DIGITS-CNN/cars/lr-investigations/sigmoid/1e-2/50_0.05/caffe_output.log

I0407 22:24:16.383111 32718 upgrade_proto.cpp:1082] Attempting to upgrade input file specified using deprecated 'solver_type' field (enum)': /mnt/bigdisk/DIGITS-TAM-1/digits/jobs/20210407-222415-72b6/solver.prototxt
I0407 22:24:16.383256 32718 upgrade_proto.cpp:1089] Successfully upgraded file specified using deprecated 'solver_type' field (enum) to 'type' field (string).
W0407 22:24:16.383261 32718 upgrade_proto.cpp:1091] Note that future Caffe releases will only support 'type' field (string) for a solver's type.
I0407 22:24:16.383323 32718 caffe.cpp:218] Using GPUs 1
I0407 22:24:16.425405 32718 caffe.cpp:223] GPU 1: GeForce RTX 2080
I0407 22:24:16.785578 32718 solver.cpp:44] Initializing solver from parameters:
test_iter: 51
test_interval: 102
base_lr: 0.01
display: 12
max_iter: 10200
lr_policy: "sigmoid"
gamma: -0.00049019611
momentum: 0.9
weight_decay: 0.0001
stepsize: 5100
snapshot: 102
snapshot_prefix: "snapshot"
solver_mode: GPU
device_id: 1
net: "train_val.prototxt"
train_state {
level: 0
stage: ""
}
type: "SGD"
I0407 22:24:16.786384 32718 solver.cpp:87] Creating training net from net file: train_val.prototxt
I0407 22:24:16.786985 32718 net.cpp:294] The NetState phase (0) differed from the phase (1) specified by a rule in layer val-data
I0407 22:24:16.786998 32718 net.cpp:294] The NetState phase (0) differed from the phase (1) specified by a rule in layer accuracy
I0407 22:24:16.787122 32718 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-TAM-1/digits/jobs/20210407-221849-d51c/mean.binaryproto"
}
data_param {
source: "/mnt/bigdisk/DIGITS-TAM-1/digits/jobs/20210407-221849-d51c/train_db"
batch_size: 128
backend: LMDB
}
}
layer {
name: "conv1"
type: "Convolution"
bottom: "data"
top: "conv1"
param {
lr_mult: 1
decay_mult: 1
}
param {
lr_mult: 2
decay_mult: 0
}
convolution_param {
num_output: 96
kernel_size: 11
stride: 4
weight_filler {
type: "gaussian"
std: 0.01
}
bias_filler {
type: "constant"
value: 0
}
}
}
layer {
name: "relu1"
type: "ReLU"
bottom: "conv1"
top: "conv1"
}
layer {
name: "norm1"
type: "LRN"
bottom: "conv1"
top: "norm1"
lrn_param {
local_size: 5
alpha: 0.0001
beta: 0.75
}
}
layer {
name: "pool1"
type: "Pooling"
bottom: "norm1"
top: "pool1"
pooling_param {
pool: MAX
kernel_size: 3
stride: 2
}
}
layer {
name: "conv2"
type: "Convolution"
bottom: "pool1"
top: "conv2"
param {
lr_mult: 1
decay_mult: 1
}
param {
lr_mult: 2
decay_mult: 0
}
convolution_param {
num_output: 256
pad: 2
kernel_size: 5
group: 2
weight_filler {
type: "gaussian"
std: 0.01
}
bias_filler {
type: "constant"
value: 0.1
}
}
}
layer {
name: "relu2"
type: "ReLU"
bottom: "conv2"
top: "conv2"
}
layer {
name: "norm2"
type: "LRN"
bottom: "conv2"
top: "norm2"
lrn_param {
local_size: 5
alpha: 0.0001
beta: 0.75
}
}
layer {
name: "pool2"
type: "Pooling"
bottom: "norm2"
top: "pool2"
pooling_param {
pool: MAX
kernel_size: 3
stride: 2
}
}
layer {
name: "conv3"
type: "Convolution"
bottom: "pool2"
top: "conv3"
param {
lr_mult: 1
decay_mult: 1
}
param {
lr_mult: 2
decay_mult: 0
}
convolution_param {
num_output: 384
pad: 1
kernel_size: 3
weight_filler {
type: "gaussian"
std: 0.01
}
bias_filler {
type: "constant"
value: 0
}
}
}
layer {
name: "relu3"
type: "ReLU"
bottom: "conv3"
top: "conv3"
}
layer {
name: "conv4"
type: "Convolution"
bottom: "conv3"
top: "conv4"
param {
lr_mult: 1
decay_mult: 1
}
param {
lr_mult: 2
decay_mult: 0
}
convolution_param {
num_output: 384
pad: 1
kernel_size: 3
group: 2
weight_filler {
type: "gaussian"
std: 0.01
}
bias_filler {
type: "constant"
value: 0.1
}
}
}
layer {
name: "relu4"
type: "ReLU"
bottom: "conv4"
top: "conv4"
}
layer {
name: "conv5"
type: "Convolution"
bottom: "conv4"
top: "conv5"
param {
lr_mult: 1
decay_mult: 1
}
param {
lr_mult: 2
decay_mult: 0
}
convolution_param {
num_output: 256
pad: 1
kernel_size: 3
group: 2
weight_filler {
type: "gaussian"
std: 0.01
}
bias_filler {
type: "constant"
value: 0.1
}
}
}
layer {
name: "relu5"
type: "ReLU"
bottom: "conv5"
top: "conv5"
}
layer {
name: "pool5"
type: "Pooling"
bottom: "conv5"
top: "pool5"
pooling_param {
pool: MAX
kernel_size: 3
stride: 2
}
}
layer {
name: "fc6"
type: "InnerProduct"
bottom: "pool5"
top: "fc6"
param {
lr_mult: 1
decay_mult: 1
}
param {
lr_mult: 2
decay_mult: 0
}
inner_product_param {
num_output: 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"
}
I0407 22:24:16.787212 32718 layer_factory.hpp:77] Creating layer train-data
I0407 22:24:16.788659 32718 db_lmdb.cpp:35] Opened lmdb /mnt/bigdisk/DIGITS-TAM-1/digits/jobs/20210407-221849-d51c/train_db
I0407 22:24:16.789345 32718 net.cpp:84] Creating Layer train-data
I0407 22:24:16.789355 32718 net.cpp:380] train-data -> data
I0407 22:24:16.789376 32718 net.cpp:380] train-data -> label
I0407 22:24:16.789386 32718 data_transformer.cpp:25] Loading mean file from: /mnt/bigdisk/DIGITS-TAM-1/digits/jobs/20210407-221849-d51c/mean.binaryproto
I0407 22:24:16.793639 32718 data_layer.cpp:45] output data size: 128,3,227,227
I0407 22:24:16.921115 32718 net.cpp:122] Setting up train-data
I0407 22:24:16.921134 32718 net.cpp:129] Top shape: 128 3 227 227 (19787136)
I0407 22:24:16.921139 32718 net.cpp:129] Top shape: 128 (128)
I0407 22:24:16.921142 32718 net.cpp:137] Memory required for data: 79149056
I0407 22:24:16.921151 32718 layer_factory.hpp:77] Creating layer conv1
I0407 22:24:16.921171 32718 net.cpp:84] Creating Layer conv1
I0407 22:24:16.921176 32718 net.cpp:406] conv1 <- data
I0407 22:24:16.921188 32718 net.cpp:380] conv1 -> conv1
I0407 22:24:17.783514 32718 net.cpp:122] Setting up conv1
I0407 22:24:17.783531 32718 net.cpp:129] Top shape: 128 96 55 55 (37171200)
I0407 22:24:17.783535 32718 net.cpp:137] Memory required for data: 227833856
I0407 22:24:17.783555 32718 layer_factory.hpp:77] Creating layer relu1
I0407 22:24:17.783565 32718 net.cpp:84] Creating Layer relu1
I0407 22:24:17.783567 32718 net.cpp:406] relu1 <- conv1
I0407 22:24:17.783573 32718 net.cpp:367] relu1 -> conv1 (in-place)
I0407 22:24:17.783888 32718 net.cpp:122] Setting up relu1
I0407 22:24:17.783897 32718 net.cpp:129] Top shape: 128 96 55 55 (37171200)
I0407 22:24:17.783900 32718 net.cpp:137] Memory required for data: 376518656
I0407 22:24:17.783903 32718 layer_factory.hpp:77] Creating layer norm1
I0407 22:24:17.783911 32718 net.cpp:84] Creating Layer norm1
I0407 22:24:17.783929 32718 net.cpp:406] norm1 <- conv1
I0407 22:24:17.783936 32718 net.cpp:380] norm1 -> norm1
I0407 22:24:17.784480 32718 net.cpp:122] Setting up norm1
I0407 22:24:17.784490 32718 net.cpp:129] Top shape: 128 96 55 55 (37171200)
I0407 22:24:17.784493 32718 net.cpp:137] Memory required for data: 525203456
I0407 22:24:17.784497 32718 layer_factory.hpp:77] Creating layer pool1
I0407 22:24:17.784504 32718 net.cpp:84] Creating Layer pool1
I0407 22:24:17.784507 32718 net.cpp:406] pool1 <- norm1
I0407 22:24:17.784512 32718 net.cpp:380] pool1 -> pool1
I0407 22:24:17.784548 32718 net.cpp:122] Setting up pool1
I0407 22:24:17.784554 32718 net.cpp:129] Top shape: 128 96 27 27 (8957952)
I0407 22:24:17.784556 32718 net.cpp:137] Memory required for data: 561035264
I0407 22:24:17.784559 32718 layer_factory.hpp:77] Creating layer conv2
I0407 22:24:17.784567 32718 net.cpp:84] Creating Layer conv2
I0407 22:24:17.784570 32718 net.cpp:406] conv2 <- pool1
I0407 22:24:17.784574 32718 net.cpp:380] conv2 -> conv2
I0407 22:24:17.792659 32718 net.cpp:122] Setting up conv2
I0407 22:24:17.792675 32718 net.cpp:129] Top shape: 128 256 27 27 (23887872)
I0407 22:24:17.792678 32718 net.cpp:137] Memory required for data: 656586752
I0407 22:24:17.792688 32718 layer_factory.hpp:77] Creating layer relu2
I0407 22:24:17.792695 32718 net.cpp:84] Creating Layer relu2
I0407 22:24:17.792698 32718 net.cpp:406] relu2 <- conv2
I0407 22:24:17.792702 32718 net.cpp:367] relu2 -> conv2 (in-place)
I0407 22:24:17.793268 32718 net.cpp:122] Setting up relu2
I0407 22:24:17.793279 32718 net.cpp:129] Top shape: 128 256 27 27 (23887872)
I0407 22:24:17.793282 32718 net.cpp:137] Memory required for data: 752138240
I0407 22:24:17.793285 32718 layer_factory.hpp:77] Creating layer norm2
I0407 22:24:17.793292 32718 net.cpp:84] Creating Layer norm2
I0407 22:24:17.793294 32718 net.cpp:406] norm2 <- conv2
I0407 22:24:17.793300 32718 net.cpp:380] norm2 -> norm2
I0407 22:24:17.793694 32718 net.cpp:122] Setting up norm2
I0407 22:24:17.793702 32718 net.cpp:129] Top shape: 128 256 27 27 (23887872)
I0407 22:24:17.793705 32718 net.cpp:137] Memory required for data: 847689728
I0407 22:24:17.793709 32718 layer_factory.hpp:77] Creating layer pool2
I0407 22:24:17.793716 32718 net.cpp:84] Creating Layer pool2
I0407 22:24:17.793720 32718 net.cpp:406] pool2 <- norm2
I0407 22:24:17.793725 32718 net.cpp:380] pool2 -> pool2
I0407 22:24:17.793751 32718 net.cpp:122] Setting up pool2
I0407 22:24:17.793756 32718 net.cpp:129] Top shape: 128 256 13 13 (5537792)
I0407 22:24:17.793759 32718 net.cpp:137] Memory required for data: 869840896
I0407 22:24:17.793761 32718 layer_factory.hpp:77] Creating layer conv3
I0407 22:24:17.793771 32718 net.cpp:84] Creating Layer conv3
I0407 22:24:17.793773 32718 net.cpp:406] conv3 <- pool2
I0407 22:24:17.793779 32718 net.cpp:380] conv3 -> conv3
I0407 22:24:17.804400 32718 net.cpp:122] Setting up conv3
I0407 22:24:17.804414 32718 net.cpp:129] Top shape: 128 384 13 13 (8306688)
I0407 22:24:17.804416 32718 net.cpp:137] Memory required for data: 903067648
I0407 22:24:17.804425 32718 layer_factory.hpp:77] Creating layer relu3
I0407 22:24:17.804432 32718 net.cpp:84] Creating Layer relu3
I0407 22:24:17.804436 32718 net.cpp:406] relu3 <- conv3
I0407 22:24:17.804440 32718 net.cpp:367] relu3 -> conv3 (in-place)
I0407 22:24:17.805012 32718 net.cpp:122] Setting up relu3
I0407 22:24:17.805022 32718 net.cpp:129] Top shape: 128 384 13 13 (8306688)
I0407 22:24:17.805024 32718 net.cpp:137] Memory required for data: 936294400
I0407 22:24:17.805027 32718 layer_factory.hpp:77] Creating layer conv4
I0407 22:24:17.805037 32718 net.cpp:84] Creating Layer conv4
I0407 22:24:17.805040 32718 net.cpp:406] conv4 <- conv3
I0407 22:24:17.805047 32718 net.cpp:380] conv4 -> conv4
I0407 22:24:17.816521 32718 net.cpp:122] Setting up conv4
I0407 22:24:17.816535 32718 net.cpp:129] Top shape: 128 384 13 13 (8306688)
I0407 22:24:17.816540 32718 net.cpp:137] Memory required for data: 969521152
I0407 22:24:17.816546 32718 layer_factory.hpp:77] Creating layer relu4
I0407 22:24:17.816555 32718 net.cpp:84] Creating Layer relu4
I0407 22:24:17.816572 32718 net.cpp:406] relu4 <- conv4
I0407 22:24:17.816578 32718 net.cpp:367] relu4 -> conv4 (in-place)
I0407 22:24:17.817134 32718 net.cpp:122] Setting up relu4
I0407 22:24:17.817145 32718 net.cpp:129] Top shape: 128 384 13 13 (8306688)
I0407 22:24:17.817148 32718 net.cpp:137] Memory required for data: 1002747904
I0407 22:24:17.817152 32718 layer_factory.hpp:77] Creating layer conv5
I0407 22:24:17.817162 32718 net.cpp:84] Creating Layer conv5
I0407 22:24:17.817165 32718 net.cpp:406] conv5 <- conv4
I0407 22:24:17.817170 32718 net.cpp:380] conv5 -> conv5
I0407 22:24:17.829231 32718 net.cpp:122] Setting up conv5
I0407 22:24:17.829248 32718 net.cpp:129] Top shape: 128 256 13 13 (5537792)
I0407 22:24:17.829252 32718 net.cpp:137] Memory required for data: 1024899072
I0407 22:24:17.829267 32718 layer_factory.hpp:77] Creating layer relu5
I0407 22:24:17.829277 32718 net.cpp:84] Creating Layer relu5
I0407 22:24:17.829282 32718 net.cpp:406] relu5 <- conv5
I0407 22:24:17.829289 32718 net.cpp:367] relu5 -> conv5 (in-place)
I0407 22:24:17.830013 32718 net.cpp:122] Setting up relu5
I0407 22:24:17.830025 32718 net.cpp:129] Top shape: 128 256 13 13 (5537792)
I0407 22:24:17.830029 32718 net.cpp:137] Memory required for data: 1047050240
I0407 22:24:17.830034 32718 layer_factory.hpp:77] Creating layer pool5
I0407 22:24:17.830041 32718 net.cpp:84] Creating Layer pool5
I0407 22:24:17.830045 32718 net.cpp:406] pool5 <- conv5
I0407 22:24:17.830054 32718 net.cpp:380] pool5 -> pool5
I0407 22:24:17.830098 32718 net.cpp:122] Setting up pool5
I0407 22:24:17.830106 32718 net.cpp:129] Top shape: 128 256 6 6 (1179648)
I0407 22:24:17.830109 32718 net.cpp:137] Memory required for data: 1051768832
I0407 22:24:17.830112 32718 layer_factory.hpp:77] Creating layer fc6
I0407 22:24:17.830124 32718 net.cpp:84] Creating Layer fc6
I0407 22:24:17.830128 32718 net.cpp:406] fc6 <- pool5
I0407 22:24:17.830134 32718 net.cpp:380] fc6 -> fc6
I0407 22:24:18.338637 32718 net.cpp:122] Setting up fc6
I0407 22:24:18.338660 32718 net.cpp:129] Top shape: 128 4096 (524288)
I0407 22:24:18.338663 32718 net.cpp:137] Memory required for data: 1053865984
I0407 22:24:18.338675 32718 layer_factory.hpp:77] Creating layer relu6
I0407 22:24:18.338685 32718 net.cpp:84] Creating Layer relu6
I0407 22:24:18.338690 32718 net.cpp:406] relu6 <- fc6
I0407 22:24:18.338697 32718 net.cpp:367] relu6 -> fc6 (in-place)
I0407 22:24:18.339670 32718 net.cpp:122] Setting up relu6
I0407 22:24:18.339682 32718 net.cpp:129] Top shape: 128 4096 (524288)
I0407 22:24:18.339686 32718 net.cpp:137] Memory required for data: 1055963136
I0407 22:24:18.339690 32718 layer_factory.hpp:77] Creating layer drop6
I0407 22:24:18.339699 32718 net.cpp:84] Creating Layer drop6
I0407 22:24:18.339702 32718 net.cpp:406] drop6 <- fc6
I0407 22:24:18.339710 32718 net.cpp:367] drop6 -> fc6 (in-place)
I0407 22:24:18.339741 32718 net.cpp:122] Setting up drop6
I0407 22:24:18.339747 32718 net.cpp:129] Top shape: 128 4096 (524288)
I0407 22:24:18.339751 32718 net.cpp:137] Memory required for data: 1058060288
I0407 22:24:18.339754 32718 layer_factory.hpp:77] Creating layer fc7
I0407 22:24:18.339766 32718 net.cpp:84] Creating Layer fc7
I0407 22:24:18.339769 32718 net.cpp:406] fc7 <- fc6
I0407 22:24:18.339776 32718 net.cpp:380] fc7 -> fc7
I0407 22:24:18.561173 32718 net.cpp:122] Setting up fc7
I0407 22:24:18.561195 32718 net.cpp:129] Top shape: 128 4096 (524288)
I0407 22:24:18.561199 32718 net.cpp:137] Memory required for data: 1060157440
I0407 22:24:18.561211 32718 layer_factory.hpp:77] Creating layer relu7
I0407 22:24:18.561223 32718 net.cpp:84] Creating Layer relu7
I0407 22:24:18.561228 32718 net.cpp:406] relu7 <- fc7
I0407 22:24:18.561236 32718 net.cpp:367] relu7 -> fc7 (in-place)
I0407 22:24:18.561866 32718 net.cpp:122] Setting up relu7
I0407 22:24:18.561875 32718 net.cpp:129] Top shape: 128 4096 (524288)
I0407 22:24:18.561878 32718 net.cpp:137] Memory required for data: 1062254592
I0407 22:24:18.561882 32718 layer_factory.hpp:77] Creating layer drop7
I0407 22:24:18.561890 32718 net.cpp:84] Creating Layer drop7
I0407 22:24:18.561910 32718 net.cpp:406] drop7 <- fc7
I0407 22:24:18.561919 32718 net.cpp:367] drop7 -> fc7 (in-place)
I0407 22:24:18.561947 32718 net.cpp:122] Setting up drop7
I0407 22:24:18.561954 32718 net.cpp:129] Top shape: 128 4096 (524288)
I0407 22:24:18.561956 32718 net.cpp:137] Memory required for data: 1064351744
I0407 22:24:18.561960 32718 layer_factory.hpp:77] Creating layer fc8
I0407 22:24:18.561969 32718 net.cpp:84] Creating Layer fc8
I0407 22:24:18.561971 32718 net.cpp:406] fc8 <- fc7
I0407 22:24:18.561978 32718 net.cpp:380] fc8 -> fc8
I0407 22:24:18.572631 32718 net.cpp:122] Setting up fc8
I0407 22:24:18.572643 32718 net.cpp:129] Top shape: 128 196 (25088)
I0407 22:24:18.572645 32718 net.cpp:137] Memory required for data: 1064452096
I0407 22:24:18.572652 32718 layer_factory.hpp:77] Creating layer loss
I0407 22:24:18.572660 32718 net.cpp:84] Creating Layer loss
I0407 22:24:18.572664 32718 net.cpp:406] loss <- fc8
I0407 22:24:18.572669 32718 net.cpp:406] loss <- label
I0407 22:24:18.572679 32718 net.cpp:380] loss -> loss
I0407 22:24:18.572688 32718 layer_factory.hpp:77] Creating layer loss
I0407 22:24:18.573508 32718 net.cpp:122] Setting up loss
I0407 22:24:18.573519 32718 net.cpp:129] Top shape: (1)
I0407 22:24:18.573523 32718 net.cpp:132]     with loss weight 1
I0407 22:24:18.573544 32718 net.cpp:137] Memory required for data: 1064452100
I0407 22:24:18.573549 32718 net.cpp:198] loss needs backward computation.
I0407 22:24:18.573556 32718 net.cpp:198] fc8 needs backward computation.
I0407 22:24:18.573560 32718 net.cpp:198] drop7 needs backward computation.
I0407 22:24:18.573563 32718 net.cpp:198] relu7 needs backward computation.
I0407 22:24:18.573566 32718 net.cpp:198] fc7 needs backward computation.
I0407 22:24:18.573570 32718 net.cpp:198] drop6 needs backward computation.
I0407 22:24:18.573575 32718 net.cpp:198] relu6 needs backward computation.
I0407 22:24:18.573577 32718 net.cpp:198] fc6 needs backward computation.
I0407 22:24:18.573581 32718 net.cpp:198] pool5 needs backward computation.
I0407 22:24:18.573585 32718 net.cpp:198] relu5 needs backward computation.
