965003 15472 solver.cpp:218] Iteration 5820 (1.2265 iter/s, 9.78396s/12 iters), loss = 3.94532
I0409 20:44:37.965056 15472 solver.cpp:237]     Train net output #0: loss = 3.94532 (* 1 = 3.94532 loss)
I0409 20:44:37.965068 15472 sgd_solver.cpp:105] Iteration 5820, lr = 0.00315733
I0409 20:44:42.808226 15472 solver.cpp:218] Iteration 5832 (2.47783 iter/s, 4.84294s/12 iters), loss = 3.86287
I0409 20:44:42.808326 15472 solver.cpp:237]     Train net output #0: loss = 3.86287 (* 1 = 3.86287 loss)
I0409 20:44:42.808337 15472 sgd_solver.cpp:105] Iteration 5832, lr = 0.00314983
I0409 20:44:47.622320 15472 solver.cpp:218] Iteration 5844 (2.49285 iter/s, 4.81377s/12 iters), loss = 3.82482
I0409 20:44:47.622370 15472 solver.cpp:237]     Train net output #0: loss = 3.82482 (* 1 = 3.82482 loss)
I0409 20:44:47.622380 15472 sgd_solver.cpp:105] Iteration 5844, lr = 0.00314235
I0409 20:44:52.436689 15472 solver.cpp:218] Iteration 5856 (2.49268 iter/s, 4.81409s/12 iters), loss = 3.84826
I0409 20:44:52.436746 15472 solver.cpp:237]     Train net output #0: loss = 3.84826 (* 1 = 3.84826 loss)
I0409 20:44:52.436759 15472 sgd_solver.cpp:105] Iteration 5856, lr = 0.00313489
I0409 20:44:56.457809 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:44:57.238687 15472 solver.cpp:218] Iteration 5868 (2.49911 iter/s, 4.80172s/12 iters), loss = 3.66827
I0409 20:44:57.238729 15472 solver.cpp:237]     Train net output #0: loss = 3.66827 (* 1 = 3.66827 loss)
I0409 20:44:57.238739 15472 sgd_solver.cpp:105] Iteration 5868, lr = 0.00312745
I0409 20:45:02.228816 15472 solver.cpp:218] Iteration 5880 (2.40488 iter/s, 4.98985s/12 iters), loss = 3.90302
I0409 20:45:02.228863 15472 solver.cpp:237]     Train net output #0: loss = 3.90302 (* 1 = 3.90302 loss)
I0409 20:45:02.228871 15472 sgd_solver.cpp:105] Iteration 5880, lr = 0.00312002
I0409 20:45:07.000537 15472 solver.cpp:218] Iteration 5892 (2.51496 iter/s, 4.77145s/12 iters), loss = 3.89435
I0409 20:45:07.000578 15472 solver.cpp:237]     Train net output #0: loss = 3.89435 (* 1 = 3.89435 loss)
I0409 20:45:07.000588 15472 sgd_solver.cpp:105] Iteration 5892, lr = 0.00311262
I0409 20:45:11.919031 15472 solver.cpp:218] Iteration 5904 (2.43991 iter/s, 4.91822s/12 iters), loss = 3.6772
I0409 20:45:11.919076 15472 solver.cpp:237]     Train net output #0: loss = 3.6772 (* 1 = 3.6772 loss)
I0409 20:45:11.919085 15472 sgd_solver.cpp:105] Iteration 5904, lr = 0.00310523
I0409 20:45:16.355922 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_5916.caffemodel
I0409 20:45:16.817167 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_5916.solverstate
I0409 20:45:17.280068 15472 solver.cpp:330] Iteration 5916, Testing net (#0)
I0409 20:45:17.280094 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:45:19.496471 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:45:21.926262 15472 solver.cpp:397]     Test net output #0: accuracy = 0.0943627
I0409 20:45:21.926314 15472 solver.cpp:397]     Test net output #1: loss = 3.8139 (* 1 = 3.8139 loss)
I0409 20:45:22.009430 15472 solver.cpp:218] Iteration 5916 (1.18931 iter/s, 10.0899s/12 iters), loss = 3.63804
I0409 20:45:22.009481 15472 solver.cpp:237]     Train net output #0: loss = 3.63804 (* 1 = 3.63804 loss)
I0409 20:45:22.009492 15472 sgd_solver.cpp:105] Iteration 5916, lr = 0.00309785
I0409 20:45:26.162545 15472 solver.cpp:218] Iteration 5928 (2.88957 iter/s, 4.15286s/12 iters), loss = 3.80241
I0409 20:45:26.162595 15472 solver.cpp:237]     Train net output #0: loss = 3.80241 (* 1 = 3.80241 loss)
I0409 20:45:26.162606 15472 sgd_solver.cpp:105] Iteration 5928, lr = 0.0030905
I0409 20:45:31.062244 15472 solver.cpp:218] Iteration 5940 (2.44927 iter/s, 4.89941s/12 iters), loss = 3.91069
I0409 20:45:31.062296 15472 solver.cpp:237]     Train net output #0: loss = 3.91069 (* 1 = 3.91069 loss)
I0409 20:45:31.062307 15472 sgd_solver.cpp:105] Iteration 5940, lr = 0.00308316
I0409 20:45:35.888348 15472 solver.cpp:218] Iteration 5952 (2.48662 iter/s, 4.82582s/12 iters), loss = 3.93862
I0409 20:45:35.888399 15472 solver.cpp:237]     Train net output #0: loss = 3.93862 (* 1 = 3.93862 loss)
I0409 20:45:35.888411 15472 sgd_solver.cpp:105] Iteration 5952, lr = 0.00307584
I0409 20:45:40.795822 15472 solver.cpp:218] Iteration 5964 (2.44539 iter/s, 4.90719s/12 iters), loss = 3.92799
I0409 20:45:40.795871 15472 solver.cpp:237]     Train net output #0: loss = 3.92799 (* 1 = 3.92799 loss)
I0409 20:45:40.795879 15472 sgd_solver.cpp:105] Iteration 5964, lr = 0.00306854
I0409 20:45:42.086414 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:45:45.686363 15472 solver.cpp:218] Iteration 5976 (2.45386 iter/s, 4.89025s/12 iters), loss = 3.63605
I0409 20:45:45.686436 15472 solver.cpp:237]     Train net output #0: loss = 3.63605 (* 1 = 3.63605 loss)
I0409 20:45:45.686452 15472 sgd_solver.cpp:105] Iteration 5976, lr = 0.00306125
I0409 20:45:50.578860 15472 solver.cpp:218] Iteration 5988 (2.45289 iter/s, 4.89219s/12 iters), loss = 3.87662
I0409 20:45:50.578989 15472 solver.cpp:237]     Train net output #0: loss = 3.87662 (* 1 = 3.87662 loss)
I0409 20:45:50.579000 15472 sgd_solver.cpp:105] Iteration 5988, lr = 0.00305398
I0409 20:45:55.447016 15472 solver.cpp:218] Iteration 6000 (2.46518 iter/s, 4.8678s/12 iters), loss = 3.73696
I0409 20:45:55.447062 15472 solver.cpp:237]     Train net output #0: loss = 3.73696 (* 1 = 3.73696 loss)
I0409 20:45:55.447072 15472 sgd_solver.cpp:105] Iteration 6000, lr = 0.00304673
I0409 20:46:00.481331 15472 solver.cpp:218] Iteration 6012 (2.38378 iter/s, 5.03403s/12 iters), loss = 3.75002
I0409 20:46:00.481391 15472 solver.cpp:237]     Train net output #0: loss = 3.75002 (* 1 = 3.75002 loss)
I0409 20:46:00.481402 15472 sgd_solver.cpp:105] Iteration 6012, lr = 0.0030395
I0409 20:46:02.456166 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_6018.caffemodel
I0409 20:46:02.942726 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_6018.solverstate
I0409 20:46:03.267807 15472 solver.cpp:330] Iteration 6018, Testing net (#0)
I0409 20:46:03.267833 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:46:05.367282 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:46:07.790813 15472 solver.cpp:397]     Test net output #0: accuracy = 0.096201
I0409 20:46:07.790863 15472 solver.cpp:397]     Test net output #1: loss = 3.7809 (* 1 = 3.7809 loss)
I0409 20:46:09.668781 15472 solver.cpp:218] Iteration 6024 (1.3062 iter/s, 9.18697s/12 iters), loss = 3.63767
I0409 20:46:09.668835 15472 solver.cpp:237]     Train net output #0: loss = 3.63767 (* 1 = 3.63767 loss)
I0409 20:46:09.668848 15472 sgd_solver.cpp:105] Iteration 6024, lr = 0.00303228
I0409 20:46:14.508042 15472 solver.cpp:218] Iteration 6036 (2.47986 iter/s, 4.83898s/12 iters), loss = 3.89512
I0409 20:46:14.508087 15472 solver.cpp:237]     Train net output #0: loss = 3.89512 (* 1 = 3.89512 loss)
I0409 20:46:14.508097 15472 sgd_solver.cpp:105] Iteration 6036, lr = 0.00302508
I0409 20:46:19.390813 15472 solver.cpp:218] Iteration 6048 (2.45776 iter/s, 4.88249s/12 iters), loss = 3.67417
I0409 20:46:19.390869 15472 solver.cpp:237]     Train net output #0: loss = 3.67417 (* 1 = 3.67417 loss)
I0409 20:46:19.390883 15472 sgd_solver.cpp:105] Iteration 6048, lr = 0.0030179
I0409 20:46:24.513234 15472 solver.cpp:218] Iteration 6060 (2.34278 iter/s, 5.12213s/12 iters), loss = 3.77566
I0409 20:46:24.513314 15472 solver.cpp:237]     Train net output #0: loss = 3.77566 (* 1 = 3.77566 loss)
I0409 20:46:24.513329 15472 sgd_solver.cpp:105] Iteration 6060, lr = 0.00301074
I0409 20:46:27.923650 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:46:29.401460 15472 solver.cpp:218] Iteration 6072 (2.45503 iter/s, 4.88793s/12 iters), loss = 3.74109
I0409 20:46:29.401513 15472 solver.cpp:237]     Train net output #0: loss = 3.74109 (* 1 = 3.74109 loss)
I0409 20:46:29.401527 15472 sgd_solver.cpp:105] Iteration 6072, lr = 0.00300359
I0409 20:46:34.248178 15472 solver.cpp:218] Iteration 6084 (2.476 iter/s, 4.84653s/12 iters), loss = 3.59602
I0409 20:46:34.248232 15472 solver.cpp:237]     Train net output #0: loss = 3.59602 (* 1 = 3.59602 loss)
I0409 20:46:34.248245 15472 sgd_solver.cpp:105] Iteration 6084, lr = 0.00299646
I0409 20:46:39.049909 15472 solver.cpp:218] Iteration 6096 (2.49919 iter/s, 4.80155s/12 iters), loss = 3.73494
I0409 20:46:39.049975 15472 solver.cpp:237]     Train net output #0: loss = 3.73494 (* 1 = 3.73494 loss)
I0409 20:46:39.049988 15472 sgd_solver.cpp:105] Iteration 6096, lr = 0.00298934
I0409 20:46:43.876971 15472 solver.cpp:218] Iteration 6108 (2.48608 iter/s, 4.82688s/12 iters), loss = 3.5873
I0409 20:46:43.877018 15472 solver.cpp:237]     Train net output #0: loss = 3.5873 (* 1 = 3.5873 loss)
I0409 20:46:43.877027 15472 sgd_solver.cpp:105] Iteration 6108, lr = 0.00298225
I0409 20:46:48.276577 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_6120.caffemodel
I0409 20:46:49.344861 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_6120.solverstate
I0409 20:46:49.670547 15472 solver.cpp:330] Iteration 6120, Testing net (#0)
I0409 20:46:49.670576 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:46:51.673936 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:46:54.082073 15472 solver.cpp:397]     Test net output #0: accuracy = 0.103554
I0409 20:46:54.082105 15472 solver.cpp:397]     Test net output #1: loss = 3.70584 (* 1 = 3.70584 loss)
I0409 20:46:54.165390 15472 solver.cpp:218] Iteration 6120 (1.1664 iter/s, 10.2881s/12 iters), loss = 3.69781
I0409 20:46:54.165431 15472 solver.cpp:237]     Train net output #0: loss = 3.69781 (* 1 = 3.69781 loss)
I0409 20:46:54.165441 15472 sgd_solver.cpp:105] Iteration 6120, lr = 0.00297517
I0409 20:46:58.301977 15472 solver.cpp:218] Iteration 6132 (2.90107 iter/s, 4.13641s/12 iters), loss = 3.77585
I0409 20:46:58.302098 15472 solver.cpp:237]     Train net output #0: loss = 3.77585 (* 1 = 3.77585 loss)
I0409 20:46:58.302109 15472 sgd_solver.cpp:105] Iteration 6132, lr = 0.0029681
I0409 20:47:03.271257 15472 solver.cpp:218] Iteration 6144 (2.41496 iter/s, 4.96902s/12 iters), loss = 3.8487
I0409 20:47:03.271307 15472 solver.cpp:237]     Train net output #0: loss = 3.8487 (* 1 = 3.8487 loss)
I0409 20:47:03.271319 15472 sgd_solver.cpp:105] Iteration 6144, lr = 0.00296105
I0409 20:47:08.179222 15472 solver.cpp:218] Iteration 6156 (2.4451 iter/s, 4.90778s/12 iters), loss = 3.47053
I0409 20:47:08.179260 15472 solver.cpp:237]     Train net output #0: loss = 3.47053 (* 1 = 3.47053 loss)
I0409 20:47:08.179270 15472 sgd_solver.cpp:105] Iteration 6156, lr = 0.00295402
I0409 20:47:13.002019 15472 solver.cpp:218] Iteration 6168 (2.48827 iter/s, 4.82262s/12 iters), loss = 3.70163
I0409 20:47:13.002063 15472 solver.cpp:237]     Train net output #0: loss = 3.70163 (* 1 = 3.70163 loss)
I0409 20:47:13.002072 15472 sgd_solver.cpp:105] Iteration 6168, lr = 0.00294701
I0409 20:47:13.580098 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:47:17.960187 15472 solver.cpp:218] Iteration 6180 (2.42034 iter/s, 4.95798s/12 iters), loss = 3.56601
I0409 20:47:17.960233 15472 solver.cpp:237]     Train net output #0: loss = 3.56601 (* 1 = 3.56601 loss)
I0409 20:47:17.960244 15472 sgd_solver.cpp:105] Iteration 6180, lr = 0.00294001
I0409 20:47:22.756057 15472 solver.cpp:218] Iteration 6192 (2.50225 iter/s, 4.79569s/12 iters), loss = 3.75808
I0409 20:47:22.756109 15472 solver.cpp:237]     Train net output #0: loss = 3.75808 (* 1 = 3.75808 loss)
I0409 20:47:22.756122 15472 sgd_solver.cpp:105] Iteration 6192, lr = 0.00293303
I0409 20:47:27.779451 15472 solver.cpp:218] Iteration 6204 (2.38892 iter/s, 5.0232s/12 iters), loss = 3.64701
I0409 20:47:27.779496 15472 solver.cpp:237]     Train net output #0: loss = 3.64701 (* 1 = 3.64701 loss)
I0409 20:47:27.779507 15472 sgd_solver.cpp:105] Iteration 6204, lr = 0.00292607
I0409 20:47:32.658736 15472 solver.cpp:218] Iteration 6216 (2.45947 iter/s, 4.8791s/12 iters), loss = 3.42628
I0409 20:47:32.658828 15472 solver.cpp:237]     Train net output #0: loss = 3.42628 (* 1 = 3.42628 loss)
I0409 20:47:32.658840 15472 sgd_solver.cpp:105] Iteration 6216, lr = 0.00291912
I0409 20:47:34.653395 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_6222.caffemodel
I0409 20:47:35.489698 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_6222.solverstate
I0409 20:47:36.149238 15472 solver.cpp:330] Iteration 6222, Testing net (#0)
I0409 20:47:36.149262 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:47:38.190260 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:47:39.194417 15472 blocking_queue.cpp:49] Waiting for data
I0409 20:47:40.646641 15472 solver.cpp:397]     Test net output #0: accuracy = 0.0980392
I0409 20:47:40.646688 15472 solver.cpp:397]     Test net output #1: loss = 3.76116 (* 1 = 3.76116 loss)
I0409 20:47:42.489145 15472 solver.cpp:218] Iteration 6228 (1.22075 iter/s, 9.83005s/12 iters), loss = 3.5172
I0409 20:47:42.489195 15472 solver.cpp:237]     Train net output #0: loss = 3.5172 (* 1 = 3.5172 loss)
I0409 20:47:42.489208 15472 sgd_solver.cpp:105] Iteration 6228, lr = 0.00291219
I0409 20:47:47.271605 15472 solver.cpp:218] Iteration 6240 (2.50927 iter/s, 4.78227s/12 iters), loss = 3.68315
I0409 20:47:47.271654 15472 solver.cpp:237]     Train net output #0: loss = 3.68315 (* 1 = 3.68315 loss)
I0409 20:47:47.271668 15472 sgd_solver.cpp:105] Iteration 6240, lr = 0.00290528
I0409 20:47:52.164131 15472 solver.cpp:218] Iteration 6252 (2.45282 iter/s, 4.89233s/12 iters), loss = 3.42794
I0409 20:47:52.164178 15472 solver.cpp:237]     Train net output #0: loss = 3.42794 (* 1 = 3.42794 loss)
I0409 20:47:52.164188 15472 sgd_solver.cpp:105] Iteration 6252, lr = 0.00289838
I0409 20:47:57.116747 15472 solver.cpp:218] Iteration 6264 (2.42306 iter/s, 4.95242s/12 iters), loss = 3.57695
I0409 20:47:57.116794 15472 solver.cpp:237]     Train net output #0: loss = 3.57695 (* 1 = 3.57695 loss)
I0409 20:47:57.116806 15472 sgd_solver.cpp:105] Iteration 6264, lr = 0.0028915
I0409 20:47:59.760821 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:48:01.947772 15472 solver.cpp:218] Iteration 6276 (2.48404 iter/s, 4.83083s/12 iters), loss = 3.42151
I0409 20:48:01.947821 15472 solver.cpp:237]     Train net output #0: loss = 3.42151 (* 1 = 3.42151 loss)
I0409 20:48:01.947834 15472 sgd_solver.cpp:105] Iteration 6276, lr = 0.00288463
I0409 20:48:06.885025 15472 solver.cpp:218] Iteration 6288 (2.4306 iter/s, 4.93706s/12 iters), loss = 3.64371
I0409 20:48:06.885175 15472 solver.cpp:237]     Train net output #0: loss = 3.64371 (* 1 = 3.64371 loss)
I0409 20:48:06.885190 15472 sgd_solver.cpp:105] Iteration 6288, lr = 0.00287779
I0409 20:48:11.743732 15472 solver.cpp:218] Iteration 6300 (2.46994 iter/s, 4.85842s/12 iters), loss = 3.48345
I0409 20:48:11.743773 15472 solver.cpp:237]     Train net output #0: loss = 3.48345 (* 1 = 3.48345 loss)
I0409 20:48:11.743782 15472 sgd_solver.cpp:105] Iteration 6300, lr = 0.00287095
I0409 20:48:16.603510 15472 solver.cpp:218] Iteration 6312 (2.46935 iter/s, 4.85959s/12 iters), loss = 3.74387
I0409 20:48:16.603555 15472 solver.cpp:237]     Train net output #0: loss = 3.74387 (* 1 = 3.74387 loss)
I0409 20:48:16.603564 15472 sgd_solver.cpp:105] Iteration 6312, lr = 0.00286414
I0409 20:48:21.036398 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_6324.caffemodel
I0409 20:48:21.474589 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_6324.solverstate
I0409 20:48:21.812497 15472 solver.cpp:330] Iteration 6324, Testing net (#0)
I0409 20:48:21.812517 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:48:23.777635 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:48:26.251893 15472 solver.cpp:397]     Test net output #0: accuracy = 0.0955882
I0409 20:48:26.251945 15472 solver.cpp:397]     Test net output #1: loss = 3.69199 (* 1 = 3.69199 loss)
I0409 20:48:26.334971 15472 solver.cpp:218] Iteration 6324 (1.23316 iter/s, 9.73114s/12 iters), loss = 3.71924
I0409 20:48:26.335021 15472 solver.cpp:237]     Train net output #0: loss = 3.71924 (* 1 = 3.71924 loss)
I0409 20:48:26.335032 15472 sgd_solver.cpp:105] Iteration 6324, lr = 0.00285734
I0409 20:48:30.483497 15472 solver.cpp:218] Iteration 6336 (2.89272 iter/s, 4.14835s/12 iters), loss = 3.5156
I0409 20:48:30.483546 15472 solver.cpp:237]     Train net output #0: loss = 3.5156 (* 1 = 3.5156 loss)
I0409 20:48:30.483556 15472 sgd_solver.cpp:105] Iteration 6336, lr = 0.00285055
I0409 20:48:35.332042 15472 solver.cpp:218] Iteration 6348 (2.47507 iter/s, 4.84835s/12 iters), loss = 3.68571
I0409 20:48:35.332091 15472 solver.cpp:237]     Train net output #0: loss = 3.68571 (* 1 = 3.68571 loss)
I0409 20:48:35.332103 15472 sgd_solver.cpp:105] Iteration 6348, lr = 0.00284379
I0409 20:48:40.136121 15472 solver.cpp:218] Iteration 6360 (2.49799 iter/s, 4.80387s/12 iters), loss = 3.67553
I0409 20:48:40.136241 15472 solver.cpp:237]     Train net output #0: loss = 3.67553 (* 1 = 3.67553 loss)
I0409 20:48:40.136255 15472 sgd_solver.cpp:105] Iteration 6360, lr = 0.00283703
I0409 20:48:44.809142 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:48:44.939960 15472 solver.cpp:218] Iteration 6372 (2.49814 iter/s, 4.80357s/12 iters), loss = 3.50172
