DIGITS-CNN/cars/architecture-investigations/conv/layers/layer1.5/kernel/9/caffe_output.log

I0428 13:45:17.816656  7476 upgrade_proto.cpp:1082] Attempting to upgrade input file specified using deprecated 'solver_type' field (enum)': /mnt/bigdisk/DIGITS-MAN-3/digits/jobs/20210428-115432-b930/solver.prototxt
I0428 13:45:17.816807  7476 upgrade_proto.cpp:1089] Successfully upgraded file specified using deprecated 'solver_type' field (enum) to 'type' field (string).
W0428 13:45:17.816812  7476 upgrade_proto.cpp:1091] Note that future Caffe releases will only support 'type' field (string) for a solver's type.
I0428 13:45:17.816869  7476 caffe.cpp:218] Using GPUs 1
I0428 13:45:17.892688  7476 caffe.cpp:223] GPU 1: GeForce GTX 1080 Ti
I0428 13:45:18.228099  7476 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: 1
net: "train_val.prototxt"
train_state {
level: 0
stage: ""
}
type: "SGD"
I0428 13:45:18.229274  7476 solver.cpp:87] Creating training net from net file: train_val.prototxt
I0428 13:45:18.231817  7476 net.cpp:294] The NetState phase (0) differed from the phase (1) specified by a rule in layer val-data
I0428 13:45:18.231833  7476 net.cpp:294] The NetState phase (0) differed from the phase (1) specified by a rule in layer accuracy
I0428 13:45:18.231977  7476 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-3/digits/jobs/20210421-230320-902c/mean.binaryproto"
}
data_param {
source: "/mnt/bigdisk/DIGITS-MAN-3/digits/jobs/20210421-230320-902c/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: "conv1.5"
type: "Convolution"
bottom: "pool1"
top: "conv1.5"
param {
lr_mult: 1
decay_mult: 1
}
param {
lr_mult: 2
decay_mult: 0
}
convolution_param {
num_output: 176
kernel_size: 9
stride: 1
weight_filler {
type: "gaussian"
std: 0.01
}
bias_filler {
type: "constant"
value: 0
}
}
}
layer {
name: "relu1.5"
type: "ReLU"
bottom: "conv1.5"
top: "conv1.5"
}
layer {
name: "norm1.5"
type: "LRN"
bottom: "conv1.5"
top: "norm1.5"
lrn_param {
local_size: 5
alpha: 0.0001
beta: 0.75
}
}
layer {
name: "pool1.5"
type: "Pooling"
bottom: "norm1.5"
top: "pool1.5"
pooling_param {
pool: MAX
kernel_size: 3
stride: 2
}
}
layer {
name: "conv2"
type: "Convolution"
bottom: "pool1.5"
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: 4096
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: 4096
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: "fc8"
type: "InnerProduct"
bottom: "fc7"
top: "fc8"
param {
lr_mult: 1
decay_mult: 1
}
param {
lr_mult: 2
decay_mult: 0
}
inner_product_param {
num_output: 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"
}
I0428 13:45:18.232069  7476 layer_factory.hpp:77] Creating layer train-data
I0428 13:45:18.234679  7476 db_lmdb.cpp:35] Opened lmdb /mnt/bigdisk/DIGITS-MAN-3/digits/jobs/20210421-230320-902c/train_db
I0428 13:45:18.235126  7476 net.cpp:84] Creating Layer train-data
I0428 13:45:18.235137  7476 net.cpp:380] train-data -> data
I0428 13:45:18.235157  7476 net.cpp:380] train-data -> label
I0428 13:45:18.235167  7476 data_transformer.cpp:25] Loading mean file from: /mnt/bigdisk/DIGITS-MAN-3/digits/jobs/20210421-230320-902c/mean.binaryproto
I0428 13:45:18.240533  7476 data_layer.cpp:45] output data size: 128,3,227,227
I0428 13:45:18.472137  7476 net.cpp:122] Setting up train-data
I0428 13:45:18.472162  7476 net.cpp:129] Top shape: 128 3 227 227 (19787136)
I0428 13:45:18.472167  7476 net.cpp:129] Top shape: 128 (128)
I0428 13:45:18.472172  7476 net.cpp:137] Memory required for data: 79149056
I0428 13:45:18.472182  7476 layer_factory.hpp:77] Creating layer conv1
I0428 13:45:18.472203  7476 net.cpp:84] Creating Layer conv1
I0428 13:45:18.472209  7476 net.cpp:406] conv1 <- data
I0428 13:45:18.472223  7476 net.cpp:380] conv1 -> conv1
I0428 13:45:19.453624  7476 net.cpp:122] Setting up conv1
I0428 13:45:19.453672  7476 net.cpp:129] Top shape: 128 96 55 55 (37171200)
I0428 13:45:19.453678  7476 net.cpp:137] Memory required for data: 227833856
I0428 13:45:19.453699  7476 layer_factory.hpp:77] Creating layer relu1
I0428 13:45:19.453711  7476 net.cpp:84] Creating Layer relu1
I0428 13:45:19.453716  7476 net.cpp:406] relu1 <- conv1
I0428 13:45:19.453723  7476 net.cpp:367] relu1 -> conv1 (in-place)
I0428 13:45:19.454077  7476 net.cpp:122] Setting up relu1
I0428 13:45:19.454087  7476 net.cpp:129] Top shape: 128 96 55 55 (37171200)
I0428 13:45:19.454092  7476 net.cpp:137] Memory required for data: 376518656
I0428 13:45:19.454095  7476 layer_factory.hpp:77] Creating layer norm1
I0428 13:45:19.454106  7476 net.cpp:84] Creating Layer norm1
I0428 13:45:19.454111  7476 net.cpp:406] norm1 <- conv1
I0428 13:45:19.454116  7476 net.cpp:380] norm1 -> norm1
I0428 13:45:19.454656  7476 net.cpp:122] Setting up norm1
I0428 13:45:19.454668  7476 net.cpp:129] Top shape: 128 96 55 55 (37171200)
I0428 13:45:19.454671  7476 net.cpp:137] Memory required for data: 525203456
I0428 13:45:19.454676  7476 layer_factory.hpp:77] Creating layer pool1
I0428 13:45:19.454685  7476 net.cpp:84] Creating Layer pool1
I0428 13:45:19.454689  7476 net.cpp:406] pool1 <- norm1
I0428 13:45:19.454695  7476 net.cpp:380] pool1 -> pool1
I0428 13:45:19.454739  7476 net.cpp:122] Setting up pool1
I0428 13:45:19.454746  7476 net.cpp:129] Top shape: 128 96 27 27 (8957952)
I0428 13:45:19.454751  7476 net.cpp:137] Memory required for data: 561035264
I0428 13:45:19.454754  7476 layer_factory.hpp:77] Creating layer conv1.5
I0428 13:45:19.454766  7476 net.cpp:84] Creating Layer conv1.5
I0428 13:45:19.454769  7476 net.cpp:406] conv1.5 <- pool1
I0428 13:45:19.454775  7476 net.cpp:380] conv1.5 -> conv1.5
I0428 13:45:19.472767  7476 net.cpp:122] Setting up conv1.5
I0428 13:45:19.472788  7476 net.cpp:129] Top shape: 128 176 19 19 (8132608)
I0428 13:45:19.472793  7476 net.cpp:137] Memory required for data: 593565696
I0428 13:45:19.472805  7476 layer_factory.hpp:77] Creating layer relu1.5
I0428 13:45:19.472815  7476 net.cpp:84] Creating Layer relu1.5
I0428 13:45:19.472820  7476 net.cpp:406] relu1.5 <- conv1.5
I0428 13:45:19.472827  7476 net.cpp:367] relu1.5 -> conv1.5 (in-place)
I0428 13:45:19.473244  7476 net.cpp:122] Setting up relu1.5
I0428 13:45:19.473253  7476 net.cpp:129] Top shape: 128 176 19 19 (8132608)
I0428 13:45:19.473256  7476 net.cpp:137] Memory required for data: 626096128
I0428 13:45:19.473260  7476 layer_factory.hpp:77] Creating layer norm1.5
I0428 13:45:19.473270  7476 net.cpp:84] Creating Layer norm1.5
I0428 13:45:19.473274  7476 net.cpp:406] norm1.5 <- conv1.5
I0428 13:45:19.473280  7476 net.cpp:380] norm1.5 -> norm1.5
I0428 13:45:19.474995  7476 net.cpp:122] Setting up norm1.5
I0428 13:45:19.475008  7476 net.cpp:129] Top shape: 128 176 19 19 (8132608)
I0428 13:45:19.475011  7476 net.cpp:137] Memory required for data: 658626560
I0428 13:45:19.475015  7476 layer_factory.hpp:77] Creating layer pool1.5
I0428 13:45:19.475026  7476 net.cpp:84] Creating Layer pool1.5
I0428 13:45:19.475031  7476 net.cpp:406] pool1.5 <- norm1.5
I0428 13:45:19.475037  7476 net.cpp:380] pool1.5 -> pool1.5
I0428 13:45:19.475073  7476 net.cpp:122] Setting up pool1.5
I0428 13:45:19.475080  7476 net.cpp:129] Top shape: 128 176 9 9 (1824768)
I0428 13:45:19.475082  7476 net.cpp:137] Memory required for data: 665925632
I0428 13:45:19.475086  7476 layer_factory.hpp:77] Creating layer conv2
I0428 13:45:19.475098  7476 net.cpp:84] Creating Layer conv2
I0428 13:45:19.475102  7476 net.cpp:406] conv2 <- pool1.5
I0428 13:45:19.475109  7476 net.cpp:380] conv2 -> conv2
I0428 13:45:19.485344  7476 net.cpp:122] Setting up conv2
I0428 13:45:19.485363  7476 net.cpp:129] Top shape: 128 256 9 9 (2654208)
I0428 13:45:19.485368  7476 net.cpp:137] Memory required for data: 676542464
I0428 13:45:19.485381  7476 layer_factory.hpp:77] Creating layer relu2
I0428 13:45:19.485389  7476 net.cpp:84] Creating Layer relu2
I0428 13:45:19.485394  7476 net.cpp:406] relu2 <- conv2
I0428 13:45:19.485400  7476 net.cpp:367] relu2 -> conv2 (in-place)
I0428 13:45:19.485973  7476 net.cpp:122] Setting up relu2
I0428 13:45:19.485983  7476 net.cpp:129] Top shape: 128 256 9 9 (2654208)
I0428 13:45:19.485987  7476 net.cpp:137] Memory required for data: 687159296
I0428 13:45:19.485991  7476 layer_factory.hpp:77] Creating layer norm2
I0428 13:45:19.485999  7476 net.cpp:84] Creating Layer norm2
I0428 13:45:19.486002  7476 net.cpp:406] norm2 <- conv2
I0428 13:45:19.486011  7476 net.cpp:380] norm2 -> norm2
I0428 13:45:19.486413  7476 net.cpp:122] Setting up norm2
I0428 13:45:19.486423  7476 net.cpp:129] Top shape: 128 256 9 9 (2654208)
I0428 13:45:19.486425  7476 net.cpp:137] Memory required for data: 697776128
I0428 13:45:19.486429  7476 layer_factory.hpp:77] Creating layer pool2
I0428 13:45:19.486436  7476 net.cpp:84] Creating Layer pool2
I0428 13:45:19.486439  7476 net.cpp:406] pool2 <- norm2
I0428 13:45:19.486446  7476 net.cpp:380] pool2 -> pool2
I0428 13:45:19.486479  7476 net.cpp:122] Setting up pool2
I0428 13:45:19.486485  7476 net.cpp:129] Top shape: 128 256 4 4 (524288)
I0428 13:45:19.486488  7476 net.cpp:137] Memory required for data: 699873280
I0428 13:45:19.486492  7476 layer_factory.hpp:77] Creating layer conv3
I0428 13:45:19.486502  7476 net.cpp:84] Creating Layer conv3
I0428 13:45:19.486505  7476 net.cpp:406] conv3 <- pool2
I0428 13:45:19.486512  7476 net.cpp:380] conv3 -> conv3
I0428 13:45:19.524003  7476 net.cpp:122] Setting up conv3
I0428 13:45:19.524022  7476 net.cpp:129] Top shape: 128 384 4 4 (786432)
I0428 13:45:19.524026  7476 net.cpp:137] Memory required for data: 703019008
I0428 13:45:19.524036  7476 layer_factory.hpp:77] Creating layer relu3
I0428 13:45:19.524046  7476 net.cpp:84] Creating Layer relu3
I0428 13:45:19.524050  7476 net.cpp:406] relu3 <- conv3
I0428 13:45:19.524057  7476 net.cpp:367] relu3 -> conv3 (in-place)
I0428 13:45:19.557474  7476 net.cpp:122] Setting up relu3
I0428 13:45:19.557494  7476 net.cpp:129] Top shape: 128 384 4 4 (786432)
I0428 13:45:19.557499  7476 net.cpp:137] Memory required for data: 706164736
I0428 13:45:19.557505  7476 layer_factory.hpp:77] Creating layer conv4
I0428 13:45:19.557523  7476 net.cpp:84] Creating Layer conv4
I0428 13:45:19.557528  7476 net.cpp:406] conv4 <- conv3
I0428 13:45:19.557538  7476 net.cpp:380] conv4 -> conv4
I0428 13:45:19.592514  7476 net.cpp:122] Setting up conv4
I0428 13:45:19.592535  7476 net.cpp:129] Top shape: 128 384 4 4 (786432)
I0428 13:45:19.592540  7476 net.cpp:137] Memory required for data: 709310464
I0428 13:45:19.592555  7476 layer_factory.hpp:77] Creating layer relu4
I0428 13:45:19.592569  7476 net.cpp:84] Creating Layer relu4
I0428 13:45:19.592574  7476 net.cpp:406] relu4 <- conv4
I0428 13:45:19.592581  7476 net.cpp:367] relu4 -> conv4 (in-place)
I0428 13:45:19.593127  7476 net.cpp:122] Setting up relu4
I0428 13:45:19.593137  7476 net.cpp:129] Top shape: 128 384 4 4 (786432)
I0428 13:45:19.593142  7476 net.cpp:137] Memory required for data: 712456192
I0428 13:45:19.593145  7476 layer_factory.hpp:77] Creating layer conv5
I0428 13:45:19.593158  7476 net.cpp:84] Creating Layer conv5
I0428 13:45:19.593163  7476 net.cpp:406] conv5 <- conv4
I0428 13:45:19.593170  7476 net.cpp:380] conv5 -> conv5
I0428 13:45:19.603315  7476 net.cpp:122] Setting up conv5
I0428 13:45:19.603334  7476 net.cpp:129] Top shape: 128 256 4 4 (524288)
I0428 13:45:19.603338  7476 net.cpp:137] Memory required for data: 714553344
I0428 13:45:19.603348  7476 layer_factory.hpp:77] Creating layer relu5
I0428 13:45:19.603358  7476 net.cpp:84] Creating Layer relu5
I0428 13:45:19.603363  7476 net.cpp:406] relu5 <- conv5
I0428 13:45:19.603370  7476 net.cpp:367] relu5 -> conv5 (in-place)
I0428 13:45:19.603994  7476 net.cpp:122] Setting up relu5
I0428 13:45:19.604005  7476 net.cpp:129] Top shape: 128 256 4 4 (524288)
I0428 13:45:19.604009  7476 net.cpp:137] Memory required for data: 716650496
I0428 13:45:19.604013  7476 layer_factory.hpp:77] Creating layer pool5
I0428 13:45:19.604020  7476 net.cpp:84] Creating Layer pool5
I0428 13:45:19.604024  7476 net.cpp:406] pool5 <- conv5
I0428 13:45:19.604030  7476 net.cpp:380] pool5 -> pool5
I0428 13:45:19.604091  7476 net.cpp:122] Setting up pool5
I0428 13:45:19.604097  7476 net.cpp:129] Top shape: 128 256 2 2 (131072)
I0428 13:45:19.604101  7476 net.cpp:137] Memory required for data: 717174784
I0428 13:45:19.604104  7476 layer_factory.hpp:77] Creating layer fc6
I0428 13:45:19.604113  7476 net.cpp:84] Creating Layer fc6
I0428 13:45:19.604116  7476 net.cpp:406] fc6 <- pool5
I0428 13:45:19.604122  7476 net.cpp:380] fc6 -> fc6
I0428 13:45:19.703140  7476 net.cpp:122] Setting up fc6
I0428 13:45:19.703158  7476 net.cpp:129] Top shape: 128 4096 (524288)
I0428 13:45:19.703162  7476 net.cpp:137] Memory required for data: 719271936
I0428 13:45:19.703172  7476 layer_factory.hpp:77] Creating layer relu6
I0428 13:45:19.703181  7476 net.cpp:84] Creating Layer relu6
I0428 13:45:19.703187  7476 net.cpp:406] relu6 <- fc6
I0428 13:45:19.703194  7476 net.cpp:367] relu6 -> fc6 (in-place)
I0428 13:45:19.703845  7476 net.cpp:122] Setting up relu6
I0428 13:45:19.703855  7476 net.cpp:129] Top shape: 128 4096 (524288)
I0428 13:45:19.703858  7476 net.cpp:137] Memory required for data: 721369088
I0428 13:45:19.703863  7476 layer_factory.hpp:77] Creating layer drop6
I0428 13:45:19.703871  7476 net.cpp:84] Creating Layer drop6
I0428 13:45:19.703876  7476 net.cpp:406] drop6 <- fc6
I0428 13:45:19.703881  7476 net.cpp:367] drop6 -> fc6 (in-place)
I0428 13:45:19.703912  7476 net.cpp:122] Setting up drop6
I0428 13:45:19.703917  7476 net.cpp:129] Top shape: 128 4096 (524288)
I0428 13:45:19.703922  7476 net.cpp:137] Memory required for data: 723466240
I0428 13:45:19.703927  7476 layer_factory.hpp:77] Creating layer fc7
I0428 13:45:19.703933  7476 net.cpp:84] Creating Layer fc7
I0428 13:45:19.703938  7476 net.cpp:406] fc7 <- fc6
I0428 13:45:19.703945  7476 net.cpp:380] fc7 -> fc7
I0428 13:45:19.877452  7476 net.cpp:122] Setting up fc7
I0428 13:45:19.877475  7476 net.cpp:129] Top shape: 128 4096 (524288)
I0428 13:45:19.877480  7476 net.cpp:137] Memory required for data: 725563392
I0428 13:45:19.877488  7476 layer_factory.hpp:77] Creating layer relu7
I0428 13:45:19.877497  7476 net.cpp:84] Creating Layer relu7
I0428 13:45:19.877502  7476 net.cpp:406] relu7 <- fc7
I0428 13:45:19.877511  7476 net.cpp:367] relu7 -> fc7 (in-place)
I0428 13:45:19.877948  7476 net.cpp:122] Setting up relu7
I0428 13:45:19.877956  7476 net.cpp:129] Top shape: 128 4096 (524288)
I0428 13:45:19.877959  7476 net.cpp:137] Memory required for data: 727660544
I0428 13:45:19.877964  7476 layer_factory.hpp:77] Creating layer drop7
I0428 13:45:19.877972  7476 net.cpp:84] Creating Layer drop7
I0428 13:45:19.877976  7476 net.cpp:406] drop7 <- fc7
I0428 13:45:19.877981  7476 net.cpp:367] drop7 -> fc7 (in-place)
I0428 13:45:19.878006  7476 net.cpp:122] Setting up drop7
I0428 13:45:19.878011  7476 net.cpp:129] Top shape: 128 4096 (524288)
I0428 13:45:19.878015  7476 net.cpp:137] Memory required for data: 729757696
I0428 13:45:19.878018  7476 layer_factory.hpp:77] Creating layer fc8
I0428 13:45:19.878026  7476 net.cpp:84] Creating Layer fc8
I0428 13:45:19.878028  7476 net.cpp:406] fc8 <- fc7
I0428 13:45:19.878036  7476 net.cpp:380] fc8 -> fc8
I0428 13:45:19.886535  7476 net.cpp:122] Setting up fc8
I0428 13:45:19.886549  7476 net.cpp:129] Top shape: 128 196 (25088)
I0428 13:45:19.886554  7476 net.cpp:137] Memory required for data: 729858048
I0428 13:45:19.886567  7476 layer_factory.hpp:77] Creating layer loss
I0428 13:45:19.886574  7476 net.cpp:84] Creating Layer loss
I0428 13:45:19.886579  7476 net.cpp:406] loss <- fc8
I0428 13:45:19.886584  7476 net.cpp:406] loss <- label
I0428 13:45:19.886591  7476 net.cpp:380] loss -> loss
I0428 13:45:19.886601  7476 layer_factory.hpp:77] Creating layer loss
I0428 13:45:19.887295  7476 net.cpp:122] Setting up loss
I0428 13:45:19.887305  7476 net.cpp:129] Top shape: (1)
I0428 13:45:19.887308  7476 net.cpp:132]     with loss weight 1
I0428 13:45:19.887326  7476 net.cpp:137] Memory required for data: 729858052
I0428 13:45:19.887331  7476 net.cpp:198] loss needs backward computation.
I0428 13:45:19.887337  7476 net.cpp:198] fc8 needs backward computation.
I0428 13:45:19.887360  7476 net.cpp:198] drop7 needs backward computation.
I0428 13:45:19.887364  7476 net.cpp:198] relu7 needs backward computation.
I0428 13:45:19.887368  7476 net.cpp:198] fc7 needs backward computation.
I0428 13:45:19.887372  7476 net.cpp:198] drop6 needs backward computation.
I0428 13:45:19.887377  7476 net.cpp:198] relu6 needs backward computation.
I0428 13:45:19.887379  7476 net.cpp:198] fc6 needs backward computation.
I0428 13:45:19.887383  7476 net.cpp:198] pool5 needs backward computation.
I0428 13:45:19.887387  7476 net.cpp:198] relu5 needs backward computation.
I0428 13:45:19.887392  7476 net.cpp:198] conv5 needs backward computation.
I0428 13:45:19.887395  7476 net.cpp:198] relu4 needs backward computation.
I0428 13:45:19.887398  7476 net.cpp:198] conv4 needs backward computation.
I0428 13:45:19.887403  7476 net.cpp:198] relu3 needs backward computation.
I0428 13:45:19.887406  7476 net.cpp:198] conv3 needs backward computation.
I0428 13:45:19.887410  7476 net.cpp:198] pool2 needs backward computation.
I0428 13:45:19.887413  7476 net.cpp:198] norm2 needs backward computation.
I0428 13:45:19.887418  7476 net.cpp:198] relu2 needs backward computation.
I0428 13:45:19.887423  7476 net.cpp:198] conv2 needs backward computation.
I0428 13:45:19.887428  7476 net.cpp:198] pool1.5 needs backward computation.
I0428 13:45:19.887431  7476 net.cpp:198] norm1.5 needs backward computation.
I0428 13:45:19.887435  7476 net.cpp:198] relu1.5 needs backward computation.
I0428 13:45:19.887439  7476 net.cpp:198] conv1.5 needs backward computation.
I0428 13:45:19.887444  7476 net.cpp:198] pool1 needs backward computation.
I0428 13:45:19.887447  7476 net.cpp:198] norm1 needs backward computation.
I0428 13:45:19.887451  7476 net.cpp:198] relu1 needs backward computation.
I0428 13:45:19.887455  7476 net.cpp:198] conv1 needs backward computation.
I0428 13:45:19.887459  7476 net.cpp:200] train-data does not need backward computation.
I0428 13:45:19.887463  7476 net.cpp:242] This network produces output loss
I0428 13:45:19.887481  7476 net.cpp:255] Network initialization done.
I0428 13:45:19.888031  7476 solver.cpp:172] Creating test net (#0) specified by net file: train_val.prototxt
I0428 13:45:19.888067  7476 net.cpp:294] The NetState phase (1) differed from the phase (0) specified by a rule in layer train-data
I0428 13:45:19.888223  7476 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-3/digits/jobs/20210421-230320-902c/mean.binaryproto"
}
data_param {
source: "/mnt/bigdisk/DIGITS-MAN-3/digits/jobs/20210421-230320-902c/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: "conv1.5"
type: "Convolution"
bottom: "pool1"
top: "conv1.5"
param {
lr_mult: 1
decay_mult: 1
}
param {
lr_mult: 2
decay_mult: 0
}
convolution_param {
num_output: 176
kernel_size: 9
stride: 1
weight_filler {
type: "gaussian"
std: 0.01
}
bias_filler {
type: "constant"
value: 0
}
}
}
layer {
name: "relu1.5"
type: "ReLU"
bottom: "conv1.5"
top: "conv1.5"
}
layer {
name: "norm1.5"
type: "LRN"
bottom: "conv1.5"
top: "norm1.5"
lrn_param {
local_size: 5
alpha: 0.0001
beta: 0.75
}
}
layer {
name: "pool1.5"
type: "Pooling"
bottom: "norm1.5"
top: "pool1.5"
pooling_param {
pool: MAX
kernel_size: 3
stride: 2
}
}
layer {
name: "conv2"
type: "Convolution"
bottom: "pool1.5"
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: 4096
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: 4096
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: "fc8"
type: "InnerProduct"
bottom: "fc7"
top: "fc8"
param {
lr_mult: 1
decay_mult: 1
}
param {
lr_mult: 2
decay_mult: 0
}
inner_product_param {
num_output: 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"
}
I0428 13:45:19.888336  7476 layer_factory.hpp:77] Creating layer val-data
I0428 13:45:19.891165  7476 db_lmdb.cpp:35] Opened lmdb /mnt/bigdisk/DIGITS-MAN-3/digits/jobs/20210421-230320-902c/val_db
I0428 13:45:19.891350  7476 net.cpp:84] Creating Layer val-data
I0428 13:45:19.891360  7476 net.cpp:380] val-data -> data
I0428 13:45:19.891368  7476 net.cpp:380] val-data -> label
I0428 13:45:19.891376  7476 data_transformer.cpp:25] Loading mean file from: /mnt/bigdisk/DIGITS-MAN-3/digits/jobs/20210421-230320-902c/mean.binaryproto
I0428 13:45:19.894812  7476 data_layer.cpp:45] output data size: 32,3,227,227
I0428 13:45:19.937503  7476 net.cpp:122] Setting up val-data
I0428 13:45:19.937525  7476 net.cpp:129] Top shape: 32 3 227 227 (4946784)
I0428 13:45:19.937530  7476 net.cpp:129] Top shape: 32 (32)
I0428 13:45:19.937533  7476 net.cpp:137] Memory required for data: 19787264
I0428 13:45:19.937539  7476 layer_factory.hpp:77] Creating layer label_val-data_1_split
I0428 13:45:19.937551  7476 net.cpp:84] Creating Layer label_val-data_1_split
I0428 13:45:19.937556  7476 net.cpp:406] label_val-data_1_split <- label
I0428 13:45:19.937562  7476 net.cpp:380] label_val-data_1_split -> label_val-data_1_split_0
I0428 13:45:19.937572  7476 net.cpp:380] label_val-data_1_split -> label_val-data_1_split_1
I0428 13:45:19.937628  7476 net.cpp:122] Setting up label_val-data_1_split
I0428 13:45:19.937634  7476 net.cpp:129] Top shape: 32 (32)
I0428 13:45:19.937638  7476 net.cpp:129] Top shape: 32 (32)
I0428 13:45:19.937641  7476 net.cpp:137] Memory required for data: 19787520
I0428 13:45:19.937644  7476 layer_factory.hpp:77] Creating layer conv1
I0428 13:45:19.937656  7476 net.cpp:84] Creating Layer conv1
I0428 13:45:19.937660  7476 net.cpp:406] conv1 <- data
I0428 13:45:19.937666  7476 net.cpp:380] conv1 -> conv1
I0428 13:45:19.987378  7476 net.cpp:122] Setting up conv1
I0428 13:45:19.987397  7476 net.cpp:129] Top shape: 32 96 55 55 (9292800)
I0428 13:45:19.987401  7476 net.cpp:137] Memory required for data: 56958720
I0428 13:45:19.987416  7476 layer_factory.hpp:77] Creating layer relu1
I0428 13:45:19.987424  7476 net.cpp:84] Creating Layer relu1
I0428 13:45:19.987429  7476 net.cpp:406] relu1 <- conv1
I0428 13:45:19.987435  7476 net.cpp:367] relu1 -> conv1 (in-place)
I0428 13:45:19.987722  7476 net.cpp:122] Setting up relu1
I0428 13:45:19.987731  7476 net.cpp:129] Top shape: 32 96 55 55 (9292800)
I0428 13:45:19.987735  7476 net.cpp:137] Memory required for data: 94129920
I0428 13:45:19.987738  7476 layer_factory.hpp:77] Creating layer norm1
I0428 13:45:19.987747  7476 net.cpp:84] Creating Layer norm1
I0428 13:45:19.987751  7476 net.cpp:406] norm1 <- conv1
I0428 13:45:19.987756  7476 net.cpp:380] norm1 -> norm1
I0428 13:45:19.996932  7476 net.cpp:122] Setting up norm1
I0428 13:45:19.996944  7476 net.cpp:129] Top shape: 32 96 55 55 (9292800)
I0428 13:45:19.996948  7476 net.cpp:137] Memory required for data: 131301120
I0428 13:45:19.996951  7476 layer_factory.hpp:77] Creating layer pool1
I0428 13:45:19.996958  7476 net.cpp:84] Creating Layer pool1
I0428 13:45:19.996963  7476 net.cpp:406] pool1 <- norm1
I0428 13:45:19.996968  7476 net.cpp:380] pool1 -> pool1
I0428 13:45:19.996999  7476 net.cpp:122] Setting up pool1
I0428 13:45:19.997004  7476 net.cpp:129] Top shape: 32 96 27 27 (2239488)
I0428 13:45:19.997007  7476 net.cpp:137] Memory required for data: 140259072
I0428 13:45:19.997010  7476 layer_factory.hpp:77] Creating layer conv1.5
I0428 13:45:19.997020  7476 net.cpp:84] Creating Layer conv1.5
I0428 13:45:19.997023  7476 net.cpp:406] conv1.5 <- pool1
I0428 13:45:19.997028  7476 net.cpp:380] conv1.5 -> conv1.5
I0428 13:45:20.023331  7476 net.cpp:122] Setting up conv1.5
I0428 13:45:20.023350  7476 net.cpp:129] Top shape: 32 176 19 19 (2033152)
I0428 13:45:20.023355  7476 net.cpp:137] Memory required for data: 148391680
I0428 13:45:20.023366  7476 layer_factory.hpp:77] Creating layer relu1.5
I0428 13:45:20.023377  7476 net.cpp:84] Creating Layer relu1.5
I0428 13:45:20.023401  7476 net.cpp:406] relu1.5 <- conv1.5
I0428 13:45:20.023407  7476 net.cpp:367] relu1.5 -> conv1.5 (in-place)
I0428 13:45:20.023769  7476 net.cpp:122] Setting up relu1.5
I0428 13:45:20.023778  7476 net.cpp:129] Top shape: 32 176 19 19 (2033152)
I0428 13:45:20.023782  7476 net.cpp:137] Memory required for data: 156524288
I0428 13:45:20.023785  7476 layer_factory.hpp:77] Creating layer norm1.5
I0428 13:45:20.023795  7476 net.cpp:84] Creating Layer norm1.5
I0428 13:45:20.023799  7476 net.cpp:406] norm1.5 <- conv1.5
I0428 13:45:20.023806  7476 net.cpp:380] norm1.5 -> norm1.5
I0428 13:45:20.024382  7476 net.cpp:122] Setting up norm1.5
I0428 13:45:20.024394  7476 net.cpp:129] Top shape: 32 176 19 19 (2033152)
I0428 13:45:20.024399  7476 net.cpp:137] Memory required for data: 164656896
I0428 13:45:20.024405  7476 layer_factory.hpp:77] Creating layer pool1.5
I0428 13:45:20.024417  7476 net.cpp:84] Creating Layer pool1.5
I0428 13:45:20.024423  7476 net.cpp:406] pool1.5 <- norm1.5
I0428 13:45:20.024430  7476 net.cpp:380] pool1.5 -> pool1.5
I0428 13:45:20.024473  7476 net.cpp:122] Setting up pool1.5
I0428 13:45:20.024480  7476 net.cpp:129] Top shape: 32 176 9 9 (456192)
I0428 13:45:20.024483  7476 net.cpp:137] Memory required for data: 166481664
I0428 13:45:20.024521  7476 layer_factory.hpp:77] Creating layer conv2
I0428 13:45:20.024534  7476 net.cpp:84] Creating Layer conv2
I0428 13:45:20.024538  7476 net.cpp:406] conv2 <- pool1.5
I0428 13:45:20.024544  7476 net.cpp:380] conv2 -> conv2
I0428 13:45:20.034204  7476 net.cpp:122] Setting up conv2
I0428 13:45:20.034222  7476 net.cpp:129] Top shape: 32 256 9 9 (663552)
I0428 13:45:20.034226  7476 net.cpp:137] Memory required for data: 169135872
I0428 13:45:20.034237  7476 layer_factory.hpp:77] Creating layer relu2
I0428 13:45:20.034246  7476 net.cpp:84] Creating Layer relu2
I0428 13:45:20.034250  7476 net.cpp:406] relu2 <- conv2
I0428 13:45:20.034256  7476 net.cpp:367] relu2 -> conv2 (in-place)
I0428 13:45:20.034768  7476 net.cpp:122] Setting up relu2
I0428 13:45:20.034778  7476 net.cpp:129] Top shape: 32 256 9 9 (663552)
I0428 13:45:20.034781  7476 net.cpp:137] Memory required for data: 171790080
I0428 13:45:20.034785  7476 layer_factory.hpp:77] Creating layer norm2
I0428 13:45:20.034792  7476 net.cpp:84] Creating Layer norm2
I0428 13:45:20.034796  7476 net.cpp:406] norm2 <- conv2
I0428 13:45:20.034803  7476 net.cpp:380] norm2 -> norm2
I0428 13:45:20.035169  7476 net.cpp:122] Setting up norm2
I0428 13:45:20.035177  7476 net.cpp:129] Top shape: 32 256 9 9 (663552)
I0428 13:45:20.035181  7476 net.cpp:137] Memory required for data: 174444288
I0428 13:45:20.035184  7476 layer_factory.hpp:77] Creating layer pool2
I0428 13:45:20.035192  7476 net.cpp:84] Creating Layer pool2
I0428 13:45:20.035194  7476 net.cpp:406] pool2 <- norm2
I0428 13:45:20.035200  7476 net.cpp:380] pool2 -> pool2
I0428 13:45:20.035231  7476 net.cpp:122] Setting up pool2
I0428 13:45:20.035238  7476 net.cpp:129] Top shape: 32 256 4 4 (131072)
I0428 13:45:20.035240  7476 net.cpp:137] Memory required for data: 174968576
I0428 13:45:20.035244  7476 layer_factory.hpp:77] Creating layer conv3
I0428 13:45:20.035254  7476 net.cpp:84] Creating Layer conv3
I0428 13:45:20.035257  7476 net.cpp:406] conv3 <- pool2
I0428 13:45:20.035262  7476 net.cpp:380] conv3 -> conv3
I0428 13:45:20.046614  7476 net.cpp:122] Setting up conv3
I0428 13:45:20.046634  7476 net.cpp:129] Top shape: 32 384 4 4 (196608)
I0428 13:45:20.046639  7476 net.cpp:137] Memory required for data: 175755008
I0428 13:45:20.046648  7476 layer_factory.hpp:77] Creating layer relu3
I0428 13:45:20.046656  7476 net.cpp:84] Creating Layer relu3
I0428 13:45:20.046660  7476 net.cpp:406] relu3 <- conv3
I0428 13:45:20.046667  7476 net.cpp:367] relu3 -> conv3 (in-place)
I0428 13:45:20.047189  7476 net.cpp:122] Setting up relu3
I0428 13:45:20.047199  7476 net.cpp:129] Top shape: 32 384 4 4 (196608)
I0428 13:45:20.047204  7476 net.cpp:137] Memory required for data: 176541440
I0428 13:45:20.047207  7476 layer_factory.hpp:77] Creating layer conv4
I0428 13:45:20.047241  7476 net.cpp:84] Creating Layer conv4
I0428 13:45:20.047245  7476 net.cpp:406] conv4 <- conv3
I0428 13:45:20.047251  7476 net.cpp:380] conv4 -> conv4
I0428 13:45:20.056877  7476 net.cpp:122] Setting up conv4
I0428 13:45:20.056893  7476 net.cpp:129] Top shape: 32 384 4 4 (196608)
I0428 13:45:20.056897  7476 net.cpp:137] Memory required for data: 177327872
I0428 13:45:20.056910  7476 layer_factory.hpp:77] Creating layer relu4
I0428 13:45:20.056917  7476 net.cpp:84] Creating Layer relu4
I0428 13:45:20.056921  7476 net.cpp:406] relu4 <- conv4
I0428 13:45:20.056928  7476 net.cpp:367] relu4 -> conv4 (in-place)
I0428 13:45:20.057426  7476 net.cpp:122] Setting up relu4
I0428 13:45:20.057437  7476 net.cpp:129] Top shape: 32 384 4 4 (196608)
I0428 13:45:20.057441  7476 net.cpp:137] Memory required for data: 178114304
I0428 13:45:20.057446  7476 layer_factory.hpp:77] Creating layer conv5
I0428 13:45:20.057456  7476 net.cpp:84] Creating Layer conv5
I0428 13:45:20.057461  7476 net.cpp:406] conv5 <- conv4
I0428 13:45:20.057466  7476 net.cpp:380] conv5 -> conv5
I0428 13:45:20.067272  7476 net.cpp:122] Setting up conv5
I0428 13:45:20.067289  7476 net.cpp:129] Top shape: 32 256 4 4 (131072)
I0428 13:45:20.067293  7476 net.cpp:137] Memory required for data: 178638592
I0428 13:45:20.067302  7476 layer_factory.hpp:77] Creating layer relu5
I0428 13:45:20.067312  7476 net.cpp:84] Creating Layer relu5
I0428 13:45:20.067315  7476 net.cpp:406] relu5 <- conv5
I0428 13:45:20.067322  7476 net.cpp:367] relu5 -> conv5 (in-place)
I0428 13:45:20.067821  7476 net.cpp:122] Setting up relu5
I0428 13:45:20.067831  7476 net.cpp:129] Top shape: 32 256 4 4 (131072)
I0428 13:45:20.067835  7476 net.cpp:137] Memory required for data: 179162880
I0428 13:45:20.067839  7476 layer_factory.hpp:77] Creating layer pool5
I0428 13:45:20.067848  7476 net.cpp:84] Creating Layer pool5
I0428 13:45:20.067852  7476 net.cpp:406] pool5 <- conv5
I0428 13:45:20.067858  7476 net.cpp:380] pool5 -> pool5
I0428 13:45:20.067900  7476 net.cpp:122] Setting up pool5
I0428 13:45:20.067907  7476 net.cpp:129] Top shape: 32 256 2 2 (32768)
I0428 13:45:20.067911  7476 net.cpp:137] Memory required for data: 179293952
I0428 13:45:20.067914  7476 layer_factory.hpp:77] Creating layer fc6
I0428 13:45:20.067921  7476 net.cpp:84] Creating Layer fc6
I0428 13:45:20.067925  7476 net.cpp:406] fc6 <- pool5
I0428 13:45:20.067930  7476 net.cpp:380] fc6 -> fc6
I0428 13:45:20.109063  7476 net.cpp:122] Setting up fc6
I0428 13:45:20.109086  7476 net.cpp:129] Top shape: 32 4096 (131072)
I0428 13:45:20.109091  7476 net.cpp:137] Memory required for data: 179818240
I0428 13:45:20.109100  7476 layer_factory.hpp:77] Creating layer relu6
I0428 13:45:20.109112  7476 net.cpp:84] Creating Layer relu6
I0428 13:45:20.109117  7476 net.cpp:406] relu6 <- fc6
I0428 13:45:20.109123  7476 net.cpp:367] relu6 -> fc6 (in-place)
I0428 13:45:20.109798  7476 net.cpp:122] Setting up relu6
I0428 13:45:20.109809  7476 net.cpp:129] Top shape: 32 4096 (131072)
I0428 13:45:20.109813  7476 net.cpp:137] Memory required for data: 180342528
I0428 13:45:20.109817  7476 layer_factory.hpp:77] Creating layer drop6
I0428 13:45:20.109824  7476 net.cpp:84] Creating Layer drop6
I0428 13:45:20.109828  7476 net.cpp:406] drop6 <- fc6
I0428 13:45:20.109833  7476 net.cpp:367] drop6 -> fc6 (in-place)
I0428 13:45:20.109858  7476 net.cpp:122] Setting up drop6
I0428 13:45:20.109863  7476 net.cpp:129] Top shape: 32 4096 (131072)
I0428 13:45:20.109867  7476 net.cpp:137] Memory required for data: 180866816
I0428 13:45:20.109871  7476 layer_factory.hpp:77] Creating layer fc7
I0428 13:45:20.109879  7476 net.cpp:84] Creating Layer fc7
I0428 13:45:20.109884  7476 net.cpp:406] fc7 <- fc6
I0428 13:45:20.109889  7476 net.cpp:380] fc7 -> fc7
I0428 13:45:20.268368  7476 net.cpp:122] Setting up fc7
I0428 13:45:20.268388  7476 net.cpp:129] Top shape: 32 4096 (131072)
I0428 13:45:20.268391  7476 net.cpp:137] Memory required for data: 181391104
I0428 13:45:20.268400  7476 layer_factory.hpp:77] Creating layer relu7
I0428 13:45:20.268410  7476 net.cpp:84] Creating Layer relu7
I0428 13:45:20.268435  7476 net.cpp:406] relu7 <- fc7
I0428 13:45:20.268442  7476 net.cpp:367] relu7 -> fc7 (in-place)
I0428 13:45:20.268888  7476 net.cpp:122] Setting up relu7
I0428 13:45:20.268898  7476 net.cpp:129] Top shape: 32 4096 (131072)
I0428 13:45:20.268900  7476 net.cpp:137] Memory required for data: 181915392
I0428 13:45:20.268904  7476 layer_factory.hpp:77] Creating layer drop7
I0428 13:45:20.268910  7476 net.cpp:84] Creating Layer drop7
I0428 13:45:20.268920  7476 net.cpp:406] drop7 <- fc7
I0428 13:45:20.268925  7476 net.cpp:367] drop7 -> fc7 (in-place)
I0428 13:45:20.268949  7476 net.cpp:122] Setting up drop7
I0428 13:45:20.268954  7476 net.cpp:129] Top shape: 32 4096 (131072)
I0428 13:45:20.268957  7476 net.cpp:137] Memory required for data: 182439680
I0428 13:45:20.268960  7476 layer_factory.hpp:77] Creating layer fc8
I0428 13:45:20.268968  7476 net.cpp:84] Creating Layer fc8
I0428 13:45:20.268972  7476 net.cpp:406] fc8 <- fc7
I0428 13:45:20.268977  7476 net.cpp:380] fc8 -> fc8
I0428 13:45:20.280522  7476 net.cpp:122] Setting up fc8
I0428 13:45:20.280539  7476 net.cpp:129] Top shape: 32 196 (6272)
I0428 13:45:20.280541  7476 net.cpp:137] Memory required for data: 182464768
I0428 13:45:20.280555  7476 layer_factory.hpp:77] Creating layer fc8_fc8_0_split
I0428 13:45:20.280563  7476 net.cpp:84] Creating Layer fc8_fc8_0_split
I0428 13:45:20.280568  7476 net.cpp:406] fc8_fc8_0_split <- fc8
I0428 13:45:20.280575  7476 net.cpp:380] fc8_fc8_0_split -> fc8_fc8_0_split_0
I0428 13:45:20.280583  7476 net.cpp:380] fc8_fc8_0_split -> fc8_fc8_0_split_1
I0428 13:45:20.280622  7476 net.cpp:122] Setting up fc8_fc8_0_split
I0428 13:45:20.280627  7476 net.cpp:129] Top shape: 32 196 (6272)
I0428 13:45:20.280629  7476 net.cpp:129] Top shape: 32 196 (6272)
I0428 13:45:20.280632  7476 net.cpp:137] Memory required for data: 182514944
I0428 13:45:20.280635  7476 layer_factory.hpp:77] Creating layer accuracy
I0428 13:45:20.280642  7476 net.cpp:84] Creating Layer accuracy
I0428 13:45:20.280644  7476 net.cpp:406] accuracy <- fc8_fc8_0_split_0
I0428 13:45:20.280649  7476 net.cpp:406] accuracy <- label_val-data_1_split_0
I0428 13:45:20.280659  7476 net.cpp:380] accuracy -> accuracy
I0428 13:45:20.280666  7476 net.cpp:122] Setting up accuracy
I0428 13:45:20.280670  7476 net.cpp:129] Top shape: (1)
I0428 13:45:20.280673  7476 net.cpp:137] Memory required for data: 182514948
I0428 13:45:20.280676  7476 layer_factory.hpp:77] Creating layer loss
I0428 13:45:20.280683  7476 net.cpp:84] Creating Layer loss
I0428 13:45:20.280685  7476 net.cpp:406] loss <- fc8_fc8_0_split_1
I0428 13:45:20.280689  7476 net.cpp:406] loss <- label_val-data_1_split_1
I0428 13:45:20.280694  7476 net.cpp:380] loss -> loss
I0428 13:45:20.280701  7476 layer_factory.hpp:77] Creating layer loss
I0428 13:45:20.281657  7476 net.cpp:122] Setting up loss
I0428 13:45:20.281666  7476 net.cpp:129] Top shape: (1)
I0428 13:45:20.281669  7476 net.cpp:132]     with loss weight 1
I0428 13:45:20.281679  7476 net.cpp:137] Memory required for data: 182514952
I0428 13:45:20.281682  7476 net.cpp:198] loss needs backward computation.
I0428 13:45:20.281687  7476 net.cpp:200] accuracy does not need backward computation.
I0428 13:45:20.281692  7476 net.cpp:198] fc8_fc8_0_split needs backward computation.
I0428 13:45:20.281694  7476 net.cpp:198] fc8 needs backward computation.
I0428 13:45:20.281698  7476 net.cpp:198] drop7 needs backward computation.
I0428 13:45:20.281702  7476 net.cpp:198] relu7 needs backward computation.
I0428 13:45:20.281705  7476 net.cpp:198] fc7 needs backward computation.
I0428 13:45:20.281708  7476 net.cpp:198] drop6 needs backward computation.
I0428 13:45:20.281711  7476 net.cpp:198] relu6 needs backward computation.
I0428 13:45:20.281715  7476 net.cpp:198] fc6 needs backward computation.
I0428 13:45:20.281718  7476 net.cpp:198] pool5 needs backward computation.
I0428 13:45:20.281723  7476 net.cpp:198] relu5 needs backward computation.
I0428 13:45:20.281726  7476 net.cpp:198] conv5 needs backward computation.
I0428 13:45:20.281730  7476 net.cpp:198] relu4 needs backward computation.
I0428 13:45:20.281750  7476 net.cpp:198] conv4 needs backward computation.
I0428 13:45:20.281754  7476 net.cpp:198] relu3 needs backward computation.
I0428 13:45:20.281756  7476 net.cpp:198] conv3 needs backward computation.
I0428 13:45:20.281760  7476 net.cpp:198] pool2 needs backward computation.
I0428 13:45:20.281764  7476 net.cpp:198] norm2 needs backward computation.
I0428 13:45:20.281767  7476 net.cpp:198] relu2 needs backward computation.
I0428 13:45:20.281770  7476 net.cpp:198] conv2 needs backward computation.
I0428 13:45:20.281774  7476 net.cpp:198] pool1.5 needs backward computation.
I0428 13:45:20.281777  7476 net.cpp:198] norm1.5 needs backward computation.
I0428 13:45:20.281781  7476 net.cpp:198] relu1.5 needs backward computation.
I0428 13:45:20.281785  7476 net.cpp:198] conv1.5 needs backward computation.
I0428 13:45:20.281787  7476 net.cpp:198] pool1 needs backward computation.
I0428 13:45:20.281791  7476 net.cpp:198] norm1 needs backward computation.
I0428 13:45:20.281795  7476 net.cpp:198] relu1 needs backward computation.
I0428 13:45:20.281800  7476 net.cpp:198] conv1 needs backward computation.
I0428 13:45:20.281803  7476 net.cpp:200] label_val-data_1_split does not need backward computation.
I0428 13:45:20.281807  7476 net.cpp:200] val-data does not need backward computation.
I0428 13:45:20.281810  7476 net.cpp:242] This network produces output accuracy
I0428 13:45:20.281813  7476 net.cpp:242] This network produces output loss
I0428 13:45:20.281831  7476 net.cpp:255] Network initialization done.
I0428 13:45:20.281905  7476 solver.cpp:56] Solver scaffolding done.