I0407 22:24:18.573588 32718 net.cpp:198] conv5 needs backward computation.
I0407 22:24:18.573592 32718 net.cpp:198] relu4 needs backward computation.
I0407 22:24:18.573596 32718 net.cpp:198] conv4 needs backward computation.
I0407 22:24:18.573599 32718 net.cpp:198] relu3 needs backward computation.
I0407 22:24:18.573603 32718 net.cpp:198] conv3 needs backward computation.
I0407 22:24:18.573607 32718 net.cpp:198] pool2 needs backward computation.
I0407 22:24:18.573611 32718 net.cpp:198] norm2 needs backward computation.
I0407 22:24:18.573614 32718 net.cpp:198] relu2 needs backward computation.
I0407 22:24:18.573618 32718 net.cpp:198] conv2 needs backward computation.
I0407 22:24:18.573622 32718 net.cpp:198] pool1 needs backward computation.
I0407 22:24:18.573626 32718 net.cpp:198] norm1 needs backward computation.
I0407 22:24:18.573629 32718 net.cpp:198] relu1 needs backward computation.
I0407 22:24:18.573633 32718 net.cpp:198] conv1 needs backward computation.
I0407 22:24:18.573637 32718 net.cpp:200] train-data does not need backward computation.
I0407 22:24:18.573640 32718 net.cpp:242] This network produces output loss
I0407 22:24:18.573658 32718 net.cpp:255] Network initialization done.
I0407 22:24:18.574224 32718 solver.cpp:172] Creating test net (#0) specified by net file: train_val.prototxt
I0407 22:24:18.574262 32718 net.cpp:294] The NetState phase (1) differed from the phase (0) specified by a rule in layer train-data
I0407 22:24:18.574456 32718 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-TAM-1/digits/jobs/20210407-221849-d51c/mean.binaryproto"
}
data_param {
source: "/mnt/bigdisk/DIGITS-TAM-1/digits/jobs/20210407-221849-d51c/val_db"
batch_size: 32
backend: LMDB
}
}
layer {
name: "conv1"
type: "Convolution"
bottom: "data"
top: "conv1"
param {
lr_mult: 1
decay_mult: 1
}
param {
lr_mult: 2
decay_mult: 0
}
convolution_param {
num_output: 96
kernel_size: 11
stride: 4
weight_filler {
type: "gaussian"
std: 0.01
}
bias_filler {
type: "constant"
value: 0
}
}
}
layer {
name: "relu1"
type: "ReLU"
bottom: "conv1"
top: "conv1"
}
layer {
name: "norm1"
type: "LRN"
bottom: "conv1"
top: "norm1"
lrn_param {
local_size: 5
alpha: 0.0001
beta: 0.75
}
}
layer {
name: "pool1"
type: "Pooling"
bottom: "norm1"
top: "pool1"
pooling_param {
pool: MAX
kernel_size: 3
stride: 2
}
}
layer {
name: "conv2"
type: "Convolution"
bottom: "pool1"
top: "conv2"
param {
lr_mult: 1
decay_mult: 1
}
param {
lr_mult: 2
decay_mult: 0
}
convolution_param {
num_output: 256
pad: 2
kernel_size: 5
group: 2
weight_filler {
type: "gaussian"
std: 0.01
}
bias_filler {
type: "constant"
value: 0.1
}
}
}
layer {
name: "relu2"
type: "ReLU"
bottom: "conv2"
top: "conv2"
}
layer {
name: "norm2"
type: "LRN"
bottom: "conv2"
top: "norm2"
lrn_param {
local_size: 5
alpha: 0.0001
beta: 0.75
}
}
layer {
name: "pool2"
type: "Pooling"
bottom: "norm2"
top: "pool2"
pooling_param {
pool: MAX
kernel_size: 3
stride: 2
}
}
layer {
name: "conv3"
type: "Convolution"
bottom: "pool2"
top: "conv3"
param {
lr_mult: 1
decay_mult: 1
}
param {
lr_mult: 2
decay_mult: 0
}
convolution_param {
num_output: 384
pad: 1
kernel_size: 3
weight_filler {
type: "gaussian"
std: 0.01
}
bias_filler {
type: "constant"
value: 0
}
}
}
layer {
name: "relu3"
type: "ReLU"
bottom: "conv3"
top: "conv3"
}
layer {
name: "conv4"
type: "Convolution"
bottom: "conv3"
top: "conv4"
param {
lr_mult: 1
decay_mult: 1
}
param {
lr_mult: 2
decay_mult: 0
}
convolution_param {
num_output: 384
pad: 1
kernel_size: 3
group: 2
weight_filler {
type: "gaussian"
std: 0.01
}
bias_filler {
type: "constant"
value: 0.1
}
}
}
layer {
name: "relu4"
type: "ReLU"
bottom: "conv4"
top: "conv4"
}
layer {
name: "conv5"
type: "Convolution"
bottom: "conv4"
top: "conv5"
param {
lr_mult: 1
decay_mult: 1
}
param {
lr_mult: 2
decay_mult: 0
}
convolution_param {
num_output: 256
pad: 1
kernel_size: 3
group: 2
weight_filler {
type: "gaussian"
std: 0.01
}
bias_filler {
type: "constant"
value: 0.1
}
}
}
layer {
name: "relu5"
type: "ReLU"
bottom: "conv5"
top: "conv5"
}
layer {
name: "pool5"
type: "Pooling"
bottom: "conv5"
top: "pool5"
pooling_param {
pool: MAX
kernel_size: 3
stride: 2
}
}
layer {
name: "fc6"
type: "InnerProduct"
bottom: "pool5"
top: "fc6"
param {
lr_mult: 1
decay_mult: 1
}
param {
lr_mult: 2
decay_mult: 0
}
inner_product_param {
num_output: 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"
}
I0407 22:24:18.574579 32718 layer_factory.hpp:77] Creating layer val-data
I0407 22:24:18.576093 32718 db_lmdb.cpp:35] Opened lmdb /mnt/bigdisk/DIGITS-TAM-1/digits/jobs/20210407-221849-d51c/val_db
I0407 22:24:18.576725 32718 net.cpp:84] Creating Layer val-data
I0407 22:24:18.576735 32718 net.cpp:380] val-data -> data
I0407 22:24:18.576747 32718 net.cpp:380] val-data -> label
I0407 22:24:18.576756 32718 data_transformer.cpp:25] Loading mean file from: /mnt/bigdisk/DIGITS-TAM-1/digits/jobs/20210407-221849-d51c/mean.binaryproto
I0407 22:24:18.581146 32718 data_layer.cpp:45] output data size: 32,3,227,227
I0407 22:24:18.619853 32718 net.cpp:122] Setting up val-data
I0407 22:24:18.619871 32718 net.cpp:129] Top shape: 32 3 227 227 (4946784)
I0407 22:24:18.619875 32718 net.cpp:129] Top shape: 32 (32)
I0407 22:24:18.619877 32718 net.cpp:137] Memory required for data: 19787264
I0407 22:24:18.619884 32718 layer_factory.hpp:77] Creating layer label_val-data_1_split
I0407 22:24:18.619895 32718 net.cpp:84] Creating Layer label_val-data_1_split
I0407 22:24:18.619899 32718 net.cpp:406] label_val-data_1_split <- label
I0407 22:24:18.619905 32718 net.cpp:380] label_val-data_1_split -> label_val-data_1_split_0
I0407 22:24:18.619915 32718 net.cpp:380] label_val-data_1_split -> label_val-data_1_split_1
I0407 22:24:18.619967 32718 net.cpp:122] Setting up label_val-data_1_split
I0407 22:24:18.619972 32718 net.cpp:129] Top shape: 32 (32)
I0407 22:24:18.619976 32718 net.cpp:129] Top shape: 32 (32)
I0407 22:24:18.619978 32718 net.cpp:137] Memory required for data: 19787520
I0407 22:24:18.619980 32718 layer_factory.hpp:77] Creating layer conv1
I0407 22:24:18.619992 32718 net.cpp:84] Creating Layer conv1
I0407 22:24:18.619994 32718 net.cpp:406] conv1 <- data
I0407 22:24:18.619999 32718 net.cpp:380] conv1 -> conv1
I0407 22:24:18.623205 32718 net.cpp:122] Setting up conv1
I0407 22:24:18.623215 32718 net.cpp:129] Top shape: 32 96 55 55 (9292800)
I0407 22:24:18.623219 32718 net.cpp:137] Memory required for data: 56958720
I0407 22:24:18.623229 32718 layer_factory.hpp:77] Creating layer relu1
I0407 22:24:18.623234 32718 net.cpp:84] Creating Layer relu1
I0407 22:24:18.623236 32718 net.cpp:406] relu1 <- conv1
I0407 22:24:18.623241 32718 net.cpp:367] relu1 -> conv1 (in-place)
I0407 22:24:18.623569 32718 net.cpp:122] Setting up relu1
I0407 22:24:18.623579 32718 net.cpp:129] Top shape: 32 96 55 55 (9292800)
I0407 22:24:18.623580 32718 net.cpp:137] Memory required for data: 94129920
I0407 22:24:18.623584 32718 layer_factory.hpp:77] Creating layer norm1
I0407 22:24:18.623591 32718 net.cpp:84] Creating Layer norm1
I0407 22:24:18.623594 32718 net.cpp:406] norm1 <- conv1
I0407 22:24:18.623600 32718 net.cpp:380] norm1 -> norm1
I0407 22:24:18.624128 32718 net.cpp:122] Setting up norm1
I0407 22:24:18.624138 32718 net.cpp:129] Top shape: 32 96 55 55 (9292800)
I0407 22:24:18.624140 32718 net.cpp:137] Memory required for data: 131301120
I0407 22:24:18.624143 32718 layer_factory.hpp:77] Creating layer pool1
I0407 22:24:18.624150 32718 net.cpp:84] Creating Layer pool1
I0407 22:24:18.624153 32718 net.cpp:406] pool1 <- norm1
I0407 22:24:18.624158 32718 net.cpp:380] pool1 -> pool1
I0407 22:24:18.624182 32718 net.cpp:122] Setting up pool1
I0407 22:24:18.624187 32718 net.cpp:129] Top shape: 32 96 27 27 (2239488)
I0407 22:24:18.624189 32718 net.cpp:137] Memory required for data: 140259072
I0407 22:24:18.624192 32718 layer_factory.hpp:77] Creating layer conv2
I0407 22:24:18.624199 32718 net.cpp:84] Creating Layer conv2
I0407 22:24:18.624202 32718 net.cpp:406] conv2 <- pool1
I0407 22:24:18.624222 32718 net.cpp:380] conv2 -> conv2
I0407 22:24:18.633859 32718 net.cpp:122] Setting up conv2
I0407 22:24:18.633873 32718 net.cpp:129] Top shape: 32 256 27 27 (5971968)
I0407 22:24:18.633877 32718 net.cpp:137] Memory required for data: 164146944
I0407 22:24:18.633885 32718 layer_factory.hpp:77] Creating layer relu2
I0407 22:24:18.633893 32718 net.cpp:84] Creating Layer relu2
I0407 22:24:18.633895 32718 net.cpp:406] relu2 <- conv2
I0407 22:24:18.633903 32718 net.cpp:367] relu2 -> conv2 (in-place)
I0407 22:24:18.634476 32718 net.cpp:122] Setting up relu2
I0407 22:24:18.634485 32718 net.cpp:129] Top shape: 32 256 27 27 (5971968)
I0407 22:24:18.634488 32718 net.cpp:137] Memory required for data: 188034816
I0407 22:24:18.634491 32718 layer_factory.hpp:77] Creating layer norm2
I0407 22:24:18.634501 32718 net.cpp:84] Creating Layer norm2
I0407 22:24:18.634505 32718 net.cpp:406] norm2 <- conv2
I0407 22:24:18.634510 32718 net.cpp:380] norm2 -> norm2
I0407 22:24:18.635298 32718 net.cpp:122] Setting up norm2
I0407 22:24:18.635308 32718 net.cpp:129] Top shape: 32 256 27 27 (5971968)
I0407 22:24:18.635311 32718 net.cpp:137] Memory required for data: 211922688
I0407 22:24:18.635314 32718 layer_factory.hpp:77] Creating layer pool2
I0407 22:24:18.635320 32718 net.cpp:84] Creating Layer pool2
I0407 22:24:18.635324 32718 net.cpp:406] pool2 <- norm2
I0407 22:24:18.635329 32718 net.cpp:380] pool2 -> pool2
I0407 22:24:18.635360 32718 net.cpp:122] Setting up pool2
I0407 22:24:18.635365 32718 net.cpp:129] Top shape: 32 256 13 13 (1384448)
I0407 22:24:18.635366 32718 net.cpp:137] Memory required for data: 217460480
I0407 22:24:18.635370 32718 layer_factory.hpp:77] Creating layer conv3
I0407 22:24:18.635380 32718 net.cpp:84] Creating Layer conv3
I0407 22:24:18.635381 32718 net.cpp:406] conv3 <- pool2
I0407 22:24:18.635386 32718 net.cpp:380] conv3 -> conv3
I0407 22:24:18.647136 32718 net.cpp:122] Setting up conv3
I0407 22:24:18.647153 32718 net.cpp:129] Top shape: 32 384 13 13 (2076672)
I0407 22:24:18.647157 32718 net.cpp:137] Memory required for data: 225767168
I0407 22:24:18.647167 32718 layer_factory.hpp:77] Creating layer relu3
I0407 22:24:18.647176 32718 net.cpp:84] Creating Layer relu3
I0407 22:24:18.647178 32718 net.cpp:406] relu3 <- conv3
I0407 22:24:18.647184 32718 net.cpp:367] relu3 -> conv3 (in-place)
I0407 22:24:18.647795 32718 net.cpp:122] Setting up relu3
I0407 22:24:18.647806 32718 net.cpp:129] Top shape: 32 384 13 13 (2076672)
I0407 22:24:18.647809 32718 net.cpp:137] Memory required for data: 234073856
I0407 22:24:18.647812 32718 layer_factory.hpp:77] Creating layer conv4
I0407 22:24:18.647822 32718 net.cpp:84] Creating Layer conv4
I0407 22:24:18.647825 32718 net.cpp:406] conv4 <- conv3
I0407 22:24:18.647831 32718 net.cpp:380] conv4 -> conv4
I0407 22:24:18.658180 32718 net.cpp:122] Setting up conv4
I0407 22:24:18.658193 32718 net.cpp:129] Top shape: 32 384 13 13 (2076672)
I0407 22:24:18.658196 32718 net.cpp:137] Memory required for data: 242380544
I0407 22:24:18.658203 32718 layer_factory.hpp:77] Creating layer relu4
I0407 22:24:18.658211 32718 net.cpp:84] Creating Layer relu4
I0407 22:24:18.658215 32718 net.cpp:406] relu4 <- conv4
I0407 22:24:18.658219 32718 net.cpp:367] relu4 -> conv4 (in-place)
I0407 22:24:18.658596 32718 net.cpp:122] Setting up relu4
I0407 22:24:18.658608 32718 net.cpp:129] Top shape: 32 384 13 13 (2076672)
I0407 22:24:18.658612 32718 net.cpp:137] Memory required for data: 250687232
I0407 22:24:18.658614 32718 layer_factory.hpp:77] Creating layer conv5
I0407 22:24:18.658623 32718 net.cpp:84] Creating Layer conv5
I0407 22:24:18.658627 32718 net.cpp:406] conv5 <- conv4
I0407 22:24:18.658632 32718 net.cpp:380] conv5 -> conv5
I0407 22:24:18.668480 32718 net.cpp:122] Setting up conv5
I0407 22:24:18.668495 32718 net.cpp:129] Top shape: 32 256 13 13 (1384448)
I0407 22:24:18.668498 32718 net.cpp:137] Memory required for data: 256225024
I0407 22:24:18.668511 32718 layer_factory.hpp:77] Creating layer relu5
I0407 22:24:18.668519 32718 net.cpp:84] Creating Layer relu5
I0407 22:24:18.668546 32718 net.cpp:406] relu5 <- conv5
I0407 22:24:18.668552 32718 net.cpp:367] relu5 -> conv5 (in-place)
I0407 22:24:18.669122 32718 net.cpp:122] Setting up relu5
I0407 22:24:18.669133 32718 net.cpp:129] Top shape: 32 256 13 13 (1384448)
I0407 22:24:18.669137 32718 net.cpp:137] Memory required for data: 261762816
I0407 22:24:18.669139 32718 layer_factory.hpp:77] Creating layer pool5
I0407 22:24:18.669149 32718 net.cpp:84] Creating Layer pool5
I0407 22:24:18.669152 32718 net.cpp:406] pool5 <- conv5
I0407 22:24:18.669157 32718 net.cpp:380] pool5 -> pool5
I0407 22:24:18.669193 32718 net.cpp:122] Setting up pool5
I0407 22:24:18.669198 32718 net.cpp:129] Top shape: 32 256 6 6 (294912)
I0407 22:24:18.669201 32718 net.cpp:137] Memory required for data: 262942464
I0407 22:24:18.669203 32718 layer_factory.hpp:77] Creating layer fc6
I0407 22:24:18.669209 32718 net.cpp:84] Creating Layer fc6
I0407 22:24:18.669212 32718 net.cpp:406] fc6 <- pool5
I0407 22:24:18.669219 32718 net.cpp:380] fc6 -> fc6
I0407 22:24:19.039827 32718 net.cpp:122] Setting up fc6
I0407 22:24:19.039849 32718 net.cpp:129] Top shape: 32 4096 (131072)
I0407 22:24:19.039851 32718 net.cpp:137] Memory required for data: 263466752
I0407 22:24:19.039860 32718 layer_factory.hpp:77] Creating layer relu6
I0407 22:24:19.039867 32718 net.cpp:84] Creating Layer relu6
I0407 22:24:19.039871 32718 net.cpp:406] relu6 <- fc6
I0407 22:24:19.039878 32718 net.cpp:367] relu6 -> fc6 (in-place)
I0407 22:24:19.040699 32718 net.cpp:122] Setting up relu6
I0407 22:24:19.040710 32718 net.cpp:129] Top shape: 32 4096 (131072)
I0407 22:24:19.040714 32718 net.cpp:137] Memory required for data: 263991040
I0407 22:24:19.040716 32718 layer_factory.hpp:77] Creating layer drop6
I0407 22:24:19.040724 32718 net.cpp:84] Creating Layer drop6
I0407 22:24:19.040726 32718 net.cpp:406] drop6 <- fc6
I0407 22:24:19.040730 32718 net.cpp:367] drop6 -> fc6 (in-place)
I0407 22:24:19.040755 32718 net.cpp:122] Setting up drop6
I0407 22:24:19.040760 32718 net.cpp:129] Top shape: 32 4096 (131072)
I0407 22:24:19.040762 32718 net.cpp:137] Memory required for data: 264515328
I0407 22:24:19.040766 32718 layer_factory.hpp:77] Creating layer fc7
I0407 22:24:19.040772 32718 net.cpp:84] Creating Layer fc7
I0407 22:24:19.040776 32718 net.cpp:406] fc7 <- fc6
I0407 22:24:19.040781 32718 net.cpp:380] fc7 -> fc7
I0407 22:24:19.195336 32718 net.cpp:122] Setting up fc7
I0407 22:24:19.195355 32718 net.cpp:129] Top shape: 32 4096 (131072)
I0407 22:24:19.195359 32718 net.cpp:137] Memory required for data: 265039616
I0407 22:24:19.195369 32718 layer_factory.hpp:77] Creating layer relu7
I0407 22:24:19.195377 32718 net.cpp:84] Creating Layer relu7
I0407 22:24:19.195381 32718 net.cpp:406] relu7 <- fc7
I0407 22:24:19.195387 32718 net.cpp:367] relu7 -> fc7 (in-place)
I0407 22:24:19.195897 32718 net.cpp:122] Setting up relu7
I0407 22:24:19.195905 32718 net.cpp:129] Top shape: 32 4096 (131072)
I0407 22:24:19.195909 32718 net.cpp:137] Memory required for data: 265563904
I0407 22:24:19.195910 32718 layer_factory.hpp:77] Creating layer drop7
I0407 22:24:19.195919 32718 net.cpp:84] Creating Layer drop7
I0407 22:24:19.195922 32718 net.cpp:406] drop7 <- fc7
I0407 22:24:19.195926 32718 net.cpp:367] drop7 -> fc7 (in-place)
I0407 22:24:19.195950 32718 net.cpp:122] Setting up drop7
I0407 22:24:19.195953 32718 net.cpp:129] Top shape: 32 4096 (131072)
I0407 22:24:19.195956 32718 net.cpp:137] Memory required for data: 266088192
I0407 22:24:19.195958 32718 layer_factory.hpp:77] Creating layer fc8
I0407 22:24:19.195966 32718 net.cpp:84] Creating Layer fc8
I0407 22:24:19.195967 32718 net.cpp:406] fc8 <- fc7
I0407 22:24:19.195973 32718 net.cpp:380] fc8 -> fc8
I0407 22:24:19.203724 32718 net.cpp:122] Setting up fc8
I0407 22:24:19.203734 32718 net.cpp:129] Top shape: 32 196 (6272)
I0407 22:24:19.203737 32718 net.cpp:137] Memory required for data: 266113280
I0407 22:24:19.203742 32718 layer_factory.hpp:77] Creating layer fc8_fc8_0_split
I0407 22:24:19.203748 32718 net.cpp:84] Creating Layer fc8_fc8_0_split
I0407 22:24:19.203750 32718 net.cpp:406] fc8_fc8_0_split <- fc8
I0407 22:24:19.203768 32718 net.cpp:380] fc8_fc8_0_split -> fc8_fc8_0_split_0
I0407 22:24:19.203774 32718 net.cpp:380] fc8_fc8_0_split -> fc8_fc8_0_split_1
I0407 22:24:19.203802 32718 net.cpp:122] Setting up fc8_fc8_0_split
I0407 22:24:19.203807 32718 net.cpp:129] Top shape: 32 196 (6272)
I0407 22:24:19.203810 32718 net.cpp:129] Top shape: 32 196 (6272)
I0407 22:24:19.203812 32718 net.cpp:137] Memory required for data: 266163456
I0407 22:24:19.203814 32718 layer_factory.hpp:77] Creating layer accuracy
I0407 22:24:19.203821 32718 net.cpp:84] Creating Layer accuracy
I0407 22:24:19.203824 32718 net.cpp:406] accuracy <- fc8_fc8_0_split_0
I0407 22:24:19.203827 32718 net.cpp:406] accuracy <- label_val-data_1_split_0
I0407 22:24:19.203832 32718 net.cpp:380] accuracy -> accuracy
I0407 22:24:19.203840 32718 net.cpp:122] Setting up accuracy
I0407 22:24:19.203842 32718 net.cpp:129] Top shape: (1)
I0407 22:24:19.203845 32718 net.cpp:137] Memory required for data: 266163460
I0407 22:24:19.203846 32718 layer_factory.hpp:77] Creating layer loss
I0407 22:24:19.203851 32718 net.cpp:84] Creating Layer loss
I0407 22:24:19.203853 32718 net.cpp:406] loss <- fc8_fc8_0_split_1
I0407 22:24:19.203856 32718 net.cpp:406] loss <- label_val-data_1_split_1
I0407 22:24:19.203860 32718 net.cpp:380] loss -> loss
I0407 22:24:19.203866 32718 layer_factory.hpp:77] Creating layer loss
I0407 22:24:19.204516 32718 net.cpp:122] Setting up loss
I0407 22:24:19.204524 32718 net.cpp:129] Top shape: (1)
I0407 22:24:19.204526 32718 net.cpp:132]     with loss weight 1
I0407 22:24:19.204535 32718 net.cpp:137] Memory required for data: 266163464
I0407 22:24:19.204538 32718 net.cpp:198] loss needs backward computation.