I0409 20:48:44.940014 15472 solver.cpp:237]     Train net output #0: loss = 3.50172 (* 1 = 3.50172 loss)
I0409 20:48:44.940026 15472 sgd_solver.cpp:105] Iteration 6372, lr = 0.0028303
I0409 20:48:49.758824 15472 solver.cpp:218] Iteration 6384 (2.49032 iter/s, 4.81867s/12 iters), loss = 3.47658
I0409 20:48:49.758865 15472 solver.cpp:237]     Train net output #0: loss = 3.47658 (* 1 = 3.47658 loss)
I0409 20:48:49.758874 15472 sgd_solver.cpp:105] Iteration 6384, lr = 0.00282358
I0409 20:48:54.586073 15472 solver.cpp:218] Iteration 6396 (2.48599 iter/s, 4.82706s/12 iters), loss = 3.36756
I0409 20:48:54.586114 15472 solver.cpp:237]     Train net output #0: loss = 3.36756 (* 1 = 3.36756 loss)
I0409 20:48:54.586124 15472 sgd_solver.cpp:105] Iteration 6396, lr = 0.00281687
I0409 20:48:59.411864 15472 solver.cpp:218] Iteration 6408 (2.48674 iter/s, 4.82559s/12 iters), loss = 3.87802
I0409 20:48:59.411914 15472 solver.cpp:237]     Train net output #0: loss = 3.87802 (* 1 = 3.87802 loss)
I0409 20:48:59.411926 15472 sgd_solver.cpp:105] Iteration 6408, lr = 0.00281019
I0409 20:49:04.317718 15472 solver.cpp:218] Iteration 6420 (2.44616 iter/s, 4.90565s/12 iters), loss = 3.47277
I0409 20:49:04.317780 15472 solver.cpp:237]     Train net output #0: loss = 3.47277 (* 1 = 3.47277 loss)
I0409 20:49:04.317795 15472 sgd_solver.cpp:105] Iteration 6420, lr = 0.00280351
I0409 20:49:06.231674 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_6426.caffemodel
I0409 20:49:06.708572 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_6426.solverstate
I0409 20:49:07.039350 15472 solver.cpp:330] Iteration 6426, Testing net (#0)
I0409 20:49:07.039378 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:49:08.958053 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:49:11.475122 15472 solver.cpp:397]     Test net output #0: accuracy = 0.119485
I0409 20:49:11.475255 15472 solver.cpp:397]     Test net output #1: loss = 3.5874 (* 1 = 3.5874 loss)
I0409 20:49:13.407749 15472 solver.cpp:218] Iteration 6432 (1.32018 iter/s, 9.0897s/12 iters), loss = 3.48331
I0409 20:49:13.407797 15472 solver.cpp:237]     Train net output #0: loss = 3.48331 (* 1 = 3.48331 loss)
I0409 20:49:13.407809 15472 sgd_solver.cpp:105] Iteration 6432, lr = 0.00279686
I0409 20:49:18.349745 15472 solver.cpp:218] Iteration 6444 (2.42827 iter/s, 4.94179s/12 iters), loss = 3.89137
I0409 20:49:18.349803 15472 solver.cpp:237]     Train net output #0: loss = 3.89137 (* 1 = 3.89137 loss)
I0409 20:49:18.349817 15472 sgd_solver.cpp:105] Iteration 6444, lr = 0.00279022
I0409 20:49:23.298650 15472 solver.cpp:218] Iteration 6456 (2.42489 iter/s, 4.94868s/12 iters), loss = 3.44881
I0409 20:49:23.298712 15472 solver.cpp:237]     Train net output #0: loss = 3.44881 (* 1 = 3.44881 loss)
I0409 20:49:23.298727 15472 sgd_solver.cpp:105] Iteration 6456, lr = 0.00278359
I0409 20:49:28.208902 15472 solver.cpp:218] Iteration 6468 (2.44397 iter/s, 4.91004s/12 iters), loss = 3.75354
I0409 20:49:28.208956 15472 solver.cpp:237]     Train net output #0: loss = 3.75354 (* 1 = 3.75354 loss)
I0409 20:49:28.208967 15472 sgd_solver.cpp:105] Iteration 6468, lr = 0.00277698
I0409 20:49:30.157022 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:49:33.111372 15472 solver.cpp:218] Iteration 6480 (2.44785 iter/s, 4.90226s/12 iters), loss = 3.52602
I0409 20:49:33.111415 15472 solver.cpp:237]     Train net output #0: loss = 3.52602 (* 1 = 3.52602 loss)
I0409 20:49:33.111425 15472 sgd_solver.cpp:105] Iteration 6480, lr = 0.00277039
I0409 20:49:38.014590 15472 solver.cpp:218] Iteration 6492 (2.44748 iter/s, 4.90301s/12 iters), loss = 3.46861
I0409 20:49:38.014652 15472 solver.cpp:237]     Train net output #0: loss = 3.46861 (* 1 = 3.46861 loss)
I0409 20:49:38.014664 15472 sgd_solver.cpp:105] Iteration 6492, lr = 0.00276381
I0409 20:49:43.061103 15472 solver.cpp:218] Iteration 6504 (2.37799 iter/s, 5.04628s/12 iters), loss = 3.22133
I0409 20:49:43.061266 15472 solver.cpp:237]     Train net output #0: loss = 3.22133 (* 1 = 3.22133 loss)
I0409 20:49:43.061277 15472 sgd_solver.cpp:105] Iteration 6504, lr = 0.00275725
I0409 20:49:47.838929 15472 solver.cpp:218] Iteration 6516 (2.51177 iter/s, 4.77751s/12 iters), loss = 3.36499
I0409 20:49:47.838985 15472 solver.cpp:237]     Train net output #0: loss = 3.36499 (* 1 = 3.36499 loss)
I0409 20:49:47.838997 15472 sgd_solver.cpp:105] Iteration 6516, lr = 0.00275071
I0409 20:49:52.197063 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_6528.caffemodel
I0409 20:49:53.081969 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_6528.solverstate
I0409 20:49:53.494346 15472 solver.cpp:330] Iteration 6528, Testing net (#0)
I0409 20:49:53.494371 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:49:55.366475 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:49:58.007333 15472 solver.cpp:397]     Test net output #0: accuracy = 0.129902
I0409 20:49:58.007375 15472 solver.cpp:397]     Test net output #1: loss = 3.4999 (* 1 = 3.4999 loss)
I0409 20:49:58.090291 15472 solver.cpp:218] Iteration 6528 (1.17062 iter/s, 10.251s/12 iters), loss = 3.47298
I0409 20:49:58.090332 15472 solver.cpp:237]     Train net output #0: loss = 3.47298 (* 1 = 3.47298 loss)
I0409 20:49:58.090342 15472 sgd_solver.cpp:105] Iteration 6528, lr = 0.00274418
I0409 20:50:02.372081 15472 solver.cpp:218] Iteration 6540 (2.80269 iter/s, 4.2816s/12 iters), loss = 3.41447
I0409 20:50:02.372141 15472 solver.cpp:237]     Train net output #0: loss = 3.41447 (* 1 = 3.41447 loss)
I0409 20:50:02.372153 15472 sgd_solver.cpp:105] Iteration 6540, lr = 0.00273766
I0409 20:50:07.182283 15472 solver.cpp:218] Iteration 6552 (2.49482 iter/s, 4.80997s/12 iters), loss = 3.6068
I0409 20:50:07.182358 15472 solver.cpp:237]     Train net output #0: loss = 3.6068 (* 1 = 3.6068 loss)
I0409 20:50:07.182375 15472 sgd_solver.cpp:105] Iteration 6552, lr = 0.00273116
I0409 20:50:12.015744 15472 solver.cpp:218] Iteration 6564 (2.48281 iter/s, 4.83324s/12 iters), loss = 3.52947
I0409 20:50:12.015787 15472 solver.cpp:237]     Train net output #0: loss = 3.52947 (* 1 = 3.52947 loss)
I0409 20:50:12.015799 15472 sgd_solver.cpp:105] Iteration 6564, lr = 0.00272468
I0409 20:50:16.177045 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:50:16.924073 15472 solver.cpp:218] Iteration 6576 (2.44493 iter/s, 4.90812s/12 iters), loss = 3.34003
I0409 20:50:16.924127 15472 solver.cpp:237]     Train net output #0: loss = 3.34003 (* 1 = 3.34003 loss)
I0409 20:50:16.924139 15472 sgd_solver.cpp:105] Iteration 6576, lr = 0.00271821
I0409 20:50:21.825119 15472 solver.cpp:218] Iteration 6588 (2.44857 iter/s, 4.90083s/12 iters), loss = 3.57261
I0409 20:50:21.825170 15472 solver.cpp:237]     Train net output #0: loss = 3.57261 (* 1 = 3.57261 loss)
I0409 20:50:21.825181 15472 sgd_solver.cpp:105] Iteration 6588, lr = 0.00271175
I0409 20:50:26.602056 15472 solver.cpp:218] Iteration 6600 (2.51218 iter/s, 4.77673s/12 iters), loss = 3.61047
I0409 20:50:26.602110 15472 solver.cpp:237]     Train net output #0: loss = 3.61047 (* 1 = 3.61047 loss)
I0409 20:50:26.602123 15472 sgd_solver.cpp:105] Iteration 6600, lr = 0.00270532
I0409 20:50:31.523869 15472 solver.cpp:218] Iteration 6612 (2.43824 iter/s, 4.92159s/12 iters), loss = 3.2655
I0409 20:50:31.523927 15472 solver.cpp:237]     Train net output #0: loss = 3.2655 (* 1 = 3.2655 loss)
I0409 20:50:31.523941 15472 sgd_solver.cpp:105] Iteration 6612, lr = 0.00269889
I0409 20:50:36.349874 15472 solver.cpp:218] Iteration 6624 (2.48664 iter/s, 4.82579s/12 iters), loss = 3.26523
I0409 20:50:36.349920 15472 solver.cpp:237]     Train net output #0: loss = 3.26523 (* 1 = 3.26523 loss)
I0409 20:50:36.349931 15472 sgd_solver.cpp:105] Iteration 6624, lr = 0.00269248
I0409 20:50:38.315269 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_6630.caffemodel
I0409 20:50:38.795487 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_6630.solverstate
I0409 20:50:39.127046 15472 solver.cpp:330] Iteration 6630, Testing net (#0)
I0409 20:50:39.127076 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:50:40.982406 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:50:43.647070 15472 solver.cpp:397]     Test net output #0: accuracy = 0.130515
I0409 20:50:43.647125 15472 solver.cpp:397]     Test net output #1: loss = 3.46486 (* 1 = 3.46486 loss)
I0409 20:50:45.553781 15472 solver.cpp:218] Iteration 6636 (1.30384 iter/s, 9.20357s/12 iters), loss = 3.39567
I0409 20:50:45.553834 15472 solver.cpp:237]     Train net output #0: loss = 3.39567 (* 1 = 3.39567 loss)
I0409 20:50:45.553846 15472 sgd_solver.cpp:105] Iteration 6636, lr = 0.00268609
I0409 20:50:50.460345 15472 solver.cpp:218] Iteration 6648 (2.44581 iter/s, 4.90635s/12 iters), loss = 3.56036
I0409 20:50:50.460487 15472 solver.cpp:237]     Train net output #0: loss = 3.56036 (* 1 = 3.56036 loss)
I0409 20:50:50.460501 15472 sgd_solver.cpp:105] Iteration 6648, lr = 0.00267971
I0409 20:50:55.326189 15472 solver.cpp:218] Iteration 6660 (2.46633 iter/s, 4.86554s/12 iters), loss = 3.52523
I0409 20:50:55.326242 15472 solver.cpp:237]     Train net output #0: loss = 3.52523 (* 1 = 3.52523 loss)
I0409 20:50:55.326254 15472 sgd_solver.cpp:105] Iteration 6660, lr = 0.00267335
I0409 20:51:00.306813 15472 solver.cpp:218] Iteration 6672 (2.40944 iter/s, 4.9804s/12 iters), loss = 3.38719
I0409 20:51:00.306862 15472 solver.cpp:237]     Train net output #0: loss = 3.38719 (* 1 = 3.38719 loss)
I0409 20:51:00.306871 15472 sgd_solver.cpp:105] Iteration 6672, lr = 0.00266701
I0409 20:51:01.622154 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:51:05.225033 15472 solver.cpp:218] Iteration 6684 (2.44001 iter/s, 4.918s/12 iters), loss = 3.38488
I0409 20:51:05.225085 15472 solver.cpp:237]     Train net output #0: loss = 3.38488 (* 1 = 3.38488 loss)
I0409 20:51:05.225098 15472 sgd_solver.cpp:105] Iteration 6684, lr = 0.00266067
I0409 20:51:10.131637 15472 solver.cpp:218] Iteration 6696 (2.44579 iter/s, 4.90638s/12 iters), loss = 3.40818
I0409 20:51:10.131693 15472 solver.cpp:237]     Train net output #0: loss = 3.40818 (* 1 = 3.40818 loss)
I0409 20:51:10.131707 15472 sgd_solver.cpp:105] Iteration 6696, lr = 0.00265436
I0409 20:51:15.000615 15472 solver.cpp:218] Iteration 6708 (2.4647 iter/s, 4.86875s/12 iters), loss = 3.45254
I0409 20:51:15.000663 15472 solver.cpp:237]     Train net output #0: loss = 3.45254 (* 1 = 3.45254 loss)
I0409 20:51:15.000674 15472 sgd_solver.cpp:105] Iteration 6708, lr = 0.00264805
I0409 20:51:19.885668 15472 solver.cpp:218] Iteration 6720 (2.45658 iter/s, 4.88483s/12 iters), loss = 3.44664
I0409 20:51:19.885726 15472 solver.cpp:237]     Train net output #0: loss = 3.44664 (* 1 = 3.44664 loss)
I0409 20:51:19.885740 15472 sgd_solver.cpp:105] Iteration 6720, lr = 0.00264177
I0409 20:51:24.395818 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_6732.caffemodel
I0409 20:51:26.445096 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_6732.solverstate
I0409 20:51:27.199656 15472 solver.cpp:330] Iteration 6732, Testing net (#0)
I0409 20:51:27.199678 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:51:29.001164 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:51:31.632453 15472 solver.cpp:397]     Test net output #0: accuracy = 0.129289
I0409 20:51:31.632479 15472 solver.cpp:397]     Test net output #1: loss = 3.454 (* 1 = 3.454 loss)
I0409 20:51:31.714905 15472 solver.cpp:218] Iteration 6732 (1.01447 iter/s, 11.8288s/12 iters), loss = 3.26234
I0409 20:51:31.714964 15472 solver.cpp:237]     Train net output #0: loss = 3.26234 (* 1 = 3.26234 loss)
I0409 20:51:31.714977 15472 sgd_solver.cpp:105] Iteration 6732, lr = 0.0026355
I0409 20:51:36.129637 15472 solver.cpp:218] Iteration 6744 (2.7183 iter/s, 4.41452s/12 iters), loss = 3.332
I0409 20:51:36.129688 15472 solver.cpp:237]     Train net output #0: loss = 3.332 (* 1 = 3.332 loss)
I0409 20:51:36.129700 15472 sgd_solver.cpp:105] Iteration 6744, lr = 0.00262924
I0409 20:51:41.473989 15472 solver.cpp:218] Iteration 6756 (2.24546 iter/s, 5.34411s/12 iters), loss = 3.23537
I0409 20:51:41.474035 15472 solver.cpp:237]     Train net output #0: loss = 3.23537 (* 1 = 3.23537 loss)
I0409 20:51:41.474046 15472 sgd_solver.cpp:105] Iteration 6756, lr = 0.002623
I0409 20:51:46.329159 15472 solver.cpp:218] Iteration 6768 (2.47171 iter/s, 4.85495s/12 iters), loss = 3.35405
I0409 20:51:46.329216 15472 solver.cpp:237]     Train net output #0: loss = 3.35405 (* 1 = 3.35405 loss)
I0409 20:51:46.329228 15472 sgd_solver.cpp:105] Iteration 6768, lr = 0.00261677
I0409 20:51:49.710175 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:51:51.155318 15472 solver.cpp:218] Iteration 6780 (2.48657 iter/s, 4.82593s/12 iters), loss = 3.49389
I0409 20:51:51.155369 15472 solver.cpp:237]     Train net output #0: loss = 3.49389 (* 1 = 3.49389 loss)
I0409 20:51:51.155381 15472 sgd_solver.cpp:105] Iteration 6780, lr = 0.00261056
I0409 20:51:56.032831 15472 solver.cpp:218] Iteration 6792 (2.46038 iter/s, 4.87729s/12 iters), loss = 3.29095
I0409 20:51:56.033013 15472 solver.cpp:237]     Train net output #0: loss = 3.29095 (* 1 = 3.29095 loss)
I0409 20:51:56.033031 15472 sgd_solver.cpp:105] Iteration 6792, lr = 0.00260436
I0409 20:52:00.880584 15472 solver.cpp:218] Iteration 6804 (2.47555 iter/s, 4.8474s/12 iters), loss = 3.43653
I0409 20:52:00.880640 15472 solver.cpp:237]     Train net output #0: loss = 3.43653 (* 1 = 3.43653 loss)
I0409 20:52:00.880653 15472 sgd_solver.cpp:105] Iteration 6804, lr = 0.00259817
I0409 20:52:05.687299 15472 solver.cpp:218] Iteration 6816 (2.49663 iter/s, 4.80649s/12 iters), loss = 3.4757
I0409 20:52:05.687355 15472 solver.cpp:237]     Train net output #0: loss = 3.4757 (* 1 = 3.4757 loss)
I0409 20:52:05.687366 15472 sgd_solver.cpp:105] Iteration 6816, lr = 0.00259201
I0409 20:52:10.560355 15472 solver.cpp:218] Iteration 6828 (2.46264 iter/s, 4.87283s/12 iters), loss = 3.27689
I0409 20:52:10.560413 15472 solver.cpp:237]     Train net output #0: loss = 3.27689 (* 1 = 3.27689 loss)
I0409 20:52:10.560426 15472 sgd_solver.cpp:105] Iteration 6828, lr = 0.00258585
I0409 20:52:12.551192 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_6834.caffemodel
I0409 20:52:13.320381 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_6834.solverstate
I0409 20:52:13.813807 15472 solver.cpp:330] Iteration 6834, Testing net (#0)
I0409 20:52:13.813838 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:52:15.439939 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:52:18.109366 15472 solver.cpp:397]     Test net output #0: accuracy = 0.143382
I0409 20:52:18.109421 15472 solver.cpp:397]     Test net output #1: loss = 3.44164 (* 1 = 3.44164 loss)
I0409 20:52:19.900936 15472 solver.cpp:218] Iteration 6840 (1.28477 iter/s, 9.3402s/12 iters), loss = 3.3475
I0409 20:52:19.900995 15472 solver.cpp:237]     Train net output #0: loss = 3.3475 (* 1 = 3.3475 loss)
I0409 20:52:19.901007 15472 sgd_solver.cpp:105] Iteration 6840, lr = 0.00257971
I0409 20:52:24.998297 15472 solver.cpp:218] Iteration 6852 (2.35427 iter/s, 5.09712s/12 iters), loss = 3.35743
I0409 20:52:24.998351 15472 solver.cpp:237]     Train net output #0: loss = 3.35743 (* 1 = 3.35743 loss)
I0409 20:52:24.998363 15472 sgd_solver.cpp:105] Iteration 6852, lr = 0.00257359
I0409 20:52:29.987263 15472 solver.cpp:218] Iteration 6864 (2.40542 iter/s, 4.98873s/12 iters), loss = 3.30061
I0409 20:52:29.987393 15472 solver.cpp:237]     Train net output #0: loss = 3.30061 (* 1 = 3.30061 loss)
I0409 20:52:29.987407 15472 sgd_solver.cpp:105] Iteration 6864, lr = 0.00256748
I0409 20:52:34.801504 15472 solver.cpp:218] Iteration 6876 (2.49276 iter/s, 4.81395s/12 iters), loss = 3.58035
I0409 20:52:34.801543 15472 solver.cpp:237]     Train net output #0: loss = 3.58035 (* 1 = 3.58035 loss)
I0409 20:52:34.801553 15472 sgd_solver.cpp:105] Iteration 6876, lr = 0.00256138
I0409 20:52:35.399392 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:52:39.622294 15472 solver.cpp:218] Iteration 6888 (2.48933 iter/s, 4.82057s/12 iters), loss = 3.219
I0409 20:52:39.622342 15472 solver.cpp:237]     Train net output #0: loss = 3.219 (* 1 = 3.219 loss)
I0409 20:52:39.622352 15472 sgd_solver.cpp:105] Iteration 6888, lr = 0.0025553
I0409 20:52:44.424100 15472 solver.cpp:218] Iteration 6900 (2.49918 iter/s, 4.80158s/12 iters), loss = 3.36256
I0409 20:52:44.424160 15472 solver.cpp:237]     Train net output #0: loss = 3.36256 (* 1 = 3.36256 loss)
I0409 20:52:44.424172 15472 sgd_solver.cpp:105] Iteration 6900, lr = 0.00254923
I0409 20:52:49.220742 15472 solver.cpp:218] Iteration 6912 (2.50187 iter/s, 4.79642s/12 iters), loss = 3.42122
I0409 20:52:49.220785 15472 solver.cpp:237]     Train net output #0: loss = 3.42122 (* 1 = 3.42122 loss)
I0409 20:52:49.220794 15472 sgd_solver.cpp:105] Iteration 6912, lr = 0.00254318
I0409 20:52:54.073649 15472 solver.cpp:218] Iteration 6924 (2.47286 iter/s, 4.85269s/12 iters), loss = 3.19476
I0409 20:52:54.073700 15472 solver.cpp:237]     Train net output #0: loss = 3.19476 (* 1 = 3.19476 loss)
I0409 20:52:54.073712 15472 sgd_solver.cpp:105] Iteration 6924, lr = 0.00253714
I0409 20:52:58.420557 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_6936.caffemodel