I0428 13:45:20.282399  7476 caffe.cpp:248] Starting Optimization
I0428 13:45:20.282408  7476 solver.cpp:272] Solving
I0428 13:45:20.282411  7476 solver.cpp:273] Learning Rate Policy: exp
I0428 13:45:20.304643  7476 solver.cpp:330] Iteration 0, Testing net (#0)
I0428 13:45:20.304662  7476 net.cpp:676] Ignoring source layer train-data
I0428 13:45:20.375517  7476 blocking_queue.cpp:49] Waiting for data
I0428 13:45:24.611348  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 13:45:24.657397  7476 solver.cpp:397]     Test net output #0: accuracy = 0.00490196
I0428 13:45:24.657428  7476 solver.cpp:397]     Test net output #1: loss = 5.2817 (* 1 = 5.2817 loss)
I0428 13:45:24.835229  7476 solver.cpp:218] Iteration 0 (1.55617e+21 iter/s, 4.55265s/12 iters), loss = 5.268
I0428 13:45:24.836833  7476 solver.cpp:237]     Train net output #0: loss = 5.268 (* 1 = 5.268 loss)
I0428 13:45:24.836858  7476 sgd_solver.cpp:105] Iteration 0, lr = 0.01
I0428 13:45:29.044793  7476 solver.cpp:218] Iteration 12 (2.85182 iter/s, 4.20783s/12 iters), loss = 5.26913
I0428 13:45:29.044836  7476 solver.cpp:237]     Train net output #0: loss = 5.26913 (* 1 = 5.26913 loss)
I0428 13:45:29.044844  7476 sgd_solver.cpp:105] Iteration 12, lr = 0.00997626
I0428 13:45:34.361970  7476 solver.cpp:218] Iteration 24 (2.25692 iter/s, 5.31698s/12 iters), loss = 5.27485
I0428 13:45:34.362011  7476 solver.cpp:237]     Train net output #0: loss = 5.27485 (* 1 = 5.27485 loss)
I0428 13:45:34.362023  7476 sgd_solver.cpp:105] Iteration 24, lr = 0.00995257
I0428 13:45:39.890369  7476 solver.cpp:218] Iteration 36 (2.17069 iter/s, 5.52819s/12 iters), loss = 5.31243
I0428 13:45:39.890413  7476 solver.cpp:237]     Train net output #0: loss = 5.31243 (* 1 = 5.31243 loss)
I0428 13:45:39.890420  7476 sgd_solver.cpp:105] Iteration 36, lr = 0.00992894
I0428 13:45:45.400569  7476 solver.cpp:218] Iteration 48 (2.17786 iter/s, 5.50999s/12 iters), loss = 5.27071
I0428 13:45:45.400609  7476 solver.cpp:237]     Train net output #0: loss = 5.27071 (* 1 = 5.27071 loss)
I0428 13:45:45.400617  7476 sgd_solver.cpp:105] Iteration 48, lr = 0.00990537
I0428 13:45:50.848086  7476 solver.cpp:218] Iteration 60 (2.20292 iter/s, 5.44731s/12 iters), loss = 5.30013
I0428 13:45:50.848248  7476 solver.cpp:237]     Train net output #0: loss = 5.30013 (* 1 = 5.30013 loss)
I0428 13:45:50.848261  7476 sgd_solver.cpp:105] Iteration 60, lr = 0.00988185
I0428 13:45:56.321079  7476 solver.cpp:218] Iteration 72 (2.19271 iter/s, 5.47268s/12 iters), loss = 5.28736
I0428 13:45:56.321125  7476 solver.cpp:237]     Train net output #0: loss = 5.28736 (* 1 = 5.28736 loss)
I0428 13:45:56.321133  7476 sgd_solver.cpp:105] Iteration 72, lr = 0.00985839
I0428 13:46:01.760422  7476 solver.cpp:218] Iteration 84 (2.20623 iter/s, 5.43913s/12 iters), loss = 5.29303
I0428 13:46:01.760476  7476 solver.cpp:237]     Train net output #0: loss = 5.29303 (* 1 = 5.29303 loss)
I0428 13:46:01.760511  7476 sgd_solver.cpp:105] Iteration 84, lr = 0.00983498
I0428 13:46:07.306560  7476 solver.cpp:218] Iteration 96 (2.16375 iter/s, 5.54593s/12 iters), loss = 5.29359
I0428 13:46:07.306592  7476 solver.cpp:237]     Train net output #0: loss = 5.29359 (* 1 = 5.29359 loss)
I0428 13:46:07.306599  7476 sgd_solver.cpp:105] Iteration 96, lr = 0.00981163
I0428 13:46:09.233454  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 13:46:09.572295  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_102.caffemodel
I0428 13:46:18.811486  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_102.solverstate
I0428 13:46:26.522773  7476 solver.cpp:330] Iteration 102, Testing net (#0)
I0428 13:46:26.522841  7476 net.cpp:676] Ignoring source layer train-data
I0428 13:46:31.047114  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 13:46:31.134656  7476 solver.cpp:397]     Test net output #0: accuracy = 0.00245098
I0428 13:46:31.134685  7476 solver.cpp:397]     Test net output #1: loss = 5.29011 (* 1 = 5.29011 loss)
I0428 13:46:33.044806  7476 solver.cpp:218] Iteration 108 (0.466246 iter/s, 25.7375s/12 iters), loss = 5.2767
I0428 13:46:33.044847  7476 solver.cpp:237]     Train net output #0: loss = 5.2767 (* 1 = 5.2767 loss)
I0428 13:46:33.044855  7476 sgd_solver.cpp:105] Iteration 108, lr = 0.00978834
I0428 13:46:38.532785  7476 solver.cpp:218] Iteration 120 (2.18668 iter/s, 5.48777s/12 iters), loss = 5.28734
I0428 13:46:38.532845  7476 solver.cpp:237]     Train net output #0: loss = 5.28734 (* 1 = 5.28734 loss)
I0428 13:46:38.532855  7476 sgd_solver.cpp:105] Iteration 120, lr = 0.0097651
I0428 13:46:43.818711  7476 solver.cpp:218] Iteration 132 (2.27027 iter/s, 5.28571s/12 iters), loss = 5.27625
I0428 13:46:43.818749  7476 solver.cpp:237]     Train net output #0: loss = 5.27625 (* 1 = 5.27625 loss)
I0428 13:46:43.818758  7476 sgd_solver.cpp:105] Iteration 132, lr = 0.00974192
I0428 13:46:49.324483  7476 solver.cpp:218] Iteration 144 (2.17961 iter/s, 5.50556s/12 iters), loss = 5.29493
I0428 13:46:49.324568  7476 solver.cpp:237]     Train net output #0: loss = 5.29493 (* 1 = 5.29493 loss)
I0428 13:46:49.324580  7476 sgd_solver.cpp:105] Iteration 144, lr = 0.00971879
I0428 13:46:54.677338  7476 solver.cpp:218] Iteration 156 (2.2419 iter/s, 5.35261s/12 iters), loss = 5.29937
I0428 13:46:54.677383  7476 solver.cpp:237]     Train net output #0: loss = 5.29937 (* 1 = 5.29937 loss)
I0428 13:46:54.677392  7476 sgd_solver.cpp:105] Iteration 156, lr = 0.00969571
I0428 13:47:00.346249  7476 solver.cpp:218] Iteration 168 (2.11689 iter/s, 5.66869s/12 iters), loss = 5.29131
I0428 13:47:00.346355  7476 solver.cpp:237]     Train net output #0: loss = 5.29131 (* 1 = 5.29131 loss)
I0428 13:47:00.346369  7476 sgd_solver.cpp:105] Iteration 168, lr = 0.00967269
I0428 13:47:05.819779  7476 solver.cpp:218] Iteration 180 (2.19248 iter/s, 5.47327s/12 iters), loss = 5.29943
I0428 13:47:05.819824  7476 solver.cpp:237]     Train net output #0: loss = 5.29943 (* 1 = 5.29943 loss)
I0428 13:47:05.819833  7476 sgd_solver.cpp:105] Iteration 180, lr = 0.00964973
I0428 13:47:11.282382  7476 solver.cpp:218] Iteration 192 (2.19684 iter/s, 5.4624s/12 iters), loss = 5.27657
I0428 13:47:11.282419  7476 solver.cpp:237]     Train net output #0: loss = 5.27657 (* 1 = 5.27657 loss)
I0428 13:47:11.282428  7476 sgd_solver.cpp:105] Iteration 192, lr = 0.00962682
I0428 13:47:15.475569  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 13:47:16.191301  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_204.caffemodel
I0428 13:47:20.985994  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_204.solverstate
I0428 13:47:27.003470  7476 solver.cpp:330] Iteration 204, Testing net (#0)
I0428 13:47:27.003492  7476 net.cpp:676] Ignoring source layer train-data
I0428 13:47:31.267009  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 13:47:31.395162  7476 solver.cpp:397]     Test net output #0: accuracy = 0.00551471
I0428 13:47:31.395190  7476 solver.cpp:397]     Test net output #1: loss = 5.28579 (* 1 = 5.28579 loss)
I0428 13:47:31.523402  7476 solver.cpp:218] Iteration 204 (0.592873 iter/s, 20.2404s/12 iters), loss = 5.26564
I0428 13:47:31.523439  7476 solver.cpp:237]     Train net output #0: loss = 5.26564 (* 1 = 5.26564 loss)
I0428 13:47:31.523447  7476 sgd_solver.cpp:105] Iteration 204, lr = 0.00960396
I0428 13:47:36.024463  7476 solver.cpp:218] Iteration 216 (2.66614 iter/s, 4.50089s/12 iters), loss = 5.30185
I0428 13:47:36.024516  7476 solver.cpp:237]     Train net output #0: loss = 5.30185 (* 1 = 5.30185 loss)
I0428 13:47:36.024525  7476 sgd_solver.cpp:105] Iteration 216, lr = 0.00958116
I0428 13:47:41.640889  7476 solver.cpp:218] Iteration 228 (2.13667 iter/s, 5.6162s/12 iters), loss = 5.27519
I0428 13:47:41.640939  7476 solver.cpp:237]     Train net output #0: loss = 5.27519 (* 1 = 5.27519 loss)
I0428 13:47:41.640952  7476 sgd_solver.cpp:105] Iteration 228, lr = 0.00955841
I0428 13:47:47.137745  7476 solver.cpp:218] Iteration 240 (2.18315 iter/s, 5.49664s/12 iters), loss = 5.26937
I0428 13:47:47.137785  7476 solver.cpp:237]     Train net output #0: loss = 5.26937 (* 1 = 5.26937 loss)
I0428 13:47:47.137794  7476 sgd_solver.cpp:105] Iteration 240, lr = 0.00953572
I0428 13:47:52.708509  7476 solver.cpp:218] Iteration 252 (2.15419 iter/s, 5.57054s/12 iters), loss = 5.27945
I0428 13:47:52.708555  7476 solver.cpp:237]     Train net output #0: loss = 5.27945 (* 1 = 5.27945 loss)
I0428 13:47:52.708564  7476 sgd_solver.cpp:105] Iteration 252, lr = 0.00951308
I0428 13:47:58.128072  7476 solver.cpp:218] Iteration 264 (2.21429 iter/s, 5.41935s/12 iters), loss = 5.26455
I0428 13:47:58.128115  7476 solver.cpp:237]     Train net output #0: loss = 5.26455 (* 1 = 5.26455 loss)
I0428 13:47:58.128124  7476 sgd_solver.cpp:105] Iteration 264, lr = 0.00949049
I0428 13:48:03.584154  7476 solver.cpp:218] Iteration 276 (2.19946 iter/s, 5.45587s/12 iters), loss = 5.27036
I0428 13:48:03.584261  7476 solver.cpp:237]     Train net output #0: loss = 5.27036 (* 1 = 5.27036 loss)
I0428 13:48:03.584273  7476 sgd_solver.cpp:105] Iteration 276, lr = 0.00946796
I0428 13:48:09.014789  7476 solver.cpp:218] Iteration 288 (2.2098 iter/s, 5.43037s/12 iters), loss = 5.29918
I0428 13:48:09.014833  7476 solver.cpp:237]     Train net output #0: loss = 5.29918 (* 1 = 5.29918 loss)
I0428 13:48:09.014842  7476 sgd_solver.cpp:105] Iteration 288, lr = 0.00944548
I0428 13:48:14.536631  7476 solver.cpp:218] Iteration 300 (2.17327 iter/s, 5.52163s/12 iters), loss = 5.27747
I0428 13:48:14.536676  7476 solver.cpp:237]     Train net output #0: loss = 5.27747 (* 1 = 5.27747 loss)
I0428 13:48:14.536685  7476 sgd_solver.cpp:105] Iteration 300, lr = 0.00942305
I0428 13:48:15.593601  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 13:48:16.723204  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_306.caffemodel
I0428 13:48:18.700567  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_306.solverstate
I0428 13:48:21.512859  7476 solver.cpp:330] Iteration 306, Testing net (#0)
I0428 13:48:21.512881  7476 net.cpp:676] Ignoring source layer train-data
I0428 13:48:25.768172  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 13:48:25.933941  7476 solver.cpp:397]     Test net output #0: accuracy = 0.00919118
I0428 13:48:25.933974  7476 solver.cpp:397]     Test net output #1: loss = 5.20036 (* 1 = 5.20036 loss)
I0428 13:48:27.834321  7476 solver.cpp:218] Iteration 312 (0.902441 iter/s, 13.2973s/12 iters), loss = 5.24452
I0428 13:48:27.834367  7476 solver.cpp:237]     Train net output #0: loss = 5.24452 (* 1 = 5.24452 loss)
I0428 13:48:27.834374  7476 sgd_solver.cpp:105] Iteration 312, lr = 0.00940068
I0428 13:48:33.329648  7476 solver.cpp:218] Iteration 324 (2.18376 iter/s, 5.49511s/12 iters), loss = 5.23672
I0428 13:48:33.329692  7476 solver.cpp:237]     Train net output #0: loss = 5.23672 (* 1 = 5.23672 loss)
I0428 13:48:33.329702  7476 sgd_solver.cpp:105] Iteration 324, lr = 0.00937836
I0428 13:48:38.901677  7476 solver.cpp:218] Iteration 336 (2.15369 iter/s, 5.57182s/12 iters), loss = 5.24073
I0428 13:48:38.901798  7476 solver.cpp:237]     Train net output #0: loss = 5.24073 (* 1 = 5.24073 loss)
I0428 13:48:38.901808  7476 sgd_solver.cpp:105] Iteration 336, lr = 0.0093561
I0428 13:48:44.419163  7476 solver.cpp:218] Iteration 348 (2.17501 iter/s, 5.51721s/12 iters), loss = 5.19246
I0428 13:48:44.419203  7476 solver.cpp:237]     Train net output #0: loss = 5.19246 (* 1 = 5.19246 loss)
I0428 13:48:44.419210  7476 sgd_solver.cpp:105] Iteration 348, lr = 0.00933388
I0428 13:48:49.857187  7476 solver.cpp:218] Iteration 360 (2.20677 iter/s, 5.43781s/12 iters), loss = 5.18458
I0428 13:48:49.857229  7476 solver.cpp:237]     Train net output #0: loss = 5.18458 (* 1 = 5.18458 loss)
I0428 13:48:49.857237  7476 sgd_solver.cpp:105] Iteration 360, lr = 0.00931172
I0428 13:48:55.332239  7476 solver.cpp:218] Iteration 372 (2.19184 iter/s, 5.47485s/12 iters), loss = 5.20041
I0428 13:48:55.332281  7476 solver.cpp:237]     Train net output #0: loss = 5.20041 (* 1 = 5.20041 loss)
I0428 13:48:55.332291  7476 sgd_solver.cpp:105] Iteration 372, lr = 0.00928961
I0428 13:49:00.727108  7476 solver.cpp:218] Iteration 384 (2.22442 iter/s, 5.39466s/12 iters), loss = 5.18071
I0428 13:49:00.727154  7476 solver.cpp:237]     Train net output #0: loss = 5.18071 (* 1 = 5.18071 loss)
I0428 13:49:00.727164  7476 sgd_solver.cpp:105] Iteration 384, lr = 0.00926756
I0428 13:49:06.038003  7476 solver.cpp:218] Iteration 396 (2.25959 iter/s, 5.31069s/12 iters), loss = 5.15011
I0428 13:49:06.038045  7476 solver.cpp:237]     Train net output #0: loss = 5.15011 (* 1 = 5.15011 loss)
I0428 13:49:06.038053  7476 sgd_solver.cpp:105] Iteration 396, lr = 0.00924556
I0428 13:49:09.496770  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 13:49:11.031702  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_408.caffemodel
I0428 13:49:15.003371  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_408.solverstate
I0428 13:49:17.017483  7476 solver.cpp:330] Iteration 408, Testing net (#0)
I0428 13:49:17.017504  7476 net.cpp:676] Ignoring source layer train-data
I0428 13:49:21.356339  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 13:49:21.565369  7476 solver.cpp:397]     Test net output #0: accuracy = 0.0067402
I0428 13:49:21.565415  7476 solver.cpp:397]     Test net output #1: loss = 5.15047 (* 1 = 5.15047 loss)
I0428 13:49:21.693892  7476 solver.cpp:218] Iteration 408 (0.766509 iter/s, 15.6554s/12 iters), loss = 5.21381
I0428 13:49:21.693944  7476 solver.cpp:237]     Train net output #0: loss = 5.21381 (* 1 = 5.21381 loss)
I0428 13:49:21.693958  7476 sgd_solver.cpp:105] Iteration 408, lr = 0.00922361
I0428 13:49:26.313083  7476 solver.cpp:218] Iteration 420 (2.59797 iter/s, 4.61899s/12 iters), loss = 5.03452
I0428 13:49:26.313133  7476 solver.cpp:237]     Train net output #0: loss = 5.03452 (* 1 = 5.03452 loss)
I0428 13:49:26.313144  7476 sgd_solver.cpp:105] Iteration 420, lr = 0.00920171
I0428 13:49:31.839879  7476 solver.cpp:218] Iteration 432 (2.17133 iter/s, 5.52658s/12 iters), loss = 4.99839
I0428 13:49:31.839918  7476 solver.cpp:237]     Train net output #0: loss = 4.99839 (* 1 = 4.99839 loss)
I0428 13:49:31.839927  7476 sgd_solver.cpp:105] Iteration 432, lr = 0.00917986
I0428 13:49:37.465984  7476 solver.cpp:218] Iteration 444 (2.133 iter/s, 5.62589s/12 iters), loss = 5.16532
I0428 13:49:37.466029  7476 solver.cpp:237]     Train net output #0: loss = 5.16532 (* 1 = 5.16532 loss)
I0428 13:49:37.466038  7476 sgd_solver.cpp:105] Iteration 444, lr = 0.00915807
I0428 13:49:42.993062  7476 solver.cpp:218] Iteration 456 (2.17121 iter/s, 5.52687s/12 iters), loss = 5.12939
I0428 13:49:42.993181  7476 solver.cpp:237]     Train net output #0: loss = 5.12939 (* 1 = 5.12939 loss)
I0428 13:49:42.993191  7476 sgd_solver.cpp:105] Iteration 456, lr = 0.00913632
I0428 13:49:48.233997  7476 solver.cpp:218] Iteration 468 (2.28979 iter/s, 5.24066s/12 iters), loss = 5.12796
I0428 13:49:48.234040  7476 solver.cpp:237]     Train net output #0: loss = 5.12796 (* 1 = 5.12796 loss)
I0428 13:49:48.234047  7476 sgd_solver.cpp:105] Iteration 468, lr = 0.00911463
I0428 13:49:53.773135  7476 solver.cpp:218] Iteration 480 (2.16648 iter/s, 5.53893s/12 iters), loss = 5.05032
I0428 13:49:53.773171  7476 solver.cpp:237]     Train net output #0: loss = 5.05032 (* 1 = 5.05032 loss)
I0428 13:49:53.773180  7476 sgd_solver.cpp:105] Iteration 480, lr = 0.00909299
I0428 13:49:59.132861  7476 solver.cpp:218] Iteration 492 (2.23901 iter/s, 5.35952s/12 iters), loss = 5.16911
I0428 13:49:59.132913  7476 solver.cpp:237]     Train net output #0: loss = 5.16911 (* 1 = 5.16911 loss)
I0428 13:49:59.132925  7476 sgd_solver.cpp:105] Iteration 492, lr = 0.0090714
I0428 13:50:04.543133  7476 solver.cpp:218] Iteration 504 (2.21809 iter/s, 5.41005s/12 iters), loss = 5.1083
I0428 13:50:04.543179  7476 solver.cpp:237]     Train net output #0: loss = 5.1083 (* 1 = 5.1083 loss)
I0428 13:50:04.543187  7476 sgd_solver.cpp:105] Iteration 504, lr = 0.00904986
I0428 13:50:04.798246  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 13:50:06.789896  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_510.caffemodel
I0428 13:50:15.438181  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_510.solverstate
I0428 13:50:28.124894  7476 solver.cpp:330] Iteration 510, Testing net (#0)
I0428 13:50:28.124915  7476 net.cpp:676] Ignoring source layer train-data
I0428 13:50:32.297575  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 13:50:32.542649  7476 solver.cpp:397]     Test net output #0: accuracy = 0.0110294
I0428 13:50:32.542713  7476 solver.cpp:397]     Test net output #1: loss = 5.09457 (* 1 = 5.09457 loss)
I0428 13:50:34.555416  7476 solver.cpp:218] Iteration 516 (0.399848 iter/s, 30.0114s/12 iters), loss = 5.00004
I0428 13:50:34.555474  7476 solver.cpp:237]     Train net output #0: loss = 5.00004 (* 1 = 5.00004 loss)
I0428 13:50:34.555486  7476 sgd_solver.cpp:105] Iteration 516, lr = 0.00902838
I0428 13:50:39.970556  7476 solver.cpp:218] Iteration 528 (2.2161 iter/s, 5.41492s/12 iters), loss = 5.07837
I0428 13:50:39.970589  7476 solver.cpp:237]     Train net output #0: loss = 5.07837 (* 1 = 5.07837 loss)
I0428 13:50:39.970598  7476 sgd_solver.cpp:105] Iteration 528, lr = 0.00900694
I0428 13:50:45.384445  7476 solver.cpp:218] Iteration 540 (2.21661 iter/s, 5.41368s/12 iters), loss = 4.98665
I0428 13:50:45.384523  7476 solver.cpp:237]     Train net output #0: loss = 4.98665 (* 1 = 4.98665 loss)
I0428 13:50:45.384538  7476 sgd_solver.cpp:105] Iteration 540, lr = 0.00898556
I0428 13:50:50.781721  7476 solver.cpp:218] Iteration 552 (2.22344 iter/s, 5.39703s/12 iters), loss = 5.13913
I0428 13:50:50.781857  7476 solver.cpp:237]     Train net output #0: loss = 5.13913 (* 1 = 5.13913 loss)
I0428 13:50:50.781872  7476 sgd_solver.cpp:105] Iteration 552, lr = 0.00896423
I0428 13:50:56.223610  7476 solver.cpp:218] Iteration 564 (2.20524 iter/s, 5.44158s/12 iters), loss = 5.14511
I0428 13:50:56.223666  7476 solver.cpp:237]     Train net output #0: loss = 5.14511 (* 1 = 5.14511 loss)
I0428 13:50:56.223676  7476 sgd_solver.cpp:105] Iteration 564, lr = 0.00894294
I0428 13:51:01.611510  7476 solver.cpp:218] Iteration 576 (2.22731 iter/s, 5.38768s/12 iters), loss = 5.06711
I0428 13:51:01.611553  7476 solver.cpp:237]     Train net output #0: loss = 5.06711 (* 1 = 5.06711 loss)
I0428 13:51:01.611562  7476 sgd_solver.cpp:105] Iteration 576, lr = 0.00892171
I0428 13:51:07.088407  7476 solver.cpp:218] Iteration 588 (2.1911 iter/s, 5.47669s/12 iters), loss = 5.07806
I0428 13:51:07.088447  7476 solver.cpp:237]     Train net output #0: loss = 5.07806 (* 1 = 5.07806 loss)
I0428 13:51:07.088457  7476 sgd_solver.cpp:105] Iteration 588, lr = 0.00890053
I0428 13:51:12.581324  7476 solver.cpp:218] Iteration 600 (2.18471 iter/s, 5.49271s/12 iters), loss = 5.09201
I0428 13:51:12.581367  7476 solver.cpp:237]     Train net output #0: loss = 5.09201 (* 1 = 5.09201 loss)
I0428 13:51:12.581377  7476 sgd_solver.cpp:105] Iteration 600, lr = 0.0088794
I0428 13:51:15.117086  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 13:51:17.443240  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_612.caffemodel
I0428 13:51:28.713251  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_612.solverstate
I0428 13:51:35.825423  7476 solver.cpp:330] Iteration 612, Testing net (#0)
I0428 13:51:35.825443  7476 net.cpp:676] Ignoring source layer train-data
I0428 13:51:39.992559  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 13:51:40.311893  7476 solver.cpp:397]     Test net output #0: accuracy = 0.0104167
I0428 13:51:40.311939  7476 solver.cpp:397]     Test net output #1: loss = 5.07039 (* 1 = 5.07039 loss)
I0428 13:51:40.440723  7476 solver.cpp:218] Iteration 612 (0.430747 iter/s, 27.8586s/12 iters), loss = 5.03155
I0428 13:51:40.440790  7476 solver.cpp:237]     Train net output #0: loss = 5.03155 (* 1 = 5.03155 loss)
I0428 13:51:40.440801  7476 sgd_solver.cpp:105] Iteration 612, lr = 0.00885831
I0428 13:51:45.064525  7476 solver.cpp:218] Iteration 624 (2.59539 iter/s, 4.62358s/12 iters), loss = 5.16649
I0428 13:51:45.064570  7476 solver.cpp:237]     Train net output #0: loss = 5.16649 (* 1 = 5.16649 loss)
I0428 13:51:45.064579  7476 sgd_solver.cpp:105] Iteration 624, lr = 0.00883728
I0428 13:51:50.673050  7476 solver.cpp:218] Iteration 636 (2.13968 iter/s, 5.60831s/12 iters), loss = 5.13071
I0428 13:51:50.673095  7476 solver.cpp:237]     Train net output #0: loss = 5.13071 (* 1 = 5.13071 loss)
I0428 13:51:50.673103  7476 sgd_solver.cpp:105] Iteration 636, lr = 0.0088163
I0428 13:51:56.022979  7476 solver.cpp:218] Iteration 648 (2.24311 iter/s, 5.34972s/12 iters), loss = 5.09121
I0428 13:51:56.023020  7476 solver.cpp:237]     Train net output #0: loss = 5.09121 (* 1 = 5.09121 loss)
I0428 13:51:56.023031  7476 sgd_solver.cpp:105] Iteration 648, lr = 0.00879537
I0428 13:52:01.489370  7476 solver.cpp:218] Iteration 660 (2.19532 iter/s, 5.46617s/12 iters), loss = 5.09963
I0428 13:52:01.489503  7476 solver.cpp:237]     Train net output #0: loss = 5.09963 (* 1 = 5.09963 loss)
I0428 13:52:01.489517  7476 sgd_solver.cpp:105] Iteration 660, lr = 0.00877449
I0428 13:52:06.998363  7476 solver.cpp:218] Iteration 672 (2.17837 iter/s, 5.5087s/12 iters), loss = 5.04408
I0428 13:52:06.998407  7476 solver.cpp:237]     Train net output #0: loss = 5.04408 (* 1 = 5.04408 loss)
I0428 13:52:06.998415  7476 sgd_solver.cpp:105] Iteration 672, lr = 0.00875366
I0428 13:52:12.590306  7476 solver.cpp:218] Iteration 684 (2.14603 iter/s, 5.59173s/12 iters), loss = 4.96204
I0428 13:52:12.590349  7476 solver.cpp:237]     Train net output #0: loss = 4.96204 (* 1 = 4.96204 loss)
I0428 13:52:12.590358  7476 sgd_solver.cpp:105] Iteration 684, lr = 0.00873287
I0428 13:52:13.416563  7476 blocking_queue.cpp:49] Waiting for data
I0428 13:52:18.166393  7476 solver.cpp:218] Iteration 696 (2.15213 iter/s, 5.57587s/12 iters), loss = 5.10696
I0428 13:52:18.166435  7476 solver.cpp:237]     Train net output #0: loss = 5.10696 (* 1 = 5.10696 loss)
I0428 13:52:18.166445  7476 sgd_solver.cpp:105] Iteration 696, lr = 0.00871214
I0428 13:52:23.227128  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 13:52:23.649524  7476 solver.cpp:218] Iteration 708 (2.18861 iter/s, 5.48292s/12 iters), loss = 5.00308
I0428 13:52:23.649569  7476 solver.cpp:237]     Train net output #0: loss = 5.00308 (* 1 = 5.00308 loss)
I0428 13:52:23.649580  7476 sgd_solver.cpp:105] Iteration 708, lr = 0.00869145
I0428 13:52:25.837158  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_714.caffemodel
I0428 13:52:27.193058  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_714.solverstate
I0428 13:52:29.119470  7476 solver.cpp:330] Iteration 714, Testing net (#0)
I0428 13:52:29.119498  7476 net.cpp:676] Ignoring source layer train-data
I0428 13:52:33.260145  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 13:52:33.613164  7476 solver.cpp:397]     Test net output #0: accuracy = 0.00919118
I0428 13:52:33.613202  7476 solver.cpp:397]     Test net output #1: loss = 5.07428 (* 1 = 5.07428 loss)
I0428 13:52:35.519147  7476 solver.cpp:218] Iteration 720 (1.01102 iter/s, 11.8692s/12 iters), loss = 5.01957
I0428 13:52:35.519199  7476 solver.cpp:237]     Train net output #0: loss = 5.01957 (* 1 = 5.01957 loss)
I0428 13:52:35.519210  7476 sgd_solver.cpp:105] Iteration 720, lr = 0.00867082
I0428 13:52:40.774505  7476 solver.cpp:218] Iteration 732 (2.28348 iter/s, 5.25514s/12 iters), loss = 5.04602
I0428 13:52:40.774555  7476 solver.cpp:237]     Train net output #0: loss = 5.04602 (* 1 = 5.04602 loss)
I0428 13:52:40.774565  7476 sgd_solver.cpp:105] Iteration 732, lr = 0.00865023
I0428 13:52:46.225575  7476 solver.cpp:218] Iteration 744 (2.20149 iter/s, 5.45085s/12 iters), loss = 4.98638
I0428 13:52:46.225622  7476 solver.cpp:237]     Train net output #0: loss = 4.98638 (* 1 = 4.98638 loss)
I0428 13:52:46.225630  7476 sgd_solver.cpp:105] Iteration 744, lr = 0.0086297
I0428 13:52:51.917245  7476 solver.cpp:218] Iteration 756 (2.10843 iter/s, 5.69145s/12 iters), loss = 4.92112
I0428 13:52:51.917284  7476 solver.cpp:237]     Train net output #0: loss = 4.92112 (* 1 = 4.92112 loss)
I0428 13:52:51.917294  7476 sgd_solver.cpp:105] Iteration 756, lr = 0.00860921
I0428 13:52:57.244771  7476 solver.cpp:218] Iteration 768 (2.25254 iter/s, 5.32732s/12 iters), loss = 5.05638
I0428 13:52:57.244813  7476 solver.cpp:237]     Train net output #0: loss = 5.05638 (* 1 = 5.05638 loss)
I0428 13:52:57.244822  7476 sgd_solver.cpp:105] Iteration 768, lr = 0.00858877
I0428 13:53:02.684273  7476 solver.cpp:218] Iteration 780 (2.20617 iter/s, 5.43929s/12 iters), loss = 4.92169
I0428 13:53:02.684314  7476 solver.cpp:237]     Train net output #0: loss = 4.92169 (* 1 = 4.92169 loss)
I0428 13:53:02.684322  7476 sgd_solver.cpp:105] Iteration 780, lr = 0.00856838
I0428 13:53:08.159471  7476 solver.cpp:218] Iteration 792 (2.19179 iter/s, 5.47499s/12 iters), loss = 4.93255
I0428 13:53:08.159585  7476 solver.cpp:237]     Train net output #0: loss = 4.93255 (* 1 = 4.93255 loss)
I0428 13:53:08.159595  7476 sgd_solver.cpp:105] Iteration 792, lr = 0.00854803
I0428 13:53:13.640136  7476 solver.cpp:218] Iteration 804 (2.18963 iter/s, 5.48038s/12 iters), loss = 5.0118
I0428 13:53:13.640182  7476 solver.cpp:237]     Train net output #0: loss = 5.0118 (* 1 = 5.0118 loss)
I0428 13:53:13.640192  7476 sgd_solver.cpp:105] Iteration 804, lr = 0.00852774
I0428 13:53:15.540458  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 13:53:18.627758  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_816.caffemodel
I0428 13:53:23.028375  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_816.solverstate
I0428 13:53:24.261202  7476 solver.cpp:330] Iteration 816, Testing net (#0)
I0428 13:53:24.261220  7476 net.cpp:676] Ignoring source layer train-data
I0428 13:53:28.510165  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 13:53:28.875213  7476 solver.cpp:397]     Test net output #0: accuracy = 0.0189951
I0428 13:53:28.875252  7476 solver.cpp:397]     Test net output #1: loss = 5.01875 (* 1 = 5.01875 loss)
I0428 13:53:29.004933  7476 solver.cpp:218] Iteration 816 (0.781031 iter/s, 15.3643s/12 iters), loss = 4.92443
I0428 13:53:29.004976  7476 solver.cpp:237]     Train net output #0: loss = 4.92443 (* 1 = 4.92443 loss)
I0428 13:53:29.004985  7476 sgd_solver.cpp:105] Iteration 816, lr = 0.00850749
I0428 13:53:33.840602  7476 solver.cpp:218] Iteration 828 (2.48166 iter/s, 4.83547s/12 iters), loss = 5.01235
I0428 13:53:33.840659  7476 solver.cpp:237]     Train net output #0: loss = 5.01235 (* 1 = 5.01235 loss)
I0428 13:53:33.840672  7476 sgd_solver.cpp:105] Iteration 828, lr = 0.00848729
I0428 13:53:39.208845  7476 solver.cpp:218] Iteration 840 (2.23546 iter/s, 5.36802s/12 iters), loss = 4.90612
I0428 13:53:39.209010  7476 solver.cpp:237]     Train net output #0: loss = 4.90612 (* 1 = 4.90612 loss)
I0428 13:53:39.209024  7476 sgd_solver.cpp:105] Iteration 840, lr = 0.00846714
I0428 13:53:44.775120  7476 solver.cpp:218] Iteration 852 (2.15597 iter/s, 5.56594s/12 iters), loss = 4.91975
I0428 13:53:44.775161  7476 solver.cpp:237]     Train net output #0: loss = 4.91975 (* 1 = 4.91975 loss)
I0428 13:53:44.775171  7476 sgd_solver.cpp:105] Iteration 852, lr = 0.00844704
I0428 13:53:50.332882  7476 solver.cpp:218] Iteration 864 (2.15923 iter/s, 5.55754s/12 iters), loss = 5.00278
I0428 13:53:50.332921  7476 solver.cpp:237]     Train net output #0: loss = 5.00278 (* 1 = 5.00278 loss)
I0428 13:53:50.332931  7476 sgd_solver.cpp:105] Iteration 864, lr = 0.00842698
I0428 13:53:55.758000  7476 solver.cpp:218] Iteration 876 (2.21202 iter/s, 5.42491s/12 iters), loss = 4.87842
I0428 13:53:55.758040  7476 solver.cpp:237]     Train net output #0: loss = 4.87842 (* 1 = 4.87842 loss)
I0428 13:53:55.758049  7476 sgd_solver.cpp:105] Iteration 876, lr = 0.00840698
I0428 13:54:01.262661  7476 solver.cpp:218] Iteration 888 (2.18006 iter/s, 5.50445s/12 iters), loss = 4.8458
I0428 13:54:01.262710  7476 solver.cpp:237]     Train net output #0: loss = 4.8458 (* 1 = 4.8458 loss)
I0428 13:54:01.262719  7476 sgd_solver.cpp:105] Iteration 888, lr = 0.00838702
I0428 13:54:06.645634  7476 solver.cpp:218] Iteration 900 (2.22934 iter/s, 5.38276s/12 iters), loss = 4.93698
I0428 13:54:06.645675  7476 solver.cpp:237]     Train net output #0: loss = 4.93698 (* 1 = 4.93698 loss)
I0428 13:54:06.645684  7476 sgd_solver.cpp:105] Iteration 900, lr = 0.0083671
I0428 13:54:10.883014  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 13:54:12.174057  7476 solver.cpp:218] Iteration 912 (2.17068 iter/s, 5.52821s/12 iters), loss = 4.85631
I0428 13:54:12.174098  7476 solver.cpp:237]     Train net output #0: loss = 4.85631 (* 1 = 4.85631 loss)
I0428 13:54:12.174105  7476 sgd_solver.cpp:105] Iteration 912, lr = 0.00834724
I0428 13:54:14.517172  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_918.caffemodel
I0428 13:54:16.704723  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_918.solverstate
I0428 13:54:19.182006  7476 solver.cpp:330] Iteration 918, Testing net (#0)
I0428 13:54:19.182024  7476 net.cpp:676] Ignoring source layer train-data
I0428 13:54:23.403702  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 13:54:23.847815  7476 solver.cpp:397]     Test net output #0: accuracy = 0.0251225
I0428 13:54:23.847844  7476 solver.cpp:397]     Test net output #1: loss = 4.95052 (* 1 = 4.95052 loss)
I0428 13:54:25.728407  7476 solver.cpp:218] Iteration 924 (0.885353 iter/s, 13.5539s/12 iters), loss = 4.9482
I0428 13:54:25.728447  7476 solver.cpp:237]     Train net output #0: loss = 4.9482 (* 1 = 4.9482 loss)
I0428 13:54:25.728456  7476 sgd_solver.cpp:105] Iteration 924, lr = 0.00832742
I0428 13:54:31.323263  7476 solver.cpp:218] Iteration 936 (2.14491 iter/s, 5.59464s/12 iters), loss = 4.79748
I0428 13:54:31.323309  7476 solver.cpp:237]     Train net output #0: loss = 4.79748 (* 1 = 4.79748 loss)
I0428 13:54:31.323318  7476 sgd_solver.cpp:105] Iteration 936, lr = 0.00830765
I0428 13:54:36.835738  7476 solver.cpp:218] Iteration 948 (2.17697 iter/s, 5.51226s/12 iters), loss = 4.79704
I0428 13:54:36.835788  7476 solver.cpp:237]     Train net output #0: loss = 4.79704 (* 1 = 4.79704 loss)
I0428 13:54:36.835798  7476 sgd_solver.cpp:105] Iteration 948, lr = 0.00828793
I0428 13:54:42.340795  7476 solver.cpp:218] Iteration 960 (2.1799 iter/s, 5.50483s/12 iters), loss = 4.91143
I0428 13:54:42.349933  7476 solver.cpp:237]     Train net output #0: loss = 4.91143 (* 1 = 4.91143 loss)
I0428 13:54:42.349946  7476 sgd_solver.cpp:105] Iteration 960, lr = 0.00826825
I0428 13:54:47.955811  7476 solver.cpp:218] Iteration 972 (2.14067 iter/s, 5.60571s/12 iters), loss = 4.9615
I0428 13:54:47.955858  7476 solver.cpp:237]     Train net output #0: loss = 4.9615 (* 1 = 4.9615 loss)
I0428 13:54:47.955868  7476 sgd_solver.cpp:105] Iteration 972, lr = 0.00824862
I0428 13:54:53.547518  7476 solver.cpp:218] Iteration 984 (2.14612 iter/s, 5.59149s/12 iters), loss = 4.80551
I0428 13:54:53.547570  7476 solver.cpp:237]     Train net output #0: loss = 4.80551 (* 1 = 4.80551 loss)
I0428 13:54:53.547583  7476 sgd_solver.cpp:105] Iteration 984, lr = 0.00822903
I0428 13:54:59.214793  7476 solver.cpp:218] Iteration 996 (2.1175 iter/s, 5.66706s/12 iters), loss = 4.84756
I0428 13:54:59.214835  7476 solver.cpp:237]     Train net output #0: loss = 4.84756 (* 1 = 4.84756 loss)
I0428 13:54:59.214845  7476 sgd_solver.cpp:105] Iteration 996, lr = 0.0082095
I0428 13:55:04.509168  7476 solver.cpp:218] Iteration 1008 (2.26664 iter/s, 5.29417s/12 iters), loss = 4.8367
I0428 13:55:04.509222  7476 solver.cpp:237]     Train net output #0: loss = 4.8367 (* 1 = 4.8367 loss)
I0428 13:55:04.509236  7476 sgd_solver.cpp:105] Iteration 1008, lr = 0.00819001
I0428 13:55:05.589186  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 13:55:09.431015  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_1020.caffemodel
I0428 13:55:11.771705  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_1020.solverstate
I0428 13:55:14.126766  7476 solver.cpp:330] Iteration 1020, Testing net (#0)
I0428 13:55:14.126847  7476 net.cpp:676] Ignoring source layer train-data
I0428 13:55:18.152281  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 13:55:18.596911  7476 solver.cpp:397]     Test net output #0: accuracy = 0.0367647
I0428 13:55:18.596951  7476 solver.cpp:397]     Test net output #1: loss = 4.79636 (* 1 = 4.79636 loss)
I0428 13:55:18.725462  7476 solver.cpp:218] Iteration 1020 (0.844129 iter/s, 14.2158s/12 iters), loss = 4.88558
I0428 13:55:18.725512  7476 solver.cpp:237]     Train net output #0: loss = 4.88558 (* 1 = 4.88558 loss)
I0428 13:55:18.725522  7476 sgd_solver.cpp:105] Iteration 1020, lr = 0.00817056
I0428 13:55:23.182019  7476 solver.cpp:218] Iteration 1032 (2.69278 iter/s, 4.45637s/12 iters), loss = 4.90581
I0428 13:55:23.182065  7476 solver.cpp:237]     Train net output #0: loss = 4.90581 (* 1 = 4.90581 loss)
I0428 13:55:23.182072  7476 sgd_solver.cpp:105] Iteration 1032, lr = 0.00815116
I0428 13:55:28.652213  7476 solver.cpp:218] Iteration 1044 (2.19379 iter/s, 5.46998s/12 iters), loss = 4.86796
I0428 13:55:28.652258  7476 solver.cpp:237]     Train net output #0: loss = 4.86796 (* 1 = 4.86796 loss)
I0428 13:55:28.652268  7476 sgd_solver.cpp:105] Iteration 1044, lr = 0.00813181
I0428 13:55:33.944775  7476 solver.cpp:218] Iteration 1056 (2.26742 iter/s, 5.29235s/12 iters), loss = 4.76688
I0428 13:55:33.944830  7476 solver.cpp:237]     Train net output #0: loss = 4.76688 (* 1 = 4.76688 loss)
I0428 13:55:33.944844  7476 sgd_solver.cpp:105] Iteration 1056, lr = 0.0081125
I0428 13:55:39.454788  7476 solver.cpp:218] Iteration 1068 (2.17794 iter/s, 5.5098s/12 iters), loss = 4.70642
I0428 13:55:39.454830  7476 solver.cpp:237]     Train net output #0: loss = 4.70642 (* 1 = 4.70642 loss)
I0428 13:55:39.454839  7476 sgd_solver.cpp:105] Iteration 1068, lr = 0.00809324
I0428 13:55:45.039420  7476 solver.cpp:218] Iteration 1080 (2.14884 iter/s, 5.58442s/12 iters), loss = 4.80772
I0428 13:55:45.041729  7476 solver.cpp:237]     Train net output #0: loss = 4.80772 (* 1 = 4.80772 loss)
I0428 13:55:45.041743  7476 sgd_solver.cpp:105] Iteration 1080, lr = 0.00807403
I0428 13:55:50.751257  7476 solver.cpp:218] Iteration 1092 (2.10181 iter/s, 5.70936s/12 iters), loss = 4.79197
I0428 13:55:50.751307  7476 solver.cpp:237]     Train net output #0: loss = 4.79197 (* 1 = 4.79197 loss)
I0428 13:55:50.751317  7476 sgd_solver.cpp:105] Iteration 1092, lr = 0.00805486
I0428 13:55:56.229183  7476 solver.cpp:218] Iteration 1104 (2.19069 iter/s, 5.47772s/12 iters), loss = 4.61329
I0428 13:55:56.229221  7476 solver.cpp:237]     Train net output #0: loss = 4.61329 (* 1 = 4.61329 loss)
I0428 13:55:56.229229  7476 sgd_solver.cpp:105] Iteration 1104, lr = 0.00803573
I0428 13:55:59.731698  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 13:56:01.770417  7476 solver.cpp:218] Iteration 1116 (2.16566 iter/s, 5.54103s/12 iters), loss = 4.7003
I0428 13:56:01.770469  7476 solver.cpp:237]     Train net output #0: loss = 4.7003 (* 1 = 4.7003 loss)
I0428 13:56:01.770479  7476 sgd_solver.cpp:105] Iteration 1116, lr = 0.00801666
I0428 13:56:03.928073  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_1122.caffemodel
I0428 13:56:09.359988  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_1122.solverstate
I0428 13:56:10.568843  7476 solver.cpp:330] Iteration 1122, Testing net (#0)
I0428 13:56:10.568862  7476 net.cpp:676] Ignoring source layer train-data
I0428 13:56:14.680430  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 13:56:15.185226  7476 solver.cpp:397]     Test net output #0: accuracy = 0.0300245
I0428 13:56:15.186241  7476 solver.cpp:397]     Test net output #1: loss = 4.68687 (* 1 = 4.68687 loss)
I0428 13:56:17.101565  7476 solver.cpp:218] Iteration 1128 (0.782745 iter/s, 15.3307s/12 iters), loss = 4.48964
I0428 13:56:17.101609  7476 solver.cpp:237]     Train net output #0: loss = 4.48964 (* 1 = 4.48964 loss)
I0428 13:56:17.101619  7476 sgd_solver.cpp:105] Iteration 1128, lr = 0.00799762
I0428 13:56:22.718004  7476 solver.cpp:218] Iteration 1140 (2.13666 iter/s, 5.61623s/12 iters), loss = 4.50305
I0428 13:56:22.718040  7476 solver.cpp:237]     Train net output #0: loss = 4.50305 (* 1 = 4.50305 loss)
I0428 13:56:22.718048  7476 sgd_solver.cpp:105] Iteration 1140, lr = 0.00797863
I0428 13:56:28.108537  7476 solver.cpp:218] Iteration 1152 (2.22622 iter/s, 5.3903s/12 iters), loss = 4.6096
I0428 13:56:28.108594  7476 solver.cpp:237]     Train net output #0: loss = 4.6096 (* 1 = 4.6096 loss)
I0428 13:56:28.108605  7476 sgd_solver.cpp:105] Iteration 1152, lr = 0.00795969
I0428 13:56:33.540661  7476 solver.cpp:218] Iteration 1164 (2.20917 iter/s, 5.43191s/12 iters), loss = 4.71041
I0428 13:56:33.540705  7476 solver.cpp:237]     Train net output #0: loss = 4.71041 (* 1 = 4.71041 loss)
I0428 13:56:33.540714  7476 sgd_solver.cpp:105] Iteration 1164, lr = 0.00794079
I0428 13:56:38.990082  7476 solver.cpp:218] Iteration 1176 (2.20215 iter/s, 5.44921s/12 iters), loss = 4.67605
I0428 13:56:38.990134  7476 solver.cpp:237]     Train net output #0: loss = 4.67605 (* 1 = 4.67605 loss)
I0428 13:56:38.990146  7476 sgd_solver.cpp:105] Iteration 1176, lr = 0.00792194
I0428 13:56:44.395037  7476 solver.cpp:218] Iteration 1188 (2.22027 iter/s, 5.40475s/12 iters), loss = 4.59166
I0428 13:56:44.395076  7476 solver.cpp:237]     Train net output #0: loss = 4.59166 (* 1 = 4.59166 loss)
I0428 13:56:44.395084  7476 sgd_solver.cpp:105] Iteration 1188, lr = 0.00790313
I0428 13:56:50.047626  7476 solver.cpp:218] Iteration 1200 (2.123 iter/s, 5.65238s/12 iters), loss = 4.69862
I0428 13:56:50.047979  7476 solver.cpp:237]     Train net output #0: loss = 4.69862 (* 1 = 4.69862 loss)
I0428 13:56:50.047991  7476 sgd_solver.cpp:105] Iteration 1200, lr = 0.00788437
I0428 13:56:55.541968  7476 solver.cpp:218] Iteration 1212 (2.18427 iter/s, 5.49382s/12 iters), loss = 4.70298
I0428 13:56:55.542026  7476 solver.cpp:237]     Train net output #0: loss = 4.70298 (* 1 = 4.70298 loss)
I0428 13:56:55.542037  7476 sgd_solver.cpp:105] Iteration 1212, lr = 0.00786565
I0428 13:56:55.828316  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 13:57:00.429406  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_1224.caffemodel
I0428 13:57:07.941735  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_1224.solverstate
I0428 13:57:14.336899  7476 solver.cpp:330] Iteration 1224, Testing net (#0)
I0428 13:57:14.336921  7476 net.cpp:676] Ignoring source layer train-data
I0428 13:57:18.517053  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 13:57:19.047811  7476 solver.cpp:397]     Test net output #0: accuracy = 0.0422794
I0428 13:57:19.047848  7476 solver.cpp:397]     Test net output #1: loss = 4.62219 (* 1 = 4.62219 loss)
I0428 13:57:19.176200  7476 solver.cpp:218] Iteration 1224 (0.507753 iter/s, 23.6335s/12 iters), loss = 4.4749
I0428 13:57:19.176241  7476 solver.cpp:237]     Train net output #0: loss = 4.4749 (* 1 = 4.4749 loss)
I0428 13:57:19.176251  7476 sgd_solver.cpp:105] Iteration 1224, lr = 0.00784697
I0428 13:57:23.834009  7476 solver.cpp:218] Iteration 1236 (2.57642 iter/s, 4.65762s/12 iters), loss = 4.52017
I0428 13:57:23.834146  7476 solver.cpp:237]     Train net output #0: loss = 4.52017 (* 1 = 4.52017 loss)
I0428 13:57:23.834154  7476 sgd_solver.cpp:105] Iteration 1236, lr = 0.00782834
I0428 13:57:29.436748  7476 solver.cpp:218] Iteration 1248 (2.14193 iter/s, 5.60243s/12 iters), loss = 4.72647
I0428 13:57:29.436792  7476 solver.cpp:237]     Train net output #0: loss = 4.72647 (* 1 = 4.72647 loss)
I0428 13:57:29.436801  7476 sgd_solver.cpp:105] Iteration 1248, lr = 0.00780976
I0428 13:57:35.062359  7476 solver.cpp:218] Iteration 1260 (2.13318 iter/s, 5.62539s/12 iters), loss = 4.73195
I0428 13:57:35.062417  7476 solver.cpp:237]     Train net output #0: loss = 4.73195 (* 1 = 4.73195 loss)
I0428 13:57:35.062430  7476 sgd_solver.cpp:105] Iteration 1260, lr = 0.00779122
I0428 13:57:40.698154  7476 solver.cpp:218] Iteration 1272 (2.12933 iter/s, 5.63557s/12 iters), loss = 4.66998
I0428 13:57:40.698207  7476 solver.cpp:237]     Train net output #0: loss = 4.66998 (* 1 = 4.66998 loss)
I0428 13:57:40.698222  7476 sgd_solver.cpp:105] Iteration 1272, lr = 0.00777272
I0428 13:57:46.371381  7476 solver.cpp:218] Iteration 1284 (2.11528 iter/s, 5.673s/12 iters), loss = 4.44058
I0428 13:57:46.371428  7476 solver.cpp:237]     Train net output #0: loss = 4.44058 (* 1 = 4.44058 loss)
I0428 13:57:46.371436  7476 sgd_solver.cpp:105] Iteration 1284, lr = 0.00775426
I0428 13:57:51.816556  7476 solver.cpp:218] Iteration 1296 (2.20387 iter/s, 5.44496s/12 iters), loss = 4.54423
I0428 13:57:51.816597  7476 solver.cpp:237]     Train net output #0: loss = 4.54423 (* 1 = 4.54423 loss)
I0428 13:57:51.816606  7476 sgd_solver.cpp:105] Iteration 1296, lr = 0.00773585
I0428 13:57:57.317875  7476 solver.cpp:218] Iteration 1308 (2.18138 iter/s, 5.50111s/12 iters), loss = 4.57265
I0428 13:57:57.317982  7476 solver.cpp:237]     Train net output #0: loss = 4.57265 (* 1 = 4.57265 loss)
I0428 13:57:57.317991  7476 sgd_solver.cpp:105] Iteration 1308, lr = 0.00771749
I0428 13:58:00.110374  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 13:58:02.887179  7476 solver.cpp:218] Iteration 1320 (2.15478 iter/s, 5.56903s/12 iters), loss = 4.61505
I0428 13:58:02.887233  7476 solver.cpp:237]     Train net output #0: loss = 4.61505 (* 1 = 4.61505 loss)
I0428 13:58:02.887243  7476 sgd_solver.cpp:105] Iteration 1320, lr = 0.00769916
I0428 13:58:05.184808  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_1326.caffemodel
I0428 13:58:09.585355  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_1326.solverstate
I0428 13:58:11.634155  7476 solver.cpp:330] Iteration 1326, Testing net (#0)
I0428 13:58:11.634182  7476 net.cpp:676] Ignoring source layer train-data
I0428 13:58:15.572476  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 13:58:16.141906  7476 solver.cpp:397]     Test net output #0: accuracy = 0.0447304
I0428 13:58:16.141935  7476 solver.cpp:397]     Test net output #1: loss = 4.47889 (* 1 = 4.47889 loss)
I0428 13:58:18.148531  7476 solver.cpp:218] Iteration 1332 (0.786325 iter/s, 15.2609s/12 iters), loss = 4.63983
I0428 13:58:18.148571  7476 solver.cpp:237]     Train net output #0: loss = 4.63983 (* 1 = 4.63983 loss)
I0428 13:58:18.148581  7476 sgd_solver.cpp:105] Iteration 1332, lr = 0.00768088
I0428 13:58:23.679931  7476 solver.cpp:218] Iteration 1344 (2.16951 iter/s, 5.53119s/12 iters), loss = 4.58624
I0428 13:58:23.679972  7476 solver.cpp:237]     Train net output #0: loss = 4.58624 (* 1 = 4.58624 loss)
I0428 13:58:23.679980  7476 sgd_solver.cpp:105] Iteration 1344, lr = 0.00766265
I0428 13:58:29.287334  7476 solver.cpp:218] Iteration 1356 (2.14011 iter/s, 5.60719s/12 iters), loss = 4.58963
I0428 13:58:29.287494  7476 solver.cpp:237]     Train net output #0: loss = 4.58963 (* 1 = 4.58963 loss)
I0428 13:58:29.287508  7476 sgd_solver.cpp:105] Iteration 1356, lr = 0.00764446
I0428 13:58:34.911141  7476 solver.cpp:218] Iteration 1368 (2.13391 iter/s, 5.62348s/12 iters), loss = 4.41563
I0428 13:58:34.911186  7476 solver.cpp:237]     Train net output #0: loss = 4.41563 (* 1 = 4.41563 loss)
I0428 13:58:34.911195  7476 sgd_solver.cpp:105] Iteration 1368, lr = 0.00762631
I0428 13:58:36.181850  7476 blocking_queue.cpp:49] Waiting for data
I0428 13:58:40.481997  7476 solver.cpp:218] Iteration 1380 (2.15415 iter/s, 5.57065s/12 iters), loss = 4.37381
I0428 13:58:40.482039  7476 solver.cpp:237]     Train net output #0: loss = 4.37381 (* 1 = 4.37381 loss)
I0428 13:58:40.482048  7476 sgd_solver.cpp:105] Iteration 1380, lr = 0.0076082
I0428 13:58:46.058779  7476 solver.cpp:218] Iteration 1392 (2.15186 iter/s, 5.57657s/12 iters), loss = 4.40022
I0428 13:58:46.058825  7476 solver.cpp:237]     Train net output #0: loss = 4.40022 (* 1 = 4.40022 loss)
I0428 13:58:46.058835  7476 sgd_solver.cpp:105] Iteration 1392, lr = 0.00759014
I0428 13:58:51.428943  7476 solver.cpp:218] Iteration 1404 (2.23465 iter/s, 5.36996s/12 iters), loss = 4.2842
I0428 13:58:51.428988  7476 solver.cpp:237]     Train net output #0: loss = 4.2842 (* 1 = 4.2842 loss)
I0428 13:58:51.429003  7476 sgd_solver.cpp:105] Iteration 1404, lr = 0.00757212
I0428 13:58:56.595147  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 13:58:56.993211  7476 solver.cpp:218] Iteration 1416 (2.1567 iter/s, 5.56406s/12 iters), loss = 4.41393
I0428 13:58:56.993255  7476 solver.cpp:237]     Train net output #0: loss = 4.41393 (* 1 = 4.41393 loss)
I0428 13:58:56.993264  7476 sgd_solver.cpp:105] Iteration 1416, lr = 0.00755414
I0428 13:59:02.458884  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_1428.caffemodel
I0428 13:59:07.305060  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_1428.solverstate
I0428 13:59:11.815503  7476 solver.cpp:330] Iteration 1428, Testing net (#0)
I0428 13:59:11.815524  7476 net.cpp:676] Ignoring source layer train-data
I0428 13:59:15.751853  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 13:59:16.373481  7476 solver.cpp:397]     Test net output #0: accuracy = 0.0594363
I0428 13:59:16.373509  7476 solver.cpp:397]     Test net output #1: loss = 4.32424 (* 1 = 4.32424 loss)
I0428 13:59:16.502327  7476 solver.cpp:218] Iteration 1428 (0.615115 iter/s, 19.5085s/12 iters), loss = 4.35125
I0428 13:59:16.502418  7476 solver.cpp:237]     Train net output #0: loss = 4.35125 (* 1 = 4.35125 loss)
I0428 13:59:16.502451  7476 sgd_solver.cpp:105] Iteration 1428, lr = 0.0075362
I0428 13:59:21.013940  7476 solver.cpp:218] Iteration 1440 (2.65993 iter/s, 4.51139s/12 iters), loss = 4.27271
I0428 13:59:21.013985  7476 solver.cpp:237]     Train net output #0: loss = 4.27271 (* 1 = 4.27271 loss)
I0428 13:59:21.013993  7476 sgd_solver.cpp:105] Iteration 1440, lr = 0.00751831
