DIGITS-CNN/cars/architecture-investigations/conv/layers/layer1.5/kernel/11/caffe_output.log

I0428 14:42:23.188472 29440 upgrade_proto.cpp:1082] Attempting to upgrade input file specified using deprecated 'solver_type' field (enum)': /mnt/bigdisk/DIGITS-MAN-3/digits/jobs/20210428-115451-c443/solver.prototxt
I0428 14:42:23.188719 29440 upgrade_proto.cpp:1089] Successfully upgraded file specified using deprecated 'solver_type' field (enum) to 'type' field (string).
W0428 14:42:23.188724 29440 upgrade_proto.cpp:1091] Note that future Caffe releases will only support 'type' field (string) for a solver's type.
I0428 14:42:23.188783 29440 caffe.cpp:218] Using GPUs 3
I0428 14:42:23.203313 29440 caffe.cpp:223] GPU 3: GeForce GTX 1080 Ti
I0428 14:42:23.601500 29440 solver.cpp:44] Initializing solver from parameters:
test_iter: 51
test_interval: 102
base_lr: 0.01
display: 12
max_iter: 10200
lr_policy: "exp"
gamma: 0.99980193
momentum: 0.9
weight_decay: 0.0001
snapshot: 102
snapshot_prefix: "snapshot"
solver_mode: GPU
device_id: 3
net: "train_val.prototxt"
train_state {
level: 0
stage: ""
}
type: "SGD"
I0428 14:42:23.694342 29440 solver.cpp:87] Creating training net from net file: train_val.prototxt
I0428 14:42:23.746060 29440 net.cpp:294] The NetState phase (0) differed from the phase (1) specified by a rule in layer val-data
I0428 14:42:23.746093 29440 net.cpp:294] The NetState phase (0) differed from the phase (1) specified by a rule in layer accuracy
I0428 14:42:23.746358 29440 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: 11
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 14:42:23.746505 29440 layer_factory.hpp:77] Creating layer train-data
I0428 14:42:23.986516 29440 db_lmdb.cpp:35] Opened lmdb /mnt/bigdisk/DIGITS-MAN-3/digits/jobs/20210421-230320-902c/train_db
I0428 14:42:23.987079 29440 net.cpp:84] Creating Layer train-data
I0428 14:42:23.987102 29440 net.cpp:380] train-data -> data
I0428 14:42:23.987129 29440 net.cpp:380] train-data -> label
I0428 14:42:23.987146 29440 data_transformer.cpp:25] Loading mean file from: /mnt/bigdisk/DIGITS-MAN-3/digits/jobs/20210421-230320-902c/mean.binaryproto
I0428 14:42:24.069190 29440 data_layer.cpp:45] output data size: 128,3,227,227
I0428 14:42:24.374904 29440 net.cpp:122] Setting up train-data
I0428 14:42:24.374927 29440 net.cpp:129] Top shape: 128 3 227 227 (19787136)
I0428 14:42:24.374933 29440 net.cpp:129] Top shape: 128 (128)
I0428 14:42:24.374936 29440 net.cpp:137] Memory required for data: 79149056
I0428 14:42:24.374944 29440 layer_factory.hpp:77] Creating layer conv1
I0428 14:42:24.374965 29440 net.cpp:84] Creating Layer conv1
I0428 14:42:24.374971 29440 net.cpp:406] conv1 <- data
I0428 14:42:24.374984 29440 net.cpp:380] conv1 -> conv1
I0428 14:42:25.104444 29440 net.cpp:122] Setting up conv1
I0428 14:42:25.104481 29440 net.cpp:129] Top shape: 128 96 55 55 (37171200)
I0428 14:42:25.104511 29440 net.cpp:137] Memory required for data: 227833856
I0428 14:42:25.104532 29440 layer_factory.hpp:77] Creating layer relu1
I0428 14:42:25.104543 29440 net.cpp:84] Creating Layer relu1
I0428 14:42:25.104548 29440 net.cpp:406] relu1 <- conv1
I0428 14:42:25.104553 29440 net.cpp:367] relu1 -> conv1 (in-place)
I0428 14:42:25.104848 29440 net.cpp:122] Setting up relu1
I0428 14:42:25.104857 29440 net.cpp:129] Top shape: 128 96 55 55 (37171200)
I0428 14:42:25.104861 29440 net.cpp:137] Memory required for data: 376518656
I0428 14:42:25.104864 29440 layer_factory.hpp:77] Creating layer norm1
I0428 14:42:25.104874 29440 net.cpp:84] Creating Layer norm1
I0428 14:42:25.104877 29440 net.cpp:406] norm1 <- conv1
I0428 14:42:25.104882 29440 net.cpp:380] norm1 -> norm1
I0428 14:42:25.105332 29440 net.cpp:122] Setting up norm1
I0428 14:42:25.105342 29440 net.cpp:129] Top shape: 128 96 55 55 (37171200)
I0428 14:42:25.105345 29440 net.cpp:137] Memory required for data: 525203456
I0428 14:42:25.105350 29440 layer_factory.hpp:77] Creating layer pool1
I0428 14:42:25.105357 29440 net.cpp:84] Creating Layer pool1
I0428 14:42:25.105360 29440 net.cpp:406] pool1 <- norm1
I0428 14:42:25.105366 29440 net.cpp:380] pool1 -> pool1
I0428 14:42:25.105402 29440 net.cpp:122] Setting up pool1
I0428 14:42:25.105408 29440 net.cpp:129] Top shape: 128 96 27 27 (8957952)
I0428 14:42:25.105412 29440 net.cpp:137] Memory required for data: 561035264
I0428 14:42:25.105415 29440 layer_factory.hpp:77] Creating layer conv1.5
I0428 14:42:25.105425 29440 net.cpp:84] Creating Layer conv1.5
I0428 14:42:25.105429 29440 net.cpp:406] conv1.5 <- pool1
I0428 14:42:25.105434 29440 net.cpp:380] conv1.5 -> conv1.5
I0428 14:42:25.126418 29440 net.cpp:122] Setting up conv1.5
I0428 14:42:25.126437 29440 net.cpp:129] Top shape: 128 176 17 17 (6510592)
I0428 14:42:25.126441 29440 net.cpp:137] Memory required for data: 587077632
I0428 14:42:25.126452 29440 layer_factory.hpp:77] Creating layer relu1.5
I0428 14:42:25.126461 29440 net.cpp:84] Creating Layer relu1.5
I0428 14:42:25.126466 29440 net.cpp:406] relu1.5 <- conv1.5
I0428 14:42:25.126471 29440 net.cpp:367] relu1.5 -> conv1.5 (in-place)
I0428 14:42:25.126751 29440 net.cpp:122] Setting up relu1.5
I0428 14:42:25.126760 29440 net.cpp:129] Top shape: 128 176 17 17 (6510592)
I0428 14:42:25.126762 29440 net.cpp:137] Memory required for data: 613120000
I0428 14:42:25.126766 29440 layer_factory.hpp:77] Creating layer norm1.5
I0428 14:42:25.126773 29440 net.cpp:84] Creating Layer norm1.5
I0428 14:42:25.126777 29440 net.cpp:406] norm1.5 <- conv1.5
I0428 14:42:25.126782 29440 net.cpp:380] norm1.5 -> norm1.5
I0428 14:42:25.149451 29440 net.cpp:122] Setting up norm1.5
I0428 14:42:25.149468 29440 net.cpp:129] Top shape: 128 176 17 17 (6510592)
I0428 14:42:25.149472 29440 net.cpp:137] Memory required for data: 639162368
I0428 14:42:25.149478 29440 layer_factory.hpp:77] Creating layer pool1.5
I0428 14:42:25.149490 29440 net.cpp:84] Creating Layer pool1.5
I0428 14:42:25.149494 29440 net.cpp:406] pool1.5 <- norm1.5
I0428 14:42:25.149502 29440 net.cpp:380] pool1.5 -> pool1.5
I0428 14:42:25.149534 29440 net.cpp:122] Setting up pool1.5
I0428 14:42:25.149539 29440 net.cpp:129] Top shape: 128 176 8 8 (1441792)
I0428 14:42:25.149542 29440 net.cpp:137] Memory required for data: 644929536
I0428 14:42:25.149546 29440 layer_factory.hpp:77] Creating layer conv2
I0428 14:42:25.149555 29440 net.cpp:84] Creating Layer conv2
I0428 14:42:25.149559 29440 net.cpp:406] conv2 <- pool1.5
I0428 14:42:25.149564 29440 net.cpp:380] conv2 -> conv2
I0428 14:42:25.211534 29440 net.cpp:122] Setting up conv2
I0428 14:42:25.211558 29440 net.cpp:129] Top shape: 128 256 8 8 (2097152)
I0428 14:42:25.211565 29440 net.cpp:137] Memory required for data: 653318144
I0428 14:42:25.211581 29440 layer_factory.hpp:77] Creating layer relu2
I0428 14:42:25.211592 29440 net.cpp:84] Creating Layer relu2
I0428 14:42:25.211599 29440 net.cpp:406] relu2 <- conv2
I0428 14:42:25.211609 29440 net.cpp:367] relu2 -> conv2 (in-place)
I0428 14:42:25.212327 29440 net.cpp:122] Setting up relu2
I0428 14:42:25.212340 29440 net.cpp:129] Top shape: 128 256 8 8 (2097152)
I0428 14:42:25.212344 29440 net.cpp:137] Memory required for data: 661706752
I0428 14:42:25.212349 29440 layer_factory.hpp:77] Creating layer norm2
I0428 14:42:25.212360 29440 net.cpp:84] Creating Layer norm2
I0428 14:42:25.212366 29440 net.cpp:406] norm2 <- conv2
I0428 14:42:25.212374 29440 net.cpp:380] norm2 -> norm2
I0428 14:42:25.212815 29440 net.cpp:122] Setting up norm2
I0428 14:42:25.212827 29440 net.cpp:129] Top shape: 128 256 8 8 (2097152)
I0428 14:42:25.212832 29440 net.cpp:137] Memory required for data: 670095360
I0428 14:42:25.212838 29440 layer_factory.hpp:77] Creating layer pool2
I0428 14:42:25.212847 29440 net.cpp:84] Creating Layer pool2
I0428 14:42:25.212852 29440 net.cpp:406] pool2 <- norm2
I0428 14:42:25.212860 29440 net.cpp:380] pool2 -> pool2
I0428 14:42:25.212905 29440 net.cpp:122] Setting up pool2
I0428 14:42:25.212913 29440 net.cpp:129] Top shape: 128 256 4 4 (524288)
I0428 14:42:25.212919 29440 net.cpp:137] Memory required for data: 672192512
I0428 14:42:25.212925 29440 layer_factory.hpp:77] Creating layer conv3
I0428 14:42:25.212939 29440 net.cpp:84] Creating Layer conv3
I0428 14:42:25.212945 29440 net.cpp:406] conv3 <- pool2
I0428 14:42:25.212954 29440 net.cpp:380] conv3 -> conv3
I0428 14:42:25.289469 29440 net.cpp:122] Setting up conv3
I0428 14:42:25.289494 29440 net.cpp:129] Top shape: 128 384 4 4 (786432)
I0428 14:42:25.289499 29440 net.cpp:137] Memory required for data: 675338240
I0428 14:42:25.289513 29440 layer_factory.hpp:77] Creating layer relu3
I0428 14:42:25.289525 29440 net.cpp:84] Creating Layer relu3
I0428 14:42:25.289533 29440 net.cpp:406] relu3 <- conv3
I0428 14:42:25.289543 29440 net.cpp:367] relu3 -> conv3 (in-place)
I0428 14:42:25.290242 29440 net.cpp:122] Setting up relu3
I0428 14:42:25.290256 29440 net.cpp:129] Top shape: 128 384 4 4 (786432)
I0428 14:42:25.290262 29440 net.cpp:137] Memory required for data: 678483968
I0428 14:42:25.290268 29440 layer_factory.hpp:77] Creating layer conv4
I0428 14:42:25.290285 29440 net.cpp:84] Creating Layer conv4
I0428 14:42:25.290292 29440 net.cpp:406] conv4 <- conv3
I0428 14:42:25.290304 29440 net.cpp:380] conv4 -> conv4
I0428 14:42:25.305485 29440 net.cpp:122] Setting up conv4
I0428 14:42:25.305508 29440 net.cpp:129] Top shape: 128 384 4 4 (786432)
I0428 14:42:25.305514 29440 net.cpp:137] Memory required for data: 681629696
I0428 14:42:25.305532 29440 layer_factory.hpp:77] Creating layer relu4
I0428 14:42:25.305552 29440 net.cpp:84] Creating Layer relu4
I0428 14:42:25.305560 29440 net.cpp:406] relu4 <- conv4
I0428 14:42:25.305572 29440 net.cpp:367] relu4 -> conv4 (in-place)
I0428 14:42:25.306288 29440 net.cpp:122] Setting up relu4
I0428 14:42:25.306303 29440 net.cpp:129] Top shape: 128 384 4 4 (786432)
I0428 14:42:25.306309 29440 net.cpp:137] Memory required for data: 684775424
I0428 14:42:25.306315 29440 layer_factory.hpp:77] Creating layer conv5
I0428 14:42:25.306330 29440 net.cpp:84] Creating Layer conv5
I0428 14:42:25.306336 29440 net.cpp:406] conv5 <- conv4
I0428 14:42:25.306345 29440 net.cpp:380] conv5 -> conv5
I0428 14:42:25.342564 29440 net.cpp:122] Setting up conv5
I0428 14:42:25.342592 29440 net.cpp:129] Top shape: 128 256 4 4 (524288)
I0428 14:42:25.342597 29440 net.cpp:137] Memory required for data: 686872576
I0428 14:42:25.342612 29440 layer_factory.hpp:77] Creating layer relu5
I0428 14:42:25.342624 29440 net.cpp:84] Creating Layer relu5
I0428 14:42:25.342630 29440 net.cpp:406] relu5 <- conv5
I0428 14:42:25.342639 29440 net.cpp:367] relu5 -> conv5 (in-place)
I0428 14:42:25.345038 29440 net.cpp:122] Setting up relu5
I0428 14:42:25.345054 29440 net.cpp:129] Top shape: 128 256 4 4 (524288)
I0428 14:42:25.345060 29440 net.cpp:137] Memory required for data: 688969728
I0428 14:42:25.345067 29440 layer_factory.hpp:77] Creating layer pool5
I0428 14:42:25.345077 29440 net.cpp:84] Creating Layer pool5
I0428 14:42:25.345082 29440 net.cpp:406] pool5 <- conv5
I0428 14:42:25.345090 29440 net.cpp:380] pool5 -> pool5
I0428 14:42:25.345172 29440 net.cpp:122] Setting up pool5
I0428 14:42:25.345181 29440 net.cpp:129] Top shape: 128 256 2 2 (131072)
I0428 14:42:25.345185 29440 net.cpp:137] Memory required for data: 689494016
I0428 14:42:25.345191 29440 layer_factory.hpp:77] Creating layer fc6
I0428 14:42:25.345203 29440 net.cpp:84] Creating Layer fc6
I0428 14:42:25.345208 29440 net.cpp:406] fc6 <- pool5
I0428 14:42:25.345217 29440 net.cpp:380] fc6 -> fc6
I0428 14:42:25.405196 29440 net.cpp:122] Setting up fc6
I0428 14:42:25.405218 29440 net.cpp:129] Top shape: 128 4096 (524288)
I0428 14:42:25.405222 29440 net.cpp:137] Memory required for data: 691591168
I0428 14:42:25.405232 29440 layer_factory.hpp:77] Creating layer relu6
I0428 14:42:25.405242 29440 net.cpp:84] Creating Layer relu6
I0428 14:42:25.405247 29440 net.cpp:406] relu6 <- fc6
I0428 14:42:25.405256 29440 net.cpp:367] relu6 -> fc6 (in-place)
I0428 14:42:25.405877 29440 net.cpp:122] Setting up relu6
I0428 14:42:25.405887 29440 net.cpp:129] Top shape: 128 4096 (524288)
I0428 14:42:25.405891 29440 net.cpp:137] Memory required for data: 693688320
I0428 14:42:25.405895 29440 layer_factory.hpp:77] Creating layer drop6
I0428 14:42:25.405903 29440 net.cpp:84] Creating Layer drop6
I0428 14:42:25.405908 29440 net.cpp:406] drop6 <- fc6
I0428 14:42:25.405915 29440 net.cpp:367] drop6 -> fc6 (in-place)
I0428 14:42:25.405943 29440 net.cpp:122] Setting up drop6
I0428 14:42:25.405951 29440 net.cpp:129] Top shape: 128 4096 (524288)
I0428 14:42:25.405956 29440 net.cpp:137] Memory required for data: 695785472
I0428 14:42:25.405961 29440 layer_factory.hpp:77] Creating layer fc7
I0428 14:42:25.405968 29440 net.cpp:84] Creating Layer fc7
I0428 14:42:25.405973 29440 net.cpp:406] fc7 <- fc6
I0428 14:42:25.405980 29440 net.cpp:380] fc7 -> fc7
I0428 14:42:25.567991 29440 net.cpp:122] Setting up fc7
I0428 14:42:25.568015 29440 net.cpp:129] Top shape: 128 4096 (524288)
I0428 14:42:25.568019 29440 net.cpp:137] Memory required for data: 697882624
I0428 14:42:25.568028 29440 layer_factory.hpp:77] Creating layer relu7
I0428 14:42:25.568038 29440 net.cpp:84] Creating Layer relu7
I0428 14:42:25.568043 29440 net.cpp:406] relu7 <- fc7
I0428 14:42:25.568049 29440 net.cpp:367] relu7 -> fc7 (in-place)
I0428 14:42:25.568468 29440 net.cpp:122] Setting up relu7
I0428 14:42:25.568476 29440 net.cpp:129] Top shape: 128 4096 (524288)
I0428 14:42:25.568481 29440 net.cpp:137] Memory required for data: 699979776
I0428 14:42:25.568511 29440 layer_factory.hpp:77] Creating layer drop7
I0428 14:42:25.568517 29440 net.cpp:84] Creating Layer drop7
I0428 14:42:25.568522 29440 net.cpp:406] drop7 <- fc7
I0428 14:42:25.568529 29440 net.cpp:367] drop7 -> fc7 (in-place)
I0428 14:42:25.568553 29440 net.cpp:122] Setting up drop7
I0428 14:42:25.568560 29440 net.cpp:129] Top shape: 128 4096 (524288)
I0428 14:42:25.568564 29440 net.cpp:137] Memory required for data: 702076928
I0428 14:42:25.568568 29440 layer_factory.hpp:77] Creating layer fc8
I0428 14:42:25.568576 29440 net.cpp:84] Creating Layer fc8
I0428 14:42:25.568580 29440 net.cpp:406] fc8 <- fc7
I0428 14:42:25.568586 29440 net.cpp:380] fc8 -> fc8
I0428 14:42:25.577450 29440 net.cpp:122] Setting up fc8
I0428 14:42:25.577468 29440 net.cpp:129] Top shape: 128 196 (25088)
I0428 14:42:25.577471 29440 net.cpp:137] Memory required for data: 702177280
I0428 14:42:25.577486 29440 layer_factory.hpp:77] Creating layer loss
I0428 14:42:25.577494 29440 net.cpp:84] Creating Layer loss
I0428 14:42:25.577498 29440 net.cpp:406] loss <- fc8
I0428 14:42:25.577504 29440 net.cpp:406] loss <- label
I0428 14:42:25.577512 29440 net.cpp:380] loss -> loss
I0428 14:42:25.577523 29440 layer_factory.hpp:77] Creating layer loss
I0428 14:42:25.578794 29440 net.cpp:122] Setting up loss
I0428 14:42:25.578802 29440 net.cpp:129] Top shape: (1)
I0428 14:42:25.578806 29440 net.cpp:132]     with loss weight 1
I0428 14:42:25.578826 29440 net.cpp:137] Memory required for data: 702177284
I0428 14:42:25.578830 29440 net.cpp:198] loss needs backward computation.
I0428 14:42:25.578837 29440 net.cpp:198] fc8 needs backward computation.
I0428 14:42:25.578861 29440 net.cpp:198] drop7 needs backward computation.
I0428 14:42:25.578864 29440 net.cpp:198] relu7 needs backward computation.
I0428 14:42:25.578868 29440 net.cpp:198] fc7 needs backward computation.
I0428 14:42:25.578871 29440 net.cpp:198] drop6 needs backward computation.
I0428 14:42:25.578876 29440 net.cpp:198] relu6 needs backward computation.
I0428 14:42:25.578879 29440 net.cpp:198] fc6 needs backward computation.
I0428 14:42:25.578883 29440 net.cpp:198] pool5 needs backward computation.
I0428 14:42:25.578886 29440 net.cpp:198] relu5 needs backward computation.
I0428 14:42:25.578891 29440 net.cpp:198] conv5 needs backward computation.
I0428 14:42:25.578894 29440 net.cpp:198] relu4 needs backward computation.
I0428 14:42:25.578898 29440 net.cpp:198] conv4 needs backward computation.
I0428 14:42:25.578902 29440 net.cpp:198] relu3 needs backward computation.
I0428 14:42:25.578905 29440 net.cpp:198] conv3 needs backward computation.
I0428 14:42:25.578908 29440 net.cpp:198] pool2 needs backward computation.
I0428 14:42:25.578912 29440 net.cpp:198] norm2 needs backward computation.
I0428 14:42:25.578917 29440 net.cpp:198] relu2 needs backward computation.
I0428 14:42:25.578919 29440 net.cpp:198] conv2 needs backward computation.
I0428 14:42:25.578923 29440 net.cpp:198] pool1.5 needs backward computation.
I0428 14:42:25.578927 29440 net.cpp:198] norm1.5 needs backward computation.
I0428 14:42:25.578930 29440 net.cpp:198] relu1.5 needs backward computation.
I0428 14:42:25.578934 29440 net.cpp:198] conv1.5 needs backward computation.
I0428 14:42:25.578938 29440 net.cpp:198] pool1 needs backward computation.
I0428 14:42:25.578941 29440 net.cpp:198] norm1 needs backward computation.
I0428 14:42:25.578945 29440 net.cpp:198] relu1 needs backward computation.
I0428 14:42:25.578949 29440 net.cpp:198] conv1 needs backward computation.
I0428 14:42:25.578953 29440 net.cpp:200] train-data does not need backward computation.
I0428 14:42:25.578956 29440 net.cpp:242] This network produces output loss
I0428 14:42:25.578974 29440 net.cpp:255] Network initialization done.
I0428 14:42:25.579517 29440 solver.cpp:172] Creating test net (#0) specified by net file: train_val.prototxt
I0428 14:42:25.579550 29440 net.cpp:294] The NetState phase (1) differed from the phase (0) specified by a rule in layer train-data
I0428 14:42:25.579703 29440 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: 11
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 14:42:25.579809 29440 layer_factory.hpp:77] Creating layer val-data
I0428 14:42:25.581405 29440 db_lmdb.cpp:35] Opened lmdb /mnt/bigdisk/DIGITS-MAN-3/digits/jobs/20210421-230320-902c/val_db
I0428 14:42:25.581588 29440 net.cpp:84] Creating Layer val-data
I0428 14:42:25.581598 29440 net.cpp:380] val-data -> data
I0428 14:42:25.581607 29440 net.cpp:380] val-data -> label
I0428 14:42:25.581614 29440 data_transformer.cpp:25] Loading mean file from: /mnt/bigdisk/DIGITS-MAN-3/digits/jobs/20210421-230320-902c/mean.binaryproto
I0428 14:42:25.585081 29440 data_layer.cpp:45] output data size: 32,3,227,227
I0428 14:42:25.617525 29440 net.cpp:122] Setting up val-data
I0428 14:42:25.617547 29440 net.cpp:129] Top shape: 32 3 227 227 (4946784)
I0428 14:42:25.617552 29440 net.cpp:129] Top shape: 32 (32)
I0428 14:42:25.617556 29440 net.cpp:137] Memory required for data: 19787264
I0428 14:42:25.617563 29440 layer_factory.hpp:77] Creating layer label_val-data_1_split
I0428 14:42:25.617575 29440 net.cpp:84] Creating Layer label_val-data_1_split
I0428 14:42:25.617579 29440 net.cpp:406] label_val-data_1_split <- label
I0428 14:42:25.617586 29440 net.cpp:380] label_val-data_1_split -> label_val-data_1_split_0
I0428 14:42:25.617595 29440 net.cpp:380] label_val-data_1_split -> label_val-data_1_split_1
I0428 14:42:25.617645 29440 net.cpp:122] Setting up label_val-data_1_split
I0428 14:42:25.617650 29440 net.cpp:129] Top shape: 32 (32)
I0428 14:42:25.617655 29440 net.cpp:129] Top shape: 32 (32)
I0428 14:42:25.617657 29440 net.cpp:137] Memory required for data: 19787520
I0428 14:42:25.617660 29440 layer_factory.hpp:77] Creating layer conv1
I0428 14:42:25.617672 29440 net.cpp:84] Creating Layer conv1
I0428 14:42:25.617676 29440 net.cpp:406] conv1 <- data
I0428 14:42:25.617681 29440 net.cpp:380] conv1 -> conv1
I0428 14:42:25.627460 29440 net.cpp:122] Setting up conv1
I0428 14:42:25.627478 29440 net.cpp:129] Top shape: 32 96 55 55 (9292800)
I0428 14:42:25.627483 29440 net.cpp:137] Memory required for data: 56958720
I0428 14:42:25.627497 29440 layer_factory.hpp:77] Creating layer relu1
I0428 14:42:25.627507 29440 net.cpp:84] Creating Layer relu1
I0428 14:42:25.627511 29440 net.cpp:406] relu1 <- conv1
I0428 14:42:25.627519 29440 net.cpp:367] relu1 -> conv1 (in-place)
I0428 14:42:25.627835 29440 net.cpp:122] Setting up relu1
I0428 14:42:25.627846 29440 net.cpp:129] Top shape: 32 96 55 55 (9292800)
I0428 14:42:25.627849 29440 net.cpp:137] Memory required for data: 94129920
I0428 14:42:25.627853 29440 layer_factory.hpp:77] Creating layer norm1
I0428 14:42:25.627864 29440 net.cpp:84] Creating Layer norm1
I0428 14:42:25.627871 29440 net.cpp:406] norm1 <- conv1
I0428 14:42:25.627876 29440 net.cpp:380] norm1 -> norm1
I0428 14:42:25.629544 29440 net.cpp:122] Setting up norm1
I0428 14:42:25.629555 29440 net.cpp:129] Top shape: 32 96 55 55 (9292800)
I0428 14:42:25.629560 29440 net.cpp:137] Memory required for data: 131301120
I0428 14:42:25.629565 29440 layer_factory.hpp:77] Creating layer pool1
I0428 14:42:25.629573 29440 net.cpp:84] Creating Layer pool1
I0428 14:42:25.629580 29440 net.cpp:406] pool1 <- norm1
I0428 14:42:25.629585 29440 net.cpp:380] pool1 -> pool1
I0428 14:42:25.629617 29440 net.cpp:122] Setting up pool1
I0428 14:42:25.629623 29440 net.cpp:129] Top shape: 32 96 27 27 (2239488)
I0428 14:42:25.629627 29440 net.cpp:137] Memory required for data: 140259072
I0428 14:42:25.629632 29440 layer_factory.hpp:77] Creating layer conv1.5
I0428 14:42:25.629642 29440 net.cpp:84] Creating Layer conv1.5
I0428 14:42:25.629647 29440 net.cpp:406] conv1.5 <- pool1
I0428 14:42:25.629654 29440 net.cpp:380] conv1.5 -> conv1.5
I0428 14:42:25.651742 29440 net.cpp:122] Setting up conv1.5
I0428 14:42:25.651764 29440 net.cpp:129] Top shape: 32 176 17 17 (1627648)
I0428 14:42:25.651768 29440 net.cpp:137] Memory required for data: 146769664
I0428 14:42:25.651782 29440 layer_factory.hpp:77] Creating layer relu1.5
I0428 14:42:25.651790 29440 net.cpp:84] Creating Layer relu1.5
I0428 14:42:25.651815 29440 net.cpp:406] relu1.5 <- conv1.5
I0428 14:42:25.651823 29440 net.cpp:367] relu1.5 -> conv1.5 (in-place)
I0428 14:42:25.652186 29440 net.cpp:122] Setting up relu1.5
I0428 14:42:25.652196 29440 net.cpp:129] Top shape: 32 176 17 17 (1627648)
I0428 14:42:25.652201 29440 net.cpp:137] Memory required for data: 153280256
I0428 14:42:25.652206 29440 layer_factory.hpp:77] Creating layer norm1.5
I0428 14:42:25.652217 29440 net.cpp:84] Creating Layer norm1.5
I0428 14:42:25.652221 29440 net.cpp:406] norm1.5 <- conv1.5
I0428 14:42:25.652227 29440 net.cpp:380] norm1.5 -> norm1.5
I0428 14:42:25.654485 29440 net.cpp:122] Setting up norm1.5
I0428 14:42:25.654495 29440 net.cpp:129] Top shape: 32 176 17 17 (1627648)
I0428 14:42:25.654500 29440 net.cpp:137] Memory required for data: 159790848
I0428 14:42:25.654502 29440 layer_factory.hpp:77] Creating layer pool1.5
I0428 14:42:25.654511 29440 net.cpp:84] Creating Layer pool1.5
I0428 14:42:25.654516 29440 net.cpp:406] pool1.5 <- norm1.5
I0428 14:42:25.654521 29440 net.cpp:380] pool1.5 -> pool1.5
I0428 14:42:25.654557 29440 net.cpp:122] Setting up pool1.5
I0428 14:42:25.654563 29440 net.cpp:129] Top shape: 32 176 8 8 (360448)
I0428 14:42:25.654567 29440 net.cpp:137] Memory required for data: 161232640
I0428 14:42:25.654572 29440 layer_factory.hpp:77] Creating layer conv2
I0428 14:42:25.654582 29440 net.cpp:84] Creating Layer conv2
I0428 14:42:25.654587 29440 net.cpp:406] conv2 <- pool1.5
I0428 14:42:25.654593 29440 net.cpp:380] conv2 -> conv2
I0428 14:42:25.664744 29440 net.cpp:122] Setting up conv2
I0428 14:42:25.664762 29440 net.cpp:129] Top shape: 32 256 8 8 (524288)
I0428 14:42:25.664767 29440 net.cpp:137] Memory required for data: 163329792
I0428 14:42:25.664778 29440 layer_factory.hpp:77] Creating layer relu2
I0428 14:42:25.664788 29440 net.cpp:84] Creating Layer relu2
I0428 14:42:25.664793 29440 net.cpp:406] relu2 <- conv2
I0428 14:42:25.664799 29440 net.cpp:367] relu2 -> conv2 (in-place)
I0428 14:42:25.665365 29440 net.cpp:122] Setting up relu2
I0428 14:42:25.665376 29440 net.cpp:129] Top shape: 32 256 8 8 (524288)
I0428 14:42:25.665380 29440 net.cpp:137] Memory required for data: 165426944
I0428 14:42:25.665383 29440 layer_factory.hpp:77] Creating layer norm2
I0428 14:42:25.665392 29440 net.cpp:84] Creating Layer norm2
I0428 14:42:25.665396 29440 net.cpp:406] norm2 <- conv2
I0428 14:42:25.665402 29440 net.cpp:380] norm2 -> norm2
I0428 14:42:25.665778 29440 net.cpp:122] Setting up norm2
I0428 14:42:25.665786 29440 net.cpp:129] Top shape: 32 256 8 8 (524288)
I0428 14:42:25.665791 29440 net.cpp:137] Memory required for data: 167524096
I0428 14:42:25.665793 29440 layer_factory.hpp:77] Creating layer pool2
I0428 14:42:25.665802 29440 net.cpp:84] Creating Layer pool2
I0428 14:42:25.665805 29440 net.cpp:406] pool2 <- norm2
I0428 14:42:25.665810 29440 net.cpp:380] pool2 -> pool2
I0428 14:42:25.665844 29440 net.cpp:122] Setting up pool2
I0428 14:42:25.665850 29440 net.cpp:129] Top shape: 32 256 4 4 (131072)
I0428 14:42:25.665853 29440 net.cpp:137] Memory required for data: 168048384
I0428 14:42:25.665856 29440 layer_factory.hpp:77] Creating layer conv3
I0428 14:42:25.665868 29440 net.cpp:84] Creating Layer conv3
I0428 14:42:25.665871 29440 net.cpp:406] conv3 <- pool2
I0428 14:42:25.665879 29440 net.cpp:380] conv3 -> conv3
I0428 14:42:25.763216 29440 net.cpp:122] Setting up conv3
I0428 14:42:25.763234 29440 net.cpp:129] Top shape: 32 384 4 4 (196608)
I0428 14:42:25.763238 29440 net.cpp:137] Memory required for data: 168834816
I0428 14:42:25.763247 29440 layer_factory.hpp:77] Creating layer relu3
I0428 14:42:25.763257 29440 net.cpp:84] Creating Layer relu3
I0428 14:42:25.763262 29440 net.cpp:406] relu3 <- conv3
I0428 14:42:25.763269 29440 net.cpp:367] relu3 -> conv3 (in-place)
I0428 14:42:25.763931 29440 net.cpp:122] Setting up relu3
I0428 14:42:25.763942 29440 net.cpp:129] Top shape: 32 384 4 4 (196608)
I0428 14:42:25.763945 29440 net.cpp:137] Memory required for data: 169621248
I0428 14:42:25.763949 29440 layer_factory.hpp:77] Creating layer conv4
I0428 14:42:25.763988 29440 net.cpp:84] Creating Layer conv4
I0428 14:42:25.763991 29440 net.cpp:406] conv4 <- conv3
I0428 14:42:25.763999 29440 net.cpp:380] conv4 -> conv4
I0428 14:42:25.840773 29440 net.cpp:122] Setting up conv4
I0428 14:42:25.840798 29440 net.cpp:129] Top shape: 32 384 4 4 (196608)
I0428 14:42:25.840806 29440 net.cpp:137] Memory required for data: 170407680
I0428 14:42:25.840824 29440 layer_factory.hpp:77] Creating layer relu4
I0428 14:42:25.840838 29440 net.cpp:84] Creating Layer relu4
I0428 14:42:25.840845 29440 net.cpp:406] relu4 <- conv4
I0428 14:42:25.840854 29440 net.cpp:367] relu4 -> conv4 (in-place)
I0428 14:42:25.841519 29440 net.cpp:122] Setting up relu4
I0428 14:42:25.841531 29440 net.cpp:129] Top shape: 32 384 4 4 (196608)
I0428 14:42:25.841537 29440 net.cpp:137] Memory required for data: 171194112
I0428 14:42:25.841543 29440 layer_factory.hpp:77] Creating layer conv5
I0428 14:42:25.841560 29440 net.cpp:84] Creating Layer conv5
I0428 14:42:25.841567 29440 net.cpp:406] conv5 <- conv4
I0428 14:42:25.841576 29440 net.cpp:380] conv5 -> conv5
I0428 14:42:25.854580 29440 net.cpp:122] Setting up conv5
I0428 14:42:25.854604 29440 net.cpp:129] Top shape: 32 256 4 4 (131072)
I0428 14:42:25.854609 29440 net.cpp:137] Memory required for data: 171718400
I0428 14:42:25.854620 29440 layer_factory.hpp:77] Creating layer relu5
I0428 14:42:25.854631 29440 net.cpp:84] Creating Layer relu5
I0428 14:42:25.854638 29440 net.cpp:406] relu5 <- conv5
I0428 14:42:25.854648 29440 net.cpp:367] relu5 -> conv5 (in-place)
I0428 14:42:25.855293 29440 net.cpp:122] Setting up relu5
I0428 14:42:25.855306 29440 net.cpp:129] Top shape: 32 256 4 4 (131072)
I0428 14:42:25.855311 29440 net.cpp:137] Memory required for data: 172242688
I0428 14:42:25.855316 29440 layer_factory.hpp:77] Creating layer pool5
I0428 14:42:25.855325 29440 net.cpp:84] Creating Layer pool5
I0428 14:42:25.855330 29440 net.cpp:406] pool5 <- conv5
I0428 14:42:25.855340 29440 net.cpp:380] pool5 -> pool5
I0428 14:42:25.855410 29440 net.cpp:122] Setting up pool5
I0428 14:42:25.855422 29440 net.cpp:129] Top shape: 32 256 2 2 (32768)
I0428 14:42:25.855427 29440 net.cpp:137] Memory required for data: 172373760
I0428 14:42:25.855432 29440 layer_factory.hpp:77] Creating layer fc6
I0428 14:42:25.855443 29440 net.cpp:84] Creating Layer fc6
I0428 14:42:25.855448 29440 net.cpp:406] fc6 <- pool5
I0428 14:42:25.855459 29440 net.cpp:380] fc6 -> fc6
I0428 14:42:25.899504 29440 net.cpp:122] Setting up fc6
I0428 14:42:25.899526 29440 net.cpp:129] Top shape: 32 4096 (131072)
I0428 14:42:25.899530 29440 net.cpp:137] Memory required for data: 172898048
I0428 14:42:25.899539 29440 layer_factory.hpp:77] Creating layer relu6
I0428 14:42:25.899549 29440 net.cpp:84] Creating Layer relu6
I0428 14:42:25.899554 29440 net.cpp:406] relu6 <- fc6
I0428 14:42:25.899561 29440 net.cpp:367] relu6 -> fc6 (in-place)
I0428 14:42:25.900246 29440 net.cpp:122] Setting up relu6
I0428 14:42:25.900255 29440 net.cpp:129] Top shape: 32 4096 (131072)
I0428 14:42:25.900259 29440 net.cpp:137] Memory required for data: 173422336
I0428 14:42:25.900264 29440 layer_factory.hpp:77] Creating layer drop6
I0428 14:42:25.900269 29440 net.cpp:84] Creating Layer drop6
I0428 14:42:25.900274 29440 net.cpp:406] drop6 <- fc6
I0428 14:42:25.900280 29440 net.cpp:367] drop6 -> fc6 (in-place)
I0428 14:42:25.900305 29440 net.cpp:122] Setting up drop6
I0428 14:42:25.900310 29440 net.cpp:129] Top shape: 32 4096 (131072)
I0428 14:42:25.900312 29440 net.cpp:137] Memory required for data: 173946624
I0428 14:42:25.900315 29440 layer_factory.hpp:77] Creating layer fc7
I0428 14:42:25.900324 29440 net.cpp:84] Creating Layer fc7
I0428 14:42:25.900328 29440 net.cpp:406] fc7 <- fc6
I0428 14:42:25.900334 29440 net.cpp:380] fc7 -> fc7
I0428 14:42:26.091887 29440 net.cpp:122] Setting up fc7
I0428 14:42:26.091910 29440 net.cpp:129] Top shape: 32 4096 (131072)
I0428 14:42:26.091915 29440 net.cpp:137] Memory required for data: 174470912
I0428 14:42:26.091924 29440 layer_factory.hpp:77] Creating layer relu7
I0428 14:42:26.091934 29440 net.cpp:84] Creating Layer relu7
I0428 14:42:26.091961 29440 net.cpp:406] relu7 <- fc7
I0428 14:42:26.091969 29440 net.cpp:367] relu7 -> fc7 (in-place)
I0428 14:42:26.092402 29440 net.cpp:122] Setting up relu7
I0428 14:42:26.092412 29440 net.cpp:129] Top shape: 32 4096 (131072)
I0428 14:42:26.092417 29440 net.cpp:137] Memory required for data: 174995200
I0428 14:42:26.092420 29440 layer_factory.hpp:77] Creating layer drop7
I0428 14:42:26.092427 29440 net.cpp:84] Creating Layer drop7
I0428 14:42:26.092432 29440 net.cpp:406] drop7 <- fc7
I0428 14:42:26.092438 29440 net.cpp:367] drop7 -> fc7 (in-place)
I0428 14:42:26.092461 29440 net.cpp:122] Setting up drop7
I0428 14:42:26.092468 29440 net.cpp:129] Top shape: 32 4096 (131072)
I0428 14:42:26.092470 29440 net.cpp:137] Memory required for data: 175519488
I0428 14:42:26.092474 29440 layer_factory.hpp:77] Creating layer fc8
I0428 14:42:26.092483 29440 net.cpp:84] Creating Layer fc8
I0428 14:42:26.092511 29440 net.cpp:406] fc8 <- fc7
I0428 14:42:26.092519 29440 net.cpp:380] fc8 -> fc8
I0428 14:42:26.100292 29440 net.cpp:122] Setting up fc8
I0428 14:42:26.100302 29440 net.cpp:129] Top shape: 32 196 (6272)
I0428 14:42:26.100306 29440 net.cpp:137] Memory required for data: 175544576
I0428 14:42:26.100318 29440 layer_factory.hpp:77] Creating layer fc8_fc8_0_split
I0428 14:42:26.100324 29440 net.cpp:84] Creating Layer fc8_fc8_0_split
I0428 14:42:26.100329 29440 net.cpp:406] fc8_fc8_0_split <- fc8
I0428 14:42:26.100334 29440 net.cpp:380] fc8_fc8_0_split -> fc8_fc8_0_split_0
I0428 14:42:26.100342 29440 net.cpp:380] fc8_fc8_0_split -> fc8_fc8_0_split_1
I0428 14:42:26.100375 29440 net.cpp:122] Setting up fc8_fc8_0_split
I0428 14:42:26.100381 29440 net.cpp:129] Top shape: 32 196 (6272)
I0428 14:42:26.100384 29440 net.cpp:129] Top shape: 32 196 (6272)
I0428 14:42:26.100387 29440 net.cpp:137] Memory required for data: 175594752
I0428 14:42:26.100391 29440 layer_factory.hpp:77] Creating layer accuracy
I0428 14:42:26.100399 29440 net.cpp:84] Creating Layer accuracy
I0428 14:42:26.100402 29440 net.cpp:406] accuracy <- fc8_fc8_0_split_0
I0428 14:42:26.100406 29440 net.cpp:406] accuracy <- label_val-data_1_split_0
I0428 14:42:26.100411 29440 net.cpp:380] accuracy -> accuracy
I0428 14:42:26.100419 29440 net.cpp:122] Setting up accuracy
I0428 14:42:26.100422 29440 net.cpp:129] Top shape: (1)
I0428 14:42:26.100425 29440 net.cpp:137] Memory required for data: 175594756
I0428 14:42:26.100430 29440 layer_factory.hpp:77] Creating layer loss
I0428 14:42:26.100438 29440 net.cpp:84] Creating Layer loss
I0428 14:42:26.100443 29440 net.cpp:406] loss <- fc8_fc8_0_split_1
I0428 14:42:26.100447 29440 net.cpp:406] loss <- label_val-data_1_split_1
I0428 14:42:26.100452 29440 net.cpp:380] loss -> loss
I0428 14:42:26.100459 29440 layer_factory.hpp:77] Creating layer loss
I0428 14:42:26.102097 29440 net.cpp:122] Setting up loss
I0428 14:42:26.102108 29440 net.cpp:129] Top shape: (1)
I0428 14:42:26.102110 29440 net.cpp:132]     with loss weight 1
I0428 14:42:26.102120 29440 net.cpp:137] Memory required for data: 175594760
I0428 14:42:26.102125 29440 net.cpp:198] loss needs backward computation.
I0428 14:42:26.102130 29440 net.cpp:200] accuracy does not need backward computation.
I0428 14:42:26.102135 29440 net.cpp:198] fc8_fc8_0_split needs backward computation.
I0428 14:42:26.102139 29440 net.cpp:198] fc8 needs backward computation.
I0428 14:42:26.102142 29440 net.cpp:198] drop7 needs backward computation.
I0428 14:42:26.102146 29440 net.cpp:198] relu7 needs backward computation.
I0428 14:42:26.102149 29440 net.cpp:198] fc7 needs backward computation.
I0428 14:42:26.102152 29440 net.cpp:198] drop6 needs backward computation.
I0428 14:42:26.102156 29440 net.cpp:198] relu6 needs backward computation.
I0428 14:42:26.102160 29440 net.cpp:198] fc6 needs backward computation.
I0428 14:42:26.102167 29440 net.cpp:198] pool5 needs backward computation.
I0428 14:42:26.102171 29440 net.cpp:198] relu5 needs backward computation.
I0428 14:42:26.102175 29440 net.cpp:198] conv5 needs backward computation.
I0428 14:42:26.102178 29440 net.cpp:198] relu4 needs backward computation.
I0428 14:42:26.102193 29440 net.cpp:198] conv4 needs backward computation.
I0428 14:42:26.102197 29440 net.cpp:198] relu3 needs backward computation.
I0428 14:42:26.102200 29440 net.cpp:198] conv3 needs backward computation.
I0428 14:42:26.102205 29440 net.cpp:198] pool2 needs backward computation.
I0428 14:42:26.102208 29440 net.cpp:198] norm2 needs backward computation.
I0428 14:42:26.102211 29440 net.cpp:198] relu2 needs backward computation.
I0428 14:42:26.102216 29440 net.cpp:198] conv2 needs backward computation.
I0428 14:42:26.102218 29440 net.cpp:198] pool1.5 needs backward computation.
I0428 14:42:26.102222 29440 net.cpp:198] norm1.5 needs backward computation.
I0428 14:42:26.102226 29440 net.cpp:198] relu1.5 needs backward computation.
I0428 14:42:26.102229 29440 net.cpp:198] conv1.5 needs backward computation.
I0428 14:42:26.102233 29440 net.cpp:198] pool1 needs backward computation.
I0428 14:42:26.102237 29440 net.cpp:198] norm1 needs backward computation.
I0428 14:42:26.102241 29440 net.cpp:198] relu1 needs backward computation.
I0428 14:42:26.102244 29440 net.cpp:198] conv1 needs backward computation.
I0428 14:42:26.102248 29440 net.cpp:200] label_val-data_1_split does not need backward computation.
I0428 14:42:26.102253 29440 net.cpp:200] val-data does not need backward computation.
I0428 14:42:26.102257 29440 net.cpp:242] This network produces output accuracy
I0428 14:42:26.102262 29440 net.cpp:242] This network produces output loss
I0428 14:42:26.102279 29440 net.cpp:255] Network initialization done.
I0428 14:42:26.102360 29440 solver.cpp:56] Solver scaffolding done.