I0409 20:52:58.863325 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_6936.solverstate
I0409 20:52:59.182153 15472 solver.cpp:330] Iteration 6936, Testing net (#0)
I0409 20:52:59.182173 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:52:59.552428 15472 blocking_queue.cpp:49] Waiting for data
I0409 20:53:00.920151 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:53:03.627578 15472 solver.cpp:397]     Test net output #0: accuracy = 0.150123
I0409 20:53:03.627606 15472 solver.cpp:397]     Test net output #1: loss = 3.34542 (* 1 = 3.34542 loss)
I0409 20:53:03.710165 15472 solver.cpp:218] Iteration 6936 (1.24531 iter/s, 9.63613s/12 iters), loss = 3.05694
I0409 20:53:03.710220 15472 solver.cpp:237]     Train net output #0: loss = 3.05694 (* 1 = 3.05694 loss)
I0409 20:53:03.710232 15472 sgd_solver.cpp:105] Iteration 6936, lr = 0.00253112
I0409 20:53:07.973599 15472 solver.cpp:218] Iteration 6948 (2.81477 iter/s, 4.26323s/12 iters), loss = 3.27471
I0409 20:53:07.973637 15472 solver.cpp:237]     Train net output #0: loss = 3.27471 (* 1 = 3.27471 loss)
I0409 20:53:07.973646 15472 sgd_solver.cpp:105] Iteration 6948, lr = 0.00252511
I0409 20:53:12.759352 15472 solver.cpp:218] Iteration 6960 (2.50756 iter/s, 4.78553s/12 iters), loss = 3.10164
I0409 20:53:12.759407 15472 solver.cpp:237]     Train net output #0: loss = 3.10164 (* 1 = 3.10164 loss)
I0409 20:53:12.759418 15472 sgd_solver.cpp:105] Iteration 6960, lr = 0.00251911
I0409 20:53:17.693509 15472 solver.cpp:218] Iteration 6972 (2.43214 iter/s, 4.93393s/12 iters), loss = 3.30347
I0409 20:53:17.693552 15472 solver.cpp:237]     Train net output #0: loss = 3.30347 (* 1 = 3.30347 loss)
I0409 20:53:17.693562 15472 sgd_solver.cpp:105] Iteration 6972, lr = 0.00251313
I0409 20:53:20.393270 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:53:22.579772 15472 solver.cpp:218] Iteration 6984 (2.45598 iter/s, 4.88603s/12 iters), loss = 3.25899
I0409 20:53:22.579833 15472 solver.cpp:237]     Train net output #0: loss = 3.25899 (* 1 = 3.25899 loss)
I0409 20:53:22.579845 15472 sgd_solver.cpp:105] Iteration 6984, lr = 0.00250717
I0409 20:53:27.464154 15472 solver.cpp:218] Iteration 6996 (2.45693 iter/s, 4.88414s/12 iters), loss = 3.19998
I0409 20:53:27.464207 15472 solver.cpp:237]     Train net output #0: loss = 3.19998 (* 1 = 3.19998 loss)
I0409 20:53:27.464219 15472 sgd_solver.cpp:105] Iteration 6996, lr = 0.00250121
I0409 20:53:32.331295 15472 solver.cpp:218] Iteration 7008 (2.46563 iter/s, 4.8669s/12 iters), loss = 3.19341
I0409 20:53:32.331454 15472 solver.cpp:237]     Train net output #0: loss = 3.19341 (* 1 = 3.19341 loss)
I0409 20:53:32.331468 15472 sgd_solver.cpp:105] Iteration 7008, lr = 0.00249528
I0409 20:53:37.180335 15472 solver.cpp:218] Iteration 7020 (2.47489 iter/s, 4.84871s/12 iters), loss = 3.47929
I0409 20:53:37.180374 15472 solver.cpp:237]     Train net output #0: loss = 3.47929 (* 1 = 3.47929 loss)
I0409 20:53:37.180384 15472 sgd_solver.cpp:105] Iteration 7020, lr = 0.00248935
I0409 20:53:42.041049 15472 solver.cpp:218] Iteration 7032 (2.46889 iter/s, 4.86049s/12 iters), loss = 3.11574
I0409 20:53:42.041108 15472 solver.cpp:237]     Train net output #0: loss = 3.11574 (* 1 = 3.11574 loss)
I0409 20:53:42.041121 15472 sgd_solver.cpp:105] Iteration 7032, lr = 0.00248344
I0409 20:53:44.059955 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_7038.caffemodel
I0409 20:53:44.520929 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_7038.solverstate
I0409 20:53:44.847487 15472 solver.cpp:330] Iteration 7038, Testing net (#0)
I0409 20:53:44.847506 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:53:46.547780 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:53:49.573277 15472 solver.cpp:397]     Test net output #0: accuracy = 0.14951
I0409 20:53:49.573310 15472 solver.cpp:397]     Test net output #1: loss = 3.33516 (* 1 = 3.33516 loss)
I0409 20:53:51.486860 15472 solver.cpp:218] Iteration 7044 (1.27046 iter/s, 9.44542s/12 iters), loss = 3.23673
I0409 20:53:51.486909 15472 solver.cpp:237]     Train net output #0: loss = 3.23673 (* 1 = 3.23673 loss)
I0409 20:53:51.486919 15472 sgd_solver.cpp:105] Iteration 7044, lr = 0.00247755
I0409 20:53:56.339826 15472 solver.cpp:218] Iteration 7056 (2.47283 iter/s, 4.85274s/12 iters), loss = 3.21405
I0409 20:53:56.339872 15472 solver.cpp:237]     Train net output #0: loss = 3.21405 (* 1 = 3.21405 loss)
I0409 20:53:56.339882 15472 sgd_solver.cpp:105] Iteration 7056, lr = 0.00247166
I0409 20:54:01.236600 15472 solver.cpp:218] Iteration 7068 (2.45071 iter/s, 4.89654s/12 iters), loss = 3.36101
I0409 20:54:01.236654 15472 solver.cpp:237]     Train net output #0: loss = 3.36101 (* 1 = 3.36101 loss)
I0409 20:54:01.236665 15472 sgd_solver.cpp:105] Iteration 7068, lr = 0.0024658
I0409 20:54:06.002202 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:54:06.104836 15472 solver.cpp:218] Iteration 7080 (2.46508 iter/s, 4.86801s/12 iters), loss = 3.25239
I0409 20:54:06.104880 15472 solver.cpp:237]     Train net output #0: loss = 3.25239 (* 1 = 3.25239 loss)
I0409 20:54:06.104890 15472 sgd_solver.cpp:105] Iteration 7080, lr = 0.00245994
I0409 20:54:10.996170 15472 solver.cpp:218] Iteration 7092 (2.45343 iter/s, 4.89111s/12 iters), loss = 3.13824
I0409 20:54:10.996223 15472 solver.cpp:237]     Train net output #0: loss = 3.13824 (* 1 = 3.13824 loss)
I0409 20:54:10.996237 15472 sgd_solver.cpp:105] Iteration 7092, lr = 0.0024541
I0409 20:54:15.903231 15472 solver.cpp:218] Iteration 7104 (2.44557 iter/s, 4.90683s/12 iters), loss = 3.01861
I0409 20:54:15.903275 15472 solver.cpp:237]     Train net output #0: loss = 3.01861 (* 1 = 3.01861 loss)
I0409 20:54:15.903283 15472 sgd_solver.cpp:105] Iteration 7104, lr = 0.00244827
I0409 20:54:20.705530 15472 solver.cpp:218] Iteration 7116 (2.49892 iter/s, 4.80207s/12 iters), loss = 3.283
I0409 20:54:20.705580 15472 solver.cpp:237]     Train net output #0: loss = 3.283 (* 1 = 3.283 loss)
I0409 20:54:20.705591 15472 sgd_solver.cpp:105] Iteration 7116, lr = 0.00244246
I0409 20:54:25.543660 15472 solver.cpp:218] Iteration 7128 (2.48042 iter/s, 4.8379s/12 iters), loss = 3.05029
I0409 20:54:25.543715 15472 solver.cpp:237]     Train net output #0: loss = 3.05029 (* 1 = 3.05029 loss)
I0409 20:54:25.543727 15472 sgd_solver.cpp:105] Iteration 7128, lr = 0.00243666
I0409 20:54:29.938446 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_7140.caffemodel
I0409 20:54:30.968451 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_7140.solverstate
I0409 20:54:31.286213 15472 solver.cpp:330] Iteration 7140, Testing net (#0)
I0409 20:54:31.286234 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:54:32.860020 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:54:35.735852 15472 solver.cpp:397]     Test net output #0: accuracy = 0.164828
I0409 20:54:35.735903 15472 solver.cpp:397]     Test net output #1: loss = 3.22237 (* 1 = 3.22237 loss)
I0409 20:54:35.818917 15472 solver.cpp:218] Iteration 7140 (1.1679 iter/s, 10.2748s/12 iters), loss = 3.06038
I0409 20:54:35.818972 15472 solver.cpp:237]     Train net output #0: loss = 3.06038 (* 1 = 3.06038 loss)
I0409 20:54:35.818984 15472 sgd_solver.cpp:105] Iteration 7140, lr = 0.00243088
I0409 20:54:39.818727 15472 solver.cpp:218] Iteration 7152 (3.00031 iter/s, 3.99959s/12 iters), loss = 3.38947
I0409 20:54:39.818897 15472 solver.cpp:237]     Train net output #0: loss = 3.38947 (* 1 = 3.38947 loss)
I0409 20:54:39.818912 15472 sgd_solver.cpp:105] Iteration 7152, lr = 0.00242511
I0409 20:54:44.663143 15472 solver.cpp:218] Iteration 7164 (2.47726 iter/s, 4.84407s/12 iters), loss = 3.22635
I0409 20:54:44.663197 15472 solver.cpp:237]     Train net output #0: loss = 3.22635 (* 1 = 3.22635 loss)
I0409 20:54:44.663208 15472 sgd_solver.cpp:105] Iteration 7164, lr = 0.00241935
I0409 20:54:49.702414 15472 solver.cpp:218] Iteration 7176 (2.38141 iter/s, 5.03903s/12 iters), loss = 3.43348
I0409 20:54:49.702467 15472 solver.cpp:237]     Train net output #0: loss = 3.43348 (* 1 = 3.43348 loss)
I0409 20:54:49.702481 15472 sgd_solver.cpp:105] Iteration 7176, lr = 0.0024136
I0409 20:54:51.740208 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:54:54.586179 15472 solver.cpp:218] Iteration 7188 (2.45724 iter/s, 4.88353s/12 iters), loss = 3.17883
I0409 20:54:54.586223 15472 solver.cpp:237]     Train net output #0: loss = 3.17883 (* 1 = 3.17883 loss)
I0409 20:54:54.586232 15472 sgd_solver.cpp:105] Iteration 7188, lr = 0.00240787
I0409 20:54:59.446789 15472 solver.cpp:218] Iteration 7200 (2.46894 iter/s, 4.86038s/12 iters), loss = 3.21064
I0409 20:54:59.446835 15472 solver.cpp:237]     Train net output #0: loss = 3.21064 (* 1 = 3.21064 loss)
I0409 20:54:59.446847 15472 sgd_solver.cpp:105] Iteration 7200, lr = 0.00240216
I0409 20:55:04.311300 15472 solver.cpp:218] Iteration 7212 (2.46696 iter/s, 4.86428s/12 iters), loss = 3.03776
I0409 20:55:04.311352 15472 solver.cpp:237]     Train net output #0: loss = 3.03776 (* 1 = 3.03776 loss)
I0409 20:55:04.311362 15472 sgd_solver.cpp:105] Iteration 7212, lr = 0.00239645
I0409 20:55:09.196650 15472 solver.cpp:218] Iteration 7224 (2.45644 iter/s, 4.88511s/12 iters), loss = 3.05133
I0409 20:55:09.196696 15472 solver.cpp:237]     Train net output #0: loss = 3.05133 (* 1 = 3.05133 loss)
I0409 20:55:09.196707 15472 sgd_solver.cpp:105] Iteration 7224, lr = 0.00239076
I0409 20:55:14.106669 15472 solver.cpp:218] Iteration 7236 (2.4441 iter/s, 4.90978s/12 iters), loss = 3.04177
I0409 20:55:14.106771 15472 solver.cpp:237]     Train net output #0: loss = 3.04177 (* 1 = 3.04177 loss)
I0409 20:55:14.106783 15472 sgd_solver.cpp:105] Iteration 7236, lr = 0.00238509
I0409 20:55:16.094879 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_7242.caffemodel
I0409 20:55:16.954934 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_7242.solverstate
I0409 20:55:17.688591 15472 solver.cpp:330] Iteration 7242, Testing net (#0)
I0409 20:55:17.688622 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:55:19.422216 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:55:22.320835 15472 solver.cpp:397]     Test net output #0: accuracy = 0.140931
I0409 20:55:22.320884 15472 solver.cpp:397]     Test net output #1: loss = 3.32811 (* 1 = 3.32811 loss)
I0409 20:55:24.229339 15472 solver.cpp:218] Iteration 7248 (1.18551 iter/s, 10.1222s/12 iters), loss = 3.23837
I0409 20:55:24.229399 15472 solver.cpp:237]     Train net output #0: loss = 3.23837 (* 1 = 3.23837 loss)
I0409 20:55:24.229413 15472 sgd_solver.cpp:105] Iteration 7248, lr = 0.00237942
I0409 20:55:29.468308 15472 solver.cpp:218] Iteration 7260 (2.29064 iter/s, 5.23871s/12 iters), loss = 3.23362
I0409 20:55:29.468349 15472 solver.cpp:237]     Train net output #0: loss = 3.23362 (* 1 = 3.23362 loss)
I0409 20:55:29.468358 15472 sgd_solver.cpp:105] Iteration 7260, lr = 0.00237378
I0409 20:55:34.462509 15472 solver.cpp:218] Iteration 7272 (2.4029 iter/s, 4.99397s/12 iters), loss = 3.13948
I0409 20:55:34.462560 15472 solver.cpp:237]     Train net output #0: loss = 3.13948 (* 1 = 3.13948 loss)
I0409 20:55:34.462575 15472 sgd_solver.cpp:105] Iteration 7272, lr = 0.00236814
I0409 20:55:38.631114 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:55:39.358306 15472 solver.cpp:218] Iteration 7284 (2.4512 iter/s, 4.89556s/12 iters), loss = 3.16869
I0409 20:55:39.358355 15472 solver.cpp:237]     Train net output #0: loss = 3.16869 (* 1 = 3.16869 loss)
I0409 20:55:39.358367 15472 sgd_solver.cpp:105] Iteration 7284, lr = 0.00236252
I0409 20:55:44.262956 15472 solver.cpp:218] Iteration 7296 (2.44678 iter/s, 4.90441s/12 iters), loss = 3.13335
I0409 20:55:44.263108 15472 solver.cpp:237]     Train net output #0: loss = 3.13335 (* 1 = 3.13335 loss)
I0409 20:55:44.263123 15472 sgd_solver.cpp:105] Iteration 7296, lr = 0.00235691
I0409 20:55:49.132591 15472 solver.cpp:218] Iteration 7308 (2.46442 iter/s, 4.8693s/12 iters), loss = 3.20577
I0409 20:55:49.132634 15472 solver.cpp:237]     Train net output #0: loss = 3.20577 (* 1 = 3.20577 loss)
I0409 20:55:49.132643 15472 sgd_solver.cpp:105] Iteration 7308, lr = 0.00235131
I0409 20:55:53.978238 15472 solver.cpp:218] Iteration 7320 (2.47657 iter/s, 4.84541s/12 iters), loss = 2.93912
I0409 20:55:53.978297 15472 solver.cpp:237]     Train net output #0: loss = 2.93912 (* 1 = 2.93912 loss)
I0409 20:55:53.978310 15472 sgd_solver.cpp:105] Iteration 7320, lr = 0.00234573
I0409 20:55:58.859325 15472 solver.cpp:218] Iteration 7332 (2.45859 iter/s, 4.88084s/12 iters), loss = 3.0887
I0409 20:55:58.859375 15472 solver.cpp:237]     Train net output #0: loss = 3.0887 (* 1 = 3.0887 loss)
I0409 20:55:58.859385 15472 sgd_solver.cpp:105] Iteration 7332, lr = 0.00234016
I0409 20:56:03.286428 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_7344.caffemodel
I0409 20:56:03.712188 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_7344.solverstate
I0409 20:56:04.026259 15472 solver.cpp:330] Iteration 7344, Testing net (#0)
I0409 20:56:04.026300 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:56:05.555054 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:56:08.481792 15472 solver.cpp:397]     Test net output #0: accuracy = 0.170956
I0409 20:56:08.481840 15472 solver.cpp:397]     Test net output #1: loss = 3.20784 (* 1 = 3.20784 loss)
I0409 20:56:08.564874 15472 solver.cpp:218] Iteration 7344 (1.23646 iter/s, 9.70514s/12 iters), loss = 3.07089
I0409 20:56:08.564924 15472 solver.cpp:237]     Train net output #0: loss = 3.07089 (* 1 = 3.07089 loss)
I0409 20:56:08.564935 15472 sgd_solver.cpp:105] Iteration 7344, lr = 0.0023346
I0409 20:56:12.664842 15472 solver.cpp:218] Iteration 7356 (2.927 iter/s, 4.09976s/12 iters), loss = 3.10839
I0409 20:56:12.664888 15472 solver.cpp:237]     Train net output #0: loss = 3.10839 (* 1 = 3.10839 loss)
I0409 20:56:12.664897 15472 sgd_solver.cpp:105] Iteration 7356, lr = 0.00232906
I0409 20:56:17.513332 15472 solver.cpp:218] Iteration 7368 (2.47512 iter/s, 4.84825s/12 iters), loss = 3.17864
I0409 20:56:17.513442 15472 solver.cpp:237]     Train net output #0: loss = 3.17864 (* 1 = 3.17864 loss)
I0409 20:56:17.513456 15472 sgd_solver.cpp:105] Iteration 7368, lr = 0.00232353
I0409 20:56:22.373561 15472 solver.cpp:218] Iteration 7380 (2.46917 iter/s, 4.85993s/12 iters), loss = 2.99362
I0409 20:56:22.373612 15472 solver.cpp:237]     Train net output #0: loss = 2.99362 (* 1 = 2.99362 loss)
I0409 20:56:22.373625 15472 sgd_solver.cpp:105] Iteration 7380, lr = 0.00231802
I0409 20:56:23.725581 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:56:27.243314 15472 solver.cpp:218] Iteration 7392 (2.46431 iter/s, 4.86951s/12 iters), loss = 3.07397
I0409 20:56:27.243367 15472 solver.cpp:237]     Train net output #0: loss = 3.07397 (* 1 = 3.07397 loss)
I0409 20:56:27.243379 15472 sgd_solver.cpp:105] Iteration 7392, lr = 0.00231251
I0409 20:56:32.050424 15472 solver.cpp:218] Iteration 7404 (2.49643 iter/s, 4.80687s/12 iters), loss = 3.02843
I0409 20:56:32.050464 15472 solver.cpp:237]     Train net output #0: loss = 3.02843 (* 1 = 3.02843 loss)
I0409 20:56:32.050473 15472 sgd_solver.cpp:105] Iteration 7404, lr = 0.00230702
I0409 20:56:36.854591 15472 solver.cpp:218] Iteration 7416 (2.49795 iter/s, 4.80394s/12 iters), loss = 2.97233
I0409 20:56:36.854643 15472 solver.cpp:237]     Train net output #0: loss = 2.97233 (* 1 = 2.97233 loss)
I0409 20:56:36.854657 15472 sgd_solver.cpp:105] Iteration 7416, lr = 0.00230154
I0409 20:56:41.665022 15472 solver.cpp:218] Iteration 7428 (2.4947 iter/s, 4.81019s/12 iters), loss = 3.05682
I0409 20:56:41.665081 15472 solver.cpp:237]     Train net output #0: loss = 3.05682 (* 1 = 3.05682 loss)
I0409 20:56:41.665093 15472 sgd_solver.cpp:105] Iteration 7428, lr = 0.00229608
I0409 20:56:46.494643 15472 solver.cpp:218] Iteration 7440 (2.48479 iter/s, 4.82937s/12 iters), loss = 3.16783
I0409 20:56:46.494693 15472 solver.cpp:237]     Train net output #0: loss = 3.16783 (* 1 = 3.16783 loss)
I0409 20:56:46.494704 15472 sgd_solver.cpp:105] Iteration 7440, lr = 0.00229063
I0409 20:56:48.427037 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_7446.caffemodel
I0409 20:56:48.882138 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_7446.solverstate
I0409 20:56:49.198273 15472 solver.cpp:330] Iteration 7446, Testing net (#0)
I0409 20:56:49.198302 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:56:50.766134 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:56:53.671670 15472 solver.cpp:397]     Test net output #0: accuracy = 0.172181
I0409 20:56:53.671717 15472 solver.cpp:397]     Test net output #1: loss = 3.19673 (* 1 = 3.19673 loss)
I0409 20:56:55.519786 15472 solver.cpp:218] Iteration 7452 (1.32968 iter/s, 9.02475s/12 iters), loss = 2.9877
I0409 20:56:55.519836 15472 solver.cpp:237]     Train net output #0: loss = 2.9877 (* 1 = 2.9877 loss)
I0409 20:56:55.519846 15472 sgd_solver.cpp:105] Iteration 7452, lr = 0.00228519
I0409 20:57:00.516326 15472 solver.cpp:218] Iteration 7464 (2.40178 iter/s, 4.9963s/12 iters), loss = 3.09372
I0409 20:57:00.516369 15472 solver.cpp:237]     Train net output #0: loss = 3.09372 (* 1 = 3.09372 loss)
I0409 20:57:00.516378 15472 sgd_solver.cpp:105] Iteration 7464, lr = 0.00227976