I0428 13:59:26.443560  7476 solver.cpp:218] Iteration 1452 (2.21018 iter/s, 5.42941s/12 iters), loss = 4.30261
I0428 13:59:26.443603  7476 solver.cpp:237]     Train net output #0: loss = 4.30261 (* 1 = 4.30261 loss)
I0428 13:59:26.443614  7476 sgd_solver.cpp:105] Iteration 1452, lr = 0.00750046
I0428 13:59:31.973119  7476 solver.cpp:218] Iteration 1464 (2.17024 iter/s, 5.52935s/12 iters), loss = 4.36069
I0428 13:59:31.973162  7476 solver.cpp:237]     Train net output #0: loss = 4.36069 (* 1 = 4.36069 loss)
I0428 13:59:31.973171  7476 sgd_solver.cpp:105] Iteration 1464, lr = 0.00748265
I0428 13:59:37.564373  7476 solver.cpp:218] Iteration 1476 (2.14629 iter/s, 5.59104s/12 iters), loss = 4.23016
I0428 13:59:37.572994  7476 solver.cpp:237]     Train net output #0: loss = 4.23016 (* 1 = 4.23016 loss)
I0428 13:59:37.573006  7476 sgd_solver.cpp:105] Iteration 1476, lr = 0.00746489
I0428 13:59:43.179977  7476 solver.cpp:218] Iteration 1488 (2.14025 iter/s, 5.60683s/12 iters), loss = 4.29654
I0428 13:59:43.180030  7476 solver.cpp:237]     Train net output #0: loss = 4.29654 (* 1 = 4.29654 loss)
I0428 13:59:43.180043  7476 sgd_solver.cpp:105] Iteration 1488, lr = 0.00744716
I0428 13:59:48.801522  7476 solver.cpp:218] Iteration 1500 (2.13473 iter/s, 5.62133s/12 iters), loss = 4.25337
I0428 13:59:48.801560  7476 solver.cpp:237]     Train net output #0: loss = 4.25337 (* 1 = 4.25337 loss)
I0428 13:59:48.801569  7476 sgd_solver.cpp:105] Iteration 1500, lr = 0.00742948
I0428 13:59:54.340548  7476 solver.cpp:218] Iteration 1512 (2.16654 iter/s, 5.53879s/12 iters), loss = 4.22252
I0428 13:59:54.340603  7476 solver.cpp:237]     Train net output #0: loss = 4.22252 (* 1 = 4.22252 loss)
I0428 13:59:54.340612  7476 sgd_solver.cpp:105] Iteration 1512, lr = 0.00741184
I0428 13:59:56.217186  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 13:59:59.730944  7476 solver.cpp:218] Iteration 1524 (2.22627 iter/s, 5.39018s/12 iters), loss = 4.15893
I0428 13:59:59.730984  7476 solver.cpp:237]     Train net output #0: loss = 4.15893 (* 1 = 4.15893 loss)
I0428 13:59:59.730994  7476 sgd_solver.cpp:105] Iteration 1524, lr = 0.00739425
I0428 14:00:01.779646  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_1530.caffemodel
I0428 14:00:07.669982  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_1530.solverstate
I0428 14:00:12.283866  7476 solver.cpp:330] Iteration 1530, Testing net (#0)
I0428 14:00:12.283892  7476 net.cpp:676] Ignoring source layer train-data
I0428 14:00:16.137377  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:00:16.820333  7476 solver.cpp:397]     Test net output #0: accuracy = 0.0667892
I0428 14:00:16.820361  7476 solver.cpp:397]     Test net output #1: loss = 4.26153 (* 1 = 4.26153 loss)
I0428 14:00:18.815941  7476 solver.cpp:218] Iteration 1536 (0.628785 iter/s, 19.0844s/12 iters), loss = 3.96466
I0428 14:00:18.815981  7476 solver.cpp:237]     Train net output #0: loss = 3.96466 (* 1 = 3.96466 loss)
I0428 14:00:18.815991  7476 sgd_solver.cpp:105] Iteration 1536, lr = 0.00737669
I0428 14:00:24.331948  7476 solver.cpp:218] Iteration 1548 (2.17557 iter/s, 5.5158s/12 iters), loss = 4.10038
I0428 14:00:24.331988  7476 solver.cpp:237]     Train net output #0: loss = 4.10038 (* 1 = 4.10038 loss)
I0428 14:00:24.331998  7476 sgd_solver.cpp:105] Iteration 1548, lr = 0.00735918
I0428 14:00:29.874897  7476 solver.cpp:218] Iteration 1560 (2.165 iter/s, 5.54274s/12 iters), loss = 4.0567
I0428 14:00:29.874949  7476 solver.cpp:237]     Train net output #0: loss = 4.0567 (* 1 = 4.0567 loss)
I0428 14:00:29.874963  7476 sgd_solver.cpp:105] Iteration 1560, lr = 0.00734171
I0428 14:00:35.503211  7476 solver.cpp:218] Iteration 1572 (2.13216 iter/s, 5.62809s/12 iters), loss = 4.25396
I0428 14:00:35.503264  7476 solver.cpp:237]     Train net output #0: loss = 4.25396 (* 1 = 4.25396 loss)
I0428 14:00:35.503274  7476 sgd_solver.cpp:105] Iteration 1572, lr = 0.00732427
I0428 14:00:41.265920  7476 solver.cpp:218] Iteration 1584 (2.08243 iter/s, 5.76249s/12 iters), loss = 4.15654
I0428 14:00:41.266031  7476 solver.cpp:237]     Train net output #0: loss = 4.15654 (* 1 = 4.15654 loss)
I0428 14:00:41.266041  7476 sgd_solver.cpp:105] Iteration 1584, lr = 0.00730688
I0428 14:00:46.728053  7476 solver.cpp:218] Iteration 1596 (2.19706 iter/s, 5.46185s/12 iters), loss = 3.90848
I0428 14:00:46.728104  7476 solver.cpp:237]     Train net output #0: loss = 3.90848 (* 1 = 3.90848 loss)
I0428 14:00:46.728121  7476 sgd_solver.cpp:105] Iteration 1596, lr = 0.00728954
I0428 14:00:52.208215  7476 solver.cpp:218] Iteration 1608 (2.1898 iter/s, 5.47995s/12 iters), loss = 4.09993
I0428 14:00:52.208264  7476 solver.cpp:237]     Train net output #0: loss = 4.09993 (* 1 = 4.09993 loss)
I0428 14:00:52.208277  7476 sgd_solver.cpp:105] Iteration 1608, lr = 0.00727223
I0428 14:00:56.656100  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:00:58.137737  7476 solver.cpp:218] Iteration 1620 (2.02385 iter/s, 5.9293s/12 iters), loss = 4.08775
I0428 14:00:58.137790  7476 solver.cpp:237]     Train net output #0: loss = 4.08775 (* 1 = 4.08775 loss)
I0428 14:00:58.137805  7476 sgd_solver.cpp:105] Iteration 1620, lr = 0.00725496
I0428 14:01:02.997467  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_1632.caffemodel
I0428 14:01:08.717046  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_1632.solverstate
I0428 14:01:12.980407  7476 solver.cpp:330] Iteration 1632, Testing net (#0)
I0428 14:01:12.980602  7476 net.cpp:676] Ignoring source layer train-data
I0428 14:01:16.836262  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:01:17.536199  7476 solver.cpp:397]     Test net output #0: accuracy = 0.0790441
I0428 14:01:17.536228  7476 solver.cpp:397]     Test net output #1: loss = 4.22128 (* 1 = 4.22128 loss)
I0428 14:01:17.664546  7476 solver.cpp:218] Iteration 1632 (0.614558 iter/s, 19.5262s/12 iters), loss = 4.09547
I0428 14:01:17.664602  7476 solver.cpp:237]     Train net output #0: loss = 4.09547 (* 1 = 4.09547 loss)
I0428 14:01:17.664613  7476 sgd_solver.cpp:105] Iteration 1632, lr = 0.00723774
I0428 14:01:22.273447  7476 solver.cpp:218] Iteration 1644 (2.60377 iter/s, 4.6087s/12 iters), loss = 4.01076
I0428 14:01:22.273512  7476 solver.cpp:237]     Train net output #0: loss = 4.01076 (* 1 = 4.01076 loss)
I0428 14:01:22.273525  7476 sgd_solver.cpp:105] Iteration 1644, lr = 0.00722056
I0428 14:01:27.968276  7476 solver.cpp:218] Iteration 1656 (2.10726 iter/s, 5.6946s/12 iters), loss = 4.06061
I0428 14:01:27.968319  7476 solver.cpp:237]     Train net output #0: loss = 4.06061 (* 1 = 4.06061 loss)
I0428 14:01:27.968329  7476 sgd_solver.cpp:105] Iteration 1656, lr = 0.00720341
I0428 14:01:33.384441  7476 solver.cpp:218] Iteration 1668 (2.21567 iter/s, 5.41596s/12 iters), loss = 4.37244
I0428 14:01:33.384516  7476 solver.cpp:237]     Train net output #0: loss = 4.37244 (* 1 = 4.37244 loss)
I0428 14:01:33.384527  7476 sgd_solver.cpp:105] Iteration 1668, lr = 0.00718631
I0428 14:01:38.869001  7476 solver.cpp:218] Iteration 1680 (2.18804 iter/s, 5.48435s/12 iters), loss = 3.98668
I0428 14:01:38.869041  7476 solver.cpp:237]     Train net output #0: loss = 3.98668 (* 1 = 3.98668 loss)
I0428 14:01:38.869050  7476 sgd_solver.cpp:105] Iteration 1680, lr = 0.00716925
I0428 14:01:44.520670  7476 solver.cpp:218] Iteration 1692 (2.12334 iter/s, 5.65147s/12 iters), loss = 4.06568
I0428 14:01:44.520783  7476 solver.cpp:237]     Train net output #0: loss = 4.06568 (* 1 = 4.06568 loss)
I0428 14:01:44.520793  7476 sgd_solver.cpp:105] Iteration 1692, lr = 0.00715223
I0428 14:01:50.281003  7476 solver.cpp:218] Iteration 1704 (2.08331 iter/s, 5.76005s/12 iters), loss = 4.17389
I0428 14:01:50.281054  7476 solver.cpp:237]     Train net output #0: loss = 4.17389 (* 1 = 4.17389 loss)
I0428 14:01:50.281062  7476 sgd_solver.cpp:105] Iteration 1704, lr = 0.00713525
I0428 14:01:55.837270  7476 solver.cpp:218] Iteration 1716 (2.15981 iter/s, 5.55605s/12 iters), loss = 4.26871
I0428 14:01:55.837327  7476 solver.cpp:237]     Train net output #0: loss = 4.26871 (* 1 = 4.26871 loss)
I0428 14:01:55.837339  7476 sgd_solver.cpp:105] Iteration 1716, lr = 0.00711831
I0428 14:01:56.976467  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:02:01.401767  7476 solver.cpp:218] Iteration 1728 (2.15662 iter/s, 5.56427s/12 iters), loss = 4.1638
I0428 14:02:01.401818  7476 solver.cpp:237]     Train net output #0: loss = 4.1638 (* 1 = 4.1638 loss)
I0428 14:02:01.401827  7476 sgd_solver.cpp:105] Iteration 1728, lr = 0.00710141
I0428 14:02:03.638393  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_1734.caffemodel
I0428 14:02:04.966064  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_1734.solverstate
I0428 14:02:06.031026  7476 solver.cpp:330] Iteration 1734, Testing net (#0)
I0428 14:02:06.031046  7476 net.cpp:676] Ignoring source layer train-data
I0428 14:02:09.977178  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:02:10.753468  7476 solver.cpp:397]     Test net output #0: accuracy = 0.0808824
I0428 14:02:10.753494  7476 solver.cpp:397]     Test net output #1: loss = 4.14724 (* 1 = 4.14724 loss)
I0428 14:02:13.006969  7476 solver.cpp:218] Iteration 1740 (1.03405 iter/s, 11.6048s/12 iters), loss = 4.11292
I0428 14:02:13.007015  7476 solver.cpp:237]     Train net output #0: loss = 4.11292 (* 1 = 4.11292 loss)
I0428 14:02:13.007025  7476 sgd_solver.cpp:105] Iteration 1740, lr = 0.00708455
I0428 14:02:18.522037  7476 solver.cpp:218] Iteration 1752 (2.17594 iter/s, 5.51486s/12 iters), loss = 4.07569
I0428 14:02:18.522154  7476 solver.cpp:237]     Train net output #0: loss = 4.07569 (* 1 = 4.07569 loss)
I0428 14:02:18.522164  7476 sgd_solver.cpp:105] Iteration 1752, lr = 0.00706773
I0428 14:02:23.957885  7476 solver.cpp:218] Iteration 1764 (2.20768 iter/s, 5.43557s/12 iters), loss = 4.05261
I0428 14:02:23.957926  7476 solver.cpp:237]     Train net output #0: loss = 4.05261 (* 1 = 4.05261 loss)
I0428 14:02:23.957934  7476 sgd_solver.cpp:105] Iteration 1764, lr = 0.00705094
I0428 14:02:29.517300  7476 solver.cpp:218] Iteration 1776 (2.15858 iter/s, 5.55921s/12 iters), loss = 3.99005
I0428 14:02:29.517341  7476 solver.cpp:237]     Train net output #0: loss = 3.99005 (* 1 = 3.99005 loss)
I0428 14:02:29.517349  7476 sgd_solver.cpp:105] Iteration 1776, lr = 0.0070342
I0428 14:02:35.171382  7476 solver.cpp:218] Iteration 1788 (2.12244 iter/s, 5.65387s/12 iters), loss = 4.03494
I0428 14:02:35.171445  7476 solver.cpp:237]     Train net output #0: loss = 4.03494 (* 1 = 4.03494 loss)
I0428 14:02:35.171458  7476 sgd_solver.cpp:105] Iteration 1788, lr = 0.0070175
I0428 14:02:40.699261  7476 solver.cpp:218] Iteration 1800 (2.1709 iter/s, 5.52766s/12 iters), loss = 4.04054
I0428 14:02:40.699306  7476 solver.cpp:237]     Train net output #0: loss = 4.04054 (* 1 = 4.04054 loss)
I0428 14:02:40.699314  7476 sgd_solver.cpp:105] Iteration 1800, lr = 0.00700084
I0428 14:02:46.771215  7476 solver.cpp:218] Iteration 1812 (1.97637 iter/s, 6.07173s/12 iters), loss = 3.93575
I0428 14:02:46.771256  7476 solver.cpp:237]     Train net output #0: loss = 3.93575 (* 1 = 3.93575 loss)
I0428 14:02:46.771265  7476 sgd_solver.cpp:105] Iteration 1812, lr = 0.00698422
I0428 14:02:50.307783  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:02:52.574702  7476 solver.cpp:218] Iteration 1824 (2.0678 iter/s, 5.80328s/12 iters), loss = 4.07652
I0428 14:02:52.574743  7476 solver.cpp:237]     Train net output #0: loss = 4.07652 (* 1 = 4.07652 loss)
I0428 14:02:52.574751  7476 sgd_solver.cpp:105] Iteration 1824, lr = 0.00696764
I0428 14:02:57.601728  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_1836.caffemodel
I0428 14:02:59.057731  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_1836.solverstate
I0428 14:03:00.105965  7476 solver.cpp:330] Iteration 1836, Testing net (#0)
I0428 14:03:00.105984  7476 net.cpp:676] Ignoring source layer train-data
I0428 14:03:04.027843  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:03:04.931035  7476 solver.cpp:397]     Test net output #0: accuracy = 0.088848
I0428 14:03:04.931068  7476 solver.cpp:397]     Test net output #1: loss = 4.05926 (* 1 = 4.05926 loss)
I0428 14:03:05.059707  7476 solver.cpp:218] Iteration 1836 (0.961183 iter/s, 12.4846s/12 iters), loss = 3.95086
I0428 14:03:05.059763  7476 solver.cpp:237]     Train net output #0: loss = 3.95086 (* 1 = 3.95086 loss)
I0428 14:03:05.059775  7476 sgd_solver.cpp:105] Iteration 1836, lr = 0.0069511
I0428 14:03:09.690404  7476 solver.cpp:218] Iteration 1848 (2.59151 iter/s, 4.6305s/12 iters), loss = 3.80607
I0428 14:03:09.690450  7476 solver.cpp:237]     Train net output #0: loss = 3.80607 (* 1 = 3.80607 loss)
I0428 14:03:09.690459  7476 sgd_solver.cpp:105] Iteration 1848, lr = 0.00693459
I0428 14:03:15.122316  7476 solver.cpp:218] Iteration 1860 (2.20925 iter/s, 5.4317s/12 iters), loss = 3.97257
I0428 14:03:15.122367  7476 solver.cpp:237]     Train net output #0: loss = 3.97257 (* 1 = 3.97257 loss)
I0428 14:03:15.122380  7476 sgd_solver.cpp:105] Iteration 1860, lr = 0.00691813
I0428 14:03:20.900177  7476 solver.cpp:218] Iteration 1872 (2.07697 iter/s, 5.77765s/12 iters), loss = 3.81841
I0428 14:03:20.900301  7476 solver.cpp:237]     Train net output #0: loss = 3.81841 (* 1 = 3.81841 loss)
I0428 14:03:20.900310  7476 sgd_solver.cpp:105] Iteration 1872, lr = 0.0069017
I0428 14:03:27.209522  7476 solver.cpp:218] Iteration 1884 (1.90203 iter/s, 6.30904s/12 iters), loss = 4.05364
I0428 14:03:27.209566  7476 solver.cpp:237]     Train net output #0: loss = 4.05364 (* 1 = 4.05364 loss)
I0428 14:03:27.209575  7476 sgd_solver.cpp:105] Iteration 1884, lr = 0.00688532
I0428 14:03:32.859298  7476 solver.cpp:218] Iteration 1896 (2.12406 iter/s, 5.64956s/12 iters), loss = 3.86886
I0428 14:03:32.859351  7476 solver.cpp:237]     Train net output #0: loss = 3.86886 (* 1 = 3.86886 loss)
I0428 14:03:32.859364  7476 sgd_solver.cpp:105] Iteration 1896, lr = 0.00686897
I0428 14:03:38.883515  7476 solver.cpp:218] Iteration 1908 (1.99203 iter/s, 6.02399s/12 iters), loss = 3.95893
I0428 14:03:38.883563  7476 solver.cpp:237]     Train net output #0: loss = 3.95893 (* 1 = 3.95893 loss)
I0428 14:03:38.883574  7476 sgd_solver.cpp:105] Iteration 1908, lr = 0.00685266
I0428 14:03:44.674647  7476 solver.cpp:218] Iteration 1920 (2.07221 iter/s, 5.79092s/12 iters), loss = 3.88948
I0428 14:03:44.674690  7476 solver.cpp:237]     Train net output #0: loss = 3.88948 (* 1 = 3.88948 loss)
I0428 14:03:44.674697  7476 sgd_solver.cpp:105] Iteration 1920, lr = 0.00683639
I0428 14:03:44.995008  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:03:50.912915  7476 solver.cpp:218] Iteration 1932 (1.92368 iter/s, 6.23805s/12 iters), loss = 3.60336
I0428 14:03:50.913022  7476 solver.cpp:237]     Train net output #0: loss = 3.60336 (* 1 = 3.60336 loss)
I0428 14:03:50.913030  7476 sgd_solver.cpp:105] Iteration 1932, lr = 0.00682016
I0428 14:03:53.137995  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_1938.caffemodel
I0428 14:03:58.150198  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_1938.solverstate
I0428 14:04:02.469264  7476 solver.cpp:330] Iteration 1938, Testing net (#0)
I0428 14:04:02.469283  7476 net.cpp:676] Ignoring source layer train-data
I0428 14:04:06.201638  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:04:07.001138  7476 solver.cpp:397]     Test net output #0: accuracy = 0.101716
I0428 14:04:07.001165  7476 solver.cpp:397]     Test net output #1: loss = 3.94002 (* 1 = 3.94002 loss)
I0428 14:04:08.992208  7476 solver.cpp:218] Iteration 1944 (0.663764 iter/s, 18.0787s/12 iters), loss = 3.77643
I0428 14:04:08.992254  7476 solver.cpp:237]     Train net output #0: loss = 3.77643 (* 1 = 3.77643 loss)
I0428 14:04:08.992262  7476 sgd_solver.cpp:105] Iteration 1944, lr = 0.00680397
I0428 14:04:14.645903  7476 solver.cpp:218] Iteration 1956 (2.12259 iter/s, 5.65348s/12 iters), loss = 3.88576
I0428 14:04:14.645959  7476 solver.cpp:237]     Train net output #0: loss = 3.88576 (* 1 = 3.88576 loss)
I0428 14:04:14.645972  7476 sgd_solver.cpp:105] Iteration 1956, lr = 0.00678782
I0428 14:04:20.322685  7476 solver.cpp:218] Iteration 1968 (2.11396 iter/s, 5.67656s/12 iters), loss = 4.21532
I0428 14:04:20.322726  7476 solver.cpp:237]     Train net output #0: loss = 4.21532 (* 1 = 4.21532 loss)
I0428 14:04:20.322733  7476 sgd_solver.cpp:105] Iteration 1968, lr = 0.0067717
I0428 14:04:25.934906  7476 solver.cpp:218] Iteration 1980 (2.13827 iter/s, 5.61202s/12 iters), loss = 3.90542
I0428 14:04:25.935032  7476 solver.cpp:237]     Train net output #0: loss = 3.90542 (* 1 = 3.90542 loss)
I0428 14:04:25.935041  7476 sgd_solver.cpp:105] Iteration 1980, lr = 0.00675562
I0428 14:04:31.614943  7476 solver.cpp:218] Iteration 1992 (2.11277 iter/s, 5.67975s/12 iters), loss = 3.97887
I0428 14:04:31.614984  7476 solver.cpp:237]     Train net output #0: loss = 3.97887 (* 1 = 3.97887 loss)
I0428 14:04:31.614992  7476 sgd_solver.cpp:105] Iteration 1992, lr = 0.00673958
I0428 14:04:37.121498  7476 solver.cpp:218] Iteration 2004 (2.1793 iter/s, 5.50635s/12 iters), loss = 3.856
I0428 14:04:37.121539  7476 solver.cpp:237]     Train net output #0: loss = 3.856 (* 1 = 3.856 loss)
I0428 14:04:37.121549  7476 sgd_solver.cpp:105] Iteration 2004, lr = 0.00672358
I0428 14:04:43.053689  7476 solver.cpp:218] Iteration 2016 (2.02294 iter/s, 5.93197s/12 iters), loss = 3.81741
I0428 14:04:43.053731  7476 solver.cpp:237]     Train net output #0: loss = 3.81741 (* 1 = 3.81741 loss)
I0428 14:04:43.053740  7476 sgd_solver.cpp:105] Iteration 2016, lr = 0.00670762
I0428 14:04:46.002336  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:04:48.820432  7476 solver.cpp:218] Iteration 2028 (2.08097 iter/s, 5.76653s/12 iters), loss = 3.92978
I0428 14:04:48.820531  7476 solver.cpp:237]     Train net output #0: loss = 3.92978 (* 1 = 3.92978 loss)
I0428 14:04:48.820544  7476 sgd_solver.cpp:105] Iteration 2028, lr = 0.00669169
I0428 14:04:53.850502  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_2040.caffemodel
I0428 14:04:59.400853  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_2040.solverstate
I0428 14:05:00.447118  7476 solver.cpp:330] Iteration 2040, Testing net (#0)
I0428 14:05:00.447136  7476 net.cpp:676] Ignoring source layer train-data
I0428 14:05:04.247838  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:05:05.092089  7476 solver.cpp:397]     Test net output #0: accuracy = 0.103554
I0428 14:05:05.092128  7476 solver.cpp:397]     Test net output #1: loss = 3.9455 (* 1 = 3.9455 loss)
I0428 14:05:05.220537  7476 solver.cpp:218] Iteration 2040 (0.731725 iter/s, 16.3996s/12 iters), loss = 3.91941
I0428 14:05:05.220578  7476 solver.cpp:237]     Train net output #0: loss = 3.91941 (* 1 = 3.91941 loss)
I0428 14:05:05.220589  7476 sgd_solver.cpp:105] Iteration 2040, lr = 0.00667581
I0428 14:05:09.909698  7476 solver.cpp:218] Iteration 2052 (2.5592 iter/s, 4.68897s/12 iters), loss = 3.81951
I0428 14:05:09.909736  7476 solver.cpp:237]     Train net output #0: loss = 3.81951 (* 1 = 3.81951 loss)
I0428 14:05:09.909746  7476 sgd_solver.cpp:105] Iteration 2052, lr = 0.00665996
I0428 14:05:11.638183  7476 blocking_queue.cpp:49] Waiting for data
I0428 14:05:15.385234  7476 solver.cpp:218] Iteration 2064 (2.19165 iter/s, 5.47534s/12 iters), loss = 3.72239
I0428 14:05:15.385280  7476 solver.cpp:237]     Train net output #0: loss = 3.72239 (* 1 = 3.72239 loss)
I0428 14:05:15.385289  7476 sgd_solver.cpp:105] Iteration 2064, lr = 0.00664414
I0428 14:05:21.065225  7476 solver.cpp:218] Iteration 2076 (2.11276 iter/s, 5.67978s/12 iters), loss = 3.70927
I0428 14:05:21.065268  7476 solver.cpp:237]     Train net output #0: loss = 3.70927 (* 1 = 3.70927 loss)
I0428 14:05:21.065276  7476 sgd_solver.cpp:105] Iteration 2076, lr = 0.00662837
I0428 14:05:26.816439  7476 solver.cpp:218] Iteration 2088 (2.08659 iter/s, 5.751s/12 iters), loss = 3.54812
I0428 14:05:26.816515  7476 solver.cpp:237]     Train net output #0: loss = 3.54812 (* 1 = 3.54812 loss)
I0428 14:05:26.816529  7476 sgd_solver.cpp:105] Iteration 2088, lr = 0.00661263
I0428 14:05:32.857582  7476 solver.cpp:218] Iteration 2100 (1.98645 iter/s, 6.04092s/12 iters), loss = 3.94485
I0428 14:05:32.857705  7476 solver.cpp:237]     Train net output #0: loss = 3.94485 (* 1 = 3.94485 loss)
I0428 14:05:32.857715  7476 sgd_solver.cpp:105] Iteration 2100, lr = 0.00659693
I0428 14:05:38.437115  7476 solver.cpp:218] Iteration 2112 (2.15083 iter/s, 5.57925s/12 iters), loss = 3.8159
I0428 14:05:38.437170  7476 solver.cpp:237]     Train net output #0: loss = 3.8159 (* 1 = 3.8159 loss)
I0428 14:05:38.437183  7476 sgd_solver.cpp:105] Iteration 2112, lr = 0.00658127
I0428 14:05:43.571561  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:05:44.095398  7476 solver.cpp:218] Iteration 2124 (2.12087 iter/s, 5.65807s/12 iters), loss = 3.83717
I0428 14:05:44.095461  7476 solver.cpp:237]     Train net output #0: loss = 3.83717 (* 1 = 3.83717 loss)
I0428 14:05:44.095472  7476 sgd_solver.cpp:105] Iteration 2124, lr = 0.00656564
I0428 14:05:49.829360  7476 solver.cpp:218] Iteration 2136 (2.09287 iter/s, 5.73375s/12 iters), loss = 3.70578
I0428 14:05:49.829403  7476 solver.cpp:237]     Train net output #0: loss = 3.70578 (* 1 = 3.70578 loss)
I0428 14:05:49.829414  7476 sgd_solver.cpp:105] Iteration 2136, lr = 0.00655006
I0428 14:05:52.075390  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_2142.caffemodel
I0428 14:05:59.458637  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_2142.solverstate
I0428 14:06:00.515769  7476 solver.cpp:330] Iteration 2142, Testing net (#0)
I0428 14:06:00.515787  7476 net.cpp:676] Ignoring source layer train-data
I0428 14:06:04.199261  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:06:05.111579  7476 solver.cpp:397]     Test net output #0: accuracy = 0.107843
I0428 14:06:05.111616  7476 solver.cpp:397]     Test net output #1: loss = 3.85071 (* 1 = 3.85071 loss)
I0428 14:06:07.160254  7476 solver.cpp:218] Iteration 2148 (0.692425 iter/s, 17.3304s/12 iters), loss = 3.60788
I0428 14:06:07.160296  7476 solver.cpp:237]     Train net output #0: loss = 3.60788 (* 1 = 3.60788 loss)
I0428 14:06:07.160306  7476 sgd_solver.cpp:105] Iteration 2148, lr = 0.00653451
I0428 14:06:12.664870  7476 solver.cpp:218] Iteration 2160 (2.18007 iter/s, 5.50441s/12 iters), loss = 3.51118
I0428 14:06:12.664912  7476 solver.cpp:237]     Train net output #0: loss = 3.51118 (* 1 = 3.51118 loss)
I0428 14:06:12.664921  7476 sgd_solver.cpp:105] Iteration 2160, lr = 0.00651899
I0428 14:06:18.248564  7476 solver.cpp:218] Iteration 2172 (2.1492 iter/s, 5.58346s/12 iters), loss = 3.75031
I0428 14:06:18.248605  7476 solver.cpp:237]     Train net output #0: loss = 3.75031 (* 1 = 3.75031 loss)
I0428 14:06:18.248613  7476 sgd_solver.cpp:105] Iteration 2172, lr = 0.00650351
I0428 14:06:23.826196  7476 solver.cpp:218] Iteration 2184 (2.15153 iter/s, 5.57743s/12 iters), loss = 3.51092
I0428 14:06:23.826244  7476 solver.cpp:237]     Train net output #0: loss = 3.51092 (* 1 = 3.51092 loss)
I0428 14:06:23.826256  7476 sgd_solver.cpp:105] Iteration 2184, lr = 0.00648807
I0428 14:06:29.637830  7476 solver.cpp:218] Iteration 2196 (2.0649 iter/s, 5.81142s/12 iters), loss = 3.5304
I0428 14:06:29.637873  7476 solver.cpp:237]     Train net output #0: loss = 3.5304 (* 1 = 3.5304 loss)
I0428 14:06:29.637882  7476 sgd_solver.cpp:105] Iteration 2196, lr = 0.00647267
I0428 14:06:35.160226  7476 solver.cpp:218] Iteration 2208 (2.17305 iter/s, 5.52219s/12 iters), loss = 3.75483
I0428 14:06:35.160346  7476 solver.cpp:237]     Train net output #0: loss = 3.75483 (* 1 = 3.75483 loss)
I0428 14:06:35.160356  7476 sgd_solver.cpp:105] Iteration 2208, lr = 0.0064573
I0428 14:06:40.898450  7476 solver.cpp:218] Iteration 2220 (2.09134 iter/s, 5.73794s/12 iters), loss = 3.49362
I0428 14:06:40.898504  7476 solver.cpp:237]     Train net output #0: loss = 3.49362 (* 1 = 3.49362 loss)
I0428 14:06:40.898516  7476 sgd_solver.cpp:105] Iteration 2220, lr = 0.00644197
I0428 14:06:42.900580  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:06:46.430626  7476 solver.cpp:218] Iteration 2232 (2.16921 iter/s, 5.53196s/12 iters), loss = 3.34676
I0428 14:06:46.430666  7476 solver.cpp:237]     Train net output #0: loss = 3.34676 (* 1 = 3.34676 loss)
I0428 14:06:46.430675  7476 sgd_solver.cpp:105] Iteration 2232, lr = 0.00642668
I0428 14:06:51.453526  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_2244.caffemodel
I0428 14:06:58.968796  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_2244.solverstate
I0428 14:07:00.069689  7476 solver.cpp:330] Iteration 2244, Testing net (#0)
I0428 14:07:00.069715  7476 net.cpp:676] Ignoring source layer train-data
I0428 14:07:03.776523  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:07:04.721611  7476 solver.cpp:397]     Test net output #0: accuracy = 0.120711
I0428 14:07:04.721637  7476 solver.cpp:397]     Test net output #1: loss = 3.77377 (* 1 = 3.77377 loss)
I0428 14:07:04.849756  7476 solver.cpp:218] Iteration 2244 (0.651516 iter/s, 18.4186s/12 iters), loss = 3.37157
I0428 14:07:04.849817  7476 solver.cpp:237]     Train net output #0: loss = 3.37157 (* 1 = 3.37157 loss)
I0428 14:07:04.849831  7476 sgd_solver.cpp:105] Iteration 2244, lr = 0.00641142
I0428 14:07:09.527627  7476 solver.cpp:218] Iteration 2256 (2.56538 iter/s, 4.67767s/12 iters), loss = 3.58063
I0428 14:07:09.527763  7476 solver.cpp:237]     Train net output #0: loss = 3.58063 (* 1 = 3.58063 loss)
I0428 14:07:09.527773  7476 sgd_solver.cpp:105] Iteration 2256, lr = 0.0063962
I0428 14:07:15.001549  7476 solver.cpp:218] Iteration 2268 (2.19233 iter/s, 5.47363s/12 iters), loss = 3.44957
I0428 14:07:15.001591  7476 solver.cpp:237]     Train net output #0: loss = 3.44957 (* 1 = 3.44957 loss)
I0428 14:07:15.001602  7476 sgd_solver.cpp:105] Iteration 2268, lr = 0.00638101
I0428 14:07:20.466572  7476 solver.cpp:218] Iteration 2280 (2.19586 iter/s, 5.46482s/12 iters), loss = 3.49819
I0428 14:07:20.466627  7476 solver.cpp:237]     Train net output #0: loss = 3.49819 (* 1 = 3.49819 loss)
I0428 14:07:20.466639  7476 sgd_solver.cpp:105] Iteration 2280, lr = 0.00636586
I0428 14:07:25.956373  7476 solver.cpp:218] Iteration 2292 (2.18596 iter/s, 5.48958s/12 iters), loss = 3.40094
I0428 14:07:25.956431  7476 solver.cpp:237]     Train net output #0: loss = 3.40094 (* 1 = 3.40094 loss)
I0428 14:07:25.956444  7476 sgd_solver.cpp:105] Iteration 2292, lr = 0.00635075
I0428 14:07:31.630988  7476 solver.cpp:218] Iteration 2304 (2.11476 iter/s, 5.67439s/12 iters), loss = 3.44094
I0428 14:07:31.631045  7476 solver.cpp:237]     Train net output #0: loss = 3.44094 (* 1 = 3.44094 loss)
I0428 14:07:31.631057  7476 sgd_solver.cpp:105] Iteration 2304, lr = 0.00633567
I0428 14:07:37.313311  7476 solver.cpp:218] Iteration 2316 (2.11189 iter/s, 5.6821s/12 iters), loss = 3.58429
I0428 14:07:37.313349  7476 solver.cpp:237]     Train net output #0: loss = 3.58429 (* 1 = 3.58429 loss)
I0428 14:07:37.313357  7476 sgd_solver.cpp:105] Iteration 2316, lr = 0.00632063
I0428 14:07:42.194756  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:07:43.417845  7476 solver.cpp:218] Iteration 2328 (1.96582 iter/s, 6.10433s/12 iters), loss = 3.46984
I0428 14:07:43.417891  7476 solver.cpp:237]     Train net output #0: loss = 3.46984 (* 1 = 3.46984 loss)
I0428 14:07:43.417901  7476 sgd_solver.cpp:105] Iteration 2328, lr = 0.00630562
I0428 14:07:49.037860  7476 solver.cpp:218] Iteration 2340 (2.13531 iter/s, 5.61981s/12 iters), loss = 3.45641
I0428 14:07:49.037922  7476 solver.cpp:237]     Train net output #0: loss = 3.45641 (* 1 = 3.45641 loss)
I0428 14:07:49.037935  7476 sgd_solver.cpp:105] Iteration 2340, lr = 0.00629065
I0428 14:07:51.212919  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_2346.caffemodel
I0428 14:07:55.953689  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_2346.solverstate
I0428 14:07:58.034308  7476 solver.cpp:330] Iteration 2346, Testing net (#0)
I0428 14:07:58.034337  7476 net.cpp:676] Ignoring source layer train-data
I0428 14:08:01.765614  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:08:02.774991  7476 solver.cpp:397]     Test net output #0: accuracy = 0.132966
I0428 14:08:02.775022  7476 solver.cpp:397]     Test net output #1: loss = 3.68486 (* 1 = 3.68486 loss)
I0428 14:08:04.772190  7476 solver.cpp:218] Iteration 2352 (0.762687 iter/s, 15.7338s/12 iters), loss = 3.3425
I0428 14:08:04.772230  7476 solver.cpp:237]     Train net output #0: loss = 3.3425 (* 1 = 3.3425 loss)
I0428 14:08:04.772239  7476 sgd_solver.cpp:105] Iteration 2352, lr = 0.00627571
I0428 14:08:10.069648  7476 solver.cpp:218] Iteration 2364 (2.26532 iter/s, 5.29726s/12 iters), loss = 3.6161
I0428 14:08:10.069694  7476 solver.cpp:237]     Train net output #0: loss = 3.6161 (* 1 = 3.6161 loss)
I0428 14:08:10.069702  7476 sgd_solver.cpp:105] Iteration 2364, lr = 0.00626081
I0428 14:08:15.726099  7476 solver.cpp:218] Iteration 2376 (2.12155 iter/s, 5.65624s/12 iters), loss = 3.58882
I0428 14:08:15.726233  7476 solver.cpp:237]     Train net output #0: loss = 3.58882 (* 1 = 3.58882 loss)
I0428 14:08:15.726245  7476 sgd_solver.cpp:105] Iteration 2376, lr = 0.00624595
I0428 14:08:21.418459  7476 solver.cpp:218] Iteration 2388 (2.1082 iter/s, 5.69205s/12 iters), loss = 3.42479
I0428 14:08:21.418514  7476 solver.cpp:237]     Train net output #0: loss = 3.42479 (* 1 = 3.42479 loss)
I0428 14:08:21.418527  7476 sgd_solver.cpp:105] Iteration 2388, lr = 0.00623112
I0428 14:08:27.160455  7476 solver.cpp:218] Iteration 2400 (2.08994 iter/s, 5.74178s/12 iters), loss = 3.42432
I0428 14:08:27.160518  7476 solver.cpp:237]     Train net output #0: loss = 3.42432 (* 1 = 3.42432 loss)
I0428 14:08:27.160529  7476 sgd_solver.cpp:105] Iteration 2400, lr = 0.00621633
I0428 14:08:32.783201  7476 solver.cpp:218] Iteration 2412 (2.13428 iter/s, 5.62252s/12 iters), loss = 3.45007
I0428 14:08:32.783241  7476 solver.cpp:237]     Train net output #0: loss = 3.45007 (* 1 = 3.45007 loss)
I0428 14:08:32.783249  7476 sgd_solver.cpp:105] Iteration 2412, lr = 0.00620157
I0428 14:08:38.385607  7476 solver.cpp:218] Iteration 2424 (2.14202 iter/s, 5.6022s/12 iters), loss = 3.56894
I0428 14:08:38.385661  7476 solver.cpp:237]     Train net output #0: loss = 3.56894 (* 1 = 3.56894 loss)
I0428 14:08:38.385675  7476 sgd_solver.cpp:105] Iteration 2424, lr = 0.00618684
I0428 14:08:39.464880  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:08:44.307968  7476 solver.cpp:218] Iteration 2436 (2.0263 iter/s, 5.92213s/12 iters), loss = 3.581
I0428 14:08:44.308012  7476 solver.cpp:237]     Train net output #0: loss = 3.581 (* 1 = 3.581 loss)
I0428 14:08:44.308022  7476 sgd_solver.cpp:105] Iteration 2436, lr = 0.00617215
I0428 14:08:49.624701  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_2448.caffemodel
I0428 14:08:52.644361  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_2448.solverstate
I0428 14:08:57.439529  7476 solver.cpp:330] Iteration 2448, Testing net (#0)
I0428 14:08:57.439553  7476 net.cpp:676] Ignoring source layer train-data
I0428 14:09:00.889348  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:09:01.885638  7476 solver.cpp:397]     Test net output #0: accuracy = 0.136029
I0428 14:09:01.885676  7476 solver.cpp:397]     Test net output #1: loss = 3.57087 (* 1 = 3.57087 loss)
I0428 14:09:02.014745  7476 solver.cpp:218] Iteration 2448 (0.677727 iter/s, 17.7063s/12 iters), loss = 3.45147
I0428 14:09:02.014801  7476 solver.cpp:237]     Train net output #0: loss = 3.45147 (* 1 = 3.45147 loss)
I0428 14:09:02.014816  7476 sgd_solver.cpp:105] Iteration 2448, lr = 0.0061575
I0428 14:09:06.487594  7476 solver.cpp:218] Iteration 2460 (2.68297 iter/s, 4.47265s/12 iters), loss = 3.39437
I0428 14:09:06.487649  7476 solver.cpp:237]     Train net output #0: loss = 3.39437 (* 1 = 3.39437 loss)
I0428 14:09:06.487661  7476 sgd_solver.cpp:105] Iteration 2460, lr = 0.00614288
I0428 14:09:12.062831  7476 solver.cpp:218] Iteration 2472 (2.15246 iter/s, 5.57502s/12 iters), loss = 3.41315
I0428 14:09:12.062877  7476 solver.cpp:237]     Train net output #0: loss = 3.41315 (* 1 = 3.41315 loss)
I0428 14:09:12.062887  7476 sgd_solver.cpp:105] Iteration 2472, lr = 0.0061283
I0428 14:09:17.538195  7476 solver.cpp:218] Iteration 2484 (2.19172 iter/s, 5.47516s/12 iters), loss = 3.34061
I0428 14:09:17.538251  7476 solver.cpp:237]     Train net output #0: loss = 3.34061 (* 1 = 3.34061 loss)
I0428 14:09:17.538264  7476 sgd_solver.cpp:105] Iteration 2484, lr = 0.00611375
I0428 14:09:22.977356  7476 solver.cpp:218] Iteration 2496 (2.20631 iter/s, 5.43895s/12 iters), loss = 3.62
I0428 14:09:22.977488  7476 solver.cpp:237]     Train net output #0: loss = 3.62 (* 1 = 3.62 loss)
I0428 14:09:22.977504  7476 sgd_solver.cpp:105] Iteration 2496, lr = 0.00609923
I0428 14:09:28.550623  7476 solver.cpp:218] Iteration 2508 (2.15325 iter/s, 5.57297s/12 iters), loss = 3.45214
I0428 14:09:28.550684  7476 solver.cpp:237]     Train net output #0: loss = 3.45214 (* 1 = 3.45214 loss)
I0428 14:09:28.550695  7476 sgd_solver.cpp:105] Iteration 2508, lr = 0.00608475
I0428 14:09:34.475473  7476 solver.cpp:218] Iteration 2520 (2.02545 iter/s, 5.92462s/12 iters), loss = 3.58294
I0428 14:09:34.475533  7476 solver.cpp:237]     Train net output #0: loss = 3.58294 (* 1 = 3.58294 loss)
I0428 14:09:34.475546  7476 sgd_solver.cpp:105] Iteration 2520, lr = 0.0060703
I0428 14:09:37.998925  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:09:40.012112  7476 solver.cpp:218] Iteration 2532 (2.16746 iter/s, 5.53642s/12 iters), loss = 3.44722
I0428 14:09:40.012168  7476 solver.cpp:237]     Train net output #0: loss = 3.44722 (* 1 = 3.44722 loss)
I0428 14:09:40.012181  7476 sgd_solver.cpp:105] Iteration 2532, lr = 0.00605589
I0428 14:09:45.585038  7476 solver.cpp:218] Iteration 2544 (2.15335 iter/s, 5.57271s/12 iters), loss = 3.30557
I0428 14:09:45.585079  7476 solver.cpp:237]     Train net output #0: loss = 3.30557 (* 1 = 3.30557 loss)
I0428 14:09:45.585088  7476 sgd_solver.cpp:105] Iteration 2544, lr = 0.00604151
I0428 14:09:48.274344  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_2550.caffemodel
I0428 14:09:49.985224  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_2550.solverstate
I0428 14:09:53.335530  7476 solver.cpp:330] Iteration 2550, Testing net (#0)
I0428 14:09:53.335677  7476 net.cpp:676] Ignoring source layer train-data
I0428 14:09:56.819715  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:09:57.959890  7476 solver.cpp:397]     Test net output #0: accuracy = 0.152574
I0428 14:09:57.959921  7476 solver.cpp:397]     Test net output #1: loss = 3.47783 (* 1 = 3.47783 loss)
I0428 14:10:00.079627  7476 solver.cpp:218] Iteration 2556 (0.82792 iter/s, 14.4942s/12 iters), loss = 3.18464
I0428 14:10:00.079675  7476 solver.cpp:237]     Train net output #0: loss = 3.18464 (* 1 = 3.18464 loss)
I0428 14:10:00.079684  7476 sgd_solver.cpp:105] Iteration 2556, lr = 0.00602717
I0428 14:10:05.814074  7476 solver.cpp:218] Iteration 2568 (2.09269 iter/s, 5.73424s/12 iters), loss = 3.41342
I0428 14:10:05.814116  7476 solver.cpp:237]     Train net output #0: loss = 3.41342 (* 1 = 3.41342 loss)
I0428 14:10:05.814128  7476 sgd_solver.cpp:105] Iteration 2568, lr = 0.00601286
I0428 14:10:11.996381  7476 solver.cpp:218] Iteration 2580 (1.94109 iter/s, 6.18209s/12 iters), loss = 3.54919
I0428 14:10:11.996423  7476 solver.cpp:237]     Train net output #0: loss = 3.54919 (* 1 = 3.54919 loss)
I0428 14:10:11.996433  7476 sgd_solver.cpp:105] Iteration 2580, lr = 0.00599858
I0428 14:10:17.769460  7476 solver.cpp:218] Iteration 2592 (2.07869 iter/s, 5.77287s/12 iters), loss = 3.4741
I0428 14:10:17.769505  7476 solver.cpp:237]     Train net output #0: loss = 3.4741 (* 1 = 3.4741 loss)
I0428 14:10:17.769515  7476 sgd_solver.cpp:105] Iteration 2592, lr = 0.00598434
I0428 14:10:23.268654  7476 solver.cpp:218] Iteration 2604 (2.18222 iter/s, 5.49899s/12 iters), loss = 3.24853
I0428 14:10:23.268697  7476 solver.cpp:237]     Train net output #0: loss = 3.24853 (* 1 = 3.24853 loss)
I0428 14:10:23.268707  7476 sgd_solver.cpp:105] Iteration 2604, lr = 0.00597013
I0428 14:10:29.329169  7476 solver.cpp:218] Iteration 2616 (1.9801 iter/s, 6.0603s/12 iters), loss = 3.46618
I0428 14:10:29.358227  7476 solver.cpp:237]     Train net output #0: loss = 3.46618 (* 1 = 3.46618 loss)
I0428 14:10:29.358242  7476 sgd_solver.cpp:105] Iteration 2616, lr = 0.00595596
I0428 14:10:35.485077  7476 solver.cpp:218] Iteration 2628 (1.95864 iter/s, 6.12669s/12 iters), loss = 3.33886
I0428 14:10:35.485117  7476 solver.cpp:237]     Train net output #0: loss = 3.33886 (* 1 = 3.33886 loss)
I0428 14:10:35.485127  7476 sgd_solver.cpp:105] Iteration 2628, lr = 0.00594182
I0428 14:10:35.865221  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:10:41.186435  7476 solver.cpp:218] Iteration 2640 (2.10484 iter/s, 5.70116s/12 iters), loss = 3.12418
I0428 14:10:41.186478  7476 solver.cpp:237]     Train net output #0: loss = 3.12418 (* 1 = 3.12418 loss)
I0428 14:10:41.186488  7476 sgd_solver.cpp:105] Iteration 2640, lr = 0.00592771
I0428 14:10:46.098791  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_2652.caffemodel
I0428 14:10:48.508388  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_2652.solverstate
I0428 14:10:49.550055  7476 solver.cpp:330] Iteration 2652, Testing net (#0)
I0428 14:10:49.550078  7476 net.cpp:676] Ignoring source layer train-data
I0428 14:10:53.021136  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:10:54.101200  7476 solver.cpp:397]     Test net output #0: accuracy = 0.15625
I0428 14:10:54.101227  7476 solver.cpp:397]     Test net output #1: loss = 3.53929 (* 1 = 3.53929 loss)
I0428 14:10:54.229658  7476 solver.cpp:218] Iteration 2652 (0.920046 iter/s, 13.0428s/12 iters), loss = 3.28494
I0428 14:10:54.229707  7476 solver.cpp:237]     Train net output #0: loss = 3.28494 (* 1 = 3.28494 loss)
I0428 14:10:54.229715  7476 sgd_solver.cpp:105] Iteration 2652, lr = 0.00591364
I0428 14:10:58.828682  7476 solver.cpp:218] Iteration 2664 (2.60935 iter/s, 4.59884s/12 iters), loss = 3.20642
I0428 14:10:58.828732  7476 solver.cpp:237]     Train net output #0: loss = 3.20642 (* 1 = 3.20642 loss)
I0428 14:10:58.828745  7476 sgd_solver.cpp:105] Iteration 2664, lr = 0.0058996
I0428 14:11:04.371620  7476 solver.cpp:218] Iteration 2676 (2.165 iter/s, 5.54273s/12 iters), loss = 3.47323
I0428 14:11:04.371795  7476 solver.cpp:237]     Train net output #0: loss = 3.47323 (* 1 = 3.47323 loss)
I0428 14:11:04.371809  7476 sgd_solver.cpp:105] Iteration 2676, lr = 0.00588559
I0428 14:11:10.096113  7476 solver.cpp:218] Iteration 2688 (2.09638 iter/s, 5.72416s/12 iters), loss = 3.32075
I0428 14:11:10.096158  7476 solver.cpp:237]     Train net output #0: loss = 3.32075 (* 1 = 3.32075 loss)
I0428 14:11:10.096168  7476 sgd_solver.cpp:105] Iteration 2688, lr = 0.00587162
I0428 14:11:15.652333  7476 solver.cpp:218] Iteration 2700 (2.15982 iter/s, 5.55601s/12 iters), loss = 3.24298
I0428 14:11:15.652386  7476 solver.cpp:237]     Train net output #0: loss = 3.24298 (* 1 = 3.24298 loss)
I0428 14:11:15.652401  7476 sgd_solver.cpp:105] Iteration 2700, lr = 0.00585768
I0428 14:11:21.073735  7476 solver.cpp:218] Iteration 2712 (2.21353 iter/s, 5.4212s/12 iters), loss = 3.27762
I0428 14:11:21.073777  7476 solver.cpp:237]     Train net output #0: loss = 3.27762 (* 1 = 3.27762 loss)
I0428 14:11:21.073786  7476 sgd_solver.cpp:105] Iteration 2712, lr = 0.00584377
I0428 14:11:27.125679  7476 solver.cpp:218] Iteration 2724 (1.98291 iter/s, 6.05172s/12 iters), loss = 3.48884
I0428 14:11:27.125721  7476 solver.cpp:237]     Train net output #0: loss = 3.48884 (* 1 = 3.48884 loss)
I0428 14:11:27.125731  7476 sgd_solver.cpp:105] Iteration 2724, lr = 0.0058299
I0428 14:11:30.142401  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:11:33.483212  7476 solver.cpp:218] Iteration 2736 (1.88759 iter/s, 6.35731s/12 iters), loss = 3.2439
I0428 14:11:33.483256  7476 solver.cpp:237]     Train net output #0: loss = 3.2439 (* 1 = 3.2439 loss)
I0428 14:11:33.483264  7476 sgd_solver.cpp:105] Iteration 2736, lr = 0.00581605
I0428 14:11:38.995389  7476 solver.cpp:218] Iteration 2748 (2.17708 iter/s, 5.51197s/12 iters), loss = 3.15617
I0428 14:11:38.995491  7476 solver.cpp:237]     Train net output #0: loss = 3.15617 (* 1 = 3.15617 loss)
I0428 14:11:38.995501  7476 sgd_solver.cpp:105] Iteration 2748, lr = 0.00580225
I0428 14:11:41.738693  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_2754.caffemodel
I0428 14:11:43.240644  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_2754.solverstate
I0428 14:11:44.326087  7476 solver.cpp:330] Iteration 2754, Testing net (#0)
I0428 14:11:44.326105  7476 net.cpp:676] Ignoring source layer train-data
I0428 14:11:47.555851  7476 blocking_queue.cpp:49] Waiting for data
I0428 14:11:47.842465  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:11:48.970700  7476 solver.cpp:397]     Test net output #0: accuracy = 0.164216
I0428 14:11:48.970738  7476 solver.cpp:397]     Test net output #1: loss = 3.3983 (* 1 = 3.3983 loss)
I0428 14:11:51.080616  7476 solver.cpp:218] Iteration 2760 (0.992983 iter/s, 12.0848s/12 iters), loss = 3.30372
I0428 14:11:51.080677  7476 solver.cpp:237]     Train net output #0: loss = 3.30372 (* 1 = 3.30372 loss)
I0428 14:11:51.080689  7476 sgd_solver.cpp:105] Iteration 2760, lr = 0.00578847
I0428 14:11:56.645220  7476 solver.cpp:218] Iteration 2772 (2.15657 iter/s, 5.56438s/12 iters), loss = 3.11125
I0428 14:11:56.645265  7476 solver.cpp:237]     Train net output #0: loss = 3.11125 (* 1 = 3.11125 loss)
I0428 14:11:56.645274  7476 sgd_solver.cpp:105] Iteration 2772, lr = 0.00577473
I0428 14:12:02.235333  7476 solver.cpp:218] Iteration 2784 (2.14672 iter/s, 5.58991s/12 iters), loss = 3.11609
I0428 14:12:02.235371  7476 solver.cpp:237]     Train net output #0: loss = 3.11609 (* 1 = 3.11609 loss)
I0428 14:12:02.235378  7476 sgd_solver.cpp:105] Iteration 2784, lr = 0.00576102
I0428 14:12:08.162356  7476 solver.cpp:218] Iteration 2796 (2.0247 iter/s, 5.92681s/12 iters), loss = 3.25915
I0428 14:12:08.162396  7476 solver.cpp:237]     Train net output #0: loss = 3.25915 (* 1 = 3.25915 loss)
I0428 14:12:08.162405  7476 sgd_solver.cpp:105] Iteration 2796, lr = 0.00574734
I0428 14:12:13.738864  7476 solver.cpp:218] Iteration 2808 (2.15196 iter/s, 5.57631s/12 iters), loss = 3.20172
I0428 14:12:13.739022  7476 solver.cpp:237]     Train net output #0: loss = 3.20172 (* 1 = 3.20172 loss)
I0428 14:12:13.739032  7476 sgd_solver.cpp:105] Iteration 2808, lr = 0.00573369
I0428 14:12:19.461751  7476 solver.cpp:218] Iteration 2820 (2.09696 iter/s, 5.72256s/12 iters), loss = 2.97079
I0428 14:12:19.461812  7476 solver.cpp:237]     Train net output #0: loss = 2.97079 (* 1 = 2.97079 loss)
I0428 14:12:19.461824  7476 sgd_solver.cpp:105] Iteration 2820, lr = 0.00572008
I0428 14:12:24.714339  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:12:25.044463  7476 solver.cpp:218] Iteration 2832 (2.14958 iter/s, 5.58249s/12 iters), loss = 3.0759
I0428 14:12:25.044536  7476 solver.cpp:237]     Train net output #0: loss = 3.0759 (* 1 = 3.0759 loss)
I0428 14:12:25.044545  7476 sgd_solver.cpp:105] Iteration 2832, lr = 0.0057065
I0428 14:12:30.641556  7476 solver.cpp:218] Iteration 2844 (2.14406 iter/s, 5.59686s/12 iters), loss = 3.06206
I0428 14:12:30.641594  7476 solver.cpp:237]     Train net output #0: loss = 3.06206 (* 1 = 3.06206 loss)
I0428 14:12:30.641603  7476 sgd_solver.cpp:105] Iteration 2844, lr = 0.00569295
I0428 14:12:35.641263  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_2856.caffemodel
I0428 14:12:37.012217  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_2856.solverstate
I0428 14:12:38.048163  7476 solver.cpp:330] Iteration 2856, Testing net (#0)
I0428 14:12:38.048183  7476 net.cpp:676] Ignoring source layer train-data
I0428 14:12:41.365267  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:12:42.569605  7476 solver.cpp:397]     Test net output #0: accuracy = 0.169118
I0428 14:12:42.569659  7476 solver.cpp:397]     Test net output #1: loss = 3.3809 (* 1 = 3.3809 loss)
I0428 14:12:42.691416  7476 solver.cpp:218] Iteration 2856 (0.995892 iter/s, 12.0495s/12 iters), loss = 2.99857
I0428 14:12:42.691460  7476 solver.cpp:237]     Train net output #0: loss = 2.99857 (* 1 = 2.99857 loss)
I0428 14:12:42.691470  7476 sgd_solver.cpp:105] Iteration 2856, lr = 0.00567944
I0428 14:12:47.327319  7476 solver.cpp:218] Iteration 2868 (2.58859 iter/s, 4.63572s/12 iters), loss = 3.17238
I0428 14:12:47.327461  7476 solver.cpp:237]     Train net output #0: loss = 3.17238 (* 1 = 3.17238 loss)
I0428 14:12:47.327471  7476 sgd_solver.cpp:105] Iteration 2868, lr = 0.00566595
I0428 14:12:53.217757  7476 solver.cpp:218] Iteration 2880 (2.03731 iter/s, 5.89013s/12 iters), loss = 2.9863
I0428 14:12:53.217797  7476 solver.cpp:237]     Train net output #0: loss = 2.9863 (* 1 = 2.9863 loss)
I0428 14:12:53.217806  7476 sgd_solver.cpp:105] Iteration 2880, lr = 0.0056525
I0428 14:12:59.257223  7476 solver.cpp:218] Iteration 2892 (1.987 iter/s, 6.03925s/12 iters), loss = 3.09402
I0428 14:12:59.257277  7476 solver.cpp:237]     Train net output #0: loss = 3.09402 (* 1 = 3.09402 loss)
I0428 14:12:59.257287  7476 sgd_solver.cpp:105] Iteration 2892, lr = 0.00563908
I0428 14:13:04.632282  7476 solver.cpp:218] Iteration 2904 (2.23262 iter/s, 5.37485s/12 iters), loss = 2.98272
I0428 14:13:04.632326  7476 solver.cpp:237]     Train net output #0: loss = 2.98272 (* 1 = 2.98272 loss)
I0428 14:13:04.632337  7476 sgd_solver.cpp:105] Iteration 2904, lr = 0.00562569
I0428 14:13:10.278460  7476 solver.cpp:218] Iteration 2916 (2.12541 iter/s, 5.64597s/12 iters), loss = 3.1447
I0428 14:13:10.278498  7476 solver.cpp:237]     Train net output #0: loss = 3.1447 (* 1 = 3.1447 loss)
I0428 14:13:10.278507  7476 sgd_solver.cpp:105] Iteration 2916, lr = 0.00561233