I0407 22:24:19.204542 32718 net.cpp:200] accuracy does not need backward computation.
I0407 22:24:19.204545 32718 net.cpp:198] fc8_fc8_0_split needs backward computation.
I0407 22:24:19.204548 32718 net.cpp:198] fc8 needs backward computation.
I0407 22:24:19.204550 32718 net.cpp:198] drop7 needs backward computation.
I0407 22:24:19.204553 32718 net.cpp:198] relu7 needs backward computation.
I0407 22:24:19.204555 32718 net.cpp:198] fc7 needs backward computation.
I0407 22:24:19.204558 32718 net.cpp:198] drop6 needs backward computation.
I0407 22:24:19.204560 32718 net.cpp:198] relu6 needs backward computation.
I0407 22:24:19.204563 32718 net.cpp:198] fc6 needs backward computation.
I0407 22:24:19.204566 32718 net.cpp:198] pool5 needs backward computation.
I0407 22:24:19.204569 32718 net.cpp:198] relu5 needs backward computation.
I0407 22:24:19.204573 32718 net.cpp:198] conv5 needs backward computation.
I0407 22:24:19.204576 32718 net.cpp:198] relu4 needs backward computation.
I0407 22:24:19.204579 32718 net.cpp:198] conv4 needs backward computation.
I0407 22:24:19.204581 32718 net.cpp:198] relu3 needs backward computation.
I0407 22:24:19.204584 32718 net.cpp:198] conv3 needs backward computation.
I0407 22:24:19.204587 32718 net.cpp:198] pool2 needs backward computation.
I0407 22:24:19.204591 32718 net.cpp:198] norm2 needs backward computation.
I0407 22:24:19.204592 32718 net.cpp:198] relu2 needs backward computation.
I0407 22:24:19.204596 32718 net.cpp:198] conv2 needs backward computation.
I0407 22:24:19.204597 32718 net.cpp:198] pool1 needs backward computation.
I0407 22:24:19.204600 32718 net.cpp:198] norm1 needs backward computation.
I0407 22:24:19.204602 32718 net.cpp:198] relu1 needs backward computation.
I0407 22:24:19.204605 32718 net.cpp:198] conv1 needs backward computation.
I0407 22:24:19.204608 32718 net.cpp:200] label_val-data_1_split does not need backward computation.
I0407 22:24:19.204612 32718 net.cpp:200] val-data does not need backward computation.
I0407 22:24:19.204613 32718 net.cpp:242] This network produces output accuracy
I0407 22:24:19.204617 32718 net.cpp:242] This network produces output loss
I0407 22:24:19.204632 32718 net.cpp:255] Network initialization done.
I0407 22:24:19.204696 32718 solver.cpp:56] Solver scaffolding done.
I0407 22:24:19.205014 32718 caffe.cpp:248] Starting Optimization
I0407 22:24:19.205030 32718 solver.cpp:272] Solving
I0407 22:24:19.205034 32718 solver.cpp:273] Learning Rate Policy: sigmoid
I0407 22:24:19.206583 32718 solver.cpp:330] Iteration 0, Testing net (#0)
I0407 22:24:19.206590 32718 net.cpp:676] Ignoring source layer train-data
I0407 22:24:19.297063 32718 blocking_queue.cpp:49] Waiting for data
I0407 22:24:23.493813 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:24:23.538113 32718 solver.cpp:397]     Test net output #0: accuracy = 0.00367647
I0407 22:24:23.538159 32718 solver.cpp:397]     Test net output #1: loss = 5.27868 (* 1 = 5.27868 loss)
I0407 22:24:23.636787 32718 solver.cpp:218] Iteration 0 (-1.10703e-43 iter/s, 4.43171s/12 iters), loss = 5.28151
I0407 22:24:23.638348 32718 solver.cpp:237]     Train net output #0: loss = 5.28151 (* 1 = 5.28151 loss)
I0407 22:24:23.638383 32718 sgd_solver.cpp:105] Iteration 0, lr = 0.00924142
I0407 22:24:27.326468 32718 solver.cpp:218] Iteration 12 (3.25371 iter/s, 3.6881s/12 iters), loss = 5.28284
I0407 22:24:27.326503 32718 solver.cpp:237]     Train net output #0: loss = 5.28284 (* 1 = 5.28284 loss)
I0407 22:24:27.326511 32718 sgd_solver.cpp:105] Iteration 12, lr = 0.00923728
I0407 22:24:32.202330 32718 solver.cpp:218] Iteration 24 (2.46114 iter/s, 4.8758s/12 iters), loss = 5.28408
I0407 22:24:32.202378 32718 solver.cpp:237]     Train net output #0: loss = 5.28408 (* 1 = 5.28408 loss)
I0407 22:24:32.202387 32718 sgd_solver.cpp:105] Iteration 24, lr = 0.00923313
I0407 22:24:37.074025 32718 solver.cpp:218] Iteration 36 (2.46324 iter/s, 4.87162s/12 iters), loss = 5.27463
I0407 22:24:37.074060 32718 solver.cpp:237]     Train net output #0: loss = 5.27463 (* 1 = 5.27463 loss)
I0407 22:24:37.074069 32718 sgd_solver.cpp:105] Iteration 36, lr = 0.00922895
I0407 22:24:41.933892 32718 solver.cpp:218] Iteration 48 (2.46923 iter/s, 4.85981s/12 iters), loss = 5.28514
I0407 22:24:41.933923 32718 solver.cpp:237]     Train net output #0: loss = 5.28514 (* 1 = 5.28514 loss)
I0407 22:24:41.933930 32718 sgd_solver.cpp:105] Iteration 48, lr = 0.00922476
I0407 22:24:46.773737 32718 solver.cpp:218] Iteration 60 (2.47944 iter/s, 4.83979s/12 iters), loss = 5.27958
I0407 22:24:46.773897 32718 solver.cpp:237]     Train net output #0: loss = 5.27958 (* 1 = 5.27958 loss)
I0407 22:24:46.773906 32718 sgd_solver.cpp:105] Iteration 60, lr = 0.00922054
I0407 22:24:51.659129 32718 solver.cpp:218] Iteration 72 (2.45639 iter/s, 4.88521s/12 iters), loss = 5.30799
I0407 22:24:51.659162 32718 solver.cpp:237]     Train net output #0: loss = 5.30799 (* 1 = 5.30799 loss)
I0407 22:24:51.659169 32718 sgd_solver.cpp:105] Iteration 72, lr = 0.0092163
I0407 22:24:56.470335 32718 solver.cpp:218] Iteration 84 (2.49421 iter/s, 4.81114s/12 iters), loss = 5.30394
I0407 22:24:56.470371 32718 solver.cpp:237]     Train net output #0: loss = 5.30394 (* 1 = 5.30394 loss)
I0407 22:24:56.470379 32718 sgd_solver.cpp:105] Iteration 84, lr = 0.00921204
I0407 22:25:01.413249 32718 solver.cpp:218] Iteration 96 (2.42775 iter/s, 4.94285s/12 iters), loss = 5.29844
I0407 22:25:01.413287 32718 solver.cpp:237]     Train net output #0: loss = 5.29844 (* 1 = 5.29844 loss)
I0407 22:25:01.413295 32718 sgd_solver.cpp:105] Iteration 96, lr = 0.00920776
I0407 22:25:03.093621 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:25:03.395242 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_102.caffemodel
I0407 22:25:06.580042 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_102.solverstate
I0407 22:25:08.940768 32718 solver.cpp:330] Iteration 102, Testing net (#0)
I0407 22:25:08.940786 32718 net.cpp:676] Ignoring source layer train-data
I0407 22:25:13.647437 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:25:13.734531 32718 solver.cpp:397]     Test net output #0: accuracy = 0.00551471
I0407 22:25:13.734575 32718 solver.cpp:397]     Test net output #1: loss = 5.28267 (* 1 = 5.28267 loss)
I0407 22:25:15.583377 32718 solver.cpp:218] Iteration 108 (0.846858 iter/s, 14.17s/12 iters), loss = 5.27993
I0407 22:25:15.583429 32718 solver.cpp:237]     Train net output #0: loss = 5.27993 (* 1 = 5.27993 loss)
I0407 22:25:15.583438 32718 sgd_solver.cpp:105] Iteration 108, lr = 0.00920346
I0407 22:25:20.529230 32718 solver.cpp:218] Iteration 120 (2.42632 iter/s, 4.94577s/12 iters), loss = 5.25569
I0407 22:25:20.529386 32718 solver.cpp:237]     Train net output #0: loss = 5.25569 (* 1 = 5.25569 loss)
I0407 22:25:20.529395 32718 sgd_solver.cpp:105] Iteration 120, lr = 0.00919914
I0407 22:25:25.475067 32718 solver.cpp:218] Iteration 132 (2.42637 iter/s, 4.94566s/12 iters), loss = 5.27161
I0407 22:25:25.475108 32718 solver.cpp:237]     Train net output #0: loss = 5.27161 (* 1 = 5.27161 loss)
I0407 22:25:25.475117 32718 sgd_solver.cpp:105] Iteration 132, lr = 0.00919479
I0407 22:25:30.485555 32718 solver.cpp:218] Iteration 144 (2.39501 iter/s, 5.01042s/12 iters), loss = 5.20478
I0407 22:25:30.485597 32718 solver.cpp:237]     Train net output #0: loss = 5.20478 (* 1 = 5.20478 loss)
I0407 22:25:30.485606 32718 sgd_solver.cpp:105] Iteration 144, lr = 0.00919043
I0407 22:25:35.626868 32718 solver.cpp:218] Iteration 156 (2.33407 iter/s, 5.14124s/12 iters), loss = 5.19318
I0407 22:25:35.626904 32718 solver.cpp:237]     Train net output #0: loss = 5.19318 (* 1 = 5.19318 loss)
I0407 22:25:35.626911 32718 sgd_solver.cpp:105] Iteration 156, lr = 0.00918604
I0407 22:25:40.659317 32718 solver.cpp:218] Iteration 168 (2.38456 iter/s, 5.03238s/12 iters), loss = 5.2191
I0407 22:25:40.659354 32718 solver.cpp:237]     Train net output #0: loss = 5.2191 (* 1 = 5.2191 loss)
I0407 22:25:40.659363 32718 sgd_solver.cpp:105] Iteration 168, lr = 0.00918163
I0407 22:25:45.628757 32718 solver.cpp:218] Iteration 180 (2.41479 iter/s, 4.96938s/12 iters), loss = 5.23434
I0407 22:25:45.628795 32718 solver.cpp:237]     Train net output #0: loss = 5.23434 (* 1 = 5.23434 loss)
I0407 22:25:45.628803 32718 sgd_solver.cpp:105] Iteration 180, lr = 0.0091772
I0407 22:25:50.596285 32718 solver.cpp:218] Iteration 192 (2.41572 iter/s, 4.96746s/12 iters), loss = 5.27376
I0407 22:25:50.596400 32718 solver.cpp:237]     Train net output #0: loss = 5.27376 (* 1 = 5.27376 loss)
I0407 22:25:50.596407 32718 sgd_solver.cpp:105] Iteration 192, lr = 0.00917275
I0407 22:25:54.409615 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:25:55.068599 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_204.caffemodel
I0407 22:25:58.716253 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_204.solverstate
I0407 22:26:01.446503 32718 solver.cpp:330] Iteration 204, Testing net (#0)
I0407 22:26:01.446521 32718 net.cpp:676] Ignoring source layer train-data
I0407 22:26:06.091583 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:26:06.227525 32718 solver.cpp:397]     Test net output #0: accuracy = 0.00796569
I0407 22:26:06.227573 32718 solver.cpp:397]     Test net output #1: loss = 5.17027 (* 1 = 5.17027 loss)
I0407 22:26:06.324076 32718 solver.cpp:218] Iteration 204 (0.762989 iter/s, 15.7276s/12 iters), loss = 5.21906
I0407 22:26:06.324115 32718 solver.cpp:237]     Train net output #0: loss = 5.21906 (* 1 = 5.21906 loss)
I0407 22:26:06.324122 32718 sgd_solver.cpp:105] Iteration 204, lr = 0.00916827
I0407 22:26:10.407316 32718 solver.cpp:218] Iteration 216 (2.93889 iter/s, 4.08317s/12 iters), loss = 5.25391
I0407 22:26:10.407352 32718 solver.cpp:237]     Train net output #0: loss = 5.25391 (* 1 = 5.25391 loss)
I0407 22:26:10.407361 32718 sgd_solver.cpp:105] Iteration 216, lr = 0.00916378
I0407 22:26:15.357630 32718 solver.cpp:218] Iteration 228 (2.42412 iter/s, 4.95025s/12 iters), loss = 5.18612
I0407 22:26:15.357666 32718 solver.cpp:237]     Train net output #0: loss = 5.18612 (* 1 = 5.18612 loss)
I0407 22:26:15.357674 32718 sgd_solver.cpp:105] Iteration 228, lr = 0.00915926
I0407 22:26:20.237766 32718 solver.cpp:218] Iteration 240 (2.45898 iter/s, 4.88007s/12 iters), loss = 5.15336
I0407 22:26:20.237805 32718 solver.cpp:237]     Train net output #0: loss = 5.15336 (* 1 = 5.15336 loss)
I0407 22:26:20.237813 32718 sgd_solver.cpp:105] Iteration 240, lr = 0.00915472
I0407 22:26:25.209236 32718 solver.cpp:218] Iteration 252 (2.41381 iter/s, 4.9714s/12 iters), loss = 5.17207
I0407 22:26:25.209373 32718 solver.cpp:237]     Train net output #0: loss = 5.17207 (* 1 = 5.17207 loss)
I0407 22:26:25.209386 32718 sgd_solver.cpp:105] Iteration 252, lr = 0.00915015
I0407 22:26:30.141425 32718 solver.cpp:218] Iteration 264 (2.43308 iter/s, 4.93203s/12 iters), loss = 5.15642
I0407 22:26:30.141463 32718 solver.cpp:237]     Train net output #0: loss = 5.15642 (* 1 = 5.15642 loss)
I0407 22:26:30.141470 32718 sgd_solver.cpp:105] Iteration 264, lr = 0.00914557
I0407 22:26:35.110734 32718 solver.cpp:218] Iteration 276 (2.41486 iter/s, 4.96924s/12 iters), loss = 5.14071
I0407 22:26:35.110778 32718 solver.cpp:237]     Train net output #0: loss = 5.14071 (* 1 = 5.14071 loss)
I0407 22:26:35.110786 32718 sgd_solver.cpp:105] Iteration 276, lr = 0.00914096
I0407 22:26:40.089454 32718 solver.cpp:218] Iteration 288 (2.4103 iter/s, 4.97864s/12 iters), loss = 5.14392
I0407 22:26:40.089499 32718 solver.cpp:237]     Train net output #0: loss = 5.14392 (* 1 = 5.14392 loss)
I0407 22:26:40.089509 32718 sgd_solver.cpp:105] Iteration 288, lr = 0.00913633
I0407 22:26:44.981642 32718 solver.cpp:218] Iteration 300 (2.45293 iter/s, 4.8921s/12 iters), loss = 5.1393
I0407 22:26:44.981696 32718 solver.cpp:237]     Train net output #0: loss = 5.1393 (* 1 = 5.1393 loss)
I0407 22:26:44.981709 32718 sgd_solver.cpp:105] Iteration 300, lr = 0.00913168
I0407 22:26:45.936774 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:26:46.952330 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_306.caffemodel
I0407 22:26:50.848235 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_306.solverstate
I0407 22:26:54.679260 32718 solver.cpp:330] Iteration 306, Testing net (#0)
I0407 22:26:54.679277 32718 net.cpp:676] Ignoring source layer train-data
I0407 22:26:59.216290 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:26:59.380146 32718 solver.cpp:397]     Test net output #0: accuracy = 0.0128676
I0407 22:26:59.380193 32718 solver.cpp:397]     Test net output #1: loss = 5.14151 (* 1 = 5.14151 loss)
I0407 22:27:01.177523 32718 solver.cpp:218] Iteration 312 (0.740935 iter/s, 16.1958s/12 iters), loss = 5.17723
I0407 22:27:01.177567 32718 solver.cpp:237]     Train net output #0: loss = 5.17723 (* 1 = 5.17723 loss)
I0407 22:27:01.177575 32718 sgd_solver.cpp:105] Iteration 312, lr = 0.009127
I0407 22:27:06.122674 32718 solver.cpp:218] Iteration 324 (2.42665 iter/s, 4.94508s/12 iters), loss = 5.11914
I0407 22:27:06.122712 32718 solver.cpp:237]     Train net output #0: loss = 5.11914 (* 1 = 5.11914 loss)
I0407 22:27:06.122720 32718 sgd_solver.cpp:105] Iteration 324, lr = 0.0091223
I0407 22:27:11.066164 32718 solver.cpp:218] Iteration 336 (2.42747 iter/s, 4.94341s/12 iters), loss = 5.16921
I0407 22:27:11.066215 32718 solver.cpp:237]     Train net output #0: loss = 5.16921 (* 1 = 5.16921 loss)
I0407 22:27:11.066227 32718 sgd_solver.cpp:105] Iteration 336, lr = 0.00911758
I0407 22:27:15.982884 32718 solver.cpp:218] Iteration 348 (2.44069 iter/s, 4.91664s/12 iters), loss = 5.14046
I0407 22:27:15.982929 32718 solver.cpp:237]     Train net output #0: loss = 5.14046 (* 1 = 5.14046 loss)
I0407 22:27:15.982939 32718 sgd_solver.cpp:105] Iteration 348, lr = 0.00911284
I0407 22:27:20.935897 32718 solver.cpp:218] Iteration 360 (2.42281 iter/s, 4.95293s/12 iters), loss = 5.09642
I0407 22:27:20.935941 32718 solver.cpp:237]     Train net output #0: loss = 5.09642 (* 1 = 5.09642 loss)
I0407 22:27:20.935950 32718 sgd_solver.cpp:105] Iteration 360, lr = 0.00910807
I0407 22:27:25.865448 32718 solver.cpp:218] Iteration 372 (2.43434 iter/s, 4.92947s/12 iters), loss = 5.21648
I0407 22:27:25.865489 32718 solver.cpp:237]     Train net output #0: loss = 5.21648 (* 1 = 5.21648 loss)
I0407 22:27:25.865496 32718 sgd_solver.cpp:105] Iteration 372, lr = 0.00910328
I0407 22:27:30.819808 32718 solver.cpp:218] Iteration 384 (2.42215 iter/s, 4.95428s/12 iters), loss = 5.12993
I0407 22:27:30.819979 32718 solver.cpp:237]     Train net output #0: loss = 5.12993 (* 1 = 5.12993 loss)
I0407 22:27:30.819990 32718 sgd_solver.cpp:105] Iteration 384, lr = 0.00909847
I0407 22:27:35.763106 32718 solver.cpp:218] Iteration 396 (2.42763 iter/s, 4.9431s/12 iters), loss = 5.18298
I0407 22:27:35.763154 32718 solver.cpp:237]     Train net output #0: loss = 5.18298 (* 1 = 5.18298 loss)
I0407 22:27:35.763164 32718 sgd_solver.cpp:105] Iteration 396, lr = 0.00909363
I0407 22:27:38.848598 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:27:40.222198 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_408.caffemodel
I0407 22:27:43.311239 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_408.solverstate
I0407 22:27:45.888810 32718 solver.cpp:330] Iteration 408, Testing net (#0)
I0407 22:27:45.888828 32718 net.cpp:676] Ignoring source layer train-data
I0407 22:27:50.503528 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:27:50.726682 32718 solver.cpp:397]     Test net output #0: accuracy = 0.0159314
I0407 22:27:50.726713 32718 solver.cpp:397]     Test net output #1: loss = 5.08323 (* 1 = 5.08323 loss)
I0407 22:27:50.823464 32718 solver.cpp:218] Iteration 408 (0.7968 iter/s, 15.0602s/12 iters), loss = 5.10292
I0407 22:27:50.823529 32718 solver.cpp:237]     Train net output #0: loss = 5.10292 (* 1 = 5.10292 loss)
I0407 22:27:50.823541 32718 sgd_solver.cpp:105] Iteration 408, lr = 0.00908877
I0407 22:27:54.985101 32718 solver.cpp:218] Iteration 420 (2.88354 iter/s, 4.16155s/12 iters), loss = 4.97895
I0407 22:27:54.985132 32718 solver.cpp:237]     Train net output #0: loss = 4.97895 (* 1 = 4.97895 loss)
I0407 22:27:54.985139 32718 sgd_solver.cpp:105] Iteration 420, lr = 0.00908389
I0407 22:27:59.961405 32718 solver.cpp:218] Iteration 432 (2.41146 iter/s, 4.97624s/12 iters), loss = 5.11216
I0407 22:27:59.961439 32718 solver.cpp:237]     Train net output #0: loss = 5.11216 (* 1 = 5.11216 loss)
I0407 22:27:59.961447 32718 sgd_solver.cpp:105] Iteration 432, lr = 0.00907898
I0407 22:28:04.903988 32718 solver.cpp:218] Iteration 444 (2.42791 iter/s, 4.94252s/12 iters), loss = 5.03089
I0407 22:28:04.904109 32718 solver.cpp:237]     Train net output #0: loss = 5.03089 (* 1 = 5.03089 loss)
I0407 22:28:04.904119 32718 sgd_solver.cpp:105] Iteration 444, lr = 0.00907405
I0407 22:28:09.870734 32718 solver.cpp:218] Iteration 456 (2.41614 iter/s, 4.9666s/12 iters), loss = 5.08423
I0407 22:28:09.870770 32718 solver.cpp:237]     Train net output #0: loss = 5.08423 (* 1 = 5.08423 loss)