I0428 14:42:26.102859 29440 caffe.cpp:248] Starting Optimization
I0428 14:42:26.102866 29440 solver.cpp:272] Solving
I0428 14:42:26.102870 29440 solver.cpp:273] Learning Rate Policy: exp
I0428 14:42:26.104298 29440 solver.cpp:330] Iteration 0, Testing net (#0)
I0428 14:42:26.104310 29440 net.cpp:676] Ignoring source layer train-data
I0428 14:42:26.182860 29440 blocking_queue.cpp:49] Waiting for data
I0428 14:42:30.527220 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:42:30.573175 29440 solver.cpp:397]     Test net output #0: accuracy = 0.00735294
I0428 14:42:30.573217 29440 solver.cpp:397]     Test net output #1: loss = 5.28137 (* 1 = 5.28137 loss)
I0428 14:42:30.759508 29440 solver.cpp:218] Iteration 0 (-2.03172e-39 iter/s, 4.6564s/12 iters), loss = 5.28307
I0428 14:42:30.761025 29440 solver.cpp:237]     Train net output #0: loss = 5.28307 (* 1 = 5.28307 loss)
I0428 14:42:30.761049 29440 sgd_solver.cpp:105] Iteration 0, lr = 0.01
I0428 14:42:35.225829 29440 solver.cpp:218] Iteration 12 (2.6878 iter/s, 4.46461s/12 iters), loss = 5.28102
I0428 14:42:35.225874 29440 solver.cpp:237]     Train net output #0: loss = 5.28102 (* 1 = 5.28102 loss)
I0428 14:42:35.225884 29440 sgd_solver.cpp:105] Iteration 12, lr = 0.00997626
I0428 14:42:40.893496 29440 solver.cpp:218] Iteration 24 (2.11738 iter/s, 5.66738s/12 iters), loss = 5.27741
I0428 14:42:40.893534 29440 solver.cpp:237]     Train net output #0: loss = 5.27741 (* 1 = 5.27741 loss)
I0428 14:42:40.893544 29440 sgd_solver.cpp:105] Iteration 24, lr = 0.00995257
I0428 14:42:46.538396 29440 solver.cpp:218] Iteration 36 (2.12592 iter/s, 5.64462s/12 iters), loss = 5.31417
I0428 14:42:46.538436 29440 solver.cpp:237]     Train net output #0: loss = 5.31417 (* 1 = 5.31417 loss)
I0428 14:42:46.538444 29440 sgd_solver.cpp:105] Iteration 36, lr = 0.00992894
I0428 14:42:52.188529 29440 solver.cpp:218] Iteration 48 (2.12395 iter/s, 5.64985s/12 iters), loss = 5.27131
I0428 14:42:52.188573 29440 solver.cpp:237]     Train net output #0: loss = 5.27131 (* 1 = 5.27131 loss)
I0428 14:42:52.188582 29440 sgd_solver.cpp:105] Iteration 48, lr = 0.00990537
I0428 14:42:57.926990 29440 solver.cpp:218] Iteration 60 (2.09126 iter/s, 5.73817s/12 iters), loss = 5.29881
I0428 14:42:57.927127 29440 solver.cpp:237]     Train net output #0: loss = 5.29881 (* 1 = 5.29881 loss)
I0428 14:42:57.927137 29440 sgd_solver.cpp:105] Iteration 60, lr = 0.00988185
I0428 14:43:03.687633 29440 solver.cpp:218] Iteration 72 (2.08324 iter/s, 5.76026s/12 iters), loss = 5.27587
I0428 14:43:03.687674 29440 solver.cpp:237]     Train net output #0: loss = 5.27587 (* 1 = 5.27587 loss)
I0428 14:43:03.687682 29440 sgd_solver.cpp:105] Iteration 72, lr = 0.00985839
I0428 14:43:09.349583 29440 solver.cpp:218] Iteration 84 (2.11952 iter/s, 5.66167s/12 iters), loss = 5.29278
I0428 14:43:09.349622 29440 solver.cpp:237]     Train net output #0: loss = 5.29278 (* 1 = 5.29278 loss)
I0428 14:43:09.349632 29440 sgd_solver.cpp:105] Iteration 84, lr = 0.00983498
I0428 14:43:15.035012 29440 solver.cpp:218] Iteration 96 (2.11076 iter/s, 5.68515s/12 iters), loss = 5.29984
I0428 14:43:15.035050 29440 solver.cpp:237]     Train net output #0: loss = 5.29984 (* 1 = 5.29984 loss)
I0428 14:43:15.035058 29440 sgd_solver.cpp:105] Iteration 96, lr = 0.00981163
I0428 14:43:16.848696 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:43:17.159327 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_102.caffemodel
I0428 14:43:20.480465 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_102.solverstate
I0428 14:43:22.405184 29440 solver.cpp:330] Iteration 102, Testing net (#0)
I0428 14:43:22.405211 29440 net.cpp:676] Ignoring source layer train-data
I0428 14:43:27.002794 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:43:27.091076 29440 solver.cpp:397]     Test net output #0: accuracy = 0.00245098
I0428 14:43:27.091117 29440 solver.cpp:397]     Test net output #1: loss = 5.29061 (* 1 = 5.29061 loss)
I0428 14:43:29.074126 29440 solver.cpp:218] Iteration 108 (0.854792 iter/s, 14.0385s/12 iters), loss = 5.28302
I0428 14:43:29.074246 29440 solver.cpp:237]     Train net output #0: loss = 5.28302 (* 1 = 5.28302 loss)
I0428 14:43:29.074257 29440 sgd_solver.cpp:105] Iteration 108, lr = 0.00978834
I0428 14:43:34.784106 29440 solver.cpp:218] Iteration 120 (2.10172 iter/s, 5.70962s/12 iters), loss = 5.28224
I0428 14:43:34.784145 29440 solver.cpp:237]     Train net output #0: loss = 5.28224 (* 1 = 5.28224 loss)
I0428 14:43:34.784153 29440 sgd_solver.cpp:105] Iteration 120, lr = 0.0097651
I0428 14:43:40.425812 29440 solver.cpp:218] Iteration 132 (2.12712 iter/s, 5.64143s/12 iters), loss = 5.28081
I0428 14:43:40.425858 29440 solver.cpp:237]     Train net output #0: loss = 5.28081 (* 1 = 5.28081 loss)
I0428 14:43:40.425866 29440 sgd_solver.cpp:105] Iteration 132, lr = 0.00974192
I0428 14:43:46.153666 29440 solver.cpp:218] Iteration 144 (2.09513 iter/s, 5.72756s/12 iters), loss = 5.29211
I0428 14:43:46.153707 29440 solver.cpp:237]     Train net output #0: loss = 5.29211 (* 1 = 5.29211 loss)
I0428 14:43:46.153717 29440 sgd_solver.cpp:105] Iteration 144, lr = 0.00971879
I0428 14:43:51.802469 29440 solver.cpp:218] Iteration 156 (2.12445 iter/s, 5.64852s/12 iters), loss = 5.30306
I0428 14:43:51.802508 29440 solver.cpp:237]     Train net output #0: loss = 5.30306 (* 1 = 5.30306 loss)
I0428 14:43:51.802517 29440 sgd_solver.cpp:105] Iteration 156, lr = 0.00969571
I0428 14:43:57.442842 29440 solver.cpp:218] Iteration 168 (2.12762 iter/s, 5.6401s/12 iters), loss = 5.29719
I0428 14:43:57.442883 29440 solver.cpp:237]     Train net output #0: loss = 5.29719 (* 1 = 5.29719 loss)
I0428 14:43:57.442893 29440 sgd_solver.cpp:105] Iteration 168, lr = 0.00967269
I0428 14:44:03.110077 29440 solver.cpp:218] Iteration 180 (2.11754 iter/s, 5.66695s/12 iters), loss = 5.29638
I0428 14:44:03.110186 29440 solver.cpp:237]     Train net output #0: loss = 5.29638 (* 1 = 5.29638 loss)
I0428 14:44:03.110195 29440 sgd_solver.cpp:105] Iteration 180, lr = 0.00964973
I0428 14:44:08.765861 29440 solver.cpp:218] Iteration 192 (2.12185 iter/s, 5.65543s/12 iters), loss = 5.2732
I0428 14:44:08.765903 29440 solver.cpp:237]     Train net output #0: loss = 5.2732 (* 1 = 5.2732 loss)
I0428 14:44:08.765913 29440 sgd_solver.cpp:105] Iteration 192, lr = 0.00962682
I0428 14:44:13.144817 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:44:13.887392 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_204.caffemodel
I0428 14:44:16.472442 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_204.solverstate
I0428 14:44:18.698989 29440 solver.cpp:330] Iteration 204, Testing net (#0)
I0428 14:44:18.699013 29440 net.cpp:676] Ignoring source layer train-data
I0428 14:44:23.141250 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:44:23.268862 29440 solver.cpp:397]     Test net output #0: accuracy = 0.00551471
I0428 14:44:23.268889 29440 solver.cpp:397]     Test net output #1: loss = 5.288 (* 1 = 5.288 loss)
I0428 14:44:23.425982 29440 solver.cpp:218] Iteration 204 (0.818583 iter/s, 14.6595s/12 iters), loss = 5.27241
I0428 14:44:23.426023 29440 solver.cpp:237]     Train net output #0: loss = 5.27241 (* 1 = 5.27241 loss)
I0428 14:44:23.426033 29440 sgd_solver.cpp:105] Iteration 204, lr = 0.00960396
I0428 14:44:28.149768 29440 solver.cpp:218] Iteration 216 (2.54047 iter/s, 4.72354s/12 iters), loss = 5.28746
I0428 14:44:28.149811 29440 solver.cpp:237]     Train net output #0: loss = 5.28746 (* 1 = 5.28746 loss)
I0428 14:44:28.149819 29440 sgd_solver.cpp:105] Iteration 216, lr = 0.00958116
I0428 14:44:33.693882 29440 solver.cpp:218] Iteration 228 (2.16457 iter/s, 5.54384s/12 iters), loss = 5.27737
I0428 14:44:33.694033 29440 solver.cpp:237]     Train net output #0: loss = 5.27737 (* 1 = 5.27737 loss)
I0428 14:44:33.694043 29440 sgd_solver.cpp:105] Iteration 228, lr = 0.00955841
I0428 14:44:39.384946 29440 solver.cpp:218] Iteration 240 (2.10871 iter/s, 5.69067s/12 iters), loss = 5.26787
I0428 14:44:39.384984 29440 solver.cpp:237]     Train net output #0: loss = 5.26787 (* 1 = 5.26787 loss)
I0428 14:44:39.384991 29440 sgd_solver.cpp:105] Iteration 240, lr = 0.00953572
I0428 14:44:44.861624 29440 solver.cpp:218] Iteration 252 (2.19122 iter/s, 5.4764s/12 iters), loss = 5.29207
I0428 14:44:44.861676 29440 solver.cpp:237]     Train net output #0: loss = 5.29207 (* 1 = 5.29207 loss)
I0428 14:44:44.861687 29440 sgd_solver.cpp:105] Iteration 252, lr = 0.00951308
I0428 14:44:50.519878 29440 solver.cpp:218] Iteration 264 (2.1209 iter/s, 5.65796s/12 iters), loss = 5.27011
I0428 14:44:50.519917 29440 solver.cpp:237]     Train net output #0: loss = 5.27011 (* 1 = 5.27011 loss)
I0428 14:44:50.519925 29440 sgd_solver.cpp:105] Iteration 264, lr = 0.00949049
I0428 14:44:56.085287 29440 solver.cpp:218] Iteration 276 (2.15628 iter/s, 5.56513s/12 iters), loss = 5.26378
I0428 14:44:56.085330 29440 solver.cpp:237]     Train net output #0: loss = 5.26378 (* 1 = 5.26378 loss)
I0428 14:44:56.085340 29440 sgd_solver.cpp:105] Iteration 276, lr = 0.00946796
I0428 14:45:01.762146 29440 solver.cpp:218] Iteration 288 (2.11395 iter/s, 5.67657s/12 iters), loss = 5.29806
I0428 14:45:01.762187 29440 solver.cpp:237]     Train net output #0: loss = 5.29806 (* 1 = 5.29806 loss)
I0428 14:45:01.762197 29440 sgd_solver.cpp:105] Iteration 288, lr = 0.00944548
I0428 14:45:07.213016 29440 solver.cpp:218] Iteration 300 (2.20159 iter/s, 5.4506s/12 iters), loss = 5.26055
I0428 14:45:07.213119 29440 solver.cpp:237]     Train net output #0: loss = 5.26055 (* 1 = 5.26055 loss)
I0428 14:45:07.213129 29440 sgd_solver.cpp:105] Iteration 300, lr = 0.00942305
I0428 14:45:08.302124 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:45:09.454792 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_306.caffemodel
I0428 14:45:15.442564 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_306.solverstate
I0428 14:45:17.476675 29440 solver.cpp:330] Iteration 306, Testing net (#0)
I0428 14:45:17.476701 29440 net.cpp:676] Ignoring source layer train-data
I0428 14:45:21.826122 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:45:22.026845 29440 solver.cpp:397]     Test net output #0: accuracy = 0.0104167
I0428 14:45:22.026873 29440 solver.cpp:397]     Test net output #1: loss = 5.21928 (* 1 = 5.21928 loss)
I0428 14:45:23.968535 29440 solver.cpp:218] Iteration 312 (0.716216 iter/s, 16.7547s/12 iters), loss = 5.24413
I0428 14:45:23.968583 29440 solver.cpp:237]     Train net output #0: loss = 5.24413 (* 1 = 5.24413 loss)
I0428 14:45:23.968592 29440 sgd_solver.cpp:105] Iteration 312, lr = 0.00940068
I0428 14:45:29.627801 29440 solver.cpp:218] Iteration 324 (2.12052 iter/s, 5.65898s/12 iters), loss = 5.25302
I0428 14:45:29.627846 29440 solver.cpp:237]     Train net output #0: loss = 5.25302 (* 1 = 5.25302 loss)
I0428 14:45:29.627856 29440 sgd_solver.cpp:105] Iteration 324, lr = 0.00937836
I0428 14:45:35.196993 29440 solver.cpp:218] Iteration 336 (2.15482 iter/s, 5.56891s/12 iters), loss = 5.23656
I0428 14:45:35.197039 29440 solver.cpp:237]     Train net output #0: loss = 5.23656 (* 1 = 5.23656 loss)
I0428 14:45:35.197048 29440 sgd_solver.cpp:105] Iteration 336, lr = 0.0093561
I0428 14:45:40.726547 29440 solver.cpp:218] Iteration 348 (2.17027 iter/s, 5.52927s/12 iters), loss = 5.21322
I0428 14:45:40.726678 29440 solver.cpp:237]     Train net output #0: loss = 5.21322 (* 1 = 5.21322 loss)
I0428 14:45:40.726688 29440 sgd_solver.cpp:105] Iteration 348, lr = 0.00933388
I0428 14:45:46.408305 29440 solver.cpp:218] Iteration 360 (2.11216 iter/s, 5.68138s/12 iters), loss = 5.16693
I0428 14:45:46.408361 29440 solver.cpp:237]     Train net output #0: loss = 5.16693 (* 1 = 5.16693 loss)
I0428 14:45:46.408375 29440 sgd_solver.cpp:105] Iteration 360, lr = 0.00931172
I0428 14:45:51.783569 29440 solver.cpp:218] Iteration 372 (2.23257 iter/s, 5.37498s/12 iters), loss = 5.21606
I0428 14:45:51.783612 29440 solver.cpp:237]     Train net output #0: loss = 5.21606 (* 1 = 5.21606 loss)
I0428 14:45:51.783622 29440 sgd_solver.cpp:105] Iteration 372, lr = 0.00928961
I0428 14:45:57.526006 29440 solver.cpp:218] Iteration 384 (2.08981 iter/s, 5.74215s/12 iters), loss = 5.18351
I0428 14:45:57.526051 29440 solver.cpp:237]     Train net output #0: loss = 5.18351 (* 1 = 5.18351 loss)
I0428 14:45:57.526062 29440 sgd_solver.cpp:105] Iteration 384, lr = 0.00926756
I0428 14:46:03.187644 29440 solver.cpp:218] Iteration 396 (2.11964 iter/s, 5.66135s/12 iters), loss = 5.14124
I0428 14:46:03.187687 29440 solver.cpp:237]     Train net output #0: loss = 5.14124 (* 1 = 5.14124 loss)
I0428 14:46:03.187696 29440 sgd_solver.cpp:105] Iteration 396, lr = 0.00924556
I0428 14:46:06.688935 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:46:08.258843 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_408.caffemodel
I0428 14:46:11.645036 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_408.solverstate
I0428 14:46:16.407445 29440 solver.cpp:330] Iteration 408, Testing net (#0)
I0428 14:46:16.407469 29440 net.cpp:676] Ignoring source layer train-data
I0428 14:46:21.122082 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:46:21.330811 29440 solver.cpp:397]     Test net output #0: accuracy = 0.00980392
I0428 14:46:21.330840 29440 solver.cpp:397]     Test net output #1: loss = 5.1692 (* 1 = 5.1692 loss)
I0428 14:46:21.479007 29440 solver.cpp:218] Iteration 408 (0.656075 iter/s, 18.2906s/12 iters), loss = 5.21766
I0428 14:46:21.479049 29440 solver.cpp:237]     Train net output #0: loss = 5.21766 (* 1 = 5.21766 loss)
I0428 14:46:21.479058 29440 sgd_solver.cpp:105] Iteration 408, lr = 0.00922361
I0428 14:46:26.067975 29440 solver.cpp:218] Iteration 420 (2.61511 iter/s, 4.58872s/12 iters), loss = 5.05101
I0428 14:46:26.068020 29440 solver.cpp:237]     Train net output #0: loss = 5.05101 (* 1 = 5.05101 loss)
I0428 14:46:26.068029 29440 sgd_solver.cpp:105] Iteration 420, lr = 0.00920171
I0428 14:46:31.728574 29440 solver.cpp:218] Iteration 432 (2.12003 iter/s, 5.66031s/12 iters), loss = 5.03242
I0428 14:46:31.728634 29440 solver.cpp:237]     Train net output #0: loss = 5.03242 (* 1 = 5.03242 loss)
I0428 14:46:31.728646 29440 sgd_solver.cpp:105] Iteration 432, lr = 0.00917986
I0428 14:46:37.242877 29440 solver.cpp:218] Iteration 444 (2.17627 iter/s, 5.51401s/12 iters), loss = 5.15427
I0428 14:46:37.242919 29440 solver.cpp:237]     Train net output #0: loss = 5.15427 (* 1 = 5.15427 loss)
I0428 14:46:37.242928 29440 sgd_solver.cpp:105] Iteration 444, lr = 0.00915807
I0428 14:46:42.972712 29440 solver.cpp:218] Iteration 456 (2.09441 iter/s, 5.72955s/12 iters), loss = 5.13772
I0428 14:46:42.972862 29440 solver.cpp:237]     Train net output #0: loss = 5.13772 (* 1 = 5.13772 loss)
I0428 14:46:42.972873 29440 sgd_solver.cpp:105] Iteration 456, lr = 0.00913632
I0428 14:46:48.536985 29440 solver.cpp:218] Iteration 468 (2.15677 iter/s, 5.56388s/12 iters), loss = 5.13441
I0428 14:46:48.537041 29440 solver.cpp:237]     Train net output #0: loss = 5.13441 (* 1 = 5.13441 loss)
I0428 14:46:48.537053 29440 sgd_solver.cpp:105] Iteration 468, lr = 0.00911463
I0428 14:46:54.170359 29440 solver.cpp:218] Iteration 480 (2.13027 iter/s, 5.63308s/12 iters), loss = 5.09039
I0428 14:46:54.170403 29440 solver.cpp:237]     Train net output #0: loss = 5.09039 (* 1 = 5.09039 loss)
I0428 14:46:54.170413 29440 sgd_solver.cpp:105] Iteration 480, lr = 0.00909299
I0428 14:46:59.804162 29440 solver.cpp:218] Iteration 492 (2.13011 iter/s, 5.63352s/12 iters), loss = 5.16206
I0428 14:46:59.804201 29440 solver.cpp:237]     Train net output #0: loss = 5.16206 (* 1 = 5.16206 loss)
I0428 14:46:59.804209 29440 sgd_solver.cpp:105] Iteration 492, lr = 0.0090714
I0428 14:47:05.507405 29440 solver.cpp:218] Iteration 504 (2.10417 iter/s, 5.70297s/12 iters), loss = 5.12331
I0428 14:47:05.507443 29440 solver.cpp:237]     Train net output #0: loss = 5.12331 (* 1 = 5.12331 loss)
I0428 14:47:05.507452 29440 sgd_solver.cpp:105] Iteration 504, lr = 0.00904986
I0428 14:47:05.784436 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:47:07.923110 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_510.caffemodel
I0428 14:47:10.206712 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_510.solverstate
I0428 14:47:12.043633 29440 solver.cpp:330] Iteration 510, Testing net (#0)
I0428 14:47:12.043656 29440 net.cpp:676] Ignoring source layer train-data
I0428 14:47:16.504266 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:47:16.777420 29440 solver.cpp:397]     Test net output #0: accuracy = 0.0134804
I0428 14:47:16.777464 29440 solver.cpp:397]     Test net output #1: loss = 5.10843 (* 1 = 5.10843 loss)
I0428 14:47:18.923414 29440 solver.cpp:218] Iteration 516 (0.894493 iter/s, 13.4154s/12 iters), loss = 5.01744
I0428 14:47:18.923461 29440 solver.cpp:237]     Train net output #0: loss = 5.01744 (* 1 = 5.01744 loss)
I0428 14:47:18.923472 29440 sgd_solver.cpp:105] Iteration 516, lr = 0.00902838
I0428 14:47:24.467445 29440 solver.cpp:218] Iteration 528 (2.1646 iter/s, 5.54375s/12 iters), loss = 5.08039
I0428 14:47:24.467486 29440 solver.cpp:237]     Train net output #0: loss = 5.08039 (* 1 = 5.08039 loss)
I0428 14:47:24.467496 29440 sgd_solver.cpp:105] Iteration 528, lr = 0.00900694
I0428 14:47:30.101410 29440 solver.cpp:218] Iteration 540 (2.13005 iter/s, 5.63368s/12 iters), loss = 5.0977
I0428 14:47:30.101455 29440 solver.cpp:237]     Train net output #0: loss = 5.0977 (* 1 = 5.0977 loss)
I0428 14:47:30.101464 29440 sgd_solver.cpp:105] Iteration 540, lr = 0.00898556
I0428 14:47:35.852229 29440 solver.cpp:218] Iteration 552 (2.08676 iter/s, 5.75053s/12 iters), loss = 5.17785
I0428 14:47:35.852274 29440 solver.cpp:237]     Train net output #0: loss = 5.17785 (* 1 = 5.17785 loss)
I0428 14:47:35.852283 29440 sgd_solver.cpp:105] Iteration 552, lr = 0.00896423
I0428 14:47:41.600181 29440 solver.cpp:218] Iteration 564 (2.08781 iter/s, 5.74766s/12 iters), loss = 5.1459
I0428 14:47:41.600221 29440 solver.cpp:237]     Train net output #0: loss = 5.1459 (* 1 = 5.1459 loss)
I0428 14:47:41.600229 29440 sgd_solver.cpp:105] Iteration 564, lr = 0.00894294
I0428 14:47:47.021204 29440 solver.cpp:218] Iteration 576 (2.21371 iter/s, 5.42075s/12 iters), loss = 5.08442
I0428 14:47:47.021428 29440 solver.cpp:237]     Train net output #0: loss = 5.08442 (* 1 = 5.08442 loss)
I0428 14:47:47.021440 29440 sgd_solver.cpp:105] Iteration 576, lr = 0.00892171
I0428 14:47:52.605573 29440 solver.cpp:218] Iteration 588 (2.14903 iter/s, 5.5839s/12 iters), loss = 5.04719
I0428 14:47:52.605626 29440 solver.cpp:237]     Train net output #0: loss = 5.04719 (* 1 = 5.04719 loss)
I0428 14:47:52.605638 29440 sgd_solver.cpp:105] Iteration 588, lr = 0.00890053
I0428 14:47:58.203536 29440 solver.cpp:218] Iteration 600 (2.14375 iter/s, 5.59767s/12 iters), loss = 5.12273
I0428 14:47:58.203591 29440 solver.cpp:237]     Train net output #0: loss = 5.12273 (* 1 = 5.12273 loss)
I0428 14:47:58.203606 29440 sgd_solver.cpp:105] Iteration 600, lr = 0.0088794
I0428 14:48:00.953759 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:48:03.530426 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_612.caffemodel
I0428 14:48:07.002296 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_612.solverstate
I0428 14:48:08.087504 29440 solver.cpp:330] Iteration 612, Testing net (#0)
I0428 14:48:08.087525 29440 net.cpp:676] Ignoring source layer train-data
I0428 14:48:12.385051 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:48:12.695477 29440 solver.cpp:397]     Test net output #0: accuracy = 0.00857843
I0428 14:48:12.695525 29440 solver.cpp:397]     Test net output #1: loss = 5.08582 (* 1 = 5.08582 loss)
I0428 14:48:12.856402 29440 solver.cpp:218] Iteration 612 (0.818989 iter/s, 14.6522s/12 iters), loss = 5.06397
I0428 14:48:12.856467 29440 solver.cpp:237]     Train net output #0: loss = 5.06397 (* 1 = 5.06397 loss)
I0428 14:48:12.856477 29440 sgd_solver.cpp:105] Iteration 612, lr = 0.00885831
I0428 14:48:17.736678 29440 solver.cpp:218] Iteration 624 (2.45902 iter/s, 4.88s/12 iters), loss = 5.18759
I0428 14:48:17.736788 29440 solver.cpp:237]     Train net output #0: loss = 5.18759 (* 1 = 5.18759 loss)
I0428 14:48:17.736799 29440 sgd_solver.cpp:105] Iteration 624, lr = 0.00883728
I0428 14:48:23.317076 29440 solver.cpp:218] Iteration 636 (2.15052 iter/s, 5.58004s/12 iters), loss = 5.14763
I0428 14:48:23.317155 29440 solver.cpp:237]     Train net output #0: loss = 5.14763 (* 1 = 5.14763 loss)
I0428 14:48:23.317167 29440 sgd_solver.cpp:105] Iteration 636, lr = 0.0088163
I0428 14:48:28.896629 29440 solver.cpp:218] Iteration 648 (2.15083 iter/s, 5.57924s/12 iters), loss = 5.10881
I0428 14:48:28.896677 29440 solver.cpp:237]     Train net output #0: loss = 5.10881 (* 1 = 5.10881 loss)
I0428 14:48:28.896685 29440 sgd_solver.cpp:105] Iteration 648, lr = 0.00879537
I0428 14:48:34.588086 29440 solver.cpp:218] Iteration 660 (2.10853 iter/s, 5.69117s/12 iters), loss = 5.12022
I0428 14:48:34.588130 29440 solver.cpp:237]     Train net output #0: loss = 5.12022 (* 1 = 5.12022 loss)
I0428 14:48:34.588138 29440 sgd_solver.cpp:105] Iteration 660, lr = 0.00877449
I0428 14:48:40.320586 29440 solver.cpp:218] Iteration 672 (2.09343 iter/s, 5.73221s/12 iters), loss = 5.08137
I0428 14:48:40.320628 29440 solver.cpp:237]     Train net output #0: loss = 5.08137 (* 1 = 5.08137 loss)
I0428 14:48:40.320638 29440 sgd_solver.cpp:105] Iteration 672, lr = 0.00875366
I0428 14:48:45.219853 29440 solver.cpp:218] Iteration 684 (2.45059 iter/s, 4.89677s/12 iters), loss = 5.02825
I0428 14:48:45.219895 29440 solver.cpp:237]     Train net output #0: loss = 5.02825 (* 1 = 5.02825 loss)
I0428 14:48:45.219903 29440 sgd_solver.cpp:105] Iteration 684, lr = 0.00873287
I0428 14:48:45.807016 29440 blocking_queue.cpp:49] Waiting for data
I0428 14:48:50.100924 29440 solver.cpp:218] Iteration 696 (2.45972 iter/s, 4.87861s/12 iters), loss = 5.06679
I0428 14:48:50.102274 29440 solver.cpp:237]     Train net output #0: loss = 5.06679 (* 1 = 5.06679 loss)
I0428 14:48:50.102284 29440 sgd_solver.cpp:105] Iteration 696, lr = 0.00871214
I0428 14:48:54.548887 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:48:55.151087 29440 solver.cpp:218] Iteration 708 (2.37731 iter/s, 5.04772s/12 iters), loss = 5.01713
I0428 14:48:55.151129 29440 solver.cpp:237]     Train net output #0: loss = 5.01713 (* 1 = 5.01713 loss)
I0428 14:48:55.151136 29440 sgd_solver.cpp:105] Iteration 708, lr = 0.00869145
I0428 14:48:57.333667 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_714.caffemodel
I0428 14:49:02.600685 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_714.solverstate
I0428 14:49:03.947074 29440 solver.cpp:330] Iteration 714, Testing net (#0)
I0428 14:49:03.947098 29440 net.cpp:676] Ignoring source layer train-data
I0428 14:49:08.775168 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:49:09.130777 29440 solver.cpp:397]     Test net output #0: accuracy = 0.0110294
I0428 14:49:09.130805 29440 solver.cpp:397]     Test net output #1: loss = 5.07846 (* 1 = 5.07846 loss)
I0428 14:49:11.020318 29440 solver.cpp:218] Iteration 720 (0.756213 iter/s, 15.8685s/12 iters), loss = 5.03251
I0428 14:49:11.020368 29440 solver.cpp:237]     Train net output #0: loss = 5.03251 (* 1 = 5.03251 loss)
I0428 14:49:11.020380 29440 sgd_solver.cpp:105] Iteration 720, lr = 0.00867082
I0428 14:49:16.805822 29440 solver.cpp:218] Iteration 732 (2.07425 iter/s, 5.78521s/12 iters), loss = 5.07949
I0428 14:49:16.805864 29440 solver.cpp:237]     Train net output #0: loss = 5.07949 (* 1 = 5.07949 loss)
I0428 14:49:16.805873 29440 sgd_solver.cpp:105] Iteration 732, lr = 0.00865023
I0428 14:49:22.002014 29440 solver.cpp:218] Iteration 744 (2.31047 iter/s, 5.19374s/12 iters), loss = 4.95759
I0428 14:49:22.002130 29440 solver.cpp:237]     Train net output #0: loss = 4.95759 (* 1 = 4.95759 loss)
I0428 14:49:22.002140 29440 sgd_solver.cpp:105] Iteration 744, lr = 0.0086297
I0428 14:49:27.228243 29440 solver.cpp:218] Iteration 756 (2.29719 iter/s, 5.22377s/12 iters), loss = 5.03546
I0428 14:49:27.228281 29440 solver.cpp:237]     Train net output #0: loss = 5.03546 (* 1 = 5.03546 loss)
I0428 14:49:27.228289 29440 sgd_solver.cpp:105] Iteration 756, lr = 0.00860921
I0428 14:49:32.899675 29440 solver.cpp:218] Iteration 768 (2.11597 iter/s, 5.67116s/12 iters), loss = 5.09208
I0428 14:49:32.899720 29440 solver.cpp:237]     Train net output #0: loss = 5.09208 (* 1 = 5.09208 loss)
I0428 14:49:32.899729 29440 sgd_solver.cpp:105] Iteration 768, lr = 0.00858877
I0428 14:49:38.207674 29440 solver.cpp:218] Iteration 780 (2.26179 iter/s, 5.30553s/12 iters), loss = 4.95952
I0428 14:49:38.207712 29440 solver.cpp:237]     Train net output #0: loss = 4.95952 (* 1 = 4.95952 loss)
I0428 14:49:38.207720 29440 sgd_solver.cpp:105] Iteration 780, lr = 0.00856838
I0428 14:49:43.540330 29440 solver.cpp:218] Iteration 792 (2.25134 iter/s, 5.33016s/12 iters), loss = 4.93161
I0428 14:49:43.540385 29440 solver.cpp:237]     Train net output #0: loss = 4.93161 (* 1 = 4.93161 loss)
I0428 14:49:43.540397 29440 sgd_solver.cpp:105] Iteration 792, lr = 0.00854803
I0428 14:49:48.874619 29440 solver.cpp:218] Iteration 804 (2.24971 iter/s, 5.33401s/12 iters), loss = 5.05733
I0428 14:49:48.874662 29440 solver.cpp:237]     Train net output #0: loss = 5.05733 (* 1 = 5.05733 loss)
I0428 14:49:48.874671 29440 sgd_solver.cpp:105] Iteration 804, lr = 0.00852774
I0428 14:49:50.737067 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:49:53.907409 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_816.caffemodel
I0428 14:49:55.878547 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_816.solverstate
I0428 14:49:58.560474 29440 solver.cpp:330] Iteration 816, Testing net (#0)
I0428 14:49:58.560520 29440 net.cpp:676] Ignoring source layer train-data
I0428 14:50:03.196460 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:50:03.583698 29440 solver.cpp:397]     Test net output #0: accuracy = 0.0183824
I0428 14:50:03.583735 29440 solver.cpp:397]     Test net output #1: loss = 5.04771 (* 1 = 5.04771 loss)
I0428 14:50:03.871599 29440 solver.cpp:218] Iteration 816 (0.800195 iter/s, 14.9963s/12 iters), loss = 4.99
I0428 14:50:03.871668 29440 solver.cpp:237]     Train net output #0: loss = 4.99 (* 1 = 4.99 loss)
I0428 14:50:03.871678 29440 sgd_solver.cpp:105] Iteration 816, lr = 0.00850749
I0428 14:50:08.742547 29440 solver.cpp:218] Iteration 828 (2.46372 iter/s, 4.87068s/12 iters), loss = 5.04737
I0428 14:50:08.742590 29440 solver.cpp:237]     Train net output #0: loss = 5.04737 (* 1 = 5.04737 loss)
I0428 14:50:08.742600 29440 sgd_solver.cpp:105] Iteration 828, lr = 0.00848729
I0428 14:50:14.341953 29440 solver.cpp:218] Iteration 840 (2.14319 iter/s, 5.59913s/12 iters), loss = 4.93996
I0428 14:50:14.341997 29440 solver.cpp:237]     Train net output #0: loss = 4.93996 (* 1 = 4.93996 loss)
I0428 14:50:14.342007 29440 sgd_solver.cpp:105] Iteration 840, lr = 0.00846714
I0428 14:50:19.530912 29440 solver.cpp:218] Iteration 852 (2.31272 iter/s, 5.1887s/12 iters), loss = 4.99836
I0428 14:50:19.530953 29440 solver.cpp:237]     Train net output #0: loss = 4.99836 (* 1 = 4.99836 loss)
I0428 14:50:19.530962 29440 sgd_solver.cpp:105] Iteration 852, lr = 0.00844704
I0428 14:50:24.782210 29440 solver.cpp:218] Iteration 864 (2.28526 iter/s, 5.25104s/12 iters), loss = 5.03346
I0428 14:50:24.782339 29440 solver.cpp:237]     Train net output #0: loss = 5.03346 (* 1 = 5.03346 loss)
I0428 14:50:24.782351 29440 sgd_solver.cpp:105] Iteration 864, lr = 0.00842698
I0428 14:50:29.806340 29440 solver.cpp:218] Iteration 876 (2.38864 iter/s, 5.02379s/12 iters), loss = 4.94084
I0428 14:50:29.806391 29440 solver.cpp:237]     Train net output #0: loss = 4.94084 (* 1 = 4.94084 loss)
I0428 14:50:29.806406 29440 sgd_solver.cpp:105] Iteration 876, lr = 0.00840698
I0428 14:50:35.033421 29440 solver.cpp:218] Iteration 888 (2.29585 iter/s, 5.22681s/12 iters), loss = 4.87092
I0428 14:50:35.033488 29440 solver.cpp:237]     Train net output #0: loss = 4.87092 (* 1 = 4.87092 loss)
I0428 14:50:35.033500 29440 sgd_solver.cpp:105] Iteration 888, lr = 0.00838702
I0428 14:50:40.236224 29440 solver.cpp:218] Iteration 900 (2.30657 iter/s, 5.20252s/12 iters), loss = 5.00009
I0428 14:50:40.236291 29440 solver.cpp:237]     Train net output #0: loss = 5.00009 (* 1 = 5.00009 loss)
I0428 14:50:40.236305 29440 sgd_solver.cpp:105] Iteration 900, lr = 0.0083671
I0428 14:50:44.305229 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:50:45.514108 29440 solver.cpp:218] Iteration 912 (2.27376 iter/s, 5.27761s/12 iters), loss = 4.89435
I0428 14:50:45.514147 29440 solver.cpp:237]     Train net output #0: loss = 4.89435 (* 1 = 4.89435 loss)
I0428 14:50:45.514156 29440 sgd_solver.cpp:105] Iteration 912, lr = 0.00834724
I0428 14:50:47.684190 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_918.caffemodel
I0428 14:50:52.977115 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_918.solverstate
I0428 14:50:57.570601 29440 solver.cpp:330] Iteration 918, Testing net (#0)
I0428 14:50:57.570777 29440 net.cpp:676] Ignoring source layer train-data
I0428 14:51:01.991631 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:51:02.461521 29440 solver.cpp:397]     Test net output #0: accuracy = 0.0281863
I0428 14:51:02.461556 29440 solver.cpp:397]     Test net output #1: loss = 4.97995 (* 1 = 4.97995 loss)
I0428 14:51:04.612172 29440 solver.cpp:218] Iteration 924 (0.628361 iter/s, 19.0973s/12 iters), loss = 4.91923
I0428 14:51:04.612226 29440 solver.cpp:237]     Train net output #0: loss = 4.91923 (* 1 = 4.91923 loss)
I0428 14:51:04.612236 29440 sgd_solver.cpp:105] Iteration 924, lr = 0.00832742
I0428 14:51:10.018258 29440 solver.cpp:218] Iteration 936 (2.21983 iter/s, 5.40581s/12 iters), loss = 4.88043
I0428 14:51:10.018303 29440 solver.cpp:237]     Train net output #0: loss = 4.88043 (* 1 = 4.88043 loss)
I0428 14:51:10.018313 29440 sgd_solver.cpp:105] Iteration 936, lr = 0.00830765
I0428 14:51:15.632398 29440 solver.cpp:218] Iteration 948 (2.13756 iter/s, 5.61387s/12 iters), loss = 4.97577
I0428 14:51:15.632438 29440 solver.cpp:237]     Train net output #0: loss = 4.97577 (* 1 = 4.97577 loss)
I0428 14:51:15.632447 29440 sgd_solver.cpp:105] Iteration 948, lr = 0.00828793
I0428 14:51:21.089675 29440 solver.cpp:218] Iteration 960 (2.19988 iter/s, 5.45483s/12 iters), loss = 4.9849
I0428 14:51:21.089716 29440 solver.cpp:237]     Train net output #0: loss = 4.9849 (* 1 = 4.9849 loss)
I0428 14:51:21.089726 29440 sgd_solver.cpp:105] Iteration 960, lr = 0.00826825
I0428 14:51:26.416325 29440 solver.cpp:218] Iteration 972 (2.25386 iter/s, 5.3242s/12 iters), loss = 4.90079
I0428 14:51:26.416366 29440 solver.cpp:237]     Train net output #0: loss = 4.90079 (* 1 = 4.90079 loss)
I0428 14:51:26.416376 29440 sgd_solver.cpp:105] Iteration 972, lr = 0.00824862
I0428 14:51:31.924310 29440 solver.cpp:218] Iteration 984 (2.17876 iter/s, 5.50772s/12 iters), loss = 4.90626
I0428 14:51:31.924443 29440 solver.cpp:237]     Train net output #0: loss = 4.90626 (* 1 = 4.90626 loss)
I0428 14:51:31.924453 29440 sgd_solver.cpp:105] Iteration 984, lr = 0.00822903
I0428 14:51:37.377096 29440 solver.cpp:218] Iteration 996 (2.20085 iter/s, 5.45244s/12 iters), loss = 4.98787
I0428 14:51:37.377138 29440 solver.cpp:237]     Train net output #0: loss = 4.98787 (* 1 = 4.98787 loss)
I0428 14:51:37.377147 29440 sgd_solver.cpp:105] Iteration 996, lr = 0.0082095
I0428 14:51:42.758550 29440 solver.cpp:218] Iteration 1008 (2.2309 iter/s, 5.37899s/12 iters), loss = 4.94723
I0428 14:51:42.758611 29440 solver.cpp:237]     Train net output #0: loss = 4.94723 (* 1 = 4.94723 loss)
I0428 14:51:42.758623 29440 sgd_solver.cpp:105] Iteration 1008, lr = 0.00819001
I0428 14:51:43.663364 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:51:47.556470 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_1020.caffemodel
I0428 14:51:49.703917 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_1020.solverstate
I0428 14:51:51.565768 29440 solver.cpp:330] Iteration 1020, Testing net (#0)
I0428 14:51:51.565788 29440 net.cpp:676] Ignoring source layer train-data
I0428 14:51:55.957998 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:51:56.477633 29440 solver.cpp:397]     Test net output #0: accuracy = 0.028799
I0428 14:51:56.477663 29440 solver.cpp:397]     Test net output #1: loss = 4.96047 (* 1 = 4.96047 loss)
I0428 14:51:56.662101 29440 solver.cpp:218] Iteration 1020 (0.863261 iter/s, 13.9008s/12 iters), loss = 5.02653
I0428 14:51:56.662169 29440 solver.cpp:237]     Train net output #0: loss = 5.02653 (* 1 = 5.02653 loss)
I0428 14:51:56.662180 29440 sgd_solver.cpp:105] Iteration 1020, lr = 0.00817056
I0428 14:52:01.476517 29440 solver.cpp:218] Iteration 1032 (2.49267 iter/s, 4.81412s/12 iters), loss = 4.97035
I0428 14:52:01.476562 29440 solver.cpp:237]     Train net output #0: loss = 4.97035 (* 1 = 4.97035 loss)
I0428 14:52:01.476574 29440 sgd_solver.cpp:105] Iteration 1032, lr = 0.00815116
I0428 14:52:06.782893 29440 solver.cpp:218] Iteration 1044 (2.26154 iter/s, 5.30612s/12 iters), loss = 4.9132
I0428 14:52:06.782991 29440 solver.cpp:237]     Train net output #0: loss = 4.9132 (* 1 = 4.9132 loss)
I0428 14:52:06.783002 29440 sgd_solver.cpp:105] Iteration 1044, lr = 0.00813181
I0428 14:52:12.124904 29440 solver.cpp:218] Iteration 1056 (2.24738 iter/s, 5.33955s/12 iters), loss = 4.74778
I0428 14:52:12.124958 29440 solver.cpp:237]     Train net output #0: loss = 4.74778 (* 1 = 4.74778 loss)
I0428 14:52:12.124969 29440 sgd_solver.cpp:105] Iteration 1056, lr = 0.0081125
I0428 14:52:17.545179 29440 solver.cpp:218] Iteration 1068 (2.21402 iter/s, 5.42s/12 iters), loss = 4.78488
I0428 14:52:17.545225 29440 solver.cpp:237]     Train net output #0: loss = 4.78488 (* 1 = 4.78488 loss)
I0428 14:52:17.545233 29440 sgd_solver.cpp:105] Iteration 1068, lr = 0.00809324
I0428 14:52:23.305075 29440 solver.cpp:218] Iteration 1080 (2.08347 iter/s, 5.75962s/12 iters), loss = 4.9734
I0428 14:52:23.305116 29440 solver.cpp:237]     Train net output #0: loss = 4.9734 (* 1 = 4.9734 loss)
I0428 14:52:23.305125 29440 sgd_solver.cpp:105] Iteration 1080, lr = 0.00807403
I0428 14:52:28.770377 29440 solver.cpp:218] Iteration 1092 (2.19666 iter/s, 5.46283s/12 iters), loss = 4.85747
I0428 14:52:28.770419 29440 solver.cpp:237]     Train net output #0: loss = 4.85747 (* 1 = 4.85747 loss)
I0428 14:52:28.770428 29440 sgd_solver.cpp:105] Iteration 1092, lr = 0.00805486
I0428 14:52:34.072865 29440 solver.cpp:218] Iteration 1104 (2.2632 iter/s, 5.30223s/12 iters), loss = 4.81959
I0428 14:52:34.072930 29440 solver.cpp:237]     Train net output #0: loss = 4.81959 (* 1 = 4.81959 loss)
I0428 14:52:34.072944 29440 sgd_solver.cpp:105] Iteration 1104, lr = 0.00803573
I0428 14:52:37.605682 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:52:39.713382 29440 solver.cpp:218] Iteration 1116 (2.12757 iter/s, 5.64024s/12 iters), loss = 4.84949
I0428 14:52:39.713415 29440 solver.cpp:237]     Train net output #0: loss = 4.84949 (* 1 = 4.84949 loss)
I0428 14:52:39.713423 29440 sgd_solver.cpp:105] Iteration 1116, lr = 0.00801666
I0428 14:52:41.756435 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_1122.caffemodel
I0428 14:52:44.109632 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_1122.solverstate
I0428 14:52:45.179466 29440 solver.cpp:330] Iteration 1122, Testing net (#0)
I0428 14:52:45.179487 29440 net.cpp:676] Ignoring source layer train-data
I0428 14:52:49.566238 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:52:50.117236 29440 solver.cpp:397]     Test net output #0: accuracy = 0.03125
I0428 14:52:50.117274 29440 solver.cpp:397]     Test net output #1: loss = 4.80633 (* 1 = 4.80633 loss)
I0428 14:52:52.231796 29440 solver.cpp:218] Iteration 1128 (0.958627 iter/s, 12.5179s/12 iters), loss = 4.56822
I0428 14:52:52.231842 29440 solver.cpp:237]     Train net output #0: loss = 4.56822 (* 1 = 4.56822 loss)
I0428 14:52:52.231849 29440 sgd_solver.cpp:105] Iteration 1128, lr = 0.00799762
I0428 14:52:57.887956 29440 solver.cpp:218] Iteration 1140 (2.12168 iter/s, 5.65589s/12 iters), loss = 4.74096
I0428 14:52:57.888000 29440 solver.cpp:237]     Train net output #0: loss = 4.74096 (* 1 = 4.74096 loss)
I0428 14:52:57.888010 29440 sgd_solver.cpp:105] Iteration 1140, lr = 0.00797863
I0428 14:53:03.585983 29440 solver.cpp:218] Iteration 1152 (2.10609 iter/s, 5.69775s/12 iters), loss = 4.83637
I0428 14:53:03.586027 29440 solver.cpp:237]     Train net output #0: loss = 4.83637 (* 1 = 4.83637 loss)
I0428 14:53:03.586040 29440 sgd_solver.cpp:105] Iteration 1152, lr = 0.00795969
I0428 14:53:08.701023 29440 solver.cpp:218] Iteration 1164 (2.34714 iter/s, 5.11261s/12 iters), loss = 4.84052
I0428 14:53:08.701138 29440 solver.cpp:237]     Train net output #0: loss = 4.84052 (* 1 = 4.84052 loss)
I0428 14:53:08.701146 29440 sgd_solver.cpp:105] Iteration 1164, lr = 0.00794079
I0428 14:53:13.912655 29440 solver.cpp:218] Iteration 1176 (2.30269 iter/s, 5.2113s/12 iters), loss = 4.74656
I0428 14:53:13.912710 29440 solver.cpp:237]     Train net output #0: loss = 4.74656 (* 1 = 4.74656 loss)
I0428 14:53:13.912722 29440 sgd_solver.cpp:105] Iteration 1176, lr = 0.00792194
I0428 14:53:19.276672 29440 solver.cpp:218] Iteration 1188 (2.23724 iter/s, 5.36374s/12 iters), loss = 4.90003
I0428 14:53:19.276717 29440 solver.cpp:237]     Train net output #0: loss = 4.90003 (* 1 = 4.90003 loss)
I0428 14:53:19.276726 29440 sgd_solver.cpp:105] Iteration 1188, lr = 0.00790313
I0428 14:53:25.056895 29440 solver.cpp:218] Iteration 1200 (2.07614 iter/s, 5.77994s/12 iters), loss = 4.75759
I0428 14:53:25.056939 29440 solver.cpp:237]     Train net output #0: loss = 4.75759 (* 1 = 4.75759 loss)
I0428 14:53:25.056948 29440 sgd_solver.cpp:105] Iteration 1200, lr = 0.00788437
I0428 14:53:30.374117 29440 solver.cpp:218] Iteration 1212 (2.25693 iter/s, 5.31696s/12 iters), loss = 4.7221
I0428 14:53:30.374166 29440 solver.cpp:237]     Train net output #0: loss = 4.7221 (* 1 = 4.7221 loss)
I0428 14:53:30.374176 29440 sgd_solver.cpp:105] Iteration 1212, lr = 0.00786565
I0428 14:53:30.660576 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:53:35.521657 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_1224.caffemodel
I0428 14:53:39.049865 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_1224.solverstate
I0428 14:53:41.122072 29440 solver.cpp:330] Iteration 1224, Testing net (#0)
I0428 14:53:41.122092 29440 net.cpp:676] Ignoring source layer train-data
I0428 14:53:45.600540 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:53:46.203689 29440 solver.cpp:397]     Test net output #0: accuracy = 0.0398284
I0428 14:53:46.203727 29440 solver.cpp:397]     Test net output #1: loss = 4.68014 (* 1 = 4.68014 loss)
I0428 14:53:46.466013 29440 solver.cpp:218] Iteration 1224 (0.745748 iter/s, 16.0912s/12 iters), loss = 4.51501
I0428 14:53:46.467554 29440 solver.cpp:237]     Train net output #0: loss = 4.51501 (* 1 = 4.51501 loss)
I0428 14:53:46.467566 29440 sgd_solver.cpp:105] Iteration 1224, lr = 0.00784697
I0428 14:53:51.092805 29440 solver.cpp:218] Iteration 1236 (2.59456 iter/s, 4.62506s/12 iters), loss = 4.75315
I0428 14:53:51.092852 29440 solver.cpp:237]     Train net output #0: loss = 4.75315 (* 1 = 4.75315 loss)
I0428 14:53:51.092861 29440 sgd_solver.cpp:105] Iteration 1236, lr = 0.00782834
I0428 14:53:56.685264 29440 solver.cpp:218] Iteration 1248 (2.14585 iter/s, 5.59219s/12 iters), loss = 4.72513
I0428 14:53:56.685308 29440 solver.cpp:237]     Train net output #0: loss = 4.72513 (* 1 = 4.72513 loss)
I0428 14:53:56.685317 29440 sgd_solver.cpp:105] Iteration 1248, lr = 0.00780976
I0428 14:54:02.263577 29440 solver.cpp:218] Iteration 1260 (2.15215 iter/s, 5.57582s/12 iters), loss = 4.80891
I0428 14:54:02.263636 29440 solver.cpp:237]     Train net output #0: loss = 4.80891 (* 1 = 4.80891 loss)
I0428 14:54:02.263648 29440 sgd_solver.cpp:105] Iteration 1260, lr = 0.00779122
I0428 14:54:07.733023 29440 solver.cpp:218] Iteration 1272 (2.19412 iter/s, 5.46917s/12 iters), loss = 4.67406
I0428 14:54:07.733067 29440 solver.cpp:237]     Train net output #0: loss = 4.67406 (* 1 = 4.67406 loss)
I0428 14:54:07.733076 29440 sgd_solver.cpp:105] Iteration 1272, lr = 0.00777272
I0428 14:54:13.326298 29440 solver.cpp:218] Iteration 1284 (2.14554 iter/s, 5.59301s/12 iters), loss = 4.66627
I0428 14:54:13.326836 29440 solver.cpp:237]     Train net output #0: loss = 4.66627 (* 1 = 4.66627 loss)
I0428 14:54:13.326846 29440 sgd_solver.cpp:105] Iteration 1284, lr = 0.00775426
I0428 14:54:18.593775 29440 solver.cpp:218] Iteration 1296 (2.27846 iter/s, 5.26673s/12 iters), loss = 4.56693
I0428 14:54:18.593819 29440 solver.cpp:237]     Train net output #0: loss = 4.56693 (* 1 = 4.56693 loss)
I0428 14:54:18.593828 29440 sgd_solver.cpp:105] Iteration 1296, lr = 0.00773585
I0428 14:54:24.125674 29440 solver.cpp:218] Iteration 1308 (2.17021 iter/s, 5.52942s/12 iters), loss = 4.80347
I0428 14:54:24.125717 29440 solver.cpp:237]     Train net output #0: loss = 4.80347 (* 1 = 4.80347 loss)
I0428 14:54:24.125727 29440 sgd_solver.cpp:105] Iteration 1308, lr = 0.00771749
I0428 14:54:26.773274 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:54:29.656543 29440 solver.cpp:218] Iteration 1320 (2.17061 iter/s, 5.52839s/12 iters), loss = 4.63062
I0428 14:54:29.656601 29440 solver.cpp:237]     Train net output #0: loss = 4.63062 (* 1 = 4.63062 loss)
I0428 14:54:29.656613 29440 sgd_solver.cpp:105] Iteration 1320, lr = 0.00769916
I0428 14:54:31.932619 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_1326.caffemodel
I0428 14:54:35.238804 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_1326.solverstate
I0428 14:54:36.291829 29440 solver.cpp:330] Iteration 1326, Testing net (#0)
I0428 14:54:36.291848 29440 net.cpp:676] Ignoring source layer train-data
I0428 14:54:40.445313 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:54:41.074620 29440 solver.cpp:397]     Test net output #0: accuracy = 0.0459559
I0428 14:54:41.074654 29440 solver.cpp:397]     Test net output #1: loss = 4.62356 (* 1 = 4.62356 loss)
I0428 14:54:43.459683 29440 solver.cpp:218] Iteration 1332 (0.869404 iter/s, 13.8026s/12 iters), loss = 4.79063
I0428 14:54:43.459987 29440 solver.cpp:237]     Train net output #0: loss = 4.79063 (* 1 = 4.79063 loss)
I0428 14:54:43.460009 29440 sgd_solver.cpp:105] Iteration 1332, lr = 0.00768088
I0428 14:54:49.029951 29440 solver.cpp:218] Iteration 1344 (2.15524 iter/s, 5.56781s/12 iters), loss = 4.7036
I0428 14:54:49.029994 29440 solver.cpp:237]     Train net output #0: loss = 4.7036 (* 1 = 4.7036 loss)
I0428 14:54:49.030002 29440 sgd_solver.cpp:105] Iteration 1344, lr = 0.00766265
I0428 14:54:53.937690 29440 solver.cpp:218] Iteration 1356 (2.44634 iter/s, 4.90528s/12 iters), loss = 4.64373
I0428 14:54:53.937731 29440 solver.cpp:237]     Train net output #0: loss = 4.64373 (* 1 = 4.64373 loss)
I0428 14:54:53.937741 29440 sgd_solver.cpp:105] Iteration 1356, lr = 0.00764446
I0428 14:54:59.634117 29440 solver.cpp:218] Iteration 1368 (2.10668 iter/s, 5.69616s/12 iters), loss = 4.46358
I0428 14:54:59.634163 29440 solver.cpp:237]     Train net output #0: loss = 4.46358 (* 1 = 4.46358 loss)
I0428 14:54:59.634172 29440 sgd_solver.cpp:105] Iteration 1368, lr = 0.00762631
I0428 14:55:02.409184 29440 blocking_queue.cpp:49] Waiting for data
I0428 14:55:04.933459 29440 solver.cpp:218] Iteration 1380 (2.26454 iter/s, 5.29908s/12 iters), loss = 4.40816
I0428 14:55:04.933499 29440 solver.cpp:237]     Train net output #0: loss = 4.40816 (* 1 = 4.40816 loss)
I0428 14:55:04.933508 29440 sgd_solver.cpp:105] Iteration 1380, lr = 0.0076082
I0428 14:55:10.381538 29440 solver.cpp:218] Iteration 1392 (2.20272 iter/s, 5.44782s/12 iters), loss = 4.55693
I0428 14:55:10.381588 29440 solver.cpp:237]     Train net output #0: loss = 4.55693 (* 1 = 4.55693 loss)
I0428 14:55:10.381599 29440 sgd_solver.cpp:105] Iteration 1392, lr = 0.00759014
I0428 14:55:16.030547 29440 solver.cpp:218] Iteration 1404 (2.1252 iter/s, 5.64654s/12 iters), loss = 4.50476
I0428 14:55:16.030647 29440 solver.cpp:237]     Train net output #0: loss = 4.50476 (* 1 = 4.50476 loss)
I0428 14:55:16.030658 29440 sgd_solver.cpp:105] Iteration 1404, lr = 0.00757212
I0428 14:55:20.724866 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:55:21.183917 29440 solver.cpp:218] Iteration 1416 (2.32968 iter/s, 5.15092s/12 iters), loss = 4.53271
I0428 14:55:21.183970 29440 solver.cpp:237]     Train net output #0: loss = 4.53271 (* 1 = 4.53271 loss)
I0428 14:55:21.183981 29440 sgd_solver.cpp:105] Iteration 1416, lr = 0.00755414
I0428 14:55:25.777477 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_1428.caffemodel
I0428 14:55:30.839200 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_1428.solverstate
I0428 14:55:32.845315 29440 solver.cpp:330] Iteration 1428, Testing net (#0)
I0428 14:55:32.845340 29440 net.cpp:676] Ignoring source layer train-data
I0428 14:55:37.033596 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:55:37.803865 29440 solver.cpp:397]     Test net output #0: accuracy = 0.0514706
I0428 14:55:37.803900 29440 solver.cpp:397]     Test net output #1: loss = 4.47059 (* 1 = 4.47059 loss)
I0428 14:55:38.153707 29440 solver.cpp:218] Iteration 1428 (0.707168 iter/s, 16.9691s/12 iters), loss = 4.59142
I0428 14:55:38.153750 29440 solver.cpp:237]     Train net output #0: loss = 4.59142 (* 1 = 4.59142 loss)
I0428 14:55:38.153761 29440 sgd_solver.cpp:105] Iteration 1428, lr = 0.0075362
I0428 14:55:42.910935 29440 solver.cpp:218] Iteration 1440 (2.52262 iter/s, 4.75696s/12 iters), loss = 4.47916
I0428 14:55:42.910979 29440 solver.cpp:237]     Train net output #0: loss = 4.47916 (* 1 = 4.47916 loss)
I0428 14:55:42.910987 29440 sgd_solver.cpp:105] Iteration 1440, lr = 0.00751831