I0428 14:13:16.210739  7476 solver.cpp:218] Iteration 2928 (2.0229 iter/s, 5.93207s/12 iters), loss = 2.78951
I0428 14:13:16.210783  7476 solver.cpp:237]     Train net output #0: loss = 2.78951 (* 1 = 2.78951 loss)
I0428 14:13:16.210791  7476 sgd_solver.cpp:105] Iteration 2928, lr = 0.00559901
I0428 14:13:18.333344  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:13:21.801069  7476 solver.cpp:218] Iteration 2940 (2.14664 iter/s, 5.59013s/12 iters), loss = 2.93893
I0428 14:13:21.801105  7476 solver.cpp:237]     Train net output #0: loss = 2.93893 (* 1 = 2.93893 loss)
I0428 14:13:21.801115  7476 sgd_solver.cpp:105] Iteration 2940, lr = 0.00558572
I0428 14:13:27.590847  7476 solver.cpp:218] Iteration 2952 (2.07269 iter/s, 5.78958s/12 iters), loss = 2.78595
I0428 14:13:27.590895  7476 solver.cpp:237]     Train net output #0: loss = 2.78595 (* 1 = 2.78595 loss)
I0428 14:13:27.590906  7476 sgd_solver.cpp:105] Iteration 2952, lr = 0.00557245
I0428 14:13:29.903000  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_2958.caffemodel
I0428 14:13:31.245594  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_2958.solverstate
I0428 14:13:32.323348  7476 solver.cpp:330] Iteration 2958, Testing net (#0)
I0428 14:13:32.323367  7476 net.cpp:676] Ignoring source layer train-data
I0428 14:13:35.643721  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:13:37.086846  7476 solver.cpp:397]     Test net output #0: accuracy = 0.164828
I0428 14:13:37.086879  7476 solver.cpp:397]     Test net output #1: loss = 3.42206 (* 1 = 3.42206 loss)
I0428 14:13:39.267627  7476 solver.cpp:218] Iteration 2964 (1.02771 iter/s, 11.6764s/12 iters), loss = 2.9501
I0428 14:13:39.267674  7476 solver.cpp:237]     Train net output #0: loss = 2.9501 (* 1 = 2.9501 loss)
I0428 14:13:39.267684  7476 sgd_solver.cpp:105] Iteration 2964, lr = 0.00555922
I0428 14:13:44.947906  7476 solver.cpp:218] Iteration 2976 (2.11265 iter/s, 5.68007s/12 iters), loss = 2.99181
I0428 14:13:44.947948  7476 solver.cpp:237]     Train net output #0: loss = 2.99181 (* 1 = 2.99181 loss)
I0428 14:13:44.947957  7476 sgd_solver.cpp:105] Iteration 2976, lr = 0.00554603
I0428 14:13:50.593045  7476 solver.cpp:218] Iteration 2988 (2.1258 iter/s, 5.64493s/12 iters), loss = 2.94579
I0428 14:13:50.593365  7476 solver.cpp:237]     Train net output #0: loss = 2.94579 (* 1 = 2.94579 loss)
I0428 14:13:50.593376  7476 sgd_solver.cpp:105] Iteration 2988, lr = 0.00553286
I0428 14:13:55.959738  7476 solver.cpp:218] Iteration 3000 (2.23621 iter/s, 5.36622s/12 iters), loss = 2.91025
I0428 14:13:55.959789  7476 solver.cpp:237]     Train net output #0: loss = 2.91025 (* 1 = 2.91025 loss)
I0428 14:13:55.959798  7476 sgd_solver.cpp:105] Iteration 3000, lr = 0.00551972
I0428 14:14:02.796864  7476 solver.cpp:218] Iteration 3012 (1.75519 iter/s, 6.83688s/12 iters), loss = 2.99627
I0428 14:14:02.796907  7476 solver.cpp:237]     Train net output #0: loss = 2.99627 (* 1 = 2.99627 loss)
I0428 14:14:02.796917  7476 sgd_solver.cpp:105] Iteration 3012, lr = 0.00550662
I0428 14:14:08.458165  7476 solver.cpp:218] Iteration 3024 (2.11973 iter/s, 5.66109s/12 iters), loss = 3.00463
I0428 14:14:08.458205  7476 solver.cpp:237]     Train net output #0: loss = 3.00463 (* 1 = 3.00463 loss)
I0428 14:14:08.458214  7476 sgd_solver.cpp:105] Iteration 3024, lr = 0.00549354
I0428 14:14:12.839341  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:14:14.077761  7476 solver.cpp:218] Iteration 3036 (2.13546 iter/s, 5.61939s/12 iters), loss = 2.85586
I0428 14:14:14.077808  7476 solver.cpp:237]     Train net output #0: loss = 2.85586 (* 1 = 2.85586 loss)
I0428 14:14:14.077818  7476 sgd_solver.cpp:105] Iteration 3036, lr = 0.0054805
I0428 14:14:19.923085  7476 solver.cpp:218] Iteration 3048 (2.053 iter/s, 5.84511s/12 iters), loss = 2.9691
I0428 14:14:19.923130  7476 solver.cpp:237]     Train net output #0: loss = 2.9691 (* 1 = 2.9691 loss)
I0428 14:14:19.923139  7476 sgd_solver.cpp:105] Iteration 3048, lr = 0.00546749
I0428 14:14:25.803489  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_3060.caffemodel
I0428 14:14:28.718765  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_3060.solverstate
I0428 14:14:29.920976  7476 solver.cpp:330] Iteration 3060, Testing net (#0)
I0428 14:14:29.920998  7476 net.cpp:676] Ignoring source layer train-data
I0428 14:14:33.263949  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:14:34.504209  7476 solver.cpp:397]     Test net output #0: accuracy = 0.191789
I0428 14:14:34.504237  7476 solver.cpp:397]     Test net output #1: loss = 3.27531 (* 1 = 3.27531 loss)
I0428 14:14:34.632673  7476 solver.cpp:218] Iteration 3060 (0.815819 iter/s, 14.7092s/12 iters), loss = 2.67056
I0428 14:14:34.632725  7476 solver.cpp:237]     Train net output #0: loss = 2.67056 (* 1 = 2.67056 loss)
I0428 14:14:34.632740  7476 sgd_solver.cpp:105] Iteration 3060, lr = 0.00545451
I0428 14:14:39.464394  7476 solver.cpp:218] Iteration 3072 (2.48369 iter/s, 4.83153s/12 iters), loss = 3.03031
I0428 14:14:39.464433  7476 solver.cpp:237]     Train net output #0: loss = 3.03031 (* 1 = 3.03031 loss)
I0428 14:14:39.464442  7476 sgd_solver.cpp:105] Iteration 3072, lr = 0.00544156
I0428 14:14:45.110939  7476 solver.cpp:218] Iteration 3084 (2.12527 iter/s, 5.64634s/12 iters), loss = 2.99343
I0428 14:14:45.110982  7476 solver.cpp:237]     Train net output #0: loss = 2.99343 (* 1 = 2.99343 loss)
I0428 14:14:45.110991  7476 sgd_solver.cpp:105] Iteration 3084, lr = 0.00542864
I0428 14:14:50.666977  7476 solver.cpp:218] Iteration 3096 (2.15989 iter/s, 5.55583s/12 iters), loss = 2.88738
I0428 14:14:50.667018  7476 solver.cpp:237]     Train net output #0: loss = 2.88738 (* 1 = 2.88738 loss)
I0428 14:14:50.667029  7476 sgd_solver.cpp:105] Iteration 3096, lr = 0.00541575
I0428 14:14:56.496737  7476 solver.cpp:218] Iteration 3108 (2.05848 iter/s, 5.82955s/12 iters), loss = 2.89725
I0428 14:14:56.497017  7476 solver.cpp:237]     Train net output #0: loss = 2.89725 (* 1 = 2.89725 loss)
I0428 14:14:56.497027  7476 sgd_solver.cpp:105] Iteration 3108, lr = 0.00540289
I0428 14:15:02.009027  7476 solver.cpp:218] Iteration 3120 (2.17713 iter/s, 5.51185s/12 iters), loss = 2.89025
I0428 14:15:02.009069  7476 solver.cpp:237]     Train net output #0: loss = 2.89025 (* 1 = 2.89025 loss)
I0428 14:15:02.009080  7476 sgd_solver.cpp:105] Iteration 3120, lr = 0.00539006
I0428 14:15:07.905395  7476 solver.cpp:218] Iteration 3132 (2.03522 iter/s, 5.89616s/12 iters), loss = 2.87307
I0428 14:15:07.905436  7476 solver.cpp:237]     Train net output #0: loss = 2.87307 (* 1 = 2.87307 loss)
I0428 14:15:07.905443  7476 sgd_solver.cpp:105] Iteration 3132, lr = 0.00537727
I0428 14:15:09.254112  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:15:14.018689  7476 solver.cpp:218] Iteration 3144 (1.963 iter/s, 6.11308s/12 iters), loss = 2.9868
I0428 14:15:14.018741  7476 solver.cpp:237]     Train net output #0: loss = 2.9868 (* 1 = 2.9868 loss)
I0428 14:15:14.018752  7476 sgd_solver.cpp:105] Iteration 3144, lr = 0.0053645
I0428 14:15:19.831852  7476 solver.cpp:218] Iteration 3156 (2.06436 iter/s, 5.81294s/12 iters), loss = 2.8822
I0428 14:15:19.831903  7476 solver.cpp:237]     Train net output #0: loss = 2.8822 (* 1 = 2.8822 loss)
I0428 14:15:19.831914  7476 sgd_solver.cpp:105] Iteration 3156, lr = 0.00535176
I0428 14:15:22.074038  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_3162.caffemodel
I0428 14:15:26.506274  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_3162.solverstate
I0428 14:15:28.099023  7476 solver.cpp:330] Iteration 3162, Testing net (#0)
I0428 14:15:28.099043  7476 net.cpp:676] Ignoring source layer train-data
I0428 14:15:31.392875  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:15:32.738314  7476 solver.cpp:397]     Test net output #0: accuracy = 0.200368
I0428 14:15:32.738340  7476 solver.cpp:397]     Test net output #1: loss = 3.23124 (* 1 = 3.23124 loss)
I0428 14:15:35.181993  7476 solver.cpp:218] Iteration 3168 (0.781775 iter/s, 15.3497s/12 iters), loss = 2.75352
I0428 14:15:35.182046  7476 solver.cpp:237]     Train net output #0: loss = 2.75352 (* 1 = 2.75352 loss)
I0428 14:15:35.182057  7476 sgd_solver.cpp:105] Iteration 3168, lr = 0.00533906
I0428 14:15:40.855450  7476 solver.cpp:218] Iteration 3180 (2.1152 iter/s, 5.67323s/12 iters), loss = 2.894
I0428 14:15:40.855509  7476 solver.cpp:237]     Train net output #0: loss = 2.894 (* 1 = 2.894 loss)
I0428 14:15:40.855522  7476 sgd_solver.cpp:105] Iteration 3180, lr = 0.00532638
I0428 14:15:46.344810  7476 solver.cpp:218] Iteration 3192 (2.18613 iter/s, 5.48915s/12 iters), loss = 2.8371
I0428 14:15:46.344852  7476 solver.cpp:237]     Train net output #0: loss = 2.8371 (* 1 = 2.8371 loss)
I0428 14:15:46.344861  7476 sgd_solver.cpp:105] Iteration 3192, lr = 0.00531374
I0428 14:15:52.189472  7476 solver.cpp:218] Iteration 3204 (2.05323 iter/s, 5.84446s/12 iters), loss = 2.92952
I0428 14:15:52.189504  7476 solver.cpp:237]     Train net output #0: loss = 2.92952 (* 1 = 2.92952 loss)
I0428 14:15:52.189513  7476 sgd_solver.cpp:105] Iteration 3204, lr = 0.00530112
I0428 14:15:57.711432  7476 solver.cpp:218] Iteration 3216 (2.17322 iter/s, 5.52176s/12 iters), loss = 2.50905
I0428 14:15:57.711706  7476 solver.cpp:237]     Train net output #0: loss = 2.50905 (* 1 = 2.50905 loss)
I0428 14:15:57.711716  7476 sgd_solver.cpp:105] Iteration 3216, lr = 0.00528853
I0428 14:16:03.250566  7476 solver.cpp:218] Iteration 3228 (2.16657 iter/s, 5.5387s/12 iters), loss = 2.79706
I0428 14:16:03.250619  7476 solver.cpp:237]     Train net output #0: loss = 2.79706 (* 1 = 2.79706 loss)
I0428 14:16:03.250631  7476 sgd_solver.cpp:105] Iteration 3228, lr = 0.00527598
I0428 14:16:07.052465  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:16:09.064675  7476 solver.cpp:218] Iteration 3240 (2.06402 iter/s, 5.81389s/12 iters), loss = 2.75921
I0428 14:16:09.064735  7476 solver.cpp:237]     Train net output #0: loss = 2.75921 (* 1 = 2.75921 loss)
I0428 14:16:09.064747  7476 sgd_solver.cpp:105] Iteration 3240, lr = 0.00526345
I0428 14:16:15.143828  7476 solver.cpp:218] Iteration 3252 (1.97403 iter/s, 6.07892s/12 iters), loss = 2.71385
I0428 14:16:15.143875  7476 solver.cpp:237]     Train net output #0: loss = 2.71385 (* 1 = 2.71385 loss)
I0428 14:16:15.143884  7476 sgd_solver.cpp:105] Iteration 3252, lr = 0.00525095
I0428 14:16:20.183151  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_3264.caffemodel
I0428 14:16:25.088500  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_3264.solverstate
I0428 14:16:27.342362  7476 solver.cpp:330] Iteration 3264, Testing net (#0)
I0428 14:16:27.342386  7476 net.cpp:676] Ignoring source layer train-data
I0428 14:16:30.690245  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:16:32.010246  7476 solver.cpp:397]     Test net output #0: accuracy = 0.215074
I0428 14:16:32.010282  7476 solver.cpp:397]     Test net output #1: loss = 3.20425 (* 1 = 3.20425 loss)
I0428 14:16:32.138662  7476 solver.cpp:218] Iteration 3264 (0.706118 iter/s, 16.9943s/12 iters), loss = 2.84897
I0428 14:16:32.138705  7476 solver.cpp:237]     Train net output #0: loss = 2.84897 (* 1 = 2.84897 loss)
I0428 14:16:32.138715  7476 sgd_solver.cpp:105] Iteration 3264, lr = 0.00523849
I0428 14:16:36.632107  7476 solver.cpp:218] Iteration 3276 (2.67066 iter/s, 4.49327s/12 iters), loss = 2.93697
I0428 14:16:36.632148  7476 solver.cpp:237]     Train net output #0: loss = 2.93697 (* 1 = 2.93697 loss)
I0428 14:16:36.632158  7476 sgd_solver.cpp:105] Iteration 3276, lr = 0.00522605
I0428 14:16:42.040859  7476 solver.cpp:218] Iteration 3288 (2.21871 iter/s, 5.40856s/12 iters), loss = 2.8934
I0428 14:16:42.040899  7476 solver.cpp:237]     Train net output #0: loss = 2.8934 (* 1 = 2.8934 loss)
I0428 14:16:42.040906  7476 sgd_solver.cpp:105] Iteration 3288, lr = 0.00521364
I0428 14:16:47.651818  7476 solver.cpp:218] Iteration 3300 (2.13875 iter/s, 5.61076s/12 iters), loss = 2.82139
I0428 14:16:47.651859  7476 solver.cpp:237]     Train net output #0: loss = 2.82139 (* 1 = 2.82139 loss)
I0428 14:16:47.651867  7476 sgd_solver.cpp:105] Iteration 3300, lr = 0.00520126
I0428 14:16:53.228139  7476 solver.cpp:218] Iteration 3312 (2.15203 iter/s, 5.57612s/12 iters), loss = 2.67762
I0428 14:16:53.228183  7476 solver.cpp:237]     Train net output #0: loss = 2.67762 (* 1 = 2.67762 loss)
I0428 14:16:53.228191  7476 sgd_solver.cpp:105] Iteration 3312, lr = 0.00518892
I0428 14:16:58.683789  7476 solver.cpp:218] Iteration 3324 (2.19963 iter/s, 5.45545s/12 iters), loss = 2.73294
I0428 14:16:58.683830  7476 solver.cpp:237]     Train net output #0: loss = 2.73294 (* 1 = 2.73294 loss)
I0428 14:16:58.683840  7476 sgd_solver.cpp:105] Iteration 3324, lr = 0.0051766
I0428 14:17:04.347038  7476 solver.cpp:218] Iteration 3336 (2.119 iter/s, 5.66304s/12 iters), loss = 2.68734
I0428 14:17:04.347142  7476 solver.cpp:237]     Train net output #0: loss = 2.68734 (* 1 = 2.68734 loss)
I0428 14:17:04.347152  7476 sgd_solver.cpp:105] Iteration 3336, lr = 0.00516431
I0428 14:17:04.858606  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:17:10.041596  7476 solver.cpp:218] Iteration 3348 (2.10738 iter/s, 5.69428s/12 iters), loss = 2.83829
I0428 14:17:10.041652  7476 solver.cpp:237]     Train net output #0: loss = 2.83829 (* 1 = 2.83829 loss)
I0428 14:17:10.041664  7476 sgd_solver.cpp:105] Iteration 3348, lr = 0.00515204
I0428 14:17:16.043716  7476 solver.cpp:218] Iteration 3360 (1.99937 iter/s, 6.0019s/12 iters), loss = 2.65186
I0428 14:17:16.043761  7476 solver.cpp:237]     Train net output #0: loss = 2.65186 (* 1 = 2.65186 loss)
I0428 14:17:16.043771  7476 sgd_solver.cpp:105] Iteration 3360, lr = 0.00513981
I0428 14:17:18.269393  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_3366.caffemodel
I0428 14:17:21.364147  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_3366.solverstate
I0428 14:17:23.271162  7476 solver.cpp:330] Iteration 3366, Testing net (#0)
I0428 14:17:23.271184  7476 net.cpp:676] Ignoring source layer train-data
I0428 14:17:26.399407  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:17:27.756150  7476 solver.cpp:397]     Test net output #0: accuracy = 0.22549
I0428 14:17:27.756183  7476 solver.cpp:397]     Test net output #1: loss = 3.05619 (* 1 = 3.05619 loss)
I0428 14:17:29.718238  7476 solver.cpp:218] Iteration 3372 (0.877571 iter/s, 13.6741s/12 iters), loss = 2.5192
I0428 14:17:29.718286  7476 solver.cpp:237]     Train net output #0: loss = 2.5192 (* 1 = 2.5192 loss)
I0428 14:17:29.718298  7476 sgd_solver.cpp:105] Iteration 3372, lr = 0.00512761
I0428 14:17:35.246827  7476 solver.cpp:218] Iteration 3384 (2.17062 iter/s, 5.52838s/12 iters), loss = 2.74471
I0428 14:17:35.246989  7476 solver.cpp:237]     Train net output #0: loss = 2.74471 (* 1 = 2.74471 loss)
I0428 14:17:35.247004  7476 sgd_solver.cpp:105] Iteration 3384, lr = 0.00511544
I0428 14:17:41.074383  7476 solver.cpp:218] Iteration 3396 (2.0593 iter/s, 5.82724s/12 iters), loss = 2.50562
I0428 14:17:41.074425  7476 solver.cpp:237]     Train net output #0: loss = 2.50562 (* 1 = 2.50562 loss)
I0428 14:17:41.074432  7476 sgd_solver.cpp:105] Iteration 3396, lr = 0.00510329
I0428 14:17:46.702212  7476 solver.cpp:218] Iteration 3408 (2.13234 iter/s, 5.62763s/12 iters), loss = 2.60152
I0428 14:17:46.702253  7476 solver.cpp:237]     Train net output #0: loss = 2.60152 (* 1 = 2.60152 loss)
I0428 14:17:46.702262  7476 sgd_solver.cpp:105] Iteration 3408, lr = 0.00509117
I0428 14:17:52.172895  7476 solver.cpp:218] Iteration 3420 (2.19359 iter/s, 5.47049s/12 iters), loss = 2.58673
I0428 14:17:52.172937  7476 solver.cpp:237]     Train net output #0: loss = 2.58673 (* 1 = 2.58673 loss)
I0428 14:17:52.172946  7476 sgd_solver.cpp:105] Iteration 3420, lr = 0.00507909
I0428 14:17:57.957952  7476 solver.cpp:218] Iteration 3432 (2.07438 iter/s, 5.78485s/12 iters), loss = 2.61761
I0428 14:17:57.957994  7476 solver.cpp:237]     Train net output #0: loss = 2.61761 (* 1 = 2.61761 loss)
I0428 14:17:57.958004  7476 sgd_solver.cpp:105] Iteration 3432, lr = 0.00506703
I0428 14:18:00.792186  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:18:03.690328  7476 solver.cpp:218] Iteration 3444 (2.09345 iter/s, 5.73217s/12 iters), loss = 2.60603
I0428 14:18:03.690369  7476 solver.cpp:237]     Train net output #0: loss = 2.60603 (* 1 = 2.60603 loss)
I0428 14:18:03.690379  7476 sgd_solver.cpp:105] Iteration 3444, lr = 0.005055
I0428 14:18:09.359342  7476 solver.cpp:218] Iteration 3456 (2.11685 iter/s, 5.66881s/12 iters), loss = 2.58802
I0428 14:18:09.359468  7476 solver.cpp:237]     Train net output #0: loss = 2.58802 (* 1 = 2.58802 loss)
I0428 14:18:09.359477  7476 sgd_solver.cpp:105] Iteration 3456, lr = 0.005043
I0428 14:18:14.225555  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_3468.caffemodel
I0428 14:18:15.635128  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_3468.solverstate
I0428 14:18:17.752183  7476 solver.cpp:330] Iteration 3468, Testing net (#0)
I0428 14:18:17.752203  7476 net.cpp:676] Ignoring source layer train-data
I0428 14:18:18.182624  7476 blocking_queue.cpp:49] Waiting for data
I0428 14:18:20.838917  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:18:22.243580  7476 solver.cpp:397]     Test net output #0: accuracy = 0.230392
I0428 14:18:22.243614  7476 solver.cpp:397]     Test net output #1: loss = 3.07701 (* 1 = 3.07701 loss)
I0428 14:18:22.371789  7476 solver.cpp:218] Iteration 3468 (0.922228 iter/s, 13.012s/12 iters), loss = 2.45174
I0428 14:18:22.371845  7476 solver.cpp:237]     Train net output #0: loss = 2.45174 (* 1 = 2.45174 loss)
I0428 14:18:22.371857  7476 sgd_solver.cpp:105] Iteration 3468, lr = 0.00503102
I0428 14:18:27.041863  7476 solver.cpp:218] Iteration 3480 (2.56966 iter/s, 4.66989s/12 iters), loss = 2.44838
I0428 14:18:27.041901  7476 solver.cpp:237]     Train net output #0: loss = 2.44838 (* 1 = 2.44838 loss)
I0428 14:18:27.041909  7476 sgd_solver.cpp:105] Iteration 3480, lr = 0.00501908
I0428 14:18:32.733996  7476 solver.cpp:218] Iteration 3492 (2.10825 iter/s, 5.69191s/12 iters), loss = 2.74162
I0428 14:18:32.734040  7476 solver.cpp:237]     Train net output #0: loss = 2.74162 (* 1 = 2.74162 loss)
I0428 14:18:32.734050  7476 sgd_solver.cpp:105] Iteration 3492, lr = 0.00500716
I0428 14:18:38.385175  7476 solver.cpp:218] Iteration 3504 (2.12353 iter/s, 5.65097s/12 iters), loss = 2.58556
I0428 14:18:38.385218  7476 solver.cpp:237]     Train net output #0: loss = 2.58556 (* 1 = 2.58556 loss)
I0428 14:18:38.385226  7476 sgd_solver.cpp:105] Iteration 3504, lr = 0.00499527
I0428 14:18:44.017272  7476 solver.cpp:218] Iteration 3516 (2.13072 iter/s, 5.63189s/12 iters), loss = 2.59476
I0428 14:18:44.017402  7476 solver.cpp:237]     Train net output #0: loss = 2.59476 (* 1 = 2.59476 loss)
I0428 14:18:44.017412  7476 sgd_solver.cpp:105] Iteration 3516, lr = 0.00498341
I0428 14:18:49.497889  7476 solver.cpp:218] Iteration 3528 (2.18965 iter/s, 5.48033s/12 iters), loss = 2.41795
I0428 14:18:49.497944  7476 solver.cpp:237]     Train net output #0: loss = 2.41795 (* 1 = 2.41795 loss)
I0428 14:18:49.497956  7476 sgd_solver.cpp:105] Iteration 3528, lr = 0.00497158
I0428 14:18:54.766196  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:18:55.093964  7476 solver.cpp:218] Iteration 3540 (2.14444 iter/s, 5.59586s/12 iters), loss = 2.44386
I0428 14:18:55.094022  7476 solver.cpp:237]     Train net output #0: loss = 2.44386 (* 1 = 2.44386 loss)
I0428 14:18:55.094033  7476 sgd_solver.cpp:105] Iteration 3540, lr = 0.00495978
I0428 14:19:00.591962  7476 solver.cpp:218] Iteration 3552 (2.1827 iter/s, 5.49778s/12 iters), loss = 2.50439
I0428 14:19:00.592020  7476 solver.cpp:237]     Train net output #0: loss = 2.50439 (* 1 = 2.50439 loss)
I0428 14:19:00.592033  7476 sgd_solver.cpp:105] Iteration 3552, lr = 0.004948
I0428 14:19:06.346355  7476 solver.cpp:218] Iteration 3564 (2.08544 iter/s, 5.75417s/12 iters), loss = 2.40882
I0428 14:19:06.346395  7476 solver.cpp:237]     Train net output #0: loss = 2.40882 (* 1 = 2.40882 loss)
I0428 14:19:06.346402  7476 sgd_solver.cpp:105] Iteration 3564, lr = 0.00493626
I0428 14:19:08.564191  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_3570.caffemodel
I0428 14:19:10.123481  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_3570.solverstate
I0428 14:19:11.153622  7476 solver.cpp:330] Iteration 3570, Testing net (#0)
I0428 14:19:11.153647  7476 net.cpp:676] Ignoring source layer train-data
I0428 14:19:14.455675  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:19:15.968708  7476 solver.cpp:397]     Test net output #0: accuracy = 0.214461
I0428 14:19:15.968740  7476 solver.cpp:397]     Test net output #1: loss = 3.25528 (* 1 = 3.25528 loss)
I0428 14:19:18.847203  7476 solver.cpp:218] Iteration 3576 (0.959964 iter/s, 12.5005s/12 iters), loss = 2.33079
I0428 14:19:18.847237  7476 solver.cpp:237]     Train net output #0: loss = 2.33079 (* 1 = 2.33079 loss)
I0428 14:19:18.847245  7476 sgd_solver.cpp:105] Iteration 3576, lr = 0.00492454
I0428 14:19:24.435199  7476 solver.cpp:218] Iteration 3588 (2.14754 iter/s, 5.58779s/12 iters), loss = 2.32444
I0428 14:19:24.435238  7476 solver.cpp:237]     Train net output #0: loss = 2.32444 (* 1 = 2.32444 loss)
I0428 14:19:24.435248  7476 sgd_solver.cpp:105] Iteration 3588, lr = 0.00491284
I0428 14:19:29.939913  7476 solver.cpp:218] Iteration 3600 (2.18003 iter/s, 5.50452s/12 iters), loss = 2.64329
I0428 14:19:29.939951  7476 solver.cpp:237]     Train net output #0: loss = 2.64329 (* 1 = 2.64329 loss)
I0428 14:19:29.939960  7476 sgd_solver.cpp:105] Iteration 3600, lr = 0.00490118
I0428 14:19:35.317169  7476 solver.cpp:218] Iteration 3612 (2.2317 iter/s, 5.37706s/12 iters), loss = 2.42847
I0428 14:19:35.317211  7476 solver.cpp:237]     Train net output #0: loss = 2.42847 (* 1 = 2.42847 loss)
I0428 14:19:35.317221  7476 sgd_solver.cpp:105] Iteration 3612, lr = 0.00488954
I0428 14:19:40.866307  7476 solver.cpp:218] Iteration 3624 (2.16258 iter/s, 5.54893s/12 iters), loss = 2.46311
I0428 14:19:40.866351  7476 solver.cpp:237]     Train net output #0: loss = 2.46311 (* 1 = 2.46311 loss)
I0428 14:19:40.866362  7476 sgd_solver.cpp:105] Iteration 3624, lr = 0.00487793
I0428 14:19:46.496649  7476 solver.cpp:218] Iteration 3636 (2.13139 iter/s, 5.63012s/12 iters), loss = 2.07489
I0428 14:19:46.496779  7476 solver.cpp:237]     Train net output #0: loss = 2.07489 (* 1 = 2.07489 loss)
I0428 14:19:46.496789  7476 sgd_solver.cpp:105] Iteration 3636, lr = 0.00486635
I0428 14:19:48.651829  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:19:52.168531  7476 solver.cpp:218] Iteration 3648 (2.11581 iter/s, 5.67158s/12 iters), loss = 2.29561
I0428 14:19:52.168572  7476 solver.cpp:237]     Train net output #0: loss = 2.29561 (* 1 = 2.29561 loss)
I0428 14:19:52.168581  7476 sgd_solver.cpp:105] Iteration 3648, lr = 0.0048548
I0428 14:19:57.635111  7476 solver.cpp:218] Iteration 3660 (2.19524 iter/s, 5.46637s/12 iters), loss = 2.22256
I0428 14:19:57.635154  7476 solver.cpp:237]     Train net output #0: loss = 2.22256 (* 1 = 2.22256 loss)
I0428 14:19:57.635164  7476 sgd_solver.cpp:105] Iteration 3660, lr = 0.00484327
I0428 14:20:02.600039  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_3672.caffemodel
I0428 14:20:03.939954  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_3672.solverstate
I0428 14:20:06.767736  7476 solver.cpp:330] Iteration 3672, Testing net (#0)
I0428 14:20:06.767762  7476 net.cpp:676] Ignoring source layer train-data
I0428 14:20:09.793018  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:20:11.355710  7476 solver.cpp:397]     Test net output #0: accuracy = 0.218137
I0428 14:20:11.355746  7476 solver.cpp:397]     Test net output #1: loss = 3.2939 (* 1 = 3.2939 loss)
I0428 14:20:11.484215  7476 solver.cpp:218] Iteration 3672 (0.866508 iter/s, 13.8487s/12 iters), loss = 2.62675
I0428 14:20:11.484269  7476 solver.cpp:237]     Train net output #0: loss = 2.62675 (* 1 = 2.62675 loss)
I0428 14:20:11.484282  7476 sgd_solver.cpp:105] Iteration 3672, lr = 0.00483177
I0428 14:20:16.005669  7476 solver.cpp:218] Iteration 3684 (2.65413 iter/s, 4.52126s/12 iters), loss = 2.35979
I0428 14:20:16.005712  7476 solver.cpp:237]     Train net output #0: loss = 2.35979 (* 1 = 2.35979 loss)
I0428 14:20:16.005720  7476 sgd_solver.cpp:105] Iteration 3684, lr = 0.0048203
I0428 14:20:21.426950  7476 solver.cpp:218] Iteration 3696 (2.21358 iter/s, 5.42108s/12 iters), loss = 2.39278
I0428 14:20:21.427095  7476 solver.cpp:237]     Train net output #0: loss = 2.39278 (* 1 = 2.39278 loss)
I0428 14:20:21.427106  7476 sgd_solver.cpp:105] Iteration 3696, lr = 0.00480886
I0428 14:20:26.939739  7476 solver.cpp:218] Iteration 3708 (2.17688 iter/s, 5.51248s/12 iters), loss = 2.37475
I0428 14:20:26.939786  7476 solver.cpp:237]     Train net output #0: loss = 2.37475 (* 1 = 2.37475 loss)
I0428 14:20:26.939797  7476 sgd_solver.cpp:105] Iteration 3708, lr = 0.00479744
I0428 14:20:32.438225  7476 solver.cpp:218] Iteration 3720 (2.1825 iter/s, 5.49828s/12 iters), loss = 2.31025
I0428 14:20:32.438267  7476 solver.cpp:237]     Train net output #0: loss = 2.31025 (* 1 = 2.31025 loss)
I0428 14:20:32.438277  7476 sgd_solver.cpp:105] Iteration 3720, lr = 0.00478605
I0428 14:20:38.251284  7476 solver.cpp:218] Iteration 3732 (2.06439 iter/s, 5.81285s/12 iters), loss = 2.2839
I0428 14:20:38.251334  7476 solver.cpp:237]     Train net output #0: loss = 2.2839 (* 1 = 2.2839 loss)
I0428 14:20:38.251345  7476 sgd_solver.cpp:105] Iteration 3732, lr = 0.00477469
I0428 14:20:42.654716  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:20:43.807459  7476 solver.cpp:218] Iteration 3744 (2.15984 iter/s, 5.55597s/12 iters), loss = 2.35301
I0428 14:20:43.807498  7476 solver.cpp:237]     Train net output #0: loss = 2.35301 (* 1 = 2.35301 loss)
I0428 14:20:43.807507  7476 sgd_solver.cpp:105] Iteration 3744, lr = 0.00476335
I0428 14:20:49.364279  7476 solver.cpp:218] Iteration 3756 (2.15959 iter/s, 5.55662s/12 iters), loss = 2.49642
I0428 14:20:49.364334  7476 solver.cpp:237]     Train net output #0: loss = 2.49642 (* 1 = 2.49642 loss)
I0428 14:20:49.364346  7476 sgd_solver.cpp:105] Iteration 3756, lr = 0.00475204
I0428 14:20:54.874147  7476 solver.cpp:218] Iteration 3768 (2.17799 iter/s, 5.50966s/12 iters), loss = 2.44758
I0428 14:20:54.874248  7476 solver.cpp:237]     Train net output #0: loss = 2.44758 (* 1 = 2.44758 loss)
I0428 14:20:54.874258  7476 sgd_solver.cpp:105] Iteration 3768, lr = 0.00474076
I0428 14:20:57.047127  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_3774.caffemodel
I0428 14:20:58.373337  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_3774.solverstate
I0428 14:20:59.444103  7476 solver.cpp:330] Iteration 3774, Testing net (#0)
I0428 14:20:59.444123  7476 net.cpp:676] Ignoring source layer train-data
I0428 14:21:02.547765  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:21:04.116236  7476 solver.cpp:397]     Test net output #0: accuracy = 0.220588
I0428 14:21:04.116273  7476 solver.cpp:397]     Test net output #1: loss = 3.41468 (* 1 = 3.41468 loss)
I0428 14:21:06.146528  7476 solver.cpp:218] Iteration 3780 (1.06459 iter/s, 11.272s/12 iters), loss = 2.27387
I0428 14:21:06.146569  7476 solver.cpp:237]     Train net output #0: loss = 2.27387 (* 1 = 2.27387 loss)
I0428 14:21:06.146579  7476 sgd_solver.cpp:105] Iteration 3780, lr = 0.00472951
I0428 14:21:11.575624  7476 solver.cpp:218] Iteration 3792 (2.21039 iter/s, 5.4289s/12 iters), loss = 2.60573
I0428 14:21:11.575665  7476 solver.cpp:237]     Train net output #0: loss = 2.60573 (* 1 = 2.60573 loss)
I0428 14:21:11.575675  7476 sgd_solver.cpp:105] Iteration 3792, lr = 0.00471828
I0428 14:21:17.036474  7476 solver.cpp:218] Iteration 3804 (2.19754 iter/s, 5.46065s/12 iters), loss = 2.36469
I0428 14:21:17.036556  7476 solver.cpp:237]     Train net output #0: loss = 2.36469 (* 1 = 2.36469 loss)
I0428 14:21:17.036568  7476 sgd_solver.cpp:105] Iteration 3804, lr = 0.00470707
I0428 14:21:22.695252  7476 solver.cpp:218] Iteration 3816 (2.12069 iter/s, 5.65853s/12 iters), loss = 2.45204
I0428 14:21:22.695293  7476 solver.cpp:237]     Train net output #0: loss = 2.45204 (* 1 = 2.45204 loss)
I0428 14:21:22.695302  7476 sgd_solver.cpp:105] Iteration 3816, lr = 0.0046959
I0428 14:21:28.511337  7476 solver.cpp:218] Iteration 3828 (2.06332 iter/s, 5.81587s/12 iters), loss = 2.60842
I0428 14:21:28.511464  7476 solver.cpp:237]     Train net output #0: loss = 2.60842 (* 1 = 2.60842 loss)
I0428 14:21:28.511476  7476 sgd_solver.cpp:105] Iteration 3828, lr = 0.00468475
I0428 14:21:34.234517  7476 solver.cpp:218] Iteration 3840 (2.09684 iter/s, 5.72289s/12 iters), loss = 2.21722
I0428 14:21:34.234560  7476 solver.cpp:237]     Train net output #0: loss = 2.21722 (* 1 = 2.21722 loss)
I0428 14:21:34.234567  7476 sgd_solver.cpp:105] Iteration 3840, lr = 0.00467363
I0428 14:21:35.446609  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:21:39.691083  7476 solver.cpp:218] Iteration 3852 (2.19927 iter/s, 5.45636s/12 iters), loss = 2.26816
I0428 14:21:39.691128  7476 solver.cpp:237]     Train net output #0: loss = 2.26816 (* 1 = 2.26816 loss)
I0428 14:21:39.691136  7476 sgd_solver.cpp:105] Iteration 3852, lr = 0.00466253
I0428 14:21:45.711572  7476 solver.cpp:218] Iteration 3864 (1.99327 iter/s, 6.02027s/12 iters), loss = 2.53104
I0428 14:21:45.711617  7476 solver.cpp:237]     Train net output #0: loss = 2.53104 (* 1 = 2.53104 loss)
I0428 14:21:45.711627  7476 sgd_solver.cpp:105] Iteration 3864, lr = 0.00465146
I0428 14:21:50.712744  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_3876.caffemodel
I0428 14:21:52.065162  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_3876.solverstate
I0428 14:21:53.246541  7476 solver.cpp:330] Iteration 3876, Testing net (#0)
I0428 14:21:53.246565  7476 net.cpp:676] Ignoring source layer train-data
I0428 14:21:56.145474  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:21:57.777853  7476 solver.cpp:397]     Test net output #0: accuracy = 0.244485
I0428 14:21:57.777886  7476 solver.cpp:397]     Test net output #1: loss = 3.16535 (* 1 = 3.16535 loss)
I0428 14:21:57.906291  7476 solver.cpp:218] Iteration 3876 (0.984063 iter/s, 12.1943s/12 iters), loss = 2.35374
I0428 14:21:57.906332  7476 solver.cpp:237]     Train net output #0: loss = 2.35374 (* 1 = 2.35374 loss)
I0428 14:21:57.906342  7476 sgd_solver.cpp:105] Iteration 3876, lr = 0.00464042
I0428 14:22:03.053689  7476 solver.cpp:218] Iteration 3888 (2.33137 iter/s, 5.1472s/12 iters), loss = 2.31022
I0428 14:22:03.053848  7476 solver.cpp:237]     Train net output #0: loss = 2.31022 (* 1 = 2.31022 loss)
I0428 14:22:03.053858  7476 sgd_solver.cpp:105] Iteration 3888, lr = 0.0046294
I0428 14:22:08.563412  7476 solver.cpp:218] Iteration 3900 (2.17809 iter/s, 5.5094s/12 iters), loss = 2.22333
I0428 14:22:08.563460  7476 solver.cpp:237]     Train net output #0: loss = 2.22333 (* 1 = 2.22333 loss)
I0428 14:22:08.563468  7476 sgd_solver.cpp:105] Iteration 3900, lr = 0.00461841
I0428 14:22:14.034494  7476 solver.cpp:218] Iteration 3912 (2.19343 iter/s, 5.47087s/12 iters), loss = 2.33739
I0428 14:22:14.034545  7476 solver.cpp:237]     Train net output #0: loss = 2.33739 (* 1 = 2.33739 loss)
I0428 14:22:14.034557  7476 sgd_solver.cpp:105] Iteration 3912, lr = 0.00460744
I0428 14:22:19.447571  7476 solver.cpp:218] Iteration 3924 (2.21694 iter/s, 5.41286s/12 iters), loss = 2.17815
I0428 14:22:19.447624  7476 solver.cpp:237]     Train net output #0: loss = 2.17815 (* 1 = 2.17815 loss)
I0428 14:22:19.447636  7476 sgd_solver.cpp:105] Iteration 3924, lr = 0.0045965
I0428 14:22:25.301314  7476 solver.cpp:218] Iteration 3936 (2.05005 iter/s, 5.85352s/12 iters), loss = 2.28918
I0428 14:22:25.301352  7476 solver.cpp:237]     Train net output #0: loss = 2.28918 (* 1 = 2.28918 loss)
I0428 14:22:25.301362  7476 sgd_solver.cpp:105] Iteration 3936, lr = 0.00458559
I0428 14:22:29.077375  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:22:30.810675  7476 solver.cpp:218] Iteration 3948 (2.17819 iter/s, 5.50916s/12 iters), loss = 2.28866
I0428 14:22:30.810726  7476 solver.cpp:237]     Train net output #0: loss = 2.28866 (* 1 = 2.28866 loss)
I0428 14:22:30.810737  7476 sgd_solver.cpp:105] Iteration 3948, lr = 0.0045747
I0428 14:22:36.285910  7476 solver.cpp:218] Iteration 3960 (2.19177 iter/s, 5.47503s/12 iters), loss = 2.42073
I0428 14:22:36.286005  7476 solver.cpp:237]     Train net output #0: loss = 2.42073 (* 1 = 2.42073 loss)
I0428 14:22:36.286015  7476 sgd_solver.cpp:105] Iteration 3960, lr = 0.00456384
I0428 14:22:41.952052  7476 solver.cpp:218] Iteration 3972 (2.11794 iter/s, 5.66588s/12 iters), loss = 2.27135
I0428 14:22:41.952108  7476 solver.cpp:237]     Train net output #0: loss = 2.27135 (* 1 = 2.27135 loss)
I0428 14:22:41.952119  7476 sgd_solver.cpp:105] Iteration 3972, lr = 0.00455301
I0428 14:22:44.110934  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_3978.caffemodel
I0428 14:22:47.324141  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_3978.solverstate
I0428 14:22:49.408830  7476 solver.cpp:330] Iteration 3978, Testing net (#0)
I0428 14:22:49.408850  7476 net.cpp:676] Ignoring source layer train-data
I0428 14:22:52.441866  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:22:54.084679  7476 solver.cpp:397]     Test net output #0: accuracy = 0.247549
I0428 14:22:54.084717  7476 solver.cpp:397]     Test net output #1: loss = 3.08086 (* 1 = 3.08086 loss)
I0428 14:22:56.058043  7476 solver.cpp:218] Iteration 3984 (0.850729 iter/s, 14.1055s/12 iters), loss = 2.18969
I0428 14:22:56.058104  7476 solver.cpp:237]     Train net output #0: loss = 2.18969 (* 1 = 2.18969 loss)
I0428 14:22:56.058118  7476 sgd_solver.cpp:105] Iteration 3984, lr = 0.0045422
I0428 14:23:01.568114  7476 solver.cpp:218] Iteration 3996 (2.17792 iter/s, 5.50984s/12 iters), loss = 2.23134
I0428 14:23:01.568169  7476 solver.cpp:237]     Train net output #0: loss = 2.23134 (* 1 = 2.23134 loss)
I0428 14:23:01.568181  7476 sgd_solver.cpp:105] Iteration 3996, lr = 0.00453141
I0428 14:23:07.152884  7476 solver.cpp:218] Iteration 4008 (2.14879 iter/s, 5.58455s/12 iters), loss = 2.45102
I0428 14:23:07.153025  7476 solver.cpp:237]     Train net output #0: loss = 2.45102 (* 1 = 2.45102 loss)
I0428 14:23:07.153036  7476 sgd_solver.cpp:105] Iteration 4008, lr = 0.00452066
I0428 14:23:12.625337  7476 solver.cpp:218] Iteration 4020 (2.19292 iter/s, 5.47215s/12 iters), loss = 2.17508
I0428 14:23:12.625394  7476 solver.cpp:237]     Train net output #0: loss = 2.17508 (* 1 = 2.17508 loss)
I0428 14:23:12.625406  7476 sgd_solver.cpp:105] Iteration 4020, lr = 0.00450992
I0428 14:23:18.105227  7476 solver.cpp:218] Iteration 4032 (2.18991 iter/s, 5.47967s/12 iters), loss = 2.32541
I0428 14:23:18.105285  7476 solver.cpp:237]     Train net output #0: loss = 2.32541 (* 1 = 2.32541 loss)
I0428 14:23:18.105298  7476 sgd_solver.cpp:105] Iteration 4032, lr = 0.00449921
I0428 14:23:23.541047  7476 solver.cpp:218] Iteration 4044 (2.20767 iter/s, 5.4356s/12 iters), loss = 2.17873
I0428 14:23:23.541093  7476 solver.cpp:237]     Train net output #0: loss = 2.17873 (* 1 = 2.17873 loss)
I0428 14:23:23.541105  7476 sgd_solver.cpp:105] Iteration 4044, lr = 0.00448853
I0428 14:23:24.090675  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:23:29.048871  7476 solver.cpp:218] Iteration 4056 (2.1788 iter/s, 5.50761s/12 iters), loss = 2.15657
I0428 14:23:29.048921  7476 solver.cpp:237]     Train net output #0: loss = 2.15657 (* 1 = 2.15657 loss)
I0428 14:23:29.048933  7476 sgd_solver.cpp:105] Iteration 4056, lr = 0.00447788
I0428 14:23:34.657807  7476 solver.cpp:218] Iteration 4068 (2.13953 iter/s, 5.60872s/12 iters), loss = 2.14154
I0428 14:23:34.657855  7476 solver.cpp:237]     Train net output #0: loss = 2.14154 (* 1 = 2.14154 loss)
I0428 14:23:34.657864  7476 sgd_solver.cpp:105] Iteration 4068, lr = 0.00446724
I0428 14:23:39.489332  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_4080.caffemodel
I0428 14:23:47.396373  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_4080.solverstate
I0428 14:23:49.450163  7476 solver.cpp:330] Iteration 4080, Testing net (#0)
I0428 14:23:49.450182  7476 net.cpp:676] Ignoring source layer train-data
I0428 14:23:52.540195  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:23:54.349167  7476 solver.cpp:397]     Test net output #0: accuracy = 0.26777
I0428 14:23:54.349196  7476 solver.cpp:397]     Test net output #1: loss = 3.01232 (* 1 = 3.01232 loss)
I0428 14:23:54.477527  7476 solver.cpp:218] Iteration 4080 (0.605475 iter/s, 19.8191s/12 iters), loss = 2.17579
I0428 14:23:54.477572  7476 solver.cpp:237]     Train net output #0: loss = 2.17579 (* 1 = 2.17579 loss)
I0428 14:23:54.477583  7476 sgd_solver.cpp:105] Iteration 4080, lr = 0.00445664
I0428 14:23:59.177162  7476 solver.cpp:218] Iteration 4092 (2.55349 iter/s, 4.69945s/12 iters), loss = 2.24219
I0428 14:23:59.177204  7476 solver.cpp:237]     Train net output #0: loss = 2.24219 (* 1 = 2.24219 loss)
I0428 14:23:59.177212  7476 sgd_solver.cpp:105] Iteration 4092, lr = 0.00444606
I0428 14:24:04.655630  7476 solver.cpp:218] Iteration 4104 (2.19048 iter/s, 5.47825s/12 iters), loss = 2.10698
I0428 14:24:04.655685  7476 solver.cpp:237]     Train net output #0: loss = 2.10698 (* 1 = 2.10698 loss)
I0428 14:24:04.655699  7476 sgd_solver.cpp:105] Iteration 4104, lr = 0.0044355
I0428 14:24:10.169337  7476 solver.cpp:218] Iteration 4116 (2.17648 iter/s, 5.51349s/12 iters), loss = 2.07534
I0428 14:24:10.172076  7476 solver.cpp:237]     Train net output #0: loss = 2.07534 (* 1 = 2.07534 loss)
I0428 14:24:10.172087  7476 sgd_solver.cpp:105] Iteration 4116, lr = 0.00442497
I0428 14:24:15.550812  7476 solver.cpp:218] Iteration 4128 (2.23107 iter/s, 5.37858s/12 iters), loss = 1.9146
I0428 14:24:15.550861  7476 solver.cpp:237]     Train net output #0: loss = 1.9146 (* 1 = 1.9146 loss)
I0428 14:24:15.550871  7476 sgd_solver.cpp:105] Iteration 4128, lr = 0.00441447
I0428 14:24:21.125252  7476 solver.cpp:218] Iteration 4140 (2.15277 iter/s, 5.57423s/12 iters), loss = 1.98465
I0428 14:24:21.125293  7476 solver.cpp:237]     Train net output #0: loss = 1.98465 (* 1 = 1.98465 loss)
I0428 14:24:21.125303  7476 sgd_solver.cpp:105] Iteration 4140, lr = 0.00440398
I0428 14:24:24.005957  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:24:26.652990  7476 solver.cpp:218] Iteration 4152 (2.17095 iter/s, 5.52753s/12 iters), loss = 1.97469
I0428 14:24:26.653045  7476 solver.cpp:237]     Train net output #0: loss = 1.97469 (* 1 = 1.97469 loss)
I0428 14:24:26.653059  7476 sgd_solver.cpp:105] Iteration 4152, lr = 0.00439353
I0428 14:24:28.769407  7476 blocking_queue.cpp:49] Waiting for data
I0428 14:24:32.618531  7476 solver.cpp:218] Iteration 4164 (2.01163 iter/s, 5.96531s/12 iters), loss = 2.13234
I0428 14:24:32.618580  7476 solver.cpp:237]     Train net output #0: loss = 2.13234 (* 1 = 2.13234 loss)
I0428 14:24:32.618589  7476 sgd_solver.cpp:105] Iteration 4164, lr = 0.0043831
I0428 14:24:38.148175  7476 solver.cpp:218] Iteration 4176 (2.1702 iter/s, 5.52943s/12 iters), loss = 1.99263
I0428 14:24:38.148211  7476 solver.cpp:237]     Train net output #0: loss = 1.99263 (* 1 = 1.99263 loss)
I0428 14:24:38.148219  7476 sgd_solver.cpp:105] Iteration 4176, lr = 0.00437269
I0428 14:24:40.325444  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_4182.caffemodel
I0428 14:24:42.529492  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_4182.solverstate
I0428 14:24:43.647292  7476 solver.cpp:330] Iteration 4182, Testing net (#0)
I0428 14:24:43.647313  7476 net.cpp:676] Ignoring source layer train-data
I0428 14:24:46.509336  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:24:48.233127  7476 solver.cpp:397]     Test net output #0: accuracy = 0.301471
I0428 14:24:48.233165  7476 solver.cpp:397]     Test net output #1: loss = 2.9191 (* 1 = 2.9191 loss)
I0428 14:24:50.115068  7476 solver.cpp:218] Iteration 4188 (1.0028 iter/s, 11.9665s/12 iters), loss = 2.25257
I0428 14:24:50.115123  7476 solver.cpp:237]     Train net output #0: loss = 2.25257 (* 1 = 2.25257 loss)
I0428 14:24:50.115134  7476 sgd_solver.cpp:105] Iteration 4188, lr = 0.00436231
I0428 14:24:55.606778  7476 solver.cpp:218] Iteration 4200 (2.1852 iter/s, 5.49149s/12 iters), loss = 2.13715
I0428 14:24:55.606822  7476 solver.cpp:237]     Train net output #0: loss = 2.13715 (* 1 = 2.13715 loss)
I0428 14:24:55.606830  7476 sgd_solver.cpp:105] Iteration 4200, lr = 0.00435195
I0428 14:25:01.186383  7476 solver.cpp:218] Iteration 4212 (2.15077 iter/s, 5.57939s/12 iters), loss = 2.0813
I0428 14:25:01.186434  7476 solver.cpp:237]     Train net output #0: loss = 2.0813 (* 1 = 2.0813 loss)
I0428 14:25:01.186445  7476 sgd_solver.cpp:105] Iteration 4212, lr = 0.00434162
I0428 14:25:06.715121  7476 solver.cpp:218] Iteration 4224 (2.17056 iter/s, 5.52852s/12 iters), loss = 2.09992
I0428 14:25:06.715179  7476 solver.cpp:237]     Train net output #0: loss = 2.09992 (* 1 = 2.09992 loss)
I0428 14:25:06.715191  7476 sgd_solver.cpp:105] Iteration 4224, lr = 0.00433131
I0428 14:25:12.145198  7476 solver.cpp:218] Iteration 4236 (2.21 iter/s, 5.42985s/12 iters), loss = 1.91374
I0428 14:25:12.145342  7476 solver.cpp:237]     Train net output #0: loss = 1.91374 (* 1 = 1.91374 loss)
I0428 14:25:12.145355  7476 sgd_solver.cpp:105] Iteration 4236, lr = 0.00432103
I0428 14:25:17.368145  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:25:17.647742  7476 solver.cpp:218] Iteration 4248 (2.18093 iter/s, 5.50224s/12 iters), loss = 1.96708
I0428 14:25:17.647781  7476 solver.cpp:237]     Train net output #0: loss = 1.96708 (* 1 = 1.96708 loss)
I0428 14:25:17.647789  7476 sgd_solver.cpp:105] Iteration 4248, lr = 0.00431077
I0428 14:25:23.102043  7476 solver.cpp:218] Iteration 4260 (2.20018 iter/s, 5.4541s/12 iters), loss = 2.19557
I0428 14:25:23.102087  7476 solver.cpp:237]     Train net output #0: loss = 2.19557 (* 1 = 2.19557 loss)
I0428 14:25:23.102097  7476 sgd_solver.cpp:105] Iteration 4260, lr = 0.00430053
I0428 14:25:28.874004  7476 solver.cpp:218] Iteration 4272 (2.07909 iter/s, 5.77175s/12 iters), loss = 2.16733
I0428 14:25:28.874047  7476 solver.cpp:237]     Train net output #0: loss = 2.16733 (* 1 = 2.16733 loss)
I0428 14:25:28.874056  7476 sgd_solver.cpp:105] Iteration 4272, lr = 0.00429032
I0428 14:25:33.947757  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_4284.caffemodel
I0428 14:25:39.635023  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_4284.solverstate
I0428 14:25:41.222409  7476 solver.cpp:330] Iteration 4284, Testing net (#0)
I0428 14:25:41.222436  7476 net.cpp:676] Ignoring source layer train-data
I0428 14:25:43.971819  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:25:45.689685  7476 solver.cpp:397]     Test net output #0: accuracy = 0.297181
I0428 14:25:45.689718  7476 solver.cpp:397]     Test net output #1: loss = 2.91275 (* 1 = 2.91275 loss)
I0428 14:25:45.818123  7476 solver.cpp:218] Iteration 4284 (0.708231 iter/s, 16.9436s/12 iters), loss = 1.84661
I0428 14:25:45.818178  7476 solver.cpp:237]     Train net output #0: loss = 1.84661 (* 1 = 1.84661 loss)
I0428 14:25:45.818193  7476 sgd_solver.cpp:105] Iteration 4284, lr = 0.00428014
I0428 14:25:50.473739  7476 solver.cpp:218] Iteration 4296 (2.57764 iter/s, 4.65542s/12 iters), loss = 2.14629
I0428 14:25:50.473786  7476 solver.cpp:237]     Train net output #0: loss = 2.14629 (* 1 = 2.14629 loss)
I0428 14:25:50.473796  7476 sgd_solver.cpp:105] Iteration 4296, lr = 0.00426998
I0428 14:25:55.994971  7476 solver.cpp:218] Iteration 4308 (2.17351 iter/s, 5.52102s/12 iters), loss = 2.36843
I0428 14:25:55.995014  7476 solver.cpp:237]     Train net output #0: loss = 2.36843 (* 1 = 2.36843 loss)
I0428 14:25:55.995023  7476 sgd_solver.cpp:105] Iteration 4308, lr = 0.00425984
I0428 14:26:01.492031  7476 solver.cpp:218] Iteration 4320 (2.18307 iter/s, 5.49685s/12 iters), loss = 1.81988
I0428 14:26:01.492072  7476 solver.cpp:237]     Train net output #0: loss = 1.81988 (* 1 = 1.81988 loss)
I0428 14:26:01.492081  7476 sgd_solver.cpp:105] Iteration 4320, lr = 0.00424972
I0428 14:26:06.948114  7476 solver.cpp:218] Iteration 4332 (2.19946 iter/s, 5.45588s/12 iters), loss = 2.27494
I0428 14:26:06.948158  7476 solver.cpp:237]     Train net output #0: loss = 2.27494 (* 1 = 2.27494 loss)
I0428 14:26:06.948168  7476 sgd_solver.cpp:105] Iteration 4332, lr = 0.00423964
I0428 14:26:12.614843  7476 solver.cpp:218] Iteration 4344 (2.1177 iter/s, 5.66651s/12 iters), loss = 1.77498
I0428 14:26:12.614897  7476 solver.cpp:237]     Train net output #0: loss = 1.77498 (* 1 = 1.77498 loss)
I0428 14:26:12.614909  7476 sgd_solver.cpp:105] Iteration 4344, lr = 0.00422957
I0428 14:26:15.138381  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:26:18.714661  7476 solver.cpp:218] Iteration 4356 (1.96735 iter/s, 6.09959s/12 iters), loss = 1.82595
I0428 14:26:18.714721  7476 solver.cpp:237]     Train net output #0: loss = 1.82595 (* 1 = 1.82595 loss)
I0428 14:26:18.714733  7476 sgd_solver.cpp:105] Iteration 4356, lr = 0.00421953
I0428 14:26:24.210240  7476 solver.cpp:218] Iteration 4368 (2.18366 iter/s, 5.49536s/12 iters), loss = 1.82281
I0428 14:26:24.210281  7476 solver.cpp:237]     Train net output #0: loss = 1.82281 (* 1 = 1.82281 loss)
I0428 14:26:24.210289  7476 sgd_solver.cpp:105] Iteration 4368, lr = 0.00420951
I0428 14:26:29.878391  7476 solver.cpp:218] Iteration 4380 (2.11717 iter/s, 5.66793s/12 iters), loss = 1.82331
I0428 14:26:29.878437  7476 solver.cpp:237]     Train net output #0: loss = 1.82331 (* 1 = 1.82331 loss)
I0428 14:26:29.878444  7476 sgd_solver.cpp:105] Iteration 4380, lr = 0.00419952
I0428 14:26:32.055734  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_4386.caffemodel
I0428 14:26:34.369465  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_4386.solverstate