I0407 22:28:09.870779 32718 sgd_solver.cpp:105] Iteration 456, lr = 0.0090691
I0407 22:28:14.820003 32718 solver.cpp:218] Iteration 468 (2.42463 iter/s, 4.9492s/12 iters), loss = 5.17968
I0407 22:28:14.820041 32718 solver.cpp:237]     Train net output #0: loss = 5.17968 (* 1 = 5.17968 loss)
I0407 22:28:14.820048 32718 sgd_solver.cpp:105] Iteration 468, lr = 0.00906412
I0407 22:28:19.729722 32718 solver.cpp:218] Iteration 480 (2.44417 iter/s, 4.90965s/12 iters), loss = 5.11416
I0407 22:28:19.729756 32718 solver.cpp:237]     Train net output #0: loss = 5.11416 (* 1 = 5.11416 loss)
I0407 22:28:19.729763 32718 sgd_solver.cpp:105] Iteration 480, lr = 0.00905912
I0407 22:28:24.707805 32718 solver.cpp:218] Iteration 492 (2.4106 iter/s, 4.97802s/12 iters), loss = 5.04651
I0407 22:28:24.707841 32718 solver.cpp:237]     Train net output #0: loss = 5.04651 (* 1 = 5.04651 loss)
I0407 22:28:24.707850 32718 sgd_solver.cpp:105] Iteration 492, lr = 0.00905409
I0407 22:28:29.665148 32718 solver.cpp:218] Iteration 504 (2.42069 iter/s, 4.95727s/12 iters), loss = 5.08827
I0407 22:28:29.665194 32718 solver.cpp:237]     Train net output #0: loss = 5.08827 (* 1 = 5.08827 loss)
I0407 22:28:29.665202 32718 sgd_solver.cpp:105] Iteration 504, lr = 0.00904904
I0407 22:28:29.902094 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:28:31.673784 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_510.caffemodel
I0407 22:28:34.796123 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_510.solverstate
I0407 22:28:37.168210 32718 solver.cpp:330] Iteration 510, Testing net (#0)
I0407 22:28:37.168334 32718 net.cpp:676] Ignoring source layer train-data
I0407 22:28:41.685386 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:28:41.944061 32718 solver.cpp:397]     Test net output #0: accuracy = 0.0171569
I0407 22:28:41.944103 32718 solver.cpp:397]     Test net output #1: loss = 5.04099 (* 1 = 5.04099 loss)
I0407 22:28:43.763214 32718 solver.cpp:218] Iteration 516 (0.851187 iter/s, 14.098s/12 iters), loss = 5.06556
I0407 22:28:43.763252 32718 solver.cpp:237]     Train net output #0: loss = 5.06556 (* 1 = 5.06556 loss)
I0407 22:28:43.763259 32718 sgd_solver.cpp:105] Iteration 516, lr = 0.00904397
I0407 22:28:48.978235 32718 solver.cpp:218] Iteration 528 (2.30108 iter/s, 5.21494s/12 iters), loss = 5.03872
I0407 22:28:48.978274 32718 solver.cpp:237]     Train net output #0: loss = 5.03872 (* 1 = 5.03872 loss)
I0407 22:28:48.978282 32718 sgd_solver.cpp:105] Iteration 528, lr = 0.00903887
I0407 22:28:53.905882 32718 solver.cpp:218] Iteration 540 (2.43528 iter/s, 4.92757s/12 iters), loss = 4.93665
I0407 22:28:53.905920 32718 solver.cpp:237]     Train net output #0: loss = 4.93665 (* 1 = 4.93665 loss)
I0407 22:28:53.905928 32718 sgd_solver.cpp:105] Iteration 540, lr = 0.00903374
I0407 22:28:58.838658 32718 solver.cpp:218] Iteration 552 (2.43274 iter/s, 4.9327s/12 iters), loss = 4.92024
I0407 22:28:58.838699 32718 solver.cpp:237]     Train net output #0: loss = 4.92024 (* 1 = 4.92024 loss)
I0407 22:28:58.838708 32718 sgd_solver.cpp:105] Iteration 552, lr = 0.0090286
I0407 22:29:03.792587 32718 solver.cpp:218] Iteration 564 (2.42236 iter/s, 4.95385s/12 iters), loss = 4.97967
I0407 22:29:03.792634 32718 solver.cpp:237]     Train net output #0: loss = 4.97967 (* 1 = 4.97967 loss)
I0407 22:29:03.792641 32718 sgd_solver.cpp:105] Iteration 564, lr = 0.00902343
I0407 22:29:08.748046 32718 solver.cpp:218] Iteration 576 (2.42161 iter/s, 4.95538s/12 iters), loss = 5.11028
I0407 22:29:08.748200 32718 solver.cpp:237]     Train net output #0: loss = 5.11028 (* 1 = 5.11028 loss)
I0407 22:29:08.748216 32718 sgd_solver.cpp:105] Iteration 576, lr = 0.00901823
I0407 22:29:13.680965 32718 solver.cpp:218] Iteration 588 (2.43272 iter/s, 4.93275s/12 iters), loss = 4.97683
I0407 22:29:13.681002 32718 solver.cpp:237]     Train net output #0: loss = 4.97683 (* 1 = 4.97683 loss)
I0407 22:29:13.681011 32718 sgd_solver.cpp:105] Iteration 588, lr = 0.00901301
I0407 22:29:18.639714 32718 solver.cpp:218] Iteration 600 (2.42 iter/s, 4.95868s/12 iters), loss = 5.03846
I0407 22:29:18.639750 32718 solver.cpp:237]     Train net output #0: loss = 5.03846 (* 1 = 5.03846 loss)
I0407 22:29:18.639757 32718 sgd_solver.cpp:105] Iteration 600, lr = 0.00900776
I0407 22:29:20.989476 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:29:23.101315 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_612.caffemodel
I0407 22:29:26.194169 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_612.solverstate
I0407 22:29:28.567265 32718 solver.cpp:330] Iteration 612, Testing net (#0)
I0407 22:29:28.567282 32718 net.cpp:676] Ignoring source layer train-data
I0407 22:29:32.814745 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:29:33.102975 32718 solver.cpp:397]     Test net output #0: accuracy = 0.0238971
I0407 22:29:33.103022 32718 solver.cpp:397]     Test net output #1: loss = 4.96967 (* 1 = 4.96967 loss)
I0407 22:29:33.197422 32718 solver.cpp:218] Iteration 612 (0.824312 iter/s, 14.5576s/12 iters), loss = 4.88951
I0407 22:29:33.197481 32718 solver.cpp:237]     Train net output #0: loss = 4.88951 (* 1 = 4.88951 loss)
I0407 22:29:33.197492 32718 sgd_solver.cpp:105] Iteration 612, lr = 0.00900249
I0407 22:29:37.286150 32718 solver.cpp:218] Iteration 624 (2.93496 iter/s, 4.08865s/12 iters), loss = 5.07539
I0407 22:29:37.286185 32718 solver.cpp:237]     Train net output #0: loss = 5.07539 (* 1 = 5.07539 loss)
I0407 22:29:37.286192 32718 sgd_solver.cpp:105] Iteration 624, lr = 0.0089972
I0407 22:29:42.217469 32718 solver.cpp:218] Iteration 636 (2.43346 iter/s, 4.93125s/12 iters), loss = 4.95142
I0407 22:29:42.217617 32718 solver.cpp:237]     Train net output #0: loss = 4.95142 (* 1 = 4.95142 loss)
I0407 22:29:42.217626 32718 sgd_solver.cpp:105] Iteration 636, lr = 0.00899188
I0407 22:29:47.165823 32718 solver.cpp:218] Iteration 648 (2.42514 iter/s, 4.94817s/12 iters), loss = 4.97638
I0407 22:29:47.165865 32718 solver.cpp:237]     Train net output #0: loss = 4.97638 (* 1 = 4.97638 loss)
I0407 22:29:47.165874 32718 sgd_solver.cpp:105] Iteration 648, lr = 0.00898654
I0407 22:29:52.121062 32718 solver.cpp:218] Iteration 660 (2.42172 iter/s, 4.95516s/12 iters), loss = 4.91372
I0407 22:29:52.121105 32718 solver.cpp:237]     Train net output #0: loss = 4.91372 (* 1 = 4.91372 loss)
I0407 22:29:52.121114 32718 sgd_solver.cpp:105] Iteration 660, lr = 0.00898117
I0407 22:29:57.103801 32718 solver.cpp:218] Iteration 672 (2.40835 iter/s, 4.98266s/12 iters), loss = 5.07359
I0407 22:29:57.103842 32718 solver.cpp:237]     Train net output #0: loss = 5.07359 (* 1 = 5.07359 loss)
I0407 22:29:57.103850 32718 sgd_solver.cpp:105] Iteration 672, lr = 0.00897577
I0407 22:30:02.064760 32718 solver.cpp:218] Iteration 684 (2.41892 iter/s, 4.96089s/12 iters), loss = 4.95895
I0407 22:30:02.064795 32718 solver.cpp:237]     Train net output #0: loss = 4.95895 (* 1 = 4.95895 loss)
I0407 22:30:02.064802 32718 sgd_solver.cpp:105] Iteration 684, lr = 0.00897035
I0407 22:30:02.838140 32718 blocking_queue.cpp:49] Waiting for data
I0407 22:30:07.018224 32718 solver.cpp:218] Iteration 696 (2.42258 iter/s, 4.95339s/12 iters), loss = 4.87447
I0407 22:30:07.018268 32718 solver.cpp:237]     Train net output #0: loss = 4.87447 (* 1 = 4.87447 loss)
I0407 22:30:07.018276 32718 sgd_solver.cpp:105] Iteration 696, lr = 0.0089649
I0407 22:30:11.589746 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:30:11.961447 32718 solver.cpp:218] Iteration 708 (2.4276 iter/s, 4.94314s/12 iters), loss = 4.96902
I0407 22:30:11.961484 32718 solver.cpp:237]     Train net output #0: loss = 4.96902 (* 1 = 4.96902 loss)
I0407 22:30:11.961491 32718 sgd_solver.cpp:105] Iteration 708, lr = 0.00895943
I0407 22:30:13.956066 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_714.caffemodel
I0407 22:30:17.058223 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_714.solverstate
I0407 22:30:19.418498 32718 solver.cpp:330] Iteration 714, Testing net (#0)
I0407 22:30:19.418515 32718 net.cpp:676] Ignoring source layer train-data
I0407 22:30:23.767225 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:30:24.088006 32718 solver.cpp:397]     Test net output #0: accuracy = 0.0373775
I0407 22:30:24.088042 32718 solver.cpp:397]     Test net output #1: loss = 4.88481 (* 1 = 4.88481 loss)
I0407 22:30:25.924598 32718 solver.cpp:218] Iteration 720 (0.859411 iter/s, 13.9631s/12 iters), loss = 4.92455
I0407 22:30:25.924633 32718 solver.cpp:237]     Train net output #0: loss = 4.92455 (* 1 = 4.92455 loss)
I0407 22:30:25.924640 32718 sgd_solver.cpp:105] Iteration 720, lr = 0.00895394
I0407 22:30:30.948073 32718 solver.cpp:218] Iteration 732 (2.38882 iter/s, 5.0234s/12 iters), loss = 4.9177
I0407 22:30:30.948118 32718 solver.cpp:237]     Train net output #0: loss = 4.9177 (* 1 = 4.9177 loss)
I0407 22:30:30.948127 32718 sgd_solver.cpp:105] Iteration 732, lr = 0.00894841
I0407 22:30:35.885535 32718 solver.cpp:218] Iteration 744 (2.43043 iter/s, 4.93739s/12 iters), loss = 4.87621
I0407 22:30:35.885565 32718 solver.cpp:237]     Train net output #0: loss = 4.87621 (* 1 = 4.87621 loss)
I0407 22:30:35.885572 32718 sgd_solver.cpp:105] Iteration 744, lr = 0.00894287
I0407 22:30:40.911123 32718 solver.cpp:218] Iteration 756 (2.38781 iter/s, 5.02552s/12 iters), loss = 4.72471
I0407 22:30:40.911162 32718 solver.cpp:237]     Train net output #0: loss = 4.72471 (* 1 = 4.72471 loss)
I0407 22:30:40.911170 32718 sgd_solver.cpp:105] Iteration 756, lr = 0.00893729
I0407 22:30:45.838423 32718 solver.cpp:218] Iteration 768 (2.43545 iter/s, 4.92723s/12 iters), loss = 4.77262
I0407 22:30:45.838572 32718 solver.cpp:237]     Train net output #0: loss = 4.77262 (* 1 = 4.77262 loss)
I0407 22:30:45.838582 32718 sgd_solver.cpp:105] Iteration 768, lr = 0.00893169
I0407 22:30:50.832401 32718 solver.cpp:218] Iteration 780 (2.40298 iter/s, 4.9938s/12 iters), loss = 4.75694
I0407 22:30:50.832442 32718 solver.cpp:237]     Train net output #0: loss = 4.75694 (* 1 = 4.75694 loss)
I0407 22:30:50.832449 32718 sgd_solver.cpp:105] Iteration 780, lr = 0.00892607
I0407 22:30:55.710747 32718 solver.cpp:218] Iteration 792 (2.45989 iter/s, 4.87827s/12 iters), loss = 4.9219
I0407 22:30:55.710789 32718 solver.cpp:237]     Train net output #0: loss = 4.9219 (* 1 = 4.9219 loss)
I0407 22:30:55.710798 32718 sgd_solver.cpp:105] Iteration 792, lr = 0.00892041
I0407 22:31:00.659060 32718 solver.cpp:218] Iteration 804 (2.4251 iter/s, 4.94824s/12 iters), loss = 4.72537
I0407 22:31:00.659093 32718 solver.cpp:237]     Train net output #0: loss = 4.72537 (* 1 = 4.72537 loss)
I0407 22:31:00.659101 32718 sgd_solver.cpp:105] Iteration 804, lr = 0.00891474
I0407 22:31:02.368046 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:31:05.103698 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_816.caffemodel
I0407 22:31:08.604208 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_816.solverstate
I0407 22:31:12.030726 32718 solver.cpp:330] Iteration 816, Testing net (#0)
I0407 22:31:12.030745 32718 net.cpp:676] Ignoring source layer train-data
I0407 22:31:16.312139 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:31:16.670011 32718 solver.cpp:397]     Test net output #0: accuracy = 0.0422794
I0407 22:31:16.670058 32718 solver.cpp:397]     Test net output #1: loss = 4.76732 (* 1 = 4.76732 loss)
I0407 22:31:16.767180 32718 solver.cpp:218] Iteration 816 (0.744971 iter/s, 16.108s/12 iters), loss = 4.66682
I0407 22:31:16.767259 32718 solver.cpp:237]     Train net output #0: loss = 4.66682 (* 1 = 4.66682 loss)
I0407 22:31:16.767274 32718 sgd_solver.cpp:105] Iteration 816, lr = 0.00890903
I0407 22:31:20.919405 32718 solver.cpp:218] Iteration 828 (2.89009 iter/s, 4.15212s/12 iters), loss = 4.67913
I0407 22:31:20.919445 32718 solver.cpp:237]     Train net output #0: loss = 4.67913 (* 1 = 4.67913 loss)
I0407 22:31:20.919453 32718 sgd_solver.cpp:105] Iteration 828, lr = 0.0089033
I0407 22:31:25.834317 32718 solver.cpp:218] Iteration 840 (2.44159 iter/s, 4.91484s/12 iters), loss = 4.70798
I0407 22:31:25.834362 32718 solver.cpp:237]     Train net output #0: loss = 4.70798 (* 1 = 4.70798 loss)
I0407 22:31:25.834370 32718 sgd_solver.cpp:105] Iteration 840, lr = 0.00889754
I0407 22:31:30.791115 32718 solver.cpp:218] Iteration 852 (2.42096 iter/s, 4.95672s/12 iters), loss = 4.66097
I0407 22:31:30.791155 32718 solver.cpp:237]     Train net output #0: loss = 4.66097 (* 1 = 4.66097 loss)
I0407 22:31:30.791163 32718 sgd_solver.cpp:105] Iteration 852, lr = 0.00889176
I0407 22:31:35.789028 32718 solver.cpp:218] Iteration 864 (2.40104 iter/s, 4.99784s/12 iters), loss = 4.70034
I0407 22:31:35.789064 32718 solver.cpp:237]     Train net output #0: loss = 4.70034 (* 1 = 4.70034 loss)
I0407 22:31:35.789072 32718 sgd_solver.cpp:105] Iteration 864, lr = 0.00888595
I0407 22:31:40.777957 32718 solver.cpp:218] Iteration 876 (2.40536 iter/s, 4.98886s/12 iters), loss = 4.72219
I0407 22:31:40.777993 32718 solver.cpp:237]     Train net output #0: loss = 4.72219 (* 1 = 4.72219 loss)
I0407 22:31:40.778000 32718 sgd_solver.cpp:105] Iteration 876, lr = 0.00888011
I0407 22:31:45.727178 32718 solver.cpp:218] Iteration 888 (2.42466 iter/s, 4.94915s/12 iters), loss = 4.78573
I0407 22:31:45.727222 32718 solver.cpp:237]     Train net output #0: loss = 4.78573 (* 1 = 4.78573 loss)
I0407 22:31:45.727231 32718 sgd_solver.cpp:105] Iteration 888, lr = 0.00887425
I0407 22:31:50.702023 32718 solver.cpp:218] Iteration 900 (2.41217 iter/s, 4.97476s/12 iters), loss = 4.63705
I0407 22:31:50.702199 32718 solver.cpp:237]     Train net output #0: loss = 4.63705 (* 1 = 4.63705 loss)
I0407 22:31:50.702210 32718 sgd_solver.cpp:105] Iteration 900, lr = 0.00886836
I0407 22:31:54.533360 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:31:55.624693 32718 solver.cpp:218] Iteration 912 (2.43781 iter/s, 4.92246s/12 iters), loss = 4.71126
I0407 22:31:55.624750 32718 solver.cpp:237]     Train net output #0: loss = 4.71126 (* 1 = 4.71126 loss)
I0407 22:31:55.624761 32718 sgd_solver.cpp:105] Iteration 912, lr = 0.00886244
I0407 22:31:57.634676 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_918.caffemodel
I0407 22:32:00.708396 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_918.solverstate
I0407 22:32:03.062122 32718 solver.cpp:330] Iteration 918, Testing net (#0)
I0407 22:32:03.062140 32718 net.cpp:676] Ignoring source layer train-data
I0407 22:32:07.493194 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:32:07.943686 32718 solver.cpp:397]     Test net output #0: accuracy = 0.0502451
I0407 22:32:07.943722 32718 solver.cpp:397]     Test net output #1: loss = 4.62058 (* 1 = 4.62058 loss)
I0407 22:32:09.751595 32718 solver.cpp:218] Iteration 924 (0.84945 iter/s, 14.1268s/12 iters), loss = 4.53119
I0407 22:32:09.751642 32718 solver.cpp:237]     Train net output #0: loss = 4.53119 (* 1 = 4.53119 loss)
I0407 22:32:09.751652 32718 sgd_solver.cpp:105] Iteration 924, lr = 0.0088565
I0407 22:32:14.689116 32718 solver.cpp:218] Iteration 936 (2.43041 iter/s, 4.93744s/12 iters), loss = 4.48044
I0407 22:32:14.689155 32718 solver.cpp:237]     Train net output #0: loss = 4.48044 (* 1 = 4.48044 loss)
I0407 22:32:14.689162 32718 sgd_solver.cpp:105] Iteration 936, lr = 0.00885053
I0407 22:32:19.634963 32718 solver.cpp:218] Iteration 948 (2.42631 iter/s, 4.94578s/12 iters), loss = 4.6518
I0407 22:32:19.634999 32718 solver.cpp:237]     Train net output #0: loss = 4.6518 (* 1 = 4.6518 loss)
I0407 22:32:19.635005 32718 sgd_solver.cpp:105] Iteration 948, lr = 0.00884453
I0407 22:32:24.580896 32718 solver.cpp:218] Iteration 960 (2.42627 iter/s, 4.94587s/12 iters), loss = 4.4934
I0407 22:32:24.581023 32718 solver.cpp:237]     Train net output #0: loss = 4.4934 (* 1 = 4.4934 loss)
I0407 22:32:24.581032 32718 sgd_solver.cpp:105] Iteration 960, lr = 0.00883851
I0407 22:32:29.538648 32718 solver.cpp:218] Iteration 972 (2.42053 iter/s, 4.95759s/12 iters), loss = 4.53482
I0407 22:32:29.538683 32718 solver.cpp:237]     Train net output #0: loss = 4.53482 (* 1 = 4.53482 loss)
I0407 22:32:29.538691 32718 sgd_solver.cpp:105] Iteration 972, lr = 0.00883245
I0407 22:32:34.459743 32718 solver.cpp:218] Iteration 984 (2.43852 iter/s, 4.92103s/12 iters), loss = 4.41537
I0407 22:32:34.459780 32718 solver.cpp:237]     Train net output #0: loss = 4.41537 (* 1 = 4.41537 loss)
I0407 22:32:34.459789 32718 sgd_solver.cpp:105] Iteration 984, lr = 0.00882638
I0407 22:32:39.427188 32718 solver.cpp:218] Iteration 996 (2.41576 iter/s, 4.96738s/12 iters), loss = 4.60143
I0407 22:32:39.427233 32718 solver.cpp:237]     Train net output #0: loss = 4.60143 (* 1 = 4.60143 loss)
I0407 22:32:39.427242 32718 sgd_solver.cpp:105] Iteration 996, lr = 0.00882027
I0407 22:32:44.373071 32718 solver.cpp:218] Iteration 1008 (2.4263 iter/s, 4.94581s/12 iters), loss = 4.31286
I0407 22:32:44.373111 32718 solver.cpp:237]     Train net output #0: loss = 4.31286 (* 1 = 4.31286 loss)
I0407 22:32:44.373118 32718 sgd_solver.cpp:105] Iteration 1008, lr = 0.00881413