I0428 14:55:48.037129 29440 solver.cpp:218] Iteration 1452 (2.34204 iter/s, 5.12373s/12 iters), loss = 4.4071
I0428 14:55:48.038023 29440 solver.cpp:237]     Train net output #0: loss = 4.4071 (* 1 = 4.4071 loss)
I0428 14:55:48.038038 29440 sgd_solver.cpp:105] Iteration 1452, lr = 0.00750046
I0428 14:55:53.862017 29440 solver.cpp:218] Iteration 1464 (2.06052 iter/s, 5.82377s/12 iters), loss = 4.57864
I0428 14:55:53.862061 29440 solver.cpp:237]     Train net output #0: loss = 4.57864 (* 1 = 4.57864 loss)
I0428 14:55:53.862069 29440 sgd_solver.cpp:105] Iteration 1464, lr = 0.00748265
I0428 14:55:59.172436 29440 solver.cpp:218] Iteration 1476 (2.26076 iter/s, 5.30795s/12 iters), loss = 4.42394
I0428 14:55:59.172480 29440 solver.cpp:237]     Train net output #0: loss = 4.42394 (* 1 = 4.42394 loss)
I0428 14:55:59.172516 29440 sgd_solver.cpp:105] Iteration 1476, lr = 0.00746489
I0428 14:56:04.503866 29440 solver.cpp:218] Iteration 1488 (2.25091 iter/s, 5.33117s/12 iters), loss = 4.23262
I0428 14:56:04.503911 29440 solver.cpp:237]     Train net output #0: loss = 4.23262 (* 1 = 4.23262 loss)
I0428 14:56:04.503921 29440 sgd_solver.cpp:105] Iteration 1488, lr = 0.00744716
I0428 14:56:10.095558 29440 solver.cpp:218] Iteration 1500 (2.14614 iter/s, 5.59142s/12 iters), loss = 4.36734
I0428 14:56:10.095600 29440 solver.cpp:237]     Train net output #0: loss = 4.36734 (* 1 = 4.36734 loss)
I0428 14:56:10.095608 29440 sgd_solver.cpp:105] Iteration 1500, lr = 0.00742948
I0428 14:56:15.553828 29440 solver.cpp:218] Iteration 1512 (2.19949 iter/s, 5.45581s/12 iters), loss = 4.3221
I0428 14:56:15.553874 29440 solver.cpp:237]     Train net output #0: loss = 4.3221 (* 1 = 4.3221 loss)
I0428 14:56:15.553884 29440 sgd_solver.cpp:105] Iteration 1512, lr = 0.00741184
I0428 14:56:17.222242 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:56:20.857291 29440 solver.cpp:218] Iteration 1524 (2.26372 iter/s, 5.301s/12 iters), loss = 4.34343
I0428 14:56:20.857671 29440 solver.cpp:237]     Train net output #0: loss = 4.34343 (* 1 = 4.34343 loss)
I0428 14:56:20.857683 29440 sgd_solver.cpp:105] Iteration 1524, lr = 0.00739425
I0428 14:56:22.639267 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_1530.caffemodel
I0428 14:56:31.162168 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_1530.solverstate
I0428 14:56:37.118389 29440 solver.cpp:330] Iteration 1530, Testing net (#0)
I0428 14:56:37.118419 29440 net.cpp:676] Ignoring source layer train-data
I0428 14:56:41.501654 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:56:42.307591 29440 solver.cpp:397]     Test net output #0: accuracy = 0.0618873
I0428 14:56:42.307627 29440 solver.cpp:397]     Test net output #1: loss = 4.39923 (* 1 = 4.39923 loss)
I0428 14:56:44.668263 29440 solver.cpp:218] Iteration 1536 (0.504036 iter/s, 23.8078s/12 iters), loss = 4.26055
I0428 14:56:44.668315 29440 solver.cpp:237]     Train net output #0: loss = 4.26055 (* 1 = 4.26055 loss)
I0428 14:56:44.668325 29440 sgd_solver.cpp:105] Iteration 1536, lr = 0.00737669
I0428 14:56:49.881500 29440 solver.cpp:218] Iteration 1548 (2.30195 iter/s, 5.21298s/12 iters), loss = 4.21995
I0428 14:56:49.881541 29440 solver.cpp:237]     Train net output #0: loss = 4.21995 (* 1 = 4.21995 loss)
I0428 14:56:49.881551 29440 sgd_solver.cpp:105] Iteration 1548, lr = 0.00735918
I0428 14:56:55.189015 29440 solver.cpp:218] Iteration 1560 (2.26105 iter/s, 5.30726s/12 iters), loss = 4.214
I0428 14:56:55.189116 29440 solver.cpp:237]     Train net output #0: loss = 4.214 (* 1 = 4.214 loss)
I0428 14:56:55.189124 29440 sgd_solver.cpp:105] Iteration 1560, lr = 0.00734171
I0428 14:57:00.701350 29440 solver.cpp:218] Iteration 1572 (2.17791 iter/s, 5.50986s/12 iters), loss = 4.52214
I0428 14:57:00.701413 29440 solver.cpp:237]     Train net output #0: loss = 4.52214 (* 1 = 4.52214 loss)
I0428 14:57:00.701426 29440 sgd_solver.cpp:105] Iteration 1572, lr = 0.00732427
I0428 14:57:06.454113 29440 solver.cpp:218] Iteration 1584 (2.08606 iter/s, 5.75248s/12 iters), loss = 4.33156
I0428 14:57:06.454155 29440 solver.cpp:237]     Train net output #0: loss = 4.33156 (* 1 = 4.33156 loss)
I0428 14:57:06.454164 29440 sgd_solver.cpp:105] Iteration 1584, lr = 0.00730688
I0428 14:57:11.750844 29440 solver.cpp:218] Iteration 1596 (2.2666 iter/s, 5.29427s/12 iters), loss = 4.2547
I0428 14:57:11.750887 29440 solver.cpp:237]     Train net output #0: loss = 4.2547 (* 1 = 4.2547 loss)
I0428 14:57:11.750896 29440 sgd_solver.cpp:105] Iteration 1596, lr = 0.00728954
I0428 14:57:17.047502 29440 solver.cpp:218] Iteration 1608 (2.26569 iter/s, 5.2964s/12 iters), loss = 4.28919
I0428 14:57:17.047550 29440 solver.cpp:237]     Train net output #0: loss = 4.28919 (* 1 = 4.28919 loss)
I0428 14:57:17.047564 29440 sgd_solver.cpp:105] Iteration 1608, lr = 0.00727223
I0428 14:57:21.363559 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:57:22.616946 29440 solver.cpp:218] Iteration 1620 (2.15472 iter/s, 5.56916s/12 iters), loss = 4.26507
I0428 14:57:22.617004 29440 solver.cpp:237]     Train net output #0: loss = 4.26507 (* 1 = 4.26507 loss)
I0428 14:57:22.617017 29440 sgd_solver.cpp:105] Iteration 1620, lr = 0.00725496
I0428 14:57:27.720999 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_1632.caffemodel
I0428 14:57:35.191718 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_1632.solverstate
I0428 14:57:38.034003 29440 solver.cpp:330] Iteration 1632, Testing net (#0)
I0428 14:57:38.034021 29440 net.cpp:676] Ignoring source layer train-data
I0428 14:57:42.032033 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:57:42.812716 29440 solver.cpp:397]     Test net output #0: accuracy = 0.0594363
I0428 14:57:42.812752 29440 solver.cpp:397]     Test net output #1: loss = 4.48926 (* 1 = 4.48926 loss)
I0428 14:57:43.169754 29440 solver.cpp:218] Iteration 1632 (0.583887 iter/s, 20.5519s/12 iters), loss = 4.42339
I0428 14:57:43.171711 29440 solver.cpp:237]     Train net output #0: loss = 4.42339 (* 1 = 4.42339 loss)
I0428 14:57:43.171725 29440 sgd_solver.cpp:105] Iteration 1632, lr = 0.00723774
I0428 14:57:47.668787 29440 solver.cpp:218] Iteration 1644 (2.66852 iter/s, 4.49687s/12 iters), loss = 4.09818
I0428 14:57:47.668841 29440 solver.cpp:237]     Train net output #0: loss = 4.09818 (* 1 = 4.09818 loss)
I0428 14:57:47.668859 29440 sgd_solver.cpp:105] Iteration 1644, lr = 0.00722056
I0428 14:57:52.873804 29440 solver.cpp:218] Iteration 1656 (2.3056 iter/s, 5.20473s/12 iters), loss = 4.38674
I0428 14:57:52.873852 29440 solver.cpp:237]     Train net output #0: loss = 4.38674 (* 1 = 4.38674 loss)
I0428 14:57:52.873860 29440 sgd_solver.cpp:105] Iteration 1656, lr = 0.00720341
I0428 14:57:58.383735 29440 solver.cpp:218] Iteration 1668 (2.178 iter/s, 5.50965s/12 iters), loss = 4.32065
I0428 14:57:58.383848 29440 solver.cpp:237]     Train net output #0: loss = 4.32065 (* 1 = 4.32065 loss)
I0428 14:57:58.383860 29440 sgd_solver.cpp:105] Iteration 1668, lr = 0.00718631
I0428 14:58:04.155944 29440 solver.cpp:218] Iteration 1680 (2.07905 iter/s, 5.77186s/12 iters), loss = 4.1751
I0428 14:58:04.155987 29440 solver.cpp:237]     Train net output #0: loss = 4.1751 (* 1 = 4.1751 loss)
I0428 14:58:04.155997 29440 sgd_solver.cpp:105] Iteration 1680, lr = 0.00716925
I0428 14:58:10.044219 29440 solver.cpp:218] Iteration 1692 (2.03805 iter/s, 5.88798s/12 iters), loss = 4.19945
I0428 14:58:10.044265 29440 solver.cpp:237]     Train net output #0: loss = 4.19945 (* 1 = 4.19945 loss)
I0428 14:58:10.044273 29440 sgd_solver.cpp:105] Iteration 1692, lr = 0.00715223
I0428 14:58:15.277005 29440 solver.cpp:218] Iteration 1704 (2.29335 iter/s, 5.23252s/12 iters), loss = 4.25778
I0428 14:58:15.277048 29440 solver.cpp:237]     Train net output #0: loss = 4.25778 (* 1 = 4.25778 loss)
I0428 14:58:15.277058 29440 sgd_solver.cpp:105] Iteration 1704, lr = 0.00713525
I0428 14:58:20.668712 29440 solver.cpp:218] Iteration 1716 (2.22575 iter/s, 5.39144s/12 iters), loss = 4.17961
I0428 14:58:20.668753 29440 solver.cpp:237]     Train net output #0: loss = 4.17961 (* 1 = 4.17961 loss)
I0428 14:58:20.668762 29440 sgd_solver.cpp:105] Iteration 1716, lr = 0.00711831
I0428 14:58:21.516726 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:58:26.085641 29440 solver.cpp:218] Iteration 1728 (2.21629 iter/s, 5.41446s/12 iters), loss = 4.22032
I0428 14:58:26.085683 29440 solver.cpp:237]     Train net output #0: loss = 4.22032 (* 1 = 4.22032 loss)
I0428 14:58:26.085692 29440 sgd_solver.cpp:105] Iteration 1728, lr = 0.00710141
I0428 14:58:28.054949 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_1734.caffemodel
I0428 14:58:32.462044 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_1734.solverstate
I0428 14:58:34.261945 29440 solver.cpp:330] Iteration 1734, Testing net (#0)
I0428 14:58:34.261965 29440 net.cpp:676] Ignoring source layer train-data
I0428 14:58:38.238072 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:58:39.032622 29440 solver.cpp:397]     Test net output #0: accuracy = 0.0625
I0428 14:58:39.032649 29440 solver.cpp:397]     Test net output #1: loss = 4.26847 (* 1 = 4.26847 loss)
I0428 14:58:41.636442 29440 solver.cpp:218] Iteration 1740 (0.771697 iter/s, 15.5501s/12 iters), loss = 4.27032
I0428 14:58:41.636512 29440 solver.cpp:237]     Train net output #0: loss = 4.27032 (* 1 = 4.27032 loss)
I0428 14:58:41.636523 29440 sgd_solver.cpp:105] Iteration 1740, lr = 0.00708455
I0428 14:58:47.221966 29440 solver.cpp:218] Iteration 1752 (2.14852 iter/s, 5.58525s/12 iters), loss = 4.38871
I0428 14:58:47.222010 29440 solver.cpp:237]     Train net output #0: loss = 4.38871 (* 1 = 4.38871 loss)
I0428 14:58:47.222020 29440 sgd_solver.cpp:105] Iteration 1752, lr = 0.00706773
I0428 14:58:52.428871 29440 solver.cpp:218] Iteration 1764 (2.30573 iter/s, 5.20442s/12 iters), loss = 4.11121
I0428 14:58:52.428921 29440 solver.cpp:237]     Train net output #0: loss = 4.11121 (* 1 = 4.11121 loss)
I0428 14:58:52.428930 29440 sgd_solver.cpp:105] Iteration 1764, lr = 0.00705094
I0428 14:58:57.836539 29440 solver.cpp:218] Iteration 1776 (2.21919 iter/s, 5.40737s/12 iters), loss = 4.08553
I0428 14:58:57.836588 29440 solver.cpp:237]     Train net output #0: loss = 4.08553 (* 1 = 4.08553 loss)
I0428 14:58:57.836601 29440 sgd_solver.cpp:105] Iteration 1776, lr = 0.0070342
I0428 14:59:03.550640 29440 solver.cpp:218] Iteration 1788 (2.10017 iter/s, 5.71381s/12 iters), loss = 4.36276
I0428 14:59:03.550809 29440 solver.cpp:237]     Train net output #0: loss = 4.36276 (* 1 = 4.36276 loss)
I0428 14:59:03.550820 29440 sgd_solver.cpp:105] Iteration 1788, lr = 0.0070175
I0428 14:59:09.142115 29440 solver.cpp:218] Iteration 1800 (2.14628 iter/s, 5.59107s/12 iters), loss = 4.24578
I0428 14:59:09.142161 29440 solver.cpp:237]     Train net output #0: loss = 4.24578 (* 1 = 4.24578 loss)
I0428 14:59:09.142170 29440 sgd_solver.cpp:105] Iteration 1800, lr = 0.00700084
I0428 14:59:14.563293 29440 solver.cpp:218] Iteration 1812 (2.21365 iter/s, 5.42091s/12 iters), loss = 4.02034
I0428 14:59:14.563333 29440 solver.cpp:237]     Train net output #0: loss = 4.02034 (* 1 = 4.02034 loss)
I0428 14:59:14.563342 29440 sgd_solver.cpp:105] Iteration 1812, lr = 0.00698422
I0428 14:59:17.944387 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:59:20.364696 29440 solver.cpp:218] Iteration 1824 (2.06857 iter/s, 5.80112s/12 iters), loss = 4.30883
I0428 14:59:20.364734 29440 solver.cpp:237]     Train net output #0: loss = 4.30883 (* 1 = 4.30883 loss)
I0428 14:59:20.364743 29440 sgd_solver.cpp:105] Iteration 1824, lr = 0.00696764
I0428 14:59:24.845352 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_1836.caffemodel
I0428 14:59:29.074271 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_1836.solverstate
I0428 14:59:30.985668 29440 solver.cpp:330] Iteration 1836, Testing net (#0)
I0428 14:59:30.985688 29440 net.cpp:676] Ignoring source layer train-data
I0428 14:59:34.953411 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 14:59:35.819890 29440 solver.cpp:397]     Test net output #0: accuracy = 0.0747549
I0428 14:59:35.819921 29440 solver.cpp:397]     Test net output #1: loss = 4.18696 (* 1 = 4.18696 loss)
I0428 14:59:36.152643 29440 solver.cpp:218] Iteration 1836 (0.760212 iter/s, 15.7851s/12 iters), loss = 4.01627
I0428 14:59:36.154845 29440 solver.cpp:237]     Train net output #0: loss = 4.01627 (* 1 = 4.01627 loss)
I0428 14:59:36.154860 29440 sgd_solver.cpp:105] Iteration 1836, lr = 0.0069511
I0428 14:59:40.832545 29440 solver.cpp:218] Iteration 1848 (2.56547 iter/s, 4.6775s/12 iters), loss = 3.9889
I0428 14:59:40.832588 29440 solver.cpp:237]     Train net output #0: loss = 3.9889 (* 1 = 3.9889 loss)
I0428 14:59:40.832598 29440 sgd_solver.cpp:105] Iteration 1848, lr = 0.00693459
I0428 14:59:46.477711 29440 solver.cpp:218] Iteration 1860 (2.12582 iter/s, 5.64489s/12 iters), loss = 4.16159
I0428 14:59:46.477749 29440 solver.cpp:237]     Train net output #0: loss = 4.16159 (* 1 = 4.16159 loss)
I0428 14:59:46.477757 29440 sgd_solver.cpp:105] Iteration 1860, lr = 0.00691813
I0428 14:59:51.764219 29440 solver.cpp:218] Iteration 1872 (2.27099 iter/s, 5.28404s/12 iters), loss = 3.95548
I0428 14:59:51.764262 29440 solver.cpp:237]     Train net output #0: loss = 3.95548 (* 1 = 3.95548 loss)
I0428 14:59:51.764271 29440 sgd_solver.cpp:105] Iteration 1872, lr = 0.0069017
I0428 14:59:57.061569 29440 solver.cpp:218] Iteration 1884 (2.2654 iter/s, 5.29708s/12 iters), loss = 4.15734
I0428 14:59:57.061614 29440 solver.cpp:237]     Train net output #0: loss = 4.15734 (* 1 = 4.15734 loss)
I0428 14:59:57.061622 29440 sgd_solver.cpp:105] Iteration 1884, lr = 0.00688532
I0428 15:00:02.882529 29440 solver.cpp:218] Iteration 1896 (2.06162 iter/s, 5.82067s/12 iters), loss = 4.05488
I0428 15:00:02.882572 29440 solver.cpp:237]     Train net output #0: loss = 4.05488 (* 1 = 4.05488 loss)
I0428 15:00:02.882580 29440 sgd_solver.cpp:105] Iteration 1896, lr = 0.00686897
I0428 15:00:08.323030 29440 solver.cpp:218] Iteration 1908 (2.20666 iter/s, 5.43809s/12 iters), loss = 4.23298
I0428 15:00:08.323168 29440 solver.cpp:237]     Train net output #0: loss = 4.23298 (* 1 = 4.23298 loss)
I0428 15:00:08.323179 29440 sgd_solver.cpp:105] Iteration 1908, lr = 0.00685266
I0428 15:00:13.915714 29440 solver.cpp:218] Iteration 1920 (2.14661 iter/s, 5.5902s/12 iters), loss = 4.19047
I0428 15:00:13.915755 29440 solver.cpp:237]     Train net output #0: loss = 4.19047 (* 1 = 4.19047 loss)
I0428 15:00:13.915762 29440 sgd_solver.cpp:105] Iteration 1920, lr = 0.00683639
I0428 15:00:13.946702 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:00:19.318984 29440 solver.cpp:218] Iteration 1932 (2.2219 iter/s, 5.40077s/12 iters), loss = 3.72603
I0428 15:00:19.319029 29440 solver.cpp:237]     Train net output #0: loss = 3.72603 (* 1 = 3.72603 loss)
I0428 15:00:19.319037 29440 sgd_solver.cpp:105] Iteration 1932, lr = 0.00682016
I0428 15:00:21.499775 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_1938.caffemodel
I0428 15:00:30.655611 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_1938.solverstate
I0428 15:00:34.265130 29440 solver.cpp:330] Iteration 1938, Testing net (#0)
I0428 15:00:34.265152 29440 net.cpp:676] Ignoring source layer train-data
I0428 15:00:38.180188 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:00:39.056556 29440 solver.cpp:397]     Test net output #0: accuracy = 0.0821078
I0428 15:00:39.056995 29440 solver.cpp:397]     Test net output #1: loss = 4.09825 (* 1 = 4.09825 loss)
I0428 15:00:41.461854 29440 solver.cpp:218] Iteration 1944 (0.541958 iter/s, 22.1419s/12 iters), loss = 3.97683
I0428 15:00:41.461900 29440 solver.cpp:237]     Train net output #0: loss = 3.97683 (* 1 = 3.97683 loss)
I0428 15:00:41.461908 29440 sgd_solver.cpp:105] Iteration 1944, lr = 0.00680397
I0428 15:00:46.775617 29440 solver.cpp:218] Iteration 1956 (2.2584 iter/s, 5.31349s/12 iters), loss = 4.03365
I0428 15:00:46.775657 29440 solver.cpp:237]     Train net output #0: loss = 4.03365 (* 1 = 4.03365 loss)
I0428 15:00:46.775666 29440 sgd_solver.cpp:105] Iteration 1956, lr = 0.00678782
I0428 15:00:52.340988 29440 solver.cpp:218] Iteration 1968 (2.15715 iter/s, 5.5629s/12 iters), loss = 4.26975
I0428 15:00:52.341028 29440 solver.cpp:237]     Train net output #0: loss = 4.26975 (* 1 = 4.26975 loss)
I0428 15:00:52.341038 29440 sgd_solver.cpp:105] Iteration 1968, lr = 0.0067717
I0428 15:00:57.730427 29440 solver.cpp:218] Iteration 1980 (2.22669 iter/s, 5.38916s/12 iters), loss = 4.03055
I0428 15:00:57.730473 29440 solver.cpp:237]     Train net output #0: loss = 4.03055 (* 1 = 4.03055 loss)
I0428 15:00:57.730484 29440 sgd_solver.cpp:105] Iteration 1980, lr = 0.00675562
I0428 15:01:03.591521 29440 solver.cpp:218] Iteration 1992 (2.0475 iter/s, 5.8608s/12 iters), loss = 4.08428
I0428 15:01:03.591559 29440 solver.cpp:237]     Train net output #0: loss = 4.08428 (* 1 = 4.08428 loss)
I0428 15:01:03.591568 29440 sgd_solver.cpp:105] Iteration 1992, lr = 0.00673958
I0428 15:01:08.871201 29440 solver.cpp:218] Iteration 2004 (2.27324 iter/s, 5.27882s/12 iters), loss = 3.94816
I0428 15:01:08.871245 29440 solver.cpp:237]     Train net output #0: loss = 3.94816 (* 1 = 3.94816 loss)
I0428 15:01:08.871254 29440 sgd_solver.cpp:105] Iteration 2004, lr = 0.00672358
I0428 15:01:13.867084 29440 solver.cpp:218] Iteration 2016 (2.40316 iter/s, 4.99342s/12 iters), loss = 4.16017
I0428 15:01:13.867260 29440 solver.cpp:237]     Train net output #0: loss = 4.16017 (* 1 = 4.16017 loss)
I0428 15:01:13.867273 29440 sgd_solver.cpp:105] Iteration 2016, lr = 0.00670762
I0428 15:01:16.588433 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:01:19.359458 29440 solver.cpp:218] Iteration 2028 (2.18501 iter/s, 5.49197s/12 iters), loss = 4.00085
I0428 15:01:19.359503 29440 solver.cpp:237]     Train net output #0: loss = 4.00085 (* 1 = 4.00085 loss)
I0428 15:01:19.359511 29440 sgd_solver.cpp:105] Iteration 2028, lr = 0.00669169
I0428 15:01:23.954900 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_2040.caffemodel
I0428 15:01:30.354509 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_2040.solverstate
I0428 15:01:34.724952 29440 solver.cpp:330] Iteration 2040, Testing net (#0)
I0428 15:01:34.724975 29440 net.cpp:676] Ignoring source layer train-data
I0428 15:01:38.688884 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:01:39.627043 29440 solver.cpp:397]     Test net output #0: accuracy = 0.0931373
I0428 15:01:39.627074 29440 solver.cpp:397]     Test net output #1: loss = 3.99301 (* 1 = 3.99301 loss)
I0428 15:01:39.967952 29440 solver.cpp:218] Iteration 2040 (0.582309 iter/s, 20.6076s/12 iters), loss = 3.97075
I0428 15:01:39.969588 29440 solver.cpp:237]     Train net output #0: loss = 3.97075 (* 1 = 3.97075 loss)
I0428 15:01:39.969597 29440 sgd_solver.cpp:105] Iteration 2040, lr = 0.00667581
I0428 15:01:45.101943 29440 solver.cpp:218] Iteration 2052 (2.33821 iter/s, 5.13213s/12 iters), loss = 4.11927
I0428 15:01:45.102135 29440 solver.cpp:237]     Train net output #0: loss = 4.11927 (* 1 = 4.11927 loss)
I0428 15:01:45.102149 29440 sgd_solver.cpp:105] Iteration 2052, lr = 0.00665996
I0428 15:01:49.818593 29440 blocking_queue.cpp:49] Waiting for data
I0428 15:01:50.527647 29440 solver.cpp:218] Iteration 2064 (2.21187 iter/s, 5.42528s/12 iters), loss = 3.95953
I0428 15:01:50.527705 29440 solver.cpp:237]     Train net output #0: loss = 3.95953 (* 1 = 3.95953 loss)
I0428 15:01:50.527719 29440 sgd_solver.cpp:105] Iteration 2064, lr = 0.00664414
I0428 15:01:55.781834 29440 solver.cpp:218] Iteration 2076 (2.28401 iter/s, 5.25391s/12 iters), loss = 3.92479
I0428 15:01:55.781879 29440 solver.cpp:237]     Train net output #0: loss = 3.92479 (* 1 = 3.92479 loss)
I0428 15:01:55.781889 29440 sgd_solver.cpp:105] Iteration 2076, lr = 0.00662837
I0428 15:02:01.377245 29440 solver.cpp:218] Iteration 2088 (2.14472 iter/s, 5.59514s/12 iters), loss = 3.91048
I0428 15:02:01.377282 29440 solver.cpp:237]     Train net output #0: loss = 3.91048 (* 1 = 3.91048 loss)
I0428 15:02:01.377291 29440 sgd_solver.cpp:105] Iteration 2088, lr = 0.00661263
I0428 15:02:06.793401 29440 solver.cpp:218] Iteration 2100 (2.2166 iter/s, 5.41369s/12 iters), loss = 3.88936
I0428 15:02:06.793442 29440 solver.cpp:237]     Train net output #0: loss = 3.88936 (* 1 = 3.88936 loss)
I0428 15:02:06.793450 29440 sgd_solver.cpp:105] Iteration 2100, lr = 0.00659693
I0428 15:02:11.862004 29440 solver.cpp:218] Iteration 2112 (2.36868 iter/s, 5.06612s/12 iters), loss = 3.86601
I0428 15:02:11.862044 29440 solver.cpp:237]     Train net output #0: loss = 3.86601 (* 1 = 3.86601 loss)
I0428 15:02:11.862054 29440 sgd_solver.cpp:105] Iteration 2112, lr = 0.00658127
I0428 15:02:16.686144 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:02:17.074836 29440 solver.cpp:218] Iteration 2124 (2.30213 iter/s, 5.21257s/12 iters), loss = 3.92678
I0428 15:02:17.074877 29440 solver.cpp:237]     Train net output #0: loss = 3.92678 (* 1 = 3.92678 loss)
I0428 15:02:17.074887 29440 sgd_solver.cpp:105] Iteration 2124, lr = 0.00656564
I0428 15:02:22.762822 29440 solver.cpp:218] Iteration 2136 (2.10982 iter/s, 5.6877s/12 iters), loss = 3.86473
I0428 15:02:22.762878 29440 solver.cpp:237]     Train net output #0: loss = 3.86473 (* 1 = 3.86473 loss)
I0428 15:02:22.762890 29440 sgd_solver.cpp:105] Iteration 2136, lr = 0.00655006
I0428 15:02:24.736291 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_2142.caffemodel
I0428 15:02:30.943748 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_2142.solverstate
I0428 15:02:36.099751 29440 solver.cpp:330] Iteration 2142, Testing net (#0)
I0428 15:02:36.099771 29440 net.cpp:676] Ignoring source layer train-data
I0428 15:02:40.089103 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:02:41.014794 29440 solver.cpp:397]     Test net output #0: accuracy = 0.088848
I0428 15:02:41.014824 29440 solver.cpp:397]     Test net output #1: loss = 3.9321 (* 1 = 3.9321 loss)
I0428 15:02:43.189492 29440 solver.cpp:218] Iteration 2148 (0.587492 iter/s, 20.4258s/12 iters), loss = 3.81198
I0428 15:02:43.189534 29440 solver.cpp:237]     Train net output #0: loss = 3.81198 (* 1 = 3.81198 loss)
I0428 15:02:43.189543 29440 sgd_solver.cpp:105] Iteration 2148, lr = 0.00653451
I0428 15:02:48.817109 29440 solver.cpp:218] Iteration 2160 (2.13245 iter/s, 5.62734s/12 iters), loss = 3.9172
I0428 15:02:48.819025 29440 solver.cpp:237]     Train net output #0: loss = 3.9172 (* 1 = 3.9172 loss)
I0428 15:02:48.819034 29440 sgd_solver.cpp:105] Iteration 2160, lr = 0.00651899
I0428 15:02:53.862627 29440 solver.cpp:218] Iteration 2172 (2.37951 iter/s, 5.04305s/12 iters), loss = 4.15012
I0428 15:02:53.862684 29440 solver.cpp:237]     Train net output #0: loss = 4.15012 (* 1 = 4.15012 loss)
I0428 15:02:53.862696 29440 sgd_solver.cpp:105] Iteration 2172, lr = 0.00650351
I0428 15:02:59.178697 29440 solver.cpp:218] Iteration 2184 (2.25743 iter/s, 5.31579s/12 iters), loss = 3.69033
I0428 15:02:59.178737 29440 solver.cpp:237]     Train net output #0: loss = 3.69033 (* 1 = 3.69033 loss)
I0428 15:02:59.178746 29440 sgd_solver.cpp:105] Iteration 2184, lr = 0.00648807
I0428 15:03:04.623042 29440 solver.cpp:218] Iteration 2196 (2.20512 iter/s, 5.44187s/12 iters), loss = 3.65926
I0428 15:03:04.623088 29440 solver.cpp:237]     Train net output #0: loss = 3.65926 (* 1 = 3.65926 loss)
I0428 15:03:04.623097 29440 sgd_solver.cpp:105] Iteration 2196, lr = 0.00647267
I0428 15:03:09.853668 29440 solver.cpp:218] Iteration 2208 (2.29526 iter/s, 5.22817s/12 iters), loss = 3.7652
I0428 15:03:09.853709 29440 solver.cpp:237]     Train net output #0: loss = 3.7652 (* 1 = 3.7652 loss)
I0428 15:03:09.853718 29440 sgd_solver.cpp:105] Iteration 2208, lr = 0.0064573
I0428 15:03:15.277114 29440 solver.cpp:218] Iteration 2220 (2.21273 iter/s, 5.42317s/12 iters), loss = 3.48491
I0428 15:03:15.277160 29440 solver.cpp:237]     Train net output #0: loss = 3.48491 (* 1 = 3.48491 loss)
I0428 15:03:15.277170 29440 sgd_solver.cpp:105] Iteration 2220, lr = 0.00644197
I0428 15:03:17.313093 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:03:21.048982 29440 solver.cpp:218] Iteration 2232 (2.07915 iter/s, 5.77158s/12 iters), loss = 3.48347
I0428 15:03:21.049082 29440 solver.cpp:237]     Train net output #0: loss = 3.48347 (* 1 = 3.48347 loss)
I0428 15:03:21.049091 29440 sgd_solver.cpp:105] Iteration 2232, lr = 0.00642668
I0428 15:03:25.698576 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_2244.caffemodel
I0428 15:03:33.159828 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_2244.solverstate
I0428 15:03:35.817257 29440 solver.cpp:330] Iteration 2244, Testing net (#0)
I0428 15:03:35.817277 29440 net.cpp:676] Ignoring source layer train-data
I0428 15:03:39.678624 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:03:40.666487 29440 solver.cpp:397]     Test net output #0: accuracy = 0.112132
I0428 15:03:40.666514 29440 solver.cpp:397]     Test net output #1: loss = 3.87682 (* 1 = 3.87682 loss)
I0428 15:03:40.994019 29440 solver.cpp:218] Iteration 2244 (0.601745 iter/s, 19.942s/12 iters), loss = 3.52912
I0428 15:03:40.995671 29440 solver.cpp:237]     Train net output #0: loss = 3.52912 (* 1 = 3.52912 loss)
I0428 15:03:40.995682 29440 sgd_solver.cpp:105] Iteration 2244, lr = 0.00641142
I0428 15:03:45.739868 29440 solver.cpp:218] Iteration 2256 (2.52951 iter/s, 4.744s/12 iters), loss = 3.92017
I0428 15:03:45.739914 29440 solver.cpp:237]     Train net output #0: loss = 3.92017 (* 1 = 3.92017 loss)
I0428 15:03:45.739923 29440 sgd_solver.cpp:105] Iteration 2256, lr = 0.0063962
I0428 15:03:51.055363 29440 solver.cpp:218] Iteration 2268 (2.2586 iter/s, 5.31303s/12 iters), loss = 3.52385
I0428 15:03:51.055502 29440 solver.cpp:237]     Train net output #0: loss = 3.52385 (* 1 = 3.52385 loss)
I0428 15:03:51.055513 29440 sgd_solver.cpp:105] Iteration 2268, lr = 0.00638101
I0428 15:03:56.852413 29440 solver.cpp:218] Iteration 2280 (2.07015 iter/s, 5.79667s/12 iters), loss = 3.76948
I0428 15:03:56.852458 29440 solver.cpp:237]     Train net output #0: loss = 3.76948 (* 1 = 3.76948 loss)
I0428 15:03:56.852468 29440 sgd_solver.cpp:105] Iteration 2280, lr = 0.00636586
I0428 15:04:02.393061 29440 solver.cpp:218] Iteration 2292 (2.16592 iter/s, 5.54037s/12 iters), loss = 3.82506
I0428 15:04:02.393098 29440 solver.cpp:237]     Train net output #0: loss = 3.82506 (* 1 = 3.82506 loss)
I0428 15:04:02.393106 29440 sgd_solver.cpp:105] Iteration 2292, lr = 0.00635075
I0428 15:04:08.051452 29440 solver.cpp:218] Iteration 2304 (2.12085 iter/s, 5.65812s/12 iters), loss = 3.77617
I0428 15:04:08.051507 29440 solver.cpp:237]     Train net output #0: loss = 3.77617 (* 1 = 3.77617 loss)
I0428 15:04:08.051522 29440 sgd_solver.cpp:105] Iteration 2304, lr = 0.00633567
I0428 15:04:13.721086 29440 solver.cpp:218] Iteration 2316 (2.11665 iter/s, 5.66935s/12 iters), loss = 3.63487
I0428 15:04:13.721125 29440 solver.cpp:237]     Train net output #0: loss = 3.63487 (* 1 = 3.63487 loss)
I0428 15:04:13.721134 29440 sgd_solver.cpp:105] Iteration 2316, lr = 0.00632063
I0428 15:04:17.654255 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:04:18.817108 29440 solver.cpp:218] Iteration 2328 (2.35591 iter/s, 5.09357s/12 iters), loss = 3.78232
I0428 15:04:18.817157 29440 solver.cpp:237]     Train net output #0: loss = 3.78232 (* 1 = 3.78232 loss)
I0428 15:04:18.817165 29440 sgd_solver.cpp:105] Iteration 2328, lr = 0.00630562
I0428 15:04:24.515333 29440 solver.cpp:218] Iteration 2340 (2.10602 iter/s, 5.69794s/12 iters), loss = 3.64025
I0428 15:04:24.515450 29440 solver.cpp:237]     Train net output #0: loss = 3.64025 (* 1 = 3.64025 loss)
I0428 15:04:24.515460 29440 sgd_solver.cpp:105] Iteration 2340, lr = 0.00629065
I0428 15:04:26.593725 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_2346.caffemodel
I0428 15:04:32.785341 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_2346.solverstate
I0428 15:04:35.269713 29440 solver.cpp:330] Iteration 2346, Testing net (#0)
I0428 15:04:35.269735 29440 net.cpp:676] Ignoring source layer train-data
I0428 15:04:39.065127 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:04:40.117993 29440 solver.cpp:397]     Test net output #0: accuracy = 0.123775
I0428 15:04:40.118019 29440 solver.cpp:397]     Test net output #1: loss = 3.81717 (* 1 = 3.81717 loss)
I0428 15:04:42.640116 29440 solver.cpp:218] Iteration 2352 (0.662185 iter/s, 18.1218s/12 iters), loss = 3.54473
I0428 15:04:42.640162 29440 solver.cpp:237]     Train net output #0: loss = 3.54473 (* 1 = 3.54473 loss)
I0428 15:04:42.640170 29440 sgd_solver.cpp:105] Iteration 2352, lr = 0.00627571
I0428 15:04:47.609647 29440 solver.cpp:218] Iteration 2364 (2.41591 iter/s, 4.96707s/12 iters), loss = 3.6707
I0428 15:04:47.609691 29440 solver.cpp:237]     Train net output #0: loss = 3.6707 (* 1 = 3.6707 loss)
I0428 15:04:47.609701 29440 sgd_solver.cpp:105] Iteration 2364, lr = 0.00626081
I0428 15:04:53.394553 29440 solver.cpp:218] Iteration 2376 (2.07447 iter/s, 5.78462s/12 iters), loss = 3.64516
I0428 15:04:53.394594 29440 solver.cpp:237]     Train net output #0: loss = 3.64516 (* 1 = 3.64516 loss)
I0428 15:04:53.394603 29440 sgd_solver.cpp:105] Iteration 2376, lr = 0.00624595
I0428 15:04:58.513311 29440 solver.cpp:218] Iteration 2388 (2.34545 iter/s, 5.11628s/12 iters), loss = 3.42238
I0428 15:04:58.513556 29440 solver.cpp:237]     Train net output #0: loss = 3.42238 (* 1 = 3.42238 loss)
I0428 15:04:58.513566 29440 sgd_solver.cpp:105] Iteration 2388, lr = 0.00623112
I0428 15:05:04.102134 29440 solver.cpp:218] Iteration 2400 (2.14732 iter/s, 5.58835s/12 iters), loss = 3.62242
I0428 15:05:04.102174 29440 solver.cpp:237]     Train net output #0: loss = 3.62242 (* 1 = 3.62242 loss)
I0428 15:05:04.102182 29440 sgd_solver.cpp:105] Iteration 2400, lr = 0.00621633
I0428 15:05:09.787595 29440 solver.cpp:218] Iteration 2412 (2.11075 iter/s, 5.68518s/12 iters), loss = 3.72469
I0428 15:05:09.787639 29440 solver.cpp:237]     Train net output #0: loss = 3.72469 (* 1 = 3.72469 loss)
I0428 15:05:09.787649 29440 sgd_solver.cpp:105] Iteration 2412, lr = 0.00620157
I0428 15:05:15.074918 29440 solver.cpp:218] Iteration 2424 (2.26969 iter/s, 5.28706s/12 iters), loss = 3.55822
I0428 15:05:15.074959 29440 solver.cpp:237]     Train net output #0: loss = 3.55822 (* 1 = 3.55822 loss)
I0428 15:05:15.074967 29440 sgd_solver.cpp:105] Iteration 2424, lr = 0.00618684
I0428 15:05:16.047492 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:05:20.472430 29440 solver.cpp:218] Iteration 2436 (2.22335 iter/s, 5.39725s/12 iters), loss = 3.75331
I0428 15:05:20.472474 29440 solver.cpp:237]     Train net output #0: loss = 3.75331 (* 1 = 3.75331 loss)
I0428 15:05:20.472512 29440 sgd_solver.cpp:105] Iteration 2436, lr = 0.00617215
I0428 15:05:25.339643 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_2448.caffemodel
I0428 15:05:27.607161 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_2448.solverstate
I0428 15:05:30.475759 29440 solver.cpp:330] Iteration 2448, Testing net (#0)
I0428 15:05:30.475848 29440 net.cpp:676] Ignoring source layer train-data
I0428 15:05:34.207321 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:05:35.397238 29440 solver.cpp:397]     Test net output #0: accuracy = 0.127451
I0428 15:05:35.397269 29440 solver.cpp:397]     Test net output #1: loss = 3.79337 (* 1 = 3.79337 loss)
I0428 15:05:35.539801 29440 solver.cpp:218] Iteration 2448 (0.796457 iter/s, 15.0667s/12 iters), loss = 3.66678
I0428 15:05:35.539850 29440 solver.cpp:237]     Train net output #0: loss = 3.66678 (* 1 = 3.66678 loss)
I0428 15:05:35.539860 29440 sgd_solver.cpp:105] Iteration 2448, lr = 0.0061575
I0428 15:05:40.209148 29440 solver.cpp:218] Iteration 2460 (2.57009 iter/s, 4.6691s/12 iters), loss = 3.64972
I0428 15:05:40.209192 29440 solver.cpp:237]     Train net output #0: loss = 3.64972 (* 1 = 3.64972 loss)
I0428 15:05:40.209201 29440 sgd_solver.cpp:105] Iteration 2460, lr = 0.00614288
I0428 15:05:46.021956 29440 solver.cpp:218] Iteration 2472 (2.06451 iter/s, 5.81253s/12 iters), loss = 3.72899
I0428 15:05:46.021999 29440 solver.cpp:237]     Train net output #0: loss = 3.72899 (* 1 = 3.72899 loss)
I0428 15:05:46.022007 29440 sgd_solver.cpp:105] Iteration 2472, lr = 0.0061283
I0428 15:05:51.504207 29440 solver.cpp:218] Iteration 2484 (2.18987 iter/s, 5.47978s/12 iters), loss = 3.52385
I0428 15:05:51.504253 29440 solver.cpp:237]     Train net output #0: loss = 3.52385 (* 1 = 3.52385 loss)
I0428 15:05:51.504262 29440 sgd_solver.cpp:105] Iteration 2484, lr = 0.00611375
I0428 15:05:57.231566 29440 solver.cpp:218] Iteration 2496 (2.09532 iter/s, 5.72705s/12 iters), loss = 3.70701
I0428 15:05:57.231602 29440 solver.cpp:237]     Train net output #0: loss = 3.70701 (* 1 = 3.70701 loss)
I0428 15:05:57.231609 29440 sgd_solver.cpp:105] Iteration 2496, lr = 0.00609923
I0428 15:06:02.346077 29440 solver.cpp:218] Iteration 2508 (2.34742 iter/s, 5.112s/12 iters), loss = 3.53418
I0428 15:06:02.346192 29440 solver.cpp:237]     Train net output #0: loss = 3.53418 (* 1 = 3.53418 loss)
I0428 15:06:02.346201 29440 sgd_solver.cpp:105] Iteration 2508, lr = 0.00608475
I0428 15:06:07.696893 29440 solver.cpp:218] Iteration 2520 (2.24279 iter/s, 5.35048s/12 iters), loss = 3.45479
I0428 15:06:07.696933 29440 solver.cpp:237]     Train net output #0: loss = 3.45479 (* 1 = 3.45479 loss)
I0428 15:06:07.696943 29440 sgd_solver.cpp:105] Iteration 2520, lr = 0.0060703
I0428 15:06:11.345341 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:06:13.488569 29440 solver.cpp:218] Iteration 2532 (2.07204 iter/s, 5.7914s/12 iters), loss = 3.70052
I0428 15:06:13.488612 29440 solver.cpp:237]     Train net output #0: loss = 3.70052 (* 1 = 3.70052 loss)
I0428 15:06:13.488621 29440 sgd_solver.cpp:105] Iteration 2532, lr = 0.00605589
I0428 15:06:18.681236 29440 solver.cpp:218] Iteration 2544 (2.31204 iter/s, 5.19021s/12 iters), loss = 3.49374
I0428 15:06:18.681282 29440 solver.cpp:237]     Train net output #0: loss = 3.49374 (* 1 = 3.49374 loss)
I0428 15:06:18.681290 29440 sgd_solver.cpp:105] Iteration 2544, lr = 0.00604151
I0428 15:06:20.808360 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_2550.caffemodel
I0428 15:06:23.505589 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_2550.solverstate
I0428 15:06:25.789698 29440 solver.cpp:330] Iteration 2550, Testing net (#0)
I0428 15:06:25.789717 29440 net.cpp:676] Ignoring source layer train-data
I0428 15:06:29.692162 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:06:30.803696 29440 solver.cpp:397]     Test net output #0: accuracy = 0.125
I0428 15:06:30.803725 29440 solver.cpp:397]     Test net output #1: loss = 3.79222 (* 1 = 3.79222 loss)
I0428 15:06:32.939007 29440 solver.cpp:218] Iteration 2556 (0.841683 iter/s, 14.2572s/12 iters), loss = 3.74561
I0428 15:06:32.939098 29440 solver.cpp:237]     Train net output #0: loss = 3.74561 (* 1 = 3.74561 loss)
I0428 15:06:32.939108 29440 sgd_solver.cpp:105] Iteration 2556, lr = 0.00602717
I0428 15:06:38.598820 29440 solver.cpp:218] Iteration 2568 (2.12033 iter/s, 5.65949s/12 iters), loss = 3.79168
I0428 15:06:38.598863 29440 solver.cpp:237]     Train net output #0: loss = 3.79168 (* 1 = 3.79168 loss)
I0428 15:06:38.598872 29440 sgd_solver.cpp:105] Iteration 2568, lr = 0.00601286
I0428 15:06:44.084148 29440 solver.cpp:218] Iteration 2580 (2.18776 iter/s, 5.48506s/12 iters), loss = 3.59518
I0428 15:06:44.084189 29440 solver.cpp:237]     Train net output #0: loss = 3.59518 (* 1 = 3.59518 loss)
I0428 15:06:44.084198 29440 sgd_solver.cpp:105] Iteration 2580, lr = 0.00599858
I0428 15:06:49.565585 29440 solver.cpp:218] Iteration 2592 (2.19019 iter/s, 5.47897s/12 iters), loss = 3.36689
I0428 15:06:49.565632 29440 solver.cpp:237]     Train net output #0: loss = 3.36689 (* 1 = 3.36689 loss)
I0428 15:06:49.565642 29440 sgd_solver.cpp:105] Iteration 2592, lr = 0.00598434
I0428 15:06:55.008956 29440 solver.cpp:218] Iteration 2604 (2.20551 iter/s, 5.44091s/12 iters), loss = 3.47772
I0428 15:06:55.008996 29440 solver.cpp:237]     Train net output #0: loss = 3.47772 (* 1 = 3.47772 loss)
I0428 15:06:55.009006 29440 sgd_solver.cpp:105] Iteration 2604, lr = 0.00597013
I0428 15:07:00.379621 29440 solver.cpp:218] Iteration 2616 (2.23447 iter/s, 5.3704s/12 iters), loss = 3.48805
I0428 15:07:00.379678 29440 solver.cpp:237]     Train net output #0: loss = 3.48805 (* 1 = 3.48805 loss)
I0428 15:07:00.379690 29440 sgd_solver.cpp:105] Iteration 2616, lr = 0.00595596
I0428 15:07:05.785568 29440 solver.cpp:218] Iteration 2628 (2.21989 iter/s, 5.40567s/12 iters), loss = 3.43547
I0428 15:07:05.785732 29440 solver.cpp:237]     Train net output #0: loss = 3.43547 (* 1 = 3.43547 loss)
I0428 15:07:05.785742 29440 sgd_solver.cpp:105] Iteration 2628, lr = 0.00594182
I0428 15:07:06.089321 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:07:11.222751 29440 solver.cpp:218] Iteration 2640 (2.20718 iter/s, 5.4368s/12 iters), loss = 3.11348
I0428 15:07:11.222795 29440 solver.cpp:237]     Train net output #0: loss = 3.11348 (* 1 = 3.11348 loss)
I0428 15:07:11.222805 29440 sgd_solver.cpp:105] Iteration 2640, lr = 0.00592771
I0428 15:07:15.934737 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_2652.caffemodel
I0428 15:07:17.330034 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_2652.solverstate
I0428 15:07:18.468070 29440 solver.cpp:330] Iteration 2652, Testing net (#0)
I0428 15:07:18.468091 29440 net.cpp:676] Ignoring source layer train-data
I0428 15:07:22.176703 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:07:23.380333 29440 solver.cpp:397]     Test net output #0: accuracy = 0.136029
I0428 15:07:23.380364 29440 solver.cpp:397]     Test net output #1: loss = 3.67223 (* 1 = 3.67223 loss)
I0428 15:07:23.561640 29440 solver.cpp:218] Iteration 2652 (0.972576 iter/s, 12.3384s/12 iters), loss = 3.46316
I0428 15:07:23.563200 29440 solver.cpp:237]     Train net output #0: loss = 3.46316 (* 1 = 3.46316 loss)
I0428 15:07:23.563210 29440 sgd_solver.cpp:105] Iteration 2652, lr = 0.00591364
I0428 15:07:28.149226 29440 solver.cpp:218] Iteration 2664 (2.61675 iter/s, 4.58584s/12 iters), loss = 3.60566
I0428 15:07:28.149271 29440 solver.cpp:237]     Train net output #0: loss = 3.60566 (* 1 = 3.60566 loss)
I0428 15:07:28.149281 29440 sgd_solver.cpp:105] Iteration 2664, lr = 0.0058996
I0428 15:07:33.474676 29440 solver.cpp:218] Iteration 2676 (2.25344 iter/s, 5.32518s/12 iters), loss = 3.58863
I0428 15:07:33.474718 29440 solver.cpp:237]     Train net output #0: loss = 3.58863 (* 1 = 3.58863 loss)
I0428 15:07:33.474727 29440 sgd_solver.cpp:105] Iteration 2676, lr = 0.00588559
I0428 15:07:38.835393 29440 solver.cpp:218] Iteration 2688 (2.23862 iter/s, 5.36045s/12 iters), loss = 3.28488
I0428 15:07:38.835492 29440 solver.cpp:237]     Train net output #0: loss = 3.28488 (* 1 = 3.28488 loss)
I0428 15:07:38.835502 29440 sgd_solver.cpp:105] Iteration 2688, lr = 0.00587162
I0428 15:07:44.590780 29440 solver.cpp:218] Iteration 2700 (2.08512 iter/s, 5.75505s/12 iters), loss = 3.31488
I0428 15:07:44.590824 29440 solver.cpp:237]     Train net output #0: loss = 3.31488 (* 1 = 3.31488 loss)
I0428 15:07:44.590832 29440 sgd_solver.cpp:105] Iteration 2700, lr = 0.00585768
I0428 15:07:49.986018 29440 solver.cpp:218] Iteration 2712 (2.22519 iter/s, 5.39279s/12 iters), loss = 3.42401
I0428 15:07:49.986060 29440 solver.cpp:237]     Train net output #0: loss = 3.42401 (* 1 = 3.42401 loss)
I0428 15:07:49.986069 29440 sgd_solver.cpp:105] Iteration 2712, lr = 0.00584377
I0428 15:07:55.347054 29440 solver.cpp:218] Iteration 2724 (2.23848 iter/s, 5.36077s/12 iters), loss = 3.50961
I0428 15:07:55.347100 29440 solver.cpp:237]     Train net output #0: loss = 3.50961 (* 1 = 3.50961 loss)
I0428 15:07:55.347107 29440 sgd_solver.cpp:105] Iteration 2724, lr = 0.0058299
I0428 15:07:57.781774 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:08:00.721617 29440 solver.cpp:218] Iteration 2736 (2.23285 iter/s, 5.3743s/12 iters), loss = 3.31297
I0428 15:08:00.721662 29440 solver.cpp:237]     Train net output #0: loss = 3.31297 (* 1 = 3.31297 loss)
I0428 15:08:00.721670 29440 sgd_solver.cpp:105] Iteration 2736, lr = 0.00581605
I0428 15:08:06.377075 29440 solver.cpp:218] Iteration 2748 (2.12195 iter/s, 5.65518s/12 iters), loss = 3.47971
I0428 15:08:06.377117 29440 solver.cpp:237]     Train net output #0: loss = 3.47971 (* 1 = 3.47971 loss)
I0428 15:08:06.377126 29440 sgd_solver.cpp:105] Iteration 2748, lr = 0.00580225
I0428 15:08:08.452203 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_2754.caffemodel
I0428 15:08:09.725463 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_2754.solverstate
I0428 15:08:10.872140 29440 solver.cpp:330] Iteration 2754, Testing net (#0)
I0428 15:08:10.872160 29440 net.cpp:676] Ignoring source layer train-data
I0428 15:08:14.568742 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:08:15.558138 29440 blocking_queue.cpp:49] Waiting for data
I0428 15:08:15.905748 29440 solver.cpp:397]     Test net output #0: accuracy = 0.155025
I0428 15:08:15.905776 29440 solver.cpp:397]     Test net output #1: loss = 3.56685 (* 1 = 3.56685 loss)
I0428 15:08:18.108626 29440 solver.cpp:218] Iteration 2760 (1.02312 iter/s, 11.7289s/12 iters), loss = 3.42672
I0428 15:08:18.108675 29440 solver.cpp:237]     Train net output #0: loss = 3.42672 (* 1 = 3.42672 loss)
I0428 15:08:18.108685 29440 sgd_solver.cpp:105] Iteration 2760, lr = 0.00578847
I0428 15:08:23.411298 29440 solver.cpp:218] Iteration 2772 (2.26313 iter/s, 5.3024s/12 iters), loss = 3.18592
I0428 15:08:23.411343 29440 solver.cpp:237]     Train net output #0: loss = 3.18592 (* 1 = 3.18592 loss)
I0428 15:08:23.411352 29440 sgd_solver.cpp:105] Iteration 2772, lr = 0.00577473
I0428 15:08:28.773299 29440 solver.cpp:218] Iteration 2784 (2.23808 iter/s, 5.36174s/12 iters), loss = 3.39698
I0428 15:08:28.773342 29440 solver.cpp:237]     Train net output #0: loss = 3.39698 (* 1 = 3.39698 loss)
I0428 15:08:28.773350 29440 sgd_solver.cpp:105] Iteration 2784, lr = 0.00576102
I0428 15:08:34.114349 29440 solver.cpp:218] Iteration 2796 (2.24686 iter/s, 5.34079s/12 iters), loss = 3.45135
I0428 15:08:34.114388 29440 solver.cpp:237]     Train net output #0: loss = 3.45135 (* 1 = 3.45135 loss)
I0428 15:08:34.114398 29440 sgd_solver.cpp:105] Iteration 2796, lr = 0.00574734
I0428 15:08:39.515856 29440 solver.cpp:218] Iteration 2808 (2.22171 iter/s, 5.40125s/12 iters), loss = 3.31663
I0428 15:08:39.515898 29440 solver.cpp:237]     Train net output #0: loss = 3.31663 (* 1 = 3.31663 loss)
I0428 15:08:39.515906 29440 sgd_solver.cpp:105] Iteration 2808, lr = 0.00573369
I0428 15:08:44.957620 29440 solver.cpp:218] Iteration 2820 (2.20527 iter/s, 5.4415s/12 iters), loss = 3.27589
I0428 15:08:44.957718 29440 solver.cpp:237]     Train net output #0: loss = 3.27589 (* 1 = 3.27589 loss)
I0428 15:08:44.957727 29440 sgd_solver.cpp:105] Iteration 2820, lr = 0.00572008
I0428 15:08:49.699432 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:08:50.248427 29440 solver.cpp:218] Iteration 2832 (2.26822 iter/s, 5.29049s/12 iters), loss = 3.40885
I0428 15:08:50.248471 29440 solver.cpp:237]     Train net output #0: loss = 3.40885 (* 1 = 3.40885 loss)
I0428 15:08:50.248481 29440 sgd_solver.cpp:105] Iteration 2832, lr = 0.0057065
I0428 15:08:55.719851 29440 solver.cpp:218] Iteration 2844 (2.19341 iter/s, 5.47093s/12 iters), loss = 3.3632
I0428 15:08:55.719897 29440 solver.cpp:237]     Train net output #0: loss = 3.3632 (* 1 = 3.3632 loss)
I0428 15:08:55.719905 29440 sgd_solver.cpp:105] Iteration 2844, lr = 0.00569295
I0428 15:09:00.705787 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_2856.caffemodel
I0428 15:09:02.076177 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_2856.solverstate
I0428 15:09:03.223353 29440 solver.cpp:330] Iteration 2856, Testing net (#0)
I0428 15:09:03.223377 29440 net.cpp:676] Ignoring source layer train-data
I0428 15:09:06.871632 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:09:08.166221 29440 solver.cpp:397]     Test net output #0: accuracy = 0.158701
I0428 15:09:08.166262 29440 solver.cpp:397]     Test net output #1: loss = 3.54977 (* 1 = 3.54977 loss)
I0428 15:09:08.307016 29440 solver.cpp:218] Iteration 2856 (0.953393 iter/s, 12.5866s/12 iters), loss = 3.17925
I0428 15:09:08.307058 29440 solver.cpp:237]     Train net output #0: loss = 3.17925 (* 1 = 3.17925 loss)
I0428 15:09:08.307068 29440 sgd_solver.cpp:105] Iteration 2856, lr = 0.00567944
I0428 15:09:12.883812 29440 solver.cpp:218] Iteration 2868 (2.62205 iter/s, 4.57656s/12 iters), loss = 3.39696
I0428 15:09:12.883849 29440 solver.cpp:237]     Train net output #0: loss = 3.39696 (* 1 = 3.39696 loss)
I0428 15:09:12.883859 29440 sgd_solver.cpp:105] Iteration 2868, lr = 0.00566595
I0428 15:09:18.128350 29440 solver.cpp:218] Iteration 2880 (2.28917 iter/s, 5.24208s/12 iters), loss = 3.43667
I0428 15:09:18.128518 29440 solver.cpp:237]     Train net output #0: loss = 3.43667 (* 1 = 3.43667 loss)
I0428 15:09:18.128530 29440 sgd_solver.cpp:105] Iteration 2880, lr = 0.0056525
I0428 15:09:23.464334 29440 solver.cpp:218] Iteration 2892 (2.24992 iter/s, 5.33352s/12 iters), loss = 3.56729
I0428 15:09:23.464380 29440 solver.cpp:237]     Train net output #0: loss = 3.56729 (* 1 = 3.56729 loss)
I0428 15:09:23.464388 29440 sgd_solver.cpp:105] Iteration 2892, lr = 0.00563908
I0428 15:09:28.811354 29440 solver.cpp:218] Iteration 2904 (2.24435 iter/s, 5.34675s/12 iters), loss = 3.35752
I0428 15:09:28.811419 29440 solver.cpp:237]     Train net output #0: loss = 3.35752 (* 1 = 3.35752 loss)
I0428 15:09:28.811431 29440 sgd_solver.cpp:105] Iteration 2904, lr = 0.00562569
I0428 15:09:34.300920 29440 solver.cpp:218] Iteration 2916 (2.18608 iter/s, 5.48928s/12 iters), loss = 3.35209
I0428 15:09:34.300966 29440 solver.cpp:237]     Train net output #0: loss = 3.35209 (* 1 = 3.35209 loss)
I0428 15:09:34.300973 29440 sgd_solver.cpp:105] Iteration 2916, lr = 0.00561233