I0409 20:57:05.340874 15472 solver.cpp:218] Iteration 7476 (2.4874 iter/s, 4.82432s/12 iters), loss = 2.85786
I0409 20:57:05.340924 15472 solver.cpp:237]     Train net output #0: loss = 2.85786 (* 1 = 2.85786 loss)
I0409 20:57:05.340935 15472 sgd_solver.cpp:105] Iteration 7476, lr = 0.00227435
I0409 20:57:08.766237 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:57:10.194530 15472 solver.cpp:218] Iteration 7488 (2.47249 iter/s, 4.85341s/12 iters), loss = 2.92309
I0409 20:57:10.194583 15472 solver.cpp:237]     Train net output #0: loss = 2.92309 (* 1 = 2.92309 loss)
I0409 20:57:10.194597 15472 sgd_solver.cpp:105] Iteration 7488, lr = 0.00226895
I0409 20:57:14.993975 15472 solver.cpp:218] Iteration 7500 (2.50042 iter/s, 4.79919s/12 iters), loss = 2.97817
I0409 20:57:14.994029 15472 solver.cpp:237]     Train net output #0: loss = 2.97817 (* 1 = 2.97817 loss)
I0409 20:57:14.994040 15472 sgd_solver.cpp:105] Iteration 7500, lr = 0.00226357
I0409 20:57:19.824999 15472 solver.cpp:218] Iteration 7512 (2.48407 iter/s, 4.83077s/12 iters), loss = 3.14076
I0409 20:57:19.825124 15472 solver.cpp:237]     Train net output #0: loss = 3.14076 (* 1 = 3.14076 loss)
I0409 20:57:19.825137 15472 sgd_solver.cpp:105] Iteration 7512, lr = 0.00225819
I0409 20:57:24.667922 15472 solver.cpp:218] Iteration 7524 (2.478 iter/s, 4.84261s/12 iters), loss = 2.92096
I0409 20:57:24.667968 15472 solver.cpp:237]     Train net output #0: loss = 2.92096 (* 1 = 2.92096 loss)
I0409 20:57:24.667977 15472 sgd_solver.cpp:105] Iteration 7524, lr = 0.00225283
I0409 20:57:29.704073 15472 solver.cpp:218] Iteration 7536 (2.38289 iter/s, 5.0359s/12 iters), loss = 2.98462
I0409 20:57:29.704133 15472 solver.cpp:237]     Train net output #0: loss = 2.98462 (* 1 = 2.98462 loss)
I0409 20:57:29.704145 15472 sgd_solver.cpp:105] Iteration 7536, lr = 0.00224748
I0409 20:57:34.144991 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_7548.caffemodel
I0409 20:57:34.581558 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_7548.solverstate
I0409 20:57:34.898406 15472 solver.cpp:330] Iteration 7548, Testing net (#0)
I0409 20:57:34.898427 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:57:36.414286 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:57:39.417213 15472 solver.cpp:397]     Test net output #0: accuracy = 0.178309
I0409 20:57:39.417254 15472 solver.cpp:397]     Test net output #1: loss = 3.14873 (* 1 = 3.14873 loss)
I0409 20:57:39.500129 15472 solver.cpp:218] Iteration 7548 (1.22504 iter/s, 9.79562s/12 iters), loss = 2.94625
I0409 20:57:39.500175 15472 solver.cpp:237]     Train net output #0: loss = 2.94625 (* 1 = 2.94625 loss)
I0409 20:57:39.500185 15472 sgd_solver.cpp:105] Iteration 7548, lr = 0.00224215
I0409 20:57:43.606804 15472 solver.cpp:218] Iteration 7560 (2.92223 iter/s, 4.10646s/12 iters), loss = 2.96504
I0409 20:57:43.606861 15472 solver.cpp:237]     Train net output #0: loss = 2.96504 (* 1 = 2.96504 loss)
I0409 20:57:43.606874 15472 sgd_solver.cpp:105] Iteration 7560, lr = 0.00223682
I0409 20:57:48.467358 15472 solver.cpp:218] Iteration 7572 (2.46898 iter/s, 4.8603s/12 iters), loss = 2.93287
I0409 20:57:48.467413 15472 solver.cpp:237]     Train net output #0: loss = 2.93287 (* 1 = 2.93287 loss)
I0409 20:57:48.467427 15472 sgd_solver.cpp:105] Iteration 7572, lr = 0.00223151
I0409 20:57:53.384465 15472 solver.cpp:218] Iteration 7584 (2.44058 iter/s, 4.91686s/12 iters), loss = 2.93055
I0409 20:57:53.384608 15472 solver.cpp:237]     Train net output #0: loss = 2.93055 (* 1 = 2.93055 loss)
I0409 20:57:53.384621 15472 sgd_solver.cpp:105] Iteration 7584, lr = 0.00222621
I0409 20:57:54.025419 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:57:58.269042 15472 solver.cpp:218] Iteration 7596 (2.45688 iter/s, 4.88424s/12 iters), loss = 2.85567
I0409 20:57:58.269091 15472 solver.cpp:237]     Train net output #0: loss = 2.85567 (* 1 = 2.85567 loss)
I0409 20:57:58.269104 15472 sgd_solver.cpp:105] Iteration 7596, lr = 0.00222093
I0409 20:58:03.107153 15472 solver.cpp:218] Iteration 7608 (2.48043 iter/s, 4.83786s/12 iters), loss = 3.10441
I0409 20:58:03.107218 15472 solver.cpp:237]     Train net output #0: loss = 3.10441 (* 1 = 3.10441 loss)
I0409 20:58:03.107231 15472 sgd_solver.cpp:105] Iteration 7608, lr = 0.00221565
I0409 20:58:07.963585 15472 solver.cpp:218] Iteration 7620 (2.47108 iter/s, 4.85618s/12 iters), loss = 2.8217
I0409 20:58:07.963635 15472 solver.cpp:237]     Train net output #0: loss = 2.8217 (* 1 = 2.8217 loss)
I0409 20:58:07.963647 15472 sgd_solver.cpp:105] Iteration 7620, lr = 0.00221039
I0409 20:58:09.152395 15472 blocking_queue.cpp:49] Waiting for data
I0409 20:58:12.855743 15472 solver.cpp:218] Iteration 7632 (2.45303 iter/s, 4.8919s/12 iters), loss = 2.80486
I0409 20:58:12.855801 15472 solver.cpp:237]     Train net output #0: loss = 2.80486 (* 1 = 2.80486 loss)
I0409 20:58:12.855813 15472 sgd_solver.cpp:105] Iteration 7632, lr = 0.00220515
I0409 20:58:17.790773 15472 solver.cpp:218] Iteration 7644 (2.43172 iter/s, 4.93477s/12 iters), loss = 2.91935
I0409 20:58:17.790819 15472 solver.cpp:237]     Train net output #0: loss = 2.91935 (* 1 = 2.91935 loss)
I0409 20:58:17.790829 15472 sgd_solver.cpp:105] Iteration 7644, lr = 0.00219991
I0409 20:58:19.768955 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_7650.caffemodel
I0409 20:58:20.231544 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_7650.solverstate
I0409 20:58:20.551265 15472 solver.cpp:330] Iteration 7650, Testing net (#0)
I0409 20:58:20.551293 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:58:21.865980 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:58:24.854794 15472 solver.cpp:397]     Test net output #0: accuracy = 0.183211
I0409 20:58:24.854912 15472 solver.cpp:397]     Test net output #1: loss = 3.13513 (* 1 = 3.13513 loss)
I0409 20:58:26.600025 15472 solver.cpp:218] Iteration 7656 (1.36226 iter/s, 8.80887s/12 iters), loss = 2.88013
I0409 20:58:26.600070 15472 solver.cpp:237]     Train net output #0: loss = 2.88013 (* 1 = 2.88013 loss)
I0409 20:58:26.600078 15472 sgd_solver.cpp:105] Iteration 7656, lr = 0.00219469
I0409 20:58:31.507138 15472 solver.cpp:218] Iteration 7668 (2.44555 iter/s, 4.90687s/12 iters), loss = 2.88028
I0409 20:58:31.507180 15472 solver.cpp:237]     Train net output #0: loss = 2.88028 (* 1 = 2.88028 loss)
I0409 20:58:31.507189 15472 sgd_solver.cpp:105] Iteration 7668, lr = 0.00218948
I0409 20:58:36.460371 15472 solver.cpp:218] Iteration 7680 (2.42278 iter/s, 4.95299s/12 iters), loss = 2.95022
I0409 20:58:36.460417 15472 solver.cpp:237]     Train net output #0: loss = 2.95022 (* 1 = 2.95022 loss)
I0409 20:58:36.460427 15472 sgd_solver.cpp:105] Iteration 7680, lr = 0.00218428
I0409 20:58:39.219143 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:58:41.401751 15472 solver.cpp:218] Iteration 7692 (2.42859 iter/s, 4.94113s/12 iters), loss = 2.82885
I0409 20:58:41.401801 15472 solver.cpp:237]     Train net output #0: loss = 2.82885 (* 1 = 2.82885 loss)
I0409 20:58:41.401813 15472 sgd_solver.cpp:105] Iteration 7692, lr = 0.00217909
I0409 20:58:46.264106 15472 solver.cpp:218] Iteration 7704 (2.46807 iter/s, 4.8621s/12 iters), loss = 2.83858
I0409 20:58:46.264168 15472 solver.cpp:237]     Train net output #0: loss = 2.83858 (* 1 = 2.83858 loss)
I0409 20:58:46.264179 15472 sgd_solver.cpp:105] Iteration 7704, lr = 0.00217392
I0409 20:58:51.222962 15472 solver.cpp:218] Iteration 7716 (2.42004 iter/s, 4.9586s/12 iters), loss = 2.92414
I0409 20:58:51.223017 15472 solver.cpp:237]     Train net output #0: loss = 2.92414 (* 1 = 2.92414 loss)
I0409 20:58:51.223032 15472 sgd_solver.cpp:105] Iteration 7716, lr = 0.00216876
I0409 20:58:56.108907 15472 solver.cpp:218] Iteration 7728 (2.45615 iter/s, 4.8857s/12 iters), loss = 2.96748
I0409 20:58:56.108971 15472 solver.cpp:237]     Train net output #0: loss = 2.96748 (* 1 = 2.96748 loss)
I0409 20:58:56.108980 15472 sgd_solver.cpp:105] Iteration 7728, lr = 0.00216361
I0409 20:59:00.931052 15472 solver.cpp:218] Iteration 7740 (2.48865 iter/s, 4.82189s/12 iters), loss = 2.95598
I0409 20:59:00.931100 15472 solver.cpp:237]     Train net output #0: loss = 2.95598 (* 1 = 2.95598 loss)
I0409 20:59:00.931111 15472 sgd_solver.cpp:105] Iteration 7740, lr = 0.00215847
I0409 20:59:05.319000 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_7752.caffemodel
I0409 20:59:05.776612 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_7752.solverstate
I0409 20:59:06.088338 15472 solver.cpp:330] Iteration 7752, Testing net (#0)
I0409 20:59:06.088358 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:59:07.452559 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:59:10.535439 15472 solver.cpp:397]     Test net output #0: accuracy = 0.184436
I0409 20:59:10.535470 15472 solver.cpp:397]     Test net output #1: loss = 3.10437 (* 1 = 3.10437 loss)
I0409 20:59:10.618150 15472 solver.cpp:218] Iteration 7752 (1.23882 iter/s, 9.68667s/12 iters), loss = 2.9038
I0409 20:59:10.618194 15472 solver.cpp:237]     Train net output #0: loss = 2.9038 (* 1 = 2.9038 loss)
I0409 20:59:10.618202 15472 sgd_solver.cpp:105] Iteration 7752, lr = 0.00215335
I0409 20:59:14.792310 15472 solver.cpp:218] Iteration 7764 (2.87498 iter/s, 4.17394s/12 iters), loss = 2.98358
I0409 20:59:14.792366 15472 solver.cpp:237]     Train net output #0: loss = 2.98358 (* 1 = 2.98358 loss)
I0409 20:59:14.792378 15472 sgd_solver.cpp:105] Iteration 7764, lr = 0.00214823
I0409 20:59:19.672283 15472 solver.cpp:218] Iteration 7776 (2.45916 iter/s, 4.87972s/12 iters), loss = 3.04053
I0409 20:59:19.672320 15472 solver.cpp:237]     Train net output #0: loss = 3.04053 (* 1 = 3.04053 loss)
I0409 20:59:19.672328 15472 sgd_solver.cpp:105] Iteration 7776, lr = 0.00214313
I0409 20:59:24.588809 15472 solver.cpp:218] Iteration 7788 (2.44086 iter/s, 4.91629s/12 iters), loss = 2.7885
I0409 20:59:24.588850 15472 solver.cpp:237]     Train net output #0: loss = 2.7885 (* 1 = 2.7885 loss)
I0409 20:59:24.588857 15472 sgd_solver.cpp:105] Iteration 7788, lr = 0.00213805
I0409 20:59:24.596920 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:59:29.521144 15472 solver.cpp:218] Iteration 7800 (2.43304 iter/s, 4.93209s/12 iters), loss = 2.89739
I0409 20:59:29.521258 15472 solver.cpp:237]     Train net output #0: loss = 2.89739 (* 1 = 2.89739 loss)
I0409 20:59:29.521270 15472 sgd_solver.cpp:105] Iteration 7800, lr = 0.00213297
I0409 20:59:34.405702 15472 solver.cpp:218] Iteration 7812 (2.45688 iter/s, 4.88425s/12 iters), loss = 2.73435
I0409 20:59:34.405757 15472 solver.cpp:237]     Train net output #0: loss = 2.73435 (* 1 = 2.73435 loss)
I0409 20:59:34.405769 15472 sgd_solver.cpp:105] Iteration 7812, lr = 0.00212791
I0409 20:59:39.383181 15472 solver.cpp:218] Iteration 7824 (2.41098 iter/s, 4.97722s/12 iters), loss = 2.99917
I0409 20:59:39.383232 15472 solver.cpp:237]     Train net output #0: loss = 2.99917 (* 1 = 2.99917 loss)
I0409 20:59:39.383245 15472 sgd_solver.cpp:105] Iteration 7824, lr = 0.00212285
I0409 20:59:44.266233 15472 solver.cpp:218] Iteration 7836 (2.45761 iter/s, 4.8828s/12 iters), loss = 2.84466
I0409 20:59:44.266294 15472 solver.cpp:237]     Train net output #0: loss = 2.84466 (* 1 = 2.84466 loss)
I0409 20:59:44.266305 15472 sgd_solver.cpp:105] Iteration 7836, lr = 0.00211781
I0409 20:59:49.224699 15472 solver.cpp:218] Iteration 7848 (2.42023 iter/s, 4.9582s/12 iters), loss = 2.93431
I0409 20:59:49.224752 15472 solver.cpp:237]     Train net output #0: loss = 2.93431 (* 1 = 2.93431 loss)
I0409 20:59:49.224764 15472 sgd_solver.cpp:105] Iteration 7848, lr = 0.00211279
I0409 20:59:51.204108 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_7854.caffemodel
I0409 20:59:52.283396 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_7854.solverstate
I0409 20:59:53.291366 15472 solver.cpp:330] Iteration 7854, Testing net (#0)
I0409 20:59:53.291383 15472 net.cpp:676] Ignoring source layer train-data
I0409 20:59:54.592103 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 20:59:57.707939 15472 solver.cpp:397]     Test net output #0: accuracy = 0.183824
I0409 20:59:57.707985 15472 solver.cpp:397]     Test net output #1: loss = 3.13746 (* 1 = 3.13746 loss)
I0409 20:59:59.601044 15472 solver.cpp:218] Iteration 7860 (1.15653 iter/s, 10.3759s/12 iters), loss = 2.92488
I0409 20:59:59.601125 15472 solver.cpp:237]     Train net output #0: loss = 2.92488 (* 1 = 2.92488 loss)
I0409 20:59:59.601136 15472 sgd_solver.cpp:105] Iteration 7860, lr = 0.00210777
I0409 21:00:04.497597 15472 solver.cpp:218] Iteration 7872 (2.45084 iter/s, 4.89627s/12 iters), loss = 2.79013
I0409 21:00:04.497648 15472 solver.cpp:237]     Train net output #0: loss = 2.79013 (* 1 = 2.79013 loss)
I0409 21:00:04.497659 15472 sgd_solver.cpp:105] Iteration 7872, lr = 0.00210277
I0409 21:00:09.335706 15472 solver.cpp:218] Iteration 7884 (2.48044 iter/s, 4.83786s/12 iters), loss = 2.9466
I0409 21:00:09.335763 15472 solver.cpp:237]     Train net output #0: loss = 2.9466 (* 1 = 2.9466 loss)
I0409 21:00:09.335777 15472 sgd_solver.cpp:105] Iteration 7884, lr = 0.00209777
I0409 21:00:11.381095 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 21:00:14.248085 15472 solver.cpp:218] Iteration 7896 (2.44294 iter/s, 4.91212s/12 iters), loss = 2.86566
I0409 21:00:14.248142 15472 solver.cpp:237]     Train net output #0: loss = 2.86566 (* 1 = 2.86566 loss)
I0409 21:00:14.248157 15472 sgd_solver.cpp:105] Iteration 7896, lr = 0.00209279
I0409 21:00:19.094682 15472 solver.cpp:218] Iteration 7908 (2.4761 iter/s, 4.84634s/12 iters), loss = 2.9829
I0409 21:00:19.094734 15472 solver.cpp:237]     Train net output #0: loss = 2.9829 (* 1 = 2.9829 loss)
I0409 21:00:19.094746 15472 sgd_solver.cpp:105] Iteration 7908, lr = 0.00208782
I0409 21:00:23.897334 15472 solver.cpp:218] Iteration 7920 (2.49875 iter/s, 4.80241s/12 iters), loss = 2.84018
I0409 21:00:23.897385 15472 solver.cpp:237]     Train net output #0: loss = 2.84018 (* 1 = 2.84018 loss)
I0409 21:00:23.897398 15472 sgd_solver.cpp:105] Iteration 7920, lr = 0.00208287
I0409 21:00:28.703316 15472 solver.cpp:218] Iteration 7932 (2.49702 iter/s, 4.80574s/12 iters), loss = 2.93184
I0409 21:00:28.703349 15472 solver.cpp:237]     Train net output #0: loss = 2.93184 (* 1 = 2.93184 loss)
I0409 21:00:28.703357 15472 sgd_solver.cpp:105] Iteration 7932, lr = 0.00207792
I0409 21:00:33.539992 15472 solver.cpp:218] Iteration 7944 (2.48116 iter/s, 4.83644s/12 iters), loss = 2.5704
I0409 21:00:33.540127 15472 solver.cpp:237]     Train net output #0: loss = 2.5704 (* 1 = 2.5704 loss)
I0409 21:00:33.540138 15472 sgd_solver.cpp:105] Iteration 7944, lr = 0.00207299
I0409 21:00:37.899848 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_7956.caffemodel
I0409 21:00:38.352582 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_7956.solverstate
I0409 21:00:38.678138 15472 solver.cpp:330] Iteration 7956, Testing net (#0)
I0409 21:00:38.678156 15472 net.cpp:676] Ignoring source layer train-data
I0409 21:00:39.943509 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 21:00:43.040805 15472 solver.cpp:397]     Test net output #0: accuracy = 0.194853
I0409 21:00:43.040854 15472 solver.cpp:397]     Test net output #1: loss = 3.14429 (* 1 = 3.14429 loss)
I0409 21:00:43.123945 15472 solver.cpp:218] Iteration 7956 (1.25216 iter/s, 9.58344s/12 iters), loss = 2.82174
I0409 21:00:43.123996 15472 solver.cpp:237]     Train net output #0: loss = 2.82174 (* 1 = 2.82174 loss)
I0409 21:00:43.124007 15472 sgd_solver.cpp:105] Iteration 7956, lr = 0.00206807
I0409 21:00:47.376828 15472 solver.cpp:218] Iteration 7968 (2.82176 iter/s, 4.25266s/12 iters), loss = 2.79984
I0409 21:00:47.376871 15472 solver.cpp:237]     Train net output #0: loss = 2.79984 (* 1 = 2.79984 loss)
I0409 21:00:47.376883 15472 sgd_solver.cpp:105] Iteration 7968, lr = 0.00206316
I0409 21:00:52.322289 15472 solver.cpp:218] Iteration 7980 (2.42659 iter/s, 4.94521s/12 iters), loss = 2.73856
I0409 21:00:52.322346 15472 solver.cpp:237]     Train net output #0: loss = 2.73856 (* 1 = 2.73856 loss)
I0409 21:00:52.322358 15472 sgd_solver.cpp:105] Iteration 7980, lr = 0.00205826
I0409 21:00:56.546766 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 21:00:57.305410 15472 solver.cpp:218] Iteration 7992 (2.40826 iter/s, 4.98286s/12 iters), loss = 2.72787
I0409 21:00:57.305462 15472 solver.cpp:237]     Train net output #0: loss = 2.72787 (* 1 = 2.72787 loss)
I0409 21:00:57.305474 15472 sgd_solver.cpp:105] Iteration 7992, lr = 0.00205337
I0409 21:01:02.314147 15472 solver.cpp:218] Iteration 8004 (2.39594 iter/s, 5.00848s/12 iters), loss = 3.07392
I0409 21:01:02.314191 15472 solver.cpp:237]     Train net output #0: loss = 3.07392 (* 1 = 3.07392 loss)
I0409 21:01:02.314201 15472 sgd_solver.cpp:105] Iteration 8004, lr = 0.0020485
I0409 21:01:07.124917 15472 solver.cpp:218] Iteration 8016 (2.49453 iter/s, 4.81053s/12 iters), loss = 2.91144
I0409 21:01:07.125025 15472 solver.cpp:237]     Train net output #0: loss = 2.91144 (* 1 = 2.91144 loss)
I0409 21:01:07.125036 15472 sgd_solver.cpp:105] Iteration 8016, lr = 0.00204363
I0409 21:01:11.942098 15472 solver.cpp:218] Iteration 8028 (2.49124 iter/s, 4.81687s/12 iters), loss = 2.7542