I0407 22:32:45.383245 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:32:48.844039 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_1020.caffemodel
I0407 22:32:54.890285 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_1020.solverstate
I0407 22:32:57.802506 32718 solver.cpp:330] Iteration 1020, Testing net (#0)
I0407 22:32:57.802522 32718 net.cpp:676] Ignoring source layer train-data
I0407 22:33:02.215684 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:33:02.701782 32718 solver.cpp:397]     Test net output #0: accuracy = 0.0637255
I0407 22:33:02.701817 32718 solver.cpp:397]     Test net output #1: loss = 4.47732 (* 1 = 4.47732 loss)
I0407 22:33:02.798458 32718 solver.cpp:218] Iteration 1020 (0.65128 iter/s, 18.4253s/12 iters), loss = 4.454
I0407 22:33:02.798504 32718 solver.cpp:237]     Train net output #0: loss = 4.454 (* 1 = 4.454 loss)
I0407 22:33:02.798513 32718 sgd_solver.cpp:105] Iteration 1020, lr = 0.00880797
I0407 22:33:06.957454 32718 solver.cpp:218] Iteration 1032 (2.88537 iter/s, 4.15892s/12 iters), loss = 4.24542
I0407 22:33:06.957494 32718 solver.cpp:237]     Train net output #0: loss = 4.24542 (* 1 = 4.24542 loss)
I0407 22:33:06.957502 32718 sgd_solver.cpp:105] Iteration 1032, lr = 0.00880178
I0407 22:33:11.864286 32718 solver.cpp:218] Iteration 1044 (2.44561 iter/s, 4.90676s/12 iters), loss = 4.35913
I0407 22:33:11.864322 32718 solver.cpp:237]     Train net output #0: loss = 4.35913 (* 1 = 4.35913 loss)
I0407 22:33:11.864329 32718 sgd_solver.cpp:105] Iteration 1044, lr = 0.00879556
I0407 22:33:16.821100 32718 solver.cpp:218] Iteration 1056 (2.42094 iter/s, 4.95674s/12 iters), loss = 4.29029
I0407 22:33:16.821141 32718 solver.cpp:237]     Train net output #0: loss = 4.29029 (* 1 = 4.29029 loss)
I0407 22:33:16.821149 32718 sgd_solver.cpp:105] Iteration 1056, lr = 0.00878932
I0407 22:33:21.757622 32718 solver.cpp:218] Iteration 1068 (2.4309 iter/s, 4.93645s/12 iters), loss = 4.24287
I0407 22:33:21.757663 32718 solver.cpp:237]     Train net output #0: loss = 4.24287 (* 1 = 4.24287 loss)
I0407 22:33:21.757671 32718 sgd_solver.cpp:105] Iteration 1068, lr = 0.00878304
I0407 22:33:26.720907 32718 solver.cpp:218] Iteration 1080 (2.41779 iter/s, 4.96321s/12 iters), loss = 4.49787
I0407 22:33:26.721024 32718 solver.cpp:237]     Train net output #0: loss = 4.49787 (* 1 = 4.49787 loss)
I0407 22:33:26.721031 32718 sgd_solver.cpp:105] Iteration 1080, lr = 0.00877674
I0407 22:33:31.710100 32718 solver.cpp:218] Iteration 1092 (2.40527 iter/s, 4.98905s/12 iters), loss = 4.2338
I0407 22:33:31.710145 32718 solver.cpp:237]     Train net output #0: loss = 4.2338 (* 1 = 4.2338 loss)
I0407 22:33:31.710152 32718 sgd_solver.cpp:105] Iteration 1092, lr = 0.00877041
I0407 22:33:36.630374 32718 solver.cpp:218] Iteration 1104 (2.43893 iter/s, 4.9202s/12 iters), loss = 4.51252
I0407 22:33:36.630409 32718 solver.cpp:237]     Train net output #0: loss = 4.51252 (* 1 = 4.51252 loss)
I0407 22:33:36.630415 32718 sgd_solver.cpp:105] Iteration 1104, lr = 0.00876406
I0407 22:33:39.748641 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:33:41.571635 32718 solver.cpp:218] Iteration 1116 (2.42856 iter/s, 4.94119s/12 iters), loss = 4.19061
I0407 22:33:41.571677 32718 solver.cpp:237]     Train net output #0: loss = 4.19061 (* 1 = 4.19061 loss)
I0407 22:33:41.571686 32718 sgd_solver.cpp:105] Iteration 1116, lr = 0.00875767
I0407 22:33:43.574613 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_1122.caffemodel
I0407 22:33:47.353632 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_1122.solverstate
I0407 22:33:50.571671 32718 solver.cpp:330] Iteration 1122, Testing net (#0)
I0407 22:33:50.571687 32718 net.cpp:676] Ignoring source layer train-data
I0407 22:33:54.725750 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:33:55.213781 32718 solver.cpp:397]     Test net output #0: accuracy = 0.0833333
I0407 22:33:55.213809 32718 solver.cpp:397]     Test net output #1: loss = 4.30915 (* 1 = 4.30915 loss)
I0407 22:33:57.018309 32718 solver.cpp:218] Iteration 1128 (0.776872 iter/s, 15.4466s/12 iters), loss = 4.00429
I0407 22:33:57.018469 32718 solver.cpp:237]     Train net output #0: loss = 4.00429 (* 1 = 4.00429 loss)
I0407 22:33:57.018479 32718 sgd_solver.cpp:105] Iteration 1128, lr = 0.00875126
I0407 22:34:01.885866 32718 solver.cpp:218] Iteration 1140 (2.4654 iter/s, 4.86737s/12 iters), loss = 4.53239
I0407 22:34:01.885906 32718 solver.cpp:237]     Train net output #0: loss = 4.53239 (* 1 = 4.53239 loss)
I0407 22:34:01.885915 32718 sgd_solver.cpp:105] Iteration 1140, lr = 0.00874481
I0407 22:34:06.744552 32718 solver.cpp:218] Iteration 1152 (2.46984 iter/s, 4.85861s/12 iters), loss = 4.23879
I0407 22:34:06.744591 32718 solver.cpp:237]     Train net output #0: loss = 4.23879 (* 1 = 4.23879 loss)
I0407 22:34:06.744599 32718 sgd_solver.cpp:105] Iteration 1152, lr = 0.00873834
I0407 22:34:11.709749 32718 solver.cpp:218] Iteration 1164 (2.41686 iter/s, 4.96513s/12 iters), loss = 4.05462
I0407 22:34:11.709791 32718 solver.cpp:237]     Train net output #0: loss = 4.05462 (* 1 = 4.05462 loss)
I0407 22:34:11.709800 32718 sgd_solver.cpp:105] Iteration 1164, lr = 0.00873184
I0407 22:34:16.635529 32718 solver.cpp:218] Iteration 1176 (2.4362 iter/s, 4.9257s/12 iters), loss = 4.46448
I0407 22:34:16.635573 32718 solver.cpp:237]     Train net output #0: loss = 4.46448 (* 1 = 4.46448 loss)
I0407 22:34:16.635581 32718 sgd_solver.cpp:105] Iteration 1176, lr = 0.00872531
I0407 22:34:21.594543 32718 solver.cpp:218] Iteration 1188 (2.41987 iter/s, 4.95894s/12 iters), loss = 4.05199
I0407 22:34:21.594588 32718 solver.cpp:237]     Train net output #0: loss = 4.05199 (* 1 = 4.05199 loss)
I0407 22:34:21.594596 32718 sgd_solver.cpp:105] Iteration 1188, lr = 0.00871876
I0407 22:34:26.540733 32718 solver.cpp:218] Iteration 1200 (2.42615 iter/s, 4.94612s/12 iters), loss = 4.23265
I0407 22:34:26.540771 32718 solver.cpp:237]     Train net output #0: loss = 4.23265 (* 1 = 4.23265 loss)
I0407 22:34:26.540779 32718 sgd_solver.cpp:105] Iteration 1200, lr = 0.00871217
I0407 22:34:31.512609 32718 solver.cpp:218] Iteration 1212 (2.41361 iter/s, 4.97181s/12 iters), loss = 4.04413
I0407 22:34:31.512722 32718 solver.cpp:237]     Train net output #0: loss = 4.04413 (* 1 = 4.04413 loss)
I0407 22:34:31.512730 32718 sgd_solver.cpp:105] Iteration 1212, lr = 0.00870556
I0407 22:34:31.777479 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:34:35.976207 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_1224.caffemodel
I0407 22:34:40.415661 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_1224.solverstate
I0407 22:34:44.576165 32718 solver.cpp:330] Iteration 1224, Testing net (#0)
I0407 22:34:44.576184 32718 net.cpp:676] Ignoring source layer train-data
I0407 22:34:48.534942 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:34:49.060077 32718 solver.cpp:397]     Test net output #0: accuracy = 0.0906863
I0407 22:34:49.060123 32718 solver.cpp:397]     Test net output #1: loss = 4.13307 (* 1 = 4.13307 loss)
I0407 22:34:49.156730 32718 solver.cpp:218] Iteration 1224 (0.68012 iter/s, 17.6439s/12 iters), loss = 4.21512
I0407 22:34:49.156772 32718 solver.cpp:237]     Train net output #0: loss = 4.21512 (* 1 = 4.21512 loss)
I0407 22:34:49.156780 32718 sgd_solver.cpp:105] Iteration 1224, lr = 0.00869892
I0407 22:34:53.220782 32718 solver.cpp:218] Iteration 1236 (2.95277 iter/s, 4.06398s/12 iters), loss = 4.17272
I0407 22:34:53.220819 32718 solver.cpp:237]     Train net output #0: loss = 4.17272 (* 1 = 4.17272 loss)
I0407 22:34:53.220827 32718 sgd_solver.cpp:105] Iteration 1236, lr = 0.00869224
I0407 22:34:58.171124 32718 solver.cpp:218] Iteration 1248 (2.42411 iter/s, 4.95027s/12 iters), loss = 4.02225
I0407 22:34:58.171166 32718 solver.cpp:237]     Train net output #0: loss = 4.02225 (* 1 = 4.02225 loss)
I0407 22:34:58.171175 32718 sgd_solver.cpp:105] Iteration 1248, lr = 0.00868554
I0407 22:35:03.117488 32718 solver.cpp:218] Iteration 1260 (2.42606 iter/s, 4.94629s/12 iters), loss = 3.89349
I0407 22:35:03.117645 32718 solver.cpp:237]     Train net output #0: loss = 3.89349 (* 1 = 3.89349 loss)
I0407 22:35:03.117653 32718 sgd_solver.cpp:105] Iteration 1260, lr = 0.00867881
I0407 22:35:08.079311 32718 solver.cpp:218] Iteration 1272 (2.41856 iter/s, 4.96164s/12 iters), loss = 3.91077
I0407 22:35:08.079355 32718 solver.cpp:237]     Train net output #0: loss = 3.91077 (* 1 = 3.91077 loss)
I0407 22:35:08.079365 32718 sgd_solver.cpp:105] Iteration 1272, lr = 0.00867205
I0407 22:35:13.026973 32718 solver.cpp:218] Iteration 1284 (2.42543 iter/s, 4.94758s/12 iters), loss = 4.23755
I0407 22:35:13.027017 32718 solver.cpp:237]     Train net output #0: loss = 4.23755 (* 1 = 4.23755 loss)
I0407 22:35:13.027025 32718 sgd_solver.cpp:105] Iteration 1284, lr = 0.00866526
I0407 22:35:17.993613 32718 solver.cpp:218] Iteration 1296 (2.41616 iter/s, 4.96656s/12 iters), loss = 4.1147
I0407 22:35:17.993657 32718 solver.cpp:237]     Train net output #0: loss = 4.1147 (* 1 = 4.1147 loss)
I0407 22:35:17.993667 32718 sgd_solver.cpp:105] Iteration 1296, lr = 0.00865845
I0407 22:35:22.929260 32718 solver.cpp:218] Iteration 1308 (2.43133 iter/s, 4.93557s/12 iters), loss = 3.96399
I0407 22:35:22.929299 32718 solver.cpp:237]     Train net output #0: loss = 3.96399 (* 1 = 3.96399 loss)
I0407 22:35:22.929307 32718 sgd_solver.cpp:105] Iteration 1308, lr = 0.0086516
I0407 22:35:25.403167 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:35:27.855942 32718 solver.cpp:218] Iteration 1320 (2.43575 iter/s, 4.92661s/12 iters), loss = 4.15645
I0407 22:35:27.855983 32718 solver.cpp:237]     Train net output #0: loss = 4.15645 (* 1 = 4.15645 loss)
I0407 22:35:27.855991 32718 sgd_solver.cpp:105] Iteration 1320, lr = 0.00864472
I0407 22:35:29.872501 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_1326.caffemodel
I0407 22:35:32.937275 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_1326.solverstate
I0407 22:35:35.965188 32718 solver.cpp:330] Iteration 1326, Testing net (#0)
I0407 22:35:35.965291 32718 net.cpp:676] Ignoring source layer train-data
I0407 22:35:39.824936 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:35:40.392226 32718 solver.cpp:397]     Test net output #0: accuracy = 0.118873
I0407 22:35:40.392258 32718 solver.cpp:397]     Test net output #1: loss = 4.00358 (* 1 = 4.00358 loss)
I0407 22:35:42.194470 32718 solver.cpp:218] Iteration 1332 (0.836912 iter/s, 14.3384s/12 iters), loss = 3.8528
I0407 22:35:42.194514 32718 solver.cpp:237]     Train net output #0: loss = 3.8528 (* 1 = 3.8528 loss)
I0407 22:35:42.194522 32718 sgd_solver.cpp:105] Iteration 1332, lr = 0.00863781
I0407 22:35:47.177865 32718 solver.cpp:218] Iteration 1344 (2.40803 iter/s, 4.98332s/12 iters), loss = 3.88903
I0407 22:35:47.177903 32718 solver.cpp:237]     Train net output #0: loss = 3.88903 (* 1 = 3.88903 loss)
I0407 22:35:47.177911 32718 sgd_solver.cpp:105] Iteration 1344, lr = 0.00863088
I0407 22:35:52.167709 32718 solver.cpp:218] Iteration 1356 (2.40492 iter/s, 4.98978s/12 iters), loss = 4.08996
I0407 22:35:52.167744 32718 solver.cpp:237]     Train net output #0: loss = 4.08996 (* 1 = 4.08996 loss)
I0407 22:35:52.167752 32718 sgd_solver.cpp:105] Iteration 1356, lr = 0.00862391
I0407 22:35:57.140062 32718 solver.cpp:218] Iteration 1368 (2.41338 iter/s, 4.97229s/12 iters), loss = 4.07329
I0407 22:35:57.140096 32718 solver.cpp:237]     Train net output #0: loss = 4.07329 (* 1 = 4.07329 loss)
I0407 22:35:57.140103 32718 sgd_solver.cpp:105] Iteration 1368, lr = 0.00861692
I0407 22:35:58.315660 32718 blocking_queue.cpp:49] Waiting for data
I0407 22:36:02.075942 32718 solver.cpp:218] Iteration 1380 (2.43121 iter/s, 4.93581s/12 iters), loss = 3.9894
I0407 22:36:02.075978 32718 solver.cpp:237]     Train net output #0: loss = 3.9894 (* 1 = 3.9894 loss)
I0407 22:36:02.075984 32718 sgd_solver.cpp:105] Iteration 1380, lr = 0.00860989
I0407 22:36:07.048841 32718 solver.cpp:218] Iteration 1392 (2.41311 iter/s, 4.97283s/12 iters), loss = 3.99872
I0407 22:36:07.048986 32718 solver.cpp:237]     Train net output #0: loss = 3.99872 (* 1 = 3.99872 loss)
I0407 22:36:07.048995 32718 sgd_solver.cpp:105] Iteration 1392, lr = 0.00860284
I0407 22:36:12.029186 32718 solver.cpp:218] Iteration 1404 (2.40956 iter/s, 4.98017s/12 iters), loss = 3.73593
I0407 22:36:12.029232 32718 solver.cpp:237]     Train net output #0: loss = 3.73593 (* 1 = 3.73593 loss)
I0407 22:36:12.029240 32718 sgd_solver.cpp:105] Iteration 1404, lr = 0.00859575
I0407 22:36:16.610504 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:36:16.953045 32718 solver.cpp:218] Iteration 1416 (2.43715 iter/s, 4.92378s/12 iters), loss = 3.75344
I0407 22:36:16.953088 32718 solver.cpp:237]     Train net output #0: loss = 3.75344 (* 1 = 3.75344 loss)
I0407 22:36:16.953097 32718 sgd_solver.cpp:105] Iteration 1416, lr = 0.00858863
I0407 22:36:21.464995 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_1428.caffemodel
I0407 22:36:24.561509 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_1428.solverstate
I0407 22:36:26.980370 32718 solver.cpp:330] Iteration 1428, Testing net (#0)
I0407 22:36:26.980387 32718 net.cpp:676] Ignoring source layer train-data
I0407 22:36:31.142457 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:36:31.796398 32718 solver.cpp:397]     Test net output #0: accuracy = 0.13174
I0407 22:36:31.796433 32718 solver.cpp:397]     Test net output #1: loss = 3.88545 (* 1 = 3.88545 loss)
I0407 22:36:31.892727 32718 solver.cpp:218] Iteration 1428 (0.803236 iter/s, 14.9396s/12 iters), loss = 3.56096
I0407 22:36:31.892762 32718 solver.cpp:237]     Train net output #0: loss = 3.56096 (* 1 = 3.56096 loss)
I0407 22:36:31.892771 32718 sgd_solver.cpp:105] Iteration 1428, lr = 0.00858149
I0407 22:36:35.954787 32718 solver.cpp:218] Iteration 1440 (2.95421 iter/s, 4.062s/12 iters), loss = 3.77939
I0407 22:36:35.954828 32718 solver.cpp:237]     Train net output #0: loss = 3.77939 (* 1 = 3.77939 loss)
I0407 22:36:35.954836 32718 sgd_solver.cpp:105] Iteration 1440, lr = 0.00857431
I0407 22:36:40.872854 32718 solver.cpp:218] Iteration 1452 (2.44002 iter/s, 4.918s/12 iters), loss = 3.65816
I0407 22:36:40.872982 32718 solver.cpp:237]     Train net output #0: loss = 3.65816 (* 1 = 3.65816 loss)
I0407 22:36:40.872990 32718 sgd_solver.cpp:105] Iteration 1452, lr = 0.00856711
I0407 22:36:45.865687 32718 solver.cpp:218] Iteration 1464 (2.40352 iter/s, 4.99267s/12 iters), loss = 3.62403
I0407 22:36:45.865729 32718 solver.cpp:237]     Train net output #0: loss = 3.62403 (* 1 = 3.62403 loss)
I0407 22:36:45.865737 32718 sgd_solver.cpp:105] Iteration 1464, lr = 0.00855987
I0407 22:36:50.852176 32718 solver.cpp:218] Iteration 1476 (2.40654 iter/s, 4.98642s/12 iters), loss = 3.5241
I0407 22:36:50.852213 32718 solver.cpp:237]     Train net output #0: loss = 3.5241 (* 1 = 3.5241 loss)
I0407 22:36:50.852221 32718 sgd_solver.cpp:105] Iteration 1476, lr = 0.00855261
I0407 22:36:55.748873 32718 solver.cpp:218] Iteration 1488 (2.45067 iter/s, 4.89663s/12 iters), loss = 3.93632
I0407 22:36:55.748914 32718 solver.cpp:237]     Train net output #0: loss = 3.93632 (* 1 = 3.93632 loss)
I0407 22:36:55.748922 32718 sgd_solver.cpp:105] Iteration 1488, lr = 0.00854531
I0407 22:37:00.690063 32718 solver.cpp:218] Iteration 1500 (2.4286 iter/s, 4.94112s/12 iters), loss = 3.66003
I0407 22:37:00.690100 32718 solver.cpp:237]     Train net output #0: loss = 3.66003 (* 1 = 3.66003 loss)
I0407 22:37:00.690109 32718 sgd_solver.cpp:105] Iteration 1500, lr = 0.00853798
I0407 22:37:05.629386 32718 solver.cpp:218] Iteration 1512 (2.42952 iter/s, 4.93925s/12 iters), loss = 3.36186
I0407 22:37:05.629434 32718 solver.cpp:237]     Train net output #0: loss = 3.36186 (* 1 = 3.36186 loss)
I0407 22:37:05.629442 32718 sgd_solver.cpp:105] Iteration 1512, lr = 0.00853062
I0407 22:37:07.391831 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:37:10.579365 32718 solver.cpp:218] Iteration 1524 (2.42429 iter/s, 4.9499s/12 iters), loss = 3.33303
I0407 22:37:10.579408 32718 solver.cpp:237]     Train net output #0: loss = 3.33303 (* 1 = 3.33303 loss)
I0407 22:37:10.579416 32718 sgd_solver.cpp:105] Iteration 1524, lr = 0.00852323
I0407 22:37:12.594175 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_1530.caffemodel
I0407 22:37:16.783109 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_1530.solverstate
I0407 22:37:19.156656 32718 solver.cpp:330] Iteration 1530, Testing net (#0)
I0407 22:37:19.156674 32718 net.cpp:676] Ignoring source layer train-data
I0407 22:37:23.224967 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:37:23.932369 32718 solver.cpp:397]     Test net output #0: accuracy = 0.140319
I0407 22:37:23.932421 32718 solver.cpp:397]     Test net output #1: loss = 3.88279 (* 1 = 3.88279 loss)
I0407 22:37:25.731144 32718 solver.cpp:218] Iteration 1536 (0.791992 iter/s, 15.1517s/12 iters), loss = 3.55672
I0407 22:37:25.731184 32718 solver.cpp:237]     Train net output #0: loss = 3.55672 (* 1 = 3.55672 loss)
I0407 22:37:25.731192 32718 sgd_solver.cpp:105] Iteration 1536, lr = 0.00851581