I0428 14:26:35.843899  7476 solver.cpp:330] Iteration 4386, Testing net (#0)
I0428 14:26:35.843921  7476 net.cpp:676] Ignoring source layer train-data
I0428 14:26:38.526576  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:26:40.514364  7476 solver.cpp:397]     Test net output #0: accuracy = 0.305147
I0428 14:26:40.514396  7476 solver.cpp:397]     Test net output #1: loss = 2.91964 (* 1 = 2.91964 loss)
I0428 14:26:42.425792  7476 solver.cpp:218] Iteration 4392 (0.956403 iter/s, 12.547s/12 iters), loss = 2.0064
I0428 14:26:42.425835  7476 solver.cpp:237]     Train net output #0: loss = 2.0064 (* 1 = 2.0064 loss)
I0428 14:26:42.425844  7476 sgd_solver.cpp:105] Iteration 4392, lr = 0.00418954
I0428 14:26:47.791234  7476 solver.cpp:218] Iteration 4404 (2.23662 iter/s, 5.36524s/12 iters), loss = 2.24518
I0428 14:26:47.791404  7476 solver.cpp:237]     Train net output #0: loss = 2.24518 (* 1 = 2.24518 loss)
I0428 14:26:47.791412  7476 sgd_solver.cpp:105] Iteration 4404, lr = 0.0041796
I0428 14:26:53.242122  7476 solver.cpp:218] Iteration 4416 (2.20161 iter/s, 5.45056s/12 iters), loss = 1.99062
I0428 14:26:53.242172  7476 solver.cpp:237]     Train net output #0: loss = 1.99062 (* 1 = 1.99062 loss)
I0428 14:26:53.242182  7476 sgd_solver.cpp:105] Iteration 4416, lr = 0.00416967
I0428 14:26:58.764801  7476 solver.cpp:218] Iteration 4428 (2.17294 iter/s, 5.52246s/12 iters), loss = 1.97976
I0428 14:26:58.764855  7476 solver.cpp:237]     Train net output #0: loss = 1.97976 (* 1 = 1.97976 loss)
I0428 14:26:58.764866  7476 sgd_solver.cpp:105] Iteration 4428, lr = 0.00415977
I0428 14:27:05.817752  7476 solver.cpp:218] Iteration 4440 (1.70148 iter/s, 7.05269s/12 iters), loss = 2.11618
I0428 14:27:05.817814  7476 solver.cpp:237]     Train net output #0: loss = 2.11618 (* 1 = 2.11618 loss)
I0428 14:27:05.817826  7476 sgd_solver.cpp:105] Iteration 4440, lr = 0.0041499
I0428 14:27:11.620649  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:27:12.773293  7476 solver.cpp:218] Iteration 4452 (1.72531 iter/s, 6.95528s/12 iters), loss = 2.07432
I0428 14:27:12.773336  7476 solver.cpp:237]     Train net output #0: loss = 2.07432 (* 1 = 2.07432 loss)
I0428 14:27:12.773344  7476 sgd_solver.cpp:105] Iteration 4452, lr = 0.00414005
I0428 14:27:18.576665  7476 solver.cpp:218] Iteration 4464 (2.06784 iter/s, 5.80316s/12 iters), loss = 2.10706
I0428 14:27:18.579087  7476 solver.cpp:237]     Train net output #0: loss = 2.10706 (* 1 = 2.10706 loss)
I0428 14:27:18.579097  7476 sgd_solver.cpp:105] Iteration 4464, lr = 0.00413022
I0428 14:27:24.066723  7476 solver.cpp:218] Iteration 4476 (2.1868 iter/s, 5.48748s/12 iters), loss = 1.74465
I0428 14:27:24.066772  7476 solver.cpp:237]     Train net output #0: loss = 1.74465 (* 1 = 1.74465 loss)
I0428 14:27:24.066781  7476 sgd_solver.cpp:105] Iteration 4476, lr = 0.00412041
I0428 14:27:29.026221  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_4488.caffemodel
I0428 14:27:31.755162  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_4488.solverstate
I0428 14:27:33.559801  7476 solver.cpp:330] Iteration 4488, Testing net (#0)
I0428 14:27:33.559823  7476 net.cpp:676] Ignoring source layer train-data
I0428 14:27:36.257061  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:27:38.054018  7476 solver.cpp:397]     Test net output #0: accuracy = 0.296569
I0428 14:27:38.054056  7476 solver.cpp:397]     Test net output #1: loss = 2.96061 (* 1 = 2.96061 loss)
I0428 14:27:38.182592  7476 solver.cpp:218] Iteration 4488 (0.850133 iter/s, 14.1154s/12 iters), loss = 1.82442
I0428 14:27:38.182641  7476 solver.cpp:237]     Train net output #0: loss = 1.82442 (* 1 = 1.82442 loss)
I0428 14:27:38.182651  7476 sgd_solver.cpp:105] Iteration 4488, lr = 0.00411063
I0428 14:27:42.675837  7476 solver.cpp:218] Iteration 4500 (2.67079 iter/s, 4.49306s/12 iters), loss = 1.85235
I0428 14:27:42.675892  7476 solver.cpp:237]     Train net output #0: loss = 1.85235 (* 1 = 1.85235 loss)
I0428 14:27:42.675904  7476 sgd_solver.cpp:105] Iteration 4500, lr = 0.00410087
I0428 14:27:48.486557  7476 solver.cpp:218] Iteration 4512 (2.06523 iter/s, 5.8105s/12 iters), loss = 2.20164
I0428 14:27:48.486603  7476 solver.cpp:237]     Train net output #0: loss = 2.20164 (* 1 = 2.20164 loss)
I0428 14:27:48.486611  7476 sgd_solver.cpp:105] Iteration 4512, lr = 0.00409113
I0428 14:27:53.974623  7476 solver.cpp:218] Iteration 4524 (2.18665 iter/s, 5.48786s/12 iters), loss = 1.91631
I0428 14:27:53.974721  7476 solver.cpp:237]     Train net output #0: loss = 1.91631 (* 1 = 1.91631 loss)
I0428 14:27:53.974730  7476 sgd_solver.cpp:105] Iteration 4524, lr = 0.00408142
I0428 14:27:59.701247  7476 solver.cpp:218] Iteration 4536 (2.09557 iter/s, 5.72636s/12 iters), loss = 2.0029
I0428 14:27:59.701287  7476 solver.cpp:237]     Train net output #0: loss = 2.0029 (* 1 = 2.0029 loss)
I0428 14:27:59.701297  7476 sgd_solver.cpp:105] Iteration 4536, lr = 0.00407173
I0428 14:28:05.391000  7476 solver.cpp:218] Iteration 4548 (2.10913 iter/s, 5.68955s/12 iters), loss = 1.68074
I0428 14:28:05.391045  7476 solver.cpp:237]     Train net output #0: loss = 1.68074 (* 1 = 1.68074 loss)
I0428 14:28:05.391053  7476 sgd_solver.cpp:105] Iteration 4548, lr = 0.00406206
I0428 14:28:06.797523  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:28:10.995901  7476 solver.cpp:218] Iteration 4560 (2.14106 iter/s, 5.60469s/12 iters), loss = 2.05157
I0428 14:28:10.995946  7476 solver.cpp:237]     Train net output #0: loss = 2.05157 (* 1 = 2.05157 loss)
I0428 14:28:10.995955  7476 sgd_solver.cpp:105] Iteration 4560, lr = 0.00405242
I0428 14:28:16.926154  7476 solver.cpp:218] Iteration 4572 (2.0236 iter/s, 5.93004s/12 iters), loss = 1.83983
I0428 14:28:16.926195  7476 solver.cpp:237]     Train net output #0: loss = 1.83983 (* 1 = 1.83983 loss)
I0428 14:28:16.926203  7476 sgd_solver.cpp:105] Iteration 4572, lr = 0.0040428
I0428 14:28:22.313443  7476 solver.cpp:218] Iteration 4584 (2.22755 iter/s, 5.3871s/12 iters), loss = 2.03616
I0428 14:28:22.313483  7476 solver.cpp:237]     Train net output #0: loss = 2.03616 (* 1 = 2.03616 loss)
I0428 14:28:22.313491  7476 sgd_solver.cpp:105] Iteration 4584, lr = 0.0040332
I0428 14:28:24.480026  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_4590.caffemodel
I0428 14:28:26.957971  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_4590.solverstate
I0428 14:28:29.446187  7476 solver.cpp:330] Iteration 4590, Testing net (#0)
I0428 14:28:29.446210  7476 net.cpp:676] Ignoring source layer train-data
I0428 14:28:32.117606  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:28:33.991458  7476 solver.cpp:397]     Test net output #0: accuracy = 0.273284
I0428 14:28:33.991487  7476 solver.cpp:397]     Test net output #1: loss = 3.0955 (* 1 = 3.0955 loss)
I0428 14:28:36.027866  7476 solver.cpp:218] Iteration 4596 (0.875018 iter/s, 13.714s/12 iters), loss = 1.99296
I0428 14:28:36.027928  7476 solver.cpp:237]     Train net output #0: loss = 1.99296 (* 1 = 1.99296 loss)
I0428 14:28:36.027941  7476 sgd_solver.cpp:105] Iteration 4596, lr = 0.00402362
I0428 14:28:41.465108  7476 solver.cpp:218] Iteration 4608 (2.20709 iter/s, 5.43702s/12 iters), loss = 1.85995
I0428 14:28:41.465171  7476 solver.cpp:237]     Train net output #0: loss = 1.85995 (* 1 = 1.85995 loss)
I0428 14:28:41.465184  7476 sgd_solver.cpp:105] Iteration 4608, lr = 0.00401407
I0428 14:28:47.102965  7476 solver.cpp:218] Iteration 4620 (2.12855 iter/s, 5.63763s/12 iters), loss = 1.87515
I0428 14:28:47.103020  7476 solver.cpp:237]     Train net output #0: loss = 1.87515 (* 1 = 1.87515 loss)
I0428 14:28:47.103032  7476 sgd_solver.cpp:105] Iteration 4620, lr = 0.00400454
I0428 14:28:52.561348  7476 solver.cpp:218] Iteration 4632 (2.19854 iter/s, 5.45817s/12 iters), loss = 1.68797
I0428 14:28:52.561393  7476 solver.cpp:237]     Train net output #0: loss = 1.68797 (* 1 = 1.68797 loss)
I0428 14:28:52.561403  7476 sgd_solver.cpp:105] Iteration 4632, lr = 0.00399503
I0428 14:28:57.967433  7476 solver.cpp:218] Iteration 4644 (2.21981 iter/s, 5.40588s/12 iters), loss = 1.7199
I0428 14:28:57.967555  7476 solver.cpp:237]     Train net output #0: loss = 1.7199 (* 1 = 1.7199 loss)
I0428 14:28:57.967566  7476 sgd_solver.cpp:105] Iteration 4644, lr = 0.00398555
I0428 14:29:01.688450  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:29:03.492852  7476 solver.cpp:218] Iteration 4656 (2.17189 iter/s, 5.52513s/12 iters), loss = 1.77062
I0428 14:29:03.492895  7476 solver.cpp:237]     Train net output #0: loss = 1.77062 (* 1 = 1.77062 loss)
I0428 14:29:03.492904  7476 sgd_solver.cpp:105] Iteration 4656, lr = 0.00397608
I0428 14:29:08.847127  7476 solver.cpp:218] Iteration 4668 (2.24128 iter/s, 5.35408s/12 iters), loss = 1.83146
I0428 14:29:08.847164  7476 solver.cpp:237]     Train net output #0: loss = 1.83146 (* 1 = 1.83146 loss)
I0428 14:29:08.847173  7476 sgd_solver.cpp:105] Iteration 4668, lr = 0.00396664
I0428 14:29:14.219133  7476 solver.cpp:218] Iteration 4680 (2.23389 iter/s, 5.37181s/12 iters), loss = 1.82015
I0428 14:29:14.219182  7476 solver.cpp:237]     Train net output #0: loss = 1.82015 (* 1 = 1.82015 loss)
I0428 14:29:14.219192  7476 sgd_solver.cpp:105] Iteration 4680, lr = 0.00395723
I0428 14:29:19.210068  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_4692.caffemodel
I0428 14:29:24.631515  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_4692.solverstate
I0428 14:29:27.656070  7476 solver.cpp:330] Iteration 4692, Testing net (#0)
I0428 14:29:27.656095  7476 net.cpp:676] Ignoring source layer train-data
I0428 14:29:30.386827  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:29:32.261020  7476 solver.cpp:397]     Test net output #0: accuracy = 0.303922
I0428 14:29:32.261066  7476 solver.cpp:397]     Test net output #1: loss = 3.00196 (* 1 = 3.00196 loss)
I0428 14:29:32.389482  7476 solver.cpp:218] Iteration 4692 (0.660436 iter/s, 18.1698s/12 iters), loss = 1.87169
I0428 14:29:32.389525  7476 solver.cpp:237]     Train net output #0: loss = 1.87169 (* 1 = 1.87169 loss)
I0428 14:29:32.389536  7476 sgd_solver.cpp:105] Iteration 4692, lr = 0.00394783
I0428 14:29:36.971298  7476 solver.cpp:218] Iteration 4704 (2.61915 iter/s, 4.58164s/12 iters), loss = 1.7738
I0428 14:29:36.971340  7476 solver.cpp:237]     Train net output #0: loss = 1.7738 (* 1 = 1.7738 loss)
I0428 14:29:36.971349  7476 sgd_solver.cpp:105] Iteration 4704, lr = 0.00393846
I0428 14:29:42.603266  7476 solver.cpp:218] Iteration 4716 (2.13077 iter/s, 5.63176s/12 iters), loss = 1.92662
I0428 14:29:42.603308  7476 solver.cpp:237]     Train net output #0: loss = 1.92662 (* 1 = 1.92662 loss)
I0428 14:29:42.603317  7476 sgd_solver.cpp:105] Iteration 4716, lr = 0.00392911
I0428 14:29:48.169559  7476 solver.cpp:218] Iteration 4728 (2.15591 iter/s, 5.56609s/12 iters), loss = 1.75379
I0428 14:29:48.169598  7476 solver.cpp:237]     Train net output #0: loss = 1.75379 (* 1 = 1.75379 loss)
I0428 14:29:48.169607  7476 sgd_solver.cpp:105] Iteration 4728, lr = 0.00391978
I0428 14:29:53.638118  7476 solver.cpp:218] Iteration 4740 (2.19444 iter/s, 5.46835s/12 iters), loss = 1.76046
I0428 14:29:53.638161  7476 solver.cpp:237]     Train net output #0: loss = 1.76046 (* 1 = 1.76046 loss)
I0428 14:29:53.638170  7476 sgd_solver.cpp:105] Iteration 4740, lr = 0.00391047
I0428 14:29:59.147744  7476 solver.cpp:218] Iteration 4752 (2.17809 iter/s, 5.50942s/12 iters), loss = 1.65898
I0428 14:29:59.147784  7476 solver.cpp:237]     Train net output #0: loss = 1.65898 (* 1 = 1.65898 loss)
I0428 14:29:59.147794  7476 sgd_solver.cpp:105] Iteration 4752, lr = 0.00390119
I0428 14:29:59.719215  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:30:04.542625  7476 solver.cpp:218] Iteration 4764 (2.22441 iter/s, 5.39468s/12 iters), loss = 1.94966
I0428 14:30:04.567950  7476 solver.cpp:237]     Train net output #0: loss = 1.94966 (* 1 = 1.94966 loss)
I0428 14:30:04.567966  7476 sgd_solver.cpp:105] Iteration 4764, lr = 0.00389193
I0428 14:30:09.834209  7476 solver.cpp:218] Iteration 4776 (2.27872 iter/s, 5.26611s/12 iters), loss = 1.5604
I0428 14:30:09.834250  7476 solver.cpp:237]     Train net output #0: loss = 1.5604 (* 1 = 1.5604 loss)
I0428 14:30:09.834260  7476 sgd_solver.cpp:105] Iteration 4776, lr = 0.00388269
I0428 14:30:15.500547  7476 solver.cpp:218] Iteration 4788 (2.11785 iter/s, 5.66613s/12 iters), loss = 1.64067
I0428 14:30:15.500584  7476 solver.cpp:237]     Train net output #0: loss = 1.64067 (* 1 = 1.64067 loss)
I0428 14:30:15.500593  7476 sgd_solver.cpp:105] Iteration 4788, lr = 0.00387347
I0428 14:30:17.692991  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_4794.caffemodel
I0428 14:30:28.225647  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_4794.solverstate
I0428 14:30:34.335381  7476 solver.cpp:330] Iteration 4794, Testing net (#0)
I0428 14:30:34.335399  7476 net.cpp:676] Ignoring source layer train-data
I0428 14:30:36.877629  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:30:38.805603  7476 solver.cpp:397]     Test net output #0: accuracy = 0.304534
I0428 14:30:38.805641  7476 solver.cpp:397]     Test net output #1: loss = 2.99617 (* 1 = 2.99617 loss)
I0428 14:30:40.795164  7476 solver.cpp:218] Iteration 4800 (0.474423 iter/s, 25.2939s/12 iters), loss = 1.8708
I0428 14:30:40.795225  7476 solver.cpp:237]     Train net output #0: loss = 1.8708 (* 1 = 1.8708 loss)
I0428 14:30:40.795236  7476 sgd_solver.cpp:105] Iteration 4800, lr = 0.00386427
I0428 14:30:46.282986  7476 solver.cpp:218] Iteration 4812 (2.18675 iter/s, 5.4876s/12 iters), loss = 2.01989
I0428 14:30:46.283031  7476 solver.cpp:237]     Train net output #0: loss = 2.01989 (* 1 = 2.01989 loss)
I0428 14:30:46.283039  7476 sgd_solver.cpp:105] Iteration 4812, lr = 0.0038551
I0428 14:30:51.693132  7476 solver.cpp:218] Iteration 4824 (2.21814 iter/s, 5.40994s/12 iters), loss = 1.66674
I0428 14:30:51.693168  7476 solver.cpp:237]     Train net output #0: loss = 1.66674 (* 1 = 1.66674 loss)
I0428 14:30:51.693176  7476 sgd_solver.cpp:105] Iteration 4824, lr = 0.00384594
I0428 14:30:56.940537  7476 solver.cpp:218] Iteration 4836 (2.28693 iter/s, 5.2472s/12 iters), loss = 1.91645
I0428 14:30:56.940579  7476 solver.cpp:237]     Train net output #0: loss = 1.91645 (* 1 = 1.91645 loss)
I0428 14:30:56.940587  7476 sgd_solver.cpp:105] Iteration 4836, lr = 0.00383681
I0428 14:30:59.139942  7476 blocking_queue.cpp:49] Waiting for data
I0428 14:31:02.393431  7476 solver.cpp:218] Iteration 4848 (2.20075 iter/s, 5.45269s/12 iters), loss = 1.88006
I0428 14:31:02.393471  7476 solver.cpp:237]     Train net output #0: loss = 1.88006 (* 1 = 1.88006 loss)
I0428 14:31:02.393481  7476 sgd_solver.cpp:105] Iteration 4848, lr = 0.0038277
I0428 14:31:05.272799  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:31:07.778628  7476 solver.cpp:218] Iteration 4860 (2.22842 iter/s, 5.38499s/12 iters), loss = 1.68699
I0428 14:31:07.778764  7476 solver.cpp:237]     Train net output #0: loss = 1.68699 (* 1 = 1.68699 loss)
I0428 14:31:07.778777  7476 sgd_solver.cpp:105] Iteration 4860, lr = 0.00381862
I0428 14:31:13.369068  7476 solver.cpp:218] Iteration 4872 (2.14664 iter/s, 5.59014s/12 iters), loss = 1.59619
I0428 14:31:13.369117  7476 solver.cpp:237]     Train net output #0: loss = 1.59619 (* 1 = 1.59619 loss)
I0428 14:31:13.369128  7476 sgd_solver.cpp:105] Iteration 4872, lr = 0.00380955
I0428 14:31:18.830214  7476 solver.cpp:218] Iteration 4884 (2.19743 iter/s, 5.46093s/12 iters), loss = 1.66561
I0428 14:31:18.830256  7476 solver.cpp:237]     Train net output #0: loss = 1.66561 (* 1 = 1.66561 loss)
I0428 14:31:18.830265  7476 sgd_solver.cpp:105] Iteration 4884, lr = 0.0038005
I0428 14:31:23.663468  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_4896.caffemodel
I0428 14:31:31.280913  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_4896.solverstate
I0428 14:31:34.772289  7476 solver.cpp:330] Iteration 4896, Testing net (#0)
I0428 14:31:34.772311  7476 net.cpp:676] Ignoring source layer train-data
I0428 14:31:37.336007  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:31:39.298970  7476 solver.cpp:397]     Test net output #0: accuracy = 0.277574
I0428 14:31:39.299078  7476 solver.cpp:397]     Test net output #1: loss = 3.09493 (* 1 = 3.09493 loss)
I0428 14:31:39.427266  7476 solver.cpp:218] Iteration 4896 (0.582625 iter/s, 20.5964s/12 iters), loss = 1.61259
I0428 14:31:39.427317  7476 solver.cpp:237]     Train net output #0: loss = 1.61259 (* 1 = 1.61259 loss)
I0428 14:31:39.427330  7476 sgd_solver.cpp:105] Iteration 4896, lr = 0.00379148
I0428 14:31:43.888450  7476 solver.cpp:218] Iteration 4908 (2.68998 iter/s, 4.461s/12 iters), loss = 1.66545
I0428 14:31:43.888509  7476 solver.cpp:237]     Train net output #0: loss = 1.66545 (* 1 = 1.66545 loss)
I0428 14:31:43.888518  7476 sgd_solver.cpp:105] Iteration 4908, lr = 0.00378248
I0428 14:31:49.328542  7476 solver.cpp:218] Iteration 4920 (2.20594 iter/s, 5.43987s/12 iters), loss = 1.85383
I0428 14:31:49.328601  7476 solver.cpp:237]     Train net output #0: loss = 1.85383 (* 1 = 1.85383 loss)
I0428 14:31:49.328613  7476 sgd_solver.cpp:105] Iteration 4920, lr = 0.0037735
I0428 14:31:54.615561  7476 solver.cpp:218] Iteration 4932 (2.2698 iter/s, 5.2868s/12 iters), loss = 1.81578
I0428 14:31:54.615605  7476 solver.cpp:237]     Train net output #0: loss = 1.81578 (* 1 = 1.81578 loss)
I0428 14:31:54.615614  7476 sgd_solver.cpp:105] Iteration 4932, lr = 0.00376454
I0428 14:32:00.194314  7476 solver.cpp:218] Iteration 4944 (2.1511 iter/s, 5.57854s/12 iters), loss = 1.567
I0428 14:32:00.194362  7476 solver.cpp:237]     Train net output #0: loss = 1.567 (* 1 = 1.567 loss)
I0428 14:32:00.194372  7476 sgd_solver.cpp:105] Iteration 4944, lr = 0.0037556
I0428 14:32:05.413758  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:32:05.662420  7476 solver.cpp:218] Iteration 4956 (2.19463 iter/s, 5.46789s/12 iters), loss = 1.73827
I0428 14:32:05.662472  7476 solver.cpp:237]     Train net output #0: loss = 1.73827 (* 1 = 1.73827 loss)
I0428 14:32:05.662482  7476 sgd_solver.cpp:105] Iteration 4956, lr = 0.00374669
I0428 14:32:11.100891  7476 solver.cpp:218] Iteration 4968 (2.20659 iter/s, 5.43826s/12 iters), loss = 1.93395
I0428 14:32:11.101023  7476 solver.cpp:237]     Train net output #0: loss = 1.93395 (* 1 = 1.93395 loss)
I0428 14:32:11.101033  7476 sgd_solver.cpp:105] Iteration 4968, lr = 0.00373779
I0428 14:32:16.595487  7476 solver.cpp:218] Iteration 4980 (2.18408 iter/s, 5.4943s/12 iters), loss = 1.807
I0428 14:32:16.595526  7476 solver.cpp:237]     Train net output #0: loss = 1.807 (* 1 = 1.807 loss)
I0428 14:32:16.595535  7476 sgd_solver.cpp:105] Iteration 4980, lr = 0.00372892
I0428 14:32:22.115870  7476 solver.cpp:218] Iteration 4992 (2.17384 iter/s, 5.52018s/12 iters), loss = 1.64443
I0428 14:32:22.115916  7476 solver.cpp:237]     Train net output #0: loss = 1.64443 (* 1 = 1.64443 loss)
I0428 14:32:22.115924  7476 sgd_solver.cpp:105] Iteration 4992, lr = 0.00372006
I0428 14:32:24.294819  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_4998.caffemodel
I0428 14:32:28.229341  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_4998.solverstate
I0428 14:32:30.643210  7476 solver.cpp:330] Iteration 4998, Testing net (#0)
I0428 14:32:30.643236  7476 net.cpp:676] Ignoring source layer train-data
I0428 14:32:33.419448  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:32:35.412065  7476 solver.cpp:397]     Test net output #0: accuracy = 0.306985
I0428 14:32:35.412124  7476 solver.cpp:397]     Test net output #1: loss = 3.0635 (* 1 = 3.0635 loss)
I0428 14:32:37.511791  7476 solver.cpp:218] Iteration 5004 (0.779451 iter/s, 15.3955s/12 iters), loss = 1.61348
I0428 14:32:37.511831  7476 solver.cpp:237]     Train net output #0: loss = 1.61348 (* 1 = 1.61348 loss)
I0428 14:32:37.511839  7476 sgd_solver.cpp:105] Iteration 5004, lr = 0.00371123
I0428 14:32:42.821575  7476 solver.cpp:218] Iteration 5016 (2.26006 iter/s, 5.30958s/12 iters), loss = 1.99462
I0428 14:32:42.821679  7476 solver.cpp:237]     Train net output #0: loss = 1.99462 (* 1 = 1.99462 loss)
I0428 14:32:42.821689  7476 sgd_solver.cpp:105] Iteration 5016, lr = 0.00370242
I0428 14:32:48.311082  7476 solver.cpp:218] Iteration 5028 (2.1861 iter/s, 5.48923s/12 iters), loss = 1.86308
I0428 14:32:48.311139  7476 solver.cpp:237]     Train net output #0: loss = 1.86308 (* 1 = 1.86308 loss)
I0428 14:32:48.311151  7476 sgd_solver.cpp:105] Iteration 5028, lr = 0.00369363
I0428 14:32:54.043541  7476 solver.cpp:218] Iteration 5040 (2.09343 iter/s, 5.73222s/12 iters), loss = 1.6254
I0428 14:32:54.043603  7476 solver.cpp:237]     Train net output #0: loss = 1.6254 (* 1 = 1.6254 loss)
I0428 14:32:54.043618  7476 sgd_solver.cpp:105] Iteration 5040, lr = 0.00368486
I0428 14:32:59.575137  7476 solver.cpp:218] Iteration 5052 (2.16944 iter/s, 5.53137s/12 iters), loss = 1.55194
I0428 14:32:59.575186  7476 solver.cpp:237]     Train net output #0: loss = 1.55194 (* 1 = 1.55194 loss)
I0428 14:32:59.575194  7476 sgd_solver.cpp:105] Iteration 5052, lr = 0.00367611
I0428 14:33:01.741346  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:33:05.235934  7476 solver.cpp:218] Iteration 5064 (2.11993 iter/s, 5.66058s/12 iters), loss = 1.64708
I0428 14:33:05.235981  7476 solver.cpp:237]     Train net output #0: loss = 1.64708 (* 1 = 1.64708 loss)
I0428 14:33:05.235989  7476 sgd_solver.cpp:105] Iteration 5064, lr = 0.00366738
I0428 14:33:10.723762  7476 solver.cpp:218] Iteration 5076 (2.18674 iter/s, 5.48762s/12 iters), loss = 1.77496
I0428 14:33:10.723798  7476 solver.cpp:237]     Train net output #0: loss = 1.77496 (* 1 = 1.77496 loss)
I0428 14:33:10.723806  7476 sgd_solver.cpp:105] Iteration 5076, lr = 0.00365868
I0428 14:33:16.217468  7476 solver.cpp:218] Iteration 5088 (2.1844 iter/s, 5.4935s/12 iters), loss = 1.67177
I0428 14:33:16.223405  7476 solver.cpp:237]     Train net output #0: loss = 1.67177 (* 1 = 1.67177 loss)
I0428 14:33:16.223417  7476 sgd_solver.cpp:105] Iteration 5088, lr = 0.00364999
I0428 14:33:21.203718  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_5100.caffemodel
I0428 14:33:23.933248  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_5100.solverstate
I0428 14:33:26.365761  7476 solver.cpp:330] Iteration 5100, Testing net (#0)
I0428 14:33:26.365782  7476 net.cpp:676] Ignoring source layer train-data
I0428 14:33:28.832680  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:33:30.925226  7476 solver.cpp:397]     Test net output #0: accuracy = 0.333333
I0428 14:33:30.925253  7476 solver.cpp:397]     Test net output #1: loss = 2.93428 (* 1 = 2.93428 loss)
I0428 14:33:31.053824  7476 solver.cpp:218] Iteration 5100 (0.80917 iter/s, 14.83s/12 iters), loss = 1.60465
I0428 14:33:31.053874  7476 solver.cpp:237]     Train net output #0: loss = 1.60465 (* 1 = 1.60465 loss)
I0428 14:33:31.053884  7476 sgd_solver.cpp:105] Iteration 5100, lr = 0.00364132
I0428 14:33:35.675817  7476 solver.cpp:218] Iteration 5112 (2.59639 iter/s, 4.6218s/12 iters), loss = 1.75871
I0428 14:33:35.675866  7476 solver.cpp:237]     Train net output #0: loss = 1.75871 (* 1 = 1.75871 loss)
I0428 14:33:35.675875  7476 sgd_solver.cpp:105] Iteration 5112, lr = 0.00363268
I0428 14:33:41.029950  7476 solver.cpp:218] Iteration 5124 (2.24135 iter/s, 5.35391s/12 iters), loss = 1.88634
I0428 14:33:41.030007  7476 solver.cpp:237]     Train net output #0: loss = 1.88634 (* 1 = 1.88634 loss)
I0428 14:33:41.030022  7476 sgd_solver.cpp:105] Iteration 5124, lr = 0.00362405
I0428 14:33:46.692189  7476 solver.cpp:218] Iteration 5136 (2.11939 iter/s, 5.66201s/12 iters), loss = 1.6526
I0428 14:33:46.692313  7476 solver.cpp:237]     Train net output #0: loss = 1.6526 (* 1 = 1.6526 loss)
I0428 14:33:46.692324  7476 sgd_solver.cpp:105] Iteration 5136, lr = 0.00361545
I0428 14:33:52.340025  7476 solver.cpp:218] Iteration 5148 (2.12482 iter/s, 5.64754s/12 iters), loss = 1.60349
I0428 14:33:52.340085  7476 solver.cpp:237]     Train net output #0: loss = 1.60349 (* 1 = 1.60349 loss)
I0428 14:33:52.340098  7476 sgd_solver.cpp:105] Iteration 5148, lr = 0.00360687
I0428 14:33:56.657526  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:33:57.715922  7476 solver.cpp:218] Iteration 5160 (2.23227 iter/s, 5.37568s/12 iters), loss = 1.55422
I0428 14:33:57.715968  7476 solver.cpp:237]     Train net output #0: loss = 1.55422 (* 1 = 1.55422 loss)
I0428 14:33:57.715978  7476 sgd_solver.cpp:105] Iteration 5160, lr = 0.0035983
I0428 14:34:03.832412  7476 solver.cpp:218] Iteration 5172 (1.96199 iter/s, 6.11625s/12 iters), loss = 1.69581
I0428 14:34:03.832706  7476 solver.cpp:237]     Train net output #0: loss = 1.69581 (* 1 = 1.69581 loss)
I0428 14:34:03.832723  7476 sgd_solver.cpp:105] Iteration 5172, lr = 0.00358976
I0428 14:34:10.670455  7476 solver.cpp:218] Iteration 5184 (1.75501 iter/s, 6.83755s/12 iters), loss = 1.4031
I0428 14:34:10.670521  7476 solver.cpp:237]     Train net output #0: loss = 1.4031 (* 1 = 1.4031 loss)
I0428 14:34:10.670533  7476 sgd_solver.cpp:105] Iteration 5184, lr = 0.00358124
I0428 14:34:16.749269  7476 solver.cpp:218] Iteration 5196 (1.97415 iter/s, 6.07857s/12 iters), loss = 1.46794
I0428 14:34:16.752192  7476 solver.cpp:237]     Train net output #0: loss = 1.46794 (* 1 = 1.46794 loss)
I0428 14:34:16.752203  7476 sgd_solver.cpp:105] Iteration 5196, lr = 0.00357273
I0428 14:34:19.085738  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_5202.caffemodel
I0428 14:34:20.469543  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_5202.solverstate
I0428 14:34:22.795459  7476 solver.cpp:330] Iteration 5202, Testing net (#0)
I0428 14:34:22.795485  7476 net.cpp:676] Ignoring source layer train-data
I0428 14:34:25.224563  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:34:27.299461  7476 solver.cpp:397]     Test net output #0: accuracy = 0.326593
I0428 14:34:27.299499  7476 solver.cpp:397]     Test net output #1: loss = 2.84762 (* 1 = 2.84762 loss)
I0428 14:34:29.257215  7476 solver.cpp:218] Iteration 5208 (0.959641 iter/s, 12.5047s/12 iters), loss = 1.53876
I0428 14:34:29.257259  7476 solver.cpp:237]     Train net output #0: loss = 1.53876 (* 1 = 1.53876 loss)
I0428 14:34:29.257269  7476 sgd_solver.cpp:105] Iteration 5208, lr = 0.00356425
I0428 14:34:34.733645  7476 solver.cpp:218] Iteration 5220 (2.19129 iter/s, 5.47622s/12 iters), loss = 1.77439
I0428 14:34:34.733682  7476 solver.cpp:237]     Train net output #0: loss = 1.77439 (* 1 = 1.77439 loss)
I0428 14:34:34.733690  7476 sgd_solver.cpp:105] Iteration 5220, lr = 0.00355579
I0428 14:34:40.373168  7476 solver.cpp:218] Iteration 5232 (2.12792 iter/s, 5.63932s/12 iters), loss = 1.74172
I0428 14:34:40.373209  7476 solver.cpp:237]     Train net output #0: loss = 1.74172 (* 1 = 1.74172 loss)
I0428 14:34:40.373217  7476 sgd_solver.cpp:105] Iteration 5232, lr = 0.00354735
I0428 14:34:45.938704  7476 solver.cpp:218] Iteration 5244 (2.15621 iter/s, 5.56533s/12 iters), loss = 1.45871
I0428 14:34:45.938741  7476 solver.cpp:237]     Train net output #0: loss = 1.45871 (* 1 = 1.45871 loss)
I0428 14:34:45.938751  7476 sgd_solver.cpp:105] Iteration 5244, lr = 0.00353892
I0428 14:34:51.447369  7476 solver.cpp:218] Iteration 5256 (2.17847 iter/s, 5.50846s/12 iters), loss = 1.1405
I0428 14:34:51.447474  7476 solver.cpp:237]     Train net output #0: loss = 1.1405 (* 1 = 1.1405 loss)
I0428 14:34:51.447482  7476 sgd_solver.cpp:105] Iteration 5256, lr = 0.00353052
I0428 14:34:52.864331  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:34:57.018064  7476 solver.cpp:218] Iteration 5268 (2.15424 iter/s, 5.57042s/12 iters), loss = 1.51513
I0428 14:34:57.018110  7476 solver.cpp:237]     Train net output #0: loss = 1.51513 (* 1 = 1.51513 loss)
I0428 14:34:57.018119  7476 sgd_solver.cpp:105] Iteration 5268, lr = 0.00352214
I0428 14:35:02.561777  7476 solver.cpp:218] Iteration 5280 (2.1647 iter/s, 5.5435s/12 iters), loss = 1.70143
I0428 14:35:02.561820  7476 solver.cpp:237]     Train net output #0: loss = 1.70143 (* 1 = 1.70143 loss)
I0428 14:35:02.561828  7476 sgd_solver.cpp:105] Iteration 5280, lr = 0.00351378
I0428 14:35:07.996348  7476 solver.cpp:218] Iteration 5292 (2.20817 iter/s, 5.43436s/12 iters), loss = 1.37956
I0428 14:35:07.996393  7476 solver.cpp:237]     Train net output #0: loss = 1.37956 (* 1 = 1.37956 loss)
I0428 14:35:07.996402  7476 sgd_solver.cpp:105] Iteration 5292, lr = 0.00350544
I0428 14:35:13.129698  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_5304.caffemodel
I0428 14:35:14.521569  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_5304.solverstate
I0428 14:35:15.569342  7476 solver.cpp:330] Iteration 5304, Testing net (#0)
I0428 14:35:15.569366  7476 net.cpp:676] Ignoring source layer train-data
I0428 14:35:17.865100  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:35:20.072777  7476 solver.cpp:397]     Test net output #0: accuracy = 0.340686
I0428 14:35:20.072806  7476 solver.cpp:397]     Test net output #1: loss = 2.90407 (* 1 = 2.90407 loss)
I0428 14:35:20.200913  7476 solver.cpp:218] Iteration 5304 (0.98327 iter/s, 12.2042s/12 iters), loss = 1.9261
I0428 14:35:20.200959  7476 solver.cpp:237]     Train net output #0: loss = 1.9261 (* 1 = 1.9261 loss)
I0428 14:35:20.200968  7476 sgd_solver.cpp:105] Iteration 5304, lr = 0.00349711
I0428 14:35:24.814787  7476 solver.cpp:218] Iteration 5316 (2.60095 iter/s, 4.61369s/12 iters), loss = 1.42789
I0428 14:35:24.814924  7476 solver.cpp:237]     Train net output #0: loss = 1.42789 (* 1 = 1.42789 loss)
I0428 14:35:24.814934  7476 sgd_solver.cpp:105] Iteration 5316, lr = 0.00348881
I0428 14:35:30.266933  7476 solver.cpp:218] Iteration 5328 (2.20109 iter/s, 5.45185s/12 iters), loss = 1.68011
I0428 14:35:30.266973  7476 solver.cpp:237]     Train net output #0: loss = 1.68011 (* 1 = 1.68011 loss)
I0428 14:35:30.266981  7476 sgd_solver.cpp:105] Iteration 5328, lr = 0.00348053
I0428 14:35:35.729279  7476 solver.cpp:218] Iteration 5340 (2.19694 iter/s, 5.46214s/12 iters), loss = 1.47225
I0428 14:35:35.729319  7476 solver.cpp:237]     Train net output #0: loss = 1.47225 (* 1 = 1.47225 loss)
I0428 14:35:35.729327  7476 sgd_solver.cpp:105] Iteration 5340, lr = 0.00347226
I0428 14:35:41.330010  7476 solver.cpp:218] Iteration 5352 (2.14266 iter/s, 5.60053s/12 iters), loss = 1.28963
I0428 14:35:41.330052  7476 solver.cpp:237]     Train net output #0: loss = 1.28963 (* 1 = 1.28963 loss)
I0428 14:35:41.330060  7476 sgd_solver.cpp:105] Iteration 5352, lr = 0.00346402
I0428 14:35:45.048044  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:35:46.810911  7476 solver.cpp:218] Iteration 5364 (2.1895 iter/s, 5.4807s/12 iters), loss = 1.54825
I0428 14:35:46.810952  7476 solver.cpp:237]     Train net output #0: loss = 1.54825 (* 1 = 1.54825 loss)
I0428 14:35:46.810961  7476 sgd_solver.cpp:105] Iteration 5364, lr = 0.0034558
I0428 14:35:52.252946  7476 solver.cpp:218] Iteration 5376 (2.20514 iter/s, 5.44184s/12 iters), loss = 1.76494
I0428 14:35:52.252991  7476 solver.cpp:237]     Train net output #0: loss = 1.76494 (* 1 = 1.76494 loss)
I0428 14:35:52.253000  7476 sgd_solver.cpp:105] Iteration 5376, lr = 0.00344759
I0428 14:35:57.753991  7476 solver.cpp:218] Iteration 5388 (2.18149 iter/s, 5.50083s/12 iters), loss = 1.72291
I0428 14:35:57.754108  7476 solver.cpp:237]     Train net output #0: loss = 1.72291 (* 1 = 1.72291 loss)
I0428 14:35:57.754119  7476 sgd_solver.cpp:105] Iteration 5388, lr = 0.00343941
I0428 14:36:03.349222  7476 solver.cpp:218] Iteration 5400 (2.14479 iter/s, 5.59495s/12 iters), loss = 1.24077
I0428 14:36:03.349269  7476 solver.cpp:237]     Train net output #0: loss = 1.24077 (* 1 = 1.24077 loss)
I0428 14:36:03.349277  7476 sgd_solver.cpp:105] Iteration 5400, lr = 0.00343124
I0428 14:36:05.614367  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_5406.caffemodel
I0428 14:36:06.956200  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_5406.solverstate
I0428 14:36:08.009204  7476 solver.cpp:330] Iteration 5406, Testing net (#0)
I0428 14:36:08.009232  7476 net.cpp:676] Ignoring source layer train-data
I0428 14:36:10.323792  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:36:12.526290  7476 solver.cpp:397]     Test net output #0: accuracy = 0.349265
I0428 14:36:12.526320  7476 solver.cpp:397]     Test net output #1: loss = 2.8108 (* 1 = 2.8108 loss)
I0428 14:36:14.491905  7476 solver.cpp:218] Iteration 5412 (1.07697 iter/s, 11.1423s/12 iters), loss = 1.40934
I0428 14:36:14.491961  7476 solver.cpp:237]     Train net output #0: loss = 1.40934 (* 1 = 1.40934 loss)
I0428 14:36:14.491972  7476 sgd_solver.cpp:105] Iteration 5412, lr = 0.00342309
I0428 14:36:19.917299  7476 solver.cpp:218] Iteration 5424 (2.21191 iter/s, 5.42518s/12 iters), loss = 1.68688
I0428 14:36:19.917340  7476 solver.cpp:237]     Train net output #0: loss = 1.68688 (* 1 = 1.68688 loss)
I0428 14:36:19.917348  7476 sgd_solver.cpp:105] Iteration 5424, lr = 0.00341497
I0428 14:36:25.514403  7476 solver.cpp:218] Iteration 5436 (2.14404 iter/s, 5.5969s/12 iters), loss = 1.57546
I0428 14:36:25.514447  7476 solver.cpp:237]     Train net output #0: loss = 1.57546 (* 1 = 1.57546 loss)
I0428 14:36:25.514457  7476 sgd_solver.cpp:105] Iteration 5436, lr = 0.00340686
I0428 14:36:31.171864  7476 solver.cpp:218] Iteration 5448 (2.12117 iter/s, 5.65725s/12 iters), loss = 1.61214
I0428 14:36:31.171980  7476 solver.cpp:237]     Train net output #0: loss = 1.61214 (* 1 = 1.61214 loss)
I0428 14:36:31.171991  7476 sgd_solver.cpp:105] Iteration 5448, lr = 0.00339877
I0428 14:36:36.848455  7476 solver.cpp:218] Iteration 5460 (2.11405 iter/s, 5.67631s/12 iters), loss = 1.60954
I0428 14:36:36.848515  7476 solver.cpp:237]     Train net output #0: loss = 1.60954 (* 1 = 1.60954 loss)
I0428 14:36:36.848523  7476 sgd_solver.cpp:105] Iteration 5460, lr = 0.0033907
I0428 14:36:37.450089  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:36:42.296137  7476 solver.cpp:218] Iteration 5472 (2.20286 iter/s, 5.44746s/12 iters), loss = 1.42157
I0428 14:36:42.296177  7476 solver.cpp:237]     Train net output #0: loss = 1.42157 (* 1 = 1.42157 loss)
I0428 14:36:42.296185  7476 sgd_solver.cpp:105] Iteration 5472, lr = 0.00338265
I0428 14:36:47.766911  7476 solver.cpp:218] Iteration 5484 (2.19355 iter/s, 5.47058s/12 iters), loss = 1.33158
I0428 14:36:47.766950  7476 solver.cpp:237]     Train net output #0: loss = 1.33158 (* 1 = 1.33158 loss)
I0428 14:36:47.766959  7476 sgd_solver.cpp:105] Iteration 5484, lr = 0.00337462
I0428 14:36:53.191273  7476 solver.cpp:218] Iteration 5496 (2.21232 iter/s, 5.42416s/12 iters), loss = 1.43742
I0428 14:36:53.191313  7476 solver.cpp:237]     Train net output #0: loss = 1.43742 (* 1 = 1.43742 loss)
I0428 14:36:53.191321  7476 sgd_solver.cpp:105] Iteration 5496, lr = 0.00336661
I0428 14:36:58.112239  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_5508.caffemodel
I0428 14:37:00.195202  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_5508.solverstate
I0428 14:37:02.074924  7476 solver.cpp:330] Iteration 5508, Testing net (#0)
I0428 14:37:02.075024  7476 net.cpp:676] Ignoring source layer train-data
I0428 14:37:04.440783  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:37:06.648818  7476 solver.cpp:397]     Test net output #0: accuracy = 0.365196
I0428 14:37:06.648864  7476 solver.cpp:397]     Test net output #1: loss = 2.80982 (* 1 = 2.80982 loss)
I0428 14:37:06.777222  7476 solver.cpp:218] Iteration 5508 (0.883292 iter/s, 13.5855s/12 iters), loss = 1.41385
I0428 14:37:06.777279  7476 solver.cpp:237]     Train net output #0: loss = 1.41385 (* 1 = 1.41385 loss)
I0428 14:37:06.777290  7476 sgd_solver.cpp:105] Iteration 5508, lr = 0.00335861
I0428 14:37:11.301317  7476 solver.cpp:218] Iteration 5520 (2.65258 iter/s, 4.5239s/12 iters), loss = 1.38316
I0428 14:37:11.301373  7476 solver.cpp:237]     Train net output #0: loss = 1.38316 (* 1 = 1.38316 loss)
I0428 14:37:11.301386  7476 sgd_solver.cpp:105] Iteration 5520, lr = 0.00335064
I0428 14:37:13.988225  7476 blocking_queue.cpp:49] Waiting for data
I0428 14:37:16.856027  7476 solver.cpp:218] Iteration 5532 (2.16041 iter/s, 5.55449s/12 iters), loss = 1.59941
I0428 14:37:16.856070  7476 solver.cpp:237]     Train net output #0: loss = 1.59941 (* 1 = 1.59941 loss)
I0428 14:37:16.856079  7476 sgd_solver.cpp:105] Iteration 5532, lr = 0.00334268
I0428 14:37:22.220996  7476 solver.cpp:218] Iteration 5544 (2.23682 iter/s, 5.36476s/12 iters), loss = 1.34161
I0428 14:37:22.221062  7476 solver.cpp:237]     Train net output #0: loss = 1.34161 (* 1 = 1.34161 loss)
I0428 14:37:22.221074  7476 sgd_solver.cpp:105] Iteration 5544, lr = 0.00333475
I0428 14:37:27.717744  7476 solver.cpp:218] Iteration 5556 (2.1832 iter/s, 5.49652s/12 iters), loss = 1.50074
I0428 14:37:27.717797  7476 solver.cpp:237]     Train net output #0: loss = 1.50074 (* 1 = 1.50074 loss)
I0428 14:37:27.717811  7476 sgd_solver.cpp:105] Iteration 5556, lr = 0.00332683
I0428 14:37:30.678828  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:37:33.247467  7476 solver.cpp:218] Iteration 5568 (2.17017 iter/s, 5.52951s/12 iters), loss = 1.37347
I0428 14:37:33.247611  7476 solver.cpp:237]     Train net output #0: loss = 1.37347 (* 1 = 1.37347 loss)
I0428 14:37:33.247622  7476 sgd_solver.cpp:105] Iteration 5568, lr = 0.00331893
I0428 14:37:38.744446  7476 solver.cpp:218] Iteration 5580 (2.18314 iter/s, 5.49668s/12 iters), loss = 1.58843
I0428 14:37:38.744513  7476 solver.cpp:237]     Train net output #0: loss = 1.58843 (* 1 = 1.58843 loss)
I0428 14:37:38.744522  7476 sgd_solver.cpp:105] Iteration 5580, lr = 0.00331105
I0428 14:37:44.358135  7476 solver.cpp:218] Iteration 5592 (2.13771 iter/s, 5.61348s/12 iters), loss = 1.28476
I0428 14:37:44.358179  7476 solver.cpp:237]     Train net output #0: loss = 1.28476 (* 1 = 1.28476 loss)
I0428 14:37:44.358187  7476 sgd_solver.cpp:105] Iteration 5592, lr = 0.00330319
I0428 14:37:49.664115  7476 solver.cpp:218] Iteration 5604 (2.26168 iter/s, 5.30578s/12 iters), loss = 1.30813
I0428 14:37:49.664155  7476 solver.cpp:237]     Train net output #0: loss = 1.30813 (* 1 = 1.30813 loss)
I0428 14:37:49.664163  7476 sgd_solver.cpp:105] Iteration 5604, lr = 0.00329535
I0428 14:37:51.783504  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_5610.caffemodel
I0428 14:37:58.802147  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_5610.solverstate
I0428 14:38:01.824388  7476 solver.cpp:330] Iteration 5610, Testing net (#0)
I0428 14:38:01.824414  7476 net.cpp:676] Ignoring source layer train-data
I0428 14:38:04.190949  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:38:06.445343  7476 solver.cpp:397]     Test net output #0: accuracy = 0.352941
I0428 14:38:06.445391  7476 solver.cpp:397]     Test net output #1: loss = 2.98063 (* 1 = 2.98063 loss)
I0428 14:38:08.438630  7476 solver.cpp:218] Iteration 5616 (0.639183 iter/s, 18.774s/12 iters), loss = 1.31757
I0428 14:38:08.438673  7476 solver.cpp:237]     Train net output #0: loss = 1.31757 (* 1 = 1.31757 loss)
I0428 14:38:08.438680  7476 sgd_solver.cpp:105] Iteration 5616, lr = 0.00328752
I0428 14:38:13.872118  7476 solver.cpp:218] Iteration 5628 (2.20861 iter/s, 5.43329s/12 iters), loss = 1.25665
I0428 14:38:13.872160  7476 solver.cpp:237]     Train net output #0: loss = 1.25665 (* 1 = 1.25665 loss)
I0428 14:38:13.872169  7476 sgd_solver.cpp:105] Iteration 5628, lr = 0.00327972
I0428 14:38:19.297341  7476 solver.cpp:218] Iteration 5640 (2.21197 iter/s, 5.42502s/12 iters), loss = 1.53601
I0428 14:38:19.297381  7476 solver.cpp:237]     Train net output #0: loss = 1.53601 (* 1 = 1.53601 loss)
I0428 14:38:19.297390  7476 sgd_solver.cpp:105] Iteration 5640, lr = 0.00327193
I0428 14:38:24.753499  7476 solver.cpp:218] Iteration 5652 (2.19943 iter/s, 5.45596s/12 iters), loss = 1.26448
I0428 14:38:24.753546  7476 solver.cpp:237]     Train net output #0: loss = 1.26448 (* 1 = 1.26448 loss)
I0428 14:38:24.753557  7476 sgd_solver.cpp:105] Iteration 5652, lr = 0.00326416
I0428 14:38:30.039700  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:38:30.256527  7476 solver.cpp:218] Iteration 5664 (2.18071 iter/s, 5.5028s/12 iters), loss = 1.07121
I0428 14:38:30.256569  7476 solver.cpp:237]     Train net output #0: loss = 1.07121 (* 1 = 1.07121 loss)
I0428 14:38:30.256579  7476 sgd_solver.cpp:105] Iteration 5664, lr = 0.00325641
I0428 14:38:35.711462  7476 solver.cpp:218] Iteration 5676 (2.19993 iter/s, 5.45473s/12 iters), loss = 1.33967
I0428 14:38:35.711573  7476 solver.cpp:237]     Train net output #0: loss = 1.33967 (* 1 = 1.33967 loss)
I0428 14:38:35.711582  7476 sgd_solver.cpp:105] Iteration 5676, lr = 0.00324868
I0428 14:38:41.132421  7476 solver.cpp:218] Iteration 5688 (2.21374 iter/s, 5.42068s/12 iters), loss = 1.22409
I0428 14:38:41.132478  7476 solver.cpp:237]     Train net output #0: loss = 1.22409 (* 1 = 1.22409 loss)
I0428 14:38:41.132519  7476 sgd_solver.cpp:105] Iteration 5688, lr = 0.00324097
I0428 14:38:46.649483  7476 solver.cpp:218] Iteration 5700 (2.17515 iter/s, 5.51685s/12 iters), loss = 1.38881
I0428 14:38:46.649520  7476 solver.cpp:237]     Train net output #0: loss = 1.38881 (* 1 = 1.38881 loss)
I0428 14:38:46.649528  7476 sgd_solver.cpp:105] Iteration 5700, lr = 0.00323328
I0428 14:38:51.624675  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_5712.caffemodel
I0428 14:38:53.719087  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_5712.solverstate
I0428 14:38:54.908442  7476 solver.cpp:330] Iteration 5712, Testing net (#0)
I0428 14:38:54.908463  7476 net.cpp:676] Ignoring source layer train-data
I0428 14:38:57.029923  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:38:59.549448  7476 solver.cpp:397]     Test net output #0: accuracy = 0.370711
I0428 14:38:59.549481  7476 solver.cpp:397]     Test net output #1: loss = 2.84518 (* 1 = 2.84518 loss)
I0428 14:38:59.677939  7476 solver.cpp:218] Iteration 5712 (0.921089 iter/s, 13.0281s/12 iters), loss = 1.27282
I0428 14:38:59.677983  7476 solver.cpp:237]     Train net output #0: loss = 1.27282 (* 1 = 1.27282 loss)
I0428 14:38:59.677992  7476 sgd_solver.cpp:105] Iteration 5712, lr = 0.0032256
I0428 14:39:04.203855  7476 solver.cpp:218] Iteration 5724 (2.6515 iter/s, 4.52574s/12 iters), loss = 1.27034
I0428 14:39:04.203892  7476 solver.cpp:237]     Train net output #0: loss = 1.27034 (* 1 = 1.27034 loss)
I0428 14:39:04.203900  7476 sgd_solver.cpp:105] Iteration 5724, lr = 0.00321794
I0428 14:39:09.596293  7476 solver.cpp:218] Iteration 5736 (2.22542 iter/s, 5.39224s/12 iters), loss = 1.49248
I0428 14:39:09.596444  7476 solver.cpp:237]     Train net output #0: loss = 1.49248 (* 1 = 1.49248 loss)
I0428 14:39:09.596457  7476 sgd_solver.cpp:105] Iteration 5736, lr = 0.0032103
I0428 14:39:15.038094  7476 solver.cpp:218] Iteration 5748 (2.20528 iter/s, 5.44149s/12 iters), loss = 1.31564
I0428 14:39:15.038137  7476 solver.cpp:237]     Train net output #0: loss = 1.31564 (* 1 = 1.31564 loss)
I0428 14:39:15.038146  7476 sgd_solver.cpp:105] Iteration 5748, lr = 0.00320268
I0428 14:39:20.468974  7476 solver.cpp:218] Iteration 5760 (2.20967 iter/s, 5.43067s/12 iters), loss = 1.07287
I0428 14:39:20.469018  7476 solver.cpp:237]     Train net output #0: loss = 1.07287 (* 1 = 1.07287 loss)
I0428 14:39:20.469027  7476 sgd_solver.cpp:105] Iteration 5760, lr = 0.00319508
I0428 14:39:22.593353  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:39:26.026096  7476 solver.cpp:218] Iteration 5772 (2.15947 iter/s, 5.55691s/12 iters), loss = 1.13276
I0428 14:39:26.026142  7476 solver.cpp:237]     Train net output #0: loss = 1.13276 (* 1 = 1.13276 loss)
I0428 14:39:26.026150  7476 sgd_solver.cpp:105] Iteration 5772, lr = 0.00318749
I0428 14:39:31.492269  7476 solver.cpp:218] Iteration 5784 (2.1954 iter/s, 5.46596s/12 iters), loss = 1.56758
I0428 14:39:31.492321  7476 solver.cpp:237]     Train net output #0: loss = 1.56758 (* 1 = 1.56758 loss)
I0428 14:39:31.492332  7476 sgd_solver.cpp:105] Iteration 5784, lr = 0.00317992
I0428 14:39:37.154873  7476 solver.cpp:218] Iteration 5796 (2.11925 iter/s, 5.66239s/12 iters), loss = 1.14895
I0428 14:39:37.154919  7476 solver.cpp:237]     Train net output #0: loss = 1.14895 (* 1 = 1.14895 loss)
I0428 14:39:37.154927  7476 sgd_solver.cpp:105] Iteration 5796, lr = 0.00317237
I0428 14:39:42.596189  7476 solver.cpp:218] Iteration 5808 (2.20543 iter/s, 5.44111s/12 iters), loss = 1.21972
I0428 14:39:42.596292  7476 solver.cpp:237]     Train net output #0: loss = 1.21972 (* 1 = 1.21972 loss)
I0428 14:39:42.596302  7476 sgd_solver.cpp:105] Iteration 5808, lr = 0.00316484
I0428 14:39:44.806291  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_5814.caffemodel
I0428 14:39:46.104403  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_5814.solverstate
I0428 14:39:47.134927  7476 solver.cpp:330] Iteration 5814, Testing net (#0)
I0428 14:39:47.134949  7476 net.cpp:676] Ignoring source layer train-data
I0428 14:39:49.213235  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:39:51.578354  7476 solver.cpp:397]     Test net output #0: accuracy = 0.373162
I0428 14:39:51.578390  7476 solver.cpp:397]     Test net output #1: loss = 2.89871 (* 1 = 2.89871 loss)
I0428 14:39:53.479377  7476 solver.cpp:218] Iteration 5820 (1.10266 iter/s, 10.8828s/12 iters), loss = 1.40304
I0428 14:39:53.479425  7476 solver.cpp:237]     Train net output #0: loss = 1.40304 (* 1 = 1.40304 loss)
I0428 14:39:53.479434  7476 sgd_solver.cpp:105] Iteration 5820, lr = 0.00315733
I0428 14:39:58.949995  7476 solver.cpp:218] Iteration 5832 (2.19362 iter/s, 5.47041s/12 iters), loss = 1.27118
I0428 14:39:58.950038  7476 solver.cpp:237]     Train net output #0: loss = 1.27118 (* 1 = 1.27118 loss)
I0428 14:39:58.950047  7476 sgd_solver.cpp:105] Iteration 5832, lr = 0.00314983
I0428 14:40:04.344236  7476 solver.cpp:218] Iteration 5844 (2.22468 iter/s, 5.39403s/12 iters), loss = 1.18161
I0428 14:40:04.344295  7476 solver.cpp:237]     Train net output #0: loss = 1.18161 (* 1 = 1.18161 loss)