I0428 15:09:39.697726 29440 solver.cpp:218] Iteration 2928 (2.22365 iter/s, 5.39653s/12 iters), loss = 2.79348
I0428 15:09:39.697778 29440 solver.cpp:237]     Train net output #0: loss = 2.79348 (* 1 = 2.79348 loss)
I0428 15:09:39.697791 29440 sgd_solver.cpp:105] Iteration 2928, lr = 0.00559901
I0428 15:09:41.608268 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:09:44.930727 29440 solver.cpp:218] Iteration 2940 (2.29327 iter/s, 5.23271s/12 iters), loss = 3.12416
I0428 15:09:44.930784 29440 solver.cpp:237]     Train net output #0: loss = 3.12416 (* 1 = 3.12416 loss)
I0428 15:09:44.930797 29440 sgd_solver.cpp:105] Iteration 2940, lr = 0.00558572
I0428 15:09:50.633741 29440 solver.cpp:218] Iteration 2952 (2.10426 iter/s, 5.70273s/12 iters), loss = 3.22194
I0428 15:09:50.633850 29440 solver.cpp:237]     Train net output #0: loss = 3.22194 (* 1 = 3.22194 loss)
I0428 15:09:50.633860 29440 sgd_solver.cpp:105] Iteration 2952, lr = 0.00557245
I0428 15:09:52.850607 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_2958.caffemodel
I0428 15:09:57.027281 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_2958.solverstate
I0428 15:09:58.095688 29440 solver.cpp:330] Iteration 2958, Testing net (#0)
I0428 15:09:58.095707 29440 net.cpp:676] Ignoring source layer train-data
I0428 15:10:01.771662 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:10:03.097071 29440 solver.cpp:397]     Test net output #0: accuracy = 0.169118
I0428 15:10:03.097101 29440 solver.cpp:397]     Test net output #1: loss = 3.55263 (* 1 = 3.55263 loss)
I0428 15:10:05.252707 29440 solver.cpp:218] Iteration 2964 (0.82101 iter/s, 14.6161s/12 iters), loss = 3.2827
I0428 15:10:05.252748 29440 solver.cpp:237]     Train net output #0: loss = 3.2827 (* 1 = 3.2827 loss)
I0428 15:10:05.252755 29440 sgd_solver.cpp:105] Iteration 2964, lr = 0.00555922
I0428 15:10:10.611656 29440 solver.cpp:218] Iteration 2976 (2.23935 iter/s, 5.35869s/12 iters), loss = 3.04906
I0428 15:10:10.611696 29440 solver.cpp:237]     Train net output #0: loss = 3.04906 (* 1 = 3.04906 loss)
I0428 15:10:10.611704 29440 sgd_solver.cpp:105] Iteration 2976, lr = 0.00554603
I0428 15:10:16.491710 29440 solver.cpp:218] Iteration 2988 (2.04089 iter/s, 5.87978s/12 iters), loss = 3.15659
I0428 15:10:16.491753 29440 solver.cpp:237]     Train net output #0: loss = 3.15659 (* 1 = 3.15659 loss)
I0428 15:10:16.491761 29440 sgd_solver.cpp:105] Iteration 2988, lr = 0.00553286
I0428 15:10:21.621492 29440 solver.cpp:218] Iteration 3000 (2.3404 iter/s, 5.12732s/12 iters), loss = 3.17365
I0428 15:10:21.622283 29440 solver.cpp:237]     Train net output #0: loss = 3.17365 (* 1 = 3.17365 loss)
I0428 15:10:21.622293 29440 sgd_solver.cpp:105] Iteration 3000, lr = 0.00551972
I0428 15:10:26.982964 29440 solver.cpp:218] Iteration 3012 (2.23861 iter/s, 5.36047s/12 iters), loss = 3.17926
I0428 15:10:26.983003 29440 solver.cpp:237]     Train net output #0: loss = 3.17926 (* 1 = 3.17926 loss)
I0428 15:10:26.983012 29440 sgd_solver.cpp:105] Iteration 3012, lr = 0.00550662
I0428 15:10:32.429976 29440 solver.cpp:218] Iteration 3024 (2.20315 iter/s, 5.44675s/12 iters), loss = 3.28
I0428 15:10:32.430019 29440 solver.cpp:237]     Train net output #0: loss = 3.28 (* 1 = 3.28 loss)
I0428 15:10:32.430028 29440 sgd_solver.cpp:105] Iteration 3024, lr = 0.00549354
I0428 15:10:36.845229 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:10:38.165892 29440 solver.cpp:218] Iteration 3036 (2.09218 iter/s, 5.73564s/12 iters), loss = 2.89977
I0428 15:10:38.165943 29440 solver.cpp:237]     Train net output #0: loss = 2.89977 (* 1 = 2.89977 loss)
I0428 15:10:38.165954 29440 sgd_solver.cpp:105] Iteration 3036, lr = 0.0054805
I0428 15:10:43.499617 29440 solver.cpp:218] Iteration 3048 (2.25087 iter/s, 5.33127s/12 iters), loss = 3.50763
I0428 15:10:43.499663 29440 solver.cpp:237]     Train net output #0: loss = 3.50763 (* 1 = 3.50763 loss)
I0428 15:10:43.499672 29440 sgd_solver.cpp:105] Iteration 3048, lr = 0.00546749
I0428 15:10:48.316956 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_3060.caffemodel
I0428 15:10:50.987884 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_3060.solverstate
I0428 15:10:52.588243 29440 solver.cpp:330] Iteration 3060, Testing net (#0)
I0428 15:10:52.588306 29440 net.cpp:676] Ignoring source layer train-data
I0428 15:10:56.056324 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:10:57.398483 29440 solver.cpp:397]     Test net output #0: accuracy = 0.158088
I0428 15:10:57.398524 29440 solver.cpp:397]     Test net output #1: loss = 3.57312 (* 1 = 3.57312 loss)
I0428 15:10:57.723572 29440 solver.cpp:218] Iteration 3060 (0.843683 iter/s, 14.2234s/12 iters), loss = 3.05064
I0428 15:10:57.725178 29440 solver.cpp:237]     Train net output #0: loss = 3.05064 (* 1 = 3.05064 loss)
I0428 15:10:57.725189 29440 sgd_solver.cpp:105] Iteration 3060, lr = 0.00545451
I0428 15:11:02.606524 29440 solver.cpp:218] Iteration 3072 (2.45844 iter/s, 4.88114s/12 iters), loss = 3.36508
I0428 15:11:02.606570 29440 solver.cpp:237]     Train net output #0: loss = 3.36508 (* 1 = 3.36508 loss)
I0428 15:11:02.606577 29440 sgd_solver.cpp:105] Iteration 3072, lr = 0.00544156
I0428 15:11:08.389175 29440 solver.cpp:218] Iteration 3084 (2.07527 iter/s, 5.78237s/12 iters), loss = 3.26564
I0428 15:11:08.389225 29440 solver.cpp:237]     Train net output #0: loss = 3.26564 (* 1 = 3.26564 loss)
I0428 15:11:08.389235 29440 sgd_solver.cpp:105] Iteration 3084, lr = 0.00542864
I0428 15:11:13.911216 29440 solver.cpp:218] Iteration 3096 (2.17322 iter/s, 5.52177s/12 iters), loss = 3.16279
I0428 15:11:13.911259 29440 solver.cpp:237]     Train net output #0: loss = 3.16279 (* 1 = 3.16279 loss)
I0428 15:11:13.911268 29440 sgd_solver.cpp:105] Iteration 3096, lr = 0.00541575
I0428 15:11:19.197571 29440 solver.cpp:218] Iteration 3108 (2.27011 iter/s, 5.2861s/12 iters), loss = 3.17628
I0428 15:11:19.197610 29440 solver.cpp:237]     Train net output #0: loss = 3.17628 (* 1 = 3.17628 loss)
I0428 15:11:19.197619 29440 sgd_solver.cpp:105] Iteration 3108, lr = 0.00540289
I0428 15:11:24.729228 29440 solver.cpp:218] Iteration 3120 (2.1703 iter/s, 5.5292s/12 iters), loss = 3.25808
I0428 15:11:24.729328 29440 solver.cpp:237]     Train net output #0: loss = 3.25808 (* 1 = 3.25808 loss)
I0428 15:11:24.729337 29440 sgd_solver.cpp:105] Iteration 3120, lr = 0.00539006
I0428 15:11:30.165616 29440 solver.cpp:218] Iteration 3132 (2.20748 iter/s, 5.43607s/12 iters), loss = 2.92442
I0428 15:11:30.165655 29440 solver.cpp:237]     Train net output #0: loss = 2.92442 (* 1 = 2.92442 loss)
I0428 15:11:30.165664 29440 sgd_solver.cpp:105] Iteration 3132, lr = 0.00537727
I0428 15:11:31.214259 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:11:35.465781 29440 solver.cpp:218] Iteration 3144 (2.26419 iter/s, 5.2999s/12 iters), loss = 3.44963
I0428 15:11:35.465821 29440 solver.cpp:237]     Train net output #0: loss = 3.44963 (* 1 = 3.44963 loss)
I0428 15:11:35.465831 29440 sgd_solver.cpp:105] Iteration 3144, lr = 0.0053645
I0428 15:11:41.051503 29440 solver.cpp:218] Iteration 3156 (2.14844 iter/s, 5.58546s/12 iters), loss = 3.30912
I0428 15:11:41.051543 29440 solver.cpp:237]     Train net output #0: loss = 3.30912 (* 1 = 3.30912 loss)
I0428 15:11:41.051554 29440 sgd_solver.cpp:105] Iteration 3156, lr = 0.00535176
I0428 15:11:43.323844 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_3162.caffemodel
I0428 15:11:44.752907 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_3162.solverstate
I0428 15:11:45.927034 29440 solver.cpp:330] Iteration 3162, Testing net (#0)
I0428 15:11:45.927060 29440 net.cpp:676] Ignoring source layer train-data
I0428 15:11:49.384819 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:11:50.774041 29440 solver.cpp:397]     Test net output #0: accuracy = 0.175858
I0428 15:11:50.774070 29440 solver.cpp:397]     Test net output #1: loss = 3.52827 (* 1 = 3.52827 loss)
I0428 15:11:52.854154 29440 solver.cpp:218] Iteration 3168 (1.01676 iter/s, 11.8021s/12 iters), loss = 2.96874
I0428 15:11:52.854197 29440 solver.cpp:237]     Train net output #0: loss = 2.96874 (* 1 = 2.96874 loss)
I0428 15:11:52.854207 29440 sgd_solver.cpp:105] Iteration 3168, lr = 0.00533906
I0428 15:11:58.189663 29440 solver.cpp:218] Iteration 3180 (2.24919 iter/s, 5.33525s/12 iters), loss = 3.33611
I0428 15:11:58.189807 29440 solver.cpp:237]     Train net output #0: loss = 3.33611 (* 1 = 3.33611 loss)
I0428 15:11:58.189817 29440 sgd_solver.cpp:105] Iteration 3180, lr = 0.00532638
I0428 15:12:03.843061 29440 solver.cpp:218] Iteration 3192 (2.12276 iter/s, 5.65303s/12 iters), loss = 3.04801
I0428 15:12:03.843107 29440 solver.cpp:237]     Train net output #0: loss = 3.04801 (* 1 = 3.04801 loss)
I0428 15:12:03.843116 29440 sgd_solver.cpp:105] Iteration 3192, lr = 0.00531374
I0428 15:12:09.233671 29440 solver.cpp:218] Iteration 3204 (2.22711 iter/s, 5.38815s/12 iters), loss = 3.25609
I0428 15:12:09.233712 29440 solver.cpp:237]     Train net output #0: loss = 3.25609 (* 1 = 3.25609 loss)
I0428 15:12:09.233721 29440 sgd_solver.cpp:105] Iteration 3204, lr = 0.00530112
I0428 15:12:14.560303 29440 solver.cpp:218] Iteration 3216 (2.25388 iter/s, 5.32416s/12 iters), loss = 2.88079
I0428 15:12:14.560343 29440 solver.cpp:237]     Train net output #0: loss = 2.88079 (* 1 = 2.88079 loss)
I0428 15:12:14.560353 29440 sgd_solver.cpp:105] Iteration 3216, lr = 0.00528853
I0428 15:12:20.393654 29440 solver.cpp:218] Iteration 3228 (2.05723 iter/s, 5.83308s/12 iters), loss = 2.97672
I0428 15:12:20.393699 29440 solver.cpp:237]     Train net output #0: loss = 2.97672 (* 1 = 2.97672 loss)
I0428 15:12:20.393712 29440 sgd_solver.cpp:105] Iteration 3228, lr = 0.00527598
I0428 15:12:23.640415 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:12:25.624539 29440 solver.cpp:218] Iteration 3240 (2.29515 iter/s, 5.22842s/12 iters), loss = 3.29627
I0428 15:12:25.624591 29440 solver.cpp:237]     Train net output #0: loss = 3.29627 (* 1 = 3.29627 loss)
I0428 15:12:25.624603 29440 sgd_solver.cpp:105] Iteration 3240, lr = 0.00526345
I0428 15:12:31.107321 29440 solver.cpp:218] Iteration 3252 (2.18965 iter/s, 5.48033s/12 iters), loss = 3.27268
I0428 15:12:31.107436 29440 solver.cpp:237]     Train net output #0: loss = 3.27268 (* 1 = 3.27268 loss)
I0428 15:12:31.107450 29440 sgd_solver.cpp:105] Iteration 3252, lr = 0.00525095
I0428 15:12:35.681228 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_3264.caffemodel
I0428 15:12:37.081616 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_3264.solverstate
I0428 15:12:38.136117 29440 solver.cpp:330] Iteration 3264, Testing net (#0)
I0428 15:12:38.136138 29440 net.cpp:676] Ignoring source layer train-data
I0428 15:12:41.529297 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:12:43.013854 29440 solver.cpp:397]     Test net output #0: accuracy = 0.16973
I0428 15:12:43.013883 29440 solver.cpp:397]     Test net output #1: loss = 3.56497 (* 1 = 3.56497 loss)
I0428 15:12:43.344532 29440 solver.cpp:218] Iteration 3264 (0.980663 iter/s, 12.2366s/12 iters), loss = 3.2406
I0428 15:12:43.346155 29440 solver.cpp:237]     Train net output #0: loss = 3.2406 (* 1 = 3.2406 loss)
I0428 15:12:43.346166 29440 sgd_solver.cpp:105] Iteration 3264, lr = 0.00523849
I0428 15:12:48.060647 29440 solver.cpp:218] Iteration 3276 (2.54545 iter/s, 4.7143s/12 iters), loss = 3.42408
I0428 15:12:48.060696 29440 solver.cpp:237]     Train net output #0: loss = 3.42408 (* 1 = 3.42408 loss)
I0428 15:12:48.060706 29440 sgd_solver.cpp:105] Iteration 3276, lr = 0.00522605
I0428 15:12:53.388031 29440 solver.cpp:218] Iteration 3288 (2.25262 iter/s, 5.32712s/12 iters), loss = 3.15041
I0428 15:12:53.388074 29440 solver.cpp:237]     Train net output #0: loss = 3.15041 (* 1 = 3.15041 loss)
I0428 15:12:53.388085 29440 sgd_solver.cpp:105] Iteration 3288, lr = 0.00521364
I0428 15:12:58.787397 29440 solver.cpp:218] Iteration 3300 (2.22259 iter/s, 5.3991s/12 iters), loss = 3.13189
I0428 15:12:58.787442 29440 solver.cpp:237]     Train net output #0: loss = 3.13189 (* 1 = 3.13189 loss)
I0428 15:12:58.787451 29440 sgd_solver.cpp:105] Iteration 3300, lr = 0.00520126
I0428 15:13:04.510541 29440 solver.cpp:218] Iteration 3312 (2.09685 iter/s, 5.72287s/12 iters), loss = 3.06597
I0428 15:13:04.510679 29440 solver.cpp:237]     Train net output #0: loss = 3.06597 (* 1 = 3.06597 loss)
I0428 15:13:04.510689 29440 sgd_solver.cpp:105] Iteration 3312, lr = 0.00518892
I0428 15:13:09.784965 29440 solver.cpp:218] Iteration 3324 (2.27528 iter/s, 5.27408s/12 iters), loss = 3.08317
I0428 15:13:09.785009 29440 solver.cpp:237]     Train net output #0: loss = 3.08317 (* 1 = 3.08317 loss)
I0428 15:13:09.785018 29440 sgd_solver.cpp:105] Iteration 3324, lr = 0.0051766
I0428 15:13:15.748142 29440 solver.cpp:218] Iteration 3336 (2.01244 iter/s, 5.9629s/12 iters), loss = 2.90269
I0428 15:13:15.748181 29440 solver.cpp:237]     Train net output #0: loss = 2.90269 (* 1 = 2.90269 loss)
I0428 15:13:15.748190 29440 sgd_solver.cpp:105] Iteration 3336, lr = 0.00516431
I0428 15:13:16.061800 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:13:21.159852 29440 solver.cpp:218] Iteration 3348 (2.2184 iter/s, 5.4093s/12 iters), loss = 2.68142
I0428 15:13:21.159893 29440 solver.cpp:237]     Train net output #0: loss = 2.68142 (* 1 = 2.68142 loss)
I0428 15:13:21.159901 29440 sgd_solver.cpp:105] Iteration 3348, lr = 0.00515204
I0428 15:13:26.337348 29440 solver.cpp:218] Iteration 3360 (2.31783 iter/s, 5.17725s/12 iters), loss = 3.08382
I0428 15:13:26.337394 29440 solver.cpp:237]     Train net output #0: loss = 3.08382 (* 1 = 3.08382 loss)
I0428 15:13:26.337404 29440 sgd_solver.cpp:105] Iteration 3360, lr = 0.00513981
I0428 15:13:28.738543 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_3366.caffemodel
I0428 15:13:32.813386 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_3366.solverstate
I0428 15:13:35.921787 29440 solver.cpp:330] Iteration 3366, Testing net (#0)
I0428 15:13:35.921890 29440 net.cpp:676] Ignoring source layer train-data
I0428 15:13:39.367444 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:13:40.921731 29440 solver.cpp:397]     Test net output #0: accuracy = 0.172794
I0428 15:13:40.921777 29440 solver.cpp:397]     Test net output #1: loss = 3.61239 (* 1 = 3.61239 loss)
I0428 15:13:42.833294 29440 solver.cpp:218] Iteration 3372 (0.727481 iter/s, 16.4953s/12 iters), loss = 3.17001
I0428 15:13:42.833341 29440 solver.cpp:237]     Train net output #0: loss = 3.17001 (* 1 = 3.17001 loss)
I0428 15:13:42.833350 29440 sgd_solver.cpp:105] Iteration 3372, lr = 0.00512761
I0428 15:13:48.111460 29440 solver.cpp:218] Iteration 3384 (2.27363 iter/s, 5.27791s/12 iters), loss = 3.35121
I0428 15:13:48.111505 29440 solver.cpp:237]     Train net output #0: loss = 3.35121 (* 1 = 3.35121 loss)
I0428 15:13:48.111513 29440 sgd_solver.cpp:105] Iteration 3384, lr = 0.00511544
I0428 15:13:53.507261 29440 solver.cpp:218] Iteration 3396 (2.22406 iter/s, 5.39554s/12 iters), loss = 2.90823
I0428 15:13:53.507308 29440 solver.cpp:237]     Train net output #0: loss = 2.90823 (* 1 = 2.90823 loss)
I0428 15:13:53.507318 29440 sgd_solver.cpp:105] Iteration 3396, lr = 0.00510329
I0428 15:13:59.253321 29440 solver.cpp:218] Iteration 3408 (2.08849 iter/s, 5.74578s/12 iters), loss = 2.92467
I0428 15:13:59.253376 29440 solver.cpp:237]     Train net output #0: loss = 2.92467 (* 1 = 2.92467 loss)
I0428 15:13:59.253386 29440 sgd_solver.cpp:105] Iteration 3408, lr = 0.00509117
I0428 15:14:04.533917 29440 solver.cpp:218] Iteration 3420 (2.27258 iter/s, 5.28033s/12 iters), loss = 2.86233
I0428 15:14:04.533959 29440 solver.cpp:237]     Train net output #0: loss = 2.86233 (* 1 = 2.86233 loss)
I0428 15:14:04.533967 29440 sgd_solver.cpp:105] Iteration 3420, lr = 0.00507909
I0428 15:14:09.818713 29440 solver.cpp:218] Iteration 3432 (2.27078 iter/s, 5.28454s/12 iters), loss = 2.98038
I0428 15:14:09.824193 29440 solver.cpp:237]     Train net output #0: loss = 2.98038 (* 1 = 2.98038 loss)
I0428 15:14:09.824203 29440 sgd_solver.cpp:105] Iteration 3432, lr = 0.00506703
I0428 15:14:12.566385 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:14:15.088732 29440 solver.cpp:218] Iteration 3444 (2.27949 iter/s, 5.26434s/12 iters), loss = 2.92126
I0428 15:14:15.088776 29440 solver.cpp:237]     Train net output #0: loss = 2.92126 (* 1 = 2.92126 loss)
I0428 15:14:15.088785 29440 sgd_solver.cpp:105] Iteration 3444, lr = 0.005055
I0428 15:14:20.522660 29440 solver.cpp:218] Iteration 3456 (2.20845 iter/s, 5.43366s/12 iters), loss = 3.00436
I0428 15:14:20.522701 29440 solver.cpp:237]     Train net output #0: loss = 3.00436 (* 1 = 3.00436 loss)
I0428 15:14:20.522709 29440 sgd_solver.cpp:105] Iteration 3456, lr = 0.005043
I0428 15:14:25.610862 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_3468.caffemodel
I0428 15:14:27.857894 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_3468.solverstate
I0428 15:14:28.958524 29440 solver.cpp:330] Iteration 3468, Testing net (#0)
I0428 15:14:28.958544 29440 net.cpp:676] Ignoring source layer train-data
I0428 15:14:31.116075 29440 blocking_queue.cpp:49] Waiting for data
I0428 15:14:32.222966 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:14:33.817929 29440 solver.cpp:397]     Test net output #0: accuracy = 0.183824
I0428 15:14:33.817958 29440 solver.cpp:397]     Test net output #1: loss = 3.50421 (* 1 = 3.50421 loss)
I0428 15:14:34.144279 29440 solver.cpp:218] Iteration 3468 (0.880989 iter/s, 13.621s/12 iters), loss = 3.04636
I0428 15:14:34.144338 29440 solver.cpp:237]     Train net output #0: loss = 3.04636 (* 1 = 3.04636 loss)
I0428 15:14:34.144351 29440 sgd_solver.cpp:105] Iteration 3468, lr = 0.00503102
I0428 15:14:38.863437 29440 solver.cpp:218] Iteration 3480 (2.54296 iter/s, 4.71891s/12 iters), loss = 2.92709
I0428 15:14:38.863476 29440 solver.cpp:237]     Train net output #0: loss = 2.92709 (* 1 = 2.92709 loss)
I0428 15:14:38.863485 29440 sgd_solver.cpp:105] Iteration 3480, lr = 0.00501908
I0428 15:14:44.182400 29440 solver.cpp:218] Iteration 3492 (2.25619 iter/s, 5.31871s/12 iters), loss = 2.93665
I0428 15:14:44.182611 29440 solver.cpp:237]     Train net output #0: loss = 2.93665 (* 1 = 2.93665 loss)
I0428 15:14:44.182621 29440 sgd_solver.cpp:105] Iteration 3492, lr = 0.00500716
I0428 15:14:49.676019 29440 solver.cpp:218] Iteration 3504 (2.18452 iter/s, 5.49319s/12 iters), loss = 3.00996
I0428 15:14:49.676061 29440 solver.cpp:237]     Train net output #0: loss = 3.00996 (* 1 = 3.00996 loss)
I0428 15:14:49.676070 29440 sgd_solver.cpp:105] Iteration 3504, lr = 0.00499527
I0428 15:14:55.409842 29440 solver.cpp:218] Iteration 3516 (2.09345 iter/s, 5.73216s/12 iters), loss = 2.92885
I0428 15:14:55.409894 29440 solver.cpp:237]     Train net output #0: loss = 2.92885 (* 1 = 2.92885 loss)
I0428 15:14:55.409906 29440 sgd_solver.cpp:105] Iteration 3516, lr = 0.00498341
I0428 15:15:00.837862 29440 solver.cpp:218] Iteration 3528 (2.21176 iter/s, 5.42554s/12 iters), loss = 3.10976
I0428 15:15:00.837906 29440 solver.cpp:237]     Train net output #0: loss = 3.10976 (* 1 = 3.10976 loss)
I0428 15:15:00.837915 29440 sgd_solver.cpp:105] Iteration 3528, lr = 0.00497158
I0428 15:15:05.806993 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:15:06.147043 29440 solver.cpp:218] Iteration 3540 (2.26035 iter/s, 5.30892s/12 iters), loss = 2.92208
I0428 15:15:06.147083 29440 solver.cpp:237]     Train net output #0: loss = 2.92208 (* 1 = 2.92208 loss)
I0428 15:15:06.147094 29440 sgd_solver.cpp:105] Iteration 3540, lr = 0.00495978
I0428 15:15:11.427994 29440 solver.cpp:218] Iteration 3552 (2.27243 iter/s, 5.28069s/12 iters), loss = 2.73868
I0428 15:15:11.428033 29440 solver.cpp:237]     Train net output #0: loss = 2.73868 (* 1 = 2.73868 loss)
I0428 15:15:11.428045 29440 sgd_solver.cpp:105] Iteration 3552, lr = 0.004948
I0428 15:15:16.658890 29440 solver.cpp:218] Iteration 3564 (2.29514 iter/s, 5.22845s/12 iters), loss = 2.88565
I0428 15:15:16.659015 29440 solver.cpp:237]     Train net output #0: loss = 2.88565 (* 1 = 2.88565 loss)
I0428 15:15:16.659026 29440 sgd_solver.cpp:105] Iteration 3564, lr = 0.00493626
I0428 15:15:18.805531 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_3570.caffemodel
I0428 15:15:22.888731 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_3570.solverstate
I0428 15:15:24.127463 29440 solver.cpp:330] Iteration 3570, Testing net (#0)
I0428 15:15:24.127490 29440 net.cpp:676] Ignoring source layer train-data
I0428 15:15:27.392477 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:15:28.956456 29440 solver.cpp:397]     Test net output #0: accuracy = 0.185049
I0428 15:15:28.956512 29440 solver.cpp:397]     Test net output #1: loss = 3.44584 (* 1 = 3.44584 loss)
I0428 15:15:31.205379 29440 solver.cpp:218] Iteration 3576 (0.82498 iter/s, 14.5458s/12 iters), loss = 2.76223
I0428 15:15:31.205421 29440 solver.cpp:237]     Train net output #0: loss = 2.76223 (* 1 = 2.76223 loss)
I0428 15:15:31.205430 29440 sgd_solver.cpp:105] Iteration 3576, lr = 0.00492454
I0428 15:15:36.507688 29440 solver.cpp:218] Iteration 3588 (2.26328 iter/s, 5.30205s/12 iters), loss = 2.66349
I0428 15:15:36.507732 29440 solver.cpp:237]     Train net output #0: loss = 2.66349 (* 1 = 2.66349 loss)
I0428 15:15:36.507741 29440 sgd_solver.cpp:105] Iteration 3588, lr = 0.00491284
I0428 15:15:41.829543 29440 solver.cpp:218] Iteration 3600 (2.25496 iter/s, 5.32159s/12 iters), loss = 2.91486
I0428 15:15:41.829600 29440 solver.cpp:237]     Train net output #0: loss = 2.91486 (* 1 = 2.91486 loss)
I0428 15:15:41.829612 29440 sgd_solver.cpp:105] Iteration 3600, lr = 0.00490118
I0428 15:15:47.528892 29440 solver.cpp:218] Iteration 3612 (2.10561 iter/s, 5.69906s/12 iters), loss = 2.74505
I0428 15:15:47.528997 29440 solver.cpp:237]     Train net output #0: loss = 2.74505 (* 1 = 2.74505 loss)
I0428 15:15:47.529006 29440 sgd_solver.cpp:105] Iteration 3612, lr = 0.00488954
I0428 15:15:52.901849 29440 solver.cpp:218] Iteration 3624 (2.23354 iter/s, 5.37264s/12 iters), loss = 2.80866
I0428 15:15:52.901887 29440 solver.cpp:237]     Train net output #0: loss = 2.80866 (* 1 = 2.80866 loss)
I0428 15:15:52.901896 29440 sgd_solver.cpp:105] Iteration 3624, lr = 0.00487793
I0428 15:15:58.494973 29440 solver.cpp:218] Iteration 3636 (2.14645 iter/s, 5.59063s/12 iters), loss = 2.57675
I0428 15:15:58.495018 29440 solver.cpp:237]     Train net output #0: loss = 2.57675 (* 1 = 2.57675 loss)
I0428 15:15:58.495025 29440 sgd_solver.cpp:105] Iteration 3636, lr = 0.00486635
I0428 15:16:00.587656 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:16:04.246721 29440 solver.cpp:218] Iteration 3648 (2.08642 iter/s, 5.75147s/12 iters), loss = 2.68516
I0428 15:16:04.246760 29440 solver.cpp:237]     Train net output #0: loss = 2.68516 (* 1 = 2.68516 loss)
I0428 15:16:04.246769 29440 sgd_solver.cpp:105] Iteration 3648, lr = 0.0048548
I0428 15:16:09.701934 29440 solver.cpp:218] Iteration 3660 (2.20073 iter/s, 5.45275s/12 iters), loss = 2.53798
I0428 15:16:09.701977 29440 solver.cpp:237]     Train net output #0: loss = 2.53798 (* 1 = 2.53798 loss)
I0428 15:16:09.701985 29440 sgd_solver.cpp:105] Iteration 3660, lr = 0.00484327
I0428 15:16:14.518366 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_3672.caffemodel
I0428 15:16:20.129456 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_3672.solverstate
I0428 15:16:24.109625 29440 solver.cpp:330] Iteration 3672, Testing net (#0)
I0428 15:16:24.109650 29440 net.cpp:676] Ignoring source layer train-data
I0428 15:16:27.397778 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:16:29.075374 29440 solver.cpp:397]     Test net output #0: accuracy = 0.20098
I0428 15:16:29.075404 29440 solver.cpp:397]     Test net output #1: loss = 3.29429 (* 1 = 3.29429 loss)
I0428 15:16:29.410729 29440 solver.cpp:218] Iteration 3672 (0.608958 iter/s, 19.7058s/12 iters), loss = 2.92691
I0428 15:16:29.412343 29440 solver.cpp:237]     Train net output #0: loss = 2.92691 (* 1 = 2.92691 loss)
I0428 15:16:29.412355 29440 sgd_solver.cpp:105] Iteration 3672, lr = 0.00483177
I0428 15:16:33.940016 29440 solver.cpp:218] Iteration 3684 (2.65048 iter/s, 4.52749s/12 iters), loss = 2.76397
I0428 15:16:33.940060 29440 solver.cpp:237]     Train net output #0: loss = 2.76397 (* 1 = 2.76397 loss)
I0428 15:16:33.940069 29440 sgd_solver.cpp:105] Iteration 3684, lr = 0.0048203
I0428 15:16:38.850358 29440 solver.cpp:218] Iteration 3696 (2.44504 iter/s, 4.9079s/12 iters), loss = 2.74796
I0428 15:16:38.850412 29440 solver.cpp:237]     Train net output #0: loss = 2.74796 (* 1 = 2.74796 loss)
I0428 15:16:38.850423 29440 sgd_solver.cpp:105] Iteration 3696, lr = 0.00480886
I0428 15:16:44.077643 29440 solver.cpp:218] Iteration 3708 (2.29576 iter/s, 5.22702s/12 iters), loss = 2.63515
I0428 15:16:44.077688 29440 solver.cpp:237]     Train net output #0: loss = 2.63515 (* 1 = 2.63515 loss)
I0428 15:16:44.077697 29440 sgd_solver.cpp:105] Iteration 3708, lr = 0.00479744
I0428 15:16:49.597631 29440 solver.cpp:218] Iteration 3720 (2.17402 iter/s, 5.51972s/12 iters), loss = 2.87874
I0428 15:16:49.597672 29440 solver.cpp:237]     Train net output #0: loss = 2.87874 (* 1 = 2.87874 loss)
I0428 15:16:49.597681 29440 sgd_solver.cpp:105] Iteration 3720, lr = 0.00478605
I0428 15:16:54.527516 29440 solver.cpp:218] Iteration 3732 (2.43533 iter/s, 4.92745s/12 iters), loss = 2.68373
I0428 15:16:54.527679 29440 solver.cpp:237]     Train net output #0: loss = 2.68373 (* 1 = 2.68373 loss)
I0428 15:16:54.527689 29440 sgd_solver.cpp:105] Iteration 3732, lr = 0.00477469
I0428 15:16:58.749495 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:16:59.892568 29440 solver.cpp:218] Iteration 3744 (2.23686 iter/s, 5.36467s/12 iters), loss = 2.44137
I0428 15:16:59.892611 29440 solver.cpp:237]     Train net output #0: loss = 2.44137 (* 1 = 2.44137 loss)
I0428 15:16:59.892621 29440 sgd_solver.cpp:105] Iteration 3744, lr = 0.00476335
I0428 15:17:05.749276 29440 solver.cpp:218] Iteration 3756 (2.04903 iter/s, 5.85643s/12 iters), loss = 2.78564
I0428 15:17:05.749321 29440 solver.cpp:237]     Train net output #0: loss = 2.78564 (* 1 = 2.78564 loss)
I0428 15:17:05.749330 29440 sgd_solver.cpp:105] Iteration 3756, lr = 0.00475204
I0428 15:17:11.151563 29440 solver.cpp:218] Iteration 3768 (2.22228 iter/s, 5.39987s/12 iters), loss = 2.44567
I0428 15:17:11.151602 29440 solver.cpp:237]     Train net output #0: loss = 2.44567 (* 1 = 2.44567 loss)
I0428 15:17:11.151610 29440 sgd_solver.cpp:105] Iteration 3768, lr = 0.00474076
I0428 15:17:13.030464 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_3774.caffemodel
I0428 15:17:16.498982 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_3774.solverstate
I0428 15:17:18.190141 29440 solver.cpp:330] Iteration 3774, Testing net (#0)
I0428 15:17:18.190161 29440 net.cpp:676] Ignoring source layer train-data
I0428 15:17:21.542016 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:17:23.257483 29440 solver.cpp:397]     Test net output #0: accuracy = 0.220588
I0428 15:17:23.257511 29440 solver.cpp:397]     Test net output #1: loss = 3.22312 (* 1 = 3.22312 loss)
I0428 15:17:25.394183 29440 solver.cpp:218] Iteration 3780 (0.842707 iter/s, 14.2398s/12 iters), loss = 2.80668
I0428 15:17:25.394322 29440 solver.cpp:237]     Train net output #0: loss = 2.80668 (* 1 = 2.80668 loss)
I0428 15:17:25.394332 29440 sgd_solver.cpp:105] Iteration 3780, lr = 0.00472951
I0428 15:17:30.769408 29440 solver.cpp:218] Iteration 3792 (2.23261 iter/s, 5.37487s/12 iters), loss = 2.49462
I0428 15:17:30.769452 29440 solver.cpp:237]     Train net output #0: loss = 2.49462 (* 1 = 2.49462 loss)
I0428 15:17:30.769464 29440 sgd_solver.cpp:105] Iteration 3792, lr = 0.00471828
I0428 15:17:36.503708 29440 solver.cpp:218] Iteration 3804 (2.09277 iter/s, 5.73403s/12 iters), loss = 2.60262
I0428 15:17:36.503752 29440 solver.cpp:237]     Train net output #0: loss = 2.60262 (* 1 = 2.60262 loss)
I0428 15:17:36.503762 29440 sgd_solver.cpp:105] Iteration 3804, lr = 0.00470707
I0428 15:17:41.971004 29440 solver.cpp:218] Iteration 3816 (2.19586 iter/s, 5.46482s/12 iters), loss = 2.55445
I0428 15:17:41.971050 29440 solver.cpp:237]     Train net output #0: loss = 2.55445 (* 1 = 2.55445 loss)
I0428 15:17:41.971058 29440 sgd_solver.cpp:105] Iteration 3816, lr = 0.0046959
I0428 15:17:47.661454 29440 solver.cpp:218] Iteration 3828 (2.1089 iter/s, 5.69018s/12 iters), loss = 2.73962
I0428 15:17:47.661499 29440 solver.cpp:237]     Train net output #0: loss = 2.73962 (* 1 = 2.73962 loss)
I0428 15:17:47.661507 29440 sgd_solver.cpp:105] Iteration 3828, lr = 0.00468475
I0428 15:17:53.284668 29440 solver.cpp:218] Iteration 3840 (2.13496 iter/s, 5.62073s/12 iters), loss = 2.5418
I0428 15:17:53.284711 29440 solver.cpp:237]     Train net output #0: loss = 2.5418 (* 1 = 2.5418 loss)
I0428 15:17:53.284720 29440 sgd_solver.cpp:105] Iteration 3840, lr = 0.00467363
I0428 15:17:54.351054 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:17:59.124562 29440 solver.cpp:218] Iteration 3852 (2.05493 iter/s, 5.83962s/12 iters), loss = 2.66749
I0428 15:17:59.124938 29440 solver.cpp:237]     Train net output #0: loss = 2.66749 (* 1 = 2.66749 loss)
I0428 15:17:59.124948 29440 sgd_solver.cpp:105] Iteration 3852, lr = 0.00466253
I0428 15:18:04.389794 29440 solver.cpp:218] Iteration 3864 (2.28017 iter/s, 5.26277s/12 iters), loss = 2.89287
I0428 15:18:04.389839 29440 solver.cpp:237]     Train net output #0: loss = 2.89287 (* 1 = 2.89287 loss)
I0428 15:18:04.389849 29440 sgd_solver.cpp:105] Iteration 3864, lr = 0.00465146
I0428 15:18:09.192101 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_3876.caffemodel
I0428 15:18:14.264663 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_3876.solverstate
I0428 15:18:15.529723 29440 solver.cpp:330] Iteration 3876, Testing net (#0)
I0428 15:18:15.529745 29440 net.cpp:676] Ignoring source layer train-data
I0428 15:18:18.680260 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:18:20.366055 29440 solver.cpp:397]     Test net output #0: accuracy = 0.230392
I0428 15:18:20.366089 29440 solver.cpp:397]     Test net output #1: loss = 3.18853 (* 1 = 3.18853 loss)
I0428 15:18:20.700328 29440 solver.cpp:218] Iteration 3876 (0.735751 iter/s, 16.3099s/12 iters), loss = 2.59437
I0428 15:18:20.701984 29440 solver.cpp:237]     Train net output #0: loss = 2.59437 (* 1 = 2.59437 loss)
I0428 15:18:20.701994 29440 sgd_solver.cpp:105] Iteration 3876, lr = 0.00464042
I0428 15:18:25.435127 29440 solver.cpp:218] Iteration 3888 (2.53542 iter/s, 4.73295s/12 iters), loss = 2.6132
I0428 15:18:25.435170 29440 solver.cpp:237]     Train net output #0: loss = 2.6132 (* 1 = 2.6132 loss)
I0428 15:18:25.435179 29440 sgd_solver.cpp:105] Iteration 3888, lr = 0.0046294
I0428 15:18:30.694634 29440 solver.cpp:218] Iteration 3900 (2.28169 iter/s, 5.25925s/12 iters), loss = 2.50235
I0428 15:18:30.694757 29440 solver.cpp:237]     Train net output #0: loss = 2.50235 (* 1 = 2.50235 loss)
I0428 15:18:30.694766 29440 sgd_solver.cpp:105] Iteration 3900, lr = 0.00461841
I0428 15:18:36.384956 29440 solver.cpp:218] Iteration 3912 (2.10975 iter/s, 5.68787s/12 iters), loss = 2.61352
I0428 15:18:36.384996 29440 solver.cpp:237]     Train net output #0: loss = 2.61352 (* 1 = 2.61352 loss)
I0428 15:18:36.385004 29440 sgd_solver.cpp:105] Iteration 3912, lr = 0.00460744
I0428 15:18:41.776376 29440 solver.cpp:218] Iteration 3924 (2.22678 iter/s, 5.38896s/12 iters), loss = 2.50989
I0428 15:18:41.776415 29440 solver.cpp:237]     Train net output #0: loss = 2.50989 (* 1 = 2.50989 loss)
I0428 15:18:41.776424 29440 sgd_solver.cpp:105] Iteration 3924, lr = 0.0045965
I0428 15:18:47.077929 29440 solver.cpp:218] Iteration 3936 (2.26454 iter/s, 5.29909s/12 iters), loss = 2.51278
I0428 15:18:47.077970 29440 solver.cpp:237]     Train net output #0: loss = 2.51278 (* 1 = 2.51278 loss)
I0428 15:18:47.077978 29440 sgd_solver.cpp:105] Iteration 3936, lr = 0.00458559
I0428 15:18:50.453691 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:18:52.371724 29440 solver.cpp:218] Iteration 3948 (2.26692 iter/s, 5.29354s/12 iters), loss = 2.899
I0428 15:18:52.371771 29440 solver.cpp:237]     Train net output #0: loss = 2.899 (* 1 = 2.899 loss)
I0428 15:18:52.371780 29440 sgd_solver.cpp:105] Iteration 3948, lr = 0.0045747
I0428 15:18:57.727756 29440 solver.cpp:218] Iteration 3960 (2.24058 iter/s, 5.35577s/12 iters), loss = 2.62747
I0428 15:18:57.727798 29440 solver.cpp:237]     Train net output #0: loss = 2.62747 (* 1 = 2.62747 loss)
I0428 15:18:57.727807 29440 sgd_solver.cpp:105] Iteration 3960, lr = 0.00456384
I0428 15:19:03.249300 29440 solver.cpp:218] Iteration 3972 (2.17341 iter/s, 5.52127s/12 iters), loss = 2.63271
I0428 15:19:03.249876 29440 solver.cpp:237]     Train net output #0: loss = 2.63271 (* 1 = 2.63271 loss)
I0428 15:19:03.249886 29440 sgd_solver.cpp:105] Iteration 3972, lr = 0.00455301
I0428 15:19:05.453986 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_3978.caffemodel
I0428 15:19:12.240429 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_3978.solverstate
I0428 15:19:15.511528 29440 solver.cpp:330] Iteration 3978, Testing net (#0)
I0428 15:19:15.511549 29440 net.cpp:676] Ignoring source layer train-data
I0428 15:19:18.735502 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:19:20.491228 29440 solver.cpp:397]     Test net output #0: accuracy = 0.224877
I0428 15:19:20.491267 29440 solver.cpp:397]     Test net output #1: loss = 3.19546 (* 1 = 3.19546 loss)
I0428 15:19:22.915159 29440 solver.cpp:218] Iteration 3984 (0.610236 iter/s, 19.6645s/12 iters), loss = 2.74797
I0428 15:19:22.915202 29440 solver.cpp:237]     Train net output #0: loss = 2.74797 (* 1 = 2.74797 loss)
I0428 15:19:22.915211 29440 sgd_solver.cpp:105] Iteration 3984, lr = 0.0045422
I0428 15:19:28.226527 29440 solver.cpp:218] Iteration 3996 (2.25941 iter/s, 5.31111s/12 iters), loss = 2.95008
I0428 15:19:28.226573 29440 solver.cpp:237]     Train net output #0: loss = 2.95008 (* 1 = 2.95008 loss)
I0428 15:19:28.226583 29440 sgd_solver.cpp:105] Iteration 3996, lr = 0.00453141
I0428 15:19:33.920269 29440 solver.cpp:218] Iteration 4008 (2.10768 iter/s, 5.69346s/12 iters), loss = 2.59614
I0428 15:19:33.920372 29440 solver.cpp:237]     Train net output #0: loss = 2.59614 (* 1 = 2.59614 loss)
I0428 15:19:33.920382 29440 sgd_solver.cpp:105] Iteration 4008, lr = 0.00452066
I0428 15:19:38.908041 29440 solver.cpp:218] Iteration 4020 (2.40707 iter/s, 4.9853s/12 iters), loss = 2.45566
I0428 15:19:38.908082 29440 solver.cpp:237]     Train net output #0: loss = 2.45566 (* 1 = 2.45566 loss)
I0428 15:19:38.908092 29440 sgd_solver.cpp:105] Iteration 4020, lr = 0.00450992
I0428 15:19:44.814445 29440 solver.cpp:218] Iteration 4032 (2.03179 iter/s, 5.90613s/12 iters), loss = 2.54775
I0428 15:19:44.814491 29440 solver.cpp:237]     Train net output #0: loss = 2.54775 (* 1 = 2.54775 loss)
I0428 15:19:44.814505 29440 sgd_solver.cpp:105] Iteration 4032, lr = 0.00449921
I0428 15:19:50.057356 29440 solver.cpp:218] Iteration 4044 (2.28989 iter/s, 5.24044s/12 iters), loss = 2.56824
I0428 15:19:50.057411 29440 solver.cpp:237]     Train net output #0: loss = 2.56824 (* 1 = 2.56824 loss)
I0428 15:19:50.057423 29440 sgd_solver.cpp:105] Iteration 4044, lr = 0.00448853
I0428 15:19:50.635816 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:19:55.704515 29440 solver.cpp:218] Iteration 4056 (2.12508 iter/s, 5.64686s/12 iters), loss = 2.51414
I0428 15:19:55.704561 29440 solver.cpp:237]     Train net output #0: loss = 2.51414 (* 1 = 2.51414 loss)
I0428 15:19:55.704569 29440 sgd_solver.cpp:105] Iteration 4056, lr = 0.00447788
I0428 15:20:00.647495 29440 solver.cpp:218] Iteration 4068 (2.42888 iter/s, 4.94055s/12 iters), loss = 2.52284
I0428 15:20:00.647541 29440 solver.cpp:237]     Train net output #0: loss = 2.52284 (* 1 = 2.52284 loss)
I0428 15:20:00.647550 29440 sgd_solver.cpp:105] Iteration 4068, lr = 0.00446724
I0428 15:20:05.678813 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_4080.caffemodel
I0428 15:20:12.809459 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_4080.solverstate
I0428 15:20:16.912350 29440 solver.cpp:330] Iteration 4080, Testing net (#0)
I0428 15:20:16.912374 29440 net.cpp:676] Ignoring source layer train-data
I0428 15:20:20.002516 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:20:21.790319 29440 solver.cpp:397]     Test net output #0: accuracy = 0.227328
I0428 15:20:21.790347 29440 solver.cpp:397]     Test net output #1: loss = 3.23554 (* 1 = 3.23554 loss)
I0428 15:20:22.122413 29440 solver.cpp:218] Iteration 4080 (0.558814 iter/s, 21.4741s/12 iters), loss = 2.3955
I0428 15:20:22.124074 29440 solver.cpp:237]     Train net output #0: loss = 2.3955 (* 1 = 2.3955 loss)
I0428 15:20:22.124085 29440 sgd_solver.cpp:105] Iteration 4080, lr = 0.00445664
I0428 15:20:26.755988 29440 solver.cpp:218] Iteration 4092 (2.59083 iter/s, 4.63173s/12 iters), loss = 2.60332
I0428 15:20:26.756033 29440 solver.cpp:237]     Train net output #0: loss = 2.60332 (* 1 = 2.60332 loss)
I0428 15:20:26.756042 29440 sgd_solver.cpp:105] Iteration 4092, lr = 0.00444606
I0428 15:20:32.736855 29440 solver.cpp:218] Iteration 4104 (2.00649 iter/s, 5.98059s/12 iters), loss = 2.587
I0428 15:20:32.736898 29440 solver.cpp:237]     Train net output #0: loss = 2.587 (* 1 = 2.587 loss)
I0428 15:20:32.736907 29440 sgd_solver.cpp:105] Iteration 4104, lr = 0.0044355
I0428 15:20:38.516420 29440 solver.cpp:218] Iteration 4116 (2.07638 iter/s, 5.77929s/12 iters), loss = 2.59215
I0428 15:20:38.516575 29440 solver.cpp:237]     Train net output #0: loss = 2.59215 (* 1 = 2.59215 loss)
I0428 15:20:38.516585 29440 sgd_solver.cpp:105] Iteration 4116, lr = 0.00442497
I0428 15:20:44.190323 29440 solver.cpp:218] Iteration 4128 (2.11509 iter/s, 5.67352s/12 iters), loss = 2.38223
I0428 15:20:44.190367 29440 solver.cpp:237]     Train net output #0: loss = 2.38223 (* 1 = 2.38223 loss)
I0428 15:20:44.190377 29440 sgd_solver.cpp:105] Iteration 4128, lr = 0.00441447
I0428 15:20:49.490902 29440 solver.cpp:218] Iteration 4140 (2.26496 iter/s, 5.2981s/12 iters), loss = 2.24757
I0428 15:20:49.490962 29440 solver.cpp:237]     Train net output #0: loss = 2.24757 (* 1 = 2.24757 loss)
I0428 15:20:49.490973 29440 sgd_solver.cpp:105] Iteration 4140, lr = 0.00440398
I0428 15:20:52.089530 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:20:54.874760 29440 solver.cpp:218] Iteration 4152 (2.229 iter/s, 5.38358s/12 iters), loss = 2.50326
I0428 15:20:54.874809 29440 solver.cpp:237]     Train net output #0: loss = 2.50326 (* 1 = 2.50326 loss)
I0428 15:20:54.874817 29440 sgd_solver.cpp:105] Iteration 4152, lr = 0.00439353
I0428 15:21:00.171144 29440 solver.cpp:218] Iteration 4164 (2.26581 iter/s, 5.29612s/12 iters), loss = 2.36938
I0428 15:21:00.171192 29440 solver.cpp:237]     Train net output #0: loss = 2.36938 (* 1 = 2.36938 loss)
I0428 15:21:00.171203 29440 sgd_solver.cpp:105] Iteration 4164, lr = 0.0043831
I0428 15:21:06.158210 29440 solver.cpp:218] Iteration 4176 (2.00442 iter/s, 5.98678s/12 iters), loss = 2.51191
I0428 15:21:06.158253 29440 solver.cpp:237]     Train net output #0: loss = 2.51191 (* 1 = 2.51191 loss)
I0428 15:21:06.158262 29440 sgd_solver.cpp:105] Iteration 4176, lr = 0.00437269
I0428 15:21:06.786911 29440 blocking_queue.cpp:49] Waiting for data
I0428 15:21:08.185181 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_4182.caffemodel
I0428 15:21:09.665760 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_4182.solverstate
I0428 15:21:10.782227 29440 solver.cpp:330] Iteration 4182, Testing net (#0)
I0428 15:21:10.782248 29440 net.cpp:676] Ignoring source layer train-data
I0428 15:21:13.794667 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:21:15.629632 29440 solver.cpp:397]     Test net output #0: accuracy = 0.229167
I0428 15:21:15.629668 29440 solver.cpp:397]     Test net output #1: loss = 3.17367 (* 1 = 3.17367 loss)
I0428 15:21:17.747283 29440 solver.cpp:218] Iteration 4188 (1.0357 iter/s, 11.5864s/12 iters), loss = 2.56467
I0428 15:21:17.747324 29440 solver.cpp:237]     Train net output #0: loss = 2.56467 (* 1 = 2.56467 loss)
I0428 15:21:17.747334 29440 sgd_solver.cpp:105] Iteration 4188, lr = 0.00436231
I0428 15:21:23.176584 29440 solver.cpp:218] Iteration 4200 (2.21033 iter/s, 5.42904s/12 iters), loss = 2.47995
I0428 15:21:23.176627 29440 solver.cpp:237]     Train net output #0: loss = 2.47995 (* 1 = 2.47995 loss)
I0428 15:21:23.176636 29440 sgd_solver.cpp:105] Iteration 4200, lr = 0.00435195
I0428 15:21:28.772472 29440 solver.cpp:218] Iteration 4212 (2.14454 iter/s, 5.59562s/12 iters), loss = 2.46927
I0428 15:21:28.772547 29440 solver.cpp:237]     Train net output #0: loss = 2.46927 (* 1 = 2.46927 loss)
I0428 15:21:28.772558 29440 sgd_solver.cpp:105] Iteration 4212, lr = 0.00434162
I0428 15:21:33.963712 29440 solver.cpp:218] Iteration 4224 (2.31171 iter/s, 5.19095s/12 iters), loss = 2.37974
I0428 15:21:33.963755 29440 solver.cpp:237]     Train net output #0: loss = 2.37974 (* 1 = 2.37974 loss)
I0428 15:21:33.963765 29440 sgd_solver.cpp:105] Iteration 4224, lr = 0.00433131
I0428 15:21:39.868943 29440 solver.cpp:218] Iteration 4236 (2.03219 iter/s, 5.90495s/12 iters), loss = 2.12106
I0428 15:21:39.869062 29440 solver.cpp:237]     Train net output #0: loss = 2.12106 (* 1 = 2.12106 loss)
I0428 15:21:39.869072 29440 sgd_solver.cpp:105] Iteration 4236, lr = 0.00432103
I0428 15:21:44.820219 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:21:45.221652 29440 solver.cpp:218] Iteration 4248 (2.24288 iter/s, 5.35026s/12 iters), loss = 2.58563
I0428 15:21:45.221698 29440 solver.cpp:237]     Train net output #0: loss = 2.58563 (* 1 = 2.58563 loss)
I0428 15:21:45.221707 29440 sgd_solver.cpp:105] Iteration 4248, lr = 0.00431077
I0428 15:21:50.664597 29440 solver.cpp:218] Iteration 4260 (2.2048 iter/s, 5.44268s/12 iters), loss = 2.5155
I0428 15:21:50.664638 29440 solver.cpp:237]     Train net output #0: loss = 2.5155 (* 1 = 2.5155 loss)
I0428 15:21:50.664646 29440 sgd_solver.cpp:105] Iteration 4260, lr = 0.00430053
I0428 15:21:56.327685 29440 solver.cpp:218] Iteration 4272 (2.11909 iter/s, 5.66282s/12 iters), loss = 2.39419
I0428 15:21:56.327725 29440 solver.cpp:237]     Train net output #0: loss = 2.39419 (* 1 = 2.39419 loss)
I0428 15:21:56.327734 29440 sgd_solver.cpp:105] Iteration 4272, lr = 0.00429032
I0428 15:22:01.501737 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_4284.caffemodel
I0428 15:22:03.021898 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_4284.solverstate
I0428 15:22:04.081749 29440 solver.cpp:330] Iteration 4284, Testing net (#0)
I0428 15:22:04.081769 29440 net.cpp:676] Ignoring source layer train-data
I0428 15:22:07.052472 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:22:08.938043 29440 solver.cpp:397]     Test net output #0: accuracy = 0.248775
I0428 15:22:08.938076 29440 solver.cpp:397]     Test net output #1: loss = 3.18416 (* 1 = 3.18416 loss)
I0428 15:22:09.120368 29440 solver.cpp:218] Iteration 4284 (0.938075 iter/s, 12.7922s/12 iters), loss = 2.24209
I0428 15:22:09.121891 29440 solver.cpp:237]     Train net output #0: loss = 2.24209 (* 1 = 2.24209 loss)
I0428 15:22:09.121901 29440 sgd_solver.cpp:105] Iteration 4284, lr = 0.00428014
I0428 15:22:14.257241 29440 solver.cpp:218] Iteration 4296 (2.33684 iter/s, 5.13515s/12 iters), loss = 2.26762
I0428 15:22:14.257367 29440 solver.cpp:237]     Train net output #0: loss = 2.26762 (* 1 = 2.26762 loss)
I0428 15:22:14.257377 29440 sgd_solver.cpp:105] Iteration 4296, lr = 0.00426998
I0428 15:22:19.971444 29440 solver.cpp:218] Iteration 4308 (2.10096 iter/s, 5.71167s/12 iters), loss = 2.27751
I0428 15:22:19.971484 29440 solver.cpp:237]     Train net output #0: loss = 2.27751 (* 1 = 2.27751 loss)
I0428 15:22:19.971493 29440 sgd_solver.cpp:105] Iteration 4308, lr = 0.00425984
I0428 15:22:25.384387 29440 solver.cpp:218] Iteration 4320 (2.21792 iter/s, 5.41047s/12 iters), loss = 2.26104
I0428 15:22:25.384431 29440 solver.cpp:237]     Train net output #0: loss = 2.26104 (* 1 = 2.26104 loss)
I0428 15:22:25.384439 29440 sgd_solver.cpp:105] Iteration 4320, lr = 0.00424972
I0428 15:22:30.533543 29440 solver.cpp:218] Iteration 4332 (2.33159 iter/s, 5.1467s/12 iters), loss = 2.34826
I0428 15:22:30.533588 29440 solver.cpp:237]     Train net output #0: loss = 2.34826 (* 1 = 2.34826 loss)
I0428 15:22:30.533597 29440 sgd_solver.cpp:105] Iteration 4332, lr = 0.00423964
I0428 15:22:35.736778 29440 solver.cpp:218] Iteration 4344 (2.30735 iter/s, 5.20077s/12 iters), loss = 2.30547
I0428 15:22:35.736819 29440 solver.cpp:237]     Train net output #0: loss = 2.30547 (* 1 = 2.30547 loss)
I0428 15:22:35.736829 29440 sgd_solver.cpp:105] Iteration 4344, lr = 0.00422957
I0428 15:22:37.791641 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:22:41.562227 29440 solver.cpp:218] Iteration 4356 (2.06002 iter/s, 5.82518s/12 iters), loss = 2.36097
I0428 15:22:41.562268 29440 solver.cpp:237]     Train net output #0: loss = 2.36097 (* 1 = 2.36097 loss)
I0428 15:22:41.562275 29440 sgd_solver.cpp:105] Iteration 4356, lr = 0.00421953
I0428 15:22:47.078919 29440 solver.cpp:218] Iteration 4368 (2.17532 iter/s, 5.51643s/12 iters), loss = 2.10568
I0428 15:22:47.079030 29440 solver.cpp:237]     Train net output #0: loss = 2.10568 (* 1 = 2.10568 loss)
I0428 15:22:47.079041 29440 sgd_solver.cpp:105] Iteration 4368, lr = 0.00420951
I0428 15:22:52.818851 29440 solver.cpp:218] Iteration 4380 (2.09074 iter/s, 5.73959s/12 iters), loss = 2.56032
I0428 15:22:52.818897 29440 solver.cpp:237]     Train net output #0: loss = 2.56032 (* 1 = 2.56032 loss)
I0428 15:22:52.818905 29440 sgd_solver.cpp:105] Iteration 4380, lr = 0.00419952
I0428 15:22:55.028690 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_4386.caffemodel
I0428 15:22:56.511602 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_4386.solverstate