I0409 21:01:11.942152 15472 solver.cpp:237]     Train net output #0: loss = 2.7542 (* 1 = 2.7542 loss)
I0409 21:01:11.942170 15472 sgd_solver.cpp:105] Iteration 8028, lr = 0.00203878
I0409 21:01:16.797449 15472 solver.cpp:218] Iteration 8040 (2.47163 iter/s, 4.8551s/12 iters), loss = 2.89451
I0409 21:01:16.797498 15472 solver.cpp:237]     Train net output #0: loss = 2.89451 (* 1 = 2.89451 loss)
I0409 21:01:16.797508 15472 sgd_solver.cpp:105] Iteration 8040, lr = 0.00203394
I0409 21:01:21.687211 15472 solver.cpp:218] Iteration 8052 (2.45424 iter/s, 4.88951s/12 iters), loss = 3.0481
I0409 21:01:21.687268 15472 solver.cpp:237]     Train net output #0: loss = 3.0481 (* 1 = 3.0481 loss)
I0409 21:01:21.687283 15472 sgd_solver.cpp:105] Iteration 8052, lr = 0.00202911
I0409 21:01:23.685814 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_8058.caffemodel
I0409 21:01:24.156023 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_8058.solverstate
I0409 21:01:24.492934 15472 solver.cpp:330] Iteration 8058, Testing net (#0)
I0409 21:01:24.492959 15472 net.cpp:676] Ignoring source layer train-data
I0409 21:01:25.736634 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 21:01:28.881295 15472 solver.cpp:397]     Test net output #0: accuracy = 0.194853
I0409 21:01:28.881345 15472 solver.cpp:397]     Test net output #1: loss = 3.1146 (* 1 = 3.1146 loss)
I0409 21:01:30.659538 15472 solver.cpp:218] Iteration 8064 (1.33751 iter/s, 8.97191s/12 iters), loss = 2.91231
I0409 21:01:30.659593 15472 solver.cpp:237]     Train net output #0: loss = 2.91231 (* 1 = 2.91231 loss)
I0409 21:01:30.659605 15472 sgd_solver.cpp:105] Iteration 8064, lr = 0.00202429
I0409 21:01:35.491655 15472 solver.cpp:218] Iteration 8076 (2.48352 iter/s, 4.83186s/12 iters), loss = 2.75227
I0409 21:01:35.491709 15472 solver.cpp:237]     Train net output #0: loss = 2.75227 (* 1 = 2.75227 loss)
I0409 21:01:35.491720 15472 sgd_solver.cpp:105] Iteration 8076, lr = 0.00201949
I0409 21:01:40.361732 15472 solver.cpp:218] Iteration 8088 (2.46416 iter/s, 4.86982s/12 iters), loss = 2.77827
I0409 21:01:40.361923 15472 solver.cpp:237]     Train net output #0: loss = 2.77827 (* 1 = 2.77827 loss)
I0409 21:01:40.361936 15472 sgd_solver.cpp:105] Iteration 8088, lr = 0.00201469
I0409 21:01:41.744861 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 21:01:45.215255 15472 solver.cpp:218] Iteration 8100 (2.47263 iter/s, 4.85314s/12 iters), loss = 2.77174
I0409 21:01:45.215299 15472 solver.cpp:237]     Train net output #0: loss = 2.77174 (* 1 = 2.77174 loss)
I0409 21:01:45.215309 15472 sgd_solver.cpp:105] Iteration 8100, lr = 0.00200991
I0409 21:01:50.126111 15472 solver.cpp:218] Iteration 8112 (2.44369 iter/s, 4.91061s/12 iters), loss = 2.88329
I0409 21:01:50.126161 15472 solver.cpp:237]     Train net output #0: loss = 2.88329 (* 1 = 2.88329 loss)
I0409 21:01:50.126173 15472 sgd_solver.cpp:105] Iteration 8112, lr = 0.00200514
I0409 21:01:55.059756 15472 solver.cpp:218] Iteration 8124 (2.4324 iter/s, 4.9334s/12 iters), loss = 2.79484
I0409 21:01:55.059794 15472 solver.cpp:237]     Train net output #0: loss = 2.79484 (* 1 = 2.79484 loss)
I0409 21:01:55.059803 15472 sgd_solver.cpp:105] Iteration 8124, lr = 0.00200038
I0409 21:01:59.958019 15472 solver.cpp:218] Iteration 8136 (2.44997 iter/s, 4.89802s/12 iters), loss = 2.70329
I0409 21:01:59.958074 15472 solver.cpp:237]     Train net output #0: loss = 2.70329 (* 1 = 2.70329 loss)
I0409 21:01:59.958086 15472 sgd_solver.cpp:105] Iteration 8136, lr = 0.00199563
I0409 21:02:04.875854 15472 solver.cpp:218] Iteration 8148 (2.44022 iter/s, 4.91758s/12 iters), loss = 2.67174
I0409 21:02:04.875892 15472 solver.cpp:237]     Train net output #0: loss = 2.67174 (* 1 = 2.67174 loss)
I0409 21:02:04.875900 15472 sgd_solver.cpp:105] Iteration 8148, lr = 0.00199089
I0409 21:02:09.225273 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_8160.caffemodel
I0409 21:02:09.702483 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_8160.solverstate
I0409 21:02:10.029505 15472 solver.cpp:330] Iteration 8160, Testing net (#0)
I0409 21:02:10.029531 15472 net.cpp:676] Ignoring source layer train-data
I0409 21:02:11.207101 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 21:02:14.547796 15472 solver.cpp:397]     Test net output #0: accuracy = 0.228554
I0409 21:02:14.547834 15472 solver.cpp:397]     Test net output #1: loss = 2.87608 (* 1 = 2.87608 loss)
I0409 21:02:14.630807 15472 solver.cpp:218] Iteration 8160 (1.2302 iter/s, 9.75451s/12 iters), loss = 2.70611
I0409 21:02:14.630864 15472 solver.cpp:237]     Train net output #0: loss = 2.70611 (* 1 = 2.70611 loss)
I0409 21:02:14.630877 15472 sgd_solver.cpp:105] Iteration 8160, lr = 0.00198616
I0409 21:02:18.773360 15472 solver.cpp:218] Iteration 8172 (2.89693 iter/s, 4.14231s/12 iters), loss = 2.76109
I0409 21:02:18.773420 15472 solver.cpp:237]     Train net output #0: loss = 2.76109 (* 1 = 2.76109 loss)
I0409 21:02:18.773432 15472 sgd_solver.cpp:105] Iteration 8172, lr = 0.00198145
I0409 21:02:23.657512 15472 solver.cpp:218] Iteration 8184 (2.45706 iter/s, 4.88389s/12 iters), loss = 2.71405
I0409 21:02:23.657557 15472 solver.cpp:237]     Train net output #0: loss = 2.71405 (* 1 = 2.71405 loss)
I0409 21:02:23.657567 15472 sgd_solver.cpp:105] Iteration 8184, lr = 0.00197674
I0409 21:02:27.067005 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 21:02:28.464888 15472 solver.cpp:218] Iteration 8196 (2.4963 iter/s, 4.80712s/12 iters), loss = 2.8047
I0409 21:02:28.464942 15472 solver.cpp:237]     Train net output #0: loss = 2.8047 (* 1 = 2.8047 loss)
I0409 21:02:28.464954 15472 sgd_solver.cpp:105] Iteration 8196, lr = 0.00197205
I0409 21:02:33.349201 15472 solver.cpp:218] Iteration 8208 (2.45697 iter/s, 4.88406s/12 iters), loss = 2.81151
I0409 21:02:33.349244 15472 solver.cpp:237]     Train net output #0: loss = 2.81151 (* 1 = 2.81151 loss)
I0409 21:02:33.349254 15472 sgd_solver.cpp:105] Iteration 8208, lr = 0.00196737
I0409 21:02:38.200851 15472 solver.cpp:218] Iteration 8220 (2.47351 iter/s, 4.8514s/12 iters), loss = 2.74225
I0409 21:02:38.200902 15472 solver.cpp:237]     Train net output #0: loss = 2.74225 (* 1 = 2.74225 loss)
I0409 21:02:38.200915 15472 sgd_solver.cpp:105] Iteration 8220, lr = 0.0019627
I0409 21:02:42.993150 15472 solver.cpp:218] Iteration 8232 (2.50415 iter/s, 4.79204s/12 iters), loss = 2.6327
I0409 21:02:42.993239 15472 solver.cpp:237]     Train net output #0: loss = 2.6327 (* 1 = 2.6327 loss)
I0409 21:02:42.993253 15472 sgd_solver.cpp:105] Iteration 8232, lr = 0.00195804
I0409 21:02:48.118342 15472 solver.cpp:218] Iteration 8244 (2.34151 iter/s, 5.1249s/12 iters), loss = 2.66865
I0409 21:02:48.118391 15472 solver.cpp:237]     Train net output #0: loss = 2.66865 (* 1 = 2.66865 loss)
I0409 21:02:48.118403 15472 sgd_solver.cpp:105] Iteration 8244, lr = 0.00195339
I0409 21:02:52.917342 15472 solver.cpp:218] Iteration 8256 (2.50065 iter/s, 4.79875s/12 iters), loss = 2.85674
I0409 21:02:52.917385 15472 solver.cpp:237]     Train net output #0: loss = 2.85674 (* 1 = 2.85674 loss)
I0409 21:02:52.917394 15472 sgd_solver.cpp:105] Iteration 8256, lr = 0.00194875
I0409 21:02:54.903015 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_8262.caffemodel
I0409 21:02:55.990818 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_8262.solverstate
I0409 21:02:56.309355 15472 solver.cpp:330] Iteration 8262, Testing net (#0)
I0409 21:02:56.309374 15472 net.cpp:676] Ignoring source layer train-data
I0409 21:02:57.606537 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 21:03:00.916851 15472 solver.cpp:397]     Test net output #0: accuracy = 0.207721
I0409 21:03:00.916901 15472 solver.cpp:397]     Test net output #1: loss = 2.96776 (* 1 = 2.96776 loss)
I0409 21:03:03.017931 15472 solver.cpp:218] Iteration 8268 (1.1881 iter/s, 10.1001s/12 iters), loss = 2.77874
I0409 21:03:03.018007 15472 solver.cpp:237]     Train net output #0: loss = 2.77874 (* 1 = 2.77874 loss)
I0409 21:03:03.018021 15472 sgd_solver.cpp:105] Iteration 8268, lr = 0.00194412
I0409 21:03:07.938052 15472 solver.cpp:218] Iteration 8280 (2.4391 iter/s, 4.91984s/12 iters), loss = 2.47795
I0409 21:03:07.938108 15472 solver.cpp:237]     Train net output #0: loss = 2.47795 (* 1 = 2.47795 loss)
I0409 21:03:07.938122 15472 sgd_solver.cpp:105] Iteration 8280, lr = 0.00193951
I0409 21:03:12.731379 15472 solver.cpp:218] Iteration 8292 (2.50362 iter/s, 4.79307s/12 iters), loss = 2.51999
I0409 21:03:12.731438 15472 solver.cpp:237]     Train net output #0: loss = 2.51999 (* 1 = 2.51999 loss)
I0409 21:03:12.731451 15472 sgd_solver.cpp:105] Iteration 8292, lr = 0.0019349
I0409 21:03:13.391034 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 21:03:17.586972 15472 solver.cpp:218] Iteration 8304 (2.47151 iter/s, 4.85532s/12 iters), loss = 2.76346
I0409 21:03:17.587026 15472 solver.cpp:237]     Train net output #0: loss = 2.76346 (* 1 = 2.76346 loss)
I0409 21:03:17.587038 15472 sgd_solver.cpp:105] Iteration 8304, lr = 0.00193031
I0409 21:03:19.242235 15472 blocking_queue.cpp:49] Waiting for data
I0409 21:03:22.590580 15472 solver.cpp:218] Iteration 8316 (2.3984 iter/s, 5.00334s/12 iters), loss = 2.74201
I0409 21:03:22.590636 15472 solver.cpp:237]     Train net output #0: loss = 2.74201 (* 1 = 2.74201 loss)
I0409 21:03:22.590648 15472 sgd_solver.cpp:105] Iteration 8316, lr = 0.00192573
I0409 21:03:27.392894 15472 solver.cpp:218] Iteration 8328 (2.49893 iter/s, 4.80206s/12 iters), loss = 2.55693
I0409 21:03:27.392952 15472 solver.cpp:237]     Train net output #0: loss = 2.55693 (* 1 = 2.55693 loss)
I0409 21:03:27.392964 15472 sgd_solver.cpp:105] Iteration 8328, lr = 0.00192115
I0409 21:03:32.159929 15472 solver.cpp:218] Iteration 8340 (2.51742 iter/s, 4.76678s/12 iters), loss = 2.3967
I0409 21:03:32.159981 15472 solver.cpp:237]     Train net output #0: loss = 2.3967 (* 1 = 2.3967 loss)
I0409 21:03:32.159991 15472 sgd_solver.cpp:105] Iteration 8340, lr = 0.00191659
I0409 21:03:36.942037 15472 solver.cpp:218] Iteration 8352 (2.50949 iter/s, 4.78185s/12 iters), loss = 2.71412
I0409 21:03:36.942098 15472 solver.cpp:237]     Train net output #0: loss = 2.71412 (* 1 = 2.71412 loss)
I0409 21:03:36.942111 15472 sgd_solver.cpp:105] Iteration 8352, lr = 0.00191204
I0409 21:03:41.268153 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_8364.caffemodel
I0409 21:03:41.724296 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_8364.solverstate
I0409 21:03:42.035563 15472 solver.cpp:330] Iteration 8364, Testing net (#0)
I0409 21:03:42.035581 15472 net.cpp:676] Ignoring source layer train-data
I0409 21:03:43.123836 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 21:03:46.395184 15472 solver.cpp:397]     Test net output #0: accuracy = 0.231618
I0409 21:03:46.395305 15472 solver.cpp:397]     Test net output #1: loss = 2.9314 (* 1 = 2.9314 loss)
I0409 21:03:46.478354 15472 solver.cpp:218] Iteration 8364 (1.25841 iter/s, 9.53587s/12 iters), loss = 2.69253
I0409 21:03:46.478407 15472 solver.cpp:237]     Train net output #0: loss = 2.69253 (* 1 = 2.69253 loss)
I0409 21:03:46.478420 15472 sgd_solver.cpp:105] Iteration 8364, lr = 0.0019075
I0409 21:03:50.652081 15472 solver.cpp:218] Iteration 8376 (2.87529 iter/s, 4.17349s/12 iters), loss = 2.42358
I0409 21:03:50.652138 15472 solver.cpp:237]     Train net output #0: loss = 2.42358 (* 1 = 2.42358 loss)
I0409 21:03:50.652151 15472 sgd_solver.cpp:105] Iteration 8376, lr = 0.00190297
I0409 21:03:55.490052 15472 solver.cpp:218] Iteration 8388 (2.48051 iter/s, 4.83771s/12 iters), loss = 2.67863
I0409 21:03:55.490093 15472 solver.cpp:237]     Train net output #0: loss = 2.67863 (* 1 = 2.67863 loss)
I0409 21:03:55.490103 15472 sgd_solver.cpp:105] Iteration 8388, lr = 0.00189846
I0409 21:03:58.357439 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 21:04:00.501737 15472 solver.cpp:218] Iteration 8400 (2.39453 iter/s, 5.01143s/12 iters), loss = 2.66674
I0409 21:04:00.501780 15472 solver.cpp:237]     Train net output #0: loss = 2.66674 (* 1 = 2.66674 loss)
I0409 21:04:00.501789 15472 sgd_solver.cpp:105] Iteration 8400, lr = 0.00189395
I0409 21:04:05.504371 15472 solver.cpp:218] Iteration 8412 (2.39886 iter/s, 5.00238s/12 iters), loss = 2.54037
I0409 21:04:05.504422 15472 solver.cpp:237]     Train net output #0: loss = 2.54037 (* 1 = 2.54037 loss)
I0409 21:04:05.504434 15472 sgd_solver.cpp:105] Iteration 8412, lr = 0.00188945
I0409 21:04:10.376538 15472 solver.cpp:218] Iteration 8424 (2.4631 iter/s, 4.87191s/12 iters), loss = 2.6925
I0409 21:04:10.376595 15472 solver.cpp:237]     Train net output #0: loss = 2.6925 (* 1 = 2.6925 loss)
I0409 21:04:10.376606 15472 sgd_solver.cpp:105] Iteration 8424, lr = 0.00188497
I0409 21:04:15.284716 15472 solver.cpp:218] Iteration 8436 (2.44503 iter/s, 4.90792s/12 iters), loss = 2.67114
I0409 21:04:15.284772 15472 solver.cpp:237]     Train net output #0: loss = 2.67114 (* 1 = 2.67114 loss)
I0409 21:04:15.284787 15472 sgd_solver.cpp:105] Iteration 8436, lr = 0.00188049
I0409 21:04:20.211556 15472 solver.cpp:218] Iteration 8448 (2.43577 iter/s, 4.92657s/12 iters), loss = 2.57032
I0409 21:04:20.211730 15472 solver.cpp:237]     Train net output #0: loss = 2.57032 (* 1 = 2.57032 loss)
I0409 21:04:20.211745 15472 sgd_solver.cpp:105] Iteration 8448, lr = 0.00187603
I0409 21:04:24.991299 15472 solver.cpp:218] Iteration 8460 (2.51079 iter/s, 4.77937s/12 iters), loss = 2.63123
I0409 21:04:24.991355 15472 solver.cpp:237]     Train net output #0: loss = 2.63123 (* 1 = 2.63123 loss)
I0409 21:04:24.991369 15472 sgd_solver.cpp:105] Iteration 8460, lr = 0.00187157
I0409 21:04:26.963688 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_8466.caffemodel
I0409 21:04:27.420083 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_8466.solverstate
I0409 21:04:28.046777 15472 solver.cpp:330] Iteration 8466, Testing net (#0)
I0409 21:04:28.046803 15472 net.cpp:676] Ignoring source layer train-data
I0409 21:04:29.164849 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 21:04:32.463353 15472 solver.cpp:397]     Test net output #0: accuracy = 0.234681
I0409 21:04:32.463397 15472 solver.cpp:397]     Test net output #1: loss = 2.86752 (* 1 = 2.86752 loss)
I0409 21:04:34.339149 15472 solver.cpp:218] Iteration 8472 (1.28378 iter/s, 9.34741s/12 iters), loss = 2.60153
I0409 21:04:34.339196 15472 solver.cpp:237]     Train net output #0: loss = 2.60153 (* 1 = 2.60153 loss)
I0409 21:04:34.339206 15472 sgd_solver.cpp:105] Iteration 8472, lr = 0.00186713
I0409 21:04:39.228615 15472 solver.cpp:218] Iteration 8484 (2.45439 iter/s, 4.88921s/12 iters), loss = 2.60119
I0409 21:04:39.228668 15472 solver.cpp:237]     Train net output #0: loss = 2.60119 (* 1 = 2.60119 loss)
I0409 21:04:39.228679 15472 sgd_solver.cpp:105] Iteration 8484, lr = 0.0018627
I0409 21:04:44.045369 15472 solver.cpp:218] Iteration 8496 (2.49144 iter/s, 4.8165s/12 iters), loss = 2.68491
I0409 21:04:44.045411 15472 solver.cpp:237]     Train net output #0: loss = 2.68491 (* 1 = 2.68491 loss)
I0409 21:04:44.045419 15472 sgd_solver.cpp:105] Iteration 8496, lr = 0.00185827
I0409 21:04:44.104946 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 21:04:49.104363 15472 solver.cpp:218] Iteration 8508 (2.37214 iter/s, 5.05873s/12 iters), loss = 2.58197
I0409 21:04:49.104418 15472 solver.cpp:237]     Train net output #0: loss = 2.58197 (* 1 = 2.58197 loss)
I0409 21:04:49.104432 15472 sgd_solver.cpp:105] Iteration 8508, lr = 0.00185386
I0409 21:04:53.883908 15472 solver.cpp:218] Iteration 8520 (2.51083 iter/s, 4.77929s/12 iters), loss = 2.48107
I0409 21:04:53.885898 15472 solver.cpp:237]     Train net output #0: loss = 2.48107 (* 1 = 2.48107 loss)
I0409 21:04:53.885911 15472 sgd_solver.cpp:105] Iteration 8520, lr = 0.00184946
I0409 21:04:58.753113 15472 solver.cpp:218] Iteration 8532 (2.46558 iter/s, 4.86701s/12 iters), loss = 2.33982
I0409 21:04:58.753167 15472 solver.cpp:237]     Train net output #0: loss = 2.33982 (* 1 = 2.33982 loss)
I0409 21:04:58.753180 15472 sgd_solver.cpp:105] Iteration 8532, lr = 0.00184507
I0409 21:05:03.602725 15472 solver.cpp:218] Iteration 8544 (2.47456 iter/s, 4.84935s/12 iters), loss = 2.52207
I0409 21:05:03.602767 15472 solver.cpp:237]     Train net output #0: loss = 2.52207 (* 1 = 2.52207 loss)
I0409 21:05:03.602777 15472 sgd_solver.cpp:105] Iteration 8544, lr = 0.00184069
I0409 21:05:08.377946 15472 solver.cpp:218] Iteration 8556 (2.5131 iter/s, 4.77497s/12 iters), loss = 2.50777
I0409 21:05:08.378012 15472 solver.cpp:237]     Train net output #0: loss = 2.50777 (* 1 = 2.50777 loss)
I0409 21:05:08.378026 15472 sgd_solver.cpp:105] Iteration 8556, lr = 0.00183632
I0409 21:05:12.718729 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_8568.caffemodel
I0409 21:05:13.190802 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_8568.solverstate
I0409 21:05:13.515281 15472 solver.cpp:330] Iteration 8568, Testing net (#0)
I0409 21:05:13.515302 15472 net.cpp:676] Ignoring source layer train-data
I0409 21:05:14.613101 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 21:05:18.024116 15472 solver.cpp:397]     Test net output #0: accuracy = 0.261029