I0407 22:37:30.692394 32718 solver.cpp:218] Iteration 1548 (2.41878 iter/s, 4.96118s/12 iters), loss = 3.67465
I0407 22:37:30.692428 32718 solver.cpp:237]     Train net output #0: loss = 3.67465 (* 1 = 3.67465 loss)
I0407 22:37:30.692435 32718 sgd_solver.cpp:105] Iteration 1548, lr = 0.00850836
I0407 22:37:35.621446 32718 solver.cpp:218] Iteration 1560 (2.43458 iter/s, 4.92898s/12 iters), loss = 3.37963
I0407 22:37:35.621481 32718 solver.cpp:237]     Train net output #0: loss = 3.37963 (* 1 = 3.37963 loss)
I0407 22:37:35.621488 32718 sgd_solver.cpp:105] Iteration 1560, lr = 0.00850088
I0407 22:37:40.582684 32718 solver.cpp:218] Iteration 1572 (2.41878 iter/s, 4.96117s/12 iters), loss = 3.41866
I0407 22:37:40.582726 32718 solver.cpp:237]     Train net output #0: loss = 3.41866 (* 1 = 3.41866 loss)
I0407 22:37:40.582733 32718 sgd_solver.cpp:105] Iteration 1572, lr = 0.00849337
I0407 22:37:45.569906 32718 solver.cpp:218] Iteration 1584 (2.40618 iter/s, 4.98715s/12 iters), loss = 3.33332
I0407 22:37:45.570041 32718 solver.cpp:237]     Train net output #0: loss = 3.33332 (* 1 = 3.33332 loss)
I0407 22:37:45.570050 32718 sgd_solver.cpp:105] Iteration 1584, lr = 0.00848583
I0407 22:37:50.490465 32718 solver.cpp:218] Iteration 1596 (2.43883 iter/s, 4.9204s/12 iters), loss = 3.3939
I0407 22:37:50.490501 32718 solver.cpp:237]     Train net output #0: loss = 3.3939 (* 1 = 3.3939 loss)
I0407 22:37:50.490509 32718 sgd_solver.cpp:105] Iteration 1596, lr = 0.00847826
I0407 22:37:55.464103 32718 solver.cpp:218] Iteration 1608 (2.41275 iter/s, 4.97357s/12 iters), loss = 3.45044
I0407 22:37:55.464148 32718 solver.cpp:237]     Train net output #0: loss = 3.45044 (* 1 = 3.45044 loss)
I0407 22:37:55.464156 32718 sgd_solver.cpp:105] Iteration 1608, lr = 0.00847065
I0407 22:37:59.337863 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:38:00.410293 32718 solver.cpp:218] Iteration 1620 (2.42615 iter/s, 4.94612s/12 iters), loss = 3.38673
I0407 22:38:00.410331 32718 solver.cpp:237]     Train net output #0: loss = 3.38673 (* 1 = 3.38673 loss)
I0407 22:38:00.410339 32718 sgd_solver.cpp:105] Iteration 1620, lr = 0.00846301
I0407 22:38:04.941121 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_1632.caffemodel
I0407 22:38:09.083304 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_1632.solverstate
I0407 22:38:12.883882 32718 solver.cpp:330] Iteration 1632, Testing net (#0)
I0407 22:38:12.883898 32718 net.cpp:676] Ignoring source layer train-data
I0407 22:38:16.870803 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:38:17.560134 32718 solver.cpp:397]     Test net output #0: accuracy = 0.177083
I0407 22:38:17.560180 32718 solver.cpp:397]     Test net output #1: loss = 3.65227 (* 1 = 3.65227 loss)
I0407 22:38:17.656723 32718 solver.cpp:218] Iteration 1632 (0.6958 iter/s, 17.2463s/12 iters), loss = 3.34763
I0407 22:38:17.656761 32718 solver.cpp:237]     Train net output #0: loss = 3.34763 (* 1 = 3.34763 loss)
I0407 22:38:17.656769 32718 sgd_solver.cpp:105] Iteration 1632, lr = 0.00845535
I0407 22:38:22.081954 32718 solver.cpp:218] Iteration 1644 (2.71176 iter/s, 4.42516s/12 iters), loss = 3.30512
I0407 22:38:22.081990 32718 solver.cpp:237]     Train net output #0: loss = 3.30512 (* 1 = 3.30512 loss)
I0407 22:38:22.081998 32718 sgd_solver.cpp:105] Iteration 1644, lr = 0.00844765
I0407 22:38:27.033080 32718 solver.cpp:218] Iteration 1656 (2.42372 iter/s, 4.95106s/12 iters), loss = 3.28916
I0407 22:38:27.033121 32718 solver.cpp:237]     Train net output #0: loss = 3.28916 (* 1 = 3.28916 loss)
I0407 22:38:27.033130 32718 sgd_solver.cpp:105] Iteration 1656, lr = 0.00843992
I0407 22:38:31.997357 32718 solver.cpp:218] Iteration 1668 (2.41731 iter/s, 4.9642s/12 iters), loss = 3.39747
I0407 22:38:31.997400 32718 solver.cpp:237]     Train net output #0: loss = 3.39747 (* 1 = 3.39747 loss)
I0407 22:38:31.997408 32718 sgd_solver.cpp:105] Iteration 1668, lr = 0.00843216
I0407 22:38:36.939025 32718 solver.cpp:218] Iteration 1680 (2.42837 iter/s, 4.94159s/12 iters), loss = 3.22271
I0407 22:38:36.939072 32718 solver.cpp:237]     Train net output #0: loss = 3.22271 (* 1 = 3.22271 loss)
I0407 22:38:36.939079 32718 sgd_solver.cpp:105] Iteration 1680, lr = 0.00842437
I0407 22:38:41.912058 32718 solver.cpp:218] Iteration 1692 (2.41305 iter/s, 4.97296s/12 iters), loss = 3.24803
I0407 22:38:41.912102 32718 solver.cpp:237]     Train net output #0: loss = 3.24803 (* 1 = 3.24803 loss)
I0407 22:38:41.912111 32718 sgd_solver.cpp:105] Iteration 1692, lr = 0.00841654
I0407 22:38:46.899116 32718 solver.cpp:218] Iteration 1704 (2.40626 iter/s, 4.98698s/12 iters), loss = 3.19492
I0407 22:38:46.899277 32718 solver.cpp:237]     Train net output #0: loss = 3.19492 (* 1 = 3.19492 loss)
I0407 22:38:46.899286 32718 sgd_solver.cpp:105] Iteration 1704, lr = 0.00840869
I0407 22:38:51.763013 32718 solver.cpp:218] Iteration 1716 (2.46725 iter/s, 4.86371s/12 iters), loss = 2.82959
I0407 22:38:51.763051 32718 solver.cpp:237]     Train net output #0: loss = 2.82959 (* 1 = 2.82959 loss)
I0407 22:38:51.763059 32718 sgd_solver.cpp:105] Iteration 1716, lr = 0.0084008
I0407 22:38:52.779515 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:38:56.749512 32718 solver.cpp:218] Iteration 1728 (2.40653 iter/s, 4.98643s/12 iters), loss = 3.52787
I0407 22:38:56.749557 32718 solver.cpp:237]     Train net output #0: loss = 3.52787 (* 1 = 3.52787 loss)
I0407 22:38:56.749564 32718 sgd_solver.cpp:105] Iteration 1728, lr = 0.00839288
I0407 22:38:58.815310 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_1734.caffemodel
I0407 22:39:02.868106 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_1734.solverstate
I0407 22:39:05.674124 32718 solver.cpp:330] Iteration 1734, Testing net (#0)
I0407 22:39:05.674141 32718 net.cpp:676] Ignoring source layer train-data
I0407 22:39:09.666818 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:39:10.448235 32718 solver.cpp:397]     Test net output #0: accuracy = 0.18076
I0407 22:39:10.448287 32718 solver.cpp:397]     Test net output #1: loss = 3.63194 (* 1 = 3.63194 loss)
I0407 22:39:12.248204 32718 solver.cpp:218] Iteration 1740 (0.774264 iter/s, 15.4986s/12 iters), loss = 2.90149
I0407 22:39:12.248245 32718 solver.cpp:237]     Train net output #0: loss = 2.90149 (* 1 = 2.90149 loss)
I0407 22:39:12.248252 32718 sgd_solver.cpp:105] Iteration 1740, lr = 0.00838493
I0407 22:39:17.151182 32718 solver.cpp:218] Iteration 1752 (2.44753 iter/s, 4.9029s/12 iters), loss = 3.00702
I0407 22:39:17.151340 32718 solver.cpp:237]     Train net output #0: loss = 3.00702 (* 1 = 3.00702 loss)
I0407 22:39:17.151350 32718 sgd_solver.cpp:105] Iteration 1752, lr = 0.00837695
I0407 22:39:22.119837 32718 solver.cpp:218] Iteration 1764 (2.41523 iter/s, 4.96846s/12 iters), loss = 2.97984
I0407 22:39:22.119880 32718 solver.cpp:237]     Train net output #0: loss = 2.97984 (* 1 = 2.97984 loss)
I0407 22:39:22.119889 32718 sgd_solver.cpp:105] Iteration 1764, lr = 0.00836894
I0407 22:39:27.080416 32718 solver.cpp:218] Iteration 1776 (2.41911 iter/s, 4.9605s/12 iters), loss = 2.79101
I0407 22:39:27.080459 32718 solver.cpp:237]     Train net output #0: loss = 2.79101 (* 1 = 2.79101 loss)
I0407 22:39:27.080467 32718 sgd_solver.cpp:105] Iteration 1776, lr = 0.00836089
I0407 22:39:32.025652 32718 solver.cpp:218] Iteration 1788 (2.42661 iter/s, 4.94516s/12 iters), loss = 3.0808
I0407 22:39:32.025696 32718 solver.cpp:237]     Train net output #0: loss = 3.0808 (* 1 = 3.0808 loss)
I0407 22:39:32.025704 32718 sgd_solver.cpp:105] Iteration 1788, lr = 0.00835281
I0407 22:39:36.891175 32718 solver.cpp:218] Iteration 1800 (2.46637 iter/s, 4.86545s/12 iters), loss = 3.02408
I0407 22:39:36.891223 32718 solver.cpp:237]     Train net output #0: loss = 3.02408 (* 1 = 3.02408 loss)
I0407 22:39:36.891232 32718 sgd_solver.cpp:105] Iteration 1800, lr = 0.0083447
I0407 22:39:41.841094 32718 solver.cpp:218] Iteration 1812 (2.42432 iter/s, 4.94984s/12 iters), loss = 3.11118
I0407 22:39:41.841143 32718 solver.cpp:237]     Train net output #0: loss = 3.11118 (* 1 = 3.11118 loss)
I0407 22:39:41.841156 32718 sgd_solver.cpp:105] Iteration 1812, lr = 0.00833656
I0407 22:39:44.906080 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:39:46.714905 32718 solver.cpp:218] Iteration 1824 (2.46218 iter/s, 4.87374s/12 iters), loss = 3.23447
I0407 22:39:46.714941 32718 solver.cpp:237]     Train net output #0: loss = 3.23447 (* 1 = 3.23447 loss)
I0407 22:39:46.714949 32718 sgd_solver.cpp:105] Iteration 1824, lr = 0.00832839
I0407 22:39:51.229867 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_1836.caffemodel
I0407 22:39:54.290324 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_1836.solverstate
I0407 22:39:56.932878 32718 solver.cpp:330] Iteration 1836, Testing net (#0)
I0407 22:39:56.932894 32718 net.cpp:676] Ignoring source layer train-data
I0407 22:40:00.893101 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:40:01.733244 32718 solver.cpp:397]     Test net output #0: accuracy = 0.218137
I0407 22:40:01.733290 32718 solver.cpp:397]     Test net output #1: loss = 3.47165 (* 1 = 3.47165 loss)
I0407 22:40:01.829928 32718 solver.cpp:218] Iteration 1836 (0.793917 iter/s, 15.1149s/12 iters), loss = 3.23783
I0407 22:40:01.829973 32718 solver.cpp:237]     Train net output #0: loss = 3.23783 (* 1 = 3.23783 loss)
I0407 22:40:01.829982 32718 sgd_solver.cpp:105] Iteration 1836, lr = 0.00832018
I0407 22:40:05.969295 32718 solver.cpp:218] Iteration 1848 (2.89904 iter/s, 4.1393s/12 iters), loss = 3.13732
I0407 22:40:05.969331 32718 solver.cpp:237]     Train net output #0: loss = 3.13732 (* 1 = 3.13732 loss)
I0407 22:40:05.969339 32718 sgd_solver.cpp:105] Iteration 1848, lr = 0.00831195
I0407 22:40:10.847697 32718 solver.cpp:218] Iteration 1860 (2.45985 iter/s, 4.87834s/12 iters), loss = 2.79795
I0407 22:40:10.847733 32718 solver.cpp:237]     Train net output #0: loss = 2.79795 (* 1 = 2.79795 loss)
I0407 22:40:10.847741 32718 sgd_solver.cpp:105] Iteration 1860, lr = 0.00830368
I0407 22:40:15.812120 32718 solver.cpp:218] Iteration 1872 (2.41723 iter/s, 4.96435s/12 iters), loss = 2.82708
I0407 22:40:15.812157 32718 solver.cpp:237]     Train net output #0: loss = 2.82708 (* 1 = 2.82708 loss)
I0407 22:40:15.812165 32718 sgd_solver.cpp:105] Iteration 1872, lr = 0.00829537
I0407 22:40:20.749773 32718 solver.cpp:218] Iteration 1884 (2.43034 iter/s, 4.93758s/12 iters), loss = 3.00315
I0407 22:40:20.749815 32718 solver.cpp:237]     Train net output #0: loss = 3.00315 (* 1 = 3.00315 loss)
I0407 22:40:20.749823 32718 sgd_solver.cpp:105] Iteration 1884, lr = 0.00828704
I0407 22:40:25.755820 32718 solver.cpp:218] Iteration 1896 (2.39714 iter/s, 5.00598s/12 iters), loss = 2.93889
I0407 22:40:25.755947 32718 solver.cpp:237]     Train net output #0: loss = 2.93889 (* 1 = 2.93889 loss)
I0407 22:40:25.755955 32718 sgd_solver.cpp:105] Iteration 1896, lr = 0.00827867
I0407 22:40:30.754053 32718 solver.cpp:218] Iteration 1908 (2.40092 iter/s, 4.99808s/12 iters), loss = 2.76566
I0407 22:40:30.754088 32718 solver.cpp:237]     Train net output #0: loss = 2.76566 (* 1 = 2.76566 loss)
I0407 22:40:30.754096 32718 sgd_solver.cpp:105] Iteration 1908, lr = 0.00827028
I0407 22:40:35.705019 32718 solver.cpp:218] Iteration 1920 (2.4238 iter/s, 4.9509s/12 iters), loss = 2.61135
I0407 22:40:35.705054 32718 solver.cpp:237]     Train net output #0: loss = 2.61135 (* 1 = 2.61135 loss)
I0407 22:40:35.705062 32718 sgd_solver.cpp:105] Iteration 1920, lr = 0.00826184
I0407 22:40:36.000196 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:40:40.610100 32718 solver.cpp:218] Iteration 1932 (2.44648 iter/s, 4.90501s/12 iters), loss = 3.06587
I0407 22:40:40.610142 32718 solver.cpp:237]     Train net output #0: loss = 3.06587 (* 1 = 3.06587 loss)
I0407 22:40:40.610152 32718 sgd_solver.cpp:105] Iteration 1932, lr = 0.00825338
I0407 22:40:42.614002 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_1938.caffemodel
I0407 22:40:45.683290 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_1938.solverstate
I0407 22:40:48.068747 32718 solver.cpp:330] Iteration 1938, Testing net (#0)
I0407 22:40:48.068766 32718 net.cpp:676] Ignoring source layer train-data
I0407 22:40:51.806787 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:40:52.615928 32718 solver.cpp:397]     Test net output #0: accuracy = 0.223039
I0407 22:40:52.615976 32718 solver.cpp:397]     Test net output #1: loss = 3.42089 (* 1 = 3.42089 loss)
I0407 22:40:54.520617 32718 solver.cpp:218] Iteration 1944 (0.862663 iter/s, 13.9104s/12 iters), loss = 3.02763
I0407 22:40:54.520656 32718 solver.cpp:237]     Train net output #0: loss = 3.02763 (* 1 = 3.02763 loss)
I0407 22:40:54.520664 32718 sgd_solver.cpp:105] Iteration 1944, lr = 0.00824489
I0407 22:40:59.459337 32718 solver.cpp:218] Iteration 1956 (2.42982 iter/s, 4.93865s/12 iters), loss = 2.82791
I0407 22:40:59.459486 32718 solver.cpp:237]     Train net output #0: loss = 2.82791 (* 1 = 2.82791 loss)
I0407 22:40:59.459496 32718 sgd_solver.cpp:105] Iteration 1956, lr = 0.00823636
I0407 22:41:04.388952 32718 solver.cpp:218] Iteration 1968 (2.43435 iter/s, 4.92944s/12 iters), loss = 2.55981
I0407 22:41:04.388998 32718 solver.cpp:237]     Train net output #0: loss = 2.55981 (* 1 = 2.55981 loss)
I0407 22:41:04.389008 32718 sgd_solver.cpp:105] Iteration 1968, lr = 0.0082278
I0407 22:41:09.339820 32718 solver.cpp:218] Iteration 1980 (2.42386 iter/s, 4.95079s/12 iters), loss = 2.6496
I0407 22:41:09.339864 32718 solver.cpp:237]     Train net output #0: loss = 2.6496 (* 1 = 2.6496 loss)
I0407 22:41:09.339874 32718 sgd_solver.cpp:105] Iteration 1980, lr = 0.0082192
I0407 22:41:14.263540 32718 solver.cpp:218] Iteration 1992 (2.43722 iter/s, 4.92364s/12 iters), loss = 2.91698
I0407 22:41:14.263586 32718 solver.cpp:237]     Train net output #0: loss = 2.91698 (* 1 = 2.91698 loss)
I0407 22:41:14.263593 32718 sgd_solver.cpp:105] Iteration 1992, lr = 0.00821058
I0407 22:41:19.214309 32718 solver.cpp:218] Iteration 2004 (2.4239 iter/s, 4.95069s/12 iters), loss = 2.69779
I0407 22:41:19.214354 32718 solver.cpp:237]     Train net output #0: loss = 2.69779 (* 1 = 2.69779 loss)
I0407 22:41:19.214361 32718 sgd_solver.cpp:105] Iteration 2004, lr = 0.00820192
I0407 22:41:24.168289 32718 solver.cpp:218] Iteration 2016 (2.42233 iter/s, 4.95391s/12 iters), loss = 2.54172
I0407 22:41:24.168330 32718 solver.cpp:237]     Train net output #0: loss = 2.54172 (* 1 = 2.54172 loss)
I0407 22:41:24.168339 32718 sgd_solver.cpp:105] Iteration 2016, lr = 0.00819323
I0407 22:41:26.664943 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:41:29.069224 32718 solver.cpp:218] Iteration 2028 (2.44855 iter/s, 4.90086s/12 iters), loss = 2.86535
I0407 22:41:29.069259 32718 solver.cpp:237]     Train net output #0: loss = 2.86535 (* 1 = 2.86535 loss)
I0407 22:41:29.069267 32718 sgd_solver.cpp:105] Iteration 2028, lr = 0.0081845
I0407 22:41:33.589319 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_2040.caffemodel
I0407 22:41:36.716941 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_2040.solverstate
I0407 22:41:39.078771 32718 solver.cpp:330] Iteration 2040, Testing net (#0)
I0407 22:41:39.078789 32718 net.cpp:676] Ignoring source layer train-data
I0407 22:41:42.940378 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:41:43.866166 32718 solver.cpp:397]     Test net output #0: accuracy = 0.261642
I0407 22:41:43.866211 32718 solver.cpp:397]     Test net output #1: loss = 3.17958 (* 1 = 3.17958 loss)
I0407 22:41:43.962868 32718 solver.cpp:218] Iteration 2040 (0.805718 iter/s, 14.8935s/12 iters), loss = 2.6993
I0407 22:41:43.962915 32718 solver.cpp:237]     Train net output #0: loss = 2.6993 (* 1 = 2.6993 loss)
I0407 22:41:43.962924 32718 sgd_solver.cpp:105] Iteration 2040, lr = 0.00817574
I0407 22:41:48.139313 32718 solver.cpp:218] Iteration 2052 (2.87331 iter/s, 4.17637s/12 iters), loss = 2.79864
I0407 22:41:48.139350 32718 solver.cpp:237]     Train net output #0: loss = 2.79864 (* 1 = 2.79864 loss)
I0407 22:41:48.139359 32718 sgd_solver.cpp:105] Iteration 2052, lr = 0.00816695
I0407 22:41:49.722298 32718 blocking_queue.cpp:49] Waiting for data
I0407 22:41:53.057178 32718 solver.cpp:218] Iteration 2064 (2.44012 iter/s, 4.9178s/12 iters), loss = 3.0835
I0407 22:41:53.057216 32718 solver.cpp:237]     Train net output #0: loss = 3.0835 (* 1 = 3.0835 loss)
I0407 22:41:53.057224 32718 sgd_solver.cpp:105] Iteration 2064, lr = 0.00815813
I0407 22:41:58.044848 32718 solver.cpp:218] Iteration 2076 (2.40597 iter/s, 4.9876s/12 iters), loss = 2.87891
I0407 22:41:58.044890 32718 solver.cpp:237]     Train net output #0: loss = 2.87891 (* 1 = 2.87891 loss)
I0407 22:41:58.044898 32718 sgd_solver.cpp:105] Iteration 2076, lr = 0.00814928
I0407 22:42:03.016794 32718 solver.cpp:218] Iteration 2088 (2.41358 iter/s, 4.97187s/12 iters), loss = 2.82663
I0407 22:42:03.016949 32718 solver.cpp:237]     Train net output #0: loss = 2.82663 (* 1 = 2.82663 loss)
I0407 22:42:03.016959 32718 sgd_solver.cpp:105] Iteration 2088, lr = 0.00814039