I0428 14:40:04.344305  7476 sgd_solver.cpp:105] Iteration 5844, lr = 0.00314235
I0428 14:40:09.960526  7476 solver.cpp:218] Iteration 5856 (2.13673 iter/s, 5.61605s/12 iters), loss = 1.181
I0428 14:40:09.960572  7476 solver.cpp:237]     Train net output #0: loss = 1.181 (* 1 = 1.181 loss)
I0428 14:40:09.960582  7476 sgd_solver.cpp:105] Iteration 5856, lr = 0.00313489
I0428 14:40:14.511687  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:40:15.420639  7476 solver.cpp:218] Iteration 5868 (2.19784 iter/s, 5.45991s/12 iters), loss = 1.24591
I0428 14:40:15.420684  7476 solver.cpp:237]     Train net output #0: loss = 1.24591 (* 1 = 1.24591 loss)
I0428 14:40:15.420693  7476 sgd_solver.cpp:105] Iteration 5868, lr = 0.00312745
I0428 14:40:20.854084  7476 solver.cpp:218] Iteration 5880 (2.20863 iter/s, 5.43324s/12 iters), loss = 1.29368
I0428 14:40:20.854122  7476 solver.cpp:237]     Train net output #0: loss = 1.29368 (* 1 = 1.29368 loss)
I0428 14:40:20.854131  7476 sgd_solver.cpp:105] Iteration 5880, lr = 0.00312002
I0428 14:40:26.251807  7476 solver.cpp:218] Iteration 5892 (2.22325 iter/s, 5.39751s/12 iters), loss = 1.31008
I0428 14:40:26.251878  7476 solver.cpp:237]     Train net output #0: loss = 1.31008 (* 1 = 1.31008 loss)
I0428 14:40:26.251889  7476 sgd_solver.cpp:105] Iteration 5892, lr = 0.00311262
I0428 14:40:31.941190  7476 solver.cpp:218] Iteration 5904 (2.10927 iter/s, 5.68918s/12 iters), loss = 1.21637
I0428 14:40:31.941233  7476 solver.cpp:237]     Train net output #0: loss = 1.21637 (* 1 = 1.21637 loss)
I0428 14:40:31.941241  7476 sgd_solver.cpp:105] Iteration 5904, lr = 0.00310523
I0428 14:40:36.867874  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_5916.caffemodel
I0428 14:40:38.268899  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_5916.solverstate
I0428 14:40:39.295332  7476 solver.cpp:330] Iteration 5916, Testing net (#0)
I0428 14:40:39.295353  7476 net.cpp:676] Ignoring source layer train-data
I0428 14:40:41.402671  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:40:43.760859  7476 solver.cpp:397]     Test net output #0: accuracy = 0.348652
I0428 14:40:43.760885  7476 solver.cpp:397]     Test net output #1: loss = 3.04144 (* 1 = 3.04144 loss)
I0428 14:40:43.889186  7476 solver.cpp:218] Iteration 5916 (1.00438 iter/s, 11.9476s/12 iters), loss = 1.37753
I0428 14:40:43.889226  7476 solver.cpp:237]     Train net output #0: loss = 1.37753 (* 1 = 1.37753 loss)
I0428 14:40:43.889235  7476 sgd_solver.cpp:105] Iteration 5916, lr = 0.00309785
I0428 14:40:48.340359  7476 solver.cpp:218] Iteration 5928 (2.69602 iter/s, 4.451s/12 iters), loss = 1.32071
I0428 14:40:48.340556  7476 solver.cpp:237]     Train net output #0: loss = 1.32071 (* 1 = 1.32071 loss)
I0428 14:40:48.340567  7476 sgd_solver.cpp:105] Iteration 5928, lr = 0.0030905
I0428 14:40:53.680984  7476 solver.cpp:218] Iteration 5940 (2.24708 iter/s, 5.34027s/12 iters), loss = 1.19986
I0428 14:40:53.681023  7476 solver.cpp:237]     Train net output #0: loss = 1.19986 (* 1 = 1.19986 loss)
I0428 14:40:53.681033  7476 sgd_solver.cpp:105] Iteration 5940, lr = 0.00308316
I0428 14:40:59.186901  7476 solver.cpp:218] Iteration 5952 (2.17955 iter/s, 5.50572s/12 iters), loss = 1.20119
I0428 14:40:59.186946  7476 solver.cpp:237]     Train net output #0: loss = 1.20119 (* 1 = 1.20119 loss)
I0428 14:40:59.186955  7476 sgd_solver.cpp:105] Iteration 5952, lr = 0.00307584
I0428 14:41:04.685613  7476 solver.cpp:218] Iteration 5964 (2.18241 iter/s, 5.4985s/12 iters), loss = 1.06977
I0428 14:41:04.685667  7476 solver.cpp:237]     Train net output #0: loss = 1.06977 (* 1 = 1.06977 loss)
I0428 14:41:04.685678  7476 sgd_solver.cpp:105] Iteration 5964, lr = 0.00306854
I0428 14:41:06.145704  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:41:10.192590  7476 solver.cpp:218] Iteration 5976 (2.17914 iter/s, 5.50676s/12 iters), loss = 1.22839
I0428 14:41:10.192637  7476 solver.cpp:237]     Train net output #0: loss = 1.22839 (* 1 = 1.22839 loss)
I0428 14:41:10.192646  7476 sgd_solver.cpp:105] Iteration 5976, lr = 0.00306125
I0428 14:41:15.795598  7476 solver.cpp:218] Iteration 5988 (2.14179 iter/s, 5.6028s/12 iters), loss = 1.32094
I0428 14:41:15.795639  7476 solver.cpp:237]     Train net output #0: loss = 1.32094 (* 1 = 1.32094 loss)
I0428 14:41:15.795647  7476 sgd_solver.cpp:105] Iteration 5988, lr = 0.00305398
I0428 14:41:21.277714  7476 solver.cpp:218] Iteration 6000 (2.18902 iter/s, 5.48191s/12 iters), loss = 1.35898
I0428 14:41:21.277813  7476 solver.cpp:237]     Train net output #0: loss = 1.35898 (* 1 = 1.35898 loss)
I0428 14:41:21.277822  7476 sgd_solver.cpp:105] Iteration 6000, lr = 0.00304673
I0428 14:41:26.831025  7476 solver.cpp:218] Iteration 6012 (2.16098 iter/s, 5.55305s/12 iters), loss = 1.32159
I0428 14:41:26.831070  7476 solver.cpp:237]     Train net output #0: loss = 1.32159 (* 1 = 1.32159 loss)
I0428 14:41:26.831079  7476 sgd_solver.cpp:105] Iteration 6012, lr = 0.0030395
I0428 14:41:29.007393  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_6018.caffemodel
I0428 14:41:30.341873  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_6018.solverstate
I0428 14:41:31.433431  7476 solver.cpp:330] Iteration 6018, Testing net (#0)
I0428 14:41:31.433450  7476 net.cpp:676] Ignoring source layer train-data
I0428 14:41:33.443229  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:41:35.918917  7476 solver.cpp:397]     Test net output #0: accuracy = 0.337623
I0428 14:41:35.918948  7476 solver.cpp:397]     Test net output #1: loss = 3.05979 (* 1 = 3.05979 loss)
I0428 14:41:38.087126  7476 solver.cpp:218] Iteration 6024 (1.06612 iter/s, 11.2557s/12 iters), loss = 1.18559
I0428 14:41:38.087170  7476 solver.cpp:237]     Train net output #0: loss = 1.18559 (* 1 = 1.18559 loss)
I0428 14:41:38.087179  7476 sgd_solver.cpp:105] Iteration 6024, lr = 0.00303228
I0428 14:41:43.642283  7476 solver.cpp:218] Iteration 6036 (2.16023 iter/s, 5.55495s/12 iters), loss = 1.37302
I0428 14:41:43.642328  7476 solver.cpp:237]     Train net output #0: loss = 1.37302 (* 1 = 1.37302 loss)
I0428 14:41:43.642336  7476 sgd_solver.cpp:105] Iteration 6036, lr = 0.00302508
I0428 14:41:49.013950  7476 solver.cpp:218] Iteration 6048 (2.23403 iter/s, 5.37146s/12 iters), loss = 1.1033
I0428 14:41:49.013991  7476 solver.cpp:237]     Train net output #0: loss = 1.1033 (* 1 = 1.1033 loss)
I0428 14:41:49.014000  7476 sgd_solver.cpp:105] Iteration 6048, lr = 0.0030179
I0428 14:41:54.507380  7476 solver.cpp:218] Iteration 6060 (2.18451 iter/s, 5.49323s/12 iters), loss = 1.16772
I0428 14:41:54.507517  7476 solver.cpp:237]     Train net output #0: loss = 1.16772 (* 1 = 1.16772 loss)
I0428 14:41:54.507525  7476 sgd_solver.cpp:105] Iteration 6060, lr = 0.00301074
I0428 14:41:58.131028  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:41:59.812949  7476 solver.cpp:218] Iteration 6072 (2.2619 iter/s, 5.30528s/12 iters), loss = 1.0951
I0428 14:41:59.812992  7476 solver.cpp:237]     Train net output #0: loss = 1.0951 (* 1 = 1.0951 loss)
I0428 14:41:59.813000  7476 sgd_solver.cpp:105] Iteration 6072, lr = 0.00300359
I0428 14:42:05.391340  7476 solver.cpp:218] Iteration 6084 (2.15124 iter/s, 5.57818s/12 iters), loss = 1.33303
I0428 14:42:05.391389  7476 solver.cpp:237]     Train net output #0: loss = 1.33303 (* 1 = 1.33303 loss)
I0428 14:42:05.391402  7476 sgd_solver.cpp:105] Iteration 6084, lr = 0.00299646
I0428 14:42:10.830999  7476 solver.cpp:218] Iteration 6096 (2.20611 iter/s, 5.43945s/12 iters), loss = 1.4057
I0428 14:42:10.831043  7476 solver.cpp:237]     Train net output #0: loss = 1.4057 (* 1 = 1.4057 loss)
I0428 14:42:10.831050  7476 sgd_solver.cpp:105] Iteration 6096, lr = 0.00298934
I0428 14:42:16.386152  7476 solver.cpp:218] Iteration 6108 (2.16024 iter/s, 5.55494s/12 iters), loss = 1.46677
I0428 14:42:16.386194  7476 solver.cpp:237]     Train net output #0: loss = 1.46677 (* 1 = 1.46677 loss)
I0428 14:42:16.386202  7476 sgd_solver.cpp:105] Iteration 6108, lr = 0.00298225
I0428 14:42:21.246822  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_6120.caffemodel
I0428 14:42:23.166268  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_6120.solverstate
I0428 14:42:24.232785  7476 solver.cpp:330] Iteration 6120, Testing net (#0)
I0428 14:42:24.232807  7476 net.cpp:676] Ignoring source layer train-data
I0428 14:42:26.248153  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:42:28.712702  7476 solver.cpp:397]     Test net output #0: accuracy = 0.344363
I0428 14:42:28.712733  7476 solver.cpp:397]     Test net output #1: loss = 3.06128 (* 1 = 3.06128 loss)
I0428 14:42:28.841060  7476 solver.cpp:218] Iteration 6120 (0.963505 iter/s, 12.4545s/12 iters), loss = 1.22789
I0428 14:42:28.841101  7476 solver.cpp:237]     Train net output #0: loss = 1.22789 (* 1 = 1.22789 loss)
I0428 14:42:28.841110  7476 sgd_solver.cpp:105] Iteration 6120, lr = 0.00297517
I0428 14:42:33.251888  7476 solver.cpp:218] Iteration 6132 (2.72069 iter/s, 4.41065s/12 iters), loss = 1.13788
I0428 14:42:33.251924  7476 solver.cpp:237]     Train net output #0: loss = 1.13788 (* 1 = 1.13788 loss)
I0428 14:42:33.251933  7476 sgd_solver.cpp:105] Iteration 6132, lr = 0.0029681
I0428 14:42:38.775308  7476 solver.cpp:218] Iteration 6144 (2.17265 iter/s, 5.52322s/12 iters), loss = 1.14658
I0428 14:42:38.775357  7476 solver.cpp:237]     Train net output #0: loss = 1.14658 (* 1 = 1.14658 loss)
I0428 14:42:38.775365  7476 sgd_solver.cpp:105] Iteration 6144, lr = 0.00296105
I0428 14:42:43.978801  7476 solver.cpp:218] Iteration 6156 (2.30624 iter/s, 5.20328s/12 iters), loss = 1.31781
I0428 14:42:43.978854  7476 solver.cpp:237]     Train net output #0: loss = 1.31781 (* 1 = 1.31781 loss)
I0428 14:42:43.978864  7476 sgd_solver.cpp:105] Iteration 6156, lr = 0.00295402
I0428 14:42:49.494140  7476 solver.cpp:218] Iteration 6168 (2.17583 iter/s, 5.51513s/12 iters), loss = 1.06392
I0428 14:42:49.494184  7476 solver.cpp:237]     Train net output #0: loss = 1.06392 (* 1 = 1.06392 loss)
I0428 14:42:49.494191  7476 sgd_solver.cpp:105] Iteration 6168, lr = 0.00294701
I0428 14:42:50.139674  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:42:55.096307  7476 solver.cpp:218] Iteration 6180 (2.14211 iter/s, 5.60196s/12 iters), loss = 1.07347
I0428 14:42:55.096347  7476 solver.cpp:237]     Train net output #0: loss = 1.07347 (* 1 = 1.07347 loss)
I0428 14:42:55.096356  7476 sgd_solver.cpp:105] Iteration 6180, lr = 0.00294001
I0428 14:43:00.535732  7476 solver.cpp:218] Iteration 6192 (2.2062 iter/s, 5.43922s/12 iters), loss = 1.14581
I0428 14:43:00.535856  7476 solver.cpp:237]     Train net output #0: loss = 1.14581 (* 1 = 1.14581 loss)
I0428 14:43:00.535866  7476 sgd_solver.cpp:105] Iteration 6192, lr = 0.00293303
I0428 14:43:05.882596  7476 solver.cpp:218] Iteration 6204 (2.24442 iter/s, 5.34659s/12 iters), loss = 1.07369
I0428 14:43:05.882634  7476 solver.cpp:237]     Train net output #0: loss = 1.07369 (* 1 = 1.07369 loss)
I0428 14:43:05.882642  7476 sgd_solver.cpp:105] Iteration 6204, lr = 0.00292607
I0428 14:43:11.219352  7476 solver.cpp:218] Iteration 6216 (2.24864 iter/s, 5.33656s/12 iters), loss = 1.17733
I0428 14:43:11.219398  7476 solver.cpp:237]     Train net output #0: loss = 1.17733 (* 1 = 1.17733 loss)
I0428 14:43:11.219405  7476 sgd_solver.cpp:105] Iteration 6216, lr = 0.00291912
I0428 14:43:13.375241  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_6222.caffemodel
I0428 14:43:18.151319  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_6222.solverstate
I0428 14:43:20.476876  7476 solver.cpp:330] Iteration 6222, Testing net (#0)
I0428 14:43:20.476895  7476 net.cpp:676] Ignoring source layer train-data
I0428 14:43:22.412613  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:43:23.768000  7476 blocking_queue.cpp:49] Waiting for data
I0428 14:43:24.967468  7476 solver.cpp:397]     Test net output #0: accuracy = 0.376838
I0428 14:43:24.967501  7476 solver.cpp:397]     Test net output #1: loss = 3.01128 (* 1 = 3.01128 loss)
I0428 14:43:26.715308  7476 solver.cpp:218] Iteration 6228 (0.774419 iter/s, 15.4955s/12 iters), loss = 1.09986
I0428 14:43:26.715348  7476 solver.cpp:237]     Train net output #0: loss = 1.09986 (* 1 = 1.09986 loss)
I0428 14:43:26.715358  7476 sgd_solver.cpp:105] Iteration 6228, lr = 0.00291219
I0428 14:43:32.055716  7476 solver.cpp:218] Iteration 6240 (2.2471 iter/s, 5.34021s/12 iters), loss = 0.956921
I0428 14:43:32.055812  7476 solver.cpp:237]     Train net output #0: loss = 0.956921 (* 1 = 0.956921 loss)
I0428 14:43:32.055821  7476 sgd_solver.cpp:105] Iteration 6240, lr = 0.00290528
I0428 14:43:37.398020  7476 solver.cpp:218] Iteration 6252 (2.24633 iter/s, 5.34205s/12 iters), loss = 0.997301
I0428 14:43:37.398067  7476 solver.cpp:237]     Train net output #0: loss = 0.997301 (* 1 = 0.997301 loss)
I0428 14:43:37.398075  7476 sgd_solver.cpp:105] Iteration 6252, lr = 0.00289838
I0428 14:43:42.866591  7476 solver.cpp:218] Iteration 6264 (2.19444 iter/s, 5.46836s/12 iters), loss = 1.16934
I0428 14:43:42.866633  7476 solver.cpp:237]     Train net output #0: loss = 1.16934 (* 1 = 1.16934 loss)
I0428 14:43:42.866642  7476 sgd_solver.cpp:105] Iteration 6264, lr = 0.0028915
I0428 14:43:45.792994  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:43:48.298983  7476 solver.cpp:218] Iteration 6276 (2.20905 iter/s, 5.43219s/12 iters), loss = 1.0085
I0428 14:43:48.299027  7476 solver.cpp:237]     Train net output #0: loss = 1.0085 (* 1 = 1.0085 loss)
I0428 14:43:48.299036  7476 sgd_solver.cpp:105] Iteration 6276, lr = 0.00288463
I0428 14:43:53.588335  7476 solver.cpp:218] Iteration 6288 (2.2688 iter/s, 5.28915s/12 iters), loss = 0.9828
I0428 14:43:53.588388  7476 solver.cpp:237]     Train net output #0: loss = 0.9828 (* 1 = 0.9828 loss)
I0428 14:43:53.588399  7476 sgd_solver.cpp:105] Iteration 6288, lr = 0.00287779
I0428 14:43:59.025208  7476 solver.cpp:218] Iteration 6300 (2.20724 iter/s, 5.43666s/12 iters), loss = 0.863569
I0428 14:43:59.025256  7476 solver.cpp:237]     Train net output #0: loss = 0.863569 (* 1 = 0.863569 loss)
I0428 14:43:59.025265  7476 sgd_solver.cpp:105] Iteration 6300, lr = 0.00287095
I0428 14:44:04.464251  7476 solver.cpp:218] Iteration 6312 (2.20636 iter/s, 5.43883s/12 iters), loss = 0.911641
I0428 14:44:04.464352  7476 solver.cpp:237]     Train net output #0: loss = 0.911641 (* 1 = 0.911641 loss)
I0428 14:44:04.464362  7476 sgd_solver.cpp:105] Iteration 6312, lr = 0.00286414
I0428 14:44:09.419471  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_6324.caffemodel
I0428 14:44:12.949410  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_6324.solverstate
I0428 14:44:15.178361  7476 solver.cpp:330] Iteration 6324, Testing net (#0)
I0428 14:44:15.178380  7476 net.cpp:676] Ignoring source layer train-data
I0428 14:44:17.074769  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:44:19.599454  7476 solver.cpp:397]     Test net output #0: accuracy = 0.367647
I0428 14:44:19.599496  7476 solver.cpp:397]     Test net output #1: loss = 3.04337 (* 1 = 3.04337 loss)
I0428 14:44:19.727654  7476 solver.cpp:218] Iteration 6324 (0.786221 iter/s, 15.2629s/12 iters), loss = 0.860073
I0428 14:44:19.727721  7476 solver.cpp:237]     Train net output #0: loss = 0.860073 (* 1 = 0.860073 loss)
I0428 14:44:19.727730  7476 sgd_solver.cpp:105] Iteration 6324, lr = 0.00285734
I0428 14:44:24.225759  7476 solver.cpp:218] Iteration 6336 (2.66791 iter/s, 4.4979s/12 iters), loss = 1.16884
I0428 14:44:24.225798  7476 solver.cpp:237]     Train net output #0: loss = 1.16884 (* 1 = 1.16884 loss)
I0428 14:44:24.225807  7476 sgd_solver.cpp:105] Iteration 6336, lr = 0.00285055
I0428 14:44:29.869555  7476 solver.cpp:218] Iteration 6348 (2.12631 iter/s, 5.64359s/12 iters), loss = 0.965387
I0428 14:44:29.869596  7476 solver.cpp:237]     Train net output #0: loss = 0.965387 (* 1 = 0.965387 loss)
I0428 14:44:29.869604  7476 sgd_solver.cpp:105] Iteration 6348, lr = 0.00284379
I0428 14:44:35.380543  7476 solver.cpp:218] Iteration 6360 (2.17755 iter/s, 5.51079s/12 iters), loss = 1.13493
I0428 14:44:35.380666  7476 solver.cpp:237]     Train net output #0: loss = 1.13493 (* 1 = 1.13493 loss)
I0428 14:44:35.380676  7476 sgd_solver.cpp:105] Iteration 6360, lr = 0.00283703
I0428 14:44:40.649716  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:44:40.828083  7476 solver.cpp:218] Iteration 6372 (2.20294 iter/s, 5.44726s/12 iters), loss = 0.97076
I0428 14:44:40.828127  7476 solver.cpp:237]     Train net output #0: loss = 0.97076 (* 1 = 0.97076 loss)
I0428 14:44:40.828136  7476 sgd_solver.cpp:105] Iteration 6372, lr = 0.0028303
I0428 14:44:46.222127  7476 solver.cpp:218] Iteration 6384 (2.22476 iter/s, 5.39384s/12 iters), loss = 1.19474
I0428 14:44:46.222165  7476 solver.cpp:237]     Train net output #0: loss = 1.19474 (* 1 = 1.19474 loss)
I0428 14:44:46.222175  7476 sgd_solver.cpp:105] Iteration 6384, lr = 0.00282358
I0428 14:44:51.680521  7476 solver.cpp:218] Iteration 6396 (2.19854 iter/s, 5.45818s/12 iters), loss = 1.10571
I0428 14:44:51.680562  7476 solver.cpp:237]     Train net output #0: loss = 1.10571 (* 1 = 1.10571 loss)
I0428 14:44:51.680572  7476 sgd_solver.cpp:105] Iteration 6396, lr = 0.00281687
I0428 14:44:57.227394  7476 solver.cpp:218] Iteration 6408 (2.16346 iter/s, 5.54667s/12 iters), loss = 0.960422
I0428 14:44:57.227437  7476 solver.cpp:237]     Train net output #0: loss = 0.960422 (* 1 = 0.960422 loss)
I0428 14:44:57.227447  7476 sgd_solver.cpp:105] Iteration 6408, lr = 0.00281019
I0428 14:45:02.862912  7476 solver.cpp:218] Iteration 6420 (2.12943 iter/s, 5.63531s/12 iters), loss = 0.990795
I0428 14:45:02.862952  7476 solver.cpp:237]     Train net output #0: loss = 0.990795 (* 1 = 0.990795 loss)
I0428 14:45:02.862962  7476 sgd_solver.cpp:105] Iteration 6420, lr = 0.00280351
I0428 14:45:05.059639  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_6426.caffemodel
I0428 14:45:09.546594  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_6426.solverstate
I0428 14:45:15.174930  7476 solver.cpp:330] Iteration 6426, Testing net (#0)
I0428 14:45:15.174950  7476 net.cpp:676] Ignoring source layer train-data
I0428 14:45:17.063298  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:45:19.642442  7476 solver.cpp:397]     Test net output #0: accuracy = 0.370711
I0428 14:45:19.642477  7476 solver.cpp:397]     Test net output #1: loss = 3.08435 (* 1 = 3.08435 loss)
I0428 14:45:21.760234  7476 solver.cpp:218] Iteration 6432 (0.63503 iter/s, 18.8968s/12 iters), loss = 0.923812
I0428 14:45:21.760274  7476 solver.cpp:237]     Train net output #0: loss = 0.923812 (* 1 = 0.923812 loss)
I0428 14:45:21.760282  7476 sgd_solver.cpp:105] Iteration 6432, lr = 0.00279686
I0428 14:45:27.290257  7476 solver.cpp:218] Iteration 6444 (2.17005 iter/s, 5.52982s/12 iters), loss = 1.00127
I0428 14:45:27.290297  7476 solver.cpp:237]     Train net output #0: loss = 1.00127 (* 1 = 1.00127 loss)
I0428 14:45:27.290305  7476 sgd_solver.cpp:105] Iteration 6444, lr = 0.00279022
I0428 14:45:32.811848  7476 solver.cpp:218] Iteration 6456 (2.17337 iter/s, 5.52138s/12 iters), loss = 0.880413
I0428 14:45:32.811888  7476 solver.cpp:237]     Train net output #0: loss = 0.880413 (* 1 = 0.880413 loss)
I0428 14:45:32.811897  7476 sgd_solver.cpp:105] Iteration 6456, lr = 0.00278359
I0428 14:45:38.160162  7476 solver.cpp:218] Iteration 6468 (2.24378 iter/s, 5.34811s/12 iters), loss = 1.04465
I0428 14:45:38.160202  7476 solver.cpp:237]     Train net output #0: loss = 1.04465 (* 1 = 1.04465 loss)
I0428 14:45:38.160212  7476 sgd_solver.cpp:105] Iteration 6468, lr = 0.00277698
I0428 14:45:40.315284  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:45:43.669108  7476 solver.cpp:218] Iteration 6480 (2.17836 iter/s, 5.50874s/12 iters), loss = 1.09099
I0428 14:45:43.669154  7476 solver.cpp:237]     Train net output #0: loss = 1.09099 (* 1 = 1.09099 loss)
I0428 14:45:43.669163  7476 sgd_solver.cpp:105] Iteration 6480, lr = 0.00277039
I0428 14:45:49.127205  7476 solver.cpp:218] Iteration 6492 (2.19865 iter/s, 5.45789s/12 iters), loss = 1.10187
I0428 14:45:49.127249  7476 solver.cpp:237]     Train net output #0: loss = 1.10187 (* 1 = 1.10187 loss)
I0428 14:45:49.127256  7476 sgd_solver.cpp:105] Iteration 6492, lr = 0.00276381
I0428 14:45:54.460868  7476 solver.cpp:218] Iteration 6504 (2.24995 iter/s, 5.33346s/12 iters), loss = 1.05585
I0428 14:45:54.460911  7476 solver.cpp:237]     Train net output #0: loss = 1.05585 (* 1 = 1.05585 loss)
I0428 14:45:54.460919  7476 sgd_solver.cpp:105] Iteration 6504, lr = 0.00275725
I0428 14:45:59.805693  7476 solver.cpp:218] Iteration 6516 (2.24525 iter/s, 5.34462s/12 iters), loss = 1.14896
I0428 14:45:59.805732  7476 solver.cpp:237]     Train net output #0: loss = 1.14896 (* 1 = 1.14896 loss)
I0428 14:45:59.805740  7476 sgd_solver.cpp:105] Iteration 6516, lr = 0.00275071
I0428 14:46:04.761863  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_6528.caffemodel
I0428 14:46:09.401145  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_6528.solverstate
I0428 14:46:12.902424  7476 solver.cpp:330] Iteration 6528, Testing net (#0)
I0428 14:46:12.902506  7476 net.cpp:676] Ignoring source layer train-data
I0428 14:46:14.692402  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:46:17.425211  7476 solver.cpp:397]     Test net output #0: accuracy = 0.372549
I0428 14:46:17.425252  7476 solver.cpp:397]     Test net output #1: loss = 3.00383 (* 1 = 3.00383 loss)
I0428 14:46:17.553568  7476 solver.cpp:218] Iteration 6528 (0.676157 iter/s, 17.7473s/12 iters), loss = 1.12794
I0428 14:46:17.553611  7476 solver.cpp:237]     Train net output #0: loss = 1.12794 (* 1 = 1.12794 loss)
I0428 14:46:17.553619  7476 sgd_solver.cpp:105] Iteration 6528, lr = 0.00274418
I0428 14:46:22.143465  7476 solver.cpp:218] Iteration 6540 (2.61454 iter/s, 4.58971s/12 iters), loss = 0.940773
I0428 14:46:22.143503  7476 solver.cpp:237]     Train net output #0: loss = 0.940773 (* 1 = 0.940773 loss)
I0428 14:46:22.143512  7476 sgd_solver.cpp:105] Iteration 6540, lr = 0.00273766
I0428 14:46:27.583403  7476 solver.cpp:218] Iteration 6552 (2.20599 iter/s, 5.43973s/12 iters), loss = 0.980502
I0428 14:46:27.583453  7476 solver.cpp:237]     Train net output #0: loss = 0.980502 (* 1 = 0.980502 loss)
I0428 14:46:27.583462  7476 sgd_solver.cpp:105] Iteration 6552, lr = 0.00273116
I0428 14:46:33.063998  7476 solver.cpp:218] Iteration 6564 (2.18963 iter/s, 5.48038s/12 iters), loss = 0.901421
I0428 14:46:33.064036  7476 solver.cpp:237]     Train net output #0: loss = 0.901421 (* 1 = 0.901421 loss)
I0428 14:46:33.064044  7476 sgd_solver.cpp:105] Iteration 6564, lr = 0.00272468
I0428 14:46:37.620790  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:46:38.481086  7476 solver.cpp:218] Iteration 6576 (2.21529 iter/s, 5.41689s/12 iters), loss = 1.08484
I0428 14:46:38.481125  7476 solver.cpp:237]     Train net output #0: loss = 1.08484 (* 1 = 1.08484 loss)
I0428 14:46:38.481133  7476 sgd_solver.cpp:105] Iteration 6576, lr = 0.00271821
I0428 14:46:44.045140  7476 solver.cpp:218] Iteration 6588 (2.15678 iter/s, 5.56384s/12 iters), loss = 1.08372
I0428 14:46:44.045341  7476 solver.cpp:237]     Train net output #0: loss = 1.08372 (* 1 = 1.08372 loss)
I0428 14:46:44.045351  7476 sgd_solver.cpp:105] Iteration 6588, lr = 0.00271175
I0428 14:46:49.473950  7476 solver.cpp:218] Iteration 6600 (2.21058 iter/s, 5.42845s/12 iters), loss = 0.819176
I0428 14:46:49.473990  7476 solver.cpp:237]     Train net output #0: loss = 0.819176 (* 1 = 0.819176 loss)
I0428 14:46:49.473999  7476 sgd_solver.cpp:105] Iteration 6600, lr = 0.00270532
I0428 14:46:54.880127  7476 solver.cpp:218] Iteration 6612 (2.21976 iter/s, 5.40598s/12 iters), loss = 0.921797
I0428 14:46:54.880167  7476 solver.cpp:237]     Train net output #0: loss = 0.921797 (* 1 = 0.921797 loss)
I0428 14:46:54.880175  7476 sgd_solver.cpp:105] Iteration 6612, lr = 0.00269889
I0428 14:47:00.295578  7476 solver.cpp:218] Iteration 6624 (2.21597 iter/s, 5.41525s/12 iters), loss = 0.885576
I0428 14:47:00.295625  7476 solver.cpp:237]     Train net output #0: loss = 0.885576 (* 1 = 0.885576 loss)
I0428 14:47:00.295634  7476 sgd_solver.cpp:105] Iteration 6624, lr = 0.00269248
I0428 14:47:02.348341  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_6630.caffemodel
I0428 14:47:06.628888  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_6630.solverstate
I0428 14:47:08.189858  7476 solver.cpp:330] Iteration 6630, Testing net (#0)
I0428 14:47:08.189883  7476 net.cpp:676] Ignoring source layer train-data
I0428 14:47:09.951843  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:47:12.601202  7476 solver.cpp:397]     Test net output #0: accuracy = 0.365809
I0428 14:47:12.601239  7476 solver.cpp:397]     Test net output #1: loss = 3.18556 (* 1 = 3.18556 loss)
I0428 14:47:14.578680  7476 solver.cpp:218] Iteration 6636 (0.84018 iter/s, 14.2827s/12 iters), loss = 1.05834
I0428 14:47:14.578774  7476 solver.cpp:237]     Train net output #0: loss = 1.05834 (* 1 = 1.05834 loss)
I0428 14:47:14.578784  7476 sgd_solver.cpp:105] Iteration 6636, lr = 0.00268609
I0428 14:47:19.995852  7476 solver.cpp:218] Iteration 6648 (2.21528 iter/s, 5.41692s/12 iters), loss = 0.751942
I0428 14:47:19.995909  7476 solver.cpp:237]     Train net output #0: loss = 0.751942 (* 1 = 0.751942 loss)
I0428 14:47:19.995920  7476 sgd_solver.cpp:105] Iteration 6648, lr = 0.00267971
I0428 14:47:25.392853  7476 solver.cpp:218] Iteration 6660 (2.22355 iter/s, 5.39679s/12 iters), loss = 0.681272
I0428 14:47:25.392899  7476 solver.cpp:237]     Train net output #0: loss = 0.681272 (* 1 = 0.681272 loss)
I0428 14:47:25.392908  7476 sgd_solver.cpp:105] Iteration 6660, lr = 0.00267335
I0428 14:47:30.923283  7476 solver.cpp:218] Iteration 6672 (2.16989 iter/s, 5.53023s/12 iters), loss = 0.872733
I0428 14:47:30.923327  7476 solver.cpp:237]     Train net output #0: loss = 0.872733 (* 1 = 0.872733 loss)
I0428 14:47:30.923337  7476 sgd_solver.cpp:105] Iteration 6672, lr = 0.00266701
I0428 14:47:32.383258  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:47:36.325629  7476 solver.cpp:218] Iteration 6684 (2.22134 iter/s, 5.40214s/12 iters), loss = 1.00447
I0428 14:47:36.325672  7476 solver.cpp:237]     Train net output #0: loss = 1.00447 (* 1 = 1.00447 loss)
I0428 14:47:36.325680  7476 sgd_solver.cpp:105] Iteration 6684, lr = 0.00266067
I0428 14:47:41.912160  7476 solver.cpp:218] Iteration 6696 (2.14811 iter/s, 5.58632s/12 iters), loss = 0.900104
I0428 14:47:41.912207  7476 solver.cpp:237]     Train net output #0: loss = 0.900104 (* 1 = 0.900104 loss)
I0428 14:47:41.912217  7476 sgd_solver.cpp:105] Iteration 6696, lr = 0.00265436
I0428 14:47:47.294111  7476 solver.cpp:218] Iteration 6708 (2.22976 iter/s, 5.38174s/12 iters), loss = 0.931748
I0428 14:47:47.294226  7476 solver.cpp:237]     Train net output #0: loss = 0.931748 (* 1 = 0.931748 loss)
I0428 14:47:47.294236  7476 sgd_solver.cpp:105] Iteration 6708, lr = 0.00264805
I0428 14:47:52.800019  7476 solver.cpp:218] Iteration 6720 (2.17959 iter/s, 5.50563s/12 iters), loss = 0.808488
I0428 14:47:52.800062  7476 solver.cpp:237]     Train net output #0: loss = 0.808488 (* 1 = 0.808488 loss)
I0428 14:47:52.800071  7476 sgd_solver.cpp:105] Iteration 6720, lr = 0.00264177
I0428 14:47:57.636977  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_6732.caffemodel
I0428 14:48:00.084293  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_6732.solverstate
I0428 14:48:01.710369  7476 solver.cpp:330] Iteration 6732, Testing net (#0)
I0428 14:48:01.710399  7476 net.cpp:676] Ignoring source layer train-data
I0428 14:48:03.529451  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:48:06.201437  7476 solver.cpp:397]     Test net output #0: accuracy = 0.366422
I0428 14:48:06.201470  7476 solver.cpp:397]     Test net output #1: loss = 3.27819 (* 1 = 3.27819 loss)
I0428 14:48:06.329564  7476 solver.cpp:218] Iteration 6732 (0.886976 iter/s, 13.5291s/12 iters), loss = 1.09321
I0428 14:48:06.329619  7476 solver.cpp:237]     Train net output #0: loss = 1.09321 (* 1 = 1.09321 loss)
I0428 14:48:06.329629  7476 sgd_solver.cpp:105] Iteration 6732, lr = 0.0026355
I0428 14:48:10.792467  7476 solver.cpp:218] Iteration 6744 (2.68895 iter/s, 4.4627s/12 iters), loss = 1.05581
I0428 14:48:10.792549  7476 solver.cpp:237]     Train net output #0: loss = 1.05581 (* 1 = 1.05581 loss)
I0428 14:48:10.792562  7476 sgd_solver.cpp:105] Iteration 6744, lr = 0.00262924
I0428 14:48:16.276394  7476 solver.cpp:218] Iteration 6756 (2.18831 iter/s, 5.48369s/12 iters), loss = 0.943056
I0428 14:48:16.276437  7476 solver.cpp:237]     Train net output #0: loss = 0.943056 (* 1 = 0.943056 loss)
I0428 14:48:16.276445  7476 sgd_solver.cpp:105] Iteration 6756, lr = 0.002623
I0428 14:48:21.751451  7476 solver.cpp:218] Iteration 6768 (2.19184 iter/s, 5.47485s/12 iters), loss = 0.991989
I0428 14:48:21.751590  7476 solver.cpp:237]     Train net output #0: loss = 0.991989 (* 1 = 0.991989 loss)
I0428 14:48:21.751601  7476 sgd_solver.cpp:105] Iteration 6768, lr = 0.00261677
I0428 14:48:25.510746  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:48:27.160724  7476 solver.cpp:218] Iteration 6780 (2.21853 iter/s, 5.40898s/12 iters), loss = 0.953592
I0428 14:48:27.160770  7476 solver.cpp:237]     Train net output #0: loss = 0.953592 (* 1 = 0.953592 loss)
I0428 14:48:27.160779  7476 sgd_solver.cpp:105] Iteration 6780, lr = 0.00261056
I0428 14:48:32.536576  7476 solver.cpp:218] Iteration 6792 (2.23229 iter/s, 5.37565s/12 iters), loss = 1.12114
I0428 14:48:32.536617  7476 solver.cpp:237]     Train net output #0: loss = 1.12114 (* 1 = 1.12114 loss)
I0428 14:48:32.536626  7476 sgd_solver.cpp:105] Iteration 6792, lr = 0.00260436
I0428 14:48:37.989827  7476 solver.cpp:218] Iteration 6804 (2.20061 iter/s, 5.45304s/12 iters), loss = 1.01492
I0428 14:48:37.989867  7476 solver.cpp:237]     Train net output #0: loss = 1.01492 (* 1 = 1.01492 loss)
I0428 14:48:37.989876  7476 sgd_solver.cpp:105] Iteration 6804, lr = 0.00259817
I0428 14:48:43.356334  7476 solver.cpp:218] Iteration 6816 (2.23618 iter/s, 5.3663s/12 iters), loss = 0.756982
I0428 14:48:43.356382  7476 solver.cpp:237]     Train net output #0: loss = 0.756982 (* 1 = 0.756982 loss)
I0428 14:48:43.356392  7476 sgd_solver.cpp:105] Iteration 6816, lr = 0.00259201
I0428 14:48:48.842079  7476 solver.cpp:218] Iteration 6828 (2.18757 iter/s, 5.48553s/12 iters), loss = 0.823173
I0428 14:48:48.842123  7476 solver.cpp:237]     Train net output #0: loss = 0.823173 (* 1 = 0.823173 loss)
I0428 14:48:48.842131  7476 sgd_solver.cpp:105] Iteration 6828, lr = 0.00258585
I0428 14:48:51.032845  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_6834.caffemodel
I0428 14:48:56.225625  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_6834.solverstate
I0428 14:49:00.566359  7476 solver.cpp:330] Iteration 6834, Testing net (#0)
I0428 14:49:00.566383  7476 net.cpp:676] Ignoring source layer train-data
I0428 14:49:02.269642  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:49:05.035089  7476 solver.cpp:397]     Test net output #0: accuracy = 0.372549
I0428 14:49:05.035116  7476 solver.cpp:397]     Test net output #1: loss = 3.20027 (* 1 = 3.20027 loss)
I0428 14:49:07.089196  7476 solver.cpp:218] Iteration 6840 (0.657658 iter/s, 18.2466s/12 iters), loss = 0.79766
I0428 14:49:07.089243  7476 solver.cpp:237]     Train net output #0: loss = 0.79766 (* 1 = 0.79766 loss)
I0428 14:49:07.089251  7476 sgd_solver.cpp:105] Iteration 6840, lr = 0.00257971
I0428 14:49:12.470527  7476 solver.cpp:218] Iteration 6852 (2.23002 iter/s, 5.38112s/12 iters), loss = 0.917336
I0428 14:49:12.470571  7476 solver.cpp:237]     Train net output #0: loss = 0.917336 (* 1 = 0.917336 loss)
I0428 14:49:12.470580  7476 sgd_solver.cpp:105] Iteration 6852, lr = 0.00257359
I0428 14:49:17.939429  7476 solver.cpp:218] Iteration 6864 (2.19431 iter/s, 5.46869s/12 iters), loss = 0.935814
I0428 14:49:17.939476  7476 solver.cpp:237]     Train net output #0: loss = 0.935814 (* 1 = 0.935814 loss)
I0428 14:49:17.939484  7476 sgd_solver.cpp:105] Iteration 6864, lr = 0.00256748
I0428 14:49:23.518409  7476 solver.cpp:218] Iteration 6876 (2.15101 iter/s, 5.57877s/12 iters), loss = 1.03972
I0428 14:49:23.518448  7476 solver.cpp:237]     Train net output #0: loss = 1.03972 (* 1 = 1.03972 loss)
I0428 14:49:23.518457  7476 sgd_solver.cpp:105] Iteration 6876, lr = 0.00256138
I0428 14:49:24.184365  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:49:29.017820  7476 solver.cpp:218] Iteration 6888 (2.18214 iter/s, 5.4992s/12 iters), loss = 0.959344
I0428 14:49:29.017953  7476 solver.cpp:237]     Train net output #0: loss = 0.959344 (* 1 = 0.959344 loss)
I0428 14:49:29.017963  7476 sgd_solver.cpp:105] Iteration 6888, lr = 0.0025553
I0428 14:49:34.586076  7476 solver.cpp:218] Iteration 6900 (2.15519 iter/s, 5.56796s/12 iters), loss = 0.913277
I0428 14:49:34.586119  7476 solver.cpp:237]     Train net output #0: loss = 0.913277 (* 1 = 0.913277 loss)
I0428 14:49:34.586127  7476 sgd_solver.cpp:105] Iteration 6900, lr = 0.00254923
I0428 14:49:40.015426  7476 solver.cpp:218] Iteration 6912 (2.21029 iter/s, 5.42914s/12 iters), loss = 0.645872
I0428 14:49:40.015473  7476 solver.cpp:237]     Train net output #0: loss = 0.645872 (* 1 = 0.645872 loss)
I0428 14:49:40.015481  7476 sgd_solver.cpp:105] Iteration 6912, lr = 0.00254318
I0428 14:49:45.650782  7476 solver.cpp:218] Iteration 6924 (2.12949 iter/s, 5.63514s/12 iters), loss = 0.947795
I0428 14:49:45.650828  7476 solver.cpp:237]     Train net output #0: loss = 0.947795 (* 1 = 0.947795 loss)
I0428 14:49:45.650838  7476 sgd_solver.cpp:105] Iteration 6924, lr = 0.00253714
I0428 14:49:50.502434  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_6936.caffemodel
I0428 14:49:53.428517  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_6936.solverstate
I0428 14:49:55.252173  7476 solver.cpp:330] Iteration 6936, Testing net (#0)
I0428 14:49:55.252202  7476 net.cpp:676] Ignoring source layer train-data
I0428 14:49:55.866423  7476 blocking_queue.cpp:49] Waiting for data
I0428 14:49:56.942462  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:49:59.825219  7476 solver.cpp:397]     Test net output #0: accuracy = 0.378064
I0428 14:49:59.825356  7476 solver.cpp:397]     Test net output #1: loss = 3.21576 (* 1 = 3.21576 loss)
I0428 14:49:59.951175  7476 solver.cpp:218] Iteration 6936 (0.839164 iter/s, 14.2999s/12 iters), loss = 0.722846
I0428 14:49:59.952699  7476 solver.cpp:237]     Train net output #0: loss = 0.722846 (* 1 = 0.722846 loss)
I0428 14:49:59.952710  7476 sgd_solver.cpp:105] Iteration 6936, lr = 0.00253112
I0428 14:50:04.612190  7476 solver.cpp:218] Iteration 6948 (2.57547 iter/s, 4.65935s/12 iters), loss = 0.765993
I0428 14:50:04.612232  7476 solver.cpp:237]     Train net output #0: loss = 0.765993 (* 1 = 0.765993 loss)
I0428 14:50:04.612241  7476 sgd_solver.cpp:105] Iteration 6948, lr = 0.00252511
I0428 14:50:10.505110  7476 solver.cpp:218] Iteration 6960 (2.03642 iter/s, 5.8927s/12 iters), loss = 0.815206
I0428 14:50:10.505156  7476 solver.cpp:237]     Train net output #0: loss = 0.815206 (* 1 = 0.815206 loss)
I0428 14:50:10.505164  7476 sgd_solver.cpp:105] Iteration 6960, lr = 0.00251911
I0428 14:50:15.841001  7476 solver.cpp:218] Iteration 6972 (2.24901 iter/s, 5.33568s/12 iters), loss = 0.826715
I0428 14:50:15.841048  7476 solver.cpp:237]     Train net output #0: loss = 0.826715 (* 1 = 0.826715 loss)
I0428 14:50:15.841058  7476 sgd_solver.cpp:105] Iteration 6972, lr = 0.00251313
I0428 14:50:18.800179  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:50:21.282984  7476 solver.cpp:218] Iteration 6984 (2.20516 iter/s, 5.44177s/12 iters), loss = 0.844983
I0428 14:50:21.283026  7476 solver.cpp:237]     Train net output #0: loss = 0.844983 (* 1 = 0.844983 loss)
I0428 14:50:21.283035  7476 sgd_solver.cpp:105] Iteration 6984, lr = 0.00250717
I0428 14:50:26.715140  7476 solver.cpp:218] Iteration 6996 (2.20915 iter/s, 5.43194s/12 iters), loss = 0.753531
I0428 14:50:26.715190  7476 solver.cpp:237]     Train net output #0: loss = 0.753531 (* 1 = 0.753531 loss)
I0428 14:50:26.715199  7476 sgd_solver.cpp:105] Iteration 6996, lr = 0.00250121
I0428 14:50:32.173727  7476 solver.cpp:218] Iteration 7008 (2.19846 iter/s, 5.45837s/12 iters), loss = 0.868858
I0428 14:50:32.174772  7476 solver.cpp:237]     Train net output #0: loss = 0.868858 (* 1 = 0.868858 loss)
I0428 14:50:32.174782  7476 sgd_solver.cpp:105] Iteration 7008, lr = 0.00249528
I0428 14:50:37.574739  7476 solver.cpp:218] Iteration 7020 (2.2223 iter/s, 5.3998s/12 iters), loss = 0.717601
I0428 14:50:37.574785  7476 solver.cpp:237]     Train net output #0: loss = 0.717601 (* 1 = 0.717601 loss)
I0428 14:50:37.574796  7476 sgd_solver.cpp:105] Iteration 7020, lr = 0.00248935
I0428 14:50:42.968477  7476 solver.cpp:218] Iteration 7032 (2.22489 iter/s, 5.39352s/12 iters), loss = 0.954306
I0428 14:50:42.968566  7476 solver.cpp:237]     Train net output #0: loss = 0.954306 (* 1 = 0.954306 loss)
I0428 14:50:42.968581  7476 sgd_solver.cpp:105] Iteration 7032, lr = 0.00248344
I0428 14:50:45.057018  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_7038.caffemodel
I0428 14:50:48.397187  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_7038.solverstate
I0428 14:50:53.013696  7476 solver.cpp:330] Iteration 7038, Testing net (#0)
I0428 14:50:53.013720  7476 net.cpp:676] Ignoring source layer train-data
I0428 14:50:54.694490  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:50:57.501272  7476 solver.cpp:397]     Test net output #0: accuracy = 0.354779
I0428 14:50:57.501312  7476 solver.cpp:397]     Test net output #1: loss = 3.2061 (* 1 = 3.2061 loss)
I0428 14:50:59.327404  7476 solver.cpp:218] Iteration 7044 (0.733569 iter/s, 16.3584s/12 iters), loss = 0.745965
I0428 14:50:59.327450  7476 solver.cpp:237]     Train net output #0: loss = 0.745965 (* 1 = 0.745965 loss)
I0428 14:50:59.327461  7476 sgd_solver.cpp:105] Iteration 7044, lr = 0.00247755
I0428 14:51:04.617558  7476 solver.cpp:218] Iteration 7056 (2.26845 iter/s, 5.28995s/12 iters), loss = 0.942438
I0428 14:51:04.617655  7476 solver.cpp:237]     Train net output #0: loss = 0.942438 (* 1 = 0.942438 loss)
I0428 14:51:04.617664  7476 sgd_solver.cpp:105] Iteration 7056, lr = 0.00247166
I0428 14:51:10.152384  7476 solver.cpp:218] Iteration 7068 (2.16819 iter/s, 5.53456s/12 iters), loss = 0.767782
I0428 14:51:10.152426  7476 solver.cpp:237]     Train net output #0: loss = 0.767782 (* 1 = 0.767782 loss)
I0428 14:51:10.152434  7476 sgd_solver.cpp:105] Iteration 7068, lr = 0.0024658
I0428 14:51:15.455479  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:51:15.599885  7476 solver.cpp:218] Iteration 7080 (2.20293 iter/s, 5.44729s/12 iters), loss = 0.873208
I0428 14:51:15.599929  7476 solver.cpp:237]     Train net output #0: loss = 0.873208 (* 1 = 0.873208 loss)
I0428 14:51:15.599938  7476 sgd_solver.cpp:105] Iteration 7080, lr = 0.00245994
I0428 14:51:21.099129  7476 solver.cpp:218] Iteration 7092 (2.1822 iter/s, 5.49904s/12 iters), loss = 0.816167
I0428 14:51:21.099176  7476 solver.cpp:237]     Train net output #0: loss = 0.816167 (* 1 = 0.816167 loss)
I0428 14:51:21.099187  7476 sgd_solver.cpp:105] Iteration 7092, lr = 0.0024541
I0428 14:51:26.521767  7476 solver.cpp:218] Iteration 7104 (2.21303 iter/s, 5.42243s/12 iters), loss = 0.912042
I0428 14:51:26.521827  7476 solver.cpp:237]     Train net output #0: loss = 0.912042 (* 1 = 0.912042 loss)
I0428 14:51:26.521839  7476 sgd_solver.cpp:105] Iteration 7104, lr = 0.00244827
I0428 14:51:31.830947  7476 solver.cpp:218] Iteration 7116 (2.26033 iter/s, 5.30896s/12 iters), loss = 0.790037
I0428 14:51:31.831001  7476 solver.cpp:237]     Train net output #0: loss = 0.790037 (* 1 = 0.790037 loss)
I0428 14:51:31.831009  7476 sgd_solver.cpp:105] Iteration 7116, lr = 0.00244246
I0428 14:51:37.569394  7476 solver.cpp:218] Iteration 7128 (2.09124 iter/s, 5.73822s/12 iters), loss = 0.921498
I0428 14:51:37.569532  7476 solver.cpp:237]     Train net output #0: loss = 0.921498 (* 1 = 0.921498 loss)
I0428 14:51:37.569543  7476 sgd_solver.cpp:105] Iteration 7128, lr = 0.00243666
I0428 14:51:42.590931  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_7140.caffemodel
I0428 14:51:44.343457  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_7140.solverstate
I0428 14:51:46.907393  7476 solver.cpp:330] Iteration 7140, Testing net (#0)
I0428 14:51:46.907411  7476 net.cpp:676] Ignoring source layer train-data
I0428 14:51:48.492998  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:51:51.494660  7476 solver.cpp:397]     Test net output #0: accuracy = 0.387868
I0428 14:51:51.494704  7476 solver.cpp:397]     Test net output #1: loss = 3.23303 (* 1 = 3.23303 loss)
I0428 14:51:51.623220  7476 solver.cpp:218] Iteration 7140 (0.853892 iter/s, 14.0533s/12 iters), loss = 0.974311
I0428 14:51:51.624750  7476 solver.cpp:237]     Train net output #0: loss = 0.974311 (* 1 = 0.974311 loss)
I0428 14:51:51.624765  7476 sgd_solver.cpp:105] Iteration 7140, lr = 0.00243088
I0428 14:51:56.104423  7476 solver.cpp:218] Iteration 7152 (2.67885 iter/s, 4.47954s/12 iters), loss = 0.76788
I0428 14:51:56.104470  7476 solver.cpp:237]     Train net output #0: loss = 0.76788 (* 1 = 0.76788 loss)
I0428 14:51:56.104478  7476 sgd_solver.cpp:105] Iteration 7152, lr = 0.00242511
I0428 14:52:01.545855  7476 solver.cpp:218] Iteration 7164 (2.20539 iter/s, 5.44121s/12 iters), loss = 0.790058
I0428 14:52:01.545917  7476 solver.cpp:237]     Train net output #0: loss = 0.790058 (* 1 = 0.790058 loss)
I0428 14:52:01.545933  7476 sgd_solver.cpp:105] Iteration 7164, lr = 0.00241935
I0428 14:52:07.201087  7476 solver.cpp:218] Iteration 7176 (2.12202 iter/s, 5.655s/12 iters), loss = 0.737753
I0428 14:52:07.201135  7476 solver.cpp:237]     Train net output #0: loss = 0.737753 (* 1 = 0.737753 loss)
I0428 14:52:07.201145  7476 sgd_solver.cpp:105] Iteration 7176, lr = 0.0024136
I0428 14:52:09.530112  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:52:12.684389  7476 solver.cpp:218] Iteration 7188 (2.18855 iter/s, 5.48309s/12 iters), loss = 0.939125
I0428 14:52:12.684428  7476 solver.cpp:237]     Train net output #0: loss = 0.939125 (* 1 = 0.939125 loss)
I0428 14:52:12.684439  7476 sgd_solver.cpp:105] Iteration 7188, lr = 0.00240787
I0428 14:52:18.178120  7476 solver.cpp:218] Iteration 7200 (2.18439 iter/s, 5.49353s/12 iters), loss = 0.80173
I0428 14:52:18.178166  7476 solver.cpp:237]     Train net output #0: loss = 0.80173 (* 1 = 0.80173 loss)
I0428 14:52:18.178175  7476 sgd_solver.cpp:105] Iteration 7200, lr = 0.00240216
I0428 14:52:23.697979  7476 solver.cpp:218] Iteration 7212 (2.17405 iter/s, 5.51964s/12 iters), loss = 0.787963
I0428 14:52:23.698037  7476 solver.cpp:237]     Train net output #0: loss = 0.787963 (* 1 = 0.787963 loss)
I0428 14:52:23.698047  7476 sgd_solver.cpp:105] Iteration 7212, lr = 0.00239645
I0428 14:52:29.241511  7476 solver.cpp:218] Iteration 7224 (2.16477 iter/s, 5.54331s/12 iters), loss = 0.998741
I0428 14:52:29.241556  7476 solver.cpp:237]     Train net output #0: loss = 0.998741 (* 1 = 0.998741 loss)
I0428 14:52:29.241566  7476 sgd_solver.cpp:105] Iteration 7224, lr = 0.00239076
I0428 14:52:34.749845  7476 solver.cpp:218] Iteration 7236 (2.1786 iter/s, 5.50812s/12 iters), loss = 1.05496
I0428 14:52:34.749894  7476 solver.cpp:237]     Train net output #0: loss = 1.05496 (* 1 = 1.05496 loss)
I0428 14:52:34.749903  7476 sgd_solver.cpp:105] Iteration 7236, lr = 0.00238509
I0428 14:52:36.837528  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_7242.caffemodel
I0428 14:52:40.392256  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_7242.solverstate
I0428 14:52:43.522565  7476 solver.cpp:330] Iteration 7242, Testing net (#0)
I0428 14:52:43.522586  7476 net.cpp:676] Ignoring source layer train-data
I0428 14:52:45.085058  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:52:48.112890  7476 solver.cpp:397]     Test net output #0: accuracy = 0.39277
I0428 14:52:48.112929  7476 solver.cpp:397]     Test net output #1: loss = 3.08994 (* 1 = 3.08994 loss)
I0428 14:52:50.061578  7476 solver.cpp:218] Iteration 7248 (0.783737 iter/s, 15.3112s/12 iters), loss = 0.825384
I0428 14:52:50.061626  7476 solver.cpp:237]     Train net output #0: loss = 0.825384 (* 1 = 0.825384 loss)
I0428 14:52:50.061635  7476 sgd_solver.cpp:105] Iteration 7248, lr = 0.00237942
I0428 14:52:55.522035  7476 solver.cpp:218] Iteration 7260 (2.19771 iter/s, 5.46024s/12 iters), loss = 0.817576
I0428 14:52:55.522083  7476 solver.cpp:237]     Train net output #0: loss = 0.817576 (* 1 = 0.817576 loss)
I0428 14:52:55.522092  7476 sgd_solver.cpp:105] Iteration 7260, lr = 0.00237378
I0428 14:53:00.953600  7476 solver.cpp:218] Iteration 7272 (2.2094 iter/s, 5.43135s/12 iters), loss = 0.717549
I0428 14:53:00.953661  7476 solver.cpp:237]     Train net output #0: loss = 0.717549 (* 1 = 0.717549 loss)
I0428 14:53:00.953675  7476 sgd_solver.cpp:105] Iteration 7272, lr = 0.00236814
I0428 14:53:05.816133  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:53:06.669508  7476 solver.cpp:218] Iteration 7284 (2.09949 iter/s, 5.71568s/12 iters), loss = 0.878258
I0428 14:53:06.669559  7476 solver.cpp:237]     Train net output #0: loss = 0.878258 (* 1 = 0.878258 loss)
I0428 14:53:06.669569  7476 sgd_solver.cpp:105] Iteration 7284, lr = 0.00236252
I0428 14:53:12.325321  7476 solver.cpp:218] Iteration 7296 (2.12179 iter/s, 5.65559s/12 iters), loss = 0.724425
I0428 14:53:12.325419  7476 solver.cpp:237]     Train net output #0: loss = 0.724425 (* 1 = 0.724425 loss)
I0428 14:53:12.325429  7476 sgd_solver.cpp:105] Iteration 7296, lr = 0.00235691
I0428 14:53:17.916779  7476 solver.cpp:218] Iteration 7308 (2.14623 iter/s, 5.59119s/12 iters), loss = 0.714595
I0428 14:53:17.916821  7476 solver.cpp:237]     Train net output #0: loss = 0.714595 (* 1 = 0.714595 loss)