I0409 21:05:18.024166 15472 solver.cpp:397]     Test net output #1: loss = 2.75585 (* 1 = 2.75585 loss)
I0409 21:05:18.107161 15472 solver.cpp:218] Iteration 8568 (1.23346 iter/s, 9.72875s/12 iters), loss = 2.5891
I0409 21:05:18.107211 15472 solver.cpp:237]     Train net output #0: loss = 2.5891 (* 1 = 2.5891 loss)
I0409 21:05:18.107223 15472 sgd_solver.cpp:105] Iteration 8568, lr = 0.00183196
I0409 21:05:22.179692 15472 solver.cpp:218] Iteration 8580 (2.94674 iter/s, 4.07229s/12 iters), loss = 2.35758
I0409 21:05:22.179765 15472 solver.cpp:237]     Train net output #0: loss = 2.35758 (* 1 = 2.35758 loss)
I0409 21:05:22.179781 15472 sgd_solver.cpp:105] Iteration 8580, lr = 0.00182761
I0409 21:05:27.025027 15472 solver.cpp:218] Iteration 8592 (2.47675 iter/s, 4.84506s/12 iters), loss = 2.54459
I0409 21:05:27.025158 15472 solver.cpp:237]     Train net output #0: loss = 2.54459 (* 1 = 2.54459 loss)
I0409 21:05:27.025169 15472 sgd_solver.cpp:105] Iteration 8592, lr = 0.00182327
I0409 21:05:29.140565 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 21:05:31.865545 15472 solver.cpp:218] Iteration 8604 (2.47924 iter/s, 4.84019s/12 iters), loss = 2.46896
I0409 21:05:31.865581 15472 solver.cpp:237]     Train net output #0: loss = 2.46896 (* 1 = 2.46896 loss)
I0409 21:05:31.865590 15472 sgd_solver.cpp:105] Iteration 8604, lr = 0.00181894
I0409 21:05:36.815212 15472 solver.cpp:218] Iteration 8616 (2.42453 iter/s, 4.94941s/12 iters), loss = 2.49977
I0409 21:05:36.815263 15472 solver.cpp:237]     Train net output #0: loss = 2.49977 (* 1 = 2.49977 loss)
I0409 21:05:36.815279 15472 sgd_solver.cpp:105] Iteration 8616, lr = 0.00181462
I0409 21:05:41.817612 15472 solver.cpp:218] Iteration 8628 (2.39897 iter/s, 5.00214s/12 iters), loss = 2.51418
I0409 21:05:41.817660 15472 solver.cpp:237]     Train net output #0: loss = 2.51418 (* 1 = 2.51418 loss)
I0409 21:05:41.817672 15472 sgd_solver.cpp:105] Iteration 8628, lr = 0.00181031
I0409 21:05:46.650703 15472 solver.cpp:218] Iteration 8640 (2.48301 iter/s, 4.83284s/12 iters), loss = 2.53397
I0409 21:05:46.650750 15472 solver.cpp:237]     Train net output #0: loss = 2.53397 (* 1 = 2.53397 loss)
I0409 21:05:46.650763 15472 sgd_solver.cpp:105] Iteration 8640, lr = 0.00180602
I0409 21:05:51.433101 15472 solver.cpp:218] Iteration 8652 (2.50935 iter/s, 4.78212s/12 iters), loss = 2.45289
I0409 21:05:51.433173 15472 solver.cpp:237]     Train net output #0: loss = 2.45289 (* 1 = 2.45289 loss)
I0409 21:05:51.433183 15472 sgd_solver.cpp:105] Iteration 8652, lr = 0.00180173
I0409 21:05:56.289305 15472 solver.cpp:218] Iteration 8664 (2.4712 iter/s, 4.85593s/12 iters), loss = 2.66746
I0409 21:05:56.289359 15472 solver.cpp:237]     Train net output #0: loss = 2.66746 (* 1 = 2.66746 loss)
I0409 21:05:56.289373 15472 sgd_solver.cpp:105] Iteration 8664, lr = 0.00179745
I0409 21:05:58.332111 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_8670.caffemodel
I0409 21:05:58.781497 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_8670.solverstate
I0409 21:05:59.098973 15472 solver.cpp:330] Iteration 8670, Testing net (#0)
I0409 21:05:59.099004 15472 net.cpp:676] Ignoring source layer train-data
I0409 21:06:00.166003 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 21:06:03.673295 15472 solver.cpp:397]     Test net output #0: accuracy = 0.245098
I0409 21:06:03.673342 15472 solver.cpp:397]     Test net output #1: loss = 2.81209 (* 1 = 2.81209 loss)
I0409 21:06:05.477828 15472 solver.cpp:218] Iteration 8676 (1.30604 iter/s, 9.18809s/12 iters), loss = 2.73254
I0409 21:06:05.477874 15472 solver.cpp:237]     Train net output #0: loss = 2.73254 (* 1 = 2.73254 loss)
I0409 21:06:05.477885 15472 sgd_solver.cpp:105] Iteration 8676, lr = 0.00179318
I0409 21:06:10.281116 15472 solver.cpp:218] Iteration 8688 (2.49842 iter/s, 4.80303s/12 iters), loss = 2.40098
I0409 21:06:10.281174 15472 solver.cpp:237]     Train net output #0: loss = 2.40098 (* 1 = 2.40098 loss)
I0409 21:06:10.281186 15472 sgd_solver.cpp:105] Iteration 8688, lr = 0.00178893
I0409 21:06:14.441779 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 21:06:15.117234 15472 solver.cpp:218] Iteration 8700 (2.48146 iter/s, 4.83585s/12 iters), loss = 2.54257
I0409 21:06:15.117275 15472 solver.cpp:237]     Train net output #0: loss = 2.54257 (* 1 = 2.54257 loss)
I0409 21:06:15.117283 15472 sgd_solver.cpp:105] Iteration 8700, lr = 0.00178468
I0409 21:06:19.960744 15472 solver.cpp:218] Iteration 8712 (2.47767 iter/s, 4.84326s/12 iters), loss = 2.88448
I0409 21:06:19.960793 15472 solver.cpp:237]     Train net output #0: loss = 2.88448 (* 1 = 2.88448 loss)
I0409 21:06:19.960805 15472 sgd_solver.cpp:105] Iteration 8712, lr = 0.00178044
I0409 21:06:24.728993 15472 solver.cpp:218] Iteration 8724 (2.51678 iter/s, 4.76799s/12 iters), loss = 2.49166
I0409 21:06:24.729041 15472 solver.cpp:237]     Train net output #0: loss = 2.49166 (* 1 = 2.49166 loss)
I0409 21:06:24.729051 15472 sgd_solver.cpp:105] Iteration 8724, lr = 0.00177621
I0409 21:06:29.521692 15472 solver.cpp:218] Iteration 8736 (2.50394 iter/s, 4.79245s/12 iters), loss = 2.3309
I0409 21:06:29.521755 15472 solver.cpp:237]     Train net output #0: loss = 2.3309 (* 1 = 2.3309 loss)
I0409 21:06:29.521764 15472 sgd_solver.cpp:105] Iteration 8736, lr = 0.001772
I0409 21:06:34.488539 15472 solver.cpp:218] Iteration 8748 (2.41615 iter/s, 4.96657s/12 iters), loss = 2.69344
I0409 21:06:34.488584 15472 solver.cpp:237]     Train net output #0: loss = 2.69344 (* 1 = 2.69344 loss)
I0409 21:06:34.488592 15472 sgd_solver.cpp:105] Iteration 8748, lr = 0.00176779
I0409 21:06:39.304572 15472 solver.cpp:218] Iteration 8760 (2.49181 iter/s, 4.81578s/12 iters), loss = 2.62728
I0409 21:06:39.304627 15472 solver.cpp:237]     Train net output #0: loss = 2.62728 (* 1 = 2.62728 loss)
I0409 21:06:39.304639 15472 sgd_solver.cpp:105] Iteration 8760, lr = 0.00176359
I0409 21:06:43.688068 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_8772.caffemodel
I0409 21:06:44.166857 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_8772.solverstate
I0409 21:06:44.493072 15472 solver.cpp:330] Iteration 8772, Testing net (#0)
I0409 21:06:44.493103 15472 net.cpp:676] Ignoring source layer train-data
I0409 21:06:45.543459 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 21:06:48.979851 15472 solver.cpp:397]     Test net output #0: accuracy = 0.241422
I0409 21:06:48.979897 15472 solver.cpp:397]     Test net output #1: loss = 2.82102 (* 1 = 2.82102 loss)
I0409 21:06:49.062901 15472 solver.cpp:218] Iteration 8772 (1.22978 iter/s, 9.75787s/12 iters), loss = 2.41777
I0409 21:06:49.062947 15472 solver.cpp:237]     Train net output #0: loss = 2.41777 (* 1 = 2.41777 loss)
I0409 21:06:49.062958 15472 sgd_solver.cpp:105] Iteration 8772, lr = 0.00175941
I0409 21:06:53.165717 15472 solver.cpp:218] Iteration 8784 (2.92498 iter/s, 4.10259s/12 iters), loss = 2.49145
I0409 21:06:53.165776 15472 solver.cpp:237]     Train net output #0: loss = 2.49145 (* 1 = 2.49145 loss)
I0409 21:06:53.165788 15472 sgd_solver.cpp:105] Iteration 8784, lr = 0.00175523
I0409 21:06:57.952911 15472 solver.cpp:218] Iteration 8796 (2.50683 iter/s, 4.78693s/12 iters), loss = 2.35989
I0409 21:06:57.952968 15472 solver.cpp:237]     Train net output #0: loss = 2.35989 (* 1 = 2.35989 loss)
I0409 21:06:57.952981 15472 sgd_solver.cpp:105] Iteration 8796, lr = 0.00175106
I0409 21:06:59.330492 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 21:07:02.725665 15472 solver.cpp:218] Iteration 8808 (2.51441 iter/s, 4.77249s/12 iters), loss = 2.41431
I0409 21:07:02.725790 15472 solver.cpp:237]     Train net output #0: loss = 2.41431 (* 1 = 2.41431 loss)
I0409 21:07:02.725802 15472 sgd_solver.cpp:105] Iteration 8808, lr = 0.0017469
I0409 21:07:07.577522 15472 solver.cpp:218] Iteration 8820 (2.47345 iter/s, 4.85153s/12 iters), loss = 2.53206
I0409 21:07:07.577564 15472 solver.cpp:237]     Train net output #0: loss = 2.53206 (* 1 = 2.53206 loss)
I0409 21:07:07.577571 15472 sgd_solver.cpp:105] Iteration 8820, lr = 0.00174276
I0409 21:07:12.472134 15472 solver.cpp:218] Iteration 8832 (2.4518 iter/s, 4.89436s/12 iters), loss = 2.33651
I0409 21:07:12.472170 15472 solver.cpp:237]     Train net output #0: loss = 2.33651 (* 1 = 2.33651 loss)
I0409 21:07:12.472179 15472 sgd_solver.cpp:105] Iteration 8832, lr = 0.00173862
I0409 21:07:17.287283 15472 solver.cpp:218] Iteration 8844 (2.49226 iter/s, 4.8149s/12 iters), loss = 2.37071
I0409 21:07:17.287328 15472 solver.cpp:237]     Train net output #0: loss = 2.37071 (* 1 = 2.37071 loss)
I0409 21:07:17.287338 15472 sgd_solver.cpp:105] Iteration 8844, lr = 0.00173449
I0409 21:07:22.138051 15472 solver.cpp:218] Iteration 8856 (2.47397 iter/s, 4.85051s/12 iters), loss = 2.45187
I0409 21:07:22.138105 15472 solver.cpp:237]     Train net output #0: loss = 2.45187 (* 1 = 2.45187 loss)
I0409 21:07:22.138116 15472 sgd_solver.cpp:105] Iteration 8856, lr = 0.00173037
I0409 21:07:27.534278 15472 solver.cpp:218] Iteration 8868 (2.22389 iter/s, 5.39595s/12 iters), loss = 2.7069
I0409 21:07:27.534322 15472 solver.cpp:237]     Train net output #0: loss = 2.7069 (* 1 = 2.7069 loss)
I0409 21:07:27.534332 15472 sgd_solver.cpp:105] Iteration 8868, lr = 0.00172626
I0409 21:07:29.596558 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_8874.caffemodel
I0409 21:07:30.047827 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_8874.solverstate
I0409 21:07:30.374650 15472 solver.cpp:330] Iteration 8874, Testing net (#0)
I0409 21:07:30.374680 15472 net.cpp:676] Ignoring source layer train-data
I0409 21:07:31.317833 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 21:07:34.781327 15472 solver.cpp:397]     Test net output #0: accuracy = 0.242034
I0409 21:07:34.781415 15472 solver.cpp:397]     Test net output #1: loss = 2.87814 (* 1 = 2.87814 loss)
I0409 21:07:36.651533 15472 solver.cpp:218] Iteration 8880 (1.31625 iter/s, 9.11683s/12 iters), loss = 2.41445
I0409 21:07:36.651579 15472 solver.cpp:237]     Train net output #0: loss = 2.41445 (* 1 = 2.41445 loss)
I0409 21:07:36.651590 15472 sgd_solver.cpp:105] Iteration 8880, lr = 0.00172217
I0409 21:07:41.495419 15472 solver.cpp:218] Iteration 8892 (2.47748 iter/s, 4.84363s/12 iters), loss = 2.27691
I0409 21:07:41.495471 15472 solver.cpp:237]     Train net output #0: loss = 2.27691 (* 1 = 2.27691 loss)
I0409 21:07:41.495486 15472 sgd_solver.cpp:105] Iteration 8892, lr = 0.00171808
I0409 21:07:45.024464 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 21:07:46.412608 15472 solver.cpp:218] Iteration 8904 (2.44055 iter/s, 4.91693s/12 iters), loss = 2.55477
I0409 21:07:46.412650 15472 solver.cpp:237]     Train net output #0: loss = 2.55477 (* 1 = 2.55477 loss)
I0409 21:07:46.412660 15472 sgd_solver.cpp:105] Iteration 8904, lr = 0.001714
I0409 21:07:51.337637 15472 solver.cpp:218] Iteration 8916 (2.43666 iter/s, 4.92478s/12 iters), loss = 2.29658
I0409 21:07:51.337682 15472 solver.cpp:237]     Train net output #0: loss = 2.29658 (* 1 = 2.29658 loss)
I0409 21:07:51.337692 15472 sgd_solver.cpp:105] Iteration 8916, lr = 0.00170993
I0409 21:07:56.199012 15472 solver.cpp:218] Iteration 8928 (2.46857 iter/s, 4.86112s/12 iters), loss = 2.45591
I0409 21:07:56.199055 15472 solver.cpp:237]     Train net output #0: loss = 2.45591 (* 1 = 2.45591 loss)
I0409 21:07:56.199065 15472 sgd_solver.cpp:105] Iteration 8928, lr = 0.00170587
I0409 21:08:01.062351 15472 solver.cpp:218] Iteration 8940 (2.46757 iter/s, 4.86309s/12 iters), loss = 2.51799
I0409 21:08:01.062418 15472 solver.cpp:237]     Train net output #0: loss = 2.51799 (* 1 = 2.51799 loss)
I0409 21:08:01.062428 15472 sgd_solver.cpp:105] Iteration 8940, lr = 0.00170182
I0409 21:08:05.931089 15472 solver.cpp:218] Iteration 8952 (2.46484 iter/s, 4.86846s/12 iters), loss = 2.4032
I0409 21:08:05.931279 15472 solver.cpp:237]     Train net output #0: loss = 2.4032 (* 1 = 2.4032 loss)
I0409 21:08:05.931290 15472 sgd_solver.cpp:105] Iteration 8952, lr = 0.00169778
I0409 21:08:10.785842 15472 solver.cpp:218] Iteration 8964 (2.47201 iter/s, 4.85435s/12 iters), loss = 2.18846
I0409 21:08:10.785900 15472 solver.cpp:237]     Train net output #0: loss = 2.18846 (* 1 = 2.18846 loss)
I0409 21:08:10.785913 15472 sgd_solver.cpp:105] Iteration 8964, lr = 0.00169375
I0409 21:08:15.172989 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_8976.caffemodel
I0409 21:08:15.597308 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_8976.solverstate
I0409 21:08:15.908380 15472 solver.cpp:330] Iteration 8976, Testing net (#0)
I0409 21:08:15.908409 15472 net.cpp:676] Ignoring source layer train-data
I0409 21:08:16.848225 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 21:08:20.332023 15472 solver.cpp:397]     Test net output #0: accuracy = 0.259191
I0409 21:08:20.332073 15472 solver.cpp:397]     Test net output #1: loss = 2.80523 (* 1 = 2.80523 loss)
I0409 21:08:20.415374 15472 solver.cpp:218] Iteration 8976 (1.24623 iter/s, 9.62907s/12 iters), loss = 2.5105
I0409 21:08:20.415431 15472 solver.cpp:237]     Train net output #0: loss = 2.5105 (* 1 = 2.5105 loss)
I0409 21:08:20.415442 15472 sgd_solver.cpp:105] Iteration 8976, lr = 0.00168973
I0409 21:08:24.559332 15472 solver.cpp:218] Iteration 8988 (2.89595 iter/s, 4.14372s/12 iters), loss = 2.10001
I0409 21:08:24.559377 15472 solver.cpp:237]     Train net output #0: loss = 2.10001 (* 1 = 2.10001 loss)
I0409 21:08:24.559386 15472 sgd_solver.cpp:105] Iteration 8988, lr = 0.00168571
I0409 21:08:26.524430 15472 blocking_queue.cpp:49] Waiting for data
I0409 21:08:29.384493 15472 solver.cpp:218] Iteration 9000 (2.48709 iter/s, 4.82491s/12 iters), loss = 2.27752
I0409 21:08:29.384536 15472 solver.cpp:237]     Train net output #0: loss = 2.27752 (* 1 = 2.27752 loss)
I0409 21:08:29.384544 15472 sgd_solver.cpp:105] Iteration 9000, lr = 0.00168171
I0409 21:08:30.217221 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 21:08:34.320320 15472 solver.cpp:218] Iteration 9012 (2.43133 iter/s, 4.93556s/12 iters), loss = 2.4177
I0409 21:08:34.320372 15472 solver.cpp:237]     Train net output #0: loss = 2.4177 (* 1 = 2.4177 loss)
I0409 21:08:34.320385 15472 sgd_solver.cpp:105] Iteration 9012, lr = 0.00167772
I0409 21:08:39.133700 15472 solver.cpp:218] Iteration 9024 (2.49319 iter/s, 4.81312s/12 iters), loss = 2.38808
I0409 21:08:39.133806 15472 solver.cpp:237]     Train net output #0: loss = 2.38808 (* 1 = 2.38808 loss)
I0409 21:08:39.133818 15472 sgd_solver.cpp:105] Iteration 9024, lr = 0.00167374
I0409 21:08:43.977037 15472 solver.cpp:218] Iteration 9036 (2.47779 iter/s, 4.84302s/12 iters), loss = 2.10913
I0409 21:08:43.977094 15472 solver.cpp:237]     Train net output #0: loss = 2.10913 (* 1 = 2.10913 loss)
I0409 21:08:43.977108 15472 sgd_solver.cpp:105] Iteration 9036, lr = 0.00166976
I0409 21:08:48.768179 15472 solver.cpp:218] Iteration 9048 (2.50476 iter/s, 4.79088s/12 iters), loss = 2.17589
I0409 21:08:48.768231 15472 solver.cpp:237]     Train net output #0: loss = 2.17589 (* 1 = 2.17589 loss)
I0409 21:08:48.768242 15472 sgd_solver.cpp:105] Iteration 9048, lr = 0.0016658
I0409 21:08:53.577994 15472 solver.cpp:218] Iteration 9060 (2.49504 iter/s, 4.80954s/12 iters), loss = 2.49335
I0409 21:08:53.578049 15472 solver.cpp:237]     Train net output #0: loss = 2.49335 (* 1 = 2.49335 loss)
I0409 21:08:53.578061 15472 sgd_solver.cpp:105] Iteration 9060, lr = 0.00166184
I0409 21:08:58.398219 15472 solver.cpp:218] Iteration 9072 (2.48964 iter/s, 4.81997s/12 iters), loss = 2.51695
I0409 21:08:58.398253 15472 solver.cpp:237]     Train net output #0: loss = 2.51695 (* 1 = 2.51695 loss)
I0409 21:08:58.398263 15472 sgd_solver.cpp:105] Iteration 9072, lr = 0.0016579
I0409 21:09:00.363338 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_9078.caffemodel
I0409 21:09:01.721515 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_9078.solverstate
I0409 21:09:02.785199 15472 solver.cpp:330] Iteration 9078, Testing net (#0)
I0409 21:09:02.785218 15472 net.cpp:676] Ignoring source layer train-data
I0409 21:09:03.698132 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 21:09:07.306983 15472 solver.cpp:397]     Test net output #0: accuracy = 0.224877
I0409 21:09:07.307030 15472 solver.cpp:397]     Test net output #1: loss = 2.92984 (* 1 = 2.92984 loss)
I0409 21:09:09.066421 15472 solver.cpp:218] Iteration 9084 (1.12489 iter/s, 10.6677s/12 iters), loss = 2.20621
I0409 21:09:09.066483 15472 solver.cpp:237]     Train net output #0: loss = 2.20621 (* 1 = 2.20621 loss)
I0409 21:09:09.066494 15472 sgd_solver.cpp:105] Iteration 9084, lr = 0.00165396
I0409 21:09:13.853049 15472 solver.cpp:218] Iteration 9096 (2.50712 iter/s, 4.78636s/12 iters), loss = 2.18805
I0409 21:09:13.853199 15472 solver.cpp:237]     Train net output #0: loss = 2.18805 (* 1 = 2.18805 loss)
I0409 21:09:13.853211 15472 sgd_solver.cpp:105] Iteration 9096, lr = 0.00165003
I0409 21:09:16.657232 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 21:09:18.626564 15472 solver.cpp:218] Iteration 9108 (2.51406 iter/s, 4.77316s/12 iters), loss = 2.06087
I0409 21:09:18.626617 15472 solver.cpp:237]     Train net output #0: loss = 2.06087 (* 1 = 2.06087 loss)
I0409 21:09:18.626628 15472 sgd_solver.cpp:105] Iteration 9108, lr = 0.00164612