I0428 15:22:57.567425 29440 solver.cpp:330] Iteration 4386, Testing net (#0)
I0428 15:22:57.567445 29440 net.cpp:676] Ignoring source layer train-data
I0428 15:23:00.569406 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:23:02.594096 29440 solver.cpp:397]     Test net output #0: accuracy = 0.235907
I0428 15:23:02.594134 29440 solver.cpp:397]     Test net output #1: loss = 3.24349 (* 1 = 3.24349 loss)
I0428 15:23:04.689116 29440 solver.cpp:218] Iteration 4392 (1.01097 iter/s, 11.8698s/12 iters), loss = 2.21411
I0428 15:23:04.689160 29440 solver.cpp:237]     Train net output #0: loss = 2.21411 (* 1 = 2.21411 loss)
I0428 15:23:04.689169 29440 sgd_solver.cpp:105] Iteration 4392, lr = 0.00418954
I0428 15:23:10.081485 29440 solver.cpp:218] Iteration 4404 (2.22548 iter/s, 5.3921s/12 iters), loss = 2.53062
I0428 15:23:10.081530 29440 solver.cpp:237]     Train net output #0: loss = 2.53062 (* 1 = 2.53062 loss)
I0428 15:23:10.081539 29440 sgd_solver.cpp:105] Iteration 4404, lr = 0.0041796
I0428 15:23:15.735767 29440 solver.cpp:218] Iteration 4416 (2.12239 iter/s, 5.65401s/12 iters), loss = 2.43045
I0428 15:23:15.735806 29440 solver.cpp:237]     Train net output #0: loss = 2.43045 (* 1 = 2.43045 loss)
I0428 15:23:15.735816 29440 sgd_solver.cpp:105] Iteration 4416, lr = 0.00416967
I0428 15:23:21.515717 29440 solver.cpp:218] Iteration 4428 (2.07703 iter/s, 5.77748s/12 iters), loss = 2.50872
I0428 15:23:21.515866 29440 solver.cpp:237]     Train net output #0: loss = 2.50872 (* 1 = 2.50872 loss)
I0428 15:23:21.515877 29440 sgd_solver.cpp:105] Iteration 4428, lr = 0.00415977
I0428 15:23:27.018988 29440 solver.cpp:218] Iteration 4440 (2.18151 iter/s, 5.50079s/12 iters), loss = 2.30924
I0428 15:23:27.019034 29440 solver.cpp:237]     Train net output #0: loss = 2.30924 (* 1 = 2.30924 loss)
I0428 15:23:27.019044 29440 sgd_solver.cpp:105] Iteration 4440, lr = 0.0041499
I0428 15:23:31.352232 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:23:32.654871 29440 solver.cpp:218] Iteration 4452 (2.12932 iter/s, 5.63561s/12 iters), loss = 2.04556
I0428 15:23:32.654917 29440 solver.cpp:237]     Train net output #0: loss = 2.04556 (* 1 = 2.04556 loss)
I0428 15:23:32.654925 29440 sgd_solver.cpp:105] Iteration 4452, lr = 0.00414005
I0428 15:23:38.157416 29440 solver.cpp:218] Iteration 4464 (2.18179 iter/s, 5.50008s/12 iters), loss = 2.31246
I0428 15:23:38.157461 29440 solver.cpp:237]     Train net output #0: loss = 2.31246 (* 1 = 2.31246 loss)
I0428 15:23:38.157470 29440 sgd_solver.cpp:105] Iteration 4464, lr = 0.00413022
I0428 15:23:43.252054 29440 solver.cpp:218] Iteration 4476 (2.35655 iter/s, 5.09219s/12 iters), loss = 2.13259
I0428 15:23:43.252091 29440 solver.cpp:237]     Train net output #0: loss = 2.13259 (* 1 = 2.13259 loss)
I0428 15:23:43.252100 29440 sgd_solver.cpp:105] Iteration 4476, lr = 0.00412041
I0428 15:23:48.244962 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_4488.caffemodel
I0428 15:23:49.841869 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_4488.solverstate
I0428 15:23:52.014493 29440 solver.cpp:330] Iteration 4488, Testing net (#0)
I0428 15:23:52.014590 29440 net.cpp:676] Ignoring source layer train-data
I0428 15:23:54.882165 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:23:56.832676 29440 solver.cpp:397]     Test net output #0: accuracy = 0.234681
I0428 15:23:56.832706 29440 solver.cpp:397]     Test net output #1: loss = 3.22078 (* 1 = 3.22078 loss)
I0428 15:23:57.171466 29440 solver.cpp:218] Iteration 4488 (0.862141 iter/s, 13.9188s/12 iters), loss = 2.24273
I0428 15:23:57.171512 29440 solver.cpp:237]     Train net output #0: loss = 2.24273 (* 1 = 2.24273 loss)
I0428 15:23:57.171522 29440 sgd_solver.cpp:105] Iteration 4488, lr = 0.00411063
I0428 15:24:01.944043 29440 solver.cpp:218] Iteration 4500 (2.51449 iter/s, 4.77234s/12 iters), loss = 2.33179
I0428 15:24:01.944089 29440 solver.cpp:237]     Train net output #0: loss = 2.33179 (* 1 = 2.33179 loss)
I0428 15:24:01.944097 29440 sgd_solver.cpp:105] Iteration 4500, lr = 0.00410087
I0428 15:24:07.835070 29440 solver.cpp:218] Iteration 4512 (2.03709 iter/s, 5.89075s/12 iters), loss = 2.55548
I0428 15:24:07.835112 29440 solver.cpp:237]     Train net output #0: loss = 2.55548 (* 1 = 2.55548 loss)
I0428 15:24:07.835120 29440 sgd_solver.cpp:105] Iteration 4512, lr = 0.00409113
I0428 15:24:13.684821 29440 solver.cpp:218] Iteration 4524 (2.05146 iter/s, 5.84948s/12 iters), loss = 2.24476
I0428 15:24:13.684864 29440 solver.cpp:237]     Train net output #0: loss = 2.24476 (* 1 = 2.24476 loss)
I0428 15:24:13.684871 29440 sgd_solver.cpp:105] Iteration 4524, lr = 0.00408142
I0428 15:24:18.976038 29440 solver.cpp:218] Iteration 4536 (2.26802 iter/s, 5.29096s/12 iters), loss = 2.48508
I0428 15:24:18.976083 29440 solver.cpp:237]     Train net output #0: loss = 2.48508 (* 1 = 2.48508 loss)
I0428 15:24:18.976091 29440 sgd_solver.cpp:105] Iteration 4536, lr = 0.00407173
I0428 15:24:24.559517 29440 solver.cpp:218] Iteration 4548 (2.15013 iter/s, 5.58105s/12 iters), loss = 1.9555
I0428 15:24:24.559649 29440 solver.cpp:237]     Train net output #0: loss = 1.9555 (* 1 = 1.9555 loss)
I0428 15:24:24.559659 29440 sgd_solver.cpp:105] Iteration 4548, lr = 0.00406206
I0428 15:24:25.611539 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:24:29.908747 29440 solver.cpp:218] Iteration 4560 (2.24434 iter/s, 5.34678s/12 iters), loss = 2.25039
I0428 15:24:29.908783 29440 solver.cpp:237]     Train net output #0: loss = 2.25039 (* 1 = 2.25039 loss)
I0428 15:24:29.908792 29440 sgd_solver.cpp:105] Iteration 4560, lr = 0.00405242
I0428 15:24:35.087370 29440 solver.cpp:218] Iteration 4572 (2.31833 iter/s, 5.17615s/12 iters), loss = 2.07345
I0428 15:24:35.087414 29440 solver.cpp:237]     Train net output #0: loss = 2.07345 (* 1 = 2.07345 loss)
I0428 15:24:35.087424 29440 sgd_solver.cpp:105] Iteration 4572, lr = 0.0040428
I0428 15:24:40.889459 29440 solver.cpp:218] Iteration 4584 (2.06832 iter/s, 5.80181s/12 iters), loss = 2.1015
I0428 15:24:40.889503 29440 solver.cpp:237]     Train net output #0: loss = 2.1015 (* 1 = 2.1015 loss)
I0428 15:24:40.889510 29440 sgd_solver.cpp:105] Iteration 4584, lr = 0.0040332
I0428 15:24:43.142110 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_4590.caffemodel
I0428 15:24:44.570080 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_4590.solverstate
I0428 15:24:45.636201 29440 solver.cpp:330] Iteration 4590, Testing net (#0)
I0428 15:24:45.636226 29440 net.cpp:676] Ignoring source layer train-data
I0428 15:24:48.548655 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:24:50.552683 29440 solver.cpp:397]     Test net output #0: accuracy = 0.26348
I0428 15:24:50.552703 29440 solver.cpp:397]     Test net output #1: loss = 3.20357 (* 1 = 3.20357 loss)
I0428 15:24:52.678846 29440 solver.cpp:218] Iteration 4596 (1.01791 iter/s, 11.7889s/12 iters), loss = 2.08352
I0428 15:24:52.678891 29440 solver.cpp:237]     Train net output #0: loss = 2.08352 (* 1 = 2.08352 loss)
I0428 15:24:52.678900 29440 sgd_solver.cpp:105] Iteration 4596, lr = 0.00402362
I0428 15:24:58.079295 29440 solver.cpp:218] Iteration 4608 (2.22305 iter/s, 5.39799s/12 iters), loss = 2.2421
I0428 15:24:58.079401 29440 solver.cpp:237]     Train net output #0: loss = 2.2421 (* 1 = 2.2421 loss)
I0428 15:24:58.079411 29440 sgd_solver.cpp:105] Iteration 4608, lr = 0.00401407
I0428 15:25:03.393265 29440 solver.cpp:218] Iteration 4620 (2.25833 iter/s, 5.31365s/12 iters), loss = 2.26613
I0428 15:25:03.393307 29440 solver.cpp:237]     Train net output #0: loss = 2.26613 (* 1 = 2.26613 loss)
I0428 15:25:03.393316 29440 sgd_solver.cpp:105] Iteration 4620, lr = 0.00400454
I0428 15:25:09.101722 29440 solver.cpp:218] Iteration 4632 (2.10224 iter/s, 5.70819s/12 iters), loss = 2.28633
I0428 15:25:09.101763 29440 solver.cpp:237]     Train net output #0: loss = 2.28633 (* 1 = 2.28633 loss)
I0428 15:25:09.101773 29440 sgd_solver.cpp:105] Iteration 4632, lr = 0.00399503
I0428 15:25:14.769191 29440 solver.cpp:218] Iteration 4644 (2.11745 iter/s, 5.6672s/12 iters), loss = 2.02871
I0428 15:25:14.769230 29440 solver.cpp:237]     Train net output #0: loss = 2.02871 (* 1 = 2.02871 loss)
I0428 15:25:14.769238 29440 sgd_solver.cpp:105] Iteration 4644, lr = 0.00398555
I0428 15:25:18.193325 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:25:20.113991 29440 solver.cpp:218] Iteration 4656 (2.24621 iter/s, 5.34234s/12 iters), loss = 2.14562
I0428 15:25:20.114030 29440 solver.cpp:237]     Train net output #0: loss = 2.14562 (* 1 = 2.14562 loss)
I0428 15:25:20.114039 29440 sgd_solver.cpp:105] Iteration 4656, lr = 0.00397608
I0428 15:25:26.028251 29440 solver.cpp:218] Iteration 4668 (2.02909 iter/s, 5.91398s/12 iters), loss = 2.15853
I0428 15:25:26.028295 29440 solver.cpp:237]     Train net output #0: loss = 2.15853 (* 1 = 2.15853 loss)
I0428 15:25:26.028304 29440 sgd_solver.cpp:105] Iteration 4668, lr = 0.00396664
I0428 15:25:31.951579 29440 solver.cpp:218] Iteration 4680 (2.02598 iter/s, 5.92305s/12 iters), loss = 2.12362
I0428 15:25:31.951725 29440 solver.cpp:237]     Train net output #0: loss = 2.12362 (* 1 = 2.12362 loss)
I0428 15:25:31.951736 29440 sgd_solver.cpp:105] Iteration 4680, lr = 0.00395723
I0428 15:25:36.284216 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_4692.caffemodel
I0428 15:25:41.454048 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_4692.solverstate
I0428 15:25:43.599952 29440 solver.cpp:330] Iteration 4692, Testing net (#0)
I0428 15:25:43.599980 29440 net.cpp:676] Ignoring source layer train-data
I0428 15:25:46.303839 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:25:48.315052 29440 solver.cpp:397]     Test net output #0: accuracy = 0.240196
I0428 15:25:48.315086 29440 solver.cpp:397]     Test net output #1: loss = 3.28912 (* 1 = 3.28912 loss)
I0428 15:25:48.634694 29440 solver.cpp:218] Iteration 4692 (0.719414 iter/s, 16.6802s/12 iters), loss = 2.1729
I0428 15:25:48.636462 29440 solver.cpp:237]     Train net output #0: loss = 2.1729 (* 1 = 2.1729 loss)
I0428 15:25:48.636474 29440 sgd_solver.cpp:105] Iteration 4692, lr = 0.00394783
I0428 15:25:53.557381 29440 solver.cpp:218] Iteration 4704 (2.43867 iter/s, 4.92072s/12 iters), loss = 2.18563
I0428 15:25:53.557425 29440 solver.cpp:237]     Train net output #0: loss = 2.18563 (* 1 = 2.18563 loss)
I0428 15:25:53.557435 29440 sgd_solver.cpp:105] Iteration 4704, lr = 0.00393846
I0428 15:25:59.054739 29440 solver.cpp:218] Iteration 4716 (2.18297 iter/s, 5.49709s/12 iters), loss = 2.15072
I0428 15:25:59.054786 29440 solver.cpp:237]     Train net output #0: loss = 2.15072 (* 1 = 2.15072 loss)
I0428 15:25:59.054795 29440 sgd_solver.cpp:105] Iteration 4716, lr = 0.00392911
I0428 15:26:04.427770 29440 solver.cpp:218] Iteration 4728 (2.23349 iter/s, 5.37277s/12 iters), loss = 2.22398
I0428 15:26:04.427884 29440 solver.cpp:237]     Train net output #0: loss = 2.22398 (* 1 = 2.22398 loss)
I0428 15:26:04.427894 29440 sgd_solver.cpp:105] Iteration 4728, lr = 0.00391978
I0428 15:26:10.033138 29440 solver.cpp:218] Iteration 4740 (2.14093 iter/s, 5.60504s/12 iters), loss = 2.1284
I0428 15:26:10.033174 29440 solver.cpp:237]     Train net output #0: loss = 2.1284 (* 1 = 2.1284 loss)
I0428 15:26:10.033182 29440 sgd_solver.cpp:105] Iteration 4740, lr = 0.00391047
I0428 15:26:15.542104 29440 solver.cpp:218] Iteration 4752 (2.17924 iter/s, 5.5065s/12 iters), loss = 2.11902
I0428 15:26:15.542156 29440 solver.cpp:237]     Train net output #0: loss = 2.11902 (* 1 = 2.11902 loss)
I0428 15:26:15.542166 29440 sgd_solver.cpp:105] Iteration 4752, lr = 0.00390119
I0428 15:26:15.973796 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:26:20.955911 29440 solver.cpp:218] Iteration 4764 (2.21667 iter/s, 5.41354s/12 iters), loss = 2.04573
I0428 15:26:20.955952 29440 solver.cpp:237]     Train net output #0: loss = 2.04573 (* 1 = 2.04573 loss)
I0428 15:26:20.955961 29440 sgd_solver.cpp:105] Iteration 4764, lr = 0.00389193
I0428 15:26:26.327666 29440 solver.cpp:218] Iteration 4776 (2.23401 iter/s, 5.3715s/12 iters), loss = 2.17585
I0428 15:26:26.327709 29440 solver.cpp:237]     Train net output #0: loss = 2.17585 (* 1 = 2.17585 loss)
I0428 15:26:26.327718 29440 sgd_solver.cpp:105] Iteration 4776, lr = 0.00388269
I0428 15:26:31.547629 29440 solver.cpp:218] Iteration 4788 (2.29996 iter/s, 5.21749s/12 iters), loss = 2.15999
I0428 15:26:31.547673 29440 solver.cpp:237]     Train net output #0: loss = 2.15999 (* 1 = 2.15999 loss)
I0428 15:26:31.547683 29440 sgd_solver.cpp:105] Iteration 4788, lr = 0.00387347
I0428 15:26:33.742202 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_4794.caffemodel
I0428 15:26:36.413092 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_4794.solverstate
I0428 15:26:38.919865 29440 solver.cpp:330] Iteration 4794, Testing net (#0)
I0428 15:26:38.919884 29440 net.cpp:676] Ignoring source layer train-data
I0428 15:26:41.612107 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:26:43.795096 29440 solver.cpp:397]     Test net output #0: accuracy = 0.244485
I0428 15:26:43.795145 29440 solver.cpp:397]     Test net output #1: loss = 3.36583 (* 1 = 3.36583 loss)
I0428 15:26:46.268167 29440 solver.cpp:218] Iteration 4800 (0.815343 iter/s, 14.7177s/12 iters), loss = 2.17011
I0428 15:26:46.268213 29440 solver.cpp:237]     Train net output #0: loss = 2.17011 (* 1 = 2.17011 loss)
I0428 15:26:46.268221 29440 sgd_solver.cpp:105] Iteration 4800, lr = 0.00386427
I0428 15:26:51.925525 29440 solver.cpp:218] Iteration 4812 (2.12205 iter/s, 5.65491s/12 iters), loss = 2.113
I0428 15:26:51.925570 29440 solver.cpp:237]     Train net output #0: loss = 2.113 (* 1 = 2.113 loss)
I0428 15:26:51.925580 29440 sgd_solver.cpp:105] Iteration 4812, lr = 0.0038551
I0428 15:26:57.246853 29440 solver.cpp:218] Iteration 4824 (2.25519 iter/s, 5.32107s/12 iters), loss = 2.23668
I0428 15:26:57.246901 29440 solver.cpp:237]     Train net output #0: loss = 2.23668 (* 1 = 2.23668 loss)
I0428 15:26:57.246909 29440 sgd_solver.cpp:105] Iteration 4824, lr = 0.00384594
I0428 15:27:02.441546 29440 solver.cpp:218] Iteration 4836 (2.31017 iter/s, 5.19443s/12 iters), loss = 2.22138
I0428 15:27:02.441599 29440 solver.cpp:237]     Train net output #0: loss = 2.22138 (* 1 = 2.22138 loss)
I0428 15:27:02.441609 29440 sgd_solver.cpp:105] Iteration 4836, lr = 0.00383681
I0428 15:27:07.742434 29440 solver.cpp:218] Iteration 4848 (2.26388 iter/s, 5.30063s/12 iters), loss = 2.10136
I0428 15:27:07.742529 29440 solver.cpp:237]     Train net output #0: loss = 2.10136 (* 1 = 2.10136 loss)
I0428 15:27:07.742538 29440 sgd_solver.cpp:105] Iteration 4848, lr = 0.0038277
I0428 15:27:10.601469 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:27:13.148444 29440 solver.cpp:218] Iteration 4860 (2.21988 iter/s, 5.4057s/12 iters), loss = 2.0892
I0428 15:27:13.148512 29440 solver.cpp:237]     Train net output #0: loss = 2.0892 (* 1 = 2.0892 loss)
I0428 15:27:13.148522 29440 sgd_solver.cpp:105] Iteration 4860, lr = 0.00381862
I0428 15:27:16.345008 29440 blocking_queue.cpp:49] Waiting for data
I0428 15:27:18.748229 29440 solver.cpp:218] Iteration 4872 (2.14305 iter/s, 5.59949s/12 iters), loss = 2.05493
I0428 15:27:18.748278 29440 solver.cpp:237]     Train net output #0: loss = 2.05493 (* 1 = 2.05493 loss)
I0428 15:27:18.748288 29440 sgd_solver.cpp:105] Iteration 4872, lr = 0.00380955
I0428 15:27:24.297677 29440 solver.cpp:218] Iteration 4884 (2.16248 iter/s, 5.54918s/12 iters), loss = 2.04115
I0428 15:27:24.297719 29440 solver.cpp:237]     Train net output #0: loss = 2.04115 (* 1 = 2.04115 loss)
I0428 15:27:24.297729 29440 sgd_solver.cpp:105] Iteration 4884, lr = 0.0038005
I0428 15:27:28.840148 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_4896.caffemodel
I0428 15:27:30.904321 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_4896.solverstate
I0428 15:27:32.329084 29440 solver.cpp:330] Iteration 4896, Testing net (#0)
I0428 15:27:32.329107 29440 net.cpp:676] Ignoring source layer train-data
I0428 15:27:35.051887 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:27:37.221065 29440 solver.cpp:397]     Test net output #0: accuracy = 0.25674
I0428 15:27:37.221096 29440 solver.cpp:397]     Test net output #1: loss = 3.27323 (* 1 = 3.27323 loss)
I0428 15:27:37.565515 29440 solver.cpp:218] Iteration 4896 (0.90463 iter/s, 13.2651s/12 iters), loss = 2.04885
I0428 15:27:37.567126 29440 solver.cpp:237]     Train net output #0: loss = 2.04885 (* 1 = 2.04885 loss)
I0428 15:27:37.567137 29440 sgd_solver.cpp:105] Iteration 4896, lr = 0.00379148
I0428 15:27:42.513273 29440 solver.cpp:218] Iteration 4908 (2.42623 iter/s, 4.94594s/12 iters), loss = 2.1563
I0428 15:27:42.513422 29440 solver.cpp:237]     Train net output #0: loss = 2.1563 (* 1 = 2.1563 loss)
I0428 15:27:42.513432 29440 sgd_solver.cpp:105] Iteration 4908, lr = 0.00378248
I0428 15:27:48.031141 29440 solver.cpp:218] Iteration 4920 (2.17572 iter/s, 5.51542s/12 iters), loss = 2.02529
I0428 15:27:48.031181 29440 solver.cpp:237]     Train net output #0: loss = 2.02529 (* 1 = 2.02529 loss)
I0428 15:27:48.031189 29440 sgd_solver.cpp:105] Iteration 4920, lr = 0.0037735
I0428 15:27:53.525934 29440 solver.cpp:218] Iteration 4932 (2.18399 iter/s, 5.49453s/12 iters), loss = 2.28954
I0428 15:27:53.525974 29440 solver.cpp:237]     Train net output #0: loss = 2.28954 (* 1 = 2.28954 loss)
I0428 15:27:53.525982 29440 sgd_solver.cpp:105] Iteration 4932, lr = 0.00376454
I0428 15:27:59.375576 29440 solver.cpp:218] Iteration 4944 (2.0515 iter/s, 5.84937s/12 iters), loss = 1.90498
I0428 15:27:59.375617 29440 solver.cpp:237]     Train net output #0: loss = 1.90498 (* 1 = 1.90498 loss)
I0428 15:27:59.375625 29440 sgd_solver.cpp:105] Iteration 4944, lr = 0.0037556
I0428 15:28:04.586035 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:28:05.027817 29440 solver.cpp:218] Iteration 4956 (2.12396 iter/s, 5.64984s/12 iters), loss = 2.03415
I0428 15:28:05.027863 29440 solver.cpp:237]     Train net output #0: loss = 2.03415 (* 1 = 2.03415 loss)
I0428 15:28:05.027874 29440 sgd_solver.cpp:105] Iteration 4956, lr = 0.00374669
I0428 15:28:10.287456 29440 solver.cpp:218] Iteration 4968 (2.28259 iter/s, 5.25718s/12 iters), loss = 2.2164
I0428 15:28:10.287500 29440 solver.cpp:237]     Train net output #0: loss = 2.2164 (* 1 = 2.2164 loss)
I0428 15:28:10.287508 29440 sgd_solver.cpp:105] Iteration 4968, lr = 0.00373779
I0428 15:28:15.633642 29440 solver.cpp:218] Iteration 4980 (2.2447 iter/s, 5.34593s/12 iters), loss = 2.06599
I0428 15:28:15.633760 29440 solver.cpp:237]     Train net output #0: loss = 2.06599 (* 1 = 2.06599 loss)
I0428 15:28:15.633771 29440 sgd_solver.cpp:105] Iteration 4980, lr = 0.00372892
I0428 15:28:20.990123 29440 solver.cpp:218] Iteration 4992 (2.24042 iter/s, 5.35615s/12 iters), loss = 1.85347
I0428 15:28:20.990166 29440 solver.cpp:237]     Train net output #0: loss = 1.85347 (* 1 = 1.85347 loss)
I0428 15:28:20.990175 29440 sgd_solver.cpp:105] Iteration 4992, lr = 0.00372006
I0428 15:28:23.214367 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_4998.caffemodel
I0428 15:28:28.689630 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_4998.solverstate
I0428 15:28:32.032187 29440 solver.cpp:330] Iteration 4998, Testing net (#0)
I0428 15:28:32.032208 29440 net.cpp:676] Ignoring source layer train-data
I0428 15:28:34.790056 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:28:36.975805 29440 solver.cpp:397]     Test net output #0: accuracy = 0.255515
I0428 15:28:36.975834 29440 solver.cpp:397]     Test net output #1: loss = 3.29295 (* 1 = 3.29295 loss)
I0428 15:28:39.238909 29440 solver.cpp:218] Iteration 5004 (0.657604 iter/s, 18.2481s/12 iters), loss = 2.45524
I0428 15:28:39.238955 29440 solver.cpp:237]     Train net output #0: loss = 2.45524 (* 1 = 2.45524 loss)
I0428 15:28:39.238965 29440 sgd_solver.cpp:105] Iteration 5004, lr = 0.00371123
I0428 15:28:44.089092 29440 solver.cpp:218] Iteration 5016 (2.47538 iter/s, 4.84773s/12 iters), loss = 2.16866
I0428 15:28:44.089136 29440 solver.cpp:237]     Train net output #0: loss = 2.16866 (* 1 = 2.16866 loss)
I0428 15:28:44.089145 29440 sgd_solver.cpp:105] Iteration 5016, lr = 0.00370242
I0428 15:28:49.776115 29440 solver.cpp:218] Iteration 5028 (2.11017 iter/s, 5.68676s/12 iters), loss = 2.14202
I0428 15:28:49.776232 29440 solver.cpp:237]     Train net output #0: loss = 2.14202 (* 1 = 2.14202 loss)
I0428 15:28:49.776242 29440 sgd_solver.cpp:105] Iteration 5028, lr = 0.00369363
I0428 15:28:55.576730 29440 solver.cpp:218] Iteration 5040 (2.06963 iter/s, 5.79813s/12 iters), loss = 2.32066
I0428 15:28:55.576773 29440 solver.cpp:237]     Train net output #0: loss = 2.32066 (* 1 = 2.32066 loss)
I0428 15:28:55.576782 29440 sgd_solver.cpp:105] Iteration 5040, lr = 0.00368486
I0428 15:29:01.146261 29440 solver.cpp:218] Iteration 5052 (2.15554 iter/s, 5.56706s/12 iters), loss = 1.77872
I0428 15:29:01.146306 29440 solver.cpp:237]     Train net output #0: loss = 1.77872 (* 1 = 1.77872 loss)
I0428 15:29:01.146315 29440 sgd_solver.cpp:105] Iteration 5052, lr = 0.00367611
I0428 15:29:03.050596 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:29:06.588162 29440 solver.cpp:218] Iteration 5064 (2.20611 iter/s, 5.43945s/12 iters), loss = 1.86176
I0428 15:29:06.588208 29440 solver.cpp:237]     Train net output #0: loss = 1.86176 (* 1 = 1.86176 loss)
I0428 15:29:06.588217 29440 sgd_solver.cpp:105] Iteration 5064, lr = 0.00366738
I0428 15:29:12.057000 29440 solver.cpp:218] Iteration 5076 (2.19436 iter/s, 5.46857s/12 iters), loss = 1.89957
I0428 15:29:12.057054 29440 solver.cpp:237]     Train net output #0: loss = 1.89957 (* 1 = 1.89957 loss)
I0428 15:29:12.057067 29440 sgd_solver.cpp:105] Iteration 5076, lr = 0.00365868
I0428 15:29:17.069263 29440 solver.cpp:218] Iteration 5088 (2.39529 iter/s, 5.00982s/12 iters), loss = 2.01005
I0428 15:29:17.069309 29440 solver.cpp:237]     Train net output #0: loss = 2.01005 (* 1 = 2.01005 loss)
I0428 15:29:17.069319 29440 sgd_solver.cpp:105] Iteration 5088, lr = 0.00364999
I0428 15:29:22.098765 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_5100.caffemodel
I0428 15:29:30.539505 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_5100.solverstate
I0428 15:29:34.895390 29440 solver.cpp:330] Iteration 5100, Testing net (#0)
I0428 15:29:34.895421 29440 net.cpp:676] Ignoring source layer train-data
I0428 15:29:37.571724 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:29:39.773151 29440 solver.cpp:397]     Test net output #0: accuracy = 0.265931
I0428 15:29:39.773180 29440 solver.cpp:397]     Test net output #1: loss = 3.2293 (* 1 = 3.2293 loss)
I0428 15:29:39.915881 29440 solver.cpp:218] Iteration 5100 (0.525263 iter/s, 22.8457s/12 iters), loss = 2.07243
I0428 15:29:39.915941 29440 solver.cpp:237]     Train net output #0: loss = 2.07243 (* 1 = 2.07243 loss)
I0428 15:29:39.915951 29440 sgd_solver.cpp:105] Iteration 5100, lr = 0.00364132
I0428 15:29:45.105358 29440 solver.cpp:218] Iteration 5112 (2.31249 iter/s, 5.18921s/12 iters), loss = 2.05048
I0428 15:29:45.105401 29440 solver.cpp:237]     Train net output #0: loss = 2.05048 (* 1 = 2.05048 loss)
I0428 15:29:45.105410 29440 sgd_solver.cpp:105] Iteration 5112, lr = 0.00363268
I0428 15:29:50.550194 29440 solver.cpp:218] Iteration 5124 (2.20403 iter/s, 5.44457s/12 iters), loss = 2.06396
I0428 15:29:50.550249 29440 solver.cpp:237]     Train net output #0: loss = 2.06396 (* 1 = 2.06396 loss)
I0428 15:29:50.550262 29440 sgd_solver.cpp:105] Iteration 5124, lr = 0.00362405
I0428 15:29:56.198527 29440 solver.cpp:218] Iteration 5136 (2.12462 iter/s, 5.64806s/12 iters), loss = 2.1111
I0428 15:29:56.199118 29440 solver.cpp:237]     Train net output #0: loss = 2.1111 (* 1 = 2.1111 loss)
I0428 15:29:56.199127 29440 sgd_solver.cpp:105] Iteration 5136, lr = 0.00361545
I0428 15:30:01.871013 29440 solver.cpp:218] Iteration 5148 (2.1164 iter/s, 5.67s/12 iters), loss = 1.89722
I0428 15:30:01.871057 29440 solver.cpp:237]     Train net output #0: loss = 1.89722 (* 1 = 1.89722 loss)
I0428 15:30:01.871065 29440 sgd_solver.cpp:105] Iteration 5148, lr = 0.00360687
I0428 15:30:06.249850 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:30:07.627324 29440 solver.cpp:218] Iteration 5160 (2.08477 iter/s, 5.75604s/12 iters), loss = 1.74849
I0428 15:30:07.627363 29440 solver.cpp:237]     Train net output #0: loss = 1.74849 (* 1 = 1.74849 loss)
I0428 15:30:07.627372 29440 sgd_solver.cpp:105] Iteration 5160, lr = 0.0035983
I0428 15:30:12.924854 29440 solver.cpp:218] Iteration 5172 (2.26626 iter/s, 5.29506s/12 iters), loss = 2.25222
I0428 15:30:12.924912 29440 solver.cpp:237]     Train net output #0: loss = 2.25222 (* 1 = 2.25222 loss)
I0428 15:30:12.924924 29440 sgd_solver.cpp:105] Iteration 5172, lr = 0.00358976
I0428 15:30:18.415328 29440 solver.cpp:218] Iteration 5184 (2.18571 iter/s, 5.4902s/12 iters), loss = 1.85497
I0428 15:30:18.415380 29440 solver.cpp:237]     Train net output #0: loss = 1.85497 (* 1 = 1.85497 loss)
I0428 15:30:18.415390 29440 sgd_solver.cpp:105] Iteration 5184, lr = 0.00358124
I0428 15:30:23.447762 29440 solver.cpp:218] Iteration 5196 (2.38569 iter/s, 5.03s/12 iters), loss = 1.83312
I0428 15:30:23.447805 29440 solver.cpp:237]     Train net output #0: loss = 1.83312 (* 1 = 1.83312 loss)
I0428 15:30:23.447814 29440 sgd_solver.cpp:105] Iteration 5196, lr = 0.00357273
I0428 15:30:25.736388 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_5202.caffemodel
I0428 15:30:31.311079 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_5202.solverstate
I0428 15:30:33.036942 29440 solver.cpp:330] Iteration 5202, Testing net (#0)
I0428 15:30:33.036963 29440 net.cpp:676] Ignoring source layer train-data
I0428 15:30:35.663861 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:30:38.026263 29440 solver.cpp:397]     Test net output #0: accuracy = 0.271446
I0428 15:30:38.026298 29440 solver.cpp:397]     Test net output #1: loss = 3.20422 (* 1 = 3.20422 loss)
I0428 15:30:39.911571 29440 solver.cpp:218] Iteration 5208 (0.728901 iter/s, 16.4631s/12 iters), loss = 1.8668
I0428 15:30:39.911617 29440 solver.cpp:237]     Train net output #0: loss = 1.8668 (* 1 = 1.8668 loss)
I0428 15:30:39.911626 29440 sgd_solver.cpp:105] Iteration 5208, lr = 0.00356425
I0428 15:30:45.272857 29440 solver.cpp:218] Iteration 5220 (2.23838 iter/s, 5.36103s/12 iters), loss = 2.10678
I0428 15:30:45.272899 29440 solver.cpp:237]     Train net output #0: loss = 2.10678 (* 1 = 2.10678 loss)
I0428 15:30:45.272907 29440 sgd_solver.cpp:105] Iteration 5220, lr = 0.00355579
I0428 15:30:50.737533 29440 solver.cpp:218] Iteration 5232 (2.19602 iter/s, 5.46442s/12 iters), loss = 1.76131
I0428 15:30:50.737573 29440 solver.cpp:237]     Train net output #0: loss = 1.76131 (* 1 = 1.76131 loss)
I0428 15:30:50.737581 29440 sgd_solver.cpp:105] Iteration 5232, lr = 0.00354735
I0428 15:30:56.088945 29440 solver.cpp:218] Iteration 5244 (2.2425 iter/s, 5.35116s/12 iters), loss = 2.07791
I0428 15:30:56.088984 29440 solver.cpp:237]     Train net output #0: loss = 2.07791 (* 1 = 2.07791 loss)
I0428 15:30:56.088994 29440 sgd_solver.cpp:105] Iteration 5244, lr = 0.00353892
I0428 15:31:01.500350 29440 solver.cpp:218] Iteration 5256 (2.21855 iter/s, 5.40894s/12 iters), loss = 1.66355
I0428 15:31:01.500459 29440 solver.cpp:237]     Train net output #0: loss = 1.66355 (* 1 = 1.66355 loss)
I0428 15:31:01.500469 29440 sgd_solver.cpp:105] Iteration 5256, lr = 0.00353052
I0428 15:31:02.715701 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:31:07.059244 29440 solver.cpp:218] Iteration 5268 (2.15967 iter/s, 5.55642s/12 iters), loss = 1.81705
I0428 15:31:07.059294 29440 solver.cpp:237]     Train net output #0: loss = 1.81705 (* 1 = 1.81705 loss)
I0428 15:31:07.059305 29440 sgd_solver.cpp:105] Iteration 5268, lr = 0.00352214
I0428 15:31:12.667135 29440 solver.cpp:218] Iteration 5280 (2.13995 iter/s, 5.60762s/12 iters), loss = 1.90106
I0428 15:31:12.667181 29440 solver.cpp:237]     Train net output #0: loss = 1.90106 (* 1 = 1.90106 loss)
I0428 15:31:12.667191 29440 sgd_solver.cpp:105] Iteration 5280, lr = 0.00351378
I0428 15:31:18.203730 29440 solver.cpp:218] Iteration 5292 (2.1675 iter/s, 5.53633s/12 iters), loss = 1.65302
I0428 15:31:18.203775 29440 solver.cpp:237]     Train net output #0: loss = 1.65302 (* 1 = 1.65302 loss)
I0428 15:31:18.203784 29440 sgd_solver.cpp:105] Iteration 5292, lr = 0.00350544
I0428 15:31:23.105813 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_5304.caffemodel
I0428 15:31:27.074326 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_5304.solverstate
I0428 15:31:30.985285 29440 solver.cpp:330] Iteration 5304, Testing net (#0)
I0428 15:31:30.985303 29440 net.cpp:676] Ignoring source layer train-data
I0428 15:31:33.453176 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:31:35.843083 29440 solver.cpp:397]     Test net output #0: accuracy = 0.29473
I0428 15:31:35.843123 29440 solver.cpp:397]     Test net output #1: loss = 3.12903 (* 1 = 3.12903 loss)
I0428 15:31:35.985560 29440 solver.cpp:218] Iteration 5304 (0.674875 iter/s, 17.7811s/12 iters), loss = 1.9119
I0428 15:31:35.985611 29440 solver.cpp:237]     Train net output #0: loss = 1.9119 (* 1 = 1.9119 loss)
I0428 15:31:35.985625 29440 sgd_solver.cpp:105] Iteration 5304, lr = 0.00349711
I0428 15:31:40.631669 29440 solver.cpp:218] Iteration 5316 (2.58296 iter/s, 4.64584s/12 iters), loss = 1.81314
I0428 15:31:40.631711 29440 solver.cpp:237]     Train net output #0: loss = 1.81314 (* 1 = 1.81314 loss)
I0428 15:31:40.631721 29440 sgd_solver.cpp:105] Iteration 5316, lr = 0.00348881
I0428 15:31:45.958386 29440 solver.cpp:218] Iteration 5328 (2.25292 iter/s, 5.32643s/12 iters), loss = 1.75395
I0428 15:31:45.958429 29440 solver.cpp:237]     Train net output #0: loss = 1.75395 (* 1 = 1.75395 loss)
I0428 15:31:45.958438 29440 sgd_solver.cpp:105] Iteration 5328, lr = 0.00348053
I0428 15:31:51.403501 29440 solver.cpp:218] Iteration 5340 (2.20393 iter/s, 5.44482s/12 iters), loss = 1.60211
I0428 15:31:51.403546 29440 solver.cpp:237]     Train net output #0: loss = 1.60211 (* 1 = 1.60211 loss)
I0428 15:31:51.403555 29440 sgd_solver.cpp:105] Iteration 5340, lr = 0.00347226
I0428 15:31:57.062594 29440 solver.cpp:218] Iteration 5352 (2.12059 iter/s, 5.65879s/12 iters), loss = 1.55615
I0428 15:31:57.062638 29440 solver.cpp:237]     Train net output #0: loss = 1.55615 (* 1 = 1.55615 loss)
I0428 15:31:57.062647 29440 sgd_solver.cpp:105] Iteration 5352, lr = 0.00346402
I0428 15:32:00.677070 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:32:02.409791 29440 solver.cpp:218] Iteration 5364 (2.24429 iter/s, 5.34691s/12 iters), loss = 1.99052
I0428 15:32:02.409834 29440 solver.cpp:237]     Train net output #0: loss = 1.99052 (* 1 = 1.99052 loss)
I0428 15:32:02.409843 29440 sgd_solver.cpp:105] Iteration 5364, lr = 0.0034558
I0428 15:32:07.737442 29440 solver.cpp:218] Iteration 5376 (2.25252 iter/s, 5.32737s/12 iters), loss = 1.98938
I0428 15:32:07.737567 29440 solver.cpp:237]     Train net output #0: loss = 1.98938 (* 1 = 1.98938 loss)
I0428 15:32:07.737576 29440 sgd_solver.cpp:105] Iteration 5376, lr = 0.00344759
I0428 15:32:13.545972 29440 solver.cpp:218] Iteration 5388 (2.06606 iter/s, 5.80814s/12 iters), loss = 1.92841
I0428 15:32:13.546020 29440 solver.cpp:237]     Train net output #0: loss = 1.92841 (* 1 = 1.92841 loss)
I0428 15:32:13.546033 29440 sgd_solver.cpp:105] Iteration 5388, lr = 0.00343941
I0428 15:32:18.648471 29440 solver.cpp:218] Iteration 5400 (2.35293 iter/s, 5.10003s/12 iters), loss = 1.74162
I0428 15:32:18.648522 29440 solver.cpp:237]     Train net output #0: loss = 1.74162 (* 1 = 1.74162 loss)
I0428 15:32:18.648535 29440 sgd_solver.cpp:105] Iteration 5400, lr = 0.00343124
I0428 15:32:20.922966 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_5406.caffemodel
I0428 15:32:30.536072 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_5406.solverstate
I0428 15:32:35.858239 29440 solver.cpp:330] Iteration 5406, Testing net (#0)
I0428 15:32:35.858268 29440 net.cpp:676] Ignoring source layer train-data
I0428 15:32:38.333547 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:32:40.810235 29440 solver.cpp:397]     Test net output #0: accuracy = 0.262255
I0428 15:32:40.810286 29440 solver.cpp:397]     Test net output #1: loss = 3.23832 (* 1 = 3.23832 loss)
I0428 15:32:42.974061 29440 solver.cpp:218] Iteration 5412 (0.49333 iter/s, 24.3245s/12 iters), loss = 2.09238
I0428 15:32:42.974107 29440 solver.cpp:237]     Train net output #0: loss = 2.09238 (* 1 = 2.09238 loss)
I0428 15:32:42.974117 29440 sgd_solver.cpp:105] Iteration 5412, lr = 0.00342309
I0428 15:32:48.304339 29440 solver.cpp:218] Iteration 5424 (2.25141 iter/s, 5.32999s/12 iters), loss = 1.83193
I0428 15:32:48.304383 29440 solver.cpp:237]     Train net output #0: loss = 1.83193 (* 1 = 1.83193 loss)
I0428 15:32:48.304391 29440 sgd_solver.cpp:105] Iteration 5424, lr = 0.00341497
I0428 15:32:53.790885 29440 solver.cpp:218] Iteration 5436 (2.18729 iter/s, 5.48625s/12 iters), loss = 1.85197
I0428 15:32:53.790930 29440 solver.cpp:237]     Train net output #0: loss = 1.85197 (* 1 = 1.85197 loss)
I0428 15:32:53.790938 29440 sgd_solver.cpp:105] Iteration 5436, lr = 0.00340686
I0428 15:32:59.109948 29440 solver.cpp:218] Iteration 5448 (2.25616 iter/s, 5.31878s/12 iters), loss = 1.68398
I0428 15:32:59.109987 29440 solver.cpp:237]     Train net output #0: loss = 1.68398 (* 1 = 1.68398 loss)
I0428 15:32:59.109997 29440 sgd_solver.cpp:105] Iteration 5448, lr = 0.00339877
I0428 15:33:04.829753 29440 solver.cpp:218] Iteration 5460 (2.09808 iter/s, 5.71951s/12 iters), loss = 1.80686
I0428 15:33:04.829797 29440 solver.cpp:237]     Train net output #0: loss = 1.80686 (* 1 = 1.80686 loss)
I0428 15:33:04.829807 29440 sgd_solver.cpp:105] Iteration 5460, lr = 0.0033907
I0428 15:33:05.378872 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:33:10.170529 29440 solver.cpp:218] Iteration 5472 (2.24698 iter/s, 5.3405s/12 iters), loss = 2.0656
I0428 15:33:10.171320 29440 solver.cpp:237]     Train net output #0: loss = 2.0656 (* 1 = 2.0656 loss)
I0428 15:33:10.171330 29440 sgd_solver.cpp:105] Iteration 5472, lr = 0.00338265
I0428 15:33:15.519258 29440 solver.cpp:218] Iteration 5484 (2.24396 iter/s, 5.3477s/12 iters), loss = 1.9452
I0428 15:33:15.519304 29440 solver.cpp:237]     Train net output #0: loss = 1.9452 (* 1 = 1.9452 loss)
I0428 15:33:15.519312 29440 sgd_solver.cpp:105] Iteration 5484, lr = 0.00337462
I0428 15:33:21.196862 29440 solver.cpp:218] Iteration 5496 (2.11368 iter/s, 5.67731s/12 iters), loss = 1.75814
I0428 15:33:21.196899 29440 solver.cpp:237]     Train net output #0: loss = 1.75814 (* 1 = 1.75814 loss)
I0428 15:33:21.196907 29440 sgd_solver.cpp:105] Iteration 5496, lr = 0.00336661
I0428 15:33:25.890895 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_5508.caffemodel
I0428 15:33:32.719137 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_5508.solverstate
I0428 15:33:38.673027 29440 solver.cpp:330] Iteration 5508, Testing net (#0)
I0428 15:33:38.673048 29440 net.cpp:676] Ignoring source layer train-data
I0428 15:33:41.128252 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:33:43.559857 29440 solver.cpp:397]     Test net output #0: accuracy = 0.286765
I0428 15:33:43.559888 29440 solver.cpp:397]     Test net output #1: loss = 3.21944 (* 1 = 3.21944 loss)
I0428 15:33:43.899137 29440 solver.cpp:218] Iteration 5508 (0.528605 iter/s, 22.7013s/12 iters), loss = 1.85993
I0428 15:33:43.900743 29440 solver.cpp:237]     Train net output #0: loss = 1.85993 (* 1 = 1.85993 loss)
I0428 15:33:43.900754 29440 sgd_solver.cpp:105] Iteration 5508, lr = 0.00335861
I0428 15:33:48.603340 29440 solver.cpp:218] Iteration 5520 (2.5519 iter/s, 4.70239s/12 iters), loss = 1.88953
I0428 15:33:48.603385 29440 solver.cpp:237]     Train net output #0: loss = 1.88953 (* 1 = 1.88953 loss)
I0428 15:33:48.603394 29440 sgd_solver.cpp:105] Iteration 5520, lr = 0.00335064
I0428 15:33:54.167786 29440 solver.cpp:218] Iteration 5532 (2.15666 iter/s, 5.56415s/12 iters), loss = 1.61547
I0428 15:33:54.167826 29440 solver.cpp:237]     Train net output #0: loss = 1.61547 (* 1 = 1.61547 loss)
I0428 15:33:54.167834 29440 sgd_solver.cpp:105] Iteration 5532, lr = 0.00334268
I0428 15:33:59.937580 29440 solver.cpp:218] Iteration 5544 (2.08071 iter/s, 5.76727s/12 iters), loss = 1.68518
I0428 15:33:59.937623 29440 solver.cpp:237]     Train net output #0: loss = 1.68518 (* 1 = 1.68518 loss)
I0428 15:33:59.937631 29440 sgd_solver.cpp:105] Iteration 5544, lr = 0.00333475
I0428 15:34:05.174944 29440 solver.cpp:218] Iteration 5556 (2.29135 iter/s, 5.23709s/12 iters), loss = 1.73332
I0428 15:34:05.174983 29440 solver.cpp:237]     Train net output #0: loss = 1.73332 (* 1 = 1.73332 loss)
I0428 15:34:05.174991 29440 sgd_solver.cpp:105] Iteration 5556, lr = 0.00332683
I0428 15:34:05.175196 29440 blocking_queue.cpp:49] Waiting for data
I0428 15:34:08.095105 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:34:10.820518 29440 solver.cpp:218] Iteration 5568 (2.12568 iter/s, 5.64526s/12 iters), loss = 1.6544
I0428 15:34:10.820562 29440 solver.cpp:237]     Train net output #0: loss = 1.6544 (* 1 = 1.6544 loss)
I0428 15:34:10.820571 29440 sgd_solver.cpp:105] Iteration 5568, lr = 0.00331893
I0428 15:34:16.007496 29440 solver.cpp:218] Iteration 5580 (2.31457 iter/s, 5.18454s/12 iters), loss = 1.79852
I0428 15:34:16.007639 29440 solver.cpp:237]     Train net output #0: loss = 1.79852 (* 1 = 1.79852 loss)
I0428 15:34:16.007648 29440 sgd_solver.cpp:105] Iteration 5580, lr = 0.00331105
I0428 15:34:21.781874 29440 solver.cpp:218] Iteration 5592 (2.07829 iter/s, 5.77398s/12 iters), loss = 1.87208
I0428 15:34:21.781921 29440 solver.cpp:237]     Train net output #0: loss = 1.87208 (* 1 = 1.87208 loss)
I0428 15:34:21.781934 29440 sgd_solver.cpp:105] Iteration 5592, lr = 0.00330319
I0428 15:34:26.796452 29440 solver.cpp:218] Iteration 5604 (2.39421 iter/s, 5.01208s/12 iters), loss = 1.61089
I0428 15:34:26.796526 29440 solver.cpp:237]     Train net output #0: loss = 1.61089 (* 1 = 1.61089 loss)
I0428 15:34:26.796536 29440 sgd_solver.cpp:105] Iteration 5604, lr = 0.00329535
I0428 15:34:29.019589 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_5610.caffemodel
I0428 15:34:35.271191 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_5610.solverstate
I0428 15:34:38.176959 29440 solver.cpp:330] Iteration 5610, Testing net (#0)
I0428 15:34:38.176982 29440 net.cpp:676] Ignoring source layer train-data
I0428 15:34:40.662694 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:34:43.097023 29440 solver.cpp:397]     Test net output #0: accuracy = 0.286152
I0428 15:34:43.097049 29440 solver.cpp:397]     Test net output #1: loss = 3.17659 (* 1 = 3.17659 loss)
I0428 15:34:45.617691 29440 solver.cpp:218] Iteration 5616 (0.637607 iter/s, 18.8204s/12 iters), loss = 1.62821
I0428 15:34:45.617735 29440 solver.cpp:237]     Train net output #0: loss = 1.62821 (* 1 = 1.62821 loss)
I0428 15:34:45.617745 29440 sgd_solver.cpp:105] Iteration 5616, lr = 0.00328752
I0428 15:34:51.004881 29440 solver.cpp:218] Iteration 5628 (2.22762 iter/s, 5.38691s/12 iters), loss = 1.84744
I0428 15:34:51.004992 29440 solver.cpp:237]     Train net output #0: loss = 1.84744 (* 1 = 1.84744 loss)
I0428 15:34:51.005002 29440 sgd_solver.cpp:105] Iteration 5628, lr = 0.00327972
I0428 15:34:56.367995 29440 solver.cpp:218] Iteration 5640 (2.23854 iter/s, 5.36063s/12 iters), loss = 1.73615
I0428 15:34:56.368041 29440 solver.cpp:237]     Train net output #0: loss = 1.73615 (* 1 = 1.73615 loss)
I0428 15:34:56.368048 29440 sgd_solver.cpp:105] Iteration 5640, lr = 0.00327193
I0428 15:35:01.681520 29440 solver.cpp:218] Iteration 5652 (2.25851 iter/s, 5.31324s/12 iters), loss = 1.50181
I0428 15:35:01.681561 29440 solver.cpp:237]     Train net output #0: loss = 1.50181 (* 1 = 1.50181 loss)
I0428 15:35:01.681569 29440 sgd_solver.cpp:105] Iteration 5652, lr = 0.00326416
I0428 15:35:06.704794 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:35:07.141897 29440 solver.cpp:218] Iteration 5664 (2.19865 iter/s, 5.45788s/12 iters), loss = 1.95614
I0428 15:35:07.141940 29440 solver.cpp:237]     Train net output #0: loss = 1.95614 (* 1 = 1.95614 loss)
I0428 15:35:07.141949 29440 sgd_solver.cpp:105] Iteration 5664, lr = 0.00325641
I0428 15:35:12.299834 29440 solver.cpp:218] Iteration 5676 (2.32765 iter/s, 5.15542s/12 iters), loss = 1.74226
I0428 15:35:12.299880 29440 solver.cpp:237]     Train net output #0: loss = 1.74226 (* 1 = 1.74226 loss)
I0428 15:35:12.299887 29440 sgd_solver.cpp:105] Iteration 5676, lr = 0.00324868
I0428 15:35:17.644448 29440 solver.cpp:218] Iteration 5688 (2.24537 iter/s, 5.34434s/12 iters), loss = 1.94369
I0428 15:35:17.644510 29440 solver.cpp:237]     Train net output #0: loss = 1.94369 (* 1 = 1.94369 loss)
I0428 15:35:17.644520 29440 sgd_solver.cpp:105] Iteration 5688, lr = 0.00324097
I0428 15:35:22.878484 29440 solver.cpp:218] Iteration 5700 (2.2928 iter/s, 5.23377s/12 iters), loss = 1.4532
I0428 15:35:22.879547 29440 solver.cpp:237]     Train net output #0: loss = 1.4532 (* 1 = 1.4532 loss)
I0428 15:35:22.879557 29440 sgd_solver.cpp:105] Iteration 5700, lr = 0.00323328
I0428 15:35:27.851889 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_5712.caffemodel
I0428 15:35:30.443333 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_5712.solverstate
I0428 15:35:37.989817 29440 solver.cpp:330] Iteration 5712, Testing net (#0)
I0428 15:35:37.989840 29440 net.cpp:676] Ignoring source layer train-data
I0428 15:35:40.360159 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:35:42.836150 29440 solver.cpp:397]     Test net output #0: accuracy = 0.309436
I0428 15:35:42.836179 29440 solver.cpp:397]     Test net output #1: loss = 3.17291 (* 1 = 3.17291 loss)
I0428 15:35:43.173964 29440 solver.cpp:218] Iteration 5712 (0.59132 iter/s, 20.2936s/12 iters), loss = 1.45204
I0428 15:35:43.175901 29440 solver.cpp:237]     Train net output #0: loss = 1.45204 (* 1 = 1.45204 loss)
I0428 15:35:43.175913 29440 sgd_solver.cpp:105] Iteration 5712, lr = 0.0032256
I0428 15:35:47.878978 29440 solver.cpp:218] Iteration 5724 (2.55163 iter/s, 4.70287s/12 iters), loss = 1.77796
I0428 15:35:47.879021 29440 solver.cpp:237]     Train net output #0: loss = 1.77796 (* 1 = 1.77796 loss)
I0428 15:35:47.879030 29440 sgd_solver.cpp:105] Iteration 5724, lr = 0.00321794
I0428 15:35:53.302637 29440 solver.cpp:218] Iteration 5736 (2.21355 iter/s, 5.42117s/12 iters), loss = 1.65641
I0428 15:35:53.302748 29440 solver.cpp:237]     Train net output #0: loss = 1.65641 (* 1 = 1.65641 loss)
I0428 15:35:53.302758 29440 sgd_solver.cpp:105] Iteration 5736, lr = 0.0032103
I0428 15:35:58.831933 29440 solver.cpp:218] Iteration 5748 (2.17124 iter/s, 5.5268s/12 iters), loss = 1.68491
I0428 15:35:58.831976 29440 solver.cpp:237]     Train net output #0: loss = 1.68491 (* 1 = 1.68491 loss)
I0428 15:35:58.831984 29440 sgd_solver.cpp:105] Iteration 5748, lr = 0.00320268
I0428 15:36:04.168443 29440 solver.cpp:218] Iteration 5760 (2.24878 iter/s, 5.33623s/12 iters), loss = 1.40092
I0428 15:36:04.168505 29440 solver.cpp:237]     Train net output #0: loss = 1.40092 (* 1 = 1.40092 loss)
I0428 15:36:04.168514 29440 sgd_solver.cpp:105] Iteration 5760, lr = 0.00319508
I0428 15:36:06.217886 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:36:09.511003 29440 solver.cpp:218] Iteration 5772 (2.24623 iter/s, 5.34228s/12 iters), loss = 1.58149
I0428 15:36:09.511044 29440 solver.cpp:237]     Train net output #0: loss = 1.58149 (* 1 = 1.58149 loss)
I0428 15:36:09.511054 29440 sgd_solver.cpp:105] Iteration 5772, lr = 0.00318749
I0428 15:36:15.354513 29440 solver.cpp:218] Iteration 5784 (2.05366 iter/s, 5.84322s/12 iters), loss = 1.84581
I0428 15:36:15.354559 29440 solver.cpp:237]     Train net output #0: loss = 1.84581 (* 1 = 1.84581 loss)
I0428 15:36:15.354569 29440 sgd_solver.cpp:105] Iteration 5784, lr = 0.00317992
I0428 15:36:20.824213 29440 solver.cpp:218] Iteration 5796 (2.19492 iter/s, 5.46716s/12 iters), loss = 1.49866
I0428 15:36:20.824257 29440 solver.cpp:237]     Train net output #0: loss = 1.49866 (* 1 = 1.49866 loss)
I0428 15:36:20.824265 29440 sgd_solver.cpp:105] Iteration 5796, lr = 0.00317237
I0428 15:36:26.336315 29440 solver.cpp:218] Iteration 5808 (2.17801 iter/s, 5.50962s/12 iters), loss = 1.46768
I0428 15:36:26.336438 29440 solver.cpp:237]     Train net output #0: loss = 1.46768 (* 1 = 1.46768 loss)
I0428 15:36:26.336448 29440 sgd_solver.cpp:105] Iteration 5808, lr = 0.00316484
I0428 15:36:28.496158 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_5814.caffemodel
I0428 15:36:34.542910 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_5814.solverstate
I0428 15:36:40.121132 29440 solver.cpp:330] Iteration 5814, Testing net (#0)
I0428 15:36:40.121157 29440 net.cpp:676] Ignoring source layer train-data
I0428 15:36:42.357887 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:36:44.864140 29440 solver.cpp:397]     Test net output #0: accuracy = 0.308211
I0428 15:36:44.864185 29440 solver.cpp:397]     Test net output #1: loss = 3.0662 (* 1 = 3.0662 loss)
I0428 15:36:47.230741 29440 solver.cpp:218] Iteration 5820 (0.574402 iter/s, 20.8913s/12 iters), loss = 1.81311
I0428 15:36:47.230783 29440 solver.cpp:237]     Train net output #0: loss = 1.81311 (* 1 = 1.81311 loss)
I0428 15:36:47.230792 29440 sgd_solver.cpp:105] Iteration 5820, lr = 0.00315733
I0428 15:36:52.592468 29440 solver.cpp:218] Iteration 5832 (2.2382 iter/s, 5.36145s/12 iters), loss = 1.59109
I0428 15:36:52.592537 29440 solver.cpp:237]     Train net output #0: loss = 1.59109 (* 1 = 1.59109 loss)
I0428 15:36:52.592546 29440 sgd_solver.cpp:105] Iteration 5832, lr = 0.00314983
I0428 15:36:57.940711 29440 solver.cpp:218] Iteration 5844 (2.24386 iter/s, 5.34794s/12 iters), loss = 1.49244