I0407 22:42:07.930090 32718 solver.cpp:218] Iteration 2100 (2.44245 iter/s, 4.91311s/12 iters), loss = 2.77649
I0407 22:42:07.930233 32718 solver.cpp:237]     Train net output #0: loss = 2.77649 (* 1 = 2.77649 loss)
I0407 22:42:07.930243 32718 sgd_solver.cpp:105] Iteration 2100, lr = 0.00813147
I0407 22:42:12.914160 32718 solver.cpp:218] Iteration 2112 (2.40776 iter/s, 4.98389s/12 iters), loss = 2.50972
I0407 22:42:12.914203 32718 solver.cpp:237]     Train net output #0: loss = 2.50972 (* 1 = 2.50972 loss)
I0407 22:42:12.914211 32718 sgd_solver.cpp:105] Iteration 2112, lr = 0.00812251
I0407 22:42:17.505807 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:42:17.819054 32718 solver.cpp:218] Iteration 2124 (2.44657 iter/s, 4.90482s/12 iters), loss = 2.62927
I0407 22:42:17.819092 32718 solver.cpp:237]     Train net output #0: loss = 2.62927 (* 1 = 2.62927 loss)
I0407 22:42:17.819099 32718 sgd_solver.cpp:105] Iteration 2124, lr = 0.00811353
I0407 22:42:22.794150 32718 solver.cpp:218] Iteration 2136 (2.41205 iter/s, 4.97501s/12 iters), loss = 2.72927
I0407 22:42:22.794210 32718 solver.cpp:237]     Train net output #0: loss = 2.72927 (* 1 = 2.72927 loss)
I0407 22:42:22.794222 32718 sgd_solver.cpp:105] Iteration 2136, lr = 0.00810451
I0407 22:42:24.826797 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_2142.caffemodel
I0407 22:42:28.642627 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_2142.solverstate
I0407 22:42:31.062428 32718 solver.cpp:330] Iteration 2142, Testing net (#0)
I0407 22:42:31.062448 32718 net.cpp:676] Ignoring source layer train-data
I0407 22:42:34.801800 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:42:35.690441 32718 solver.cpp:397]     Test net output #0: accuracy = 0.273897
I0407 22:42:35.690490 32718 solver.cpp:397]     Test net output #1: loss = 3.09691 (* 1 = 3.09691 loss)
I0407 22:42:37.515280 32718 solver.cpp:218] Iteration 2148 (0.815161 iter/s, 14.721s/12 iters), loss = 2.55026
I0407 22:42:37.515323 32718 solver.cpp:237]     Train net output #0: loss = 2.55026 (* 1 = 2.55026 loss)
I0407 22:42:37.515332 32718 sgd_solver.cpp:105] Iteration 2148, lr = 0.00809545
I0407 22:42:42.398885 32718 solver.cpp:218] Iteration 2160 (2.45724 iter/s, 4.88354s/12 iters), loss = 2.60942
I0407 22:42:42.399041 32718 solver.cpp:237]     Train net output #0: loss = 2.60942 (* 1 = 2.60942 loss)
I0407 22:42:42.399050 32718 sgd_solver.cpp:105] Iteration 2160, lr = 0.00808637
I0407 22:42:47.330412 32718 solver.cpp:218] Iteration 2172 (2.43341 iter/s, 4.93134s/12 iters), loss = 2.30428
I0407 22:42:47.330447 32718 solver.cpp:237]     Train net output #0: loss = 2.30428 (* 1 = 2.30428 loss)
I0407 22:42:47.330456 32718 sgd_solver.cpp:105] Iteration 2172, lr = 0.00807725
I0407 22:42:52.281327 32718 solver.cpp:218] Iteration 2184 (2.42383 iter/s, 4.95085s/12 iters), loss = 2.39968
I0407 22:42:52.281364 32718 solver.cpp:237]     Train net output #0: loss = 2.39968 (* 1 = 2.39968 loss)
I0407 22:42:52.281373 32718 sgd_solver.cpp:105] Iteration 2184, lr = 0.0080681
I0407 22:42:57.207302 32718 solver.cpp:218] Iteration 2196 (2.4361 iter/s, 4.92591s/12 iters), loss = 2.70207
I0407 22:42:57.207340 32718 solver.cpp:237]     Train net output #0: loss = 2.70207 (* 1 = 2.70207 loss)
I0407 22:42:57.207347 32718 sgd_solver.cpp:105] Iteration 2196, lr = 0.00805891
I0407 22:43:02.154894 32718 solver.cpp:218] Iteration 2208 (2.42546 iter/s, 4.94752s/12 iters), loss = 2.61052
I0407 22:43:02.154933 32718 solver.cpp:237]     Train net output #0: loss = 2.61052 (* 1 = 2.61052 loss)
I0407 22:43:02.154942 32718 sgd_solver.cpp:105] Iteration 2208, lr = 0.00804969
I0407 22:43:07.101501 32718 solver.cpp:218] Iteration 2220 (2.42594 iter/s, 4.94654s/12 iters), loss = 2.17268
I0407 22:43:07.101539 32718 solver.cpp:237]     Train net output #0: loss = 2.17268 (* 1 = 2.17268 loss)
I0407 22:43:07.101547 32718 sgd_solver.cpp:105] Iteration 2220, lr = 0.00804044
I0407 22:43:08.816439 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:43:11.900522 32718 solver.cpp:218] Iteration 2232 (2.50055 iter/s, 4.79895s/12 iters), loss = 2.23806
I0407 22:43:11.900564 32718 solver.cpp:237]     Train net output #0: loss = 2.23806 (* 1 = 2.23806 loss)
I0407 22:43:11.900573 32718 sgd_solver.cpp:105] Iteration 2232, lr = 0.00803116
I0407 22:43:16.409530 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_2244.caffemodel
I0407 22:43:19.516086 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_2244.solverstate
I0407 22:43:21.890020 32718 solver.cpp:330] Iteration 2244, Testing net (#0)
I0407 22:43:21.890038 32718 net.cpp:676] Ignoring source layer train-data
I0407 22:43:25.494992 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:43:26.432512 32718 solver.cpp:397]     Test net output #0: accuracy = 0.275735
I0407 22:43:26.432559 32718 solver.cpp:397]     Test net output #1: loss = 3.14145 (* 1 = 3.14145 loss)
I0407 22:43:26.529023 32718 solver.cpp:218] Iteration 2244 (0.820322 iter/s, 14.6284s/12 iters), loss = 2.51047
I0407 22:43:26.529064 32718 solver.cpp:237]     Train net output #0: loss = 2.51047 (* 1 = 2.51047 loss)
I0407 22:43:26.529072 32718 sgd_solver.cpp:105] Iteration 2244, lr = 0.00802184
I0407 22:43:30.630712 32718 solver.cpp:218] Iteration 2256 (2.92568 iter/s, 4.10162s/12 iters), loss = 2.56473
I0407 22:43:30.630753 32718 solver.cpp:237]     Train net output #0: loss = 2.56473 (* 1 = 2.56473 loss)
I0407 22:43:30.630760 32718 sgd_solver.cpp:105] Iteration 2256, lr = 0.00801249
I0407 22:43:35.586402 32718 solver.cpp:218] Iteration 2268 (2.42149 iter/s, 4.95562s/12 iters), loss = 2.47402
I0407 22:43:35.586441 32718 solver.cpp:237]     Train net output #0: loss = 2.47402 (* 1 = 2.47402 loss)
I0407 22:43:35.586447 32718 sgd_solver.cpp:105] Iteration 2268, lr = 0.0080031
I0407 22:43:40.544595 32718 solver.cpp:218] Iteration 2280 (2.42027 iter/s, 4.95813s/12 iters), loss = 2.46765
I0407 22:43:40.544629 32718 solver.cpp:237]     Train net output #0: loss = 2.46765 (* 1 = 2.46765 loss)
I0407 22:43:40.544636 32718 sgd_solver.cpp:105] Iteration 2280, lr = 0.00799369
I0407 22:43:45.568703 32718 solver.cpp:218] Iteration 2292 (2.38852 iter/s, 5.02404s/12 iters), loss = 2.70167
I0407 22:43:45.568743 32718 solver.cpp:237]     Train net output #0: loss = 2.70167 (* 1 = 2.70167 loss)
I0407 22:43:45.568753 32718 sgd_solver.cpp:105] Iteration 2292, lr = 0.00798424
I0407 22:43:50.497917 32718 solver.cpp:218] Iteration 2304 (2.4345 iter/s, 4.92914s/12 iters), loss = 2.03311
I0407 22:43:50.498087 32718 solver.cpp:237]     Train net output #0: loss = 2.03311 (* 1 = 2.03311 loss)
I0407 22:43:50.498096 32718 sgd_solver.cpp:105] Iteration 2304, lr = 0.00797475
I0407 22:43:55.487020 32718 solver.cpp:218] Iteration 2316 (2.40534 iter/s, 4.9889s/12 iters), loss = 2.3157
I0407 22:43:55.487062 32718 solver.cpp:237]     Train net output #0: loss = 2.3157 (* 1 = 2.3157 loss)
I0407 22:43:55.487071 32718 sgd_solver.cpp:105] Iteration 2316, lr = 0.00796523
I0407 22:43:59.383646 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:44:00.408308 32718 solver.cpp:218] Iteration 2328 (2.43842 iter/s, 4.92122s/12 iters), loss = 2.35682
I0407 22:44:00.408345 32718 solver.cpp:237]     Train net output #0: loss = 2.35682 (* 1 = 2.35682 loss)
I0407 22:44:00.408354 32718 sgd_solver.cpp:105] Iteration 2328, lr = 0.00795568
I0407 22:44:05.285558 32718 solver.cpp:218] Iteration 2340 (2.46044 iter/s, 4.87718s/12 iters), loss = 2.53659
I0407 22:44:05.285599 32718 solver.cpp:237]     Train net output #0: loss = 2.53659 (* 1 = 2.53659 loss)
I0407 22:44:05.285609 32718 sgd_solver.cpp:105] Iteration 2340, lr = 0.0079461
I0407 22:44:07.204391 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_2346.caffemodel
I0407 22:44:10.265241 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_2346.solverstate
I0407 22:44:12.679236 32718 solver.cpp:330] Iteration 2346, Testing net (#0)
I0407 22:44:12.679255 32718 net.cpp:676] Ignoring source layer train-data
I0407 22:44:16.243657 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:44:17.212121 32718 solver.cpp:397]     Test net output #0: accuracy = 0.287377
I0407 22:44:17.212164 32718 solver.cpp:397]     Test net output #1: loss = 3.20327 (* 1 = 3.20327 loss)
I0407 22:44:18.997097 32718 solver.cpp:218] Iteration 2352 (0.875182 iter/s, 13.7114s/12 iters), loss = 2.38624
I0407 22:44:18.997138 32718 solver.cpp:237]     Train net output #0: loss = 2.38624 (* 1 = 2.38624 loss)
I0407 22:44:18.997145 32718 sgd_solver.cpp:105] Iteration 2352, lr = 0.00793648
I0407 22:44:23.917798 32718 solver.cpp:218] Iteration 2364 (2.43871 iter/s, 4.92063s/12 iters), loss = 2.38485
I0407 22:44:23.917935 32718 solver.cpp:237]     Train net output #0: loss = 2.38485 (* 1 = 2.38485 loss)
I0407 22:44:23.917945 32718 sgd_solver.cpp:105] Iteration 2364, lr = 0.00792683
I0407 22:44:28.874704 32718 solver.cpp:218] Iteration 2376 (2.42095 iter/s, 4.95673s/12 iters), loss = 2.56849
I0407 22:44:28.874750 32718 solver.cpp:237]     Train net output #0: loss = 2.56849 (* 1 = 2.56849 loss)
I0407 22:44:28.874758 32718 sgd_solver.cpp:105] Iteration 2376, lr = 0.00791715
I0407 22:44:33.807695 32718 solver.cpp:218] Iteration 2388 (2.43264 iter/s, 4.93291s/12 iters), loss = 2.2302
I0407 22:44:33.807737 32718 solver.cpp:237]     Train net output #0: loss = 2.2302 (* 1 = 2.2302 loss)
I0407 22:44:33.807745 32718 sgd_solver.cpp:105] Iteration 2388, lr = 0.00790743
I0407 22:44:38.764708 32718 solver.cpp:218] Iteration 2400 (2.42085 iter/s, 4.95694s/12 iters), loss = 2.49791
I0407 22:44:38.764744 32718 solver.cpp:237]     Train net output #0: loss = 2.49791 (* 1 = 2.49791 loss)
I0407 22:44:38.764752 32718 sgd_solver.cpp:105] Iteration 2400, lr = 0.00789768
I0407 22:44:43.708248 32718 solver.cpp:218] Iteration 2412 (2.42744 iter/s, 4.94347s/12 iters), loss = 2.23743
I0407 22:44:43.708289 32718 solver.cpp:237]     Train net output #0: loss = 2.23743 (* 1 = 2.23743 loss)
I0407 22:44:43.708298 32718 sgd_solver.cpp:105] Iteration 2412, lr = 0.0078879
I0407 22:44:48.674571 32718 solver.cpp:218] Iteration 2424 (2.41631 iter/s, 4.96625s/12 iters), loss = 2.15714
I0407 22:44:48.674612 32718 solver.cpp:237]     Train net output #0: loss = 2.15714 (* 1 = 2.15714 loss)
I0407 22:44:48.674620 32718 sgd_solver.cpp:105] Iteration 2424, lr = 0.00787808
I0407 22:44:49.720793 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:44:53.530654 32718 solver.cpp:218] Iteration 2436 (2.47116 iter/s, 4.85602s/12 iters), loss = 2.08225
I0407 22:44:53.530690 32718 solver.cpp:237]     Train net output #0: loss = 2.08225 (* 1 = 2.08225 loss)
I0407 22:44:53.530699 32718 sgd_solver.cpp:105] Iteration 2436, lr = 0.00786823
I0407 22:44:58.040483 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_2448.caffemodel
I0407 22:45:01.106876 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_2448.solverstate
I0407 22:45:03.463975 32718 solver.cpp:330] Iteration 2448, Testing net (#0)
I0407 22:45:03.463997 32718 net.cpp:676] Ignoring source layer train-data
I0407 22:45:07.205883 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:45:08.286285 32718 solver.cpp:397]     Test net output #0: accuracy = 0.293505
I0407 22:45:08.286332 32718 solver.cpp:397]     Test net output #1: loss = 3.04017 (* 1 = 3.04017 loss)
I0407 22:45:08.383071 32718 solver.cpp:218] Iteration 2448 (0.807955 iter/s, 14.8523s/12 iters), loss = 2.14156
I0407 22:45:08.383111 32718 solver.cpp:237]     Train net output #0: loss = 2.14156 (* 1 = 2.14156 loss)
I0407 22:45:08.383121 32718 sgd_solver.cpp:105] Iteration 2448, lr = 0.00785835
I0407 22:45:12.498962 32718 solver.cpp:218] Iteration 2460 (2.91558 iter/s, 4.11582s/12 iters), loss = 2.14884
I0407 22:45:12.499007 32718 solver.cpp:237]     Train net output #0: loss = 2.14884 (* 1 = 2.14884 loss)
I0407 22:45:12.499015 32718 sgd_solver.cpp:105] Iteration 2460, lr = 0.00784843
I0407 22:45:17.440325 32718 solver.cpp:218] Iteration 2472 (2.42852 iter/s, 4.94129s/12 iters), loss = 2.17274
I0407 22:45:17.440368 32718 solver.cpp:237]     Train net output #0: loss = 2.17274 (* 1 = 2.17274 loss)
I0407 22:45:17.440377 32718 sgd_solver.cpp:105] Iteration 2472, lr = 0.00783848
I0407 22:45:22.412487 32718 solver.cpp:218] Iteration 2484 (2.41347 iter/s, 4.97208s/12 iters), loss = 2.23155
I0407 22:45:22.412531 32718 solver.cpp:237]     Train net output #0: loss = 2.23155 (* 1 = 2.23155 loss)
I0407 22:45:22.412540 32718 sgd_solver.cpp:105] Iteration 2484, lr = 0.0078285
I0407 22:45:27.356674 32718 solver.cpp:218] Iteration 2496 (2.42713 iter/s, 4.94411s/12 iters), loss = 2.21864
I0407 22:45:27.356719 32718 solver.cpp:237]     Train net output #0: loss = 2.21864 (* 1 = 2.21864 loss)
I0407 22:45:27.356729 32718 sgd_solver.cpp:105] Iteration 2496, lr = 0.00781848
I0407 22:45:32.314697 32718 solver.cpp:218] Iteration 2508 (2.42036 iter/s, 4.95794s/12 iters), loss = 2.23207
I0407 22:45:32.314826 32718 solver.cpp:237]     Train net output #0: loss = 2.23207 (* 1 = 2.23207 loss)
I0407 22:45:32.314836 32718 sgd_solver.cpp:105] Iteration 2508, lr = 0.00780843
I0407 22:45:37.281098 32718 solver.cpp:218] Iteration 2520 (2.41631 iter/s, 4.96625s/12 iters), loss = 1.9509
I0407 22:45:37.281134 32718 solver.cpp:237]     Train net output #0: loss = 1.9509 (* 1 = 1.9509 loss)
I0407 22:45:37.281141 32718 sgd_solver.cpp:105] Iteration 2520, lr = 0.00779835
I0407 22:45:40.440004 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:45:42.194798 32718 solver.cpp:218] Iteration 2532 (2.44218 iter/s, 4.91364s/12 iters), loss = 2.04131
I0407 22:45:42.194837 32718 solver.cpp:237]     Train net output #0: loss = 2.04131 (* 1 = 2.04131 loss)
I0407 22:45:42.194845 32718 sgd_solver.cpp:105] Iteration 2532, lr = 0.00778824
I0407 22:45:47.168742 32718 solver.cpp:218] Iteration 2544 (2.4126 iter/s, 4.97388s/12 iters), loss = 1.95753
I0407 22:45:47.168781 32718 solver.cpp:237]     Train net output #0: loss = 1.95753 (* 1 = 1.95753 loss)
I0407 22:45:47.168788 32718 sgd_solver.cpp:105] Iteration 2544, lr = 0.00777809
I0407 22:45:49.177714 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_2550.caffemodel
I0407 22:45:52.290043 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_2550.solverstate
I0407 22:45:54.684875 32718 solver.cpp:330] Iteration 2550, Testing net (#0)
I0407 22:45:54.684891 32718 net.cpp:676] Ignoring source layer train-data
I0407 22:45:58.100086 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:45:59.144412 32718 solver.cpp:397]     Test net output #0: accuracy = 0.314338
I0407 22:45:59.144445 32718 solver.cpp:397]     Test net output #1: loss = 3.03587 (* 1 = 3.03587 loss)
I0407 22:46:00.882170 32718 solver.cpp:218] Iteration 2556 (0.875061 iter/s, 13.7133s/12 iters), loss = 2.20723
I0407 22:46:00.882208 32718 solver.cpp:237]     Train net output #0: loss = 2.20723 (* 1 = 2.20723 loss)
I0407 22:46:00.882216 32718 sgd_solver.cpp:105] Iteration 2556, lr = 0.0077679
I0407 22:46:05.827148 32718 solver.cpp:218] Iteration 2568 (2.42674 iter/s, 4.94491s/12 iters), loss = 1.91202
I0407 22:46:05.827329 32718 solver.cpp:237]     Train net output #0: loss = 1.91202 (* 1 = 1.91202 loss)
I0407 22:46:05.827339 32718 sgd_solver.cpp:105] Iteration 2568, lr = 0.00775769
I0407 22:46:10.761914 32718 solver.cpp:218] Iteration 2580 (2.43183 iter/s, 4.93456s/12 iters), loss = 1.85921
I0407 22:46:10.761947 32718 solver.cpp:237]     Train net output #0: loss = 1.85921 (* 1 = 1.85921 loss)
I0407 22:46:10.761955 32718 sgd_solver.cpp:105] Iteration 2580, lr = 0.00774744
I0407 22:46:15.721495 32718 solver.cpp:218] Iteration 2592 (2.41959 iter/s, 4.95952s/12 iters), loss = 1.99507
I0407 22:46:15.721527 32718 solver.cpp:237]     Train net output #0: loss = 1.99507 (* 1 = 1.99507 loss)
I0407 22:46:15.721534 32718 sgd_solver.cpp:105] Iteration 2592, lr = 0.00773716
I0407 22:46:20.649705 32718 solver.cpp:218] Iteration 2604 (2.43499 iter/s, 4.92815s/12 iters), loss = 1.89453
I0407 22:46:20.649745 32718 solver.cpp:237]     Train net output #0: loss = 1.89453 (* 1 = 1.89453 loss)
I0407 22:46:20.649753 32718 sgd_solver.cpp:105] Iteration 2604, lr = 0.00772684
I0407 22:46:25.631953 32718 solver.cpp:218] Iteration 2616 (2.40859 iter/s, 4.98217s/12 iters), loss = 1.88389
I0407 22:46:25.631997 32718 solver.cpp:237]     Train net output #0: loss = 1.88389 (* 1 = 1.88389 loss)
I0407 22:46:25.632005 32718 sgd_solver.cpp:105] Iteration 2616, lr = 0.00771649
I0407 22:46:30.619535 32718 solver.cpp:218] Iteration 2628 (2.40601 iter/s, 4.98751s/12 iters), loss = 1.76822
I0407 22:46:30.619576 32718 solver.cpp:237]     Train net output #0: loss = 1.76822 (* 1 = 1.76822 loss)
I0407 22:46:30.619586 32718 sgd_solver.cpp:105] Iteration 2628, lr = 0.00770611
I0407 22:46:31.036866 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:46:35.522833 32718 solver.cpp:218] Iteration 2640 (2.44737 iter/s, 4.90323s/12 iters), loss = 2.1878
I0407 22:46:35.522876 32718 solver.cpp:237]     Train net output #0: loss = 2.1878 (* 1 = 2.1878 loss)
I0407 22:46:35.522884 32718 sgd_solver.cpp:105] Iteration 2640, lr = 0.0076957