I0409 21:09:23.508204 15472 solver.cpp:218] Iteration 9120 (2.45832 iter/s, 4.88137s/12 iters), loss = 2.40846
I0409 21:09:23.508262 15472 solver.cpp:237]     Train net output #0: loss = 2.40846 (* 1 = 2.40846 loss)
I0409 21:09:23.508275 15472 sgd_solver.cpp:105] Iteration 9120, lr = 0.00164221
I0409 21:09:28.310257 15472 solver.cpp:218] Iteration 9132 (2.49907 iter/s, 4.80178s/12 iters), loss = 2.38798
I0409 21:09:28.310309 15472 solver.cpp:237]     Train net output #0: loss = 2.38798 (* 1 = 2.38798 loss)
I0409 21:09:28.310319 15472 sgd_solver.cpp:105] Iteration 9132, lr = 0.00163831
I0409 21:09:33.162459 15472 solver.cpp:218] Iteration 9144 (2.47324 iter/s, 4.85194s/12 iters), loss = 2.60746
I0409 21:09:33.162501 15472 solver.cpp:237]     Train net output #0: loss = 2.60746 (* 1 = 2.60746 loss)
I0409 21:09:33.162509 15472 sgd_solver.cpp:105] Iteration 9144, lr = 0.00163442
I0409 21:09:38.051213 15472 solver.cpp:218] Iteration 9156 (2.45474 iter/s, 4.8885s/12 iters), loss = 2.303
I0409 21:09:38.051254 15472 solver.cpp:237]     Train net output #0: loss = 2.303 (* 1 = 2.303 loss)
I0409 21:09:38.051261 15472 sgd_solver.cpp:105] Iteration 9156, lr = 0.00163054
I0409 21:09:42.967532 15472 solver.cpp:218] Iteration 9168 (2.44098 iter/s, 4.91606s/12 iters), loss = 2.1801
I0409 21:09:42.967599 15472 solver.cpp:237]     Train net output #0: loss = 2.1801 (* 1 = 2.1801 loss)
I0409 21:09:42.967613 15472 sgd_solver.cpp:105] Iteration 9168, lr = 0.00162667
I0409 21:09:47.406719 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_9180.caffemodel
I0409 21:09:47.889425 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_9180.solverstate
I0409 21:09:48.217383 15472 solver.cpp:330] Iteration 9180, Testing net (#0)
I0409 21:09:48.217401 15472 net.cpp:676] Ignoring source layer train-data
I0409 21:09:49.088059 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 21:09:52.817001 15472 solver.cpp:397]     Test net output #0: accuracy = 0.262868
I0409 21:09:52.817042 15472 solver.cpp:397]     Test net output #1: loss = 2.71006 (* 1 = 2.71006 loss)
I0409 21:09:52.899940 15472 solver.cpp:218] Iteration 9180 (1.20823 iter/s, 9.93192s/12 iters), loss = 2.51182
I0409 21:09:52.899986 15472 solver.cpp:237]     Train net output #0: loss = 2.51182 (* 1 = 2.51182 loss)
I0409 21:09:52.899996 15472 sgd_solver.cpp:105] Iteration 9180, lr = 0.00162281
I0409 21:09:56.948312 15472 solver.cpp:218] Iteration 9192 (2.96432 iter/s, 4.04814s/12 iters), loss = 2.43823
I0409 21:09:56.948369 15472 solver.cpp:237]     Train net output #0: loss = 2.43823 (* 1 = 2.43823 loss)
I0409 21:09:56.948381 15472 sgd_solver.cpp:105] Iteration 9192, lr = 0.00161895
I0409 21:10:01.725682 15472 solver.cpp:218] Iteration 9204 (2.51198 iter/s, 4.77711s/12 iters), loss = 2.3324
I0409 21:10:01.725728 15472 solver.cpp:237]     Train net output #0: loss = 2.3324 (* 1 = 2.3324 loss)
I0409 21:10:01.725739 15472 sgd_solver.cpp:105] Iteration 9204, lr = 0.00161511
I0409 21:10:01.793282 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 21:10:06.649109 15472 solver.cpp:218] Iteration 9216 (2.43746 iter/s, 4.92317s/12 iters), loss = 2.49551
I0409 21:10:06.649155 15472 solver.cpp:237]     Train net output #0: loss = 2.49551 (* 1 = 2.49551 loss)
I0409 21:10:06.649164 15472 sgd_solver.cpp:105] Iteration 9216, lr = 0.00161128
I0409 21:10:11.472723 15472 solver.cpp:218] Iteration 9228 (2.48789 iter/s, 4.82336s/12 iters), loss = 2.31643
I0409 21:10:11.472764 15472 solver.cpp:237]     Train net output #0: loss = 2.31643 (* 1 = 2.31643 loss)
I0409 21:10:11.472774 15472 sgd_solver.cpp:105] Iteration 9228, lr = 0.00160745
I0409 21:10:16.305781 15472 solver.cpp:218] Iteration 9240 (2.48303 iter/s, 4.83281s/12 iters), loss = 2.21141
I0409 21:10:16.305830 15472 solver.cpp:237]     Train net output #0: loss = 2.21141 (* 1 = 2.21141 loss)
I0409 21:10:16.305840 15472 sgd_solver.cpp:105] Iteration 9240, lr = 0.00160363
I0409 21:10:21.133802 15472 solver.cpp:218] Iteration 9252 (2.48563 iter/s, 4.82776s/12 iters), loss = 2.38715
I0409 21:10:21.133937 15472 solver.cpp:237]     Train net output #0: loss = 2.38715 (* 1 = 2.38715 loss)
I0409 21:10:21.133949 15472 sgd_solver.cpp:105] Iteration 9252, lr = 0.00159983
I0409 21:10:25.912770 15472 solver.cpp:218] Iteration 9264 (2.51119 iter/s, 4.77862s/12 iters), loss = 2.2682
I0409 21:10:25.912822 15472 solver.cpp:237]     Train net output #0: loss = 2.2682 (* 1 = 2.2682 loss)
I0409 21:10:25.912834 15472 sgd_solver.cpp:105] Iteration 9264, lr = 0.00159603
I0409 21:10:30.743319 15472 solver.cpp:218] Iteration 9276 (2.48433 iter/s, 4.83028s/12 iters), loss = 2.34008
I0409 21:10:30.743378 15472 solver.cpp:237]     Train net output #0: loss = 2.34008 (* 1 = 2.34008 loss)
I0409 21:10:30.743391 15472 sgd_solver.cpp:105] Iteration 9276, lr = 0.00159224
I0409 21:10:32.699031 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_9282.caffemodel
I0409 21:10:33.183740 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_9282.solverstate
I0409 21:10:33.513715 15472 solver.cpp:330] Iteration 9282, Testing net (#0)
I0409 21:10:33.513741 15472 net.cpp:676] Ignoring source layer train-data
I0409 21:10:34.379801 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 21:10:37.976719 15472 solver.cpp:397]     Test net output #0: accuracy = 0.261642
I0409 21:10:37.976755 15472 solver.cpp:397]     Test net output #1: loss = 2.72332 (* 1 = 2.72332 loss)
I0409 21:10:39.796841 15472 solver.cpp:218] Iteration 9288 (1.32552 iter/s, 9.05307s/12 iters), loss = 2.28903
I0409 21:10:39.796900 15472 solver.cpp:237]     Train net output #0: loss = 2.28903 (* 1 = 2.28903 loss)
I0409 21:10:39.796911 15472 sgd_solver.cpp:105] Iteration 9288, lr = 0.00158846
I0409 21:10:44.594774 15472 solver.cpp:218] Iteration 9300 (2.50121 iter/s, 4.79767s/12 iters), loss = 2.31163
I0409 21:10:44.594818 15472 solver.cpp:237]     Train net output #0: loss = 2.31163 (* 1 = 2.31163 loss)
I0409 21:10:44.594827 15472 sgd_solver.cpp:105] Iteration 9300, lr = 0.00158469
I0409 21:10:46.704085 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 21:10:49.396018 15472 solver.cpp:218] Iteration 9312 (2.49949 iter/s, 4.80099s/12 iters), loss = 2.33946
I0409 21:10:49.396077 15472 solver.cpp:237]     Train net output #0: loss = 2.33946 (* 1 = 2.33946 loss)
I0409 21:10:49.396091 15472 sgd_solver.cpp:105] Iteration 9312, lr = 0.00158092
I0409 21:10:54.201805 15472 solver.cpp:218] Iteration 9324 (2.49713 iter/s, 4.80552s/12 iters), loss = 2.34513
I0409 21:10:54.201921 15472 solver.cpp:237]     Train net output #0: loss = 2.34513 (* 1 = 2.34513 loss)
I0409 21:10:54.201931 15472 sgd_solver.cpp:105] Iteration 9324, lr = 0.00157717
I0409 21:10:58.999292 15472 solver.cpp:218] Iteration 9336 (2.50148 iter/s, 4.79717s/12 iters), loss = 2.22965
I0409 21:10:58.999339 15472 solver.cpp:237]     Train net output #0: loss = 2.22965 (* 1 = 2.22965 loss)
I0409 21:10:58.999351 15472 sgd_solver.cpp:105] Iteration 9336, lr = 0.00157343
I0409 21:11:03.837990 15472 solver.cpp:218] Iteration 9348 (2.48014 iter/s, 4.83843s/12 iters), loss = 2.04736
I0409 21:11:03.838044 15472 solver.cpp:237]     Train net output #0: loss = 2.04736 (* 1 = 2.04736 loss)
I0409 21:11:03.838057 15472 sgd_solver.cpp:105] Iteration 9348, lr = 0.00156969
I0409 21:11:08.671149 15472 solver.cpp:218] Iteration 9360 (2.48299 iter/s, 4.83289s/12 iters), loss = 2.1106
I0409 21:11:08.671211 15472 solver.cpp:237]     Train net output #0: loss = 2.1106 (* 1 = 2.1106 loss)
I0409 21:11:08.671226 15472 sgd_solver.cpp:105] Iteration 9360, lr = 0.00156596
I0409 21:11:13.433166 15472 solver.cpp:218] Iteration 9372 (2.52008 iter/s, 4.76175s/12 iters), loss = 2.3269
I0409 21:11:13.433223 15472 solver.cpp:237]     Train net output #0: loss = 2.3269 (* 1 = 2.3269 loss)
I0409 21:11:13.433233 15472 sgd_solver.cpp:105] Iteration 9372, lr = 0.00156225
I0409 21:11:17.761570 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_9384.caffemodel
I0409 21:11:18.220019 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_9384.solverstate
I0409 21:11:18.543293 15472 solver.cpp:330] Iteration 9384, Testing net (#0)
I0409 21:11:18.543320 15472 net.cpp:676] Ignoring source layer train-data
I0409 21:11:19.253213 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 21:11:22.895401 15472 solver.cpp:397]     Test net output #0: accuracy = 0.234681
I0409 21:11:22.895452 15472 solver.cpp:397]     Test net output #1: loss = 2.76652 (* 1 = 2.76652 loss)
I0409 21:11:22.978663 15472 solver.cpp:218] Iteration 9384 (1.2572 iter/s, 9.54504s/12 iters), loss = 2.32671
I0409 21:11:22.978713 15472 solver.cpp:237]     Train net output #0: loss = 2.32671 (* 1 = 2.32671 loss)
I0409 21:11:22.978724 15472 sgd_solver.cpp:105] Iteration 9384, lr = 0.00155854
I0409 21:11:27.088492 15472 solver.cpp:218] Iteration 9396 (2.91999 iter/s, 4.1096s/12 iters), loss = 2.10958
I0409 21:11:27.088595 15472 solver.cpp:237]     Train net output #0: loss = 2.10958 (* 1 = 2.10958 loss)
I0409 21:11:27.088605 15472 sgd_solver.cpp:105] Iteration 9396, lr = 0.00155484
I0409 21:11:31.324856 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 21:11:31.952080 15472 solver.cpp:218] Iteration 9408 (2.46748 iter/s, 4.86326s/12 iters), loss = 2.13004
I0409 21:11:31.952145 15472 solver.cpp:237]     Train net output #0: loss = 2.13004 (* 1 = 2.13004 loss)
I0409 21:11:31.952162 15472 sgd_solver.cpp:105] Iteration 9408, lr = 0.00155114
I0409 21:11:36.760452 15472 solver.cpp:218] Iteration 9420 (2.49579 iter/s, 4.8081s/12 iters), loss = 2.11993
I0409 21:11:36.760494 15472 solver.cpp:237]     Train net output #0: loss = 2.11993 (* 1 = 2.11993 loss)
I0409 21:11:36.760505 15472 sgd_solver.cpp:105] Iteration 9420, lr = 0.00154746
I0409 21:11:41.581512 15472 solver.cpp:218] Iteration 9432 (2.48921 iter/s, 4.82081s/12 iters), loss = 2.39566
I0409 21:11:41.581557 15472 solver.cpp:237]     Train net output #0: loss = 2.39566 (* 1 = 2.39566 loss)
I0409 21:11:41.581566 15472 sgd_solver.cpp:105] Iteration 9432, lr = 0.00154379
I0409 21:11:46.428269 15472 solver.cpp:218] Iteration 9444 (2.47601 iter/s, 4.8465s/12 iters), loss = 2.06684
I0409 21:11:46.428318 15472 solver.cpp:237]     Train net output #0: loss = 2.06684 (* 1 = 2.06684 loss)
I0409 21:11:46.428329 15472 sgd_solver.cpp:105] Iteration 9444, lr = 0.00154012
I0409 21:11:51.256006 15472 solver.cpp:218] Iteration 9456 (2.48577 iter/s, 4.82747s/12 iters), loss = 2.4459
I0409 21:11:51.256057 15472 solver.cpp:237]     Train net output #0: loss = 2.4459 (* 1 = 2.4459 loss)
I0409 21:11:51.256069 15472 sgd_solver.cpp:105] Iteration 9456, lr = 0.00153647
I0409 21:11:56.075840 15472 solver.cpp:218] Iteration 9468 (2.48985 iter/s, 4.81958s/12 iters), loss = 2.10577
I0409 21:11:56.075883 15472 solver.cpp:237]     Train net output #0: loss = 2.10577 (* 1 = 2.10577 loss)
I0409 21:11:56.075892 15472 sgd_solver.cpp:105] Iteration 9468, lr = 0.00153282
I0409 21:12:00.939445 15472 solver.cpp:218] Iteration 9480 (2.46744 iter/s, 4.86334s/12 iters), loss = 2.18032
I0409 21:12:00.939601 15472 solver.cpp:237]     Train net output #0: loss = 2.18032 (* 1 = 2.18032 loss)
I0409 21:12:00.939613 15472 sgd_solver.cpp:105] Iteration 9480, lr = 0.00152918
I0409 21:12:02.956156 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_9486.caffemodel
I0409 21:12:03.408257 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_9486.solverstate
I0409 21:12:03.719533 15472 solver.cpp:330] Iteration 9486, Testing net (#0)
I0409 21:12:03.719552 15472 net.cpp:676] Ignoring source layer train-data
I0409 21:12:04.448674 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 21:12:08.144588 15472 solver.cpp:397]     Test net output #0: accuracy = 0.245098
I0409 21:12:08.144618 15472 solver.cpp:397]     Test net output #1: loss = 2.81671 (* 1 = 2.81671 loss)
I0409 21:12:09.897976 15472 solver.cpp:218] Iteration 9492 (1.33959 iter/s, 8.95799s/12 iters), loss = 2.09042
I0409 21:12:09.898023 15472 solver.cpp:237]     Train net output #0: loss = 2.09042 (* 1 = 2.09042 loss)
I0409 21:12:09.898034 15472 sgd_solver.cpp:105] Iteration 9492, lr = 0.00152555
I0409 21:12:14.935822 15472 solver.cpp:218] Iteration 9504 (2.38209 iter/s, 5.03758s/12 iters), loss = 2.09316
I0409 21:12:14.935863 15472 solver.cpp:237]     Train net output #0: loss = 2.09316 (* 1 = 2.09316 loss)
I0409 21:12:14.935871 15472 sgd_solver.cpp:105] Iteration 9504, lr = 0.00152193
I0409 21:12:16.382448 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 21:12:19.773810 15472 solver.cpp:218] Iteration 9516 (2.4805 iter/s, 4.83774s/12 iters), loss = 1.86112
I0409 21:12:19.773856 15472 solver.cpp:237]     Train net output #0: loss = 1.86112 (* 1 = 1.86112 loss)
I0409 21:12:19.773865 15472 sgd_solver.cpp:105] Iteration 9516, lr = 0.00151831
I0409 21:12:24.594148 15472 solver.cpp:218] Iteration 9528 (2.48958 iter/s, 4.82008s/12 iters), loss = 2.17638
I0409 21:12:24.594195 15472 solver.cpp:237]     Train net output #0: loss = 2.17638 (* 1 = 2.17638 loss)
I0409 21:12:24.594205 15472 sgd_solver.cpp:105] Iteration 9528, lr = 0.00151471
I0409 21:12:29.420037 15472 solver.cpp:218] Iteration 9540 (2.48672 iter/s, 4.82563s/12 iters), loss = 2.04661
I0409 21:12:29.420094 15472 solver.cpp:237]     Train net output #0: loss = 2.04661 (* 1 = 2.04661 loss)
I0409 21:12:29.420105 15472 sgd_solver.cpp:105] Iteration 9540, lr = 0.00151111
I0409 21:12:34.353456 15472 solver.cpp:218] Iteration 9552 (2.43253 iter/s, 4.93315s/12 iters), loss = 1.94419
I0409 21:12:34.353550 15472 solver.cpp:237]     Train net output #0: loss = 1.94419 (* 1 = 1.94419 loss)
I0409 21:12:34.353560 15472 sgd_solver.cpp:105] Iteration 9552, lr = 0.00150752
I0409 21:12:39.162931 15472 solver.cpp:218] Iteration 9564 (2.49523 iter/s, 4.80917s/12 iters), loss = 2.19092
I0409 21:12:39.162989 15472 solver.cpp:237]     Train net output #0: loss = 2.19092 (* 1 = 2.19092 loss)
I0409 21:12:39.163008 15472 sgd_solver.cpp:105] Iteration 9564, lr = 0.00150395
I0409 21:12:43.994899 15472 solver.cpp:218] Iteration 9576 (2.4836 iter/s, 4.8317s/12 iters), loss = 2.25071
I0409 21:12:43.994940 15472 solver.cpp:237]     Train net output #0: loss = 2.25071 (* 1 = 2.25071 loss)
I0409 21:12:43.994948 15472 sgd_solver.cpp:105] Iteration 9576, lr = 0.00150037
I0409 21:12:48.411454 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_9588.caffemodel
I0409 21:12:48.874384 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_9588.solverstate
I0409 21:12:49.190917 15472 solver.cpp:330] Iteration 9588, Testing net (#0)
I0409 21:12:49.190935 15472 net.cpp:676] Ignoring source layer train-data
I0409 21:12:49.807303 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 21:12:53.657059 15472 solver.cpp:397]     Test net output #0: accuracy = 0.261029
I0409 21:12:53.657088 15472 solver.cpp:397]     Test net output #1: loss = 2.74824 (* 1 = 2.74824 loss)
I0409 21:12:53.733364 15472 solver.cpp:218] Iteration 9588 (1.23228 iter/s, 9.73801s/12 iters), loss = 2.21783
I0409 21:12:53.733407 15472 solver.cpp:237]     Train net output #0: loss = 2.21783 (* 1 = 2.21783 loss)
I0409 21:12:53.733414 15472 sgd_solver.cpp:105] Iteration 9588, lr = 0.00149681
I0409 21:12:58.069422 15472 solver.cpp:218] Iteration 9600 (2.76764 iter/s, 4.33582s/12 iters), loss = 2.19751
I0409 21:12:58.069478 15472 solver.cpp:237]     Train net output #0: loss = 2.19751 (* 1 = 2.19751 loss)
I0409 21:12:58.069489 15472 sgd_solver.cpp:105] Iteration 9600, lr = 0.00149326
I0409 21:13:01.541260 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 21:13:02.861800 15472 solver.cpp:218] Iteration 9612 (2.50412 iter/s, 4.79211s/12 iters), loss = 2.20074
I0409 21:13:02.861852 15472 solver.cpp:237]     Train net output #0: loss = 2.20074 (* 1 = 2.20074 loss)
I0409 21:13:02.861865 15472 sgd_solver.cpp:105] Iteration 9612, lr = 0.00148971
I0409 21:13:07.679244 15472 solver.cpp:218] Iteration 9624 (2.49108 iter/s, 4.81718s/12 iters), loss = 2.21097
I0409 21:13:07.681085 15472 solver.cpp:237]     Train net output #0: loss = 2.21097 (* 1 = 2.21097 loss)
I0409 21:13:07.681097 15472 sgd_solver.cpp:105] Iteration 9624, lr = 0.00148618
I0409 21:13:12.431077 15472 solver.cpp:218] Iteration 9636 (2.52643 iter/s, 4.74979s/12 iters), loss = 2.16439
I0409 21:13:12.431119 15472 solver.cpp:237]     Train net output #0: loss = 2.16439 (* 1 = 2.16439 loss)
I0409 21:13:12.431129 15472 sgd_solver.cpp:105] Iteration 9636, lr = 0.00148265
I0409 21:13:17.248576 15472 solver.cpp:218] Iteration 9648 (2.49105 iter/s, 4.81725s/12 iters), loss = 2.05901
I0409 21:13:17.248626 15472 solver.cpp:237]     Train net output #0: loss = 2.05901 (* 1 = 2.05901 loss)
I0409 21:13:17.248638 15472 sgd_solver.cpp:105] Iteration 9648, lr = 0.00147913
I0409 21:13:22.108585 15472 solver.cpp:218] Iteration 9660 (2.46927 iter/s, 4.85975s/12 iters), loss = 2.17899
I0409 21:13:22.108633 15472 solver.cpp:237]     Train net output #0: loss = 2.17899 (* 1 = 2.17899 loss)
I0409 21:13:22.108641 15472 sgd_solver.cpp:105] Iteration 9660, lr = 0.00147562
I0409 21:13:26.903280 15472 solver.cpp:218] Iteration 9672 (2.5029 iter/s, 4.79444s/12 iters), loss = 2.01489
I0409 21:13:26.903327 15472 solver.cpp:237]     Train net output #0: loss = 2.01489 (* 1 = 2.01489 loss)