I0428 15:36:57.943552 29440 solver.cpp:237]     Train net output #0: loss = 1.49244 (* 1 = 1.49244 loss)
I0428 15:36:57.943562 29440 sgd_solver.cpp:105] Iteration 5844, lr = 0.00314235
I0428 15:37:03.510584 29440 solver.cpp:218] Iteration 5856 (2.15564 iter/s, 5.56679s/12 iters), loss = 1.70648
I0428 15:37:03.510629 29440 solver.cpp:237]     Train net output #0: loss = 1.70648 (* 1 = 1.70648 loss)
I0428 15:37:03.510638 29440 sgd_solver.cpp:105] Iteration 5856, lr = 0.00313489
I0428 15:37:07.794131 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:37:08.928890 29440 solver.cpp:218] Iteration 5868 (2.21572 iter/s, 5.41584s/12 iters), loss = 1.5703
I0428 15:37:08.928936 29440 solver.cpp:237]     Train net output #0: loss = 1.5703 (* 1 = 1.5703 loss)
I0428 15:37:08.928946 29440 sgd_solver.cpp:105] Iteration 5868, lr = 0.00312745
I0428 15:37:14.432451 29440 solver.cpp:218] Iteration 5880 (2.1814 iter/s, 5.50107s/12 iters), loss = 1.63926
I0428 15:37:14.432521 29440 solver.cpp:237]     Train net output #0: loss = 1.63926 (* 1 = 1.63926 loss)
I0428 15:37:14.432533 29440 sgd_solver.cpp:105] Iteration 5880, lr = 0.00312002
I0428 15:37:19.736378 29440 solver.cpp:218] Iteration 5892 (2.26353 iter/s, 5.30146s/12 iters), loss = 1.50832
I0428 15:37:19.736418 29440 solver.cpp:237]     Train net output #0: loss = 1.50832 (* 1 = 1.50832 loss)
I0428 15:37:19.736428 29440 sgd_solver.cpp:105] Iteration 5892, lr = 0.00311262
I0428 15:37:25.128450 29440 solver.cpp:218] Iteration 5904 (2.22652 iter/s, 5.38959s/12 iters), loss = 1.5252
I0428 15:37:25.128520 29440 solver.cpp:237]     Train net output #0: loss = 1.5252 (* 1 = 1.5252 loss)
I0428 15:37:25.128533 29440 sgd_solver.cpp:105] Iteration 5904, lr = 0.00310523
I0428 15:37:29.783860 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_5916.caffemodel
I0428 15:37:31.315874 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_5916.solverstate
I0428 15:37:36.817183 29440 solver.cpp:330] Iteration 5916, Testing net (#0)
I0428 15:37:36.817205 29440 net.cpp:676] Ignoring source layer train-data
I0428 15:37:39.035964 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:37:41.597936 29440 solver.cpp:397]     Test net output #0: accuracy = 0.310049
I0428 15:37:41.597965 29440 solver.cpp:397]     Test net output #1: loss = 3.05101 (* 1 = 3.05101 loss)
I0428 15:37:41.795504 29440 solver.cpp:218] Iteration 5916 (0.72011 iter/s, 16.6641s/12 iters), loss = 1.74422
I0428 15:37:41.795547 29440 solver.cpp:237]     Train net output #0: loss = 1.74422 (* 1 = 1.74422 loss)
I0428 15:37:41.795558 29440 sgd_solver.cpp:105] Iteration 5916, lr = 0.00309785
I0428 15:37:46.723023 29440 solver.cpp:218] Iteration 5928 (2.43543 iter/s, 4.92726s/12 iters), loss = 1.48824
I0428 15:37:46.723060 29440 solver.cpp:237]     Train net output #0: loss = 1.48824 (* 1 = 1.48824 loss)
I0428 15:37:46.723069 29440 sgd_solver.cpp:105] Iteration 5928, lr = 0.0030905
I0428 15:37:52.166285 29440 solver.cpp:218] Iteration 5940 (2.20556 iter/s, 5.44078s/12 iters), loss = 1.48354
I0428 15:37:52.166325 29440 solver.cpp:237]     Train net output #0: loss = 1.48354 (* 1 = 1.48354 loss)
I0428 15:37:52.166334 29440 sgd_solver.cpp:105] Iteration 5940, lr = 0.00308316
I0428 15:37:57.469444 29440 solver.cpp:218] Iteration 5952 (2.26387 iter/s, 5.30067s/12 iters), loss = 1.91551
I0428 15:37:57.469482 29440 solver.cpp:237]     Train net output #0: loss = 1.91551 (* 1 = 1.91551 loss)
I0428 15:37:57.469492 29440 sgd_solver.cpp:105] Iteration 5952, lr = 0.00307584
I0428 15:38:03.218722 29440 solver.cpp:218] Iteration 5964 (2.08812 iter/s, 5.74679s/12 iters), loss = 1.31621
I0428 15:38:03.218863 29440 solver.cpp:237]     Train net output #0: loss = 1.31621 (* 1 = 1.31621 loss)
I0428 15:38:03.218873 29440 sgd_solver.cpp:105] Iteration 5964, lr = 0.00306854
I0428 15:38:04.464615 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:38:08.188942 29440 solver.cpp:218] Iteration 5976 (2.41558 iter/s, 4.96776s/12 iters), loss = 1.5893
I0428 15:38:08.189007 29440 solver.cpp:237]     Train net output #0: loss = 1.5893 (* 1 = 1.5893 loss)
I0428 15:38:08.189020 29440 sgd_solver.cpp:105] Iteration 5976, lr = 0.00306125
I0428 15:38:13.608724 29440 solver.cpp:218] Iteration 5988 (2.21423 iter/s, 5.41949s/12 iters), loss = 1.71097
I0428 15:38:13.608770 29440 solver.cpp:237]     Train net output #0: loss = 1.71097 (* 1 = 1.71097 loss)
I0428 15:38:13.608779 29440 sgd_solver.cpp:105] Iteration 5988, lr = 0.00305398
I0428 15:38:19.387379 29440 solver.cpp:218] Iteration 6000 (2.07671 iter/s, 5.77836s/12 iters), loss = 1.2138
I0428 15:38:19.387423 29440 solver.cpp:237]     Train net output #0: loss = 1.2138 (* 1 = 1.2138 loss)
I0428 15:38:19.387430 29440 sgd_solver.cpp:105] Iteration 6000, lr = 0.00304673
I0428 15:38:24.811153 29440 solver.cpp:218] Iteration 6012 (2.21349 iter/s, 5.4213s/12 iters), loss = 1.4727
I0428 15:38:24.811198 29440 solver.cpp:237]     Train net output #0: loss = 1.4727 (* 1 = 1.4727 loss)
I0428 15:38:24.811206 29440 sgd_solver.cpp:105] Iteration 6012, lr = 0.0030395
I0428 15:38:26.732988 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_6018.caffemodel
I0428 15:38:28.173105 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_6018.solverstate
I0428 15:38:31.635443 29440 solver.cpp:330] Iteration 6018, Testing net (#0)
I0428 15:38:31.635465 29440 net.cpp:676] Ignoring source layer train-data
I0428 15:38:33.858327 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:38:36.557760 29440 solver.cpp:397]     Test net output #0: accuracy = 0.314951
I0428 15:38:36.557796 29440 solver.cpp:397]     Test net output #1: loss = 3.08347 (* 1 = 3.08347 loss)
I0428 15:38:39.077841 29440 solver.cpp:218] Iteration 6024 (0.841288 iter/s, 14.2638s/12 iters), loss = 1.32574
I0428 15:38:39.077888 29440 solver.cpp:237]     Train net output #0: loss = 1.32574 (* 1 = 1.32574 loss)
I0428 15:38:39.077896 29440 sgd_solver.cpp:105] Iteration 6024, lr = 0.00303228
I0428 15:38:44.334851 29440 solver.cpp:218] Iteration 6036 (2.28279 iter/s, 5.25673s/12 iters), loss = 1.41979
I0428 15:38:44.334913 29440 solver.cpp:237]     Train net output #0: loss = 1.41979 (* 1 = 1.41979 loss)
I0428 15:38:44.334925 29440 sgd_solver.cpp:105] Iteration 6036, lr = 0.00302508
I0428 15:38:49.577001 29440 solver.cpp:218] Iteration 6048 (2.28926 iter/s, 5.24186s/12 iters), loss = 1.08793
I0428 15:38:49.577064 29440 solver.cpp:237]     Train net output #0: loss = 1.08793 (* 1 = 1.08793 loss)
I0428 15:38:49.577076 29440 sgd_solver.cpp:105] Iteration 6048, lr = 0.0030179
I0428 15:38:55.217222 29440 solver.cpp:218] Iteration 6060 (2.12769 iter/s, 5.63993s/12 iters), loss = 1.40068
I0428 15:38:55.217262 29440 solver.cpp:237]     Train net output #0: loss = 1.40068 (* 1 = 1.40068 loss)
I0428 15:38:55.217270 29440 sgd_solver.cpp:105] Iteration 6060, lr = 0.00301074
I0428 15:38:58.867173 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:39:00.561309 29440 solver.cpp:218] Iteration 6072 (2.24652 iter/s, 5.3416s/12 iters), loss = 1.7673
I0428 15:39:00.561362 29440 solver.cpp:237]     Train net output #0: loss = 1.7673 (* 1 = 1.7673 loss)
I0428 15:39:00.561375 29440 sgd_solver.cpp:105] Iteration 6072, lr = 0.00300359
I0428 15:39:06.157363 29440 solver.cpp:218] Iteration 6084 (2.14448 iter/s, 5.59577s/12 iters), loss = 1.61622
I0428 15:39:06.159518 29440 solver.cpp:237]     Train net output #0: loss = 1.61622 (* 1 = 1.61622 loss)
I0428 15:39:06.159529 29440 sgd_solver.cpp:105] Iteration 6084, lr = 0.00299646
I0428 15:39:11.511638 29440 solver.cpp:218] Iteration 6096 (2.24312 iter/s, 5.3497s/12 iters), loss = 1.78258
I0428 15:39:11.511685 29440 solver.cpp:237]     Train net output #0: loss = 1.78258 (* 1 = 1.78258 loss)
I0428 15:39:11.511693 29440 sgd_solver.cpp:105] Iteration 6096, lr = 0.00298934
I0428 15:39:16.876086 29440 solver.cpp:218] Iteration 6108 (2.23798 iter/s, 5.36198s/12 iters), loss = 1.6259
I0428 15:39:16.876129 29440 solver.cpp:237]     Train net output #0: loss = 1.6259 (* 1 = 1.6259 loss)
I0428 15:39:16.876139 29440 sgd_solver.cpp:105] Iteration 6108, lr = 0.00298225
I0428 15:39:21.583142 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_6120.caffemodel
I0428 15:39:22.946844 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_6120.solverstate
I0428 15:39:24.011868 29440 solver.cpp:330] Iteration 6120, Testing net (#0)
I0428 15:39:24.011888 29440 net.cpp:676] Ignoring source layer train-data
I0428 15:39:26.294066 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:39:29.029824 29440 solver.cpp:397]     Test net output #0: accuracy = 0.313726
I0428 15:39:29.029861 29440 solver.cpp:397]     Test net output #1: loss = 3.19624 (* 1 = 3.19624 loss)
I0428 15:39:29.176048 29440 solver.cpp:218] Iteration 6120 (0.975831 iter/s, 12.2972s/12 iters), loss = 1.39742
I0428 15:39:29.176096 29440 solver.cpp:237]     Train net output #0: loss = 1.39742 (* 1 = 1.39742 loss)
I0428 15:39:29.176106 29440 sgd_solver.cpp:105] Iteration 6120, lr = 0.00297517
I0428 15:39:33.965672 29440 solver.cpp:218] Iteration 6132 (2.50555 iter/s, 4.78937s/12 iters), loss = 1.6149
I0428 15:39:33.965718 29440 solver.cpp:237]     Train net output #0: loss = 1.6149 (* 1 = 1.6149 loss)
I0428 15:39:33.965730 29440 sgd_solver.cpp:105] Iteration 6132, lr = 0.0029681
I0428 15:39:39.755113 29440 solver.cpp:218] Iteration 6144 (2.07284 iter/s, 5.78915s/12 iters), loss = 1.53065
I0428 15:39:39.755237 29440 solver.cpp:237]     Train net output #0: loss = 1.53065 (* 1 = 1.53065 loss)
I0428 15:39:39.755247 29440 sgd_solver.cpp:105] Iteration 6144, lr = 0.00296105
I0428 15:39:45.017169 29440 solver.cpp:218] Iteration 6156 (2.28063 iter/s, 5.2617s/12 iters), loss = 1.54571
I0428 15:39:45.017213 29440 solver.cpp:237]     Train net output #0: loss = 1.54571 (* 1 = 1.54571 loss)
I0428 15:39:45.017222 29440 sgd_solver.cpp:105] Iteration 6156, lr = 0.00295402
I0428 15:39:50.375433 29440 solver.cpp:218] Iteration 6168 (2.23965 iter/s, 5.35799s/12 iters), loss = 1.6821
I0428 15:39:50.375483 29440 solver.cpp:237]     Train net output #0: loss = 1.6821 (* 1 = 1.6821 loss)
I0428 15:39:50.375491 29440 sgd_solver.cpp:105] Iteration 6168, lr = 0.00294701
I0428 15:39:50.953037 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:39:55.787487 29440 solver.cpp:218] Iteration 6180 (2.21739 iter/s, 5.41177s/12 iters), loss = 1.44057
I0428 15:39:55.787535 29440 solver.cpp:237]     Train net output #0: loss = 1.44057 (* 1 = 1.44057 loss)
I0428 15:39:55.787544 29440 sgd_solver.cpp:105] Iteration 6180, lr = 0.00294001
I0428 15:40:01.374469 29440 solver.cpp:218] Iteration 6192 (2.14796 iter/s, 5.58669s/12 iters), loss = 1.39172
I0428 15:40:01.374513 29440 solver.cpp:237]     Train net output #0: loss = 1.39172 (* 1 = 1.39172 loss)
I0428 15:40:01.374521 29440 sgd_solver.cpp:105] Iteration 6192, lr = 0.00293303
I0428 15:40:06.753212 29440 solver.cpp:218] Iteration 6204 (2.23112 iter/s, 5.37847s/12 iters), loss = 1.25847
I0428 15:40:06.753266 29440 solver.cpp:237]     Train net output #0: loss = 1.25847 (* 1 = 1.25847 loss)
I0428 15:40:06.753276 29440 sgd_solver.cpp:105] Iteration 6204, lr = 0.00292607
I0428 15:40:12.050580 29440 solver.cpp:218] Iteration 6216 (2.2654 iter/s, 5.29709s/12 iters), loss = 1.41719
I0428 15:40:12.050760 29440 solver.cpp:237]     Train net output #0: loss = 1.41719 (* 1 = 1.41719 loss)
I0428 15:40:12.050772 29440 sgd_solver.cpp:105] Iteration 6216, lr = 0.00291912
I0428 15:40:14.290215 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_6222.caffemodel
I0428 15:40:16.764433 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_6222.solverstate
I0428 15:40:17.848387 29440 solver.cpp:330] Iteration 6222, Testing net (#0)
I0428 15:40:17.848415 29440 net.cpp:676] Ignoring source layer train-data
I0428 15:40:19.925583 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:40:22.752755 29440 solver.cpp:397]     Test net output #0: accuracy = 0.316789
I0428 15:40:22.752789 29440 solver.cpp:397]     Test net output #1: loss = 3.13457 (* 1 = 3.13457 loss)
I0428 15:40:25.158318 29440 solver.cpp:218] Iteration 6228 (0.915538 iter/s, 13.107s/12 iters), loss = 1.335
I0428 15:40:25.158356 29440 solver.cpp:237]     Train net output #0: loss = 1.335 (* 1 = 1.335 loss)
I0428 15:40:25.158366 29440 sgd_solver.cpp:105] Iteration 6228, lr = 0.00291219
I0428 15:40:30.307456 29440 solver.cpp:218] Iteration 6240 (2.3316 iter/s, 5.14668s/12 iters), loss = 1.42949
I0428 15:40:30.307494 29440 solver.cpp:237]     Train net output #0: loss = 1.42949 (* 1 = 1.42949 loss)
I0428 15:40:30.307503 29440 sgd_solver.cpp:105] Iteration 6240, lr = 0.00290528
I0428 15:40:33.642220 29440 blocking_queue.cpp:49] Waiting for data
I0428 15:40:35.682612 29440 solver.cpp:218] Iteration 6252 (2.23352 iter/s, 5.37268s/12 iters), loss = 1.26075
I0428 15:40:35.682658 29440 solver.cpp:237]     Train net output #0: loss = 1.26075 (* 1 = 1.26075 loss)
I0428 15:40:35.682667 29440 sgd_solver.cpp:105] Iteration 6252, lr = 0.00289838
I0428 15:40:41.031785 29440 solver.cpp:218] Iteration 6264 (2.24345 iter/s, 5.3489s/12 iters), loss = 1.52854
I0428 15:40:41.031831 29440 solver.cpp:237]     Train net output #0: loss = 1.52854 (* 1 = 1.52854 loss)
I0428 15:40:41.031841 29440 sgd_solver.cpp:105] Iteration 6264, lr = 0.0028915
I0428 15:40:43.971334 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:40:46.596846 29440 solver.cpp:218] Iteration 6276 (2.15642 iter/s, 5.56478s/12 iters), loss = 1.52444
I0428 15:40:46.596889 29440 solver.cpp:237]     Train net output #0: loss = 1.52444 (* 1 = 1.52444 loss)
I0428 15:40:46.596899 29440 sgd_solver.cpp:105] Iteration 6276, lr = 0.00288463
I0428 15:40:51.648722 29440 solver.cpp:218] Iteration 6288 (2.37651 iter/s, 5.04942s/12 iters), loss = 1.42118
I0428 15:40:51.648766 29440 solver.cpp:237]     Train net output #0: loss = 1.42118 (* 1 = 1.42118 loss)
I0428 15:40:51.648774 29440 sgd_solver.cpp:105] Iteration 6288, lr = 0.00287779
I0428 15:40:57.354462 29440 solver.cpp:218] Iteration 6300 (2.10325 iter/s, 5.70546s/12 iters), loss = 1.40059
I0428 15:40:57.354508 29440 solver.cpp:237]     Train net output #0: loss = 1.40059 (* 1 = 1.40059 loss)
I0428 15:40:57.354516 29440 sgd_solver.cpp:105] Iteration 6300, lr = 0.00287095
I0428 15:41:02.726619 29440 solver.cpp:218] Iteration 6312 (2.23458 iter/s, 5.37014s/12 iters), loss = 1.32448
I0428 15:41:02.726667 29440 solver.cpp:237]     Train net output #0: loss = 1.32448 (* 1 = 1.32448 loss)
I0428 15:41:02.726676 29440 sgd_solver.cpp:105] Iteration 6312, lr = 0.00286414
I0428 15:41:07.558147 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_6324.caffemodel
I0428 15:41:09.068130 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_6324.solverstate
I0428 15:41:10.129750 29440 solver.cpp:330] Iteration 6324, Testing net (#0)
I0428 15:41:10.129770 29440 net.cpp:676] Ignoring source layer train-data
I0428 15:41:12.171335 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:41:14.919445 29440 solver.cpp:397]     Test net output #0: accuracy = 0.304534
I0428 15:41:14.919602 29440 solver.cpp:397]     Test net output #1: loss = 3.16389 (* 1 = 3.16389 loss)
I0428 15:41:15.285789 29440 solver.cpp:218] Iteration 6324 (0.955519 iter/s, 12.5586s/12 iters), loss = 1.25131
I0428 15:41:15.287439 29440 solver.cpp:237]     Train net output #0: loss = 1.25131 (* 1 = 1.25131 loss)
I0428 15:41:15.287451 29440 sgd_solver.cpp:105] Iteration 6324, lr = 0.00285734
I0428 15:41:19.995074 29440 solver.cpp:218] Iteration 6336 (2.54916 iter/s, 4.70743s/12 iters), loss = 1.39631
I0428 15:41:19.995127 29440 solver.cpp:237]     Train net output #0: loss = 1.39631 (* 1 = 1.39631 loss)
I0428 15:41:19.995139 29440 sgd_solver.cpp:105] Iteration 6336, lr = 0.00285055
I0428 15:41:24.972406 29440 solver.cpp:218] Iteration 6348 (2.41212 iter/s, 4.97489s/12 iters), loss = 1.45267
I0428 15:41:24.972451 29440 solver.cpp:237]     Train net output #0: loss = 1.45267 (* 1 = 1.45267 loss)
I0428 15:41:24.972460 29440 sgd_solver.cpp:105] Iteration 6348, lr = 0.00284379
I0428 15:41:30.499528 29440 solver.cpp:218] Iteration 6360 (2.17122 iter/s, 5.52685s/12 iters), loss = 1.11228
I0428 15:41:30.499572 29440 solver.cpp:237]     Train net output #0: loss = 1.11228 (* 1 = 1.11228 loss)
I0428 15:41:30.499581 29440 sgd_solver.cpp:105] Iteration 6360, lr = 0.00283703
I0428 15:41:35.617403 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:41:35.817948 29440 solver.cpp:218] Iteration 6372 (2.25642 iter/s, 5.31815s/12 iters), loss = 1.6622
I0428 15:41:35.818007 29440 solver.cpp:237]     Train net output #0: loss = 1.6622 (* 1 = 1.6622 loss)
I0428 15:41:35.818018 29440 sgd_solver.cpp:105] Iteration 6372, lr = 0.0028303
I0428 15:41:41.275586 29440 solver.cpp:218] Iteration 6384 (2.19887 iter/s, 5.45735s/12 iters), loss = 1.34789
I0428 15:41:41.275630 29440 solver.cpp:237]     Train net output #0: loss = 1.34789 (* 1 = 1.34789 loss)
I0428 15:41:41.275640 29440 sgd_solver.cpp:105] Iteration 6384, lr = 0.00282358
I0428 15:41:46.949286 29440 solver.cpp:218] Iteration 6396 (2.11513 iter/s, 5.67342s/12 iters), loss = 1.52336
I0428 15:41:46.949429 29440 solver.cpp:237]     Train net output #0: loss = 1.52336 (* 1 = 1.52336 loss)
I0428 15:41:46.949440 29440 sgd_solver.cpp:105] Iteration 6396, lr = 0.00281687
I0428 15:41:52.493177 29440 solver.cpp:218] Iteration 6408 (2.16551 iter/s, 5.54141s/12 iters), loss = 1.08112
I0428 15:41:52.493221 29440 solver.cpp:237]     Train net output #0: loss = 1.08112 (* 1 = 1.08112 loss)
I0428 15:41:52.493232 29440 sgd_solver.cpp:105] Iteration 6408, lr = 0.00281019
I0428 15:41:58.168992 29440 solver.cpp:218] Iteration 6420 (2.11434 iter/s, 5.67553s/12 iters), loss = 1.25264
I0428 15:41:58.169036 29440 solver.cpp:237]     Train net output #0: loss = 1.25264 (* 1 = 1.25264 loss)
I0428 15:41:58.169045 29440 sgd_solver.cpp:105] Iteration 6420, lr = 0.00280351
I0428 15:42:00.287091 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_6426.caffemodel
I0428 15:42:03.964210 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_6426.solverstate
I0428 15:42:05.643167 29440 solver.cpp:330] Iteration 6426, Testing net (#0)
I0428 15:42:05.643188 29440 net.cpp:676] Ignoring source layer train-data
I0428 15:42:07.682744 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:42:10.643805 29440 solver.cpp:397]     Test net output #0: accuracy = 0.329657
I0428 15:42:10.643836 29440 solver.cpp:397]     Test net output #1: loss = 3.11125 (* 1 = 3.11125 loss)
I0428 15:42:12.679548 29440 solver.cpp:218] Iteration 6432 (0.82702 iter/s, 14.5099s/12 iters), loss = 1.46762
I0428 15:42:12.679595 29440 solver.cpp:237]     Train net output #0: loss = 1.46762 (* 1 = 1.46762 loss)
I0428 15:42:12.679603 29440 sgd_solver.cpp:105] Iteration 6432, lr = 0.00279686
I0428 15:42:18.030697 29440 solver.cpp:218] Iteration 6444 (2.24262 iter/s, 5.35088s/12 iters), loss = 1.41819
I0428 15:42:18.030818 29440 solver.cpp:237]     Train net output #0: loss = 1.41819 (* 1 = 1.41819 loss)
I0428 15:42:18.030829 29440 sgd_solver.cpp:105] Iteration 6444, lr = 0.00279022
I0428 15:42:23.437489 29440 solver.cpp:218] Iteration 6456 (2.21957 iter/s, 5.40645s/12 iters), loss = 1.39332
I0428 15:42:23.437530 29440 solver.cpp:237]     Train net output #0: loss = 1.39332 (* 1 = 1.39332 loss)
I0428 15:42:23.437537 29440 sgd_solver.cpp:105] Iteration 6456, lr = 0.00278359
I0428 15:42:28.945030 29440 solver.cpp:218] Iteration 6468 (2.17894 iter/s, 5.50727s/12 iters), loss = 1.1916
I0428 15:42:28.945075 29440 solver.cpp:237]     Train net output #0: loss = 1.1916 (* 1 = 1.1916 loss)
I0428 15:42:28.945083 29440 sgd_solver.cpp:105] Iteration 6468, lr = 0.00277698
I0428 15:42:30.800658 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:42:34.242244 29440 solver.cpp:218] Iteration 6480 (2.2664 iter/s, 5.29474s/12 iters), loss = 1.27916
I0428 15:42:34.242293 29440 solver.cpp:237]     Train net output #0: loss = 1.27916 (* 1 = 1.27916 loss)
I0428 15:42:34.242305 29440 sgd_solver.cpp:105] Iteration 6480, lr = 0.00277039
I0428 15:42:39.509449 29440 solver.cpp:218] Iteration 6492 (2.27932 iter/s, 5.26474s/12 iters), loss = 1.42441
I0428 15:42:39.509496 29440 solver.cpp:237]     Train net output #0: loss = 1.42441 (* 1 = 1.42441 loss)
I0428 15:42:39.509505 29440 sgd_solver.cpp:105] Iteration 6492, lr = 0.00276381
I0428 15:42:44.826954 29440 solver.cpp:218] Iteration 6504 (2.25774 iter/s, 5.31504s/12 iters), loss = 1.57901
I0428 15:42:44.826994 29440 solver.cpp:237]     Train net output #0: loss = 1.57901 (* 1 = 1.57901 loss)
I0428 15:42:44.827004 29440 sgd_solver.cpp:105] Iteration 6504, lr = 0.00275725
I0428 15:42:49.680877 29440 solver.cpp:218] Iteration 6516 (2.47347 iter/s, 4.85148s/12 iters), loss = 1.15429
I0428 15:42:49.681003 29440 solver.cpp:237]     Train net output #0: loss = 1.15429 (* 1 = 1.15429 loss)
I0428 15:42:49.681015 29440 sgd_solver.cpp:105] Iteration 6516, lr = 0.00275071
I0428 15:42:54.690907 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_6528.caffemodel
I0428 15:42:57.717161 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_6528.solverstate
I0428 15:43:00.005141 29440 solver.cpp:330] Iteration 6528, Testing net (#0)
I0428 15:43:00.005159 29440 net.cpp:676] Ignoring source layer train-data
I0428 15:43:02.068886 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:43:04.970314 29440 solver.cpp:397]     Test net output #0: accuracy = 0.322917
I0428 15:43:04.970346 29440 solver.cpp:397]     Test net output #1: loss = 3.23059 (* 1 = 3.23059 loss)
I0428 15:43:05.110337 29440 solver.cpp:218] Iteration 6528 (0.77777 iter/s, 15.4287s/12 iters), loss = 1.39559
I0428 15:43:05.110383 29440 solver.cpp:237]     Train net output #0: loss = 1.39559 (* 1 = 1.39559 loss)
I0428 15:43:05.110394 29440 sgd_solver.cpp:105] Iteration 6528, lr = 0.00274418
I0428 15:43:10.205472 29440 solver.cpp:218] Iteration 6540 (2.35531 iter/s, 5.09488s/12 iters), loss = 1.53335
I0428 15:43:10.205513 29440 solver.cpp:237]     Train net output #0: loss = 1.53335 (* 1 = 1.53335 loss)
I0428 15:43:10.205523 29440 sgd_solver.cpp:105] Iteration 6540, lr = 0.00273766
I0428 15:43:15.421972 29440 solver.cpp:218] Iteration 6552 (2.30149 iter/s, 5.21402s/12 iters), loss = 1.51402
I0428 15:43:15.422024 29440 solver.cpp:237]     Train net output #0: loss = 1.51402 (* 1 = 1.51402 loss)
I0428 15:43:15.422036 29440 sgd_solver.cpp:105] Iteration 6552, lr = 0.00273116
I0428 15:43:20.928817 29440 solver.cpp:218] Iteration 6564 (2.17922 iter/s, 5.50657s/12 iters), loss = 1.19958
I0428 15:43:20.928905 29440 solver.cpp:237]     Train net output #0: loss = 1.19958 (* 1 = 1.19958 loss)
I0428 15:43:20.928915 29440 sgd_solver.cpp:105] Iteration 6564, lr = 0.00272468
I0428 15:43:25.303654 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:43:26.381533 29440 solver.cpp:218] Iteration 6576 (2.20174 iter/s, 5.45023s/12 iters), loss = 1.45166
I0428 15:43:26.381573 29440 solver.cpp:237]     Train net output #0: loss = 1.45166 (* 1 = 1.45166 loss)
I0428 15:43:26.381582 29440 sgd_solver.cpp:105] Iteration 6576, lr = 0.00271821
I0428 15:43:31.721048 29440 solver.cpp:218] Iteration 6588 (2.24845 iter/s, 5.33702s/12 iters), loss = 1.38614
I0428 15:43:31.721101 29440 solver.cpp:237]     Train net output #0: loss = 1.38614 (* 1 = 1.38614 loss)
I0428 15:43:31.721109 29440 sgd_solver.cpp:105] Iteration 6588, lr = 0.00271175
I0428 15:43:36.906580 29440 solver.cpp:218] Iteration 6600 (2.31425 iter/s, 5.18527s/12 iters), loss = 1.291
I0428 15:43:36.906620 29440 solver.cpp:237]     Train net output #0: loss = 1.291 (* 1 = 1.291 loss)
I0428 15:43:36.906628 29440 sgd_solver.cpp:105] Iteration 6600, lr = 0.00270532
I0428 15:43:42.036089 29440 solver.cpp:218] Iteration 6612 (2.34053 iter/s, 5.12705s/12 iters), loss = 1.25524
I0428 15:43:42.036147 29440 solver.cpp:237]     Train net output #0: loss = 1.25524 (* 1 = 1.25524 loss)
I0428 15:43:42.036159 29440 sgd_solver.cpp:105] Iteration 6612, lr = 0.00269889
I0428 15:43:47.477908 29440 solver.cpp:218] Iteration 6624 (2.20526 iter/s, 5.44154s/12 iters), loss = 1.24135
I0428 15:43:47.477954 29440 solver.cpp:237]     Train net output #0: loss = 1.24135 (* 1 = 1.24135 loss)
I0428 15:43:47.477962 29440 sgd_solver.cpp:105] Iteration 6624, lr = 0.00269248
I0428 15:43:49.366904 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_6630.caffemodel
I0428 15:43:52.306756 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_6630.solverstate
I0428 15:43:53.918603 29440 solver.cpp:330] Iteration 6630, Testing net (#0)
I0428 15:43:53.918627 29440 net.cpp:676] Ignoring source layer train-data
I0428 15:43:55.901230 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:43:58.766995 29440 solver.cpp:397]     Test net output #0: accuracy = 0.300858
I0428 15:43:58.767024 29440 solver.cpp:397]     Test net output #1: loss = 3.2658 (* 1 = 3.2658 loss)
I0428 15:44:01.329692 29440 solver.cpp:218] Iteration 6636 (0.866351 iter/s, 13.8512s/12 iters), loss = 1.42367
I0428 15:44:01.329737 29440 solver.cpp:237]     Train net output #0: loss = 1.42367 (* 1 = 1.42367 loss)
I0428 15:44:01.329747 29440 sgd_solver.cpp:105] Iteration 6636, lr = 0.00268609
I0428 15:44:06.670322 29440 solver.cpp:218] Iteration 6648 (2.24797 iter/s, 5.33815s/12 iters), loss = 1.20882
I0428 15:44:06.670368 29440 solver.cpp:237]     Train net output #0: loss = 1.20882 (* 1 = 1.20882 loss)
I0428 15:44:06.670377 29440 sgd_solver.cpp:105] Iteration 6648, lr = 0.00267971
I0428 15:44:12.119314 29440 solver.cpp:218] Iteration 6660 (2.20235 iter/s, 5.44872s/12 iters), loss = 1.53089
I0428 15:44:12.119352 29440 solver.cpp:237]     Train net output #0: loss = 1.53089 (* 1 = 1.53089 loss)
I0428 15:44:12.119361 29440 sgd_solver.cpp:105] Iteration 6660, lr = 0.00267335
I0428 15:44:17.603938 29440 solver.cpp:218] Iteration 6672 (2.18804 iter/s, 5.48436s/12 iters), loss = 1.30096
I0428 15:44:17.603986 29440 solver.cpp:237]     Train net output #0: loss = 1.30096 (* 1 = 1.30096 loss)
I0428 15:44:17.603996 29440 sgd_solver.cpp:105] Iteration 6672, lr = 0.00266701
I0428 15:44:19.061147 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:44:22.847069 29440 solver.cpp:218] Iteration 6684 (2.28882 iter/s, 5.24287s/12 iters), loss = 1.35864
I0428 15:44:22.847162 29440 solver.cpp:237]     Train net output #0: loss = 1.35864 (* 1 = 1.35864 loss)
I0428 15:44:22.847172 29440 sgd_solver.cpp:105] Iteration 6684, lr = 0.00266067
I0428 15:44:28.214100 29440 solver.cpp:218] Iteration 6696 (2.236 iter/s, 5.36672s/12 iters), loss = 1.43034
I0428 15:44:28.214139 29440 solver.cpp:237]     Train net output #0: loss = 1.43034 (* 1 = 1.43034 loss)
I0428 15:44:28.214148 29440 sgd_solver.cpp:105] Iteration 6696, lr = 0.00265436
I0428 15:44:33.524544 29440 solver.cpp:218] Iteration 6708 (2.25981 iter/s, 5.31018s/12 iters), loss = 1.41983
I0428 15:44:33.524587 29440 solver.cpp:237]     Train net output #0: loss = 1.41983 (* 1 = 1.41983 loss)
I0428 15:44:33.524597 29440 sgd_solver.cpp:105] Iteration 6708, lr = 0.00264805
I0428 15:44:39.431679 29440 solver.cpp:218] Iteration 6720 (2.03154 iter/s, 5.90685s/12 iters), loss = 1.58319
I0428 15:44:39.431722 29440 solver.cpp:237]     Train net output #0: loss = 1.58319 (* 1 = 1.58319 loss)
I0428 15:44:39.431731 29440 sgd_solver.cpp:105] Iteration 6720, lr = 0.00264177
I0428 15:44:44.743026 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_6732.caffemodel
I0428 15:44:49.763911 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_6732.solverstate
I0428 15:44:51.213282 29440 solver.cpp:330] Iteration 6732, Testing net (#0)
I0428 15:44:51.213304 29440 net.cpp:676] Ignoring source layer train-data
I0428 15:44:53.248124 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:44:56.309645 29440 solver.cpp:397]     Test net output #0: accuracy = 0.308824
I0428 15:44:56.309682 29440 solver.cpp:397]     Test net output #1: loss = 3.06736 (* 1 = 3.06736 loss)
I0428 15:44:56.680457 29440 solver.cpp:218] Iteration 6732 (0.695731 iter/s, 17.2481s/12 iters), loss = 1.28168
I0428 15:44:56.682094 29440 solver.cpp:237]     Train net output #0: loss = 1.28168 (* 1 = 1.28168 loss)
I0428 15:44:56.682111 29440 sgd_solver.cpp:105] Iteration 6732, lr = 0.0026355
I0428 15:45:01.818948 29440 solver.cpp:218] Iteration 6744 (2.33616 iter/s, 5.13664s/12 iters), loss = 1.26427
I0428 15:45:01.818994 29440 solver.cpp:237]     Train net output #0: loss = 1.26427 (* 1 = 1.26427 loss)
I0428 15:45:01.819003 29440 sgd_solver.cpp:105] Iteration 6744, lr = 0.00262924
I0428 15:45:07.375978 29440 solver.cpp:218] Iteration 6756 (2.15953 iter/s, 5.55676s/12 iters), loss = 1.25062
I0428 15:45:07.376024 29440 solver.cpp:237]     Train net output #0: loss = 1.25062 (* 1 = 1.25062 loss)
I0428 15:45:07.376034 29440 sgd_solver.cpp:105] Iteration 6756, lr = 0.002623
I0428 15:45:12.381410 29440 solver.cpp:218] Iteration 6768 (2.39858 iter/s, 5.00297s/12 iters), loss = 1.15896
I0428 15:45:12.381474 29440 solver.cpp:237]     Train net output #0: loss = 1.15896 (* 1 = 1.15896 loss)
I0428 15:45:12.381490 29440 sgd_solver.cpp:105] Iteration 6768, lr = 0.00261677
I0428 15:45:16.210577 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:45:18.040385 29440 solver.cpp:218] Iteration 6780 (2.12064 iter/s, 5.65868s/12 iters), loss = 1.50514
I0428 15:45:18.040426 29440 solver.cpp:237]     Train net output #0: loss = 1.50514 (* 1 = 1.50514 loss)
I0428 15:45:18.040436 29440 sgd_solver.cpp:105] Iteration 6780, lr = 0.00261056
I0428 15:45:22.859858 29440 solver.cpp:218] Iteration 6792 (2.49002 iter/s, 4.81923s/12 iters), loss = 1.32393
I0428 15:45:22.859910 29440 solver.cpp:237]     Train net output #0: loss = 1.32393 (* 1 = 1.32393 loss)
I0428 15:45:22.859921 29440 sgd_solver.cpp:105] Iteration 6792, lr = 0.00260436
I0428 15:45:28.170128 29440 solver.cpp:218] Iteration 6804 (2.25989 iter/s, 5.31s/12 iters), loss = 1.40687
I0428 15:45:28.170243 29440 solver.cpp:237]     Train net output #0: loss = 1.40687 (* 1 = 1.40687 loss)
I0428 15:45:28.170254 29440 sgd_solver.cpp:105] Iteration 6804, lr = 0.00259817
I0428 15:45:33.607959 29440 solver.cpp:218] Iteration 6816 (2.2069 iter/s, 5.43749s/12 iters), loss = 1.35757
I0428 15:45:33.608004 29440 solver.cpp:237]     Train net output #0: loss = 1.35757 (* 1 = 1.35757 loss)
I0428 15:45:33.608014 29440 sgd_solver.cpp:105] Iteration 6816, lr = 0.00259201
I0428 15:45:39.312068 29440 solver.cpp:218] Iteration 6828 (2.10385 iter/s, 5.70383s/12 iters), loss = 1.24502
I0428 15:45:39.312109 29440 solver.cpp:237]     Train net output #0: loss = 1.24502 (* 1 = 1.24502 loss)
I0428 15:45:39.312119 29440 sgd_solver.cpp:105] Iteration 6828, lr = 0.00258585
I0428 15:45:41.223589 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_6834.caffemodel
I0428 15:45:46.042615 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_6834.solverstate
I0428 15:45:49.039166 29440 solver.cpp:330] Iteration 6834, Testing net (#0)
I0428 15:45:49.039194 29440 net.cpp:676] Ignoring source layer train-data
I0428 15:45:50.899581 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:45:54.045543 29440 solver.cpp:397]     Test net output #0: accuracy = 0.332108
I0428 15:45:54.045576 29440 solver.cpp:397]     Test net output #1: loss = 3.12516 (* 1 = 3.12516 loss)
I0428 15:45:56.414141 29440 solver.cpp:218] Iteration 6840 (0.701699 iter/s, 17.1014s/12 iters), loss = 1.25312
I0428 15:45:56.414187 29440 solver.cpp:237]     Train net output #0: loss = 1.25312 (* 1 = 1.25312 loss)
I0428 15:45:56.414197 29440 sgd_solver.cpp:105] Iteration 6840, lr = 0.00257971
I0428 15:46:01.590377 29440 solver.cpp:218] Iteration 6852 (2.31939 iter/s, 5.17378s/12 iters), loss = 1.21101
I0428 15:46:01.590651 29440 solver.cpp:237]     Train net output #0: loss = 1.21101 (* 1 = 1.21101 loss)
I0428 15:46:01.590659 29440 sgd_solver.cpp:105] Iteration 6852, lr = 0.00257359
I0428 15:46:06.707091 29440 solver.cpp:218] Iteration 6864 (2.34638 iter/s, 5.11426s/12 iters), loss = 1.27374
I0428 15:46:06.707135 29440 solver.cpp:237]     Train net output #0: loss = 1.27374 (* 1 = 1.27374 loss)
I0428 15:46:06.707145 29440 sgd_solver.cpp:105] Iteration 6864, lr = 0.00256748
I0428 15:46:12.112540 29440 solver.cpp:218] Iteration 6876 (2.2201 iter/s, 5.40516s/12 iters), loss = 1.3098
I0428 15:46:12.112598 29440 solver.cpp:237]     Train net output #0: loss = 1.3098 (* 1 = 1.3098 loss)
I0428 15:46:12.112612 29440 sgd_solver.cpp:105] Iteration 6876, lr = 0.00256138
I0428 15:46:12.752369 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:46:17.644915 29440 solver.cpp:218] Iteration 6888 (2.16916 iter/s, 5.5321s/12 iters), loss = 1.1691
I0428 15:46:17.644951 29440 solver.cpp:237]     Train net output #0: loss = 1.1691 (* 1 = 1.1691 loss)
I0428 15:46:17.644961 29440 sgd_solver.cpp:105] Iteration 6888, lr = 0.0025553
I0428 15:46:23.293840 29440 solver.cpp:218] Iteration 6900 (2.1244 iter/s, 5.64866s/12 iters), loss = 1.04017
I0428 15:46:23.293884 29440 solver.cpp:237]     Train net output #0: loss = 1.04017 (* 1 = 1.04017 loss)
I0428 15:46:23.293892 29440 sgd_solver.cpp:105] Iteration 6900, lr = 0.00254923
I0428 15:46:28.987597 29440 solver.cpp:218] Iteration 6912 (2.10768 iter/s, 5.69348s/12 iters), loss = 1.08999
I0428 15:46:28.987643 29440 solver.cpp:237]     Train net output #0: loss = 1.08999 (* 1 = 1.08999 loss)
I0428 15:46:28.987651 29440 sgd_solver.cpp:105] Iteration 6912, lr = 0.00254318
I0428 15:46:34.192610 29440 solver.cpp:218] Iteration 6924 (2.30656 iter/s, 5.20255s/12 iters), loss = 1.15557
I0428 15:46:34.192706 29440 solver.cpp:237]     Train net output #0: loss = 1.15557 (* 1 = 1.15557 loss)
I0428 15:46:34.192716 29440 sgd_solver.cpp:105] Iteration 6924, lr = 0.00253714
I0428 15:46:38.937662 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_6936.caffemodel
I0428 15:46:45.419137 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_6936.solverstate
I0428 15:46:47.555393 29440 solver.cpp:330] Iteration 6936, Testing net (#0)
I0428 15:46:47.555413 29440 net.cpp:676] Ignoring source layer train-data
I0428 15:46:49.412472 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:46:52.572304 29440 solver.cpp:397]     Test net output #0: accuracy = 0.32598
I0428 15:46:52.572335 29440 solver.cpp:397]     Test net output #1: loss = 3.23032 (* 1 = 3.23032 loss)
I0428 15:46:52.722972 29440 solver.cpp:218] Iteration 6936 (0.647614 iter/s, 18.5296s/12 iters), loss = 1.30821
I0428 15:46:52.723039 29440 solver.cpp:237]     Train net output #0: loss = 1.30821 (* 1 = 1.30821 loss)
I0428 15:46:52.723048 29440 sgd_solver.cpp:105] Iteration 6936, lr = 0.00253112
I0428 15:46:53.669566 29440 blocking_queue.cpp:49] Waiting for data
I0428 15:46:57.323120 29440 solver.cpp:218] Iteration 6948 (2.60876 iter/s, 4.59989s/12 iters), loss = 1.15926
I0428 15:46:57.323164 29440 solver.cpp:237]     Train net output #0: loss = 1.15926 (* 1 = 1.15926 loss)
I0428 15:46:57.323173 29440 sgd_solver.cpp:105] Iteration 6948, lr = 0.00252511
I0428 15:47:02.438323 29440 solver.cpp:218] Iteration 6960 (2.34607 iter/s, 5.11494s/12 iters), loss = 1.31515
I0428 15:47:02.438378 29440 solver.cpp:237]     Train net output #0: loss = 1.31515 (* 1 = 1.31515 loss)
I0428 15:47:02.438390 29440 sgd_solver.cpp:105] Iteration 6960, lr = 0.00251911
I0428 15:47:07.777070 29440 solver.cpp:218] Iteration 6972 (2.24783 iter/s, 5.33847s/12 iters), loss = 1.04272
I0428 15:47:07.781656 29440 solver.cpp:237]     Train net output #0: loss = 1.04272 (* 1 = 1.04272 loss)
I0428 15:47:07.781666 29440 sgd_solver.cpp:105] Iteration 6972, lr = 0.00251313
I0428 15:47:10.621400 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:47:13.082602 29440 solver.cpp:218] Iteration 6984 (2.26384 iter/s, 5.30072s/12 iters), loss = 1.04756
I0428 15:47:13.082651 29440 solver.cpp:237]     Train net output #0: loss = 1.04756 (* 1 = 1.04756 loss)
I0428 15:47:13.082660 29440 sgd_solver.cpp:105] Iteration 6984, lr = 0.00250717
I0428 15:47:19.028755 29440 solver.cpp:218] Iteration 6996 (2.01821 iter/s, 5.94586s/12 iters), loss = 1.18317
I0428 15:47:19.028797 29440 solver.cpp:237]     Train net output #0: loss = 1.18317 (* 1 = 1.18317 loss)
I0428 15:47:19.028806 29440 sgd_solver.cpp:105] Iteration 6996, lr = 0.00250121
I0428 15:47:24.672433 29440 solver.cpp:218] Iteration 7008 (2.12719 iter/s, 5.64125s/12 iters), loss = 1.06958
I0428 15:47:24.672480 29440 solver.cpp:237]     Train net output #0: loss = 1.06958 (* 1 = 1.06958 loss)
I0428 15:47:24.672514 29440 sgd_solver.cpp:105] Iteration 7008, lr = 0.00249528
I0428 15:47:29.696014 29440 solver.cpp:218] Iteration 7020 (2.3899 iter/s, 5.02113s/12 iters), loss = 0.759928
I0428 15:47:29.696063 29440 solver.cpp:237]     Train net output #0: loss = 0.759928 (* 1 = 0.759928 loss)
I0428 15:47:29.696072 29440 sgd_solver.cpp:105] Iteration 7020, lr = 0.00248935
I0428 15:47:35.310014 29440 solver.cpp:218] Iteration 7032 (2.13762 iter/s, 5.61372s/12 iters), loss = 1.20293
I0428 15:47:35.310060 29440 solver.cpp:237]     Train net output #0: loss = 1.20293 (* 1 = 1.20293 loss)
I0428 15:47:35.310068 29440 sgd_solver.cpp:105] Iteration 7032, lr = 0.00248344
I0428 15:47:37.239621 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_7038.caffemodel
I0428 15:47:45.141000 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_7038.solverstate
I0428 15:47:48.325376 29440 solver.cpp:330] Iteration 7038, Testing net (#0)
I0428 15:47:48.325397 29440 net.cpp:676] Ignoring source layer train-data
I0428 15:47:50.092602 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:47:53.179286 29440 solver.cpp:397]     Test net output #0: accuracy = 0.338848
I0428 15:47:53.179316 29440 solver.cpp:397]     Test net output #1: loss = 3.10849 (* 1 = 3.10849 loss)
I0428 15:47:55.683580 29440 solver.cpp:218] Iteration 7044 (0.589086 iter/s, 20.3705s/12 iters), loss = 1.07697
I0428 15:47:55.683626 29440 solver.cpp:237]     Train net output #0: loss = 1.07697 (* 1 = 1.07697 loss)
I0428 15:47:55.683636 29440 sgd_solver.cpp:105] Iteration 7044, lr = 0.00247755
I0428 15:48:01.025312 29440 solver.cpp:218] Iteration 7056 (2.24658 iter/s, 5.34146s/12 iters), loss = 1.27424
I0428 15:48:01.025358 29440 solver.cpp:237]     Train net output #0: loss = 1.27424 (* 1 = 1.27424 loss)
I0428 15:48:01.025367 29440 sgd_solver.cpp:105] Iteration 7056, lr = 0.00247166
I0428 15:48:06.426842 29440 solver.cpp:218] Iteration 7068 (2.2217 iter/s, 5.40126s/12 iters), loss = 0.835219
I0428 15:48:06.426883 29440 solver.cpp:237]     Train net output #0: loss = 0.835219 (* 1 = 0.835219 loss)
I0428 15:48:06.426892 29440 sgd_solver.cpp:105] Iteration 7068, lr = 0.0024658
I0428 15:48:12.130152 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:48:12.297869 29440 solver.cpp:218] Iteration 7080 (2.04404 iter/s, 5.87074s/12 iters), loss = 1.05782
I0428 15:48:12.297930 29440 solver.cpp:237]     Train net output #0: loss = 1.05782 (* 1 = 1.05782 loss)
I0428 15:48:12.297943 29440 sgd_solver.cpp:105] Iteration 7080, lr = 0.00245994
I0428 15:48:17.720471 29440 solver.cpp:218] Iteration 7092 (2.21308 iter/s, 5.42232s/12 iters), loss = 1.06973
I0428 15:48:17.720669 29440 solver.cpp:237]     Train net output #0: loss = 1.06973 (* 1 = 1.06973 loss)
I0428 15:48:17.720681 29440 sgd_solver.cpp:105] Iteration 7092, lr = 0.0024541
I0428 15:48:22.975365 29440 solver.cpp:218] Iteration 7104 (2.28376 iter/s, 5.25448s/12 iters), loss = 1.22834
I0428 15:48:22.975404 29440 solver.cpp:237]     Train net output #0: loss = 1.22834 (* 1 = 1.22834 loss)
I0428 15:48:22.975414 29440 sgd_solver.cpp:105] Iteration 7104, lr = 0.00244827
I0428 15:48:28.260354 29440 solver.cpp:218] Iteration 7116 (2.2707 iter/s, 5.28472s/12 iters), loss = 1.15555
I0428 15:48:28.260411 29440 solver.cpp:237]     Train net output #0: loss = 1.15555 (* 1 = 1.15555 loss)
I0428 15:48:28.260422 29440 sgd_solver.cpp:105] Iteration 7116, lr = 0.00244246
I0428 15:48:33.586789 29440 solver.cpp:218] Iteration 7128 (2.25303 iter/s, 5.32616s/12 iters), loss = 1.21368
I0428 15:48:33.586831 29440 solver.cpp:237]     Train net output #0: loss = 1.21368 (* 1 = 1.21368 loss)
I0428 15:48:33.586839 29440 sgd_solver.cpp:105] Iteration 7128, lr = 0.00243666
I0428 15:48:38.196533 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_7140.caffemodel
I0428 15:48:41.869618 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_7140.solverstate
I0428 15:48:44.434459 29440 solver.cpp:330] Iteration 7140, Testing net (#0)
I0428 15:48:44.434480 29440 net.cpp:676] Ignoring source layer train-data
I0428 15:48:46.217818 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:48:49.293550 29440 solver.cpp:397]     Test net output #0: accuracy = 0.314951
I0428 15:48:49.293678 29440 solver.cpp:397]     Test net output #1: loss = 3.35104 (* 1 = 3.35104 loss)
I0428 15:48:49.470687 29440 solver.cpp:218] Iteration 7140 (0.755618 iter/s, 15.881s/12 iters), loss = 1.36367
I0428 15:48:49.470752 29440 solver.cpp:237]     Train net output #0: loss = 1.36367 (* 1 = 1.36367 loss)
I0428 15:48:49.470763 29440 sgd_solver.cpp:105] Iteration 7140, lr = 0.00243088
I0428 15:48:54.509502 29440 solver.cpp:218] Iteration 7152 (2.38164 iter/s, 5.03854s/12 iters), loss = 1.06492
I0428 15:48:54.509544 29440 solver.cpp:237]     Train net output #0: loss = 1.06492 (* 1 = 1.06492 loss)
I0428 15:48:54.509553 29440 sgd_solver.cpp:105] Iteration 7152, lr = 0.00242511
I0428 15:48:59.887643 29440 solver.cpp:218] Iteration 7164 (2.23228 iter/s, 5.37568s/12 iters), loss = 1.13355
I0428 15:48:59.887692 29440 solver.cpp:237]     Train net output #0: loss = 1.13355 (* 1 = 1.13355 loss)
I0428 15:48:59.887701 29440 sgd_solver.cpp:105] Iteration 7164, lr = 0.00241935
I0428 15:49:05.464653 29440 solver.cpp:218] Iteration 7176 (2.1518 iter/s, 5.57674s/12 iters), loss = 0.998491
I0428 15:49:05.464691 29440 solver.cpp:237]     Train net output #0: loss = 0.998491 (* 1 = 0.998491 loss)
I0428 15:49:05.464700 29440 sgd_solver.cpp:105] Iteration 7176, lr = 0.0024136
I0428 15:49:07.645179 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:49:10.780180 29440 solver.cpp:218] Iteration 7188 (2.25859 iter/s, 5.31305s/12 iters), loss = 1.12959
I0428 15:49:10.780228 29440 solver.cpp:237]     Train net output #0: loss = 1.12959 (* 1 = 1.12959 loss)
I0428 15:49:10.780238 29440 sgd_solver.cpp:105] Iteration 7188, lr = 0.00240787
I0428 15:49:16.811646 29440 solver.cpp:218] Iteration 7200 (1.98966 iter/s, 6.03117s/12 iters), loss = 1.25547
I0428 15:49:16.811692 29440 solver.cpp:237]     Train net output #0: loss = 1.25547 (* 1 = 1.25547 loss)
I0428 15:49:16.811700 29440 sgd_solver.cpp:105] Iteration 7200, lr = 0.00240216
I0428 15:49:22.190623 29440 solver.cpp:218] Iteration 7212 (2.23102 iter/s, 5.3787s/12 iters), loss = 0.952896
I0428 15:49:22.190780 29440 solver.cpp:237]     Train net output #0: loss = 0.952896 (* 1 = 0.952896 loss)
I0428 15:49:22.190793 29440 sgd_solver.cpp:105] Iteration 7212, lr = 0.00239645
I0428 15:49:27.855595 29440 solver.cpp:218] Iteration 7224 (2.11843 iter/s, 5.66458s/12 iters), loss = 1.0866
I0428 15:49:27.855654 29440 solver.cpp:237]     Train net output #0: loss = 1.0866 (* 1 = 1.0866 loss)
I0428 15:49:27.855664 29440 sgd_solver.cpp:105] Iteration 7224, lr = 0.00239076
I0428 15:49:33.257639 29440 solver.cpp:218] Iteration 7236 (2.22149 iter/s, 5.40177s/12 iters), loss = 1.15562
I0428 15:49:33.257684 29440 solver.cpp:237]     Train net output #0: loss = 1.15562 (* 1 = 1.15562 loss)
I0428 15:49:33.257695 29440 sgd_solver.cpp:105] Iteration 7236, lr = 0.00238509
I0428 15:49:35.393981 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_7242.caffemodel
I0428 15:49:36.955458 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_7242.solverstate
I0428 15:49:41.840102 29440 solver.cpp:330] Iteration 7242, Testing net (#0)
I0428 15:49:41.840123 29440 net.cpp:676] Ignoring source layer train-data
I0428 15:49:43.506742 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:49:46.696959 29440 solver.cpp:397]     Test net output #0: accuracy = 0.315564
I0428 15:49:46.696992 29440 solver.cpp:397]     Test net output #1: loss = 3.27307 (* 1 = 3.27307 loss)
I0428 15:49:49.436209 29440 solver.cpp:218] Iteration 7248 (0.741753 iter/s, 16.1779s/12 iters), loss = 0.968626
I0428 15:49:49.436252 29440 solver.cpp:237]     Train net output #0: loss = 0.968626 (* 1 = 0.968626 loss)
I0428 15:49:49.436261 29440 sgd_solver.cpp:105] Iteration 7248, lr = 0.00237942
I0428 15:49:54.828833 29440 solver.cpp:218] Iteration 7260 (2.22628 iter/s, 5.39015s/12 iters), loss = 1.26449
I0428 15:49:54.828930 29440 solver.cpp:237]     Train net output #0: loss = 1.26449 (* 1 = 1.26449 loss)
I0428 15:49:54.828941 29440 sgd_solver.cpp:105] Iteration 7260, lr = 0.00237378
I0428 15:50:00.567804 29440 solver.cpp:218] Iteration 7272 (2.09109 iter/s, 5.73864s/12 iters), loss = 1.14007
I0428 15:50:00.567847 29440 solver.cpp:237]     Train net output #0: loss = 1.14007 (* 1 = 1.14007 loss)
I0428 15:50:00.567855 29440 sgd_solver.cpp:105] Iteration 7272, lr = 0.00236814
I0428 15:50:05.071561 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:50:05.894552 29440 solver.cpp:218] Iteration 7284 (2.25289 iter/s, 5.32648s/12 iters), loss = 1.3501
I0428 15:50:05.894596 29440 solver.cpp:237]     Train net output #0: loss = 1.3501 (* 1 = 1.3501 loss)
I0428 15:50:05.894605 29440 sgd_solver.cpp:105] Iteration 7284, lr = 0.00236252
I0428 15:50:11.528321 29440 solver.cpp:218] Iteration 7296 (2.13012 iter/s, 5.63349s/12 iters), loss = 1.06039
I0428 15:50:11.528363 29440 solver.cpp:237]     Train net output #0: loss = 1.06039 (* 1 = 1.06039 loss)
I0428 15:50:11.528373 29440 sgd_solver.cpp:105] Iteration 7296, lr = 0.00235691
I0428 15:50:16.799921 29440 solver.cpp:218] Iteration 7308 (2.27646 iter/s, 5.27134s/12 iters), loss = 0.929493
I0428 15:50:16.799959 29440 solver.cpp:237]     Train net output #0: loss = 0.929493 (* 1 = 0.929493 loss)
I0428 15:50:16.799968 29440 sgd_solver.cpp:105] Iteration 7308, lr = 0.00235131