I0428 14:53:17.916829  7476 sgd_solver.cpp:105] Iteration 7308, lr = 0.00235131
I0428 14:53:23.480773  7476 solver.cpp:218] Iteration 7320 (2.1568 iter/s, 5.56379s/12 iters), loss = 0.614944
I0428 14:53:23.480815  7476 solver.cpp:237]     Train net output #0: loss = 0.614944 (* 1 = 0.614944 loss)
I0428 14:53:23.480824  7476 sgd_solver.cpp:105] Iteration 7320, lr = 0.00234573
I0428 14:53:29.033584  7476 solver.cpp:218] Iteration 7332 (2.16115 iter/s, 5.5526s/12 iters), loss = 0.819085
I0428 14:53:29.033634  7476 solver.cpp:237]     Train net output #0: loss = 0.819085 (* 1 = 0.819085 loss)
I0428 14:53:29.033644  7476 sgd_solver.cpp:105] Iteration 7332, lr = 0.00234016
I0428 14:53:34.009873  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_7344.caffemodel
I0428 14:53:37.741394  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_7344.solverstate
I0428 14:53:39.442904  7476 solver.cpp:330] Iteration 7344, Testing net (#0)
I0428 14:53:39.442924  7476 net.cpp:676] Ignoring source layer train-data
I0428 14:53:41.027768  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:53:44.188550  7476 solver.cpp:397]     Test net output #0: accuracy = 0.379902
I0428 14:53:44.188711  7476 solver.cpp:397]     Test net output #1: loss = 3.25399 (* 1 = 3.25399 loss)
I0428 14:53:44.316711  7476 solver.cpp:218] Iteration 7344 (0.785204 iter/s, 15.2826s/12 iters), loss = 0.81732
I0428 14:53:44.316754  7476 solver.cpp:237]     Train net output #0: loss = 0.81732 (* 1 = 0.81732 loss)
I0428 14:53:44.316763  7476 sgd_solver.cpp:105] Iteration 7344, lr = 0.0023346
I0428 14:53:48.897424  7476 solver.cpp:218] Iteration 7356 (2.61978 iter/s, 4.58053s/12 iters), loss = 0.566715
I0428 14:53:48.897470  7476 solver.cpp:237]     Train net output #0: loss = 0.566715 (* 1 = 0.566715 loss)
I0428 14:53:48.897480  7476 sgd_solver.cpp:105] Iteration 7356, lr = 0.00232906
I0428 14:53:54.329043  7476 solver.cpp:218] Iteration 7368 (2.20937 iter/s, 5.43141s/12 iters), loss = 0.797635
I0428 14:53:54.329089  7476 solver.cpp:237]     Train net output #0: loss = 0.797635 (* 1 = 0.797635 loss)
I0428 14:53:54.329098  7476 sgd_solver.cpp:105] Iteration 7368, lr = 0.00232353
I0428 14:53:59.807862  7476 solver.cpp:218] Iteration 7380 (2.19034 iter/s, 5.4786s/12 iters), loss = 0.582417
I0428 14:53:59.807919  7476 solver.cpp:237]     Train net output #0: loss = 0.582417 (* 1 = 0.582417 loss)
I0428 14:53:59.807932  7476 sgd_solver.cpp:105] Iteration 7380, lr = 0.00231802
I0428 14:54:01.306967  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:54:05.446103  7476 solver.cpp:218] Iteration 7392 (2.12841 iter/s, 5.63801s/12 iters), loss = 0.717229
I0428 14:54:05.446162  7476 solver.cpp:237]     Train net output #0: loss = 0.717229 (* 1 = 0.717229 loss)
I0428 14:54:05.446175  7476 sgd_solver.cpp:105] Iteration 7392, lr = 0.00231251
I0428 14:54:10.829879  7476 solver.cpp:218] Iteration 7404 (2.22901 iter/s, 5.38356s/12 iters), loss = 0.796338
I0428 14:54:10.829922  7476 solver.cpp:237]     Train net output #0: loss = 0.796338 (* 1 = 0.796338 loss)
I0428 14:54:10.829931  7476 sgd_solver.cpp:105] Iteration 7404, lr = 0.00230702
I0428 14:54:16.363497  7476 solver.cpp:218] Iteration 7416 (2.16865 iter/s, 5.5334s/12 iters), loss = 0.742198
I0428 14:54:16.363649  7476 solver.cpp:237]     Train net output #0: loss = 0.742198 (* 1 = 0.742198 loss)
I0428 14:54:16.363665  7476 sgd_solver.cpp:105] Iteration 7416, lr = 0.00230154
I0428 14:54:21.754130  7476 solver.cpp:218] Iteration 7428 (2.22621 iter/s, 5.39032s/12 iters), loss = 0.767749
I0428 14:54:21.754194  7476 solver.cpp:237]     Train net output #0: loss = 0.767749 (* 1 = 0.767749 loss)
I0428 14:54:21.754209  7476 sgd_solver.cpp:105] Iteration 7428, lr = 0.00229608
I0428 14:54:27.283531  7476 solver.cpp:218] Iteration 7440 (2.17031 iter/s, 5.52918s/12 iters), loss = 0.587068
I0428 14:54:27.283579  7476 solver.cpp:237]     Train net output #0: loss = 0.587068 (* 1 = 0.587068 loss)
I0428 14:54:27.283588  7476 sgd_solver.cpp:105] Iteration 7440, lr = 0.00229063
I0428 14:54:29.466995  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_7446.caffemodel
I0428 14:54:31.489058  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_7446.solverstate
I0428 14:54:34.682660  7476 solver.cpp:330] Iteration 7446, Testing net (#0)
I0428 14:54:34.682684  7476 net.cpp:676] Ignoring source layer train-data
I0428 14:54:36.134407  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:54:39.218060  7476 solver.cpp:397]     Test net output #0: accuracy = 0.389706
I0428 14:54:39.218101  7476 solver.cpp:397]     Test net output #1: loss = 3.07456 (* 1 = 3.07456 loss)
I0428 14:54:41.262320  7476 solver.cpp:218] Iteration 7452 (0.858471 iter/s, 13.9783s/12 iters), loss = 0.772387
I0428 14:54:41.262368  7476 solver.cpp:237]     Train net output #0: loss = 0.772387 (* 1 = 0.772387 loss)
I0428 14:54:41.262377  7476 sgd_solver.cpp:105] Iteration 7452, lr = 0.00228519
I0428 14:54:46.816538  7476 solver.cpp:218] Iteration 7464 (2.16061 iter/s, 5.554s/12 iters), loss = 0.561127
I0428 14:54:46.816685  7476 solver.cpp:237]     Train net output #0: loss = 0.561127 (* 1 = 0.561127 loss)
I0428 14:54:46.816696  7476 sgd_solver.cpp:105] Iteration 7464, lr = 0.00227976
I0428 14:54:52.301482  7476 solver.cpp:218] Iteration 7476 (2.18793 iter/s, 5.48463s/12 iters), loss = 0.658671
I0428 14:54:52.301525  7476 solver.cpp:237]     Train net output #0: loss = 0.658671 (* 1 = 0.658671 loss)
I0428 14:54:52.301535  7476 sgd_solver.cpp:105] Iteration 7476, lr = 0.00227435
I0428 14:54:55.943274  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:54:57.600205  7476 solver.cpp:218] Iteration 7488 (2.26478 iter/s, 5.29852s/12 iters), loss = 0.739568
I0428 14:54:57.600251  7476 solver.cpp:237]     Train net output #0: loss = 0.739568 (* 1 = 0.739568 loss)
I0428 14:54:57.600260  7476 sgd_solver.cpp:105] Iteration 7488, lr = 0.00226895
I0428 14:55:03.518787  7476 solver.cpp:218] Iteration 7500 (2.02759 iter/s, 5.91835s/12 iters), loss = 0.613555
I0428 14:55:03.518841  7476 solver.cpp:237]     Train net output #0: loss = 0.613555 (* 1 = 0.613555 loss)
I0428 14:55:03.518852  7476 sgd_solver.cpp:105] Iteration 7500, lr = 0.00226357
I0428 14:55:09.126904  7476 solver.cpp:218] Iteration 7512 (2.13984 iter/s, 5.6079s/12 iters), loss = 0.506662
I0428 14:55:09.126955  7476 solver.cpp:237]     Train net output #0: loss = 0.506662 (* 1 = 0.506662 loss)
I0428 14:55:09.126963  7476 sgd_solver.cpp:105] Iteration 7512, lr = 0.00225819
I0428 14:55:14.603541  7476 solver.cpp:218] Iteration 7524 (2.19121 iter/s, 5.47642s/12 iters), loss = 0.659934
I0428 14:55:14.603581  7476 solver.cpp:237]     Train net output #0: loss = 0.659934 (* 1 = 0.659934 loss)
I0428 14:55:14.603590  7476 sgd_solver.cpp:105] Iteration 7524, lr = 0.00225283
I0428 14:55:20.103844  7476 solver.cpp:218] Iteration 7536 (2.18178 iter/s, 5.50009s/12 iters), loss = 0.41383
I0428 14:55:20.103976  7476 solver.cpp:237]     Train net output #0: loss = 0.41383 (* 1 = 0.41383 loss)
I0428 14:55:20.103989  7476 sgd_solver.cpp:105] Iteration 7536, lr = 0.00224748
I0428 14:55:25.100721  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_7548.caffemodel
I0428 14:55:30.624225  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_7548.solverstate
I0428 14:55:32.379009  7476 solver.cpp:330] Iteration 7548, Testing net (#0)
I0428 14:55:32.379029  7476 net.cpp:676] Ignoring source layer train-data
I0428 14:55:33.860576  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:55:37.035714  7476 solver.cpp:397]     Test net output #0: accuracy = 0.403186
I0428 14:55:37.035746  7476 solver.cpp:397]     Test net output #1: loss = 3.14592 (* 1 = 3.14592 loss)
I0428 14:55:37.165531  7476 solver.cpp:218] Iteration 7548 (0.703355 iter/s, 17.0611s/12 iters), loss = 0.620544
I0428 14:55:37.165580  7476 solver.cpp:237]     Train net output #0: loss = 0.620544 (* 1 = 0.620544 loss)
I0428 14:55:37.165591  7476 sgd_solver.cpp:105] Iteration 7548, lr = 0.00224215
I0428 14:55:41.836375  7476 solver.cpp:218] Iteration 7560 (2.56924 iter/s, 4.67065s/12 iters), loss = 0.626288
I0428 14:55:41.836419  7476 solver.cpp:237]     Train net output #0: loss = 0.626288 (* 1 = 0.626288 loss)
I0428 14:55:41.836427  7476 sgd_solver.cpp:105] Iteration 7560, lr = 0.00223682
I0428 14:55:47.217998  7476 solver.cpp:218] Iteration 7572 (2.2299 iter/s, 5.38141s/12 iters), loss = 0.589936
I0428 14:55:47.218044  7476 solver.cpp:237]     Train net output #0: loss = 0.589936 (* 1 = 0.589936 loss)
I0428 14:55:47.218053  7476 sgd_solver.cpp:105] Iteration 7572, lr = 0.00223151
I0428 14:55:52.639503  7476 solver.cpp:218] Iteration 7584 (2.21349 iter/s, 5.42129s/12 iters), loss = 0.622053
I0428 14:55:52.639652  7476 solver.cpp:237]     Train net output #0: loss = 0.622053 (* 1 = 0.622053 loss)
I0428 14:55:52.639662  7476 sgd_solver.cpp:105] Iteration 7584, lr = 0.00222621
I0428 14:55:53.318732  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:55:58.059825  7476 solver.cpp:218] Iteration 7596 (2.21402 iter/s, 5.42002s/12 iters), loss = 0.458902
I0428 14:55:58.059865  7476 solver.cpp:237]     Train net output #0: loss = 0.458902 (* 1 = 0.458902 loss)
I0428 14:55:58.059873  7476 sgd_solver.cpp:105] Iteration 7596, lr = 0.00222093
I0428 14:56:03.602702  7476 solver.cpp:218] Iteration 7608 (2.16502 iter/s, 5.54267s/12 iters), loss = 0.624061
I0428 14:56:03.602747  7476 solver.cpp:237]     Train net output #0: loss = 0.624061 (* 1 = 0.624061 loss)
I0428 14:56:03.602756  7476 sgd_solver.cpp:105] Iteration 7608, lr = 0.00221565
I0428 14:56:09.149734  7476 solver.cpp:218] Iteration 7620 (2.1634 iter/s, 5.54681s/12 iters), loss = 0.665508
I0428 14:56:09.149791  7476 solver.cpp:237]     Train net output #0: loss = 0.665508 (* 1 = 0.665508 loss)
I0428 14:56:09.149801  7476 sgd_solver.cpp:105] Iteration 7620, lr = 0.00221039
I0428 14:56:11.672047  7476 blocking_queue.cpp:49] Waiting for data
I0428 14:56:14.542315  7476 solver.cpp:218] Iteration 7632 (2.22537 iter/s, 5.39236s/12 iters), loss = 0.691908
I0428 14:56:14.542357  7476 solver.cpp:237]     Train net output #0: loss = 0.691908 (* 1 = 0.691908 loss)
I0428 14:56:14.542367  7476 sgd_solver.cpp:105] Iteration 7632, lr = 0.00220515
I0428 14:56:19.901471  7476 solver.cpp:218] Iteration 7644 (2.23924 iter/s, 5.35895s/12 iters), loss = 0.539516
I0428 14:56:19.901515  7476 solver.cpp:237]     Train net output #0: loss = 0.539516 (* 1 = 0.539516 loss)
I0428 14:56:19.901525  7476 sgd_solver.cpp:105] Iteration 7644, lr = 0.00219991
I0428 14:56:22.068239  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_7650.caffemodel
I0428 14:56:29.563757  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_7650.solverstate
I0428 14:56:36.928861  7476 solver.cpp:330] Iteration 7650, Testing net (#0)
I0428 14:56:36.928889  7476 net.cpp:676] Ignoring source layer train-data
I0428 14:56:38.408917  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:56:41.540334  7476 solver.cpp:397]     Test net output #0: accuracy = 0.382966
I0428 14:56:41.540367  7476 solver.cpp:397]     Test net output #1: loss = 3.26066 (* 1 = 3.26066 loss)
I0428 14:56:43.495931  7476 solver.cpp:218] Iteration 7656 (0.508609 iter/s, 23.5938s/12 iters), loss = 0.565396
I0428 14:56:43.495996  7476 solver.cpp:237]     Train net output #0: loss = 0.565396 (* 1 = 0.565396 loss)
I0428 14:56:43.496008  7476 sgd_solver.cpp:105] Iteration 7656, lr = 0.00219469
I0428 14:56:48.895066  7476 solver.cpp:218] Iteration 7668 (2.22267 iter/s, 5.3989s/12 iters), loss = 0.483676
I0428 14:56:48.895115  7476 solver.cpp:237]     Train net output #0: loss = 0.483676 (* 1 = 0.483676 loss)
I0428 14:56:48.895126  7476 sgd_solver.cpp:105] Iteration 7668, lr = 0.00218948
I0428 14:56:54.446162  7476 solver.cpp:218] Iteration 7680 (2.16182 iter/s, 5.55089s/12 iters), loss = 0.669687
I0428 14:56:54.446202  7476 solver.cpp:237]     Train net output #0: loss = 0.669687 (* 1 = 0.669687 loss)
I0428 14:56:54.446210  7476 sgd_solver.cpp:105] Iteration 7680, lr = 0.00218428
I0428 14:56:57.485817  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:56:59.957129  7476 solver.cpp:218] Iteration 7692 (2.17756 iter/s, 5.51076s/12 iters), loss = 0.717566
I0428 14:56:59.957448  7476 solver.cpp:237]     Train net output #0: loss = 0.717566 (* 1 = 0.717566 loss)
I0428 14:56:59.957458  7476 sgd_solver.cpp:105] Iteration 7692, lr = 0.00217909
I0428 14:57:05.477689  7476 solver.cpp:218] Iteration 7704 (2.17389 iter/s, 5.52007s/12 iters), loss = 0.495471
I0428 14:57:05.477749  7476 solver.cpp:237]     Train net output #0: loss = 0.495471 (* 1 = 0.495471 loss)
I0428 14:57:05.477763  7476 sgd_solver.cpp:105] Iteration 7704, lr = 0.00217392
I0428 14:57:11.109313  7476 solver.cpp:218] Iteration 7716 (2.13091 iter/s, 5.6314s/12 iters), loss = 0.537202
I0428 14:57:11.109356  7476 solver.cpp:237]     Train net output #0: loss = 0.537202 (* 1 = 0.537202 loss)
I0428 14:57:11.109366  7476 sgd_solver.cpp:105] Iteration 7716, lr = 0.00216876
I0428 14:57:16.613373  7476 solver.cpp:218] Iteration 7728 (2.18029 iter/s, 5.50384s/12 iters), loss = 0.41698
I0428 14:57:16.613440  7476 solver.cpp:237]     Train net output #0: loss = 0.41698 (* 1 = 0.41698 loss)
I0428 14:57:16.613452  7476 sgd_solver.cpp:105] Iteration 7728, lr = 0.00216361
I0428 14:57:22.079836  7476 solver.cpp:218] Iteration 7740 (2.1953 iter/s, 5.46623s/12 iters), loss = 0.724618
I0428 14:57:22.079881  7476 solver.cpp:237]     Train net output #0: loss = 0.724618 (* 1 = 0.724618 loss)
I0428 14:57:22.079890  7476 sgd_solver.cpp:105] Iteration 7740, lr = 0.00215847
I0428 14:57:26.994025  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_7752.caffemodel
I0428 14:57:34.924226  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_7752.solverstate
I0428 14:57:37.945597  7476 solver.cpp:330] Iteration 7752, Testing net (#0)
I0428 14:57:37.945624  7476 net.cpp:676] Ignoring source layer train-data
I0428 14:57:39.496223  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:57:42.627044  7476 solver.cpp:397]     Test net output #0: accuracy = 0.402574
I0428 14:57:42.627085  7476 solver.cpp:397]     Test net output #1: loss = 3.22371 (* 1 = 3.22371 loss)
I0428 14:57:42.751387  7476 solver.cpp:218] Iteration 7752 (0.580527 iter/s, 20.6709s/12 iters), loss = 0.476027
I0428 14:57:42.751443  7476 solver.cpp:237]     Train net output #0: loss = 0.476027 (* 1 = 0.476027 loss)
I0428 14:57:42.751458  7476 sgd_solver.cpp:105] Iteration 7752, lr = 0.00215335
I0428 14:57:47.321563  7476 solver.cpp:218] Iteration 7764 (2.62584 iter/s, 4.56997s/12 iters), loss = 0.514915
I0428 14:57:47.321610  7476 solver.cpp:237]     Train net output #0: loss = 0.514915 (* 1 = 0.514915 loss)
I0428 14:57:47.321619  7476 sgd_solver.cpp:105] Iteration 7764, lr = 0.00214823
I0428 14:57:53.043923  7476 solver.cpp:218] Iteration 7776 (2.09712 iter/s, 5.72212s/12 iters), loss = 0.555569
I0428 14:57:53.043967  7476 solver.cpp:237]     Train net output #0: loss = 0.555569 (* 1 = 0.555569 loss)
I0428 14:57:53.043977  7476 sgd_solver.cpp:105] Iteration 7776, lr = 0.00214313
I0428 14:57:58.494657  7476 solver.cpp:218] Iteration 7788 (2.20163 iter/s, 5.4505s/12 iters), loss = 0.544733
I0428 14:57:58.494712  7476 solver.cpp:237]     Train net output #0: loss = 0.544733 (* 1 = 0.544733 loss)
I0428 14:57:58.494724  7476 sgd_solver.cpp:105] Iteration 7788, lr = 0.00213805
I0428 14:57:58.502362  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:58:03.935356  7476 solver.cpp:218] Iteration 7800 (2.20569 iter/s, 5.44046s/12 iters), loss = 0.600085
I0428 14:58:03.935408  7476 solver.cpp:237]     Train net output #0: loss = 0.600085 (* 1 = 0.600085 loss)
I0428 14:58:03.935421  7476 sgd_solver.cpp:105] Iteration 7800, lr = 0.00213297
I0428 14:58:09.315686  7476 solver.cpp:218] Iteration 7812 (2.23044 iter/s, 5.3801s/12 iters), loss = 0.628254
I0428 14:58:09.315795  7476 solver.cpp:237]     Train net output #0: loss = 0.628254 (* 1 = 0.628254 loss)
I0428 14:58:09.315805  7476 sgd_solver.cpp:105] Iteration 7812, lr = 0.00212791
I0428 14:58:14.741983  7476 solver.cpp:218] Iteration 7824 (2.21157 iter/s, 5.42601s/12 iters), loss = 0.515718
I0428 14:58:14.742027  7476 solver.cpp:237]     Train net output #0: loss = 0.515718 (* 1 = 0.515718 loss)
I0428 14:58:14.742035  7476 sgd_solver.cpp:105] Iteration 7824, lr = 0.00212285
I0428 14:58:20.107231  7476 solver.cpp:218] Iteration 7836 (2.23671 iter/s, 5.36503s/12 iters), loss = 0.495535
I0428 14:58:20.107276  7476 solver.cpp:237]     Train net output #0: loss = 0.495535 (* 1 = 0.495535 loss)
I0428 14:58:20.107285  7476 sgd_solver.cpp:105] Iteration 7836, lr = 0.00211781
I0428 14:58:25.550400  7476 solver.cpp:218] Iteration 7848 (2.20469 iter/s, 5.44295s/12 iters), loss = 0.594387
I0428 14:58:25.550437  7476 solver.cpp:237]     Train net output #0: loss = 0.594387 (* 1 = 0.594387 loss)
I0428 14:58:25.550446  7476 sgd_solver.cpp:105] Iteration 7848, lr = 0.00211279
I0428 14:58:27.730950  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_7854.caffemodel
I0428 14:58:32.217612  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_7854.solverstate
I0428 14:58:33.998081  7476 solver.cpp:330] Iteration 7854, Testing net (#0)
I0428 14:58:33.998101  7476 net.cpp:676] Ignoring source layer train-data
I0428 14:58:35.288625  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:58:38.471359  7476 solver.cpp:397]     Test net output #0: accuracy = 0.409314
I0428 14:58:38.471396  7476 solver.cpp:397]     Test net output #1: loss = 3.10184 (* 1 = 3.10184 loss)
I0428 14:58:40.466799  7476 solver.cpp:218] Iteration 7860 (0.80451 iter/s, 14.9159s/12 iters), loss = 0.616651
I0428 14:58:40.466987  7476 solver.cpp:237]     Train net output #0: loss = 0.616651 (* 1 = 0.616651 loss)
I0428 14:58:40.466998  7476 sgd_solver.cpp:105] Iteration 7860, lr = 0.00210777
I0428 14:58:45.884675  7476 solver.cpp:218] Iteration 7872 (2.21504 iter/s, 5.41751s/12 iters), loss = 0.502696
I0428 14:58:45.884727  7476 solver.cpp:237]     Train net output #0: loss = 0.502696 (* 1 = 0.502696 loss)
I0428 14:58:45.884737  7476 sgd_solver.cpp:105] Iteration 7872, lr = 0.00210277
I0428 14:58:51.382243  7476 solver.cpp:218] Iteration 7884 (2.18287 iter/s, 5.49735s/12 iters), loss = 0.529104
I0428 14:58:51.382282  7476 solver.cpp:237]     Train net output #0: loss = 0.529104 (* 1 = 0.529104 loss)
I0428 14:58:51.382292  7476 sgd_solver.cpp:105] Iteration 7884, lr = 0.00209777
I0428 14:58:53.766746  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:58:56.878739  7476 solver.cpp:218] Iteration 7896 (2.1833 iter/s, 5.49628s/12 iters), loss = 0.755278
I0428 14:58:56.878782  7476 solver.cpp:237]     Train net output #0: loss = 0.755278 (* 1 = 0.755278 loss)
I0428 14:58:56.878790  7476 sgd_solver.cpp:105] Iteration 7896, lr = 0.00209279
I0428 14:59:02.331445  7476 solver.cpp:218] Iteration 7908 (2.20083 iter/s, 5.45249s/12 iters), loss = 0.469467
I0428 14:59:02.331488  7476 solver.cpp:237]     Train net output #0: loss = 0.469467 (* 1 = 0.469467 loss)
I0428 14:59:02.331496  7476 sgd_solver.cpp:105] Iteration 7908, lr = 0.00208782
I0428 14:59:07.740015  7476 solver.cpp:218] Iteration 7920 (2.21879 iter/s, 5.40835s/12 iters), loss = 0.493039
I0428 14:59:07.740059  7476 solver.cpp:237]     Train net output #0: loss = 0.493039 (* 1 = 0.493039 loss)
I0428 14:59:07.740068  7476 sgd_solver.cpp:105] Iteration 7920, lr = 0.00208287
I0428 14:59:13.381201  7476 solver.cpp:218] Iteration 7932 (2.1273 iter/s, 5.64096s/12 iters), loss = 0.508372
I0428 14:59:13.381312  7476 solver.cpp:237]     Train net output #0: loss = 0.508372 (* 1 = 0.508372 loss)
I0428 14:59:13.381323  7476 sgd_solver.cpp:105] Iteration 7932, lr = 0.00207792
I0428 14:59:18.932044  7476 solver.cpp:218] Iteration 7944 (2.16194 iter/s, 5.55056s/12 iters), loss = 0.761352
I0428 14:59:18.932087  7476 solver.cpp:237]     Train net output #0: loss = 0.761352 (* 1 = 0.761352 loss)
I0428 14:59:18.932097  7476 sgd_solver.cpp:105] Iteration 7944, lr = 0.00207299
I0428 14:59:23.878531  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_7956.caffemodel
I0428 14:59:28.838415  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_7956.solverstate
I0428 14:59:30.758018  7476 solver.cpp:330] Iteration 7956, Testing net (#0)
I0428 14:59:30.758036  7476 net.cpp:676] Ignoring source layer train-data
I0428 14:59:32.028606  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:59:35.337595  7476 solver.cpp:397]     Test net output #0: accuracy = 0.415441
I0428 14:59:35.337627  7476 solver.cpp:397]     Test net output #1: loss = 3.22083 (* 1 = 3.22083 loss)
I0428 14:59:35.460551  7476 solver.cpp:218] Iteration 7956 (0.726042 iter/s, 16.528s/12 iters), loss = 0.535074
I0428 14:59:35.460615  7476 solver.cpp:237]     Train net output #0: loss = 0.535074 (* 1 = 0.535074 loss)
I0428 14:59:35.460628  7476 sgd_solver.cpp:105] Iteration 7956, lr = 0.00206807
I0428 14:59:40.132033  7476 solver.cpp:218] Iteration 7968 (2.56889 iter/s, 4.67127s/12 iters), loss = 0.689274
I0428 14:59:40.132078  7476 solver.cpp:237]     Train net output #0: loss = 0.689274 (* 1 = 0.689274 loss)
I0428 14:59:40.132087  7476 sgd_solver.cpp:105] Iteration 7968, lr = 0.00206316
I0428 14:59:45.639185  7476 solver.cpp:218] Iteration 7980 (2.17907 iter/s, 5.50693s/12 iters), loss = 0.488555
I0428 14:59:45.639323  7476 solver.cpp:237]     Train net output #0: loss = 0.488555 (* 1 = 0.488555 loss)
I0428 14:59:45.639333  7476 sgd_solver.cpp:105] Iteration 7980, lr = 0.00205826
I0428 14:59:50.339344  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:59:51.193104  7476 solver.cpp:218] Iteration 7992 (2.16076 iter/s, 5.55361s/12 iters), loss = 0.466812
I0428 14:59:51.193145  7476 solver.cpp:237]     Train net output #0: loss = 0.466812 (* 1 = 0.466812 loss)
I0428 14:59:51.193154  7476 sgd_solver.cpp:105] Iteration 7992, lr = 0.00205337
I0428 14:59:56.465669  7476 solver.cpp:218] Iteration 8004 (2.27602 iter/s, 5.27235s/12 iters), loss = 0.502394
I0428 14:59:56.465713  7476 solver.cpp:237]     Train net output #0: loss = 0.502394 (* 1 = 0.502394 loss)
I0428 14:59:56.465725  7476 sgd_solver.cpp:105] Iteration 8004, lr = 0.0020485
I0428 15:00:01.879709  7476 solver.cpp:218] Iteration 8016 (2.21655 iter/s, 5.41382s/12 iters), loss = 0.496887
I0428 15:00:01.879752  7476 solver.cpp:237]     Train net output #0: loss = 0.496887 (* 1 = 0.496887 loss)
I0428 15:00:01.879761  7476 sgd_solver.cpp:105] Iteration 8016, lr = 0.00204363
I0428 15:00:07.364182  7476 solver.cpp:218] Iteration 8028 (2.18808 iter/s, 5.48425s/12 iters), loss = 0.675149
I0428 15:00:07.364230  7476 solver.cpp:237]     Train net output #0: loss = 0.675149 (* 1 = 0.675149 loss)
I0428 15:00:07.364243  7476 sgd_solver.cpp:105] Iteration 8028, lr = 0.00203878
I0428 15:00:12.820776  7476 solver.cpp:218] Iteration 8040 (2.19926 iter/s, 5.45638s/12 iters), loss = 0.657129
I0428 15:00:12.820819  7476 solver.cpp:237]     Train net output #0: loss = 0.657129 (* 1 = 0.657129 loss)
I0428 15:00:12.820828  7476 sgd_solver.cpp:105] Iteration 8040, lr = 0.00203394
I0428 15:00:18.257437  7476 solver.cpp:218] Iteration 8052 (2.20733 iter/s, 5.43644s/12 iters), loss = 0.593908
I0428 15:00:18.257563  7476 solver.cpp:237]     Train net output #0: loss = 0.593908 (* 1 = 0.593908 loss)
I0428 15:00:18.257575  7476 sgd_solver.cpp:105] Iteration 8052, lr = 0.00202911
I0428 15:00:20.420397  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_8058.caffemodel
I0428 15:00:30.368584  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_8058.solverstate
I0428 15:00:34.017133  7476 solver.cpp:330] Iteration 8058, Testing net (#0)
I0428 15:00:34.017154  7476 net.cpp:676] Ignoring source layer train-data
I0428 15:00:35.285778  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:00:38.494733  7476 solver.cpp:397]     Test net output #0: accuracy = 0.414216
I0428 15:00:38.494766  7476 solver.cpp:397]     Test net output #1: loss = 3.18333 (* 1 = 3.18333 loss)
I0428 15:00:40.409230  7476 solver.cpp:218] Iteration 8064 (0.541736 iter/s, 22.151s/12 iters), loss = 0.456146
I0428 15:00:40.409273  7476 solver.cpp:237]     Train net output #0: loss = 0.456146 (* 1 = 0.456146 loss)
I0428 15:00:40.409282  7476 sgd_solver.cpp:105] Iteration 8064, lr = 0.00202429
I0428 15:00:45.833755  7476 solver.cpp:218] Iteration 8076 (2.21226 iter/s, 5.42431s/12 iters), loss = 0.569568
I0428 15:00:45.833797  7476 solver.cpp:237]     Train net output #0: loss = 0.569568 (* 1 = 0.569568 loss)
I0428 15:00:45.833806  7476 sgd_solver.cpp:105] Iteration 8076, lr = 0.00201949
I0428 15:00:51.325249  7476 solver.cpp:218] Iteration 8088 (2.18529 iter/s, 5.49127s/12 iters), loss = 0.659988
I0428 15:00:51.325387  7476 solver.cpp:237]     Train net output #0: loss = 0.659988 (* 1 = 0.659988 loss)
I0428 15:00:51.325397  7476 sgd_solver.cpp:105] Iteration 8088, lr = 0.00201469
I0428 15:00:52.872617  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:00:56.694875  7476 solver.cpp:218] Iteration 8100 (2.23492 iter/s, 5.36931s/12 iters), loss = 0.628014
I0428 15:00:56.694917  7476 solver.cpp:237]     Train net output #0: loss = 0.628014 (* 1 = 0.628014 loss)
I0428 15:00:56.694927  7476 sgd_solver.cpp:105] Iteration 8100, lr = 0.00200991
I0428 15:01:02.147150  7476 solver.cpp:218] Iteration 8112 (2.20101 iter/s, 5.45205s/12 iters), loss = 0.50871
I0428 15:01:02.147197  7476 solver.cpp:237]     Train net output #0: loss = 0.50871 (* 1 = 0.50871 loss)
I0428 15:01:02.147207  7476 sgd_solver.cpp:105] Iteration 8112, lr = 0.00200514
I0428 15:01:07.729785  7476 solver.cpp:218] Iteration 8124 (2.14961 iter/s, 5.58241s/12 iters), loss = 0.687445
I0428 15:01:07.729827  7476 solver.cpp:237]     Train net output #0: loss = 0.687445 (* 1 = 0.687445 loss)
I0428 15:01:07.729835  7476 sgd_solver.cpp:105] Iteration 8124, lr = 0.00200038
I0428 15:01:13.078768  7476 solver.cpp:218] Iteration 8136 (2.24351 iter/s, 5.34876s/12 iters), loss = 0.478295
I0428 15:01:13.078819  7476 solver.cpp:237]     Train net output #0: loss = 0.478295 (* 1 = 0.478295 loss)
I0428 15:01:13.078827  7476 sgd_solver.cpp:105] Iteration 8136, lr = 0.00199563
I0428 15:01:18.568153  7476 solver.cpp:218] Iteration 8148 (2.18613 iter/s, 5.48916s/12 iters), loss = 0.409402
I0428 15:01:18.568197  7476 solver.cpp:237]     Train net output #0: loss = 0.409402 (* 1 = 0.409402 loss)
I0428 15:01:18.568207  7476 sgd_solver.cpp:105] Iteration 8148, lr = 0.00199089
I0428 15:01:23.438105  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_8160.caffemodel
I0428 15:01:29.396008  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_8160.solverstate
I0428 15:01:33.375892  7476 solver.cpp:330] Iteration 8160, Testing net (#0)
I0428 15:01:33.375922  7476 net.cpp:676] Ignoring source layer train-data
I0428 15:01:34.755966  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:01:38.119263  7476 solver.cpp:397]     Test net output #0: accuracy = 0.433211
I0428 15:01:38.119295  7476 solver.cpp:397]     Test net output #1: loss = 3.18134 (* 1 = 3.18134 loss)
I0428 15:01:38.247210  7476 solver.cpp:218] Iteration 8160 (0.609805 iter/s, 19.6784s/12 iters), loss = 0.326681
I0428 15:01:38.247257  7476 solver.cpp:237]     Train net output #0: loss = 0.326681 (* 1 = 0.326681 loss)
I0428 15:01:38.247268  7476 sgd_solver.cpp:105] Iteration 8160, lr = 0.00198616
I0428 15:01:42.894696  7476 solver.cpp:218] Iteration 8172 (2.58215 iter/s, 4.64729s/12 iters), loss = 0.480526
I0428 15:01:42.894735  7476 solver.cpp:237]     Train net output #0: loss = 0.480526 (* 1 = 0.480526 loss)
I0428 15:01:42.894744  7476 sgd_solver.cpp:105] Iteration 8172, lr = 0.00198145
I0428 15:01:48.314311  7476 solver.cpp:218] Iteration 8184 (2.21427 iter/s, 5.4194s/12 iters), loss = 0.634473
I0428 15:01:48.314355  7476 solver.cpp:237]     Train net output #0: loss = 0.634473 (* 1 = 0.634473 loss)
I0428 15:01:48.314364  7476 sgd_solver.cpp:105] Iteration 8184, lr = 0.00197674
I0428 15:01:52.164875  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:01:53.775137  7476 solver.cpp:218] Iteration 8196 (2.19755 iter/s, 5.46062s/12 iters), loss = 0.390167
I0428 15:01:53.775243  7476 solver.cpp:237]     Train net output #0: loss = 0.390167 (* 1 = 0.390167 loss)
I0428 15:01:53.775254  7476 sgd_solver.cpp:105] Iteration 8196, lr = 0.00197205
I0428 15:01:59.179868  7476 solver.cpp:218] Iteration 8208 (2.22039 iter/s, 5.40446s/12 iters), loss = 0.37656
I0428 15:01:59.179910  7476 solver.cpp:237]     Train net output #0: loss = 0.37656 (* 1 = 0.37656 loss)
I0428 15:01:59.179917  7476 sgd_solver.cpp:105] Iteration 8208, lr = 0.00196737
I0428 15:02:04.648339  7476 solver.cpp:218] Iteration 8220 (2.19448 iter/s, 5.46826s/12 iters), loss = 0.648319
I0428 15:02:04.648375  7476 solver.cpp:237]     Train net output #0: loss = 0.648319 (* 1 = 0.648319 loss)
I0428 15:02:04.648382  7476 sgd_solver.cpp:105] Iteration 8220, lr = 0.0019627
I0428 15:02:10.048064  7476 solver.cpp:218] Iteration 8232 (2.22242 iter/s, 5.39952s/12 iters), loss = 0.649307
I0428 15:02:10.048110  7476 solver.cpp:237]     Train net output #0: loss = 0.649307 (* 1 = 0.649307 loss)
I0428 15:02:10.048120  7476 sgd_solver.cpp:105] Iteration 8232, lr = 0.00195804
I0428 15:02:15.563699  7476 solver.cpp:218] Iteration 8244 (2.17572 iter/s, 5.51542s/12 iters), loss = 0.313535
I0428 15:02:15.563738  7476 solver.cpp:237]     Train net output #0: loss = 0.313535 (* 1 = 0.313535 loss)
I0428 15:02:15.563746  7476 sgd_solver.cpp:105] Iteration 8244, lr = 0.00195339
I0428 15:02:20.850656  7476 solver.cpp:218] Iteration 8256 (2.26983 iter/s, 5.28675s/12 iters), loss = 0.565386
I0428 15:02:20.850697  7476 solver.cpp:237]     Train net output #0: loss = 0.565386 (* 1 = 0.565386 loss)
I0428 15:02:20.850704  7476 sgd_solver.cpp:105] Iteration 8256, lr = 0.00194875
I0428 15:02:23.027782  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_8262.caffemodel
I0428 15:02:26.474179  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_8262.solverstate
I0428 15:02:34.924916  7476 solver.cpp:330] Iteration 8262, Testing net (#0)
I0428 15:02:34.924942  7476 net.cpp:676] Ignoring source layer train-data
I0428 15:02:36.053846  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:02:39.469385  7476 solver.cpp:397]     Test net output #0: accuracy = 0.403186
I0428 15:02:39.469413  7476 solver.cpp:397]     Test net output #1: loss = 3.37932 (* 1 = 3.37932 loss)
I0428 15:02:41.431219  7476 solver.cpp:218] Iteration 8268 (0.583093 iter/s, 20.5799s/12 iters), loss = 0.452319
I0428 15:02:41.431267  7476 solver.cpp:237]     Train net output #0: loss = 0.452319 (* 1 = 0.452319 loss)
I0428 15:02:41.431275  7476 sgd_solver.cpp:105] Iteration 8268, lr = 0.00194412
I0428 15:02:46.910989  7476 solver.cpp:218] Iteration 8280 (2.18996 iter/s, 5.47955s/12 iters), loss = 0.434519
I0428 15:02:46.911038  7476 solver.cpp:237]     Train net output #0: loss = 0.434519 (* 1 = 0.434519 loss)
I0428 15:02:46.911048  7476 sgd_solver.cpp:105] Iteration 8280, lr = 0.00193951
I0428 15:02:52.333696  7476 solver.cpp:218] Iteration 8292 (2.21301 iter/s, 5.42249s/12 iters), loss = 0.44126
I0428 15:02:52.333745  7476 solver.cpp:237]     Train net output #0: loss = 0.44126 (* 1 = 0.44126 loss)
I0428 15:02:52.333755  7476 sgd_solver.cpp:105] Iteration 8292, lr = 0.0019349
I0428 15:02:53.050565  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:02:57.722687  7476 solver.cpp:218] Iteration 8304 (2.22685 iter/s, 5.38877s/12 iters), loss = 0.616253
I0428 15:02:57.722802  7476 solver.cpp:237]     Train net output #0: loss = 0.616253 (* 1 = 0.616253 loss)
I0428 15:02:57.722813  7476 sgd_solver.cpp:105] Iteration 8304, lr = 0.00193031
I0428 15:03:00.829391  7476 blocking_queue.cpp:49] Waiting for data
I0428 15:03:03.163039  7476 solver.cpp:218] Iteration 8316 (2.20586 iter/s, 5.44006s/12 iters), loss = 0.469576
I0428 15:03:03.163082  7476 solver.cpp:237]     Train net output #0: loss = 0.469576 (* 1 = 0.469576 loss)
I0428 15:03:03.163091  7476 sgd_solver.cpp:105] Iteration 8316, lr = 0.00192573
I0428 15:03:08.467612  7476 solver.cpp:218] Iteration 8328 (2.26229 iter/s, 5.30436s/12 iters), loss = 0.511124
I0428 15:03:08.467661  7476 solver.cpp:237]     Train net output #0: loss = 0.511124 (* 1 = 0.511124 loss)
I0428 15:03:08.467675  7476 sgd_solver.cpp:105] Iteration 8328, lr = 0.00192115
I0428 15:03:13.952900  7476 solver.cpp:218] Iteration 8340 (2.18776 iter/s, 5.48506s/12 iters), loss = 0.490954
I0428 15:03:13.952946  7476 solver.cpp:237]     Train net output #0: loss = 0.490954 (* 1 = 0.490954 loss)
I0428 15:03:13.952955  7476 sgd_solver.cpp:105] Iteration 8340, lr = 0.00191659
I0428 15:03:19.297425  7476 solver.cpp:218] Iteration 8352 (2.24538 iter/s, 5.34431s/12 iters), loss = 0.441199
I0428 15:03:19.297466  7476 solver.cpp:237]     Train net output #0: loss = 0.441199 (* 1 = 0.441199 loss)
I0428 15:03:19.297477  7476 sgd_solver.cpp:105] Iteration 8352, lr = 0.00191204
I0428 15:03:24.066963  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_8364.caffemodel
I0428 15:03:25.442842  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_8364.solverstate
I0428 15:03:31.965288  7476 solver.cpp:330] Iteration 8364, Testing net (#0)
I0428 15:03:31.965389  7476 net.cpp:676] Ignoring source layer train-data
I0428 15:03:33.057171  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:03:36.409489  7476 solver.cpp:397]     Test net output #0: accuracy = 0.420956
I0428 15:03:36.409515  7476 solver.cpp:397]     Test net output #1: loss = 3.26194 (* 1 = 3.26194 loss)
I0428 15:03:36.537212  7476 solver.cpp:218] Iteration 8364 (0.696086 iter/s, 17.2392s/12 iters), loss = 0.470085
I0428 15:03:36.537259  7476 solver.cpp:237]     Train net output #0: loss = 0.470085 (* 1 = 0.470085 loss)
I0428 15:03:36.537268  7476 sgd_solver.cpp:105] Iteration 8364, lr = 0.0019075
I0428 15:03:41.132215  7476 solver.cpp:218] Iteration 8376 (2.61164 iter/s, 4.59482s/12 iters), loss = 0.579016
I0428 15:03:41.132254  7476 solver.cpp:237]     Train net output #0: loss = 0.579016 (* 1 = 0.579016 loss)
I0428 15:03:41.132266  7476 sgd_solver.cpp:105] Iteration 8376, lr = 0.00190297
I0428 15:03:46.635119  7476 solver.cpp:218] Iteration 8388 (2.18075 iter/s, 5.50269s/12 iters), loss = 0.461268
I0428 15:03:46.635164  7476 solver.cpp:237]     Train net output #0: loss = 0.461268 (* 1 = 0.461268 loss)
I0428 15:03:46.635172  7476 sgd_solver.cpp:105] Iteration 8388, lr = 0.00189846
I0428 15:03:49.631968  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:03:52.018501  7476 solver.cpp:218] Iteration 8400 (2.22917 iter/s, 5.38317s/12 iters), loss = 0.386372
I0428 15:03:52.018546  7476 solver.cpp:237]     Train net output #0: loss = 0.386372 (* 1 = 0.386372 loss)
I0428 15:03:52.018555  7476 sgd_solver.cpp:105] Iteration 8400, lr = 0.00189395
I0428 15:03:57.399189  7476 solver.cpp:218] Iteration 8412 (2.23029 iter/s, 5.38047s/12 iters), loss = 0.36208
I0428 15:03:57.399238  7476 solver.cpp:237]     Train net output #0: loss = 0.36208 (* 1 = 0.36208 loss)
I0428 15:03:57.399248  7476 sgd_solver.cpp:105] Iteration 8412, lr = 0.00188945
I0428 15:04:02.841078  7476 solver.cpp:218] Iteration 8424 (2.20521 iter/s, 5.44167s/12 iters), loss = 0.369408
I0428 15:04:02.841188  7476 solver.cpp:237]     Train net output #0: loss = 0.369408 (* 1 = 0.369408 loss)
I0428 15:04:02.841197  7476 sgd_solver.cpp:105] Iteration 8424, lr = 0.00188497
I0428 15:04:08.145310  7476 solver.cpp:218] Iteration 8436 (2.26246 iter/s, 5.30396s/12 iters), loss = 0.627104
I0428 15:04:08.145356  7476 solver.cpp:237]     Train net output #0: loss = 0.627104 (* 1 = 0.627104 loss)
I0428 15:04:08.145365  7476 sgd_solver.cpp:105] Iteration 8436, lr = 0.00188049
I0428 15:04:13.572384  7476 solver.cpp:218] Iteration 8448 (2.21123 iter/s, 5.42686s/12 iters), loss = 0.743652
I0428 15:04:13.572432  7476 solver.cpp:237]     Train net output #0: loss = 0.743652 (* 1 = 0.743652 loss)
I0428 15:04:13.572441  7476 sgd_solver.cpp:105] Iteration 8448, lr = 0.00187603
I0428 15:04:18.970021  7476 solver.cpp:218] Iteration 8460 (2.22329 iter/s, 5.39742s/12 iters), loss = 0.641842
I0428 15:04:18.970064  7476 solver.cpp:237]     Train net output #0: loss = 0.641842 (* 1 = 0.641842 loss)
I0428 15:04:18.970073  7476 sgd_solver.cpp:105] Iteration 8460, lr = 0.00187157
I0428 15:04:21.114598  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_8466.caffemodel
I0428 15:04:25.112205  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_8466.solverstate
I0428 15:04:31.843545  7476 solver.cpp:330] Iteration 8466, Testing net (#0)
I0428 15:04:31.843566  7476 net.cpp:676] Ignoring source layer train-data
I0428 15:04:32.976183  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:04:36.465867  7476 solver.cpp:397]     Test net output #0: accuracy = 0.42402
I0428 15:04:36.465898  7476 solver.cpp:397]     Test net output #1: loss = 3.20924 (* 1 = 3.20924 loss)
I0428 15:04:38.338357  7476 solver.cpp:218] Iteration 8472 (0.619587 iter/s, 19.3677s/12 iters), loss = 0.58584
I0428 15:04:38.338397  7476 solver.cpp:237]     Train net output #0: loss = 0.58584 (* 1 = 0.58584 loss)
I0428 15:04:38.338407  7476 sgd_solver.cpp:105] Iteration 8472, lr = 0.00186713
I0428 15:04:43.855546  7476 solver.cpp:218] Iteration 8484 (2.17511 iter/s, 5.51697s/12 iters), loss = 0.522646
I0428 15:04:43.855587  7476 solver.cpp:237]     Train net output #0: loss = 0.522646 (* 1 = 0.522646 loss)
I0428 15:04:43.855595  7476 sgd_solver.cpp:105] Iteration 8484, lr = 0.0018627
I0428 15:04:49.237592  7476 solver.cpp:218] Iteration 8496 (2.22972 iter/s, 5.38183s/12 iters), loss = 0.387334
I0428 15:04:49.237632  7476 solver.cpp:237]     Train net output #0: loss = 0.387334 (* 1 = 0.387334 loss)
I0428 15:04:49.237641  7476 sgd_solver.cpp:105] Iteration 8496, lr = 0.00185827
I0428 15:04:49.275216  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:04:54.664896  7476 solver.cpp:218] Iteration 8508 (2.21113 iter/s, 5.42709s/12 iters), loss = 0.490503
I0428 15:04:54.664942  7476 solver.cpp:237]     Train net output #0: loss = 0.490503 (* 1 = 0.490503 loss)
I0428 15:04:54.664950  7476 sgd_solver.cpp:105] Iteration 8508, lr = 0.00185386
I0428 15:05:00.119411  7476 solver.cpp:218] Iteration 8520 (2.2001 iter/s, 5.4543s/12 iters), loss = 0.397853
I0428 15:05:00.119452  7476 solver.cpp:237]     Train net output #0: loss = 0.397853 (* 1 = 0.397853 loss)
I0428 15:05:00.119460  7476 sgd_solver.cpp:105] Iteration 8520, lr = 0.00184946
I0428 15:05:05.492868  7476 solver.cpp:218] Iteration 8532 (2.23329 iter/s, 5.37324s/12 iters), loss = 0.614213
I0428 15:05:05.493005  7476 solver.cpp:237]     Train net output #0: loss = 0.614213 (* 1 = 0.614213 loss)
I0428 15:05:05.493018  7476 sgd_solver.cpp:105] Iteration 8532, lr = 0.00184507
I0428 15:05:10.890492  7476 solver.cpp:218] Iteration 8544 (2.22333 iter/s, 5.39732s/12 iters), loss = 0.711543
I0428 15:05:10.890550  7476 solver.cpp:237]     Train net output #0: loss = 0.711543 (* 1 = 0.711543 loss)
I0428 15:05:10.890563  7476 sgd_solver.cpp:105] Iteration 8544, lr = 0.00184069
I0428 15:05:16.278282  7476 solver.cpp:218] Iteration 8556 (2.22735 iter/s, 5.38756s/12 iters), loss = 0.496109
I0428 15:05:16.278335  7476 solver.cpp:237]     Train net output #0: loss = 0.496109 (* 1 = 0.496109 loss)
I0428 15:05:16.278347  7476 sgd_solver.cpp:105] Iteration 8556, lr = 0.00183632
I0428 15:05:21.081401  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_8568.caffemodel
I0428 15:05:22.833102  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_8568.solverstate
I0428 15:05:25.735080  7476 solver.cpp:330] Iteration 8568, Testing net (#0)
I0428 15:05:25.735110  7476 net.cpp:676] Ignoring source layer train-data
I0428 15:05:26.720806  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:05:30.123718  7476 solver.cpp:397]     Test net output #0: accuracy = 0.393995
I0428 15:05:30.123745  7476 solver.cpp:397]     Test net output #1: loss = 3.34436 (* 1 = 3.34436 loss)
I0428 15:05:30.250800  7476 solver.cpp:218] Iteration 8568 (0.858857 iter/s, 13.9721s/12 iters), loss = 0.522805
I0428 15:05:30.250859  7476 solver.cpp:237]     Train net output #0: loss = 0.522805 (* 1 = 0.522805 loss)
I0428 15:05:30.250870  7476 sgd_solver.cpp:105] Iteration 8568, lr = 0.00183196
I0428 15:05:34.645519  7476 solver.cpp:218] Iteration 8580 (2.73068 iter/s, 4.39452s/12 iters), loss = 0.566949
I0428 15:05:34.645561  7476 solver.cpp:237]     Train net output #0: loss = 0.566949 (* 1 = 0.566949 loss)
I0428 15:05:34.645570  7476 sgd_solver.cpp:105] Iteration 8580, lr = 0.00182761
I0428 15:05:40.077719  7476 solver.cpp:218] Iteration 8592 (2.20914 iter/s, 5.43199s/12 iters), loss = 0.554791
I0428 15:05:40.077859  7476 solver.cpp:237]     Train net output #0: loss = 0.554791 (* 1 = 0.554791 loss)
I0428 15:05:40.077869  7476 sgd_solver.cpp:105] Iteration 8592, lr = 0.00182327
I0428 15:05:42.429461  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:05:45.470772  7476 solver.cpp:218] Iteration 8604 (2.22521 iter/s, 5.39274s/12 iters), loss = 0.479958
I0428 15:05:45.470813  7476 solver.cpp:237]     Train net output #0: loss = 0.479958 (* 1 = 0.479958 loss)
I0428 15:05:45.470821  7476 sgd_solver.cpp:105] Iteration 8604, lr = 0.00181894
I0428 15:05:50.931867  7476 solver.cpp:218] Iteration 8616 (2.19745 iter/s, 5.46088s/12 iters), loss = 0.560066
I0428 15:05:50.931910  7476 solver.cpp:237]     Train net output #0: loss = 0.560066 (* 1 = 0.560066 loss)
I0428 15:05:50.931919  7476 sgd_solver.cpp:105] Iteration 8616, lr = 0.00181462
I0428 15:05:56.420734  7476 solver.cpp:218] Iteration 8628 (2.18633 iter/s, 5.48866s/12 iters), loss = 0.488887
I0428 15:05:56.420773  7476 solver.cpp:237]     Train net output #0: loss = 0.488887 (* 1 = 0.488887 loss)
I0428 15:05:56.420781  7476 sgd_solver.cpp:105] Iteration 8628, lr = 0.00181031
I0428 15:06:02.027067  7476 solver.cpp:218] Iteration 8640 (2.14052 iter/s, 5.60612s/12 iters), loss = 0.652664
I0428 15:06:02.027108  7476 solver.cpp:237]     Train net output #0: loss = 0.652664 (* 1 = 0.652664 loss)
I0428 15:06:02.027118  7476 sgd_solver.cpp:105] Iteration 8640, lr = 0.00180602
I0428 15:06:07.449584  7476 solver.cpp:218] Iteration 8652 (2.21308 iter/s, 5.4223s/12 iters), loss = 0.34884
I0428 15:06:07.449641  7476 solver.cpp:237]     Train net output #0: loss = 0.34884 (* 1 = 0.34884 loss)
I0428 15:06:07.449652  7476 sgd_solver.cpp:105] Iteration 8652, lr = 0.00180173
I0428 15:06:13.011982  7476 solver.cpp:218] Iteration 8664 (2.15743 iter/s, 5.56217s/12 iters), loss = 0.499829
I0428 15:06:13.012109  7476 solver.cpp:237]     Train net output #0: loss = 0.499829 (* 1 = 0.499829 loss)
I0428 15:06:13.012120  7476 sgd_solver.cpp:105] Iteration 8664, lr = 0.00179745
I0428 15:06:15.178083  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_8670.caffemodel
I0428 15:06:16.484916  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_8670.solverstate
I0428 15:06:17.511149  7476 solver.cpp:330] Iteration 8670, Testing net (#0)
I0428 15:06:17.511168  7476 net.cpp:676] Ignoring source layer train-data
I0428 15:06:18.467968  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:06:22.079270  7476 solver.cpp:397]     Test net output #0: accuracy = 0.407476
I0428 15:06:22.079308  7476 solver.cpp:397]     Test net output #1: loss = 3.29947 (* 1 = 3.29947 loss)
I0428 15:06:24.009673  7476 solver.cpp:218] Iteration 8676 (1.09118 iter/s, 10.9972s/12 iters), loss = 0.674196
I0428 15:06:24.009713  7476 solver.cpp:237]     Train net output #0: loss = 0.674196 (* 1 = 0.674196 loss)
I0428 15:06:24.009721  7476 sgd_solver.cpp:105] Iteration 8676, lr = 0.00179318
I0428 15:06:29.366802  7476 solver.cpp:218] Iteration 8688 (2.24009 iter/s, 5.35692s/12 iters), loss = 0.506422
I0428 15:06:29.366850  7476 solver.cpp:237]     Train net output #0: loss = 0.506422 (* 1 = 0.506422 loss)
I0428 15:06:29.366860  7476 sgd_solver.cpp:105] Iteration 8688, lr = 0.00178893
I0428 15:06:33.982904  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:06:34.740989  7476 solver.cpp:218] Iteration 8700 (2.23298 iter/s, 5.37397s/12 iters), loss = 0.505813
I0428 15:06:34.741030  7476 solver.cpp:237]     Train net output #0: loss = 0.505813 (* 1 = 0.505813 loss)
I0428 15:06:34.741041  7476 sgd_solver.cpp:105] Iteration 8700, lr = 0.00178468
I0428 15:06:40.011404  7476 solver.cpp:218] Iteration 8712 (2.27695 iter/s, 5.2702s/12 iters), loss = 0.38759
I0428 15:06:40.011446  7476 solver.cpp:237]     Train net output #0: loss = 0.38759 (* 1 = 0.38759 loss)
I0428 15:06:40.011456  7476 sgd_solver.cpp:105] Iteration 8712, lr = 0.00178044
I0428 15:06:45.357620  7476 solver.cpp:218] Iteration 8724 (2.24467 iter/s, 5.346s/12 iters), loss = 0.361828
I0428 15:06:45.358307  7476 solver.cpp:237]     Train net output #0: loss = 0.361828 (* 1 = 0.361828 loss)
I0428 15:06:45.358320  7476 sgd_solver.cpp:105] Iteration 8724, lr = 0.00177621
I0428 15:06:50.785269  7476 solver.cpp:218] Iteration 8736 (2.21125 iter/s, 5.4268s/12 iters), loss = 0.447482
I0428 15:06:50.785308  7476 solver.cpp:237]     Train net output #0: loss = 0.447482 (* 1 = 0.447482 loss)
I0428 15:06:50.785316  7476 sgd_solver.cpp:105] Iteration 8736, lr = 0.001772
I0428 15:06:56.182668  7476 solver.cpp:218] Iteration 8748 (2.22338 iter/s, 5.39719s/12 iters), loss = 0.330498
I0428 15:06:56.182708  7476 solver.cpp:237]     Train net output #0: loss = 0.330498 (* 1 = 0.330498 loss)
I0428 15:06:56.182716  7476 sgd_solver.cpp:105] Iteration 8748, lr = 0.00176779
I0428 15:07:01.462361  7476 solver.cpp:218] Iteration 8760 (2.27295 iter/s, 5.27948s/12 iters), loss = 0.385245
I0428 15:07:01.462404  7476 solver.cpp:237]     Train net output #0: loss = 0.385245 (* 1 = 0.385245 loss)
I0428 15:07:01.462414  7476 sgd_solver.cpp:105] Iteration 8760, lr = 0.00176359
I0428 15:07:06.359439  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_8772.caffemodel