I0409 21:13:26.903337 15472 sgd_solver.cpp:105] Iteration 9672, lr = 0.00147211
I0409 21:13:31.726900 15472 solver.cpp:218] Iteration 9684 (2.48789 iter/s, 4.82336s/12 iters), loss = 2.08043
I0409 21:13:31.726949 15472 solver.cpp:237]     Train net output #0: loss = 2.08043 (* 1 = 2.08043 loss)
I0409 21:13:31.726961 15472 sgd_solver.cpp:105] Iteration 9684, lr = 0.00146862
I0409 21:13:33.698783 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_9690.caffemodel
I0409 21:13:34.124857 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_9690.solverstate
I0409 21:13:34.434937 15472 solver.cpp:330] Iteration 9690, Testing net (#0)
I0409 21:13:34.434958 15472 net.cpp:676] Ignoring source layer train-data
I0409 21:13:35.046108 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 21:13:37.703647 15472 blocking_queue.cpp:49] Waiting for data
I0409 21:13:38.952137 15472 solver.cpp:397]     Test net output #0: accuracy = 0.273897
I0409 21:13:38.952165 15472 solver.cpp:397]     Test net output #1: loss = 2.72428 (* 1 = 2.72428 loss)
I0409 21:13:40.863687 15472 solver.cpp:218] Iteration 9696 (1.31343 iter/s, 9.13635s/12 iters), loss = 1.83758
I0409 21:13:40.863731 15472 solver.cpp:237]     Train net output #0: loss = 1.83758 (* 1 = 1.83758 loss)
I0409 21:13:40.863740 15472 sgd_solver.cpp:105] Iteration 9696, lr = 0.00146513
I0409 21:13:45.724570 15472 solver.cpp:218] Iteration 9708 (2.46882 iter/s, 4.86063s/12 iters), loss = 2.01632
I0409 21:13:45.724611 15472 solver.cpp:237]     Train net output #0: loss = 2.01632 (* 1 = 2.01632 loss)
I0409 21:13:45.724619 15472 sgd_solver.cpp:105] Iteration 9708, lr = 0.00146165
I0409 21:13:46.483325 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 21:13:50.725584 15472 solver.cpp:218] Iteration 9720 (2.39964 iter/s, 5.00075s/12 iters), loss = 2.13231
I0409 21:13:50.725638 15472 solver.cpp:237]     Train net output #0: loss = 2.13231 (* 1 = 2.13231 loss)
I0409 21:13:50.725651 15472 sgd_solver.cpp:105] Iteration 9720, lr = 0.00145818
I0409 21:13:55.516631 15472 solver.cpp:218] Iteration 9732 (2.50481 iter/s, 4.79078s/12 iters), loss = 2.50381
I0409 21:13:55.516690 15472 solver.cpp:237]     Train net output #0: loss = 2.50381 (* 1 = 2.50381 loss)
I0409 21:13:55.516702 15472 sgd_solver.cpp:105] Iteration 9732, lr = 0.00145472
I0409 21:14:00.268402 15472 solver.cpp:218] Iteration 9744 (2.52552 iter/s, 4.7515s/12 iters), loss = 2.18394
I0409 21:14:00.268448 15472 solver.cpp:237]     Train net output #0: loss = 2.18394 (* 1 = 2.18394 loss)
I0409 21:14:00.268458 15472 sgd_solver.cpp:105] Iteration 9744, lr = 0.00145127
I0409 21:14:05.060056 15472 solver.cpp:218] Iteration 9756 (2.50449 iter/s, 4.7914s/12 iters), loss = 2.06022
I0409 21:14:05.060096 15472 solver.cpp:237]     Train net output #0: loss = 2.06022 (* 1 = 2.06022 loss)
I0409 21:14:05.060106 15472 sgd_solver.cpp:105] Iteration 9756, lr = 0.00144782
I0409 21:14:09.889272 15472 solver.cpp:218] Iteration 9768 (2.48501 iter/s, 4.82896s/12 iters), loss = 2.35815
I0409 21:14:09.890249 15472 solver.cpp:237]     Train net output #0: loss = 2.35815 (* 1 = 2.35815 loss)
I0409 21:14:09.890259 15472 sgd_solver.cpp:105] Iteration 9768, lr = 0.00144438
I0409 21:14:14.690213 15472 solver.cpp:218] Iteration 9780 (2.50013 iter/s, 4.79976s/12 iters), loss = 2.20807
I0409 21:14:14.690254 15472 solver.cpp:237]     Train net output #0: loss = 2.20807 (* 1 = 2.20807 loss)
I0409 21:14:14.690263 15472 sgd_solver.cpp:105] Iteration 9780, lr = 0.00144095
I0409 21:14:19.132316 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_9792.caffemodel
I0409 21:14:19.597806 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_9792.solverstate
I0409 21:14:19.923949 15472 solver.cpp:330] Iteration 9792, Testing net (#0)
I0409 21:14:19.923976 15472 net.cpp:676] Ignoring source layer train-data
I0409 21:14:20.467113 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 21:14:24.429551 15472 solver.cpp:397]     Test net output #0: accuracy = 0.261029
I0409 21:14:24.429600 15472 solver.cpp:397]     Test net output #1: loss = 2.70966 (* 1 = 2.70966 loss)
I0409 21:14:24.512542 15472 solver.cpp:218] Iteration 9792 (1.22176 iter/s, 9.82187s/12 iters), loss = 2.04923
I0409 21:14:24.512593 15472 solver.cpp:237]     Train net output #0: loss = 2.04923 (* 1 = 2.04923 loss)
I0409 21:14:24.512604 15472 sgd_solver.cpp:105] Iteration 9792, lr = 0.00143753
I0409 21:14:28.618500 15472 solver.cpp:218] Iteration 9804 (2.92275 iter/s, 4.10573s/12 iters), loss = 2.21753
I0409 21:14:28.618553 15472 solver.cpp:237]     Train net output #0: loss = 2.21753 (* 1 = 2.21753 loss)
I0409 21:14:28.618566 15472 sgd_solver.cpp:105] Iteration 9804, lr = 0.00143412
I0409 21:14:31.662709 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 21:14:33.620971 15472 solver.cpp:218] Iteration 9816 (2.39894 iter/s, 5.0022s/12 iters), loss = 2.21247
I0409 21:14:33.621022 15472 solver.cpp:237]     Train net output #0: loss = 2.21247 (* 1 = 2.21247 loss)
I0409 21:14:33.621033 15472 sgd_solver.cpp:105] Iteration 9816, lr = 0.00143072
I0409 21:14:38.430523 15472 solver.cpp:218] Iteration 9828 (2.49517 iter/s, 4.80929s/12 iters), loss = 2.24522
I0409 21:14:38.430567 15472 solver.cpp:237]     Train net output #0: loss = 2.24522 (* 1 = 2.24522 loss)
I0409 21:14:38.430577 15472 sgd_solver.cpp:105] Iteration 9828, lr = 0.00142732
I0409 21:14:43.398365 15472 solver.cpp:218] Iteration 9840 (2.41567 iter/s, 4.96758s/12 iters), loss = 1.93749
I0409 21:14:43.398527 15472 solver.cpp:237]     Train net output #0: loss = 1.93749 (* 1 = 1.93749 loss)
I0409 21:14:43.398541 15472 sgd_solver.cpp:105] Iteration 9840, lr = 0.00142393
I0409 21:14:48.222788 15472 solver.cpp:218] Iteration 9852 (2.48753 iter/s, 4.82406s/12 iters), loss = 2.15355
I0409 21:14:48.222836 15472 solver.cpp:237]     Train net output #0: loss = 2.15355 (* 1 = 2.15355 loss)
I0409 21:14:48.222846 15472 sgd_solver.cpp:105] Iteration 9852, lr = 0.00142055
I0409 21:14:53.102546 15472 solver.cpp:218] Iteration 9864 (2.45927 iter/s, 4.8795s/12 iters), loss = 2.0043
I0409 21:14:53.102586 15472 solver.cpp:237]     Train net output #0: loss = 2.0043 (* 1 = 2.0043 loss)
I0409 21:14:53.102596 15472 sgd_solver.cpp:105] Iteration 9864, lr = 0.00141718
I0409 21:14:57.900619 15472 solver.cpp:218] Iteration 9876 (2.50114 iter/s, 4.79782s/12 iters), loss = 2.14864
I0409 21:14:57.900665 15472 solver.cpp:237]     Train net output #0: loss = 2.14864 (* 1 = 2.14864 loss)
I0409 21:14:57.900676 15472 sgd_solver.cpp:105] Iteration 9876, lr = 0.00141381
I0409 21:15:02.751994 15472 solver.cpp:218] Iteration 9888 (2.47366 iter/s, 4.85112s/12 iters), loss = 2.22272
I0409 21:15:02.752033 15472 solver.cpp:237]     Train net output #0: loss = 2.22272 (* 1 = 2.22272 loss)
I0409 21:15:02.752043 15472 sgd_solver.cpp:105] Iteration 9888, lr = 0.00141045
I0409 21:15:04.716022 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_9894.caffemodel
I0409 21:15:05.172425 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_9894.solverstate
I0409 21:15:05.486603 15472 solver.cpp:330] Iteration 9894, Testing net (#0)
I0409 21:15:05.486627 15472 net.cpp:676] Ignoring source layer train-data
I0409 21:15:06.049715 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 21:15:09.932145 15472 solver.cpp:397]     Test net output #0: accuracy = 0.314951
I0409 21:15:09.932209 15472 solver.cpp:397]     Test net output #1: loss = 2.56893 (* 1 = 2.56893 loss)
I0409 21:15:11.712576 15472 solver.cpp:218] Iteration 9900 (1.33926 iter/s, 8.96016s/12 iters), loss = 2.11091
I0409 21:15:11.712636 15472 solver.cpp:237]     Train net output #0: loss = 2.11091 (* 1 = 2.11091 loss)
I0409 21:15:11.712648 15472 sgd_solver.cpp:105] Iteration 9900, lr = 0.00140711
I0409 21:15:16.558662 15472 solver.cpp:218] Iteration 9912 (2.47637 iter/s, 4.84581s/12 iters), loss = 1.94353
I0409 21:15:16.558780 15472 solver.cpp:237]     Train net output #0: loss = 1.94353 (* 1 = 1.94353 loss)
I0409 21:15:16.558794 15472 sgd_solver.cpp:105] Iteration 9912, lr = 0.00140377
I0409 21:15:16.656950 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 21:15:21.332937 15472 solver.cpp:218] Iteration 9924 (2.51364 iter/s, 4.77395s/12 iters), loss = 1.96048
I0409 21:15:21.332993 15472 solver.cpp:237]     Train net output #0: loss = 1.96048 (* 1 = 1.96048 loss)
I0409 21:15:21.333006 15472 sgd_solver.cpp:105] Iteration 9924, lr = 0.00140043
I0409 21:15:26.174078 15472 solver.cpp:218] Iteration 9936 (2.47889 iter/s, 4.84087s/12 iters), loss = 2.09397
I0409 21:15:26.174134 15472 solver.cpp:237]     Train net output #0: loss = 2.09397 (* 1 = 2.09397 loss)
I0409 21:15:26.174145 15472 sgd_solver.cpp:105] Iteration 9936, lr = 0.00139711
I0409 21:15:31.093331 15472 solver.cpp:218] Iteration 9948 (2.43953 iter/s, 4.91899s/12 iters), loss = 1.94448
I0409 21:15:31.093369 15472 solver.cpp:237]     Train net output #0: loss = 1.94448 (* 1 = 1.94448 loss)
I0409 21:15:31.093379 15472 sgd_solver.cpp:105] Iteration 9948, lr = 0.00139379
I0409 21:15:35.883074 15472 solver.cpp:218] Iteration 9960 (2.50549 iter/s, 4.78949s/12 iters), loss = 1.91977
I0409 21:15:35.883126 15472 solver.cpp:237]     Train net output #0: loss = 1.91977 (* 1 = 1.91977 loss)
I0409 21:15:35.883139 15472 sgd_solver.cpp:105] Iteration 9960, lr = 0.00139048
I0409 21:15:40.679121 15472 solver.cpp:218] Iteration 9972 (2.5022 iter/s, 4.79578s/12 iters), loss = 2.10221
I0409 21:15:40.679179 15472 solver.cpp:237]     Train net output #0: loss = 2.10221 (* 1 = 2.10221 loss)
I0409 21:15:40.679193 15472 sgd_solver.cpp:105] Iteration 9972, lr = 0.00138718
I0409 21:15:45.618677 15472 solver.cpp:218] Iteration 9984 (2.4295 iter/s, 4.93928s/12 iters), loss = 2.08823
I0409 21:15:45.618733 15472 solver.cpp:237]     Train net output #0: loss = 2.08823 (* 1 = 2.08823 loss)
I0409 21:15:45.618746 15472 sgd_solver.cpp:105] Iteration 9984, lr = 0.00138389
I0409 21:15:49.998620 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_9996.caffemodel
I0409 21:15:50.471662 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_9996.solverstate
I0409 21:15:51.654055 15472 solver.cpp:330] Iteration 9996, Testing net (#0)
I0409 21:15:51.654083 15472 net.cpp:676] Ignoring source layer train-data
I0409 21:15:52.175645 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 21:15:56.173226 15472 solver.cpp:397]     Test net output #0: accuracy = 0.287377
I0409 21:15:56.173276 15472 solver.cpp:397]     Test net output #1: loss = 2.60998 (* 1 = 2.60998 loss)
I0409 21:15:56.256238 15472 solver.cpp:218] Iteration 9996 (1.12813 iter/s, 10.6371s/12 iters), loss = 2.04609
I0409 21:15:56.256289 15472 solver.cpp:237]     Train net output #0: loss = 2.04609 (* 1 = 2.04609 loss)
I0409 21:15:56.256300 15472 sgd_solver.cpp:105] Iteration 9996, lr = 0.0013806
I0409 21:16:00.353504 15472 solver.cpp:218] Iteration 10008 (2.92895 iter/s, 4.09703s/12 iters), loss = 1.78947
I0409 21:16:00.353550 15472 solver.cpp:237]     Train net output #0: loss = 1.78947 (* 1 = 1.78947 loss)
I0409 21:16:00.353560 15472 sgd_solver.cpp:105] Iteration 10008, lr = 0.00137732
I0409 21:16:02.517192 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 21:16:05.180034 15472 solver.cpp:218] Iteration 10020 (2.4864 iter/s, 4.82626s/12 iters), loss = 2.01901
I0409 21:16:05.180090 15472 solver.cpp:237]     Train net output #0: loss = 2.01901 (* 1 = 2.01901 loss)
I0409 21:16:05.180102 15472 sgd_solver.cpp:105] Iteration 10020, lr = 0.00137405
I0409 21:16:10.235458 15472 solver.cpp:218] Iteration 10032 (2.37382 iter/s, 5.05515s/12 iters), loss = 2.04564
I0409 21:16:10.235512 15472 solver.cpp:237]     Train net output #0: loss = 2.04564 (* 1 = 2.04564 loss)
I0409 21:16:10.235524 15472 sgd_solver.cpp:105] Iteration 10032, lr = 0.00137079
I0409 21:16:15.047453 15472 solver.cpp:218] Iteration 10044 (2.4939 iter/s, 4.81173s/12 iters), loss = 2.18109
I0409 21:16:15.047498 15472 solver.cpp:237]     Train net output #0: loss = 2.18109 (* 1 = 2.18109 loss)
I0409 21:16:15.047508 15472 sgd_solver.cpp:105] Iteration 10044, lr = 0.00136754
I0409 21:16:19.948101 15472 solver.cpp:218] Iteration 10056 (2.44879 iter/s, 4.90038s/12 iters), loss = 1.87002
I0409 21:16:19.948150 15472 solver.cpp:237]     Train net output #0: loss = 1.87002 (* 1 = 1.87002 loss)
I0409 21:16:19.948161 15472 sgd_solver.cpp:105] Iteration 10056, lr = 0.00136429
I0409 21:16:24.757699 15472 solver.cpp:218] Iteration 10068 (2.49515 iter/s, 4.80934s/12 iters), loss = 2.18594
I0409 21:16:24.757812 15472 solver.cpp:237]     Train net output #0: loss = 2.18594 (* 1 = 2.18594 loss)
I0409 21:16:24.757827 15472 sgd_solver.cpp:105] Iteration 10068, lr = 0.00136105
I0409 21:16:29.569622 15472 solver.cpp:218] Iteration 10080 (2.49397 iter/s, 4.8116s/12 iters), loss = 2.02895
I0409 21:16:29.569665 15472 solver.cpp:237]     Train net output #0: loss = 2.02895 (* 1 = 2.02895 loss)
I0409 21:16:29.569675 15472 sgd_solver.cpp:105] Iteration 10080, lr = 0.00135782
I0409 21:16:34.336225 15472 solver.cpp:218] Iteration 10092 (2.51765 iter/s, 4.76635s/12 iters), loss = 2.03168
I0409 21:16:34.336277 15472 solver.cpp:237]     Train net output #0: loss = 2.03168 (* 1 = 2.03168 loss)
I0409 21:16:34.336290 15472 sgd_solver.cpp:105] Iteration 10092, lr = 0.0013546
I0409 21:16:36.278524 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_10098.caffemodel
I0409 21:16:36.727607 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_10098.solverstate
I0409 21:16:37.049227 15472 solver.cpp:330] Iteration 10098, Testing net (#0)
I0409 21:16:37.049255 15472 net.cpp:676] Ignoring source layer train-data
I0409 21:16:37.472508 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 21:16:41.447966 15472 solver.cpp:397]     Test net output #0: accuracy = 0.308211
I0409 21:16:41.448017 15472 solver.cpp:397]     Test net output #1: loss = 2.65596 (* 1 = 2.65596 loss)
I0409 21:16:43.149170 15472 solver.cpp:218] Iteration 10104 (1.3617 iter/s, 8.81251s/12 iters), loss = 1.73197
I0409 21:16:43.149231 15472 solver.cpp:237]     Train net output #0: loss = 1.73197 (* 1 = 1.73197 loss)
I0409 21:16:43.149243 15472 sgd_solver.cpp:105] Iteration 10104, lr = 0.00135138
I0409 21:16:47.338699 15476 data_layer.cpp:73] Restarting data prefetching from start.
I0409 21:16:47.937481 15472 solver.cpp:218] Iteration 10116 (2.50624 iter/s, 4.78804s/12 iters), loss = 1.86349
I0409 21:16:47.937531 15472 solver.cpp:237]     Train net output #0: loss = 1.86349 (* 1 = 1.86349 loss)
I0409 21:16:47.937544 15472 sgd_solver.cpp:105] Iteration 10116, lr = 0.00134817
I0409 21:16:52.835671 15472 solver.cpp:218] Iteration 10128 (2.45002 iter/s, 4.89792s/12 iters), loss = 2.08204
I0409 21:16:52.835726 15472 solver.cpp:237]     Train net output #0: loss = 2.08204 (* 1 = 2.08204 loss)
I0409 21:16:52.835738 15472 sgd_solver.cpp:105] Iteration 10128, lr = 0.00134497
I0409 21:16:57.649873 15472 solver.cpp:218] Iteration 10140 (2.49276 iter/s, 4.81394s/12 iters), loss = 2.104
I0409 21:16:57.650019 15472 solver.cpp:237]     Train net output #0: loss = 2.104 (* 1 = 2.104 loss)
I0409 21:16:57.650033 15472 sgd_solver.cpp:105] Iteration 10140, lr = 0.00134178
I0409 21:17:02.412457 15472 solver.cpp:218] Iteration 10152 (2.51983 iter/s, 4.76223s/12 iters), loss = 1.82418
I0409 21:17:02.412513 15472 solver.cpp:237]     Train net output #0: loss = 1.82418 (* 1 = 1.82418 loss)
I0409 21:17:02.412525 15472 sgd_solver.cpp:105] Iteration 10152, lr = 0.00133859
I0409 21:17:07.195978 15472 solver.cpp:218] Iteration 10164 (2.50875 iter/s, 4.78325s/12 iters), loss = 2.10124
I0409 21:17:07.196038 15472 solver.cpp:237]     Train net output #0: loss = 2.10124 (* 1 = 2.10124 loss)
I0409 21:17:07.196049 15472 sgd_solver.cpp:105] Iteration 10164, lr = 0.00133541
I0409 21:17:11.969314 15472 solver.cpp:218] Iteration 10176 (2.51411 iter/s, 4.77307s/12 iters), loss = 1.98662
I0409 21:17:11.969367 15472 solver.cpp:237]     Train net output #0: loss = 1.98662 (* 1 = 1.98662 loss)
I0409 21:17:11.969378 15472 sgd_solver.cpp:105] Iteration 10176, lr = 0.00133224
I0409 21:17:16.758219 15472 solver.cpp:218] Iteration 10188 (2.50593 iter/s, 4.78864s/12 iters), loss = 1.82618
I0409 21:17:16.758272 15472 solver.cpp:237]     Train net output #0: loss = 1.82618 (* 1 = 1.82618 loss)
I0409 21:17:16.758285 15472 sgd_solver.cpp:105] Iteration 10188, lr = 0.00132908
I0409 21:17:21.104993 15472 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_10200.caffemodel
I0409 21:17:21.991124 15472 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_10200.solverstate
I0409 21:17:22.553282 15472 solver.cpp:310] Iteration 10200, loss = 1.98113
I0409 21:17:22.553318 15472 solver.cpp:330] Iteration 10200, Testing net (#0)
I0409 21:17:22.553328 15472 net.cpp:676] Ignoring source layer train-data
I0409 21:17:22.961777 15477 data_layer.cpp:73] Restarting data prefetching from start.
I0409 21:17:26.977030 15472 solver.cpp:397]     Test net output #0: accuracy = 0.308211
I0409 21:17:26.977063 15472 solver.cpp:397]     Test net output #1: loss = 2.56913 (* 1 = 2.56913 loss)
I0409 21:17:26.977072 15472 solver.cpp:315] Optimization Done.
I0409 21:17:26.977077 15472 caffe.cpp:259] Optimization Done.