I0428 15:50:22.104357 29440 solver.cpp:218] Iteration 7320 (2.26237 iter/s, 5.30417s/12 iters), loss = 0.981951
I0428 15:50:22.104416 29440 solver.cpp:237]     Train net output #0: loss = 0.981951 (* 1 = 0.981951 loss)
I0428 15:50:22.104429 29440 sgd_solver.cpp:105] Iteration 7320, lr = 0.00234573
I0428 15:50:27.597800 29440 solver.cpp:218] Iteration 7332 (2.18454 iter/s, 5.49314s/12 iters), loss = 0.987527
I0428 15:50:27.597955 29440 solver.cpp:237]     Train net output #0: loss = 0.987527 (* 1 = 0.987527 loss)
I0428 15:50:27.597970 29440 sgd_solver.cpp:105] Iteration 7332, lr = 0.00234016
I0428 15:50:32.514035 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_7344.caffemodel
I0428 15:50:34.035223 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_7344.solverstate
I0428 15:50:35.097048 29440 solver.cpp:330] Iteration 7344, Testing net (#0)
I0428 15:50:35.097074 29440 net.cpp:676] Ignoring source layer train-data
I0428 15:50:36.848779 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:50:40.172155 29440 solver.cpp:397]     Test net output #0: accuracy = 0.309436
I0428 15:50:40.172194 29440 solver.cpp:397]     Test net output #1: loss = 3.47308 (* 1 = 3.47308 loss)
I0428 15:50:40.403470 29440 solver.cpp:218] Iteration 7344 (0.937133 iter/s, 12.805s/12 iters), loss = 1.29045
I0428 15:50:40.403517 29440 solver.cpp:237]     Train net output #0: loss = 1.29045 (* 1 = 1.29045 loss)
I0428 15:50:40.403527 29440 sgd_solver.cpp:105] Iteration 7344, lr = 0.0023346
I0428 15:50:45.087965 29440 solver.cpp:218] Iteration 7356 (2.56178 iter/s, 4.68425s/12 iters), loss = 1.015
I0428 15:50:45.088008 29440 solver.cpp:237]     Train net output #0: loss = 1.015 (* 1 = 1.015 loss)
I0428 15:50:45.088017 29440 sgd_solver.cpp:105] Iteration 7356, lr = 0.00232906
I0428 15:50:50.838008 29440 solver.cpp:218] Iteration 7368 (2.08704 iter/s, 5.74976s/12 iters), loss = 1.00404
I0428 15:50:50.838061 29440 solver.cpp:237]     Train net output #0: loss = 1.00404 (* 1 = 1.00404 loss)
I0428 15:50:50.838071 29440 sgd_solver.cpp:105] Iteration 7368, lr = 0.00232353
I0428 15:50:56.306123 29440 solver.cpp:218] Iteration 7380 (2.19465 iter/s, 5.46784s/12 iters), loss = 0.821369
I0428 15:50:56.306167 29440 solver.cpp:237]     Train net output #0: loss = 0.821369 (* 1 = 0.821369 loss)
I0428 15:50:56.306197 29440 sgd_solver.cpp:105] Iteration 7380, lr = 0.00231802
I0428 15:50:57.914942 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:51:01.620451 29440 solver.cpp:218] Iteration 7392 (2.25816 iter/s, 5.31406s/12 iters), loss = 1.01183
I0428 15:51:01.620535 29440 solver.cpp:237]     Train net output #0: loss = 1.01183 (* 1 = 1.01183 loss)
I0428 15:51:01.620545 29440 sgd_solver.cpp:105] Iteration 7392, lr = 0.00231251
I0428 15:51:07.227356 29440 solver.cpp:218] Iteration 7404 (2.14034 iter/s, 5.6066s/12 iters), loss = 1.12687
I0428 15:51:07.227399 29440 solver.cpp:237]     Train net output #0: loss = 1.12687 (* 1 = 1.12687 loss)
I0428 15:51:07.227407 29440 sgd_solver.cpp:105] Iteration 7404, lr = 0.00230702
I0428 15:51:12.363391 29440 solver.cpp:218] Iteration 7416 (2.33755 iter/s, 5.13357s/12 iters), loss = 0.911577
I0428 15:51:12.363431 29440 solver.cpp:237]     Train net output #0: loss = 0.911577 (* 1 = 0.911577 loss)
I0428 15:51:12.363441 29440 sgd_solver.cpp:105] Iteration 7416, lr = 0.00230154
I0428 15:51:17.662684 29440 solver.cpp:218] Iteration 7428 (2.26551 iter/s, 5.29683s/12 iters), loss = 1.06299
I0428 15:51:17.662730 29440 solver.cpp:237]     Train net output #0: loss = 1.06299 (* 1 = 1.06299 loss)
I0428 15:51:17.662739 29440 sgd_solver.cpp:105] Iteration 7428, lr = 0.00229608
I0428 15:51:23.165668 29440 solver.cpp:218] Iteration 7440 (2.18074 iter/s, 5.50271s/12 iters), loss = 1.16201
I0428 15:51:23.165709 29440 solver.cpp:237]     Train net output #0: loss = 1.16201 (* 1 = 1.16201 loss)
I0428 15:51:23.165719 29440 sgd_solver.cpp:105] Iteration 7440, lr = 0.00229063
I0428 15:51:25.119498 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_7446.caffemodel
I0428 15:51:26.501983 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_7446.solverstate
I0428 15:51:27.567036 29440 solver.cpp:330] Iteration 7446, Testing net (#0)
I0428 15:51:27.567061 29440 net.cpp:676] Ignoring source layer train-data
I0428 15:51:29.356029 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:51:32.599345 29440 solver.cpp:397]     Test net output #0: accuracy = 0.318627
I0428 15:51:32.599387 29440 solver.cpp:397]     Test net output #1: loss = 3.47313 (* 1 = 3.47313 loss)
I0428 15:51:34.827610 29440 solver.cpp:218] Iteration 7452 (1.02903 iter/s, 11.6614s/12 iters), loss = 1.29557
I0428 15:51:34.827657 29440 solver.cpp:237]     Train net output #0: loss = 1.29557 (* 1 = 1.29557 loss)
I0428 15:51:34.827667 29440 sgd_solver.cpp:105] Iteration 7452, lr = 0.00228519
I0428 15:51:40.170138 29440 solver.cpp:218] Iteration 7464 (2.24624 iter/s, 5.34226s/12 iters), loss = 0.904612
I0428 15:51:40.170181 29440 solver.cpp:237]     Train net output #0: loss = 0.904612 (* 1 = 0.904612 loss)
I0428 15:51:40.170189 29440 sgd_solver.cpp:105] Iteration 7464, lr = 0.00227976
I0428 15:51:45.169595 29440 solver.cpp:218] Iteration 7476 (2.4014 iter/s, 4.99709s/12 iters), loss = 0.818725
I0428 15:51:45.169637 29440 solver.cpp:237]     Train net output #0: loss = 0.818725 (* 1 = 0.818725 loss)
I0428 15:51:45.169646 29440 sgd_solver.cpp:105] Iteration 7476, lr = 0.00227435
I0428 15:51:49.098915 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:51:50.922705 29440 solver.cpp:218] Iteration 7488 (2.08593 iter/s, 5.75283s/12 iters), loss = 1.03829
I0428 15:51:50.922750 29440 solver.cpp:237]     Train net output #0: loss = 1.03829 (* 1 = 1.03829 loss)
I0428 15:51:50.922760 29440 sgd_solver.cpp:105] Iteration 7488, lr = 0.00226895
I0428 15:51:57.303390 29440 solver.cpp:218] Iteration 7500 (1.88141 iter/s, 6.37819s/12 iters), loss = 0.982427
I0428 15:51:57.303434 29440 solver.cpp:237]     Train net output #0: loss = 0.982427 (* 1 = 0.982427 loss)
I0428 15:51:57.303443 29440 sgd_solver.cpp:105] Iteration 7500, lr = 0.00226357
I0428 15:52:02.529700 29440 solver.cpp:218] Iteration 7512 (2.29619 iter/s, 5.22605s/12 iters), loss = 1.05568
I0428 15:52:02.539276 29440 solver.cpp:237]     Train net output #0: loss = 1.05568 (* 1 = 1.05568 loss)
I0428 15:52:02.539288 29440 sgd_solver.cpp:105] Iteration 7512, lr = 0.00225819
I0428 15:52:07.788266 29440 solver.cpp:218] Iteration 7524 (2.28625 iter/s, 5.24878s/12 iters), loss = 0.874257
I0428 15:52:07.788311 29440 solver.cpp:237]     Train net output #0: loss = 0.874257 (* 1 = 0.874257 loss)
I0428 15:52:07.788321 29440 sgd_solver.cpp:105] Iteration 7524, lr = 0.00225283
I0428 15:52:13.214798 29440 solver.cpp:218] Iteration 7536 (2.21147 iter/s, 5.42626s/12 iters), loss = 0.820808
I0428 15:52:13.214839 29440 solver.cpp:237]     Train net output #0: loss = 0.820808 (* 1 = 0.820808 loss)
I0428 15:52:13.214848 29440 sgd_solver.cpp:105] Iteration 7536, lr = 0.00224748
I0428 15:52:17.989512 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_7548.caffemodel
I0428 15:52:20.301995 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_7548.solverstate
I0428 15:52:21.369565 29440 solver.cpp:330] Iteration 7548, Testing net (#0)
I0428 15:52:21.369585 29440 net.cpp:676] Ignoring source layer train-data
I0428 15:52:22.948535 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:52:26.487272 29440 solver.cpp:397]     Test net output #0: accuracy = 0.324755
I0428 15:52:26.487301 29440 solver.cpp:397]     Test net output #1: loss = 3.30802 (* 1 = 3.30802 loss)
I0428 15:52:26.851668 29440 solver.cpp:218] Iteration 7548 (0.880004 iter/s, 13.6363s/12 iters), loss = 1.18977
I0428 15:52:26.851713 29440 solver.cpp:237]     Train net output #0: loss = 1.18977 (* 1 = 1.18977 loss)
I0428 15:52:26.851722 29440 sgd_solver.cpp:105] Iteration 7548, lr = 0.00224215
I0428 15:52:31.638525 29440 solver.cpp:218] Iteration 7560 (2.507 iter/s, 4.7866s/12 iters), loss = 0.87438
I0428 15:52:31.638576 29440 solver.cpp:237]     Train net output #0: loss = 0.87438 (* 1 = 0.87438 loss)
I0428 15:52:31.638586 29440 sgd_solver.cpp:105] Iteration 7560, lr = 0.00223682
I0428 15:52:37.326889 29440 solver.cpp:218] Iteration 7572 (2.10969 iter/s, 5.68805s/12 iters), loss = 0.962622
I0428 15:52:37.326983 29440 solver.cpp:237]     Train net output #0: loss = 0.962622 (* 1 = 0.962622 loss)
I0428 15:52:37.326993 29440 sgd_solver.cpp:105] Iteration 7572, lr = 0.00223151
I0428 15:52:42.666950 29440 solver.cpp:218] Iteration 7584 (2.24821 iter/s, 5.33758s/12 iters), loss = 1.217
I0428 15:52:42.666993 29440 solver.cpp:237]     Train net output #0: loss = 1.217 (* 1 = 1.217 loss)
I0428 15:52:42.667002 29440 sgd_solver.cpp:105] Iteration 7584, lr = 0.00222621
I0428 15:52:43.125774 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:52:47.856986 29440 solver.cpp:218] Iteration 7596 (2.31322 iter/s, 5.18757s/12 iters), loss = 0.6906
I0428 15:52:47.857031 29440 solver.cpp:237]     Train net output #0: loss = 0.6906 (* 1 = 0.6906 loss)
I0428 15:52:47.857039 29440 sgd_solver.cpp:105] Iteration 7596, lr = 0.00222093
I0428 15:52:53.466445 29440 solver.cpp:218] Iteration 7608 (2.13935 iter/s, 5.60918s/12 iters), loss = 1.10825
I0428 15:52:53.466485 29440 solver.cpp:237]     Train net output #0: loss = 1.10825 (* 1 = 1.10825 loss)
I0428 15:52:53.466493 29440 sgd_solver.cpp:105] Iteration 7608, lr = 0.00221565
I0428 15:52:59.236361 29440 solver.cpp:218] Iteration 7620 (2.07985 iter/s, 5.76964s/12 iters), loss = 0.839169
I0428 15:52:59.236408 29440 solver.cpp:237]     Train net output #0: loss = 0.839169 (* 1 = 0.839169 loss)
I0428 15:52:59.236418 29440 sgd_solver.cpp:105] Iteration 7620, lr = 0.00221039
I0428 15:53:04.735770 29440 solver.cpp:218] Iteration 7632 (2.18216 iter/s, 5.49914s/12 iters), loss = 1.05504
I0428 15:53:04.735816 29440 solver.cpp:237]     Train net output #0: loss = 1.05504 (* 1 = 1.05504 loss)
I0428 15:53:04.735826 29440 sgd_solver.cpp:105] Iteration 7632, lr = 0.00220515
I0428 15:53:10.131970 29440 solver.cpp:218] Iteration 7644 (2.22389 iter/s, 5.39594s/12 iters), loss = 1.01026
I0428 15:53:10.132089 29440 solver.cpp:237]     Train net output #0: loss = 1.01026 (* 1 = 1.01026 loss)
I0428 15:53:10.132099 29440 sgd_solver.cpp:105] Iteration 7644, lr = 0.00219991
I0428 15:53:12.122313 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_7650.caffemodel
I0428 15:53:13.661761 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_7650.solverstate
I0428 15:53:14.719914 29440 solver.cpp:330] Iteration 7650, Testing net (#0)
I0428 15:53:14.719944 29440 net.cpp:676] Ignoring source layer train-data
I0428 15:53:16.150492 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:53:17.345335 29440 blocking_queue.cpp:49] Waiting for data
I0428 15:53:19.668462 29440 solver.cpp:397]     Test net output #0: accuracy = 0.346814
I0428 15:53:19.668509 29440 solver.cpp:397]     Test net output #1: loss = 3.3515 (* 1 = 3.3515 loss)
I0428 15:53:21.781736 29440 solver.cpp:218] Iteration 7656 (1.03012 iter/s, 11.6492s/12 iters), loss = 0.81184
I0428 15:53:21.781802 29440 solver.cpp:237]     Train net output #0: loss = 0.81184 (* 1 = 0.81184 loss)
I0428 15:53:21.781814 29440 sgd_solver.cpp:105] Iteration 7656, lr = 0.00219469
I0428 15:53:27.285830 29440 solver.cpp:218] Iteration 7668 (2.18031 iter/s, 5.50381s/12 iters), loss = 0.917883
I0428 15:53:27.285874 29440 solver.cpp:237]     Train net output #0: loss = 0.917883 (* 1 = 0.917883 loss)
I0428 15:53:27.285883 29440 sgd_solver.cpp:105] Iteration 7668, lr = 0.00218948
I0428 15:53:32.497501 29440 solver.cpp:218] Iteration 7680 (2.30361 iter/s, 5.20922s/12 iters), loss = 0.927989
I0428 15:53:32.497545 29440 solver.cpp:237]     Train net output #0: loss = 0.927989 (* 1 = 0.927989 loss)
I0428 15:53:32.497552 29440 sgd_solver.cpp:105] Iteration 7680, lr = 0.00218428
I0428 15:53:35.417910 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:53:37.922320 29440 solver.cpp:218] Iteration 7692 (2.21216 iter/s, 5.42455s/12 iters), loss = 0.958572
I0428 15:53:37.922364 29440 solver.cpp:237]     Train net output #0: loss = 0.958572 (* 1 = 0.958572 loss)
I0428 15:53:37.922374 29440 sgd_solver.cpp:105] Iteration 7692, lr = 0.00217909
I0428 15:53:43.428978 29440 solver.cpp:218] Iteration 7704 (2.17929 iter/s, 5.50639s/12 iters), loss = 0.714838
I0428 15:53:43.429078 29440 solver.cpp:237]     Train net output #0: loss = 0.714838 (* 1 = 0.714838 loss)
I0428 15:53:43.429091 29440 sgd_solver.cpp:105] Iteration 7704, lr = 0.00217392
I0428 15:53:49.309772 29440 solver.cpp:218] Iteration 7716 (2.04066 iter/s, 5.88045s/12 iters), loss = 0.772859
I0428 15:53:49.309821 29440 solver.cpp:237]     Train net output #0: loss = 0.772859 (* 1 = 0.772859 loss)
I0428 15:53:49.309832 29440 sgd_solver.cpp:105] Iteration 7716, lr = 0.00216876
I0428 15:53:54.963182 29440 solver.cpp:218] Iteration 7728 (2.12272 iter/s, 5.65313s/12 iters), loss = 0.659697
I0428 15:53:54.963229 29440 solver.cpp:237]     Train net output #0: loss = 0.659697 (* 1 = 0.659697 loss)
I0428 15:53:54.963238 29440 sgd_solver.cpp:105] Iteration 7728, lr = 0.00216361
I0428 15:54:00.298575 29440 solver.cpp:218] Iteration 7740 (2.24924 iter/s, 5.33512s/12 iters), loss = 0.78506
I0428 15:54:00.298622 29440 solver.cpp:237]     Train net output #0: loss = 0.78506 (* 1 = 0.78506 loss)
I0428 15:54:00.298631 29440 sgd_solver.cpp:105] Iteration 7740, lr = 0.00215847
I0428 15:54:04.957702 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_7752.caffemodel
I0428 15:54:07.269310 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_7752.solverstate
I0428 15:54:09.658219 29440 solver.cpp:330] Iteration 7752, Testing net (#0)
I0428 15:54:09.658241 29440 net.cpp:676] Ignoring source layer train-data
I0428 15:54:11.219063 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:54:14.629264 29440 solver.cpp:397]     Test net output #0: accuracy = 0.309436
I0428 15:54:14.629386 29440 solver.cpp:397]     Test net output #1: loss = 3.51555 (* 1 = 3.51555 loss)
I0428 15:54:14.993170 29440 solver.cpp:218] Iteration 7752 (0.816662 iter/s, 14.694s/12 iters), loss = 1.0562
I0428 15:54:14.994812 29440 solver.cpp:237]     Train net output #0: loss = 1.0562 (* 1 = 1.0562 loss)
I0428 15:54:14.994824 29440 sgd_solver.cpp:105] Iteration 7752, lr = 0.00215335
I0428 15:54:19.576030 29440 solver.cpp:218] Iteration 7764 (2.6195 iter/s, 4.58103s/12 iters), loss = 1.25061
I0428 15:54:19.576073 29440 solver.cpp:237]     Train net output #0: loss = 1.25061 (* 1 = 1.25061 loss)
I0428 15:54:19.576082 29440 sgd_solver.cpp:105] Iteration 7764, lr = 0.00214823
I0428 15:54:25.103664 29440 solver.cpp:218] Iteration 7776 (2.17188 iter/s, 5.52517s/12 iters), loss = 0.77144
I0428 15:54:25.103710 29440 solver.cpp:237]     Train net output #0: loss = 0.77144 (* 1 = 0.77144 loss)
I0428 15:54:25.103719 29440 sgd_solver.cpp:105] Iteration 7776, lr = 0.00214313
I0428 15:54:30.542703 29440 solver.cpp:218] Iteration 7788 (2.20638 iter/s, 5.43877s/12 iters), loss = 1.12917
I0428 15:54:30.542750 29440 solver.cpp:237]     Train net output #0: loss = 1.12917 (* 1 = 1.12917 loss)
I0428 15:54:30.542759 29440 sgd_solver.cpp:105] Iteration 7788, lr = 0.00213805
I0428 15:54:30.550040 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:54:36.239073 29440 solver.cpp:218] Iteration 7800 (2.10671 iter/s, 5.69609s/12 iters), loss = 0.818567
I0428 15:54:36.239118 29440 solver.cpp:237]     Train net output #0: loss = 0.818567 (* 1 = 0.818567 loss)
I0428 15:54:36.239126 29440 sgd_solver.cpp:105] Iteration 7800, lr = 0.00213297
I0428 15:54:41.594288 29440 solver.cpp:218] Iteration 7812 (2.24092 iter/s, 5.35495s/12 iters), loss = 0.887493
I0428 15:54:41.594358 29440 solver.cpp:237]     Train net output #0: loss = 0.887493 (* 1 = 0.887493 loss)
I0428 15:54:41.594372 29440 sgd_solver.cpp:105] Iteration 7812, lr = 0.00212791
I0428 15:54:46.671967 29440 solver.cpp:218] Iteration 7824 (2.36341 iter/s, 5.0774s/12 iters), loss = 0.930317
I0428 15:54:46.672072 29440 solver.cpp:237]     Train net output #0: loss = 0.930317 (* 1 = 0.930317 loss)
I0428 15:54:46.672086 29440 sgd_solver.cpp:105] Iteration 7824, lr = 0.00212285
I0428 15:54:52.143007 29440 solver.cpp:218] Iteration 7836 (2.1935 iter/s, 5.47071s/12 iters), loss = 0.825444
I0428 15:54:52.143070 29440 solver.cpp:237]     Train net output #0: loss = 0.825444 (* 1 = 0.825444 loss)
I0428 15:54:52.143083 29440 sgd_solver.cpp:105] Iteration 7836, lr = 0.00211781
I0428 15:54:57.690671 29440 solver.cpp:218] Iteration 7848 (2.16319 iter/s, 5.54737s/12 iters), loss = 0.824859
I0428 15:54:57.690716 29440 solver.cpp:237]     Train net output #0: loss = 0.824859 (* 1 = 0.824859 loss)
I0428 15:54:57.690724 29440 sgd_solver.cpp:105] Iteration 7848, lr = 0.00211279
I0428 15:54:59.819896 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_7854.caffemodel
I0428 15:55:05.612674 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_7854.solverstate
I0428 15:55:08.076884 29440 solver.cpp:330] Iteration 7854, Testing net (#0)
I0428 15:55:08.076905 29440 net.cpp:676] Ignoring source layer train-data
I0428 15:55:09.452457 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:55:12.992043 29440 solver.cpp:397]     Test net output #0: accuracy = 0.351103
I0428 15:55:12.992071 29440 solver.cpp:397]     Test net output #1: loss = 3.28648 (* 1 = 3.28648 loss)
I0428 15:55:15.145793 29440 solver.cpp:218] Iteration 7860 (0.687506 iter/s, 17.4544s/12 iters), loss = 1.10367
I0428 15:55:15.145849 29440 solver.cpp:237]     Train net output #0: loss = 1.10367 (* 1 = 1.10367 loss)
I0428 15:55:15.145860 29440 sgd_solver.cpp:105] Iteration 7860, lr = 0.00210777
I0428 15:55:20.605890 29440 solver.cpp:218] Iteration 7872 (2.19788 iter/s, 5.45982s/12 iters), loss = 0.71509
I0428 15:55:20.610623 29440 solver.cpp:237]     Train net output #0: loss = 0.71509 (* 1 = 0.71509 loss)
I0428 15:55:20.610638 29440 sgd_solver.cpp:105] Iteration 7872, lr = 0.00210277
I0428 15:55:25.595861 29440 solver.cpp:218] Iteration 7884 (2.4072 iter/s, 4.98504s/12 iters), loss = 0.605078
I0428 15:55:25.595907 29440 solver.cpp:237]     Train net output #0: loss = 0.605078 (* 1 = 0.605078 loss)
I0428 15:55:25.595916 29440 sgd_solver.cpp:105] Iteration 7884, lr = 0.00209777
I0428 15:55:28.150063 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:55:31.395095 29440 solver.cpp:218] Iteration 7896 (2.06934 iter/s, 5.79895s/12 iters), loss = 0.921824
I0428 15:55:31.395140 29440 solver.cpp:237]     Train net output #0: loss = 0.921824 (* 1 = 0.921824 loss)
I0428 15:55:31.395149 29440 sgd_solver.cpp:105] Iteration 7896, lr = 0.00209279
I0428 15:55:36.887878 29440 solver.cpp:218] Iteration 7908 (2.18479 iter/s, 5.49251s/12 iters), loss = 1.04851
I0428 15:55:36.887934 29440 solver.cpp:237]     Train net output #0: loss = 1.04851 (* 1 = 1.04851 loss)
I0428 15:55:36.887945 29440 sgd_solver.cpp:105] Iteration 7908, lr = 0.00208782
I0428 15:55:42.259745 29440 solver.cpp:218] Iteration 7920 (2.23489 iter/s, 5.36938s/12 iters), loss = 0.705195
I0428 15:55:42.259799 29440 solver.cpp:237]     Train net output #0: loss = 0.705195 (* 1 = 0.705195 loss)
I0428 15:55:42.259814 29440 sgd_solver.cpp:105] Iteration 7920, lr = 0.00208287
I0428 15:55:47.481070 29440 solver.cpp:218] Iteration 7932 (2.2984 iter/s, 5.22103s/12 iters), loss = 0.888436
I0428 15:55:47.481114 29440 solver.cpp:237]     Train net output #0: loss = 0.888436 (* 1 = 0.888436 loss)
I0428 15:55:47.481123 29440 sgd_solver.cpp:105] Iteration 7932, lr = 0.00207792
I0428 15:55:53.263008 29440 solver.cpp:218] Iteration 7944 (2.07553 iter/s, 5.78166s/12 iters), loss = 0.855723
I0428 15:55:53.263482 29440 solver.cpp:237]     Train net output #0: loss = 0.855723 (* 1 = 0.855723 loss)
I0428 15:55:53.263492 29440 sgd_solver.cpp:105] Iteration 7944, lr = 0.00207299
I0428 15:55:58.106858 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_7956.caffemodel
I0428 15:55:59.636166 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_7956.solverstate
I0428 15:56:01.765450 29440 solver.cpp:330] Iteration 7956, Testing net (#0)
I0428 15:56:01.765476 29440 net.cpp:676] Ignoring source layer train-data
I0428 15:56:03.230648 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:56:06.937801 29440 solver.cpp:397]     Test net output #0: accuracy = 0.359069
I0428 15:56:06.937835 29440 solver.cpp:397]     Test net output #1: loss = 3.24825 (* 1 = 3.24825 loss)
I0428 15:56:07.087741 29440 solver.cpp:218] Iteration 7956 (0.868185 iter/s, 13.8219s/12 iters), loss = 0.757688
I0428 15:56:07.087797 29440 solver.cpp:237]     Train net output #0: loss = 0.757688 (* 1 = 0.757688 loss)
I0428 15:56:07.087810 29440 sgd_solver.cpp:105] Iteration 7956, lr = 0.00206807
I0428 15:56:12.088467 29440 solver.cpp:218] Iteration 7968 (2.39978 iter/s, 5.00046s/12 iters), loss = 1.13753
I0428 15:56:12.088532 29440 solver.cpp:237]     Train net output #0: loss = 1.13753 (* 1 = 1.13753 loss)
I0428 15:56:12.088541 29440 sgd_solver.cpp:105] Iteration 7968, lr = 0.00206316
I0428 15:56:17.478310 29440 solver.cpp:218] Iteration 7980 (2.22744 iter/s, 5.38736s/12 iters), loss = 0.78172
I0428 15:56:17.478346 29440 solver.cpp:237]     Train net output #0: loss = 0.78172 (* 1 = 0.78172 loss)
I0428 15:56:17.478354 29440 sgd_solver.cpp:105] Iteration 7980, lr = 0.00205826
I0428 15:56:21.984795 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:56:22.761822 29440 solver.cpp:218] Iteration 7992 (2.27133 iter/s, 5.28326s/12 iters), loss = 0.920023
I0428 15:56:22.761883 29440 solver.cpp:237]     Train net output #0: loss = 0.920023 (* 1 = 0.920023 loss)
I0428 15:56:22.761895 29440 sgd_solver.cpp:105] Iteration 7992, lr = 0.00205337
I0428 15:56:27.907114 29440 solver.cpp:218] Iteration 8004 (2.33236 iter/s, 5.14501s/12 iters), loss = 0.825885
I0428 15:56:27.911576 29440 solver.cpp:237]     Train net output #0: loss = 0.825885 (* 1 = 0.825885 loss)
I0428 15:56:27.911587 29440 sgd_solver.cpp:105] Iteration 8004, lr = 0.0020485
I0428 15:56:33.302832 29440 solver.cpp:218] Iteration 8016 (2.22591 iter/s, 5.39104s/12 iters), loss = 1.00037
I0428 15:56:33.302878 29440 solver.cpp:237]     Train net output #0: loss = 1.00037 (* 1 = 1.00037 loss)
I0428 15:56:33.302887 29440 sgd_solver.cpp:105] Iteration 8016, lr = 0.00204363
I0428 15:56:38.681864 29440 solver.cpp:218] Iteration 8028 (2.2319 iter/s, 5.37657s/12 iters), loss = 0.669802
I0428 15:56:38.681910 29440 solver.cpp:237]     Train net output #0: loss = 0.669802 (* 1 = 0.669802 loss)
I0428 15:56:38.681919 29440 sgd_solver.cpp:105] Iteration 8028, lr = 0.00203878
I0428 15:56:44.067665 29440 solver.cpp:218] Iteration 8040 (2.2291 iter/s, 5.38334s/12 iters), loss = 0.952809
I0428 15:56:44.067713 29440 solver.cpp:237]     Train net output #0: loss = 0.952809 (* 1 = 0.952809 loss)
I0428 15:56:44.067721 29440 sgd_solver.cpp:105] Iteration 8040, lr = 0.00203394
I0428 15:56:49.480077 29440 solver.cpp:218] Iteration 8052 (2.21814 iter/s, 5.40993s/12 iters), loss = 0.835378
I0428 15:56:49.480123 29440 solver.cpp:237]     Train net output #0: loss = 0.835378 (* 1 = 0.835378 loss)
I0428 15:56:49.480132 29440 sgd_solver.cpp:105] Iteration 8052, lr = 0.00202911
I0428 15:56:51.750495 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_8058.caffemodel
I0428 15:56:53.236411 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_8058.solverstate
I0428 15:56:54.295246 29440 solver.cpp:330] Iteration 8058, Testing net (#0)
I0428 15:56:54.295265 29440 net.cpp:676] Ignoring source layer train-data
I0428 15:56:55.639034 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:56:59.298344 29440 solver.cpp:397]     Test net output #0: accuracy = 0.352328
I0428 15:56:59.298458 29440 solver.cpp:397]     Test net output #1: loss = 3.33321 (* 1 = 3.33321 loss)
I0428 15:57:01.520469 29440 solver.cpp:218] Iteration 8064 (0.996688 iter/s, 12.0399s/12 iters), loss = 0.656013
I0428 15:57:01.520556 29440 solver.cpp:237]     Train net output #0: loss = 0.656013 (* 1 = 0.656013 loss)
I0428 15:57:01.520566 29440 sgd_solver.cpp:105] Iteration 8064, lr = 0.00202429
I0428 15:57:06.848870 29440 solver.cpp:218] Iteration 8076 (2.25313 iter/s, 5.32592s/12 iters), loss = 0.64283
I0428 15:57:06.848917 29440 solver.cpp:237]     Train net output #0: loss = 0.64283 (* 1 = 0.64283 loss)
I0428 15:57:06.848927 29440 sgd_solver.cpp:105] Iteration 8076, lr = 0.00201949
I0428 15:57:12.222235 29440 solver.cpp:218] Iteration 8088 (2.23335 iter/s, 5.3731s/12 iters), loss = 0.68104
I0428 15:57:12.222282 29440 solver.cpp:237]     Train net output #0: loss = 0.68104 (* 1 = 0.68104 loss)
I0428 15:57:12.222292 29440 sgd_solver.cpp:105] Iteration 8088, lr = 0.00201469
I0428 15:57:13.748651 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:57:18.093142 29440 solver.cpp:218] Iteration 8100 (2.04408 iter/s, 5.87062s/12 iters), loss = 0.669804
I0428 15:57:18.093185 29440 solver.cpp:237]     Train net output #0: loss = 0.669804 (* 1 = 0.669804 loss)
I0428 15:57:18.093194 29440 sgd_solver.cpp:105] Iteration 8100, lr = 0.00200991
I0428 15:57:23.279232 29440 solver.cpp:218] Iteration 8112 (2.31498 iter/s, 5.18362s/12 iters), loss = 0.930965
I0428 15:57:23.279292 29440 solver.cpp:237]     Train net output #0: loss = 0.930965 (* 1 = 0.930965 loss)
I0428 15:57:23.279306 29440 sgd_solver.cpp:105] Iteration 8112, lr = 0.00200514
I0428 15:57:28.614519 29440 solver.cpp:218] Iteration 8124 (2.24929 iter/s, 5.33501s/12 iters), loss = 0.644363
I0428 15:57:28.614564 29440 solver.cpp:237]     Train net output #0: loss = 0.644363 (* 1 = 0.644363 loss)
I0428 15:57:28.614573 29440 sgd_solver.cpp:105] Iteration 8124, lr = 0.00200038
I0428 15:57:34.445554 29440 solver.cpp:218] Iteration 8136 (2.05805 iter/s, 5.83075s/12 iters), loss = 0.836655
I0428 15:57:34.445708 29440 solver.cpp:237]     Train net output #0: loss = 0.836655 (* 1 = 0.836655 loss)
I0428 15:57:34.445719 29440 sgd_solver.cpp:105] Iteration 8136, lr = 0.00199563
I0428 15:57:39.702847 29440 solver.cpp:218] Iteration 8148 (2.28362 iter/s, 5.25482s/12 iters), loss = 0.921595
I0428 15:57:39.702896 29440 solver.cpp:237]     Train net output #0: loss = 0.921595 (* 1 = 0.921595 loss)
I0428 15:57:39.702908 29440 sgd_solver.cpp:105] Iteration 8148, lr = 0.00199089
I0428 15:57:44.618664 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_8160.caffemodel
I0428 15:57:47.189963 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_8160.solverstate
I0428 15:57:51.323276 29440 solver.cpp:330] Iteration 8160, Testing net (#0)
I0428 15:57:51.323302 29440 net.cpp:676] Ignoring source layer train-data
I0428 15:57:52.638872 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:57:56.310551 29440 solver.cpp:397]     Test net output #0: accuracy = 0.370711
I0428 15:57:56.310581 29440 solver.cpp:397]     Test net output #1: loss = 3.25945 (* 1 = 3.25945 loss)
I0428 15:57:56.461664 29440 solver.cpp:218] Iteration 8160 (0.716071 iter/s, 16.7581s/12 iters), loss = 0.647678
I0428 15:57:56.461719 29440 solver.cpp:237]     Train net output #0: loss = 0.647678 (* 1 = 0.647678 loss)
I0428 15:57:56.461731 29440 sgd_solver.cpp:105] Iteration 8160, lr = 0.00198616
I0428 15:58:01.386971 29440 solver.cpp:218] Iteration 8172 (2.43652 iter/s, 4.92505s/12 iters), loss = 0.744634
I0428 15:58:01.387018 29440 solver.cpp:237]     Train net output #0: loss = 0.744634 (* 1 = 0.744634 loss)
I0428 15:58:01.387027 29440 sgd_solver.cpp:105] Iteration 8172, lr = 0.00198145
I0428 15:58:07.072844 29440 solver.cpp:218] Iteration 8184 (2.11141 iter/s, 5.6834s/12 iters), loss = 0.783353
I0428 15:58:07.072954 29440 solver.cpp:237]     Train net output #0: loss = 0.783353 (* 1 = 0.783353 loss)
I0428 15:58:07.072964 29440 sgd_solver.cpp:105] Iteration 8184, lr = 0.00197674
I0428 15:58:10.649739 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:58:12.532025 29440 solver.cpp:218] Iteration 8196 (2.19913 iter/s, 5.4567s/12 iters), loss = 0.672777
I0428 15:58:12.532068 29440 solver.cpp:237]     Train net output #0: loss = 0.672777 (* 1 = 0.672777 loss)
I0428 15:58:12.532076 29440 sgd_solver.cpp:105] Iteration 8196, lr = 0.00197205
I0428 15:58:17.865084 29440 solver.cpp:218] Iteration 8208 (2.25023 iter/s, 5.33279s/12 iters), loss = 0.848416
I0428 15:58:17.865128 29440 solver.cpp:237]     Train net output #0: loss = 0.848416 (* 1 = 0.848416 loss)
I0428 15:58:17.865136 29440 sgd_solver.cpp:105] Iteration 8208, lr = 0.00196737
I0428 15:58:23.248098 29440 solver.cpp:218] Iteration 8220 (2.22934 iter/s, 5.38275s/12 iters), loss = 0.702825
I0428 15:58:23.248139 29440 solver.cpp:237]     Train net output #0: loss = 0.702825 (* 1 = 0.702825 loss)
I0428 15:58:23.248147 29440 sgd_solver.cpp:105] Iteration 8220, lr = 0.0019627
I0428 15:58:28.531193 29440 solver.cpp:218] Iteration 8232 (2.27151 iter/s, 5.28283s/12 iters), loss = 0.685603
I0428 15:58:28.531236 29440 solver.cpp:237]     Train net output #0: loss = 0.685603 (* 1 = 0.685603 loss)
I0428 15:58:28.531245 29440 sgd_solver.cpp:105] Iteration 8232, lr = 0.00195804
I0428 15:58:33.745121 29440 solver.cpp:218] Iteration 8244 (2.30164 iter/s, 5.21367s/12 iters), loss = 0.582904
I0428 15:58:33.745162 29440 solver.cpp:237]     Train net output #0: loss = 0.582904 (* 1 = 0.582904 loss)
I0428 15:58:33.745172 29440 sgd_solver.cpp:105] Iteration 8244, lr = 0.00195339
I0428 15:58:39.046089 29440 solver.cpp:218] Iteration 8256 (2.26385 iter/s, 5.30071s/12 iters), loss = 0.714491
I0428 15:58:39.046229 29440 solver.cpp:237]     Train net output #0: loss = 0.714491 (* 1 = 0.714491 loss)
I0428 15:58:39.046239 29440 sgd_solver.cpp:105] Iteration 8256, lr = 0.00194875
I0428 15:58:41.211701 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_8262.caffemodel
I0428 15:58:43.391181 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_8262.solverstate
I0428 15:58:45.479349 29440 solver.cpp:330] Iteration 8262, Testing net (#0)
I0428 15:58:45.479372 29440 net.cpp:676] Ignoring source layer train-data
I0428 15:58:46.711077 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:58:50.288915 29440 solver.cpp:397]     Test net output #0: accuracy = 0.356618
I0428 15:58:50.288952 29440 solver.cpp:397]     Test net output #1: loss = 3.39495 (* 1 = 3.39495 loss)
I0428 15:58:52.302975 29440 solver.cpp:218] Iteration 8268 (0.905235 iter/s, 13.2562s/12 iters), loss = 0.864364
I0428 15:58:52.303020 29440 solver.cpp:237]     Train net output #0: loss = 0.864364 (* 1 = 0.864364 loss)
I0428 15:58:52.303030 29440 sgd_solver.cpp:105] Iteration 8268, lr = 0.00194412
I0428 15:58:57.930588 29440 solver.cpp:218] Iteration 8280 (2.13245 iter/s, 5.62734s/12 iters), loss = 0.781394
I0428 15:58:57.930629 29440 solver.cpp:237]     Train net output #0: loss = 0.781394 (* 1 = 0.781394 loss)
I0428 15:58:57.930639 29440 sgd_solver.cpp:105] Iteration 8280, lr = 0.00193951
I0428 15:59:03.355191 29440 solver.cpp:218] Iteration 8292 (2.21315 iter/s, 5.42215s/12 iters), loss = 0.662791
I0428 15:59:03.355233 29440 solver.cpp:237]     Train net output #0: loss = 0.662791 (* 1 = 0.662791 loss)
I0428 15:59:03.355243 29440 sgd_solver.cpp:105] Iteration 8292, lr = 0.0019349
I0428 15:59:03.847775 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:59:08.636781 29440 solver.cpp:218] Iteration 8304 (2.27288 iter/s, 5.27965s/12 iters), loss = 0.847784
I0428 15:59:08.636819 29440 solver.cpp:237]     Train net output #0: loss = 0.847784 (* 1 = 0.847784 loss)
I0428 15:59:08.636829 29440 sgd_solver.cpp:105] Iteration 8304, lr = 0.00193031
I0428 15:59:13.909598 29440 solver.cpp:218] Iteration 8316 (2.27688 iter/s, 5.27036s/12 iters), loss = 0.962354
I0428 15:59:13.909701 29440 solver.cpp:237]     Train net output #0: loss = 0.962354 (* 1 = 0.962354 loss)
I0428 15:59:13.909713 29440 sgd_solver.cpp:105] Iteration 8316, lr = 0.00192573
I0428 15:59:19.268594 29440 solver.cpp:218] Iteration 8328 (2.24026 iter/s, 5.35651s/12 iters), loss = 0.646841
I0428 15:59:19.268642 29440 solver.cpp:237]     Train net output #0: loss = 0.646841 (* 1 = 0.646841 loss)
I0428 15:59:19.268652 29440 sgd_solver.cpp:105] Iteration 8328, lr = 0.00192115
I0428 15:59:24.676440 29440 solver.cpp:218] Iteration 8340 (2.22001 iter/s, 5.40538s/12 iters), loss = 0.583381
I0428 15:59:24.676538 29440 solver.cpp:237]     Train net output #0: loss = 0.583381 (* 1 = 0.583381 loss)
I0428 15:59:24.676551 29440 sgd_solver.cpp:105] Iteration 8340, lr = 0.00191659
I0428 15:59:29.681984 29440 solver.cpp:218] Iteration 8352 (2.39746 iter/s, 5.00529s/12 iters), loss = 0.757926
I0428 15:59:29.682027 29440 solver.cpp:237]     Train net output #0: loss = 0.757926 (* 1 = 0.757926 loss)
I0428 15:59:29.682036 29440 sgd_solver.cpp:105] Iteration 8352, lr = 0.00191204
I0428 15:59:34.530433 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_8364.caffemodel
I0428 15:59:36.006372 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_8364.solverstate
I0428 15:59:37.102627 29440 solver.cpp:330] Iteration 8364, Testing net (#0)
I0428 15:59:37.102644 29440 net.cpp:676] Ignoring source layer train-data
I0428 15:59:37.282744 29440 blocking_queue.cpp:49] Waiting for data
I0428 15:59:38.349200 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:59:42.341293 29440 solver.cpp:397]     Test net output #0: accuracy = 0.332721
I0428 15:59:42.341331 29440 solver.cpp:397]     Test net output #1: loss = 3.44642 (* 1 = 3.44642 loss)
I0428 15:59:42.512190 29440 solver.cpp:218] Iteration 8364 (0.935332 iter/s, 12.8297s/12 iters), loss = 0.692279
I0428 15:59:42.512238 29440 solver.cpp:237]     Train net output #0: loss = 0.692279 (* 1 = 0.692279 loss)
I0428 15:59:42.512248 29440 sgd_solver.cpp:105] Iteration 8364, lr = 0.0019075
I0428 15:59:47.606714 29440 solver.cpp:218] Iteration 8376 (2.35559 iter/s, 5.09426s/12 iters), loss = 0.692004
I0428 15:59:47.606868 29440 solver.cpp:237]     Train net output #0: loss = 0.692004 (* 1 = 0.692004 loss)
I0428 15:59:47.606879 29440 sgd_solver.cpp:105] Iteration 8376, lr = 0.00190297
I0428 15:59:52.923020 29440 solver.cpp:218] Iteration 8388 (2.25825 iter/s, 5.31385s/12 iters), loss = 0.75602
I0428 15:59:52.923063 29440 solver.cpp:237]     Train net output #0: loss = 0.75602 (* 1 = 0.75602 loss)
I0428 15:59:52.923074 29440 sgd_solver.cpp:105] Iteration 8388, lr = 0.00189846
I0428 15:59:55.532691 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 15:59:58.232802 29440 solver.cpp:218] Iteration 8400 (2.26103 iter/s, 5.30733s/12 iters), loss = 0.562879
I0428 15:59:58.232846 29440 solver.cpp:237]     Train net output #0: loss = 0.562879 (* 1 = 0.562879 loss)
I0428 15:59:58.232853 29440 sgd_solver.cpp:105] Iteration 8400, lr = 0.00189395
I0428 16:00:03.239058 29440 solver.cpp:218] Iteration 8412 (2.39818 iter/s, 5.0038s/12 iters), loss = 0.792083
I0428 16:00:03.239106 29440 solver.cpp:237]     Train net output #0: loss = 0.792083 (* 1 = 0.792083 loss)
I0428 16:00:03.239115 29440 sgd_solver.cpp:105] Iteration 8412, lr = 0.00188945
I0428 16:00:08.794615 29440 solver.cpp:218] Iteration 8424 (2.16011 iter/s, 5.55528s/12 iters), loss = 0.792303
I0428 16:00:08.794661 29440 solver.cpp:237]     Train net output #0: loss = 0.792303 (* 1 = 0.792303 loss)
I0428 16:00:08.794670 29440 sgd_solver.cpp:105] Iteration 8424, lr = 0.00188497
I0428 16:00:14.096421 29440 solver.cpp:218] Iteration 8436 (2.26349 iter/s, 5.30154s/12 iters), loss = 0.657448
I0428 16:00:14.096467 29440 solver.cpp:237]     Train net output #0: loss = 0.657448 (* 1 = 0.657448 loss)
I0428 16:00:14.096477 29440 sgd_solver.cpp:105] Iteration 8436, lr = 0.00188049
I0428 16:00:19.841485 29440 solver.cpp:218] Iteration 8448 (2.08885 iter/s, 5.74479s/12 iters), loss = 0.634586
I0428 16:00:19.841611 29440 solver.cpp:237]     Train net output #0: loss = 0.634586 (* 1 = 0.634586 loss)
I0428 16:00:19.841621 29440 sgd_solver.cpp:105] Iteration 8448, lr = 0.00187603
I0428 16:00:25.310971 29440 solver.cpp:218] Iteration 8460 (2.19413 iter/s, 5.46914s/12 iters), loss = 0.787094
I0428 16:00:25.311015 29440 solver.cpp:237]     Train net output #0: loss = 0.787094 (* 1 = 0.787094 loss)
I0428 16:00:25.311024 29440 sgd_solver.cpp:105] Iteration 8460, lr = 0.00187157
I0428 16:00:27.199937 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_8466.caffemodel
I0428 16:00:29.900775 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_8466.solverstate
I0428 16:00:30.956143 29440 solver.cpp:330] Iteration 8466, Testing net (#0)
I0428 16:00:30.956164 29440 net.cpp:676] Ignoring source layer train-data
I0428 16:00:32.095077 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 16:00:35.874097 29440 solver.cpp:397]     Test net output #0: accuracy = 0.347426
I0428 16:00:35.874126 29440 solver.cpp:397]     Test net output #1: loss = 3.50983 (* 1 = 3.50983 loss)
I0428 16:00:38.259202 29440 solver.cpp:218] Iteration 8472 (0.926807 iter/s, 12.9477s/12 iters), loss = 0.917034
I0428 16:00:38.259246 29440 solver.cpp:237]     Train net output #0: loss = 0.917034 (* 1 = 0.917034 loss)
I0428 16:00:38.259254 29440 sgd_solver.cpp:105] Iteration 8472, lr = 0.00186713
I0428 16:00:43.814721 29440 solver.cpp:218] Iteration 8484 (2.16087 iter/s, 5.55333s/12 iters), loss = 0.956437
I0428 16:00:43.814760 29440 solver.cpp:237]     Train net output #0: loss = 0.956437 (* 1 = 0.956437 loss)
I0428 16:00:43.814769 29440 sgd_solver.cpp:105] Iteration 8484, lr = 0.0018627
I0428 16:00:49.360627 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 16:00:49.586127 29440 solver.cpp:218] Iteration 8496 (2.07932 iter/s, 5.77113s/12 iters), loss = 0.715007
I0428 16:00:49.586172 29440 solver.cpp:237]     Train net output #0: loss = 0.715007 (* 1 = 0.715007 loss)
I0428 16:00:49.586182 29440 sgd_solver.cpp:105] Iteration 8496, lr = 0.00185827
I0428 16:00:54.990070 29440 solver.cpp:218] Iteration 8508 (2.22161 iter/s, 5.40148s/12 iters), loss = 0.669828
I0428 16:00:54.990255 29440 solver.cpp:237]     Train net output #0: loss = 0.669828 (* 1 = 0.669828 loss)
I0428 16:00:54.990265 29440 sgd_solver.cpp:105] Iteration 8508, lr = 0.00185386
I0428 16:01:00.159027 29440 solver.cpp:218] Iteration 8520 (2.32266 iter/s, 5.16649s/12 iters), loss = 0.918519
I0428 16:01:00.159086 29440 solver.cpp:237]     Train net output #0: loss = 0.918519 (* 1 = 0.918519 loss)
I0428 16:01:00.159098 29440 sgd_solver.cpp:105] Iteration 8520, lr = 0.00184946
I0428 16:01:05.603361 29440 solver.cpp:218] Iteration 8532 (2.20424 iter/s, 5.44406s/12 iters), loss = 0.691656
I0428 16:01:05.603406 29440 solver.cpp:237]     Train net output #0: loss = 0.691656 (* 1 = 0.691656 loss)
I0428 16:01:05.603415 29440 sgd_solver.cpp:105] Iteration 8532, lr = 0.00184507
I0428 16:01:11.109033 29440 solver.cpp:218] Iteration 8544 (2.17968 iter/s, 5.5054s/12 iters), loss = 0.68363
I0428 16:01:11.109079 29440 solver.cpp:237]     Train net output #0: loss = 0.68363 (* 1 = 0.68363 loss)
I0428 16:01:11.109088 29440 sgd_solver.cpp:105] Iteration 8544, lr = 0.00184069
I0428 16:01:16.711110 29440 solver.cpp:218] Iteration 8556 (2.14217 iter/s, 5.6018s/12 iters), loss = 0.734161
I0428 16:01:16.711155 29440 solver.cpp:237]     Train net output #0: loss = 0.734161 (* 1 = 0.734161 loss)
I0428 16:01:16.711163 29440 sgd_solver.cpp:105] Iteration 8556, lr = 0.00183632
I0428 16:01:21.356333 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_8568.caffemodel
I0428 16:01:24.549207 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_8568.solverstate
I0428 16:01:27.660322 29440 solver.cpp:330] Iteration 8568, Testing net (#0)
I0428 16:01:27.660400 29440 net.cpp:676] Ignoring source layer train-data
I0428 16:01:28.751946 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 16:01:32.525759 29440 solver.cpp:397]     Test net output #0: accuracy = 0.359681
I0428 16:01:32.525789 29440 solver.cpp:397]     Test net output #1: loss = 3.39827 (* 1 = 3.39827 loss)
I0428 16:01:32.765877 29440 solver.cpp:218] Iteration 8568 (0.747473 iter/s, 16.0541s/12 iters), loss = 0.733333
I0428 16:01:32.765924 29440 solver.cpp:237]     Train net output #0: loss = 0.733333 (* 1 = 0.733333 loss)
I0428 16:01:32.765934 29440 sgd_solver.cpp:105] Iteration 8568, lr = 0.00183196
I0428 16:01:37.804059 29440 solver.cpp:218] Iteration 8580 (2.38193 iter/s, 5.03792s/12 iters), loss = 0.718044
I0428 16:01:37.804112 29440 solver.cpp:237]     Train net output #0: loss = 0.718044 (* 1 = 0.718044 loss)
I0428 16:01:37.804126 29440 sgd_solver.cpp:105] Iteration 8580, lr = 0.00182761
I0428 16:01:43.173141 29440 solver.cpp:218] Iteration 8592 (2.23602 iter/s, 5.36669s/12 iters), loss = 0.739739
I0428 16:01:43.173179 29440 solver.cpp:237]     Train net output #0: loss = 0.739739 (* 1 = 0.739739 loss)
I0428 16:01:43.173188 29440 sgd_solver.cpp:105] Iteration 8592, lr = 0.00182327
I0428 16:01:45.475883 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 16:01:48.491555 29440 solver.cpp:218] Iteration 8604 (2.25642 iter/s, 5.31816s/12 iters), loss = 0.776243
I0428 16:01:48.491601 29440 solver.cpp:237]     Train net output #0: loss = 0.776243 (* 1 = 0.776243 loss)
I0428 16:01:48.491611 29440 sgd_solver.cpp:105] Iteration 8604, lr = 0.00181894
I0428 16:01:53.890400 29440 solver.cpp:218] Iteration 8616 (2.22281 iter/s, 5.39858s/12 iters), loss = 0.904793
I0428 16:01:53.890447 29440 solver.cpp:237]     Train net output #0: loss = 0.904793 (* 1 = 0.904793 loss)
I0428 16:01:53.890456 29440 sgd_solver.cpp:105] Iteration 8616, lr = 0.00181462
I0428 16:01:59.357214 29440 solver.cpp:218] Iteration 8628 (2.19517 iter/s, 5.46654s/12 iters), loss = 0.846294
I0428 16:01:59.357352 29440 solver.cpp:237]     Train net output #0: loss = 0.846294 (* 1 = 0.846294 loss)
I0428 16:01:59.357362 29440 sgd_solver.cpp:105] Iteration 8628, lr = 0.00181031
I0428 16:02:05.275532 29440 solver.cpp:218] Iteration 8640 (2.02773 iter/s, 5.91794s/12 iters), loss = 0.983144
I0428 16:02:05.275571 29440 solver.cpp:237]     Train net output #0: loss = 0.983144 (* 1 = 0.983144 loss)
I0428 16:02:05.275579 29440 sgd_solver.cpp:105] Iteration 8640, lr = 0.00180602
I0428 16:02:10.660861 29440 solver.cpp:218] Iteration 8652 (2.22838 iter/s, 5.38507s/12 iters), loss = 0.679865
I0428 16:02:10.660903 29440 solver.cpp:237]     Train net output #0: loss = 0.679865 (* 1 = 0.679865 loss)
I0428 16:02:10.660912 29440 sgd_solver.cpp:105] Iteration 8652, lr = 0.00180173
I0428 16:02:16.362102 29440 solver.cpp:218] Iteration 8664 (2.10491 iter/s, 5.70097s/12 iters), loss = 0.771529
I0428 16:02:16.362145 29440 solver.cpp:237]     Train net output #0: loss = 0.771529 (* 1 = 0.771529 loss)
I0428 16:02:16.362154 29440 sgd_solver.cpp:105] Iteration 8664, lr = 0.00179745
I0428 16:02:18.260753 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_8670.caffemodel
I0428 16:02:22.495652 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_8670.solverstate
I0428 16:02:25.604013 29440 solver.cpp:330] Iteration 8670, Testing net (#0)
I0428 16:02:25.604036 29440 net.cpp:676] Ignoring source layer train-data
I0428 16:02:26.676041 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 16:02:30.634611 29440 solver.cpp:397]     Test net output #0: accuracy = 0.363358
I0428 16:02:30.634703 29440 solver.cpp:397]     Test net output #1: loss = 3.41348 (* 1 = 3.41348 loss)
I0428 16:02:32.894402 29440 solver.cpp:218] Iteration 8676 (0.725978 iter/s, 16.5294s/12 iters), loss = 0.834931
I0428 16:02:32.894449 29440 solver.cpp:237]     Train net output #0: loss = 0.834931 (* 1 = 0.834931 loss)
I0428 16:02:32.894459 29440 sgd_solver.cpp:105] Iteration 8676, lr = 0.00179318
I0428 16:02:38.305014 29440 solver.cpp:218] Iteration 8688 (2.21797 iter/s, 5.41035s/12 iters), loss = 0.908333
I0428 16:02:38.305058 29440 solver.cpp:237]     Train net output #0: loss = 0.908333 (* 1 = 0.908333 loss)
I0428 16:02:38.305066 29440 sgd_solver.cpp:105] Iteration 8688, lr = 0.00178893
I0428 16:02:43.174876 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 16:02:44.163002 29440 solver.cpp:218] Iteration 8700 (2.04858 iter/s, 5.8577s/12 iters), loss = 0.647357
I0428 16:02:44.163054 29440 solver.cpp:237]     Train net output #0: loss = 0.647357 (* 1 = 0.647357 loss)
I0428 16:02:44.163066 29440 sgd_solver.cpp:105] Iteration 8700, lr = 0.00178468
I0428 16:02:49.029242 29440 solver.cpp:218] Iteration 8712 (2.46721 iter/s, 4.8638s/12 iters), loss = 0.759619
I0428 16:02:49.029294 29440 solver.cpp:237]     Train net output #0: loss = 0.759619 (* 1 = 0.759619 loss)
I0428 16:02:49.029306 29440 sgd_solver.cpp:105] Iteration 8712, lr = 0.00178044
I0428 16:02:54.882019 29440 solver.cpp:218] Iteration 8724 (2.05041 iter/s, 5.85249s/12 iters), loss = 0.91368
I0428 16:02:54.882058 29440 solver.cpp:237]     Train net output #0: loss = 0.91368 (* 1 = 0.91368 loss)
I0428 16:02:54.882067 29440 sgd_solver.cpp:105] Iteration 8724, lr = 0.00177621
I0428 16:03:00.179092 29440 solver.cpp:218] Iteration 8736 (2.26645 iter/s, 5.29462s/12 iters), loss = 0.62448
I0428 16:03:00.179132 29440 solver.cpp:237]     Train net output #0: loss = 0.62448 (* 1 = 0.62448 loss)
I0428 16:03:00.179141 29440 sgd_solver.cpp:105] Iteration 8736, lr = 0.001772
I0428 16:03:05.448987 29440 solver.cpp:218] Iteration 8748 (2.2772 iter/s, 5.26963s/12 iters), loss = 0.671449
I0428 16:03:05.449118 29440 solver.cpp:237]     Train net output #0: loss = 0.671449 (* 1 = 0.671449 loss)
I0428 16:03:05.449129 29440 sgd_solver.cpp:105] Iteration 8748, lr = 0.00176779
I0428 16:03:11.044147 29440 solver.cpp:218] Iteration 8760 (2.14485 iter/s, 5.5948s/12 iters), loss = 0.715239
I0428 16:03:11.044188 29440 solver.cpp:237]     Train net output #0: loss = 0.715239 (* 1 = 0.715239 loss)
I0428 16:03:11.044198 29440 sgd_solver.cpp:105] Iteration 8760, lr = 0.00176359
I0428 16:03:16.010358 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_8772.caffemodel
I0428 16:03:23.645202 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_8772.solverstate