I0407 22:46:39.952543 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_2652.caffemodel
I0407 22:46:44.212869 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_2652.solverstate
I0407 22:46:46.976444 32718 solver.cpp:330] Iteration 2652, Testing net (#0)
I0407 22:46:46.976464 32718 net.cpp:676] Ignoring source layer train-data
I0407 22:46:50.691797 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:46:51.774435 32718 solver.cpp:397]     Test net output #0: accuracy = 0.321691
I0407 22:46:51.774482 32718 solver.cpp:397]     Test net output #1: loss = 2.84975 (* 1 = 2.84975 loss)
I0407 22:46:51.871850 32718 solver.cpp:218] Iteration 2652 (0.733994 iter/s, 16.3489s/12 iters), loss = 2.00297
I0407 22:46:51.871896 32718 solver.cpp:237]     Train net output #0: loss = 2.00297 (* 1 = 2.00297 loss)
I0407 22:46:51.871906 32718 sgd_solver.cpp:105] Iteration 2652, lr = 0.00768525
I0407 22:46:55.999414 32718 solver.cpp:218] Iteration 2664 (2.90734 iter/s, 4.12749s/12 iters), loss = 1.80672
I0407 22:46:55.999452 32718 solver.cpp:237]     Train net output #0: loss = 1.80672 (* 1 = 1.80672 loss)
I0407 22:46:55.999460 32718 sgd_solver.cpp:105] Iteration 2664, lr = 0.00767477
I0407 22:47:00.879544 32718 solver.cpp:218] Iteration 2676 (2.45899 iter/s, 4.88005s/12 iters), loss = 1.83451
I0407 22:47:00.879585 32718 solver.cpp:237]     Train net output #0: loss = 1.83451 (* 1 = 1.83451 loss)
I0407 22:47:00.879593 32718 sgd_solver.cpp:105] Iteration 2676, lr = 0.00766425
I0407 22:47:05.845233 32718 solver.cpp:218] Iteration 2688 (2.41662 iter/s, 4.96562s/12 iters), loss = 2.34123
I0407 22:47:05.845273 32718 solver.cpp:237]     Train net output #0: loss = 2.34123 (* 1 = 2.34123 loss)
I0407 22:47:05.845279 32718 sgd_solver.cpp:105] Iteration 2688, lr = 0.00765371
I0407 22:47:10.751636 32718 solver.cpp:218] Iteration 2700 (2.44582 iter/s, 4.90633s/12 iters), loss = 2.12386
I0407 22:47:10.751770 32718 solver.cpp:237]     Train net output #0: loss = 2.12386 (* 1 = 2.12386 loss)
I0407 22:47:10.751777 32718 sgd_solver.cpp:105] Iteration 2700, lr = 0.00764313
I0407 22:47:15.734814 32718 solver.cpp:218] Iteration 2712 (2.40818 iter/s, 4.98302s/12 iters), loss = 2.02711
I0407 22:47:15.734853 32718 solver.cpp:237]     Train net output #0: loss = 2.02711 (* 1 = 2.02711 loss)
I0407 22:47:15.734860 32718 sgd_solver.cpp:105] Iteration 2712, lr = 0.00763251
I0407 22:47:20.700928 32718 solver.cpp:218] Iteration 2724 (2.41641 iter/s, 4.96605s/12 iters), loss = 1.58037
I0407 22:47:20.700968 32718 solver.cpp:237]     Train net output #0: loss = 1.58037 (* 1 = 1.58037 loss)
I0407 22:47:20.700976 32718 sgd_solver.cpp:105] Iteration 2724, lr = 0.00762187
I0407 22:47:23.233436 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:47:25.624454 32718 solver.cpp:218] Iteration 2736 (2.43731 iter/s, 4.92346s/12 iters), loss = 1.64928
I0407 22:47:25.624486 32718 solver.cpp:237]     Train net output #0: loss = 1.64928 (* 1 = 1.64928 loss)
I0407 22:47:25.624493 32718 sgd_solver.cpp:105] Iteration 2736, lr = 0.00761119
I0407 22:47:30.553952 32718 solver.cpp:218] Iteration 2748 (2.43435 iter/s, 4.92944s/12 iters), loss = 1.75914
I0407 22:47:30.553987 32718 solver.cpp:237]     Train net output #0: loss = 1.75914 (* 1 = 1.75914 loss)
I0407 22:47:30.553993 32718 sgd_solver.cpp:105] Iteration 2748, lr = 0.00760048
I0407 22:47:32.554169 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_2754.caffemodel
I0407 22:47:35.626262 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_2754.solverstate
I0407 22:47:38.021140 32718 solver.cpp:330] Iteration 2754, Testing net (#0)
I0407 22:47:38.021158 32718 net.cpp:676] Ignoring source layer train-data
I0407 22:47:41.372622 32718 blocking_queue.cpp:49] Waiting for data
I0407 22:47:41.633307 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:47:42.778806 32718 solver.cpp:397]     Test net output #0: accuracy = 0.327206
I0407 22:47:42.778862 32718 solver.cpp:397]     Test net output #1: loss = 2.96049 (* 1 = 2.96049 loss)
I0407 22:47:44.596410 32718 solver.cpp:218] Iteration 2760 (0.854557 iter/s, 14.0424s/12 iters), loss = 1.86
I0407 22:47:44.596446 32718 solver.cpp:237]     Train net output #0: loss = 1.86 (* 1 = 1.86 loss)
I0407 22:47:44.596454 32718 sgd_solver.cpp:105] Iteration 2760, lr = 0.00758973
I0407 22:47:49.536922 32718 solver.cpp:218] Iteration 2772 (2.42893 iter/s, 4.94045s/12 iters), loss = 1.92635
I0407 22:47:49.536952 32718 solver.cpp:237]     Train net output #0: loss = 1.92635 (* 1 = 1.92635 loss)
I0407 22:47:49.536957 32718 sgd_solver.cpp:105] Iteration 2772, lr = 0.00757896
I0407 22:47:54.518067 32718 solver.cpp:218] Iteration 2784 (2.40912 iter/s, 4.98108s/12 iters), loss = 2.12919
I0407 22:47:54.518111 32718 solver.cpp:237]     Train net output #0: loss = 2.12919 (* 1 = 2.12919 loss)
I0407 22:47:54.518119 32718 sgd_solver.cpp:105] Iteration 2784, lr = 0.00756815
I0407 22:47:59.482039 32718 solver.cpp:218] Iteration 2796 (2.41746 iter/s, 4.96389s/12 iters), loss = 1.9212
I0407 22:47:59.482082 32718 solver.cpp:237]     Train net output #0: loss = 1.9212 (* 1 = 1.9212 loss)
I0407 22:47:59.482091 32718 sgd_solver.cpp:105] Iteration 2796, lr = 0.0075573
I0407 22:48:04.387315 32718 solver.cpp:218] Iteration 2808 (2.44638 iter/s, 4.90521s/12 iters), loss = 2.01565
I0407 22:48:04.387356 32718 solver.cpp:237]     Train net output #0: loss = 2.01565 (* 1 = 2.01565 loss)
I0407 22:48:04.387364 32718 sgd_solver.cpp:105] Iteration 2808, lr = 0.00754643
I0407 22:48:09.339597 32718 solver.cpp:218] Iteration 2820 (2.42317 iter/s, 4.9522s/12 iters), loss = 1.71174
I0407 22:48:09.339658 32718 solver.cpp:237]     Train net output #0: loss = 1.71174 (* 1 = 1.71174 loss)
I0407 22:48:09.339671 32718 sgd_solver.cpp:105] Iteration 2820, lr = 0.00753552
I0407 22:48:13.971949 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:48:14.258059 32718 solver.cpp:218] Iteration 2832 (2.43983 iter/s, 4.91838s/12 iters), loss = 1.60749
I0407 22:48:14.258097 32718 solver.cpp:237]     Train net output #0: loss = 1.60749 (* 1 = 1.60749 loss)
I0407 22:48:14.258105 32718 sgd_solver.cpp:105] Iteration 2832, lr = 0.00752458
I0407 22:48:19.193333 32718 solver.cpp:218] Iteration 2844 (2.43151 iter/s, 4.93521s/12 iters), loss = 2.12352
I0407 22:48:19.193369 32718 solver.cpp:237]     Train net output #0: loss = 2.12352 (* 1 = 2.12352 loss)
I0407 22:48:19.193378 32718 sgd_solver.cpp:105] Iteration 2844, lr = 0.00751361
I0407 22:48:23.679064 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_2856.caffemodel
I0407 22:48:26.797078 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_2856.solverstate
I0407 22:48:29.223147 32718 solver.cpp:330] Iteration 2856, Testing net (#0)
I0407 22:48:29.223163 32718 net.cpp:676] Ignoring source layer train-data
I0407 22:48:32.682091 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:48:33.853653 32718 solver.cpp:397]     Test net output #0: accuracy = 0.33027
I0407 22:48:33.853699 32718 solver.cpp:397]     Test net output #1: loss = 2.91847 (* 1 = 2.91847 loss)
I0407 22:48:33.950435 32718 solver.cpp:218] Iteration 2856 (0.813173 iter/s, 14.757s/12 iters), loss = 1.60135
I0407 22:48:33.950476 32718 solver.cpp:237]     Train net output #0: loss = 1.60135 (* 1 = 1.60135 loss)
I0407 22:48:33.950484 32718 sgd_solver.cpp:105] Iteration 2856, lr = 0.0075026
I0407 22:48:38.088042 32718 solver.cpp:218] Iteration 2868 (2.90027 iter/s, 4.13754s/12 iters), loss = 1.70943
I0407 22:48:38.088079 32718 solver.cpp:237]     Train net output #0: loss = 1.70943 (* 1 = 1.70943 loss)
I0407 22:48:38.088086 32718 sgd_solver.cpp:105] Iteration 2868, lr = 0.00749156
I0407 22:48:43.039762 32718 solver.cpp:218] Iteration 2880 (2.42344 iter/s, 4.95164s/12 iters), loss = 1.91921
I0407 22:48:43.039809 32718 solver.cpp:237]     Train net output #0: loss = 1.91921 (* 1 = 1.91921 loss)
I0407 22:48:43.039817 32718 sgd_solver.cpp:105] Iteration 2880, lr = 0.00748049
I0407 22:48:48.000026 32718 solver.cpp:218] Iteration 2892 (2.41926 iter/s, 4.96019s/12 iters), loss = 2.16244
I0407 22:48:48.000150 32718 solver.cpp:237]     Train net output #0: loss = 2.16244 (* 1 = 2.16244 loss)
I0407 22:48:48.000159 32718 sgd_solver.cpp:105] Iteration 2892, lr = 0.00746939
I0407 22:48:52.938210 32718 solver.cpp:218] Iteration 2904 (2.43012 iter/s, 4.93804s/12 iters), loss = 1.60149
I0407 22:48:52.938246 32718 solver.cpp:237]     Train net output #0: loss = 1.60149 (* 1 = 1.60149 loss)
I0407 22:48:52.938253 32718 sgd_solver.cpp:105] Iteration 2904, lr = 0.00745825
I0407 22:48:57.919857 32718 solver.cpp:218] Iteration 2916 (2.40887 iter/s, 4.98158s/12 iters), loss = 1.3959
I0407 22:48:57.919891 32718 solver.cpp:237]     Train net output #0: loss = 1.3959 (* 1 = 1.3959 loss)
I0407 22:48:57.919899 32718 sgd_solver.cpp:105] Iteration 2916, lr = 0.00744709
I0407 22:49:02.885677 32718 solver.cpp:218] Iteration 2928 (2.41655 iter/s, 4.96575s/12 iters), loss = 1.1766
I0407 22:49:02.885715 32718 solver.cpp:237]     Train net output #0: loss = 1.1766 (* 1 = 1.1766 loss)
I0407 22:49:02.885725 32718 sgd_solver.cpp:105] Iteration 2928, lr = 0.00743589
I0407 22:49:04.684976 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:49:07.786634 32718 solver.cpp:218] Iteration 2940 (2.44854 iter/s, 4.90089s/12 iters), loss = 2.06634
I0407 22:49:07.786670 32718 solver.cpp:237]     Train net output #0: loss = 2.06634 (* 1 = 2.06634 loss)
I0407 22:49:07.786677 32718 sgd_solver.cpp:105] Iteration 2940, lr = 0.00742466
I0407 22:49:12.755826 32718 solver.cpp:218] Iteration 2952 (2.41491 iter/s, 4.96912s/12 iters), loss = 1.74018
I0407 22:49:12.755869 32718 solver.cpp:237]     Train net output #0: loss = 1.74018 (* 1 = 1.74018 loss)
I0407 22:49:12.755877 32718 sgd_solver.cpp:105] Iteration 2952, lr = 0.00741339
I0407 22:49:14.755533 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_2958.caffemodel
I0407 22:49:17.865705 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_2958.solverstate
I0407 22:49:20.234443 32718 solver.cpp:330] Iteration 2958, Testing net (#0)
I0407 22:49:20.234560 32718 net.cpp:676] Ignoring source layer train-data
I0407 22:49:23.756774 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:49:25.071326 32718 solver.cpp:397]     Test net output #0: accuracy = 0.348039
I0407 22:49:25.071370 32718 solver.cpp:397]     Test net output #1: loss = 2.84812 (* 1 = 2.84812 loss)
I0407 22:49:26.865653 32718 solver.cpp:218] Iteration 2964 (0.850477 iter/s, 14.1097s/12 iters), loss = 1.95347
I0407 22:49:26.865695 32718 solver.cpp:237]     Train net output #0: loss = 1.95347 (* 1 = 1.95347 loss)
I0407 22:49:26.865705 32718 sgd_solver.cpp:105] Iteration 2964, lr = 0.0074021
I0407 22:49:31.821452 32718 solver.cpp:218] Iteration 2976 (2.42144 iter/s, 4.95572s/12 iters), loss = 1.40954
I0407 22:49:31.821497 32718 solver.cpp:237]     Train net output #0: loss = 1.40954 (* 1 = 1.40954 loss)
I0407 22:49:31.821506 32718 sgd_solver.cpp:105] Iteration 2976, lr = 0.00739077
I0407 22:49:36.746484 32718 solver.cpp:218] Iteration 2988 (2.43657 iter/s, 4.92496s/12 iters), loss = 1.85518
I0407 22:49:36.746521 32718 solver.cpp:237]     Train net output #0: loss = 1.85518 (* 1 = 1.85518 loss)
I0407 22:49:36.746529 32718 sgd_solver.cpp:105] Iteration 2988, lr = 0.00737941
I0407 22:49:41.719965 32718 solver.cpp:218] Iteration 3000 (2.41283 iter/s, 4.97342s/12 iters), loss = 1.63097
I0407 22:49:41.720006 32718 solver.cpp:237]     Train net output #0: loss = 1.63097 (* 1 = 1.63097 loss)
I0407 22:49:41.720014 32718 sgd_solver.cpp:105] Iteration 3000, lr = 0.00736802
I0407 22:49:46.627593 32718 solver.cpp:218] Iteration 3012 (2.44521 iter/s, 4.90755s/12 iters), loss = 1.36271
I0407 22:49:46.627638 32718 solver.cpp:237]     Train net output #0: loss = 1.36271 (* 1 = 1.36271 loss)
I0407 22:49:46.627646 32718 sgd_solver.cpp:105] Iteration 3012, lr = 0.0073566
I0407 22:49:51.593964 32718 solver.cpp:218] Iteration 3024 (2.41629 iter/s, 4.9663s/12 iters), loss = 1.34942
I0407 22:49:51.594095 32718 solver.cpp:237]     Train net output #0: loss = 1.34942 (* 1 = 1.34942 loss)
I0407 22:49:51.594105 32718 sgd_solver.cpp:105] Iteration 3024, lr = 0.00734514
I0407 22:49:55.493008 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:49:56.496172 32718 solver.cpp:218] Iteration 3036 (2.44796 iter/s, 4.90205s/12 iters), loss = 1.55244
I0407 22:49:56.496214 32718 solver.cpp:237]     Train net output #0: loss = 1.55244 (* 1 = 1.55244 loss)
I0407 22:49:56.496222 32718 sgd_solver.cpp:105] Iteration 3036, lr = 0.00733365
I0407 22:50:01.475682 32718 solver.cpp:218] Iteration 3048 (2.40991 iter/s, 4.97944s/12 iters), loss = 1.44558
I0407 22:50:01.475719 32718 solver.cpp:237]     Train net output #0: loss = 1.44558 (* 1 = 1.44558 loss)
I0407 22:50:01.475728 32718 sgd_solver.cpp:105] Iteration 3048, lr = 0.00732214
I0407 22:50:05.961091 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_3060.caffemodel
I0407 22:50:09.029673 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_3060.solverstate
I0407 22:50:11.385665 32718 solver.cpp:330] Iteration 3060, Testing net (#0)
I0407 22:50:11.385684 32718 net.cpp:676] Ignoring source layer train-data
I0407 22:50:14.826184 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:50:16.170195 32718 solver.cpp:397]     Test net output #0: accuracy = 0.335172
I0407 22:50:16.170222 32718 solver.cpp:397]     Test net output #1: loss = 2.98123 (* 1 = 2.98123 loss)
I0407 22:50:16.266790 32718 solver.cpp:218] Iteration 3060 (0.811303 iter/s, 14.791s/12 iters), loss = 1.5745
I0407 22:50:16.266836 32718 solver.cpp:237]     Train net output #0: loss = 1.5745 (* 1 = 1.5745 loss)
I0407 22:50:16.266844 32718 sgd_solver.cpp:105] Iteration 3060, lr = 0.00731059
I0407 22:50:20.371874 32718 solver.cpp:218] Iteration 3072 (2.92326 iter/s, 4.10501s/12 iters), loss = 1.8227
I0407 22:50:20.371918 32718 solver.cpp:237]     Train net output #0: loss = 1.8227 (* 1 = 1.8227 loss)
I0407 22:50:20.371927 32718 sgd_solver.cpp:105] Iteration 3072, lr = 0.007299
I0407 22:50:25.329890 32718 solver.cpp:218] Iteration 3084 (2.42036 iter/s, 4.95794s/12 iters), loss = 1.50174
I0407 22:50:25.330070 32718 solver.cpp:237]     Train net output #0: loss = 1.50174 (* 1 = 1.50174 loss)
I0407 22:50:25.330080 32718 sgd_solver.cpp:105] Iteration 3084, lr = 0.00728739
I0407 22:50:30.263015 32718 solver.cpp:218] Iteration 3096 (2.43263 iter/s, 4.93292s/12 iters), loss = 1.39029
I0407 22:50:30.263051 32718 solver.cpp:237]     Train net output #0: loss = 1.39029 (* 1 = 1.39029 loss)
I0407 22:50:30.263057 32718 sgd_solver.cpp:105] Iteration 3096, lr = 0.00727575
I0407 22:50:35.159974 32718 solver.cpp:218] Iteration 3108 (2.45053 iter/s, 4.89689s/12 iters), loss = 1.81271
I0407 22:50:35.160015 32718 solver.cpp:237]     Train net output #0: loss = 1.81271 (* 1 = 1.81271 loss)
I0407 22:50:35.160023 32718 sgd_solver.cpp:105] Iteration 3108, lr = 0.00726407
I0407 22:50:40.087992 32718 solver.cpp:218] Iteration 3120 (2.43509 iter/s, 4.92795s/12 iters), loss = 1.44887
I0407 22:50:40.088027 32718 solver.cpp:237]     Train net output #0: loss = 1.44887 (* 1 = 1.44887 loss)
I0407 22:50:40.088033 32718 sgd_solver.cpp:105] Iteration 3120, lr = 0.00725237
I0407 22:50:45.043825 32718 solver.cpp:218] Iteration 3132 (2.42142 iter/s, 4.95577s/12 iters), loss = 1.34386
I0407 22:50:45.043862 32718 solver.cpp:237]     Train net output #0: loss = 1.34386 (* 1 = 1.34386 loss)
I0407 22:50:45.043871 32718 sgd_solver.cpp:105] Iteration 3132, lr = 0.00724063
I0407 22:50:46.115679 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:50:49.935926 32718 solver.cpp:218] Iteration 3144 (2.45297 iter/s, 4.89204s/12 iters), loss = 1.23913
I0407 22:50:49.935961 32718 solver.cpp:237]     Train net output #0: loss = 1.23913 (* 1 = 1.23913 loss)
I0407 22:50:49.935968 32718 sgd_solver.cpp:105] Iteration 3144, lr = 0.00722886
I0407 22:50:54.895689 32718 solver.cpp:218] Iteration 3156 (2.4195 iter/s, 4.9597s/12 iters), loss = 1.55236
I0407 22:50:54.895725 32718 solver.cpp:237]     Train net output #0: loss = 1.55236 (* 1 = 1.55236 loss)
I0407 22:50:54.895732 32718 sgd_solver.cpp:105] Iteration 3156, lr = 0.00721706
I0407 22:50:56.883164 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_3162.caffemodel
I0407 22:51:00.017869 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_3162.solverstate
I0407 22:51:02.380328 32718 solver.cpp:330] Iteration 3162, Testing net (#0)
I0407 22:51:02.380345 32718 net.cpp:676] Ignoring source layer train-data
I0407 22:51:05.759923 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:51:07.153890 32718 solver.cpp:397]     Test net output #0: accuracy = 0.323529
I0407 22:51:07.153928 32718 solver.cpp:397]     Test net output #1: loss = 2.9741 (* 1 = 2.9741 loss)
I0407 22:51:08.957288 32718 solver.cpp:218] Iteration 3168 (0.853394 iter/s, 14.0615s/12 iters), loss = 1.21294
I0407 22:51:08.957330 32718 solver.cpp:237]     Train net output #0: loss = 1.21294 (* 1 = 1.21294 loss)
I0407 22:51:08.957340 32718 sgd_solver.cpp:105] Iteration 3168, lr = 0.00720523
I0407 22:51:13