I0428 15:07:07.706965  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_8772.solverstate
I0428 15:07:08.741757  7476 solver.cpp:330] Iteration 8772, Testing net (#0)
I0428 15:07:08.741775  7476 net.cpp:676] Ignoring source layer train-data
I0428 15:07:09.769414  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:07:13.291844  7476 solver.cpp:397]     Test net output #0: accuracy = 0.409926
I0428 15:07:13.291883  7476 solver.cpp:397]     Test net output #1: loss = 3.31295 (* 1 = 3.31295 loss)
I0428 15:07:13.419816  7476 solver.cpp:218] Iteration 8772 (1.00359 iter/s, 11.9571s/12 iters), loss = 0.520827
I0428 15:07:13.419867  7476 solver.cpp:237]     Train net output #0: loss = 0.520827 (* 1 = 0.520827 loss)
I0428 15:07:13.419878  7476 sgd_solver.cpp:105] Iteration 8772, lr = 0.00175941
I0428 15:07:17.859356  7476 solver.cpp:218] Iteration 8784 (2.7031 iter/s, 4.43935s/12 iters), loss = 0.388678
I0428 15:07:17.859475  7476 solver.cpp:237]     Train net output #0: loss = 0.388678 (* 1 = 0.388678 loss)
I0428 15:07:17.859484  7476 sgd_solver.cpp:105] Iteration 8784, lr = 0.00175523
I0428 15:07:23.282071  7476 solver.cpp:218] Iteration 8796 (2.21303 iter/s, 5.42243s/12 iters), loss = 0.296768
I0428 15:07:23.282111  7476 solver.cpp:237]     Train net output #0: loss = 0.296768 (* 1 = 0.296768 loss)
I0428 15:07:23.282120  7476 sgd_solver.cpp:105] Iteration 8796, lr = 0.00175106
I0428 15:07:24.762364  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:07:28.595275  7476 solver.cpp:218] Iteration 8808 (2.25862 iter/s, 5.31299s/12 iters), loss = 0.536527
I0428 15:07:28.595327  7476 solver.cpp:237]     Train net output #0: loss = 0.536527 (* 1 = 0.536527 loss)
I0428 15:07:28.595336  7476 sgd_solver.cpp:105] Iteration 8808, lr = 0.0017469
I0428 15:07:34.010649  7476 solver.cpp:218] Iteration 8820 (2.216 iter/s, 5.41515s/12 iters), loss = 0.443298
I0428 15:07:34.010697  7476 solver.cpp:237]     Train net output #0: loss = 0.443298 (* 1 = 0.443298 loss)
I0428 15:07:34.010706  7476 sgd_solver.cpp:105] Iteration 8820, lr = 0.00174276
I0428 15:07:39.541761  7476 solver.cpp:218] Iteration 8832 (2.16963 iter/s, 5.53089s/12 iters), loss = 0.441086
I0428 15:07:39.541805  7476 solver.cpp:237]     Train net output #0: loss = 0.441086 (* 1 = 0.441086 loss)
I0428 15:07:39.541815  7476 sgd_solver.cpp:105] Iteration 8832, lr = 0.00173862
I0428 15:07:44.889045  7476 solver.cpp:218] Iteration 8844 (2.24422 iter/s, 5.34707s/12 iters), loss = 0.402091
I0428 15:07:44.889096  7476 solver.cpp:237]     Train net output #0: loss = 0.402091 (* 1 = 0.402091 loss)
I0428 15:07:44.889108  7476 sgd_solver.cpp:105] Iteration 8844, lr = 0.00173449
I0428 15:07:50.148253  7476 solver.cpp:218] Iteration 8856 (2.28181 iter/s, 5.25899s/12 iters), loss = 0.307776
I0428 15:07:50.150552  7476 solver.cpp:237]     Train net output #0: loss = 0.307776 (* 1 = 0.307776 loss)
I0428 15:07:50.150564  7476 sgd_solver.cpp:105] Iteration 8856, lr = 0.00173037
I0428 15:07:55.764552  7476 solver.cpp:218] Iteration 8868 (2.13758 iter/s, 5.61383s/12 iters), loss = 0.485104
I0428 15:07:55.764595  7476 solver.cpp:237]     Train net output #0: loss = 0.485104 (* 1 = 0.485104 loss)
I0428 15:07:55.764602  7476 sgd_solver.cpp:105] Iteration 8868, lr = 0.00172626
I0428 15:07:57.925920  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_8874.caffemodel
I0428 15:07:59.281978  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_8874.solverstate
I0428 15:08:00.313450  7476 solver.cpp:330] Iteration 8874, Testing net (#0)
I0428 15:08:00.313469  7476 net.cpp:676] Ignoring source layer train-data
I0428 15:08:01.276716  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:08:04.864447  7476 solver.cpp:397]     Test net output #0: accuracy = 0.411765
I0428 15:08:04.864475  7476 solver.cpp:397]     Test net output #1: loss = 3.42293 (* 1 = 3.42293 loss)
I0428 15:08:06.850910  7476 solver.cpp:218] Iteration 8880 (1.08245 iter/s, 11.086s/12 iters), loss = 0.322793
I0428 15:08:06.850956  7476 solver.cpp:237]     Train net output #0: loss = 0.322793 (* 1 = 0.322793 loss)
I0428 15:08:06.850965  7476 sgd_solver.cpp:105] Iteration 8880, lr = 0.00172217
I0428 15:08:12.203902  7476 solver.cpp:218] Iteration 8892 (2.24183 iter/s, 5.35278s/12 iters), loss = 0.530404
I0428 15:08:12.203953  7476 solver.cpp:237]     Train net output #0: loss = 0.530404 (* 1 = 0.530404 loss)
I0428 15:08:12.203965  7476 sgd_solver.cpp:105] Iteration 8892, lr = 0.00171808
I0428 15:08:16.080551  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:08:17.709121  7476 solver.cpp:218] Iteration 8904 (2.17984 iter/s, 5.50499s/12 iters), loss = 0.403464
I0428 15:08:17.709170  7476 solver.cpp:237]     Train net output #0: loss = 0.403464 (* 1 = 0.403464 loss)
I0428 15:08:17.709180  7476 sgd_solver.cpp:105] Iteration 8904, lr = 0.001714
I0428 15:08:23.119573  7476 solver.cpp:218] Iteration 8916 (2.21802 iter/s, 5.41023s/12 iters), loss = 0.534558
I0428 15:08:23.119709  7476 solver.cpp:237]     Train net output #0: loss = 0.534558 (* 1 = 0.534558 loss)
I0428 15:08:23.119717  7476 sgd_solver.cpp:105] Iteration 8916, lr = 0.00170993
I0428 15:08:28.524852  7476 solver.cpp:218] Iteration 8928 (2.22017 iter/s, 5.40498s/12 iters), loss = 0.402597
I0428 15:08:28.524894  7476 solver.cpp:237]     Train net output #0: loss = 0.402597 (* 1 = 0.402597 loss)
I0428 15:08:28.524902  7476 sgd_solver.cpp:105] Iteration 8928, lr = 0.00170587
I0428 15:08:33.870532  7476 solver.cpp:218] Iteration 8940 (2.24489 iter/s, 5.34547s/12 iters), loss = 0.401579
I0428 15:08:33.870580  7476 solver.cpp:237]     Train net output #0: loss = 0.401579 (* 1 = 0.401579 loss)
I0428 15:08:33.870589  7476 sgd_solver.cpp:105] Iteration 8940, lr = 0.00170182
I0428 15:08:39.263505  7476 solver.cpp:218] Iteration 8952 (2.22521 iter/s, 5.39276s/12 iters), loss = 0.401737
I0428 15:08:39.263548  7476 solver.cpp:237]     Train net output #0: loss = 0.401737 (* 1 = 0.401737 loss)
I0428 15:08:39.263562  7476 sgd_solver.cpp:105] Iteration 8952, lr = 0.00169778
I0428 15:08:44.723556  7476 solver.cpp:218] Iteration 8964 (2.19787 iter/s, 5.45984s/12 iters), loss = 0.310152
I0428 15:08:44.723593  7476 solver.cpp:237]     Train net output #0: loss = 0.310152 (* 1 = 0.310152 loss)
I0428 15:08:44.723603  7476 sgd_solver.cpp:105] Iteration 8964, lr = 0.00169375
I0428 15:08:49.720985  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_8976.caffemodel
I0428 15:08:52.188421  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_8976.solverstate
I0428 15:08:53.595443  7476 solver.cpp:330] Iteration 8976, Testing net (#0)
I0428 15:08:53.595531  7476 net.cpp:676] Ignoring source layer train-data
I0428 15:08:54.429563  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:08:58.162384  7476 solver.cpp:397]     Test net output #0: accuracy = 0.403799
I0428 15:08:58.162415  7476 solver.cpp:397]     Test net output #1: loss = 3.45662 (* 1 = 3.45662 loss)
I0428 15:08:58.290163  7476 solver.cpp:218] Iteration 8976 (0.884554 iter/s, 13.5662s/12 iters), loss = 0.49513
I0428 15:08:58.290205  7476 solver.cpp:237]     Train net output #0: loss = 0.49513 (* 1 = 0.49513 loss)
I0428 15:08:58.290213  7476 sgd_solver.cpp:105] Iteration 8976, lr = 0.00168973
I0428 15:09:02.947793  7476 solver.cpp:218] Iteration 8988 (2.57653 iter/s, 4.65743s/12 iters), loss = 0.527524
I0428 15:09:02.947847  7476 solver.cpp:237]     Train net output #0: loss = 0.527524 (* 1 = 0.527524 loss)
I0428 15:09:02.947858  7476 sgd_solver.cpp:105] Iteration 8988, lr = 0.00168571
I0428 15:09:06.504386  7476 blocking_queue.cpp:49] Waiting for data
I0428 15:09:08.389288  7476 solver.cpp:218] Iteration 9000 (2.20537 iter/s, 5.44127s/12 iters), loss = 0.302545
I0428 15:09:08.389345  7476 solver.cpp:237]     Train net output #0: loss = 0.302545 (* 1 = 0.302545 loss)
I0428 15:09:08.389358  7476 sgd_solver.cpp:105] Iteration 9000, lr = 0.00168171
I0428 15:09:09.141935  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:09:13.833446  7476 solver.cpp:218] Iteration 9012 (2.20429 iter/s, 5.44393s/12 iters), loss = 0.452375
I0428 15:09:13.833492  7476 solver.cpp:237]     Train net output #0: loss = 0.452375 (* 1 = 0.452375 loss)
I0428 15:09:13.833499  7476 sgd_solver.cpp:105] Iteration 9012, lr = 0.00167772
I0428 15:09:19.254956  7476 solver.cpp:218] Iteration 9024 (2.2135 iter/s, 5.42129s/12 iters), loss = 0.31091
I0428 15:09:19.255007  7476 solver.cpp:237]     Train net output #0: loss = 0.31091 (* 1 = 0.31091 loss)
I0428 15:09:19.255017  7476 sgd_solver.cpp:105] Iteration 9024, lr = 0.00167374
I0428 15:09:24.651883  7476 solver.cpp:218] Iteration 9036 (2.22358 iter/s, 5.39671s/12 iters), loss = 0.250763
I0428 15:09:24.652021  7476 solver.cpp:237]     Train net output #0: loss = 0.250763 (* 1 = 0.250763 loss)
I0428 15:09:24.652031  7476 sgd_solver.cpp:105] Iteration 9036, lr = 0.00166976
I0428 15:09:30.050433  7476 solver.cpp:218] Iteration 9048 (2.22294 iter/s, 5.39825s/12 iters), loss = 0.519322
I0428 15:09:30.050482  7476 solver.cpp:237]     Train net output #0: loss = 0.519322 (* 1 = 0.519322 loss)
I0428 15:09:30.050493  7476 sgd_solver.cpp:105] Iteration 9048, lr = 0.0016658
I0428 15:09:35.385942  7476 solver.cpp:218] Iteration 9060 (2.24917 iter/s, 5.3353s/12 iters), loss = 0.277082
I0428 15:09:35.385984  7476 solver.cpp:237]     Train net output #0: loss = 0.277082 (* 1 = 0.277082 loss)
I0428 15:09:35.385993  7476 sgd_solver.cpp:105] Iteration 9060, lr = 0.00166184
I0428 15:09:40.785935  7476 solver.cpp:218] Iteration 9072 (2.22231 iter/s, 5.39979s/12 iters), loss = 0.328602
I0428 15:09:40.785980  7476 solver.cpp:237]     Train net output #0: loss = 0.328602 (* 1 = 0.328602 loss)
I0428 15:09:40.785989  7476 sgd_solver.cpp:105] Iteration 9072, lr = 0.0016579
I0428 15:09:43.004050  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_9078.caffemodel
I0428 15:09:45.256678  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_9078.solverstate
I0428 15:09:52.693331  7476 solver.cpp:330] Iteration 9078, Testing net (#0)
I0428 15:09:52.693353  7476 net.cpp:676] Ignoring source layer train-data
I0428 15:09:53.536206  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:09:57.125517  7476 solver.cpp:397]     Test net output #0: accuracy = 0.415441
I0428 15:09:57.125602  7476 solver.cpp:397]     Test net output #1: loss = 3.51521 (* 1 = 3.51521 loss)
I0428 15:09:59.107522  7476 solver.cpp:218] Iteration 9084 (0.654986 iter/s, 18.321s/12 iters), loss = 0.421501
I0428 15:09:59.107566  7476 solver.cpp:237]     Train net output #0: loss = 0.421501 (* 1 = 0.421501 loss)
I0428 15:09:59.107575  7476 sgd_solver.cpp:105] Iteration 9084, lr = 0.00165396
I0428 15:10:04.433013  7476 solver.cpp:218] Iteration 9096 (2.2534 iter/s, 5.32528s/12 iters), loss = 0.370428
I0428 15:10:04.433056  7476 solver.cpp:237]     Train net output #0: loss = 0.370428 (* 1 = 0.370428 loss)
I0428 15:10:04.433064  7476 sgd_solver.cpp:105] Iteration 9096, lr = 0.00165003
I0428 15:10:07.665123  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:10:09.939972  7476 solver.cpp:218] Iteration 9108 (2.17915 iter/s, 5.50674s/12 iters), loss = 0.335979
I0428 15:10:09.940016  7476 solver.cpp:237]     Train net output #0: loss = 0.335979 (* 1 = 0.335979 loss)
I0428 15:10:09.940024  7476 sgd_solver.cpp:105] Iteration 9108, lr = 0.00164612
I0428 15:10:15.365334  7476 solver.cpp:218] Iteration 9120 (2.21192 iter/s, 5.42515s/12 iters), loss = 0.246229
I0428 15:10:15.365372  7476 solver.cpp:237]     Train net output #0: loss = 0.246229 (* 1 = 0.246229 loss)
I0428 15:10:15.365381  7476 sgd_solver.cpp:105] Iteration 9120, lr = 0.00164221
I0428 15:10:20.847975  7476 solver.cpp:218] Iteration 9132 (2.18881 iter/s, 5.48243s/12 iters), loss = 0.401534
I0428 15:10:20.848019  7476 solver.cpp:237]     Train net output #0: loss = 0.401534 (* 1 = 0.401534 loss)
I0428 15:10:20.848027  7476 sgd_solver.cpp:105] Iteration 9132, lr = 0.00163831
I0428 15:10:26.127624  7476 solver.cpp:218] Iteration 9144 (2.27297 iter/s, 5.27944s/12 iters), loss = 0.441634
I0428 15:10:26.127671  7476 solver.cpp:237]     Train net output #0: loss = 0.441634 (* 1 = 0.441634 loss)
I0428 15:10:26.127681  7476 sgd_solver.cpp:105] Iteration 9144, lr = 0.00163442
I0428 15:10:31.673843  7476 solver.cpp:218] Iteration 9156 (2.16372 iter/s, 5.54599s/12 iters), loss = 0.536432
I0428 15:10:31.675194  7476 solver.cpp:237]     Train net output #0: loss = 0.536432 (* 1 = 0.536432 loss)
I0428 15:10:31.675204  7476 sgd_solver.cpp:105] Iteration 9156, lr = 0.00163054
I0428 15:10:37.081919  7476 solver.cpp:218] Iteration 9168 (2.21953 iter/s, 5.40656s/12 iters), loss = 0.223614
I0428 15:10:37.081962  7476 solver.cpp:237]     Train net output #0: loss = 0.223614 (* 1 = 0.223614 loss)
I0428 15:10:37.081971  7476 sgd_solver.cpp:105] Iteration 9168, lr = 0.00162667
I0428 15:10:42.004236  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_9180.caffemodel
I0428 15:10:44.983492  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_9180.solverstate
I0428 15:10:47.273994  7476 solver.cpp:330] Iteration 9180, Testing net (#0)
I0428 15:10:47.274014  7476 net.cpp:676] Ignoring source layer train-data
I0428 15:10:48.039170  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:10:51.850154  7476 solver.cpp:397]     Test net output #0: accuracy = 0.421569
I0428 15:10:51.850183  7476 solver.cpp:397]     Test net output #1: loss = 3.39418 (* 1 = 3.39418 loss)
I0428 15:10:51.977972  7476 solver.cpp:218] Iteration 9180 (0.805608 iter/s, 14.8956s/12 iters), loss = 0.45781
I0428 15:10:51.978010  7476 solver.cpp:237]     Train net output #0: loss = 0.45781 (* 1 = 0.45781 loss)
I0428 15:10:51.978019  7476 sgd_solver.cpp:105] Iteration 9180, lr = 0.00162281
I0428 15:10:56.550683  7476 solver.cpp:218] Iteration 9192 (2.62437 iter/s, 4.57252s/12 iters), loss = 0.355423
I0428 15:10:56.550729  7476 solver.cpp:237]     Train net output #0: loss = 0.355423 (* 1 = 0.355423 loss)
I0428 15:10:56.550737  7476 sgd_solver.cpp:105] Iteration 9192, lr = 0.00161895
I0428 15:11:01.966184  7476 solver.cpp:218] Iteration 9204 (2.21595 iter/s, 5.41528s/12 iters), loss = 0.577303
I0428 15:11:01.966323  7476 solver.cpp:237]     Train net output #0: loss = 0.577303 (* 1 = 0.577303 loss)
I0428 15:11:01.966336  7476 sgd_solver.cpp:105] Iteration 9204, lr = 0.00161511
I0428 15:11:02.035233  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:11:07.437681  7476 solver.cpp:218] Iteration 9216 (2.19331 iter/s, 5.47119s/12 iters), loss = 0.357985
I0428 15:11:07.437726  7476 solver.cpp:237]     Train net output #0: loss = 0.357985 (* 1 = 0.357985 loss)
I0428 15:11:07.437736  7476 sgd_solver.cpp:105] Iteration 9216, lr = 0.00161128
I0428 15:11:12.862025  7476 solver.cpp:218] Iteration 9228 (2.21234 iter/s, 5.42413s/12 iters), loss = 0.358457
I0428 15:11:12.862071  7476 solver.cpp:237]     Train net output #0: loss = 0.358457 (* 1 = 0.358457 loss)
I0428 15:11:12.862080  7476 sgd_solver.cpp:105] Iteration 9228, lr = 0.00160745
I0428 15:11:18.274844  7476 solver.cpp:218] Iteration 9240 (2.21705 iter/s, 5.41261s/12 iters), loss = 0.443188
I0428 15:11:18.274889  7476 solver.cpp:237]     Train net output #0: loss = 0.443188 (* 1 = 0.443188 loss)
I0428 15:11:18.274897  7476 sgd_solver.cpp:105] Iteration 9240, lr = 0.00160363
I0428 15:11:23.763679  7476 solver.cpp:218] Iteration 9252 (2.18634 iter/s, 5.48862s/12 iters), loss = 0.51934
I0428 15:11:23.763726  7476 solver.cpp:237]     Train net output #0: loss = 0.51934 (* 1 = 0.51934 loss)
I0428 15:11:23.763736  7476 sgd_solver.cpp:105] Iteration 9252, lr = 0.00159983
I0428 15:11:29.189433  7476 solver.cpp:218] Iteration 9264 (2.21176 iter/s, 5.42553s/12 iters), loss = 0.334901
I0428 15:11:29.189477  7476 solver.cpp:237]     Train net output #0: loss = 0.334901 (* 1 = 0.334901 loss)
I0428 15:11:29.189486  7476 sgd_solver.cpp:105] Iteration 9264, lr = 0.00159603
I0428 15:11:34.576635  7476 solver.cpp:218] Iteration 9276 (2.22759 iter/s, 5.38699s/12 iters), loss = 0.370314
I0428 15:11:34.577020  7476 solver.cpp:237]     Train net output #0: loss = 0.370314 (* 1 = 0.370314 loss)
I0428 15:11:34.577030  7476 sgd_solver.cpp:105] Iteration 9276, lr = 0.00159224
I0428 15:11:36.790747  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_9282.caffemodel
I0428 15:11:40.175710  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_9282.solverstate
I0428 15:11:41.233872  7476 solver.cpp:330] Iteration 9282, Testing net (#0)
I0428 15:11:41.233899  7476 net.cpp:676] Ignoring source layer train-data
I0428 15:11:41.962405  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:11:45.665132  7476 solver.cpp:397]     Test net output #0: accuracy = 0.428309
I0428 15:11:45.665160  7476 solver.cpp:397]     Test net output #1: loss = 3.40102 (* 1 = 3.40102 loss)
I0428 15:11:47.604197  7476 solver.cpp:218] Iteration 9288 (0.921178 iter/s, 13.0268s/12 iters), loss = 0.317838
I0428 15:11:47.604241  7476 solver.cpp:237]     Train net output #0: loss = 0.317838 (* 1 = 0.317838 loss)
I0428 15:11:47.604249  7476 sgd_solver.cpp:105] Iteration 9288, lr = 0.00158846
I0428 15:11:52.892987  7476 solver.cpp:218] Iteration 9300 (2.26904 iter/s, 5.28858s/12 iters), loss = 0.495993
I0428 15:11:52.893030  7476 solver.cpp:237]     Train net output #0: loss = 0.495993 (* 1 = 0.495993 loss)
I0428 15:11:52.893039  7476 sgd_solver.cpp:105] Iteration 9300, lr = 0.00158469
I0428 15:11:55.281883  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:11:58.416640  7476 solver.cpp:218] Iteration 9312 (2.17256 iter/s, 5.52344s/12 iters), loss = 0.456576
I0428 15:11:58.416687  7476 solver.cpp:237]     Train net output #0: loss = 0.456576 (* 1 = 0.456576 loss)
I0428 15:11:58.416695  7476 sgd_solver.cpp:105] Iteration 9312, lr = 0.00158092
I0428 15:12:03.786514  7476 solver.cpp:218] Iteration 9324 (2.23478 iter/s, 5.36966s/12 iters), loss = 0.414565
I0428 15:12:03.786561  7476 solver.cpp:237]     Train net output #0: loss = 0.414565 (* 1 = 0.414565 loss)
I0428 15:12:03.786571  7476 sgd_solver.cpp:105] Iteration 9324, lr = 0.00157717
I0428 15:12:09.339957  7476 solver.cpp:218] Iteration 9336 (2.16091 iter/s, 5.55322s/12 iters), loss = 0.506449
I0428 15:12:09.340068  7476 solver.cpp:237]     Train net output #0: loss = 0.506449 (* 1 = 0.506449 loss)
I0428 15:12:09.340077  7476 sgd_solver.cpp:105] Iteration 9336, lr = 0.00157343
I0428 15:12:14.647800  7476 solver.cpp:218] Iteration 9348 (2.26092 iter/s, 5.30756s/12 iters), loss = 0.396756
I0428 15:12:14.647843  7476 solver.cpp:237]     Train net output #0: loss = 0.396756 (* 1 = 0.396756 loss)
I0428 15:12:14.647851  7476 sgd_solver.cpp:105] Iteration 9348, lr = 0.00156969
I0428 15:12:20.087867  7476 solver.cpp:218] Iteration 9360 (2.20594 iter/s, 5.43986s/12 iters), loss = 0.572469
I0428 15:12:20.087905  7476 solver.cpp:237]     Train net output #0: loss = 0.572469 (* 1 = 0.572469 loss)
I0428 15:12:20.087914  7476 sgd_solver.cpp:105] Iteration 9360, lr = 0.00156596
I0428 15:12:25.366724  7476 solver.cpp:218] Iteration 9372 (2.2733 iter/s, 5.27866s/12 iters), loss = 0.430608
I0428 15:12:25.366760  7476 solver.cpp:237]     Train net output #0: loss = 0.430608 (* 1 = 0.430608 loss)
I0428 15:12:25.366768  7476 sgd_solver.cpp:105] Iteration 9372, lr = 0.00156225
I0428 15:12:30.257107  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_9384.caffemodel
I0428 15:12:33.110787  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_9384.solverstate
I0428 15:12:35.720742  7476 solver.cpp:330] Iteration 9384, Testing net (#0)
I0428 15:12:35.720760  7476 net.cpp:676] Ignoring source layer train-data
I0428 15:12:36.482915  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:12:40.448099  7476 solver.cpp:397]     Test net output #0: accuracy = 0.420343
I0428 15:12:40.452203  7476 solver.cpp:397]     Test net output #1: loss = 3.4068 (* 1 = 3.4068 loss)
I0428 15:12:40.578115  7476 solver.cpp:218] Iteration 9384 (0.788907 iter/s, 15.2109s/12 iters), loss = 0.353733
I0428 15:12:40.578167  7476 solver.cpp:237]     Train net output #0: loss = 0.353733 (* 1 = 0.353733 loss)
I0428 15:12:40.578178  7476 sgd_solver.cpp:105] Iteration 9384, lr = 0.00155854
I0428 15:12:45.192515  7476 solver.cpp:218] Iteration 9396 (2.60069 iter/s, 4.61417s/12 iters), loss = 0.409633
I0428 15:12:45.192574  7476 solver.cpp:237]     Train net output #0: loss = 0.409633 (* 1 = 0.409633 loss)
I0428 15:12:45.192586  7476 sgd_solver.cpp:105] Iteration 9396, lr = 0.00155484
I0428 15:12:49.851267  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:12:50.571842  7476 solver.cpp:218] Iteration 9408 (2.23086 iter/s, 5.3791s/12 iters), loss = 0.301186
I0428 15:12:50.571893  7476 solver.cpp:237]     Train net output #0: loss = 0.301186 (* 1 = 0.301186 loss)
I0428 15:12:50.571907  7476 sgd_solver.cpp:105] Iteration 9408, lr = 0.00155114
I0428 15:12:55.935036  7476 solver.cpp:218] Iteration 9420 (2.23756 iter/s, 5.36298s/12 iters), loss = 0.232371
I0428 15:12:55.935099  7476 solver.cpp:237]     Train net output #0: loss = 0.232371 (* 1 = 0.232371 loss)
I0428 15:12:55.935112  7476 sgd_solver.cpp:105] Iteration 9420, lr = 0.00154746
I0428 15:13:01.318799  7476 solver.cpp:218] Iteration 9432 (2.22902 iter/s, 5.38353s/12 iters), loss = 0.22226
I0428 15:13:01.318846  7476 solver.cpp:237]     Train net output #0: loss = 0.22226 (* 1 = 0.22226 loss)
I0428 15:13:01.318858  7476 sgd_solver.cpp:105] Iteration 9432, lr = 0.00154379
I0428 15:13:06.715581  7476 solver.cpp:218] Iteration 9444 (2.22363 iter/s, 5.39657s/12 iters), loss = 0.422786
I0428 15:13:06.715622  7476 solver.cpp:237]     Train net output #0: loss = 0.422786 (* 1 = 0.422786 loss)
I0428 15:13:06.715631  7476 sgd_solver.cpp:105] Iteration 9444, lr = 0.00154012
I0428 15:13:11.902825  7476 solver.cpp:218] Iteration 9456 (2.31346 iter/s, 5.18704s/12 iters), loss = 0.360143
I0428 15:13:11.902940  7476 solver.cpp:237]     Train net output #0: loss = 0.360143 (* 1 = 0.360143 loss)
I0428 15:13:11.902951  7476 sgd_solver.cpp:105] Iteration 9456, lr = 0.00153647
I0428 15:13:17.172888  7476 solver.cpp:218] Iteration 9468 (2.27713 iter/s, 5.26978s/12 iters), loss = 0.319685
I0428 15:13:17.172928  7476 solver.cpp:237]     Train net output #0: loss = 0.319685 (* 1 = 0.319685 loss)
I0428 15:13:17.172937  7476 sgd_solver.cpp:105] Iteration 9468, lr = 0.00153282
I0428 15:13:22.626114  7476 solver.cpp:218] Iteration 9480 (2.20062 iter/s, 5.45302s/12 iters), loss = 0.276009
I0428 15:13:22.626155  7476 solver.cpp:237]     Train net output #0: loss = 0.276009 (* 1 = 0.276009 loss)
I0428 15:13:22.626163  7476 sgd_solver.cpp:105] Iteration 9480, lr = 0.00152918
I0428 15:13:24.786947  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_9486.caffemodel
I0428 15:13:30.872540  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_9486.solverstate
I0428 15:13:33.468080  7476 solver.cpp:330] Iteration 9486, Testing net (#0)
I0428 15:13:33.468103  7476 net.cpp:676] Ignoring source layer train-data
I0428 15:13:34.130839  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:13:37.886031  7476 solver.cpp:397]     Test net output #0: accuracy = 0.417892
I0428 15:13:37.886070  7476 solver.cpp:397]     Test net output #1: loss = 3.5795 (* 1 = 3.5795 loss)
I0428 15:13:39.934751  7476 solver.cpp:218] Iteration 9492 (0.693318 iter/s, 17.3081s/12 iters), loss = 0.223379
I0428 15:13:39.934800  7476 solver.cpp:237]     Train net output #0: loss = 0.223379 (* 1 = 0.223379 loss)
I0428 15:13:39.934808  7476 sgd_solver.cpp:105] Iteration 9492, lr = 0.00152555
I0428 15:13:45.445390  7476 solver.cpp:218] Iteration 9504 (2.17769 iter/s, 5.51042s/12 iters), loss = 0.293012
I0428 15:13:45.445535  7476 solver.cpp:237]     Train net output #0: loss = 0.293012 (* 1 = 0.293012 loss)
I0428 15:13:45.445545  7476 sgd_solver.cpp:105] Iteration 9504, lr = 0.00152193
I0428 15:13:47.035588  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:13:50.859483  7476 solver.cpp:218] Iteration 9516 (2.21657 iter/s, 5.41378s/12 iters), loss = 0.222076
I0428 15:13:50.859524  7476 solver.cpp:237]     Train net output #0: loss = 0.222076 (* 1 = 0.222076 loss)
I0428 15:13:50.859532  7476 sgd_solver.cpp:105] Iteration 9516, lr = 0.00151831
I0428 15:13:56.132912  7476 solver.cpp:218] Iteration 9528 (2.27565 iter/s, 5.27323s/12 iters), loss = 0.416639
I0428 15:13:56.132961  7476 solver.cpp:237]     Train net output #0: loss = 0.416639 (* 1 = 0.416639 loss)
I0428 15:13:56.132970  7476 sgd_solver.cpp:105] Iteration 9528, lr = 0.00151471
I0428 15:14:01.537005  7476 solver.cpp:218] Iteration 9540 (2.22063 iter/s, 5.40387s/12 iters), loss = 0.457419
I0428 15:14:01.537052  7476 solver.cpp:237]     Train net output #0: loss = 0.457419 (* 1 = 0.457419 loss)
I0428 15:14:01.537062  7476 sgd_solver.cpp:105] Iteration 9540, lr = 0.00151111
I0428 15:14:06.852566  7476 solver.cpp:218] Iteration 9552 (2.25762 iter/s, 5.31534s/12 iters), loss = 0.218793
I0428 15:14:06.852607  7476 solver.cpp:237]     Train net output #0: loss = 0.218793 (* 1 = 0.218793 loss)
I0428 15:14:06.852617  7476 sgd_solver.cpp:105] Iteration 9552, lr = 0.00150752
I0428 15:14:12.372417  7476 solver.cpp:218] Iteration 9564 (2.17406 iter/s, 5.51964s/12 iters), loss = 0.3204
I0428 15:14:12.372460  7476 solver.cpp:237]     Train net output #0: loss = 0.3204 (* 1 = 0.3204 loss)
I0428 15:14:12.372470  7476 sgd_solver.cpp:105] Iteration 9564, lr = 0.00150395
I0428 15:14:17.737123  7476 solver.cpp:218] Iteration 9576 (2.23693 iter/s, 5.3645s/12 iters), loss = 0.333437
I0428 15:14:17.737315  7476 solver.cpp:237]     Train net output #0: loss = 0.333437 (* 1 = 0.333437 loss)
I0428 15:14:17.737325  7476 sgd_solver.cpp:105] Iteration 9576, lr = 0.00150037
I0428 15:14:22.468921  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_9588.caffemodel
I0428 15:14:25.685660  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_9588.solverstate
I0428 15:14:27.229105  7476 solver.cpp:330] Iteration 9588, Testing net (#0)
I0428 15:14:27.229123  7476 net.cpp:676] Ignoring source layer train-data
I0428 15:14:27.821745  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:14:31.624656  7476 solver.cpp:397]     Test net output #0: accuracy = 0.427696
I0428 15:14:31.624686  7476 solver.cpp:397]     Test net output #1: loss = 3.41723 (* 1 = 3.41723 loss)
I0428 15:14:31.752895  7476 solver.cpp:218] Iteration 9588 (0.856215 iter/s, 14.0152s/12 iters), loss = 0.295922
I0428 15:14:31.752941  7476 solver.cpp:237]     Train net output #0: loss = 0.295922 (* 1 = 0.295922 loss)
I0428 15:14:31.752951  7476 sgd_solver.cpp:105] Iteration 9588, lr = 0.00149681
I0428 15:14:36.245658  7476 solver.cpp:218] Iteration 9600 (2.67108 iter/s, 4.49257s/12 iters), loss = 0.200638
I0428 15:14:36.245708  7476 solver.cpp:237]     Train net output #0: loss = 0.200638 (* 1 = 0.200638 loss)
I0428 15:14:36.245719  7476 sgd_solver.cpp:105] Iteration 9600, lr = 0.00149326
I0428 15:14:40.113194  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:14:41.627256  7476 solver.cpp:218] Iteration 9612 (2.22991 iter/s, 5.38138s/12 iters), loss = 0.375512
I0428 15:14:41.627300  7476 solver.cpp:237]     Train net output #0: loss = 0.375512 (* 1 = 0.375512 loss)
I0428 15:14:41.627310  7476 sgd_solver.cpp:105] Iteration 9612, lr = 0.00148971
I0428 15:14:47.064231  7476 solver.cpp:218] Iteration 9624 (2.2072 iter/s, 5.43676s/12 iters), loss = 0.475519
I0428 15:14:47.064275  7476 solver.cpp:237]     Train net output #0: loss = 0.475519 (* 1 = 0.475519 loss)
I0428 15:14:47.064283  7476 sgd_solver.cpp:105] Iteration 9624, lr = 0.00148618
I0428 15:14:52.435899  7476 solver.cpp:218] Iteration 9636 (2.23403 iter/s, 5.37146s/12 iters), loss = 0.209016
I0428 15:14:52.436041  7476 solver.cpp:237]     Train net output #0: loss = 0.209016 (* 1 = 0.209016 loss)
I0428 15:14:52.436051  7476 sgd_solver.cpp:105] Iteration 9636, lr = 0.00148265
I0428 15:14:57.765708  7476 solver.cpp:218] Iteration 9648 (2.25162 iter/s, 5.3295s/12 iters), loss = 0.224356
I0428 15:14:57.765769  7476 solver.cpp:237]     Train net output #0: loss = 0.224356 (* 1 = 0.224356 loss)
I0428 15:14:57.765781  7476 sgd_solver.cpp:105] Iteration 9648, lr = 0.00147913
I0428 15:15:03.112535  7476 solver.cpp:218] Iteration 9660 (2.24443 iter/s, 5.34657s/12 iters), loss = 0.264834
I0428 15:15:03.112581  7476 solver.cpp:237]     Train net output #0: loss = 0.264834 (* 1 = 0.264834 loss)
I0428 15:15:03.112591  7476 sgd_solver.cpp:105] Iteration 9660, lr = 0.00147562
I0428 15:15:08.577148  7476 solver.cpp:218] Iteration 9672 (2.19603 iter/s, 5.4644s/12 iters), loss = 0.269942
I0428 15:15:08.577195  7476 solver.cpp:237]     Train net output #0: loss = 0.269942 (* 1 = 0.269942 loss)
I0428 15:15:08.577203  7476 sgd_solver.cpp:105] Iteration 9672, lr = 0.00147211
I0428 15:15:13.991834  7476 solver.cpp:218] Iteration 9684 (2.21629 iter/s, 5.41447s/12 iters), loss = 0.334847
I0428 15:15:13.991892  7476 solver.cpp:237]     Train net output #0: loss = 0.334847 (* 1 = 0.334847 loss)
I0428 15:15:13.991904  7476 sgd_solver.cpp:105] Iteration 9684, lr = 0.00146862
I0428 15:15:16.133509  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_9690.caffemodel
I0428 15:15:20.468031  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_9690.solverstate
I0428 15:15:22.977341  7476 solver.cpp:330] Iteration 9690, Testing net (#0)
I0428 15:15:22.977452  7476 net.cpp:676] Ignoring source layer train-data
I0428 15:15:23.583950  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:15:26.671028  7476 blocking_queue.cpp:49] Waiting for data
I0428 15:15:27.799582  7476 solver.cpp:397]     Test net output #0: accuracy = 0.422181
I0428 15:15:27.799612  7476 solver.cpp:397]     Test net output #1: loss = 3.50786 (* 1 = 3.50786 loss)
I0428 15:15:29.572017  7476 solver.cpp:218] Iteration 9696 (0.770235 iter/s, 15.5797s/12 iters), loss = 0.227446
I0428 15:15:29.572067  7476 solver.cpp:237]     Train net output #0: loss = 0.227446 (* 1 = 0.227446 loss)
I0428 15:15:29.572082  7476 sgd_solver.cpp:105] Iteration 9696, lr = 0.00146513
I0428 15:15:35.065434  7476 solver.cpp:218] Iteration 9708 (2.18452 iter/s, 5.4932s/12 iters), loss = 0.277942
I0428 15:15:35.065479  7476 solver.cpp:237]     Train net output #0: loss = 0.277942 (* 1 = 0.277942 loss)
I0428 15:15:35.065486  7476 sgd_solver.cpp:105] Iteration 9708, lr = 0.00146165
I0428 15:15:35.914561  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:15:40.624704  7476 solver.cpp:218] Iteration 9720 (2.15864 iter/s, 5.55905s/12 iters), loss = 0.272576
I0428 15:15:40.624742  7476 solver.cpp:237]     Train net output #0: loss = 0.272576 (* 1 = 0.272576 loss)
I0428 15:15:40.624752  7476 sgd_solver.cpp:105] Iteration 9720, lr = 0.00145818
I0428 15:15:46.010265  7476 solver.cpp:218] Iteration 9732 (2.22827 iter/s, 5.38535s/12 iters), loss = 0.342114
I0428 15:15:46.010306  7476 solver.cpp:237]     Train net output #0: loss = 0.342114 (* 1 = 0.342114 loss)
I0428 15:15:46.010315  7476 sgd_solver.cpp:105] Iteration 9732, lr = 0.00145472
I0428 15:15:51.384552  7476 solver.cpp:218] Iteration 9744 (2.23294 iter/s, 5.37408s/12 iters), loss = 0.215374
I0428 15:15:51.384600  7476 solver.cpp:237]     Train net output #0: loss = 0.215374 (* 1 = 0.215374 loss)
I0428 15:15:51.384609  7476 sgd_solver.cpp:105] Iteration 9744, lr = 0.00145127
I0428 15:15:56.863080  7476 solver.cpp:218] Iteration 9756 (2.19046 iter/s, 5.47831s/12 iters), loss = 0.291063
I0428 15:15:56.869720  7476 solver.cpp:237]     Train net output #0: loss = 0.291063 (* 1 = 0.291063 loss)
I0428 15:15:56.869730  7476 sgd_solver.cpp:105] Iteration 9756, lr = 0.00144782
I0428 15:16:02.316905  7476 solver.cpp:218] Iteration 9768 (2.20304 iter/s, 5.44702s/12 iters), loss = 0.274393
I0428 15:16:02.316946  7476 solver.cpp:237]     Train net output #0: loss = 0.274393 (* 1 = 0.274393 loss)
I0428 15:16:02.316954  7476 sgd_solver.cpp:105] Iteration 9768, lr = 0.00144438
I0428 15:16:07.767606  7476 solver.cpp:218] Iteration 9780 (2.20164 iter/s, 5.45049s/12 iters), loss = 0.27707
I0428 15:16:07.767654  7476 solver.cpp:237]     Train net output #0: loss = 0.27707 (* 1 = 0.27707 loss)
I0428 15:16:07.767663  7476 sgd_solver.cpp:105] Iteration 9780, lr = 0.00144095
I0428 15:16:12.689340  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_9792.caffemodel
I0428 15:16:19.471357  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_9792.solverstate
I0428 15:16:23.144014  7476 solver.cpp:330] Iteration 9792, Testing net (#0)
I0428 15:16:23.144032  7476 net.cpp:676] Ignoring source layer train-data
I0428 15:16:23.659677  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:16:27.531107  7476 solver.cpp:397]     Test net output #0: accuracy = 0.429534
I0428 15:16:27.531199  7476 solver.cpp:397]     Test net output #1: loss = 3.61635 (* 1 = 3.61635 loss)
I0428 15:16:27.659044  7476 solver.cpp:218] Iteration 9792 (0.603293 iter/s, 19.8908s/12 iters), loss = 0.25295
I0428 15:16:27.659094  7476 solver.cpp:237]     Train net output #0: loss = 0.25295 (* 1 = 0.25295 loss)
I0428 15:16:27.659107  7476 sgd_solver.cpp:105] Iteration 9792, lr = 0.00143753
I0428 15:16:32.137246  7476 solver.cpp:218] Iteration 9804 (2.67976 iter/s, 4.47801s/12 iters), loss = 0.198628
I0428 15:16:32.137293  7476 solver.cpp:237]     Train net output #0: loss = 0.198628 (* 1 = 0.198628 loss)
I0428 15:16:32.137303  7476 sgd_solver.cpp:105] Iteration 9804, lr = 0.00143412
I0428 15:16:35.333222  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:16:37.545089  7476 solver.cpp:218] Iteration 9816 (2.21909 iter/s, 5.40763s/12 iters), loss = 0.254103
I0428 15:16:37.545131  7476 solver.cpp:237]     Train net output #0: loss = 0.254103 (* 1 = 0.254103 loss)
I0428 15:16:37.545140  7476 sgd_solver.cpp:105] Iteration 9816, lr = 0.00143072
I0428 15:16:42.928555  7476 solver.cpp:218] Iteration 9828 (2.22913 iter/s, 5.38326s/12 iters), loss = 0.509056
I0428 15:16:42.928597  7476 solver.cpp:237]     Train net output #0: loss = 0.509056 (* 1 = 0.509056 loss)
I0428 15:16:42.928606  7476 sgd_solver.cpp:105] Iteration 9828, lr = 0.00142732
I0428 15:16:48.222693  7476 solver.cpp:218] Iteration 9840 (2.26675 iter/s, 5.29393s/12 iters), loss = 0.257153
I0428 15:16:48.222743  7476 solver.cpp:237]     Train net output #0: loss = 0.257153 (* 1 = 0.257153 loss)
I0428 15:16:48.222754  7476 sgd_solver.cpp:105] Iteration 9840, lr = 0.00142393
I0428 15:16:53.505755  7476 solver.cpp:218] Iteration 9852 (2.2715 iter/s, 5.28285s/12 iters), loss = 0.244088
I0428 15:16:53.505800  7476 solver.cpp:237]     Train net output #0: loss = 0.244088 (* 1 = 0.244088 loss)
I0428 15:16:53.505808  7476 sgd_solver.cpp:105] Iteration 9852, lr = 0.00142055
I0428 15:16:58.649520  7476 solver.cpp:218] Iteration 9864 (2.33301 iter/s, 5.14356s/12 iters), loss = 0.282765
I0428 15:16:58.649613  7476 solver.cpp:237]     Train net output #0: loss = 0.282765 (* 1 = 0.282765 loss)
I0428 15:16:58.649622  7476 sgd_solver.cpp:105] Iteration 9864, lr = 0.00141718
I0428 15:17:03.961812  7476 solver.cpp:218] Iteration 9876 (2.25902 iter/s, 5.31204s/12 iters), loss = 0.262383
I0428 15:17:03.961855  7476 solver.cpp:237]     Train net output #0: loss = 0.262383 (* 1 = 0.262383 loss)
I0428 15:17:03.961863  7476 sgd_solver.cpp:105] Iteration 9876, lr = 0.00141381
I0428 15:17:09.416461  7476 solver.cpp:218] Iteration 9888 (2.20004 iter/s, 5.45443s/12 iters), loss = 0.332703
I0428 15:17:09.416540  7476 solver.cpp:237]     Train net output #0: loss = 0.332703 (* 1 = 0.332703 loss)
I0428 15:17:09.416550  7476 sgd_solver.cpp:105] Iteration 9888, lr = 0.00141045
I0428 15:17:11.573401  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_9894.caffemodel
I0428 15:17:15.739605  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_9894.solverstate
I0428 15:17:17.788300  7476 solver.cpp:330] Iteration 9894, Testing net (#0)
I0428 15:17:17.788323  7476 net.cpp:676] Ignoring source layer train-data
I0428 15:17:18.288383  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:17:22.519440  7476 solver.cpp:397]     Test net output #0: accuracy = 0.431985
I0428 15:17:22.519476  7476 solver.cpp:397]     Test net output #1: loss = 3.52845 (* 1 = 3.52845 loss)
I0428 15:17:24.551033  7476 solver.cpp:218] Iteration 9900 (0.792914 iter/s, 15.1341s/12 iters), loss = 0.398395
I0428 15:17:24.551079  7476 solver.cpp:237]     Train net output #0: loss = 0.398395 (* 1 = 0.398395 loss)
I0428 15:17:24.551087  7476 sgd_solver.cpp:105] Iteration 9900, lr = 0.00140711
I0428 15:17:29.924856  7476 solver.cpp:218] Iteration 9912 (2.23313 iter/s, 5.37361s/12 iters), loss = 0.256887
I0428 15:17:29.925253  7476 solver.cpp:237]     Train net output #0: loss = 0.256887 (* 1 = 0.256887 loss)
I0428 15:17:29.925264  7476 sgd_solver.cpp:105] Iteration 9912, lr = 0.00140377
I0428 15:17:30.025761  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:17:35.279094  7476 solver.cpp:218] Iteration 9924 (2.24145 iter/s, 5.35368s/12 iters), loss = 0.231568
I0428 15:17:35.279137  7476 solver.cpp:237]     Train net output #0: loss = 0.231568 (* 1 = 0.231568 loss)
I0428 15:17:35.279146  7476 sgd_solver.cpp:105] Iteration 9924, lr = 0.00140043
I0428 15:17:40.715157  7476 solver.cpp:218] Iteration 9936 (2.20757 iter/s, 5.43585s/12 iters), loss = 0.27379
I0428 15:17:40.715205  7476 solver.cpp:237]     Train net output #0: loss = 0.27379 (* 1 = 0.27379 loss)
I0428 15:17:40.715214  7476 sgd_solver.cpp:105] Iteration 9936, lr = 0.00139711
I0428 15:17:46.153894  7476 solver.cpp:218] Iteration 9948 (2.20648 iter/s, 5.43852s/12 iters), loss = 0.221924
I0428 15:17:46.153935  7476 solver.cpp:237]     Train net output #0: loss = 0.221924 (* 1 = 0.221924 loss)
I0428 15:17:46.153944  7476 sgd_solver.cpp:105] Iteration 9948, lr = 0.00139379
I0428 15:17:51.495455  7476 solver.cpp:218] Iteration 9960 (2.24662 iter/s, 5.34135s/12 iters), loss = 0.210502
I0428 15:17:51.495507  7476 solver.cpp:237]     Train net output #0: loss = 0.210502 (* 1 = 0.210502 loss)
I0428 15:17:51.495517  7476 sgd_solver.cpp:105] Iteration 9960, lr = 0.00139048
I0428 15:17:56.898768  7476 solver.cpp:218] Iteration 9972 (2.22095 iter/s, 5.40309s/12 iters), loss = 0.378149
I0428 15:17:56.898811  7476 solver.cpp:237]     Train net output #0: loss = 0.378149 (* 1 = 0.378149 loss)
I0428 15:17:56.898820  7476 sgd_solver.cpp:105] Iteration 9972, lr = 0.00138718
I0428 15:18:02.326162  7476 solver.cpp:218] Iteration 9984 (2.21109 iter/s, 5.42718s/12 iters), loss = 0.250859
I0428 15:18:02.326257  7476 solver.cpp:237]     Train net output #0: loss = 0.250859 (* 1 = 0.250859 loss)
I0428 15:18:02.326269  7476 sgd_solver.cpp:105] Iteration 9984, lr = 0.00138389
I0428 15:18:07.248700  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_9996.caffemodel
I0428 15:18:11.595252  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_9996.solverstate
I0428 15:18:14.701411  7476 solver.cpp:330] Iteration 9996, Testing net (#0)
I0428 15:18:14.701436  7476 net.cpp:676] Ignoring source layer train-data
I0428 15:18:15.248929  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:18:19.540896  7476 solver.cpp:397]     Test net output #0: accuracy = 0.425858
I0428 15:18:19.540923  7476 solver.cpp:397]     Test net output #1: loss = 3.58767 (* 1 = 3.58767 loss)
I0428 15:18:19.667093  7476 solver.cpp:218] Iteration 9996 (0.692028 iter/s, 17.3403s/12 iters), loss = 0.266173
I0428 15:18:19.667129  7476 solver.cpp:237]     Train net output #0: loss = 0.266173 (* 1 = 0.266173 loss)
I0428 15:18:19.667137  7476 sgd_solver.cpp:105] Iteration 9996, lr = 0.0013806
I0428 15:18:24.144479  7476 solver.cpp:218] Iteration 10008 (2.68024 iter/s, 4.47721s/12 iters), loss = 0.382708
I0428 15:18:24.144542  7476 solver.cpp:237]     Train net output #0: loss = 0.382708 (* 1 = 0.382708 loss)
I0428 15:18:24.144551  7476 sgd_solver.cpp:105] Iteration 10008, lr = 0.00137732
I0428 15:18:26.559501  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:18:29.598440  7476 solver.cpp:218] Iteration 10020 (2.20033 iter/s, 5.45373s/12 iters), loss = 0.321912
I0428 15:18:29.598484  7476 solver.cpp:237]     Train net output #0: loss = 0.321912 (* 1 = 0.321912 loss)
I0428 15:18:29.598495  7476 sgd_solver.cpp:105] Iteration 10020, lr = 0.00137405
I0428 15:18:35.112527  7476 solver.cpp:218] Iteration 10032 (2.17633 iter/s, 5.51388s/12 iters), loss = 0.433991
I0428 15:18:35.112663  7476 solver.cpp:237]     Train net output #0: loss = 0.433991 (* 1 = 0.433991 loss)
I0428 15:18:35.112673  7476 sgd_solver.cpp:105] Iteration 10032, lr = 0.00137079
I0428 15:18:40.520848  7476 solver.cpp:218] Iteration 10044 (2.21893 iter/s, 5.40802s/12 iters), loss = 0.235058
I0428 15:18:40.520889  7476 solver.cpp:237]     Train net output #0: loss = 0.235058 (* 1 = 0.235058 loss)
I0428 15:18:40.520900  7476 sgd_solver.cpp:105] Iteration 10044, lr = 0.00136754
I0428 15:18:45.798173  7476 solver.cpp:218] Iteration 10056 (2.27397 iter/s, 5.27712s/12 iters), loss = 0.179497
I0428 15:18:45.798216  7476 solver.cpp:237]     Train net output #0: loss = 0.179497 (* 1 = 0.179497 loss)
I0428 15:18:45.798226  7476 sgd_solver.cpp:105] Iteration 10056, lr = 0.00136429
I0428 15:18:51.231324  7476 solver.cpp:218] Iteration 10068 (2.20875 iter/s, 5.43294s/12 iters), loss = 0.265343
I0428 15:18:51.231374  7476 solver.cpp:237]     Train net output #0: loss = 0.265343 (* 1 = 0.265343 loss)
I0428 15:18:51.231384  7476 sgd_solver.cpp:105] Iteration 10068, lr = 0.00136105
I0428 15:18:56.633281  7476 solver.cpp:218] Iteration 10080 (2.22151 iter/s, 5.40174s/12 iters), loss = 0.258999
I0428 15:18:56.633322  7476 solver.cpp:237]     Train net output #0: loss = 0.258999 (* 1 = 0.258999 loss)
I0428 15:18:56.633329  7476 sgd_solver.cpp:105] Iteration 10080, lr = 0.00135782
I0428 15:19:01.890398  7476 solver.cpp:218] Iteration 10092 (2.28271 iter/s, 5.25692s/12 iters), loss = 0.287033
I0428 15:19:01.890441  7476 solver.cpp:237]     Train net output #0: loss = 0.287033 (* 1 = 0.287033 loss)
I0428 15:19:01.890450  7476 sgd_solver.cpp:105] Iteration 10092, lr = 0.0013546
I0428 15:19:04.092674  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_10098.caffemodel
I0428 15:19:11.396445  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_10098.solverstate
I0428 15:19:14.726929  7476 solver.cpp:330] Iteration 10098, Testing net (#0)
I0428 15:19:14.726953  7476 net.cpp:676] Ignoring source layer train-data
I0428 15:19:15.154875  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:19:19.232933  7476 solver.cpp:397]     Test net output #0: accuracy = 0.438726
I0428 15:19:19.232961  7476 solver.cpp:397]     Test net output #1: loss = 3.55243 (* 1 = 3.55243 loss)
I0428 15:19:21.170523  7476 solver.cpp:218] Iteration 10104 (0.622422 iter/s, 19.2795s/12 iters), loss = 0.135128
I0428 15:19:21.170572  7476 solver.cpp:237]     Train net output #0: loss = 0.135128 (* 1 = 0.135128 loss)
I0428 15:19:21.170581  7476 sgd_solver.cpp:105] Iteration 10104, lr = 0.00135138
I0428 15:19:25.897578  7495 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:19:26.585280  7476 solver.cpp:218] Iteration 10116 (2.21625 iter/s, 5.41454s/12 iters), loss = 0.287995
I0428 15:19:26.585325  7476 solver.cpp:237]     Train net output #0: loss = 0.287995 (* 1 = 0.287995 loss)
I0428 15:19:26.585332  7476 sgd_solver.cpp:105] Iteration 10116, lr = 0.00134817
I0428 15:19:31.878334  7476 solver.cpp:218] Iteration 10128 (2.26721 iter/s, 5.29284s/12 iters), loss = 0.192799
I0428 15:19:31.878381  7476 solver.cpp:237]     Train net output #0: loss = 0.192799 (* 1 = 0.192799 loss)
I0428 15:19:31.878389  7476 sgd_solver.cpp:105] Iteration 10128, lr = 0.00134497
I0428 15:19:37.307123  7476 solver.cpp:218] Iteration 10140 (2.21052 iter/s, 5.42858s/12 iters), loss = 0.240804
I0428 15:19:37.307166  7476 solver.cpp:237]     Train net output #0: loss = 0.240804 (* 1 = 0.240804 loss)
I0428 15:19:37.307174  7476 sgd_solver.cpp:105] Iteration 10140, lr = 0.00134178
I0428 15:19:42.652444  7476 solver.cpp:218] Iteration 10152 (2.24504 iter/s, 5.34511s/12 iters), loss = 0.247219
I0428 15:19:42.652601  7476 solver.cpp:237]     Train net output #0: loss = 0.247219 (* 1 = 0.247219 loss)
I0428 15:19:42.652609  7476 sgd_solver.cpp:105] Iteration 10152, lr = 0.00133859
I0428 15:19:48.059309  7476 solver.cpp:218] Iteration 10164 (2.21953 iter/s, 5.40655s/12 iters), loss = 0.318665
I0428 15:19:48.059352  7476 solver.cpp:237]     Train net output #0: loss = 0.318665 (* 1 = 0.318665 loss)
I0428 15:19:48.059360  7476 sgd_solver.cpp:105] Iteration 10164, lr = 0.00133541
I0428 15:19:53.264312  7476 solver.cpp:218] Iteration 10176 (2.30557 iter/s, 5.2048s/12 iters), loss = 0.238416
I0428 15:19:53.264354  7476 solver.cpp:237]     Train net output #0: loss = 0.238416 (* 1 = 0.238416 loss)
I0428 15:19:53.264362  7476 sgd_solver.cpp:105] Iteration 10176, lr = 0.00133224
I0428 15:19:58.781590  7476 solver.cpp:218] Iteration 10188 (2.17507 iter/s, 5.51706s/12 iters), loss = 0.165624
I0428 15:19:58.781631  7476 solver.cpp:237]     Train net output #0: loss = 0.165624 (* 1 = 0.165624 loss)
I0428 15:19:58.781641  7476 sgd_solver.cpp:105] Iteration 10188, lr = 0.00132908
I0428 15:20:03.651192  7476 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_10200.caffemodel
I0428 15:20:10.064640  7476 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_10200.solverstate
I0428 15:20:13.196739  7476 solver.cpp:310] Iteration 10200, loss = 0.273665
I0428 15:20:13.198506  7476 solver.cpp:330] Iteration 10200, Testing net (#0)
I0428 15:20:13.198513  7476 net.cpp:676] Ignoring source layer train-data
I0428 15:20:13.636046  7520 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:20:17.946107  7476 solver.cpp:397]     Test net output #0: accuracy = 0.432598
I0428 15:20:17.946135  7476 solver.cpp:397]     Test net output #1: loss = 3.59043 (* 1 = 3.59043 loss)
I0428 15:20:17.946139  7476 solver.cpp:315] Optimization Done.
I0428 15:20:17.946143  7476 caffe.cpp:259] Optimization Done.