I0428 16:03:28.060374 29440 solver.cpp:330] Iteration 8772, Testing net (#0)
I0428 16:03:28.060391 29440 net.cpp:676] Ignoring source layer train-data
I0428 16:03:29.071724 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 16:03:33.087707 29440 solver.cpp:397]     Test net output #0: accuracy = 0.349265
I0428 16:03:33.087739 29440 solver.cpp:397]     Test net output #1: loss = 3.43787 (* 1 = 3.43787 loss)
I0428 16:03:33.438324 29440 solver.cpp:218] Iteration 8772 (0.535875 iter/s, 22.3933s/12 iters), loss = 0.793457
I0428 16:03:33.439945 29440 solver.cpp:237]     Train net output #0: loss = 0.793457 (* 1 = 0.793457 loss)
I0428 16:03:33.439956 29440 sgd_solver.cpp:105] Iteration 8772, lr = 0.00175941
I0428 16:03:37.694959 29440 solver.cpp:218] Iteration 8784 (2.82032 iter/s, 4.25484s/12 iters), loss = 0.72356
I0428 16:03:37.695053 29440 solver.cpp:237]     Train net output #0: loss = 0.72356 (* 1 = 0.72356 loss)
I0428 16:03:37.695062 29440 sgd_solver.cpp:105] Iteration 8784, lr = 0.00175523
I0428 16:03:43.401690 29440 solver.cpp:218] Iteration 8796 (2.1029 iter/s, 5.7064s/12 iters), loss = 0.580154
I0428 16:03:43.401738 29440 solver.cpp:237]     Train net output #0: loss = 0.580154 (* 1 = 0.580154 loss)
I0428 16:03:43.401749 29440 sgd_solver.cpp:105] Iteration 8796, lr = 0.00175106
I0428 16:03:44.859970 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 16:03:48.933331 29440 solver.cpp:218] Iteration 8808 (2.16944 iter/s, 5.53137s/12 iters), loss = 0.92457
I0428 16:03:48.933375 29440 solver.cpp:237]     Train net output #0: loss = 0.92457 (* 1 = 0.92457 loss)
I0428 16:03:48.933384 29440 sgd_solver.cpp:105] Iteration 8808, lr = 0.0017469
I0428 16:03:54.025003 29440 solver.cpp:218] Iteration 8820 (2.35793 iter/s, 5.08921s/12 iters), loss = 0.869038
I0428 16:03:54.025048 29440 solver.cpp:237]     Train net output #0: loss = 0.869038 (* 1 = 0.869038 loss)
I0428 16:03:54.025058 29440 sgd_solver.cpp:105] Iteration 8820, lr = 0.00174276
I0428 16:03:59.680693 29440 solver.cpp:218] Iteration 8832 (2.12186 iter/s, 5.65541s/12 iters), loss = 0.553506
I0428 16:03:59.680737 29440 solver.cpp:237]     Train net output #0: loss = 0.553506 (* 1 = 0.553506 loss)
I0428 16:03:59.680747 29440 sgd_solver.cpp:105] Iteration 8832, lr = 0.00173862
I0428 16:04:04.971794 29440 solver.cpp:218] Iteration 8844 (2.26902 iter/s, 5.28862s/12 iters), loss = 0.870638
I0428 16:04:04.971844 29440 solver.cpp:237]     Train net output #0: loss = 0.870638 (* 1 = 0.870638 loss)
I0428 16:04:04.971853 29440 sgd_solver.cpp:105] Iteration 8844, lr = 0.00173449
I0428 16:04:10.776881 29440 solver.cpp:218] Iteration 8856 (2.06725 iter/s, 5.8048s/12 iters), loss = 0.57634
I0428 16:04:10.776978 29440 solver.cpp:237]     Train net output #0: loss = 0.57634 (* 1 = 0.57634 loss)
I0428 16:04:10.776990 29440 sgd_solver.cpp:105] Iteration 8856, lr = 0.00173037
I0428 16:04:16.194825 29440 solver.cpp:218] Iteration 8868 (2.21499 iter/s, 5.41763s/12 iters), loss = 0.755607
I0428 16:04:16.194871 29440 solver.cpp:237]     Train net output #0: loss = 0.755607 (* 1 = 0.755607 loss)
I0428 16:04:16.194880 29440 sgd_solver.cpp:105] Iteration 8868, lr = 0.00172626
I0428 16:04:18.407320 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_8874.caffemodel
I0428 16:04:26.060876 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_8874.solverstate
I0428 16:04:33.836838 29440 solver.cpp:330] Iteration 8874, Testing net (#0)
I0428 16:04:33.836865 29440 net.cpp:676] Ignoring source layer train-data
I0428 16:04:34.798724 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 16:04:38.802561 29440 solver.cpp:397]     Test net output #0: accuracy = 0.370098
I0428 16:04:38.802588 29440 solver.cpp:397]     Test net output #1: loss = 3.26485 (* 1 = 3.26485 loss)
I0428 16:04:41.116906 29440 solver.cpp:218] Iteration 8880 (0.48152 iter/s, 24.9211s/12 iters), loss = 0.598806
I0428 16:04:41.117045 29440 solver.cpp:237]     Train net output #0: loss = 0.598806 (* 1 = 0.598806 loss)
I0428 16:04:41.117056 29440 sgd_solver.cpp:105] Iteration 8880, lr = 0.00172217
I0428 16:04:46.392616 29440 solver.cpp:218] Iteration 8892 (2.27473 iter/s, 5.27536s/12 iters), loss = 0.791327
I0428 16:04:46.392655 29440 solver.cpp:237]     Train net output #0: loss = 0.791327 (* 1 = 0.791327 loss)
I0428 16:04:46.392663 29440 sgd_solver.cpp:105] Iteration 8892, lr = 0.00171808
I0428 16:04:50.188427 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 16:04:51.710408 29440 solver.cpp:218] Iteration 8904 (2.25669 iter/s, 5.31753s/12 iters), loss = 0.721285
I0428 16:04:51.710446 29440 solver.cpp:237]     Train net output #0: loss = 0.721285 (* 1 = 0.721285 loss)
I0428 16:04:51.710455 29440 sgd_solver.cpp:105] Iteration 8904, lr = 0.001714
I0428 16:04:57.406581 29440 solver.cpp:218] Iteration 8916 (2.10678 iter/s, 5.6959s/12 iters), loss = 0.659093
I0428 16:04:57.406635 29440 solver.cpp:237]     Train net output #0: loss = 0.659093 (* 1 = 0.659093 loss)
I0428 16:04:57.406646 29440 sgd_solver.cpp:105] Iteration 8916, lr = 0.00170993
I0428 16:05:02.796252 29440 solver.cpp:218] Iteration 8928 (2.22659 iter/s, 5.3894s/12 iters), loss = 0.711125
I0428 16:05:02.796295 29440 solver.cpp:237]     Train net output #0: loss = 0.711125 (* 1 = 0.711125 loss)
I0428 16:05:02.796303 29440 sgd_solver.cpp:105] Iteration 8928, lr = 0.00170587
I0428 16:05:08.207887 29440 solver.cpp:218] Iteration 8940 (2.21755 iter/s, 5.41137s/12 iters), loss = 0.673961
I0428 16:05:08.207942 29440 solver.cpp:237]     Train net output #0: loss = 0.673961 (* 1 = 0.673961 loss)
I0428 16:05:08.207952 29440 sgd_solver.cpp:105] Iteration 8940, lr = 0.00170182
I0428 16:05:13.472136 29440 solver.cpp:218] Iteration 8952 (2.27964 iter/s, 5.26398s/12 iters), loss = 0.473387
I0428 16:05:13.472270 29440 solver.cpp:237]     Train net output #0: loss = 0.473387 (* 1 = 0.473387 loss)
I0428 16:05:13.472281 29440 sgd_solver.cpp:105] Iteration 8952, lr = 0.00169778
I0428 16:05:18.765522 29440 solver.cpp:218] Iteration 8964 (2.26713 iter/s, 5.29304s/12 iters), loss = 0.561228
I0428 16:05:18.765563 29440 solver.cpp:237]     Train net output #0: loss = 0.561228 (* 1 = 0.561228 loss)
I0428 16:05:18.765570 29440 sgd_solver.cpp:105] Iteration 8964, lr = 0.00169375
I0428 16:05:23.772075 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_8976.caffemodel
I0428 16:05:26.458205 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_8976.solverstate
I0428 16:05:33.163975 29440 solver.cpp:330] Iteration 8976, Testing net (#0)
I0428 16:05:33.163996 29440 net.cpp:676] Ignoring source layer train-data
I0428 16:05:34.102203 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 16:05:38.029132 29440 solver.cpp:397]     Test net output #0: accuracy = 0.378064
I0428 16:05:38.029170 29440 solver.cpp:397]     Test net output #1: loss = 3.25091 (* 1 = 3.25091 loss)
I0428 16:05:38.171104 29440 solver.cpp:218] Iteration 8976 (0.618404 iter/s, 19.4048s/12 iters), loss = 0.758056
I0428 16:05:38.171151 29440 solver.cpp:237]     Train net output #0: loss = 0.758056 (* 1 = 0.758056 loss)
I0428 16:05:38.171164 29440 sgd_solver.cpp:105] Iteration 8976, lr = 0.00168973
I0428 16:05:43.276530 29440 solver.cpp:218] Iteration 8988 (2.35056 iter/s, 5.10516s/12 iters), loss = 0.744685
I0428 16:05:43.276574 29440 solver.cpp:237]     Train net output #0: loss = 0.744685 (* 1 = 0.744685 loss)
I0428 16:05:43.276583 29440 sgd_solver.cpp:105] Iteration 8988, lr = 0.00168571
I0428 16:05:48.745437 29440 solver.cpp:218] Iteration 9000 (2.19522 iter/s, 5.46643s/12 iters), loss = 0.561226
I0428 16:05:48.745590 29440 solver.cpp:237]     Train net output #0: loss = 0.561226 (* 1 = 0.561226 loss)
I0428 16:05:48.745604 29440 sgd_solver.cpp:105] Iteration 9000, lr = 0.00168171
I0428 16:05:49.451149 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 16:05:54.065563 29440 solver.cpp:218] Iteration 9012 (2.25574 iter/s, 5.31976s/12 iters), loss = 0.722425
I0428 16:05:54.065619 29440 solver.cpp:237]     Train net output #0: loss = 0.722425 (* 1 = 0.722425 loss)
I0428 16:05:54.065630 29440 sgd_solver.cpp:105] Iteration 9012, lr = 0.00167772
I0428 16:05:59.485620 29440 solver.cpp:218] Iteration 9024 (2.21411 iter/s, 5.41978s/12 iters), loss = 0.690618
I0428 16:05:59.485673 29440 solver.cpp:237]     Train net output #0: loss = 0.690618 (* 1 = 0.690618 loss)
I0428 16:05:59.485688 29440 sgd_solver.cpp:105] Iteration 9024, lr = 0.00167374
I0428 16:06:04.963477 29440 solver.cpp:218] Iteration 9036 (2.19075 iter/s, 5.47758s/12 iters), loss = 0.595002
I0428 16:06:04.963529 29440 solver.cpp:237]     Train net output #0: loss = 0.595002 (* 1 = 0.595002 loss)
I0428 16:06:04.963541 29440 sgd_solver.cpp:105] Iteration 9036, lr = 0.00166976
I0428 16:06:10.235126 29440 solver.cpp:218] Iteration 9048 (2.27644 iter/s, 5.27138s/12 iters), loss = 0.812039
I0428 16:06:10.235169 29440 solver.cpp:237]     Train net output #0: loss = 0.812039 (* 1 = 0.812039 loss)
I0428 16:06:10.235179 29440 sgd_solver.cpp:105] Iteration 9048, lr = 0.0016658
I0428 16:06:12.772692 29440 blocking_queue.cpp:49] Waiting for data
I0428 16:06:15.589395 29440 solver.cpp:218] Iteration 9060 (2.24223 iter/s, 5.35181s/12 iters), loss = 0.558131
I0428 16:06:15.589437 29440 solver.cpp:237]     Train net output #0: loss = 0.558131 (* 1 = 0.558131 loss)
I0428 16:06:15.589447 29440 sgd_solver.cpp:105] Iteration 9060, lr = 0.00166184
I0428 16:06:21.016988 29440 solver.cpp:218] Iteration 9072 (2.21103 iter/s, 5.42733s/12 iters), loss = 0.590123
I0428 16:06:21.018985 29440 solver.cpp:237]     Train net output #0: loss = 0.590123 (* 1 = 0.590123 loss)
I0428 16:06:21.018997 29440 sgd_solver.cpp:105] Iteration 9072, lr = 0.0016579
I0428 16:06:22.975085 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_9078.caffemodel
I0428 16:06:25.196708 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_9078.solverstate
I0428 16:06:27.222709 29440 solver.cpp:330] Iteration 9078, Testing net (#0)
I0428 16:06:27.222736 29440 net.cpp:676] Ignoring source layer train-data
I0428 16:06:28.086280 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 16:06:32.292205 29440 solver.cpp:397]     Test net output #0: accuracy = 0.384804
I0428 16:06:32.292240 29440 solver.cpp:397]     Test net output #1: loss = 3.24803 (* 1 = 3.24803 loss)
I0428 16:06:34.422880 29440 solver.cpp:218] Iteration 9084 (0.89531 iter/s, 13.4032s/12 iters), loss = 0.66985
I0428 16:06:34.422924 29440 solver.cpp:237]     Train net output #0: loss = 0.66985 (* 1 = 0.66985 loss)
I0428 16:06:34.422933 29440 sgd_solver.cpp:105] Iteration 9084, lr = 0.00165396
I0428 16:06:39.918148 29440 solver.cpp:218] Iteration 9096 (2.1838 iter/s, 5.495s/12 iters), loss = 0.659194
I0428 16:06:39.918193 29440 solver.cpp:237]     Train net output #0: loss = 0.659194 (* 1 = 0.659194 loss)
I0428 16:06:39.918202 29440 sgd_solver.cpp:105] Iteration 9096, lr = 0.00165003
I0428 16:06:43.305986 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 16:06:45.743050 29440 solver.cpp:218] Iteration 9108 (2.06022 iter/s, 5.82462s/12 iters), loss = 0.523387
I0428 16:06:45.743090 29440 solver.cpp:237]     Train net output #0: loss = 0.523387 (* 1 = 0.523387 loss)
I0428 16:06:45.743099 29440 sgd_solver.cpp:105] Iteration 9108, lr = 0.00164612
I0428 16:06:51.118602 29440 solver.cpp:218] Iteration 9120 (2.23244 iter/s, 5.37529s/12 iters), loss = 0.484876
I0428 16:06:51.118750 29440 solver.cpp:237]     Train net output #0: loss = 0.484876 (* 1 = 0.484876 loss)
I0428 16:06:51.118760 29440 sgd_solver.cpp:105] Iteration 9120, lr = 0.00164221
I0428 16:06:56.316861 29440 solver.cpp:218] Iteration 9132 (2.30862 iter/s, 5.1979s/12 iters), loss = 0.695
I0428 16:06:56.316902 29440 solver.cpp:237]     Train net output #0: loss = 0.695 (* 1 = 0.695 loss)
I0428 16:06:56.316910 29440 sgd_solver.cpp:105] Iteration 9132, lr = 0.00163831
I0428 16:07:01.620648 29440 solver.cpp:218] Iteration 9144 (2.26264 iter/s, 5.30353s/12 iters), loss = 0.54619
I0428 16:07:01.620697 29440 solver.cpp:237]     Train net output #0: loss = 0.54619 (* 1 = 0.54619 loss)
I0428 16:07:01.620707 29440 sgd_solver.cpp:105] Iteration 9144, lr = 0.00163442
I0428 16:07:07.077936 29440 solver.cpp:218] Iteration 9156 (2.199 iter/s, 5.45702s/12 iters), loss = 0.544047
I0428 16:07:07.077981 29440 solver.cpp:237]     Train net output #0: loss = 0.544047 (* 1 = 0.544047 loss)
I0428 16:07:07.077991 29440 sgd_solver.cpp:105] Iteration 9156, lr = 0.00163054
I0428 16:07:12.459401 29440 solver.cpp:218] Iteration 9168 (2.22999 iter/s, 5.3812s/12 iters), loss = 0.466621
I0428 16:07:12.459448 29440 solver.cpp:237]     Train net output #0: loss = 0.466621 (* 1 = 0.466621 loss)
I0428 16:07:12.459457 29440 sgd_solver.cpp:105] Iteration 9168, lr = 0.00162667
I0428 16:07:17.213933 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_9180.caffemodel
I0428 16:07:19.233469 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_9180.solverstate
I0428 16:07:21.415706 29440 solver.cpp:330] Iteration 9180, Testing net (#0)
I0428 16:07:21.415789 29440 net.cpp:676] Ignoring source layer train-data
I0428 16:07:22.300242 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 16:07:26.610294 29440 solver.cpp:397]     Test net output #0: accuracy = 0.400735
I0428 16:07:26.610325 29440 solver.cpp:397]     Test net output #1: loss = 3.3434 (* 1 = 3.3434 loss)
I0428 16:07:26.757448 29440 solver.cpp:218] Iteration 9180 (0.83931 iter/s, 14.2975s/12 iters), loss = 0.476561
I0428 16:07:26.757501 29440 solver.cpp:237]     Train net output #0: loss = 0.476561 (* 1 = 0.476561 loss)
I0428 16:07:26.757512 29440 sgd_solver.cpp:105] Iteration 9180, lr = 0.00162281
I0428 16:07:31.553220 29440 solver.cpp:218] Iteration 9192 (2.50234 iter/s, 4.79552s/12 iters), loss = 0.555023
I0428 16:07:31.553277 29440 solver.cpp:237]     Train net output #0: loss = 0.555023 (* 1 = 0.555023 loss)
I0428 16:07:31.553288 29440 sgd_solver.cpp:105] Iteration 9192, lr = 0.00161895
I0428 16:07:36.753234 29440 solver.cpp:218] Iteration 9204 (2.30781 iter/s, 5.19973s/12 iters), loss = 0.572327
I0428 16:07:36.753275 29440 solver.cpp:237]     Train net output #0: loss = 0.572327 (* 1 = 0.572327 loss)
I0428 16:07:36.753284 29440 sgd_solver.cpp:105] Iteration 9204, lr = 0.00161511
I0428 16:07:36.822186 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 16:07:42.356820 29440 solver.cpp:218] Iteration 9216 (2.14159 iter/s, 5.60332s/12 iters), loss = 0.452807
I0428 16:07:42.356868 29440 solver.cpp:237]     Train net output #0: loss = 0.452807 (* 1 = 0.452807 loss)
I0428 16:07:42.356878 29440 sgd_solver.cpp:105] Iteration 9216, lr = 0.00161128
I0428 16:07:47.746989 29440 solver.cpp:218] Iteration 9228 (2.22729 iter/s, 5.38771s/12 iters), loss = 0.88245
I0428 16:07:47.747031 29440 solver.cpp:237]     Train net output #0: loss = 0.88245 (* 1 = 0.88245 loss)
I0428 16:07:47.747043 29440 sgd_solver.cpp:105] Iteration 9228, lr = 0.00160745
I0428 16:07:52.950609 29440 solver.cpp:218] Iteration 9240 (2.3072 iter/s, 5.20111s/12 iters), loss = 0.455019
I0428 16:07:52.950752 29440 solver.cpp:237]     Train net output #0: loss = 0.455019 (* 1 = 0.455019 loss)
I0428 16:07:52.950762 29440 sgd_solver.cpp:105] Iteration 9240, lr = 0.00160363
I0428 16:07:58.404278 29440 solver.cpp:218] Iteration 9252 (2.2005 iter/s, 5.4533s/12 iters), loss = 0.59186
I0428 16:07:58.404340 29440 solver.cpp:237]     Train net output #0: loss = 0.59186 (* 1 = 0.59186 loss)
I0428 16:07:58.404353 29440 sgd_solver.cpp:105] Iteration 9252, lr = 0.00159983
I0428 16:08:03.647503 29440 solver.cpp:218] Iteration 9264 (2.28879 iter/s, 5.24295s/12 iters), loss = 0.550139
I0428 16:08:03.647545 29440 solver.cpp:237]     Train net output #0: loss = 0.550139 (* 1 = 0.550139 loss)
I0428 16:08:03.647554 29440 sgd_solver.cpp:105] Iteration 9264, lr = 0.00159603
I0428 16:08:09.152344 29440 solver.cpp:218] Iteration 9276 (2.18001 iter/s, 5.50457s/12 iters), loss = 0.728118
I0428 16:08:09.152388 29440 solver.cpp:237]     Train net output #0: loss = 0.728118 (* 1 = 0.728118 loss)
I0428 16:08:09.152397 29440 sgd_solver.cpp:105] Iteration 9276, lr = 0.00159224
I0428 16:08:11.111874 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_9282.caffemodel
I0428 16:08:14.390360 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_9282.solverstate
I0428 16:08:22.112105 29440 solver.cpp:330] Iteration 9282, Testing net (#0)
I0428 16:08:22.112128 29440 net.cpp:676] Ignoring source layer train-data
I0428 16:08:22.850961 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 16:08:26.824286 29440 solver.cpp:397]     Test net output #0: accuracy = 0.400123
I0428 16:08:26.825047 29440 solver.cpp:397]     Test net output #1: loss = 3.36574 (* 1 = 3.36574 loss)
I0428 16:08:29.331187 29440 solver.cpp:218] Iteration 9288 (0.594706 iter/s, 20.178s/12 iters), loss = 0.601469
I0428 16:08:29.331235 29440 solver.cpp:237]     Train net output #0: loss = 0.601469 (* 1 = 0.601469 loss)
I0428 16:08:29.331244 29440 sgd_solver.cpp:105] Iteration 9288, lr = 0.00158846
I0428 16:08:35.222136 29440 solver.cpp:218] Iteration 9300 (2.03712 iter/s, 5.89066s/12 iters), loss = 0.698297
I0428 16:08:35.222184 29440 solver.cpp:237]     Train net output #0: loss = 0.698297 (* 1 = 0.698297 loss)
I0428 16:08:35.222193 29440 sgd_solver.cpp:105] Iteration 9300, lr = 0.00158469
I0428 16:08:37.489022 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 16:08:40.474658 29440 solver.cpp:218] Iteration 9312 (2.28473 iter/s, 5.25226s/12 iters), loss = 0.382666
I0428 16:08:40.474701 29440 solver.cpp:237]     Train net output #0: loss = 0.382666 (* 1 = 0.382666 loss)
I0428 16:08:40.474710 29440 sgd_solver.cpp:105] Iteration 9312, lr = 0.00158092
I0428 16:08:46.089587 29440 solver.cpp:218] Iteration 9324 (2.13726 iter/s, 5.61465s/12 iters), loss = 0.745257
I0428 16:08:46.089632 29440 solver.cpp:237]     Train net output #0: loss = 0.745257 (* 1 = 0.745257 loss)
I0428 16:08:46.089643 29440 sgd_solver.cpp:105] Iteration 9324, lr = 0.00157717
I0428 16:08:51.277886 29440 solver.cpp:218] Iteration 9336 (2.31399 iter/s, 5.18586s/12 iters), loss = 1.07714
I0428 16:08:51.277933 29440 solver.cpp:237]     Train net output #0: loss = 1.07714 (* 1 = 1.07714 loss)
I0428 16:08:51.277942 29440 sgd_solver.cpp:105] Iteration 9336, lr = 0.00157343
I0428 16:08:56.761334 29440 solver.cpp:218] Iteration 9348 (2.18852 iter/s, 5.48315s/12 iters), loss = 0.627287
I0428 16:08:56.761394 29440 solver.cpp:237]     Train net output #0: loss = 0.627287 (* 1 = 0.627287 loss)
I0428 16:08:56.761405 29440 sgd_solver.cpp:105] Iteration 9348, lr = 0.00156969
I0428 16:09:02.115177 29440 solver.cpp:218] Iteration 9360 (2.2415 iter/s, 5.35357s/12 iters), loss = 0.518798
I0428 16:09:02.115288 29440 solver.cpp:237]     Train net output #0: loss = 0.518798 (* 1 = 0.518798 loss)
I0428 16:09:02.115300 29440 sgd_solver.cpp:105] Iteration 9360, lr = 0.00156596
I0428 16:09:07.478322 29440 solver.cpp:218] Iteration 9372 (2.23851 iter/s, 5.3607s/12 iters), loss = 0.604468
I0428 16:09:07.478363 29440 solver.cpp:237]     Train net output #0: loss = 0.604468 (* 1 = 0.604468 loss)
I0428 16:09:07.478371 29440 sgd_solver.cpp:105] Iteration 9372, lr = 0.00156225
I0428 16:09:12.290762 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_9384.caffemodel
I0428 16:09:13.765873 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_9384.solverstate
I0428 16:09:14.831856 29440 solver.cpp:330] Iteration 9384, Testing net (#0)
I0428 16:09:14.831881 29440 net.cpp:676] Ignoring source layer train-data
I0428 16:09:15.641718 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 16:09:19.969177 29440 solver.cpp:397]     Test net output #0: accuracy = 0.387255
I0428 16:09:19.969213 29440 solver.cpp:397]     Test net output #1: loss = 3.41396 (* 1 = 3.41396 loss)
I0428 16:09:20.128055 29440 solver.cpp:218] Iteration 9384 (0.948843 iter/s, 12.647s/12 iters), loss = 0.571366
I0428 16:09:20.129577 29440 solver.cpp:237]     Train net output #0: loss = 0.571366 (* 1 = 0.571366 loss)
I0428 16:09:20.129596 29440 sgd_solver.cpp:105] Iteration 9384, lr = 0.00155854
I0428 16:09:25.187316 29440 solver.cpp:218] Iteration 9396 (2.37269 iter/s, 5.05754s/12 iters), loss = 0.591712
I0428 16:09:25.187357 29440 solver.cpp:237]     Train net output #0: loss = 0.591712 (* 1 = 0.591712 loss)
I0428 16:09:25.187366 29440 sgd_solver.cpp:105] Iteration 9396, lr = 0.00155484
I0428 16:09:29.519440 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 16:09:30.524957 29440 solver.cpp:218] Iteration 9408 (2.24923 iter/s, 5.33516s/12 iters), loss = 0.430265
I0428 16:09:30.525003 29440 solver.cpp:237]     Train net output #0: loss = 0.430265 (* 1 = 0.430265 loss)
I0428 16:09:30.525012 29440 sgd_solver.cpp:105] Iteration 9408, lr = 0.00155114
I0428 16:09:35.559482 29440 solver.cpp:218] Iteration 9420 (2.3847 iter/s, 5.03208s/12 iters), loss = 0.46421
I0428 16:09:35.559643 29440 solver.cpp:237]     Train net output #0: loss = 0.46421 (* 1 = 0.46421 loss)
I0428 16:09:35.559655 29440 sgd_solver.cpp:105] Iteration 9420, lr = 0.00154746
I0428 16:09:40.863837 29440 solver.cpp:218] Iteration 9432 (2.26245 iter/s, 5.30398s/12 iters), loss = 0.880232
I0428 16:09:40.863876 29440 solver.cpp:237]     Train net output #0: loss = 0.880232 (* 1 = 0.880232 loss)
I0428 16:09:40.863885 29440 sgd_solver.cpp:105] Iteration 9432, lr = 0.00154379
I0428 16:09:46.714572 29440 solver.cpp:218] Iteration 9444 (2.05112 iter/s, 5.85046s/12 iters), loss = 0.435653
I0428 16:09:46.714615 29440 solver.cpp:237]     Train net output #0: loss = 0.435653 (* 1 = 0.435653 loss)
I0428 16:09:46.714625 29440 sgd_solver.cpp:105] Iteration 9444, lr = 0.00154012
I0428 16:09:52.230813 29440 solver.cpp:218] Iteration 9456 (2.17637 iter/s, 5.51377s/12 iters), loss = 0.748119
I0428 16:09:52.230852 29440 solver.cpp:237]     Train net output #0: loss = 0.748119 (* 1 = 0.748119 loss)
I0428 16:09:52.230861 29440 sgd_solver.cpp:105] Iteration 9456, lr = 0.00153647
I0428 16:09:57.618980 29440 solver.cpp:218] Iteration 9468 (2.22813 iter/s, 5.38569s/12 iters), loss = 0.408566
I0428 16:09:57.619032 29440 solver.cpp:237]     Train net output #0: loss = 0.408566 (* 1 = 0.408566 loss)
I0428 16:09:57.619043 29440 sgd_solver.cpp:105] Iteration 9468, lr = 0.00153282
I0428 16:10:03.102725 29440 solver.cpp:218] Iteration 9480 (2.1884 iter/s, 5.48347s/12 iters), loss = 0.367626
I0428 16:10:03.102768 29440 solver.cpp:237]     Train net output #0: loss = 0.367626 (* 1 = 0.367626 loss)
I0428 16:10:03.102777 29440 sgd_solver.cpp:105] Iteration 9480, lr = 0.00152918
I0428 16:10:05.389336 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_9486.caffemodel
I0428 16:10:08.010892 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_9486.solverstate
I0428 16:10:09.112597 29440 solver.cpp:330] Iteration 9486, Testing net (#0)
I0428 16:10:09.112618 29440 net.cpp:676] Ignoring source layer train-data
I0428 16:10:09.817477 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 16:10:14.019403 29440 solver.cpp:397]     Test net output #0: accuracy = 0.395221
I0428 16:10:14.019430 29440 solver.cpp:397]     Test net output #1: loss = 3.43668 (* 1 = 3.43668 loss)
I0428 16:10:16.356937 29440 solver.cpp:218] Iteration 9492 (0.90541 iter/s, 13.2537s/12 iters), loss = 0.645804
I0428 16:10:16.356982 29440 solver.cpp:237]     Train net output #0: loss = 0.645804 (* 1 = 0.645804 loss)
I0428 16:10:16.356990 29440 sgd_solver.cpp:105] Iteration 9492, lr = 0.00152555
I0428 16:10:21.677317 29440 solver.cpp:218] Iteration 9504 (2.25559 iter/s, 5.32012s/12 iters), loss = 0.676754
I0428 16:10:21.677359 29440 solver.cpp:237]     Train net output #0: loss = 0.676754 (* 1 = 0.676754 loss)
I0428 16:10:21.677368 29440 sgd_solver.cpp:105] Iteration 9504, lr = 0.00152193
I0428 16:10:23.207826 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 16:10:27.023583 29440 solver.cpp:218] Iteration 9516 (2.24466 iter/s, 5.34601s/12 iters), loss = 0.60607
I0428 16:10:27.023625 29440 solver.cpp:237]     Train net output #0: loss = 0.60607 (* 1 = 0.60607 loss)
I0428 16:10:27.023634 29440 sgd_solver.cpp:105] Iteration 9516, lr = 0.00151831
I0428 16:10:32.148629 29440 solver.cpp:218] Iteration 9528 (2.34255 iter/s, 5.12263s/12 iters), loss = 0.671743
I0428 16:10:32.148669 29440 solver.cpp:237]     Train net output #0: loss = 0.671743 (* 1 = 0.671743 loss)
I0428 16:10:32.148679 29440 sgd_solver.cpp:105] Iteration 9528, lr = 0.00151471
I0428 16:10:37.912704 29440 solver.cpp:218] Iteration 9540 (2.08196 iter/s, 5.7638s/12 iters), loss = 0.528653
I0428 16:10:37.912752 29440 solver.cpp:237]     Train net output #0: loss = 0.528653 (* 1 = 0.528653 loss)
I0428 16:10:37.912760 29440 sgd_solver.cpp:105] Iteration 9540, lr = 0.00151111
I0428 16:10:43.223098 29440 solver.cpp:218] Iteration 9552 (2.25983 iter/s, 5.31013s/12 iters), loss = 0.501785
I0428 16:10:43.223238 29440 solver.cpp:237]     Train net output #0: loss = 0.501785 (* 1 = 0.501785 loss)
I0428 16:10:43.223249 29440 sgd_solver.cpp:105] Iteration 9552, lr = 0.00150752
I0428 16:10:48.839488 29440 solver.cpp:218] Iteration 9564 (2.13754 iter/s, 5.61392s/12 iters), loss = 0.461077
I0428 16:10:48.839531 29440 solver.cpp:237]     Train net output #0: loss = 0.461077 (* 1 = 0.461077 loss)
I0428 16:10:48.839540 29440 sgd_solver.cpp:105] Iteration 9564, lr = 0.00150395
I0428 16:10:54.162248 29440 solver.cpp:218] Iteration 9576 (2.25458 iter/s, 5.3225s/12 iters), loss = 0.410976
I0428 16:10:54.162292 29440 solver.cpp:237]     Train net output #0: loss = 0.410976 (* 1 = 0.410976 loss)
I0428 16:10:54.162300 29440 sgd_solver.cpp:105] Iteration 9576, lr = 0.00150037
I0428 16:10:58.546087 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_9588.caffemodel
I0428 16:11:05.598877 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_9588.solverstate
I0428 16:11:07.287992 29440 solver.cpp:330] Iteration 9588, Testing net (#0)
I0428 16:11:07.288014 29440 net.cpp:676] Ignoring source layer train-data
I0428 16:11:07.954241 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 16:11:14.994277 29440 solver.cpp:397]     Test net output #0: accuracy = 0.395833
I0428 16:11:15.020567 29440 solver.cpp:397]     Test net output #1: loss = 3.40241 (* 1 = 3.40241 loss)
I0428 16:11:15.402267 29440 solver.cpp:218] Iteration 9588 (0.565053 iter/s, 21.237s/12 iters), loss = 0.344215
I0428 16:11:15.403915 29440 solver.cpp:237]     Train net output #0: loss = 0.344215 (* 1 = 0.344215 loss)
I0428 16:11:15.403931 29440 sgd_solver.cpp:105] Iteration 9588, lr = 0.00149681
I0428 16:11:23.600425 29440 solver.cpp:218] Iteration 9600 (1.4641 iter/s, 8.19619s/12 iters), loss = 0.4297
I0428 16:11:23.600632 29440 solver.cpp:237]     Train net output #0: loss = 0.4297 (* 1 = 0.4297 loss)
I0428 16:11:23.600648 29440 sgd_solver.cpp:105] Iteration 9600, lr = 0.00149326
I0428 16:11:29.982892 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 16:11:32.328572 29440 solver.cpp:218] Iteration 9612 (1.37532 iter/s, 8.72523s/12 iters), loss = 0.551347
I0428 16:11:32.328632 29440 solver.cpp:237]     Train net output #0: loss = 0.551347 (* 1 = 0.551347 loss)
I0428 16:11:32.328645 29440 sgd_solver.cpp:105] Iteration 9612, lr = 0.00148971
I0428 16:11:40.149283 29440 solver.cpp:218] Iteration 9624 (1.53446 iter/s, 7.82033s/12 iters), loss = 0.514613
I0428 16:11:40.149344 29440 solver.cpp:237]     Train net output #0: loss = 0.514613 (* 1 = 0.514613 loss)
I0428 16:11:40.149356 29440 sgd_solver.cpp:105] Iteration 9624, lr = 0.00148618
I0428 16:11:47.778169 29440 solver.cpp:218] Iteration 9636 (1.57304 iter/s, 7.62852s/12 iters), loss = 0.490601
I0428 16:11:47.779577 29440 solver.cpp:237]     Train net output #0: loss = 0.490601 (* 1 = 0.490601 loss)
I0428 16:11:47.779592 29440 sgd_solver.cpp:105] Iteration 9636, lr = 0.00148265
I0428 16:11:54.238426 29440 solver.cpp:218] Iteration 9648 (1.85823 iter/s, 6.45776s/12 iters), loss = 0.48176
I0428 16:11:54.238471 29440 solver.cpp:237]     Train net output #0: loss = 0.48176 (* 1 = 0.48176 loss)
I0428 16:11:54.238481 29440 sgd_solver.cpp:105] Iteration 9648, lr = 0.00147913
I0428 16:12:01.499315 29440 solver.cpp:218] Iteration 9660 (1.65277 iter/s, 7.26055s/12 iters), loss = 0.571071
I0428 16:12:01.499370 29440 solver.cpp:237]     Train net output #0: loss = 0.571071 (* 1 = 0.571071 loss)
I0428 16:12:01.499382 29440 sgd_solver.cpp:105] Iteration 9660, lr = 0.00147562
I0428 16:12:08.143635 29440 solver.cpp:218] Iteration 9672 (1.80614 iter/s, 6.64399s/12 iters), loss = 0.321952
I0428 16:12:08.143694 29440 solver.cpp:237]     Train net output #0: loss = 0.321952 (* 1 = 0.321952 loss)
I0428 16:12:08.143707 29440 sgd_solver.cpp:105] Iteration 9672, lr = 0.00147211
I0428 16:12:15.306919 29440 solver.cpp:218] Iteration 9684 (1.67529 iter/s, 7.16294s/12 iters), loss = 0.461148
I0428 16:12:15.306975 29440 solver.cpp:237]     Train net output #0: loss = 0.461148 (* 1 = 0.461148 loss)
I0428 16:12:15.306988 29440 sgd_solver.cpp:105] Iteration 9684, lr = 0.00146862
I0428 16:12:17.706727 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_9690.caffemodel
I0428 16:12:20.302603 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_9690.solverstate
I0428 16:12:24.997548 29440 solver.cpp:330] Iteration 9690, Testing net (#0)
I0428 16:12:24.997573 29440 net.cpp:676] Ignoring source layer train-data
I0428 16:12:25.875885 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 16:12:31.625404 29440 solver.cpp:397]     Test net output #0: accuracy = 0.38848
I0428 16:12:31.625442 29440 solver.cpp:397]     Test net output #1: loss = 3.47715 (* 1 = 3.47715 loss)
I0428 16:12:34.309613 29440 solver.cpp:218] Iteration 9696 (0.631589 iter/s, 18.9997s/12 iters), loss = 0.580254
I0428 16:12:34.309667 29440 solver.cpp:237]     Train net output #0: loss = 0.580254 (* 1 = 0.580254 loss)
I0428 16:12:34.309679 29440 sgd_solver.cpp:105] Iteration 9696, lr = 0.00146513
I0428 16:12:41.543184 29440 solver.cpp:218] Iteration 9708 (1.65901 iter/s, 7.23322s/12 iters), loss = 0.358023
I0428 16:12:41.543241 29440 solver.cpp:237]     Train net output #0: loss = 0.358023 (* 1 = 0.358023 loss)
I0428 16:12:41.543253 29440 sgd_solver.cpp:105] Iteration 9708, lr = 0.00146165
I0428 16:12:42.607061 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 16:12:48.087942 29440 solver.cpp:218] Iteration 9720 (1.83362 iter/s, 6.54444s/12 iters), loss = 0.355731
I0428 16:12:48.087983 29440 solver.cpp:237]     Train net output #0: loss = 0.355731 (* 1 = 0.355731 loss)
I0428 16:12:48.087991 29440 sgd_solver.cpp:105] Iteration 9720, lr = 0.00145818
I0428 16:12:53.373790 29440 solver.cpp:218] Iteration 9732 (2.27127 iter/s, 5.28338s/12 iters), loss = 0.522413
I0428 16:12:53.373919 29440 solver.cpp:237]     Train net output #0: loss = 0.522413 (* 1 = 0.522413 loss)
I0428 16:12:53.373929 29440 sgd_solver.cpp:105] Iteration 9732, lr = 0.00145472
I0428 16:12:58.803174 29440 solver.cpp:218] Iteration 9744 (2.21034 iter/s, 5.42904s/12 iters), loss = 0.421209
I0428 16:12:58.803220 29440 solver.cpp:237]     Train net output #0: loss = 0.421209 (* 1 = 0.421209 loss)
I0428 16:12:58.803228 29440 sgd_solver.cpp:105] Iteration 9744, lr = 0.00145127
I0428 16:12:58.803445 29440 blocking_queue.cpp:49] Waiting for data
I0428 16:13:04.348800 29440 solver.cpp:218] Iteration 9756 (2.16397 iter/s, 5.54536s/12 iters), loss = 0.45722
I0428 16:13:04.348845 29440 solver.cpp:237]     Train net output #0: loss = 0.45722 (* 1 = 0.45722 loss)
I0428 16:13:04.348855 29440 sgd_solver.cpp:105] Iteration 9756, lr = 0.00144782
I0428 16:13:10.014465 29440 solver.cpp:218] Iteration 9768 (2.11812 iter/s, 5.66539s/12 iters), loss = 0.501325
I0428 16:13:10.014504 29440 solver.cpp:237]     Train net output #0: loss = 0.501325 (* 1 = 0.501325 loss)
I0428 16:13:10.014513 29440 sgd_solver.cpp:105] Iteration 9768, lr = 0.00144438
I0428 16:13:15.477210 29440 solver.cpp:218] Iteration 9780 (2.19769 iter/s, 5.46027s/12 iters), loss = 0.556945
I0428 16:13:15.477248 29440 solver.cpp:237]     Train net output #0: loss = 0.556945 (* 1 = 0.556945 loss)
I0428 16:13:15.477257 29440 sgd_solver.cpp:105] Iteration 9780, lr = 0.00144095
I0428 16:13:20.358050 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_9792.caffemodel
I0428 16:13:22.612193 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_9792.solverstate
I0428 16:13:23.680663 29440 solver.cpp:330] Iteration 9792, Testing net (#0)
I0428 16:13:23.680796 29440 net.cpp:676] Ignoring source layer train-data
I0428 16:13:24.235348 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 16:13:28.632616 29440 solver.cpp:397]     Test net output #0: accuracy = 0.387868
I0428 16:13:28.632647 29440 solver.cpp:397]     Test net output #1: loss = 3.61144 (* 1 = 3.61144 loss)
I0428 16:13:28.993175 29440 solver.cpp:218] Iteration 9792 (0.887876 iter/s, 13.5154s/12 iters), loss = 0.646513
I0428 16:13:28.994796 29440 solver.cpp:237]     Train net output #0: loss = 0.646513 (* 1 = 0.646513 loss)
I0428 16:13:28.994810 29440 sgd_solver.cpp:105] Iteration 9792, lr = 0.00143753
I0428 16:13:33.327648 29440 solver.cpp:218] Iteration 9804 (2.76965 iter/s, 4.33268s/12 iters), loss = 0.293359
I0428 16:13:33.327692 29440 solver.cpp:237]     Train net output #0: loss = 0.293359 (* 1 = 0.293359 loss)
I0428 16:13:33.327702 29440 sgd_solver.cpp:105] Iteration 9804, lr = 0.00143412
I0428 16:13:36.439083 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 16:13:38.908696 29440 solver.cpp:218] Iteration 9816 (2.15024 iter/s, 5.58078s/12 iters), loss = 0.241809
I0428 16:13:38.908735 29440 solver.cpp:237]     Train net output #0: loss = 0.241809 (* 1 = 0.241809 loss)
I0428 16:13:38.908742 29440 sgd_solver.cpp:105] Iteration 9816, lr = 0.00143072
I0428 16:13:44.447340 29440 solver.cpp:218] Iteration 9828 (2.1667 iter/s, 5.53838s/12 iters), loss = 0.380261
I0428 16:13:44.447383 29440 solver.cpp:237]     Train net output #0: loss = 0.380261 (* 1 = 0.380261 loss)
I0428 16:13:44.447393 29440 sgd_solver.cpp:105] Iteration 9828, lr = 0.00142732
I0428 16:13:49.614558 29440 solver.cpp:218] Iteration 9840 (2.32245 iter/s, 5.16696s/12 iters), loss = 0.451515
I0428 16:13:49.614599 29440 solver.cpp:237]     Train net output #0: loss = 0.451515 (* 1 = 0.451515 loss)
I0428 16:13:49.614609 29440 sgd_solver.cpp:105] Iteration 9840, lr = 0.00142393
I0428 16:13:54.963871 29440 solver.cpp:218] Iteration 9852 (2.24339 iter/s, 5.34906s/12 iters), loss = 0.568111
I0428 16:13:54.963976 29440 solver.cpp:237]     Train net output #0: loss = 0.568111 (* 1 = 0.568111 loss)
I0428 16:13:54.963985 29440 sgd_solver.cpp:105] Iteration 9852, lr = 0.00142055
I0428 16:14:00.448601 29440 solver.cpp:218] Iteration 9864 (2.18802 iter/s, 5.4844s/12 iters), loss = 0.412682
I0428 16:14:00.448642 29440 solver.cpp:237]     Train net output #0: loss = 0.412682 (* 1 = 0.412682 loss)
I0428 16:14:00.448652 29440 sgd_solver.cpp:105] Iteration 9864, lr = 0.00141718
I0428 16:14:05.993106 29440 solver.cpp:218] Iteration 9876 (2.16441 iter/s, 5.54424s/12 iters), loss = 0.441842
I0428 16:14:05.993156 29440 solver.cpp:237]     Train net output #0: loss = 0.441842 (* 1 = 0.441842 loss)
I0428 16:14:05.993165 29440 sgd_solver.cpp:105] Iteration 9876, lr = 0.00141381
I0428 16:14:11.477075 29440 solver.cpp:218] Iteration 9888 (2.18831 iter/s, 5.48369s/12 iters), loss = 0.485103
I0428 16:14:11.477131 29440 solver.cpp:237]     Train net output #0: loss = 0.485103 (* 1 = 0.485103 loss)
I0428 16:14:11.477145 29440 sgd_solver.cpp:105] Iteration 9888, lr = 0.00141045
I0428 16:14:13.493152 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_9894.caffemodel
I0428 16:14:15.922197 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_9894.solverstate
I0428 16:14:17.009232 29440 solver.cpp:330] Iteration 9894, Testing net (#0)
I0428 16:14:17.009255 29440 net.cpp:676] Ignoring source layer train-data
I0428 16:14:17.545539 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 16:14:21.840550 29440 solver.cpp:397]     Test net output #0: accuracy = 0.384804
I0428 16:14:21.840580 29440 solver.cpp:397]     Test net output #1: loss = 3.62325 (* 1 = 3.62325 loss)
I0428 16:14:24.438931 29440 solver.cpp:218] Iteration 9900 (0.925833 iter/s, 12.9613s/12 iters), loss = 0.562203
I0428 16:14:24.438987 29440 solver.cpp:237]     Train net output #0: loss = 0.562203 (* 1 = 0.562203 loss)
I0428 16:14:24.438998 29440 sgd_solver.cpp:105] Iteration 9900, lr = 0.00140711
I0428 16:14:29.829473 29440 solver.cpp:218] Iteration 9912 (2.22624 iter/s, 5.39026s/12 iters), loss = 0.375976
I0428 16:14:29.830996 29440 solver.cpp:237]     Train net output #0: loss = 0.375976 (* 1 = 0.375976 loss)
I0428 16:14:29.831008 29440 sgd_solver.cpp:105] Iteration 9912, lr = 0.00140377
I0428 16:14:29.940217 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 16:14:35.779479 29440 solver.cpp:218] Iteration 9924 (2.0174 iter/s, 5.94825s/12 iters), loss = 0.523446
I0428 16:14:35.779541 29440 solver.cpp:237]     Train net output #0: loss = 0.523446 (* 1 = 0.523446 loss)
I0428 16:14:35.779554 29440 sgd_solver.cpp:105] Iteration 9924, lr = 0.00140043
I0428 16:14:41.214498 29440 solver.cpp:218] Iteration 9936 (2.20802 iter/s, 5.43474s/12 iters), loss = 0.237904
I0428 16:14:41.214541 29440 solver.cpp:237]     Train net output #0: loss = 0.237904 (* 1 = 0.237904 loss)
I0428 16:14:41.214552 29440 sgd_solver.cpp:105] Iteration 9936, lr = 0.00139711
I0428 16:14:46.682482 29440 solver.cpp:218] Iteration 9948 (2.19559 iter/s, 5.4655s/12 iters), loss = 0.367972
I0428 16:14:46.682538 29440 solver.cpp:237]     Train net output #0: loss = 0.367972 (* 1 = 0.367972 loss)
I0428 16:14:46.682548 29440 sgd_solver.cpp:105] Iteration 9948, lr = 0.00139379
I0428 16:14:52.370959 29440 solver.cpp:218] Iteration 9960 (2.10963 iter/s, 5.68819s/12 iters), loss = 0.48538
I0428 16:14:52.371001 29440 solver.cpp:237]     Train net output #0: loss = 0.48538 (* 1 = 0.48538 loss)
I0428 16:14:52.371009 29440 sgd_solver.cpp:105] Iteration 9960, lr = 0.00139048
I0428 16:14:58.163854 29440 solver.cpp:218] Iteration 9972 (2.0716 iter/s, 5.79262s/12 iters), loss = 0.328753
I0428 16:14:58.163893 29440 solver.cpp:237]     Train net output #0: loss = 0.328753 (* 1 = 0.328753 loss)
I0428 16:14:58.163904 29440 sgd_solver.cpp:105] Iteration 9972, lr = 0.00138718
I0428 16:15:03.476796 29440 solver.cpp:218] Iteration 9984 (2.25968 iter/s, 5.31048s/12 iters), loss = 0.428215
I0428 16:15:03.476910 29440 solver.cpp:237]     Train net output #0: loss = 0.428215 (* 1 = 0.428215 loss)
I0428 16:15:03.476919 29440 sgd_solver.cpp:105] Iteration 9984, lr = 0.00138389
I0428 16:15:08.459105 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_9996.caffemodel
I0428 16:15:09.887904 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_9996.solverstate
I0428 16:15:10.963244 29440 solver.cpp:330] Iteration 9996, Testing net (#0)
I0428 16:15:10.963264 29440 net.cpp:676] Ignoring source layer train-data
I0428 16:15:11.468924 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 16:15:15.849758 29440 solver.cpp:397]     Test net output #0: accuracy = 0.394608
I0428 16:15:15.849786 29440 solver.cpp:397]     Test net output #1: loss = 3.54174 (* 1 = 3.54174 loss)
I0428 16:15:16.176147 29440 solver.cpp:218] Iteration 9996 (0.944975 iter/s, 12.6987s/12 iters), loss = 0.404813
I0428 16:15:16.177799 29440 solver.cpp:237]     Train net output #0: loss = 0.404813 (* 1 = 0.404813 loss)
I0428 16:15:16.177810 29440 sgd_solver.cpp:105] Iteration 9996, lr = 0.0013806
I0428 16:15:21.037305 29440 solver.cpp:218] Iteration 10008 (2.46949 iter/s, 4.85931s/12 iters), loss = 0.443772
I0428 16:15:21.037351 29440 solver.cpp:237]     Train net output #0: loss = 0.443772 (* 1 = 0.443772 loss)
I0428 16:15:21.037361 29440 sgd_solver.cpp:105] Iteration 10008, lr = 0.00137732
I0428 16:15:23.030406 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 16:15:26.340977 29440 solver.cpp:218] Iteration 10020 (2.26363 iter/s, 5.30122s/12 iters), loss = 0.296928
I0428 16:15:26.341020 29440 solver.cpp:237]     Train net output #0: loss = 0.296928 (* 1 = 0.296928 loss)
I0428 16:15:26.341029 29440 sgd_solver.cpp:105] Iteration 10020, lr = 0.00137405
I0428 16:15:31.888942 29440 solver.cpp:218] Iteration 10032 (2.16392 iter/s, 5.5455s/12 iters), loss = 0.385309
I0428 16:15:31.888983 29440 solver.cpp:237]     Train net output #0: loss = 0.385309 (* 1 = 0.385309 loss)
I0428 16:15:31.888991 29440 sgd_solver.cpp:105] Iteration 10032, lr = 0.00137079
I0428 16:15:37.272855 29440 solver.cpp:218] Iteration 10044 (2.22989 iter/s, 5.38144s/12 iters), loss = 0.406743
I0428 16:15:37.274648 29440 solver.cpp:237]     Train net output #0: loss = 0.406743 (* 1 = 0.406743 loss)
I0428 16:15:37.274665 29440 sgd_solver.cpp:105] Iteration 10044, lr = 0.00136754
I0428 16:15:42.956212 29440 solver.cpp:218] Iteration 10056 (2.11217 iter/s, 5.68135s/12 iters), loss = 0.755374
I0428 16:15:42.956252 29440 solver.cpp:237]     Train net output #0: loss = 0.755374 (* 1 = 0.755374 loss)
I0428 16:15:42.956261 29440 sgd_solver.cpp:105] Iteration 10056, lr = 0.00136429
I0428 16:15:48.301972 29440 solver.cpp:218] Iteration 10068 (2.24581 iter/s, 5.34328s/12 iters), loss = 0.388262
I0428 16:15:48.302021 29440 solver.cpp:237]     Train net output #0: loss = 0.388262 (* 1 = 0.388262 loss)
I0428 16:15:48.302031 29440 sgd_solver.cpp:105] Iteration 10068, lr = 0.00136105
I0428 16:15:53.281659 29440 solver.cpp:218] Iteration 10080 (2.41098 iter/s, 4.97724s/12 iters), loss = 0.383427
I0428 16:15:53.281704 29440 solver.cpp:237]     Train net output #0: loss = 0.383427 (* 1 = 0.383427 loss)
I0428 16:15:53.281713 29440 sgd_solver.cpp:105] Iteration 10080, lr = 0.00135782
I0428 16:15:58.837549 29440 solver.cpp:218] Iteration 10092 (2.15998 iter/s, 5.55562s/12 iters), loss = 0.493419
I0428 16:15:58.837595 29440 solver.cpp:237]     Train net output #0: loss = 0.493419 (* 1 = 0.493419 loss)
I0428 16:15:58.837606 29440 sgd_solver.cpp:105] Iteration 10092, lr = 0.0013546
I0428 16:16:00.863129 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_10098.caffemodel
I0428 16:16:08.272894 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_10098.solverstate
I0428 16:16:10.885205 29440 solver.cpp:330] Iteration 10098, Testing net (#0)
I0428 16:16:10.885224 29440 net.cpp:676] Ignoring source layer train-data
I0428 16:16:11.344861 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 16:16:15.727747 29440 solver.cpp:397]     Test net output #0: accuracy = 0.39277
I0428 16:16:15.727782 29440 solver.cpp:397]     Test net output #1: loss = 3.62758 (* 1 = 3.62758 loss)
I0428 16:16:18.018039 29440 solver.cpp:218] Iteration 10104 (0.625661 iter/s, 19.1797s/12 iters), loss = 0.36959
I0428 16:16:18.018093 29440 solver.cpp:237]     Train net output #0: loss = 0.36959 (* 1 = 0.36959 loss)
I0428 16:16:18.018105 29440 sgd_solver.cpp:105] Iteration 10104, lr = 0.00135138
I0428 16:16:22.578518 29462 data_layer.cpp:73] Restarting data prefetching from start.
I0428 16:16:23.433126 29440 solver.cpp:218] Iteration 10116 (2.21614 iter/s, 5.41481s/12 iters), loss = 0.28516
I0428 16:16:23.433182 29440 solver.cpp:237]     Train net output #0: loss = 0.28516 (* 1 = 0.28516 loss)
I0428 16:16:23.433195 29440 sgd_solver.cpp:105] Iteration 10116, lr = 0.00134817
I0428 16:16:28.930997 29440 solver.cpp:218] Iteration 10128 (2.18343 iter/s, 5.49593s/12 iters), loss = 0.399156
I0428 16:16:28.931042 29440 solver.cpp:237]     Train net output #0: loss = 0.399156 (* 1 = 0.399156 loss)
I0428 16:16:28.931052 29440 sgd_solver.cpp:105] Iteration 10128, lr = 0.00134497
I0428 16:16:34.637199 29440 solver.cpp:218] Iteration 10140 (2.10308 iter/s, 5.70593s/12 iters), loss = 0.359811
I0428 16:16:34.637250 29440 solver.cpp:237]     Train net output #0: loss = 0.359811 (* 1 = 0.359811 loss)
I0428 16:16:34.637262 29440 sgd_solver.cpp:105] Iteration 10140, lr = 0.00134178
I0428 16:16:39.614187 29440 solver.cpp:218] Iteration 10152 (2.41122 iter/s, 4.97673s/12 iters), loss = 0.340898
I0428 16:16:39.614339 29440 solver.cpp:237]     Train net output #0: loss = 0.340898 (* 1 = 0.340898 loss)
I0428 16:16:39.614351 29440 sgd_solver.cpp:105] Iteration 10152, lr = 0.00133859
I0428 16:16:45.297369 29440 solver.cpp:218] Iteration 10164 (2.11163 iter/s, 5.68281s/12 iters), loss = 0.337708
I0428 16:16:45.297410 29440 solver.cpp:237]     Train net output #0: loss = 0.337708 (* 1 = 0.337708 loss)
I0428 16:16:45.297420 29440 sgd_solver.cpp:105] Iteration 10164, lr = 0.00133541
I0428 16:16:50.249455 29440 solver.cpp:218] Iteration 10176 (2.42442 iter/s, 4.94964s/12 iters), loss = 0.328213
I0428 16:16:50.249511 29440 solver.cpp:237]     Train net output #0: loss = 0.328213 (* 1 = 0.328213 loss)
I0428 16:16:50.249522 29440 sgd_solver.cpp:105] Iteration 10176, lr = 0.00133224
I0428 16:16:56.267200 29440 solver.cpp:218] Iteration 10188 (1.9942 iter/s, 6.01745s/12 iters), loss = 0.288204
I0428 16:16:56.267246 29440 solver.cpp:237]     Train net output #0: loss = 0.288204 (* 1 = 0.288204 loss)
I0428 16:16:56.267254 29440 sgd_solver.cpp:105] Iteration 10188, lr = 0.00132908
I0428 16:17:02.194170 29440 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_10200.caffemodel
I0428 16:17:05.477823 29440 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_10200.solverstate
I0428 16:17:07.557597 29440 solver.cpp:310] Iteration 10200, loss = 0.308579
I0428 16:17:07.557622 29440 solver.cpp:330] Iteration 10200, Testing net (#0)
I0428 16:17:07.557627 29440 net.cpp:676] Ignoring source layer train-data
I0428 16:17:08.031697 29481 data_layer.cpp:73] Restarting data prefetching from start.
I0428 16:17:12.707418 29440 solver.cpp:397]     Test net output #0: accuracy = 0.392157
I0428 16:17:12.711521 29440 solver.cpp:397]     Test net output #1: loss = 3.65174 (* 1 = 3.65174 loss)
I0428 16:17:12.711529 29440 solver.cpp:315] Optimization Done.
I0428 16:17:12.711534 29440 caffe.cpp:259] Optimization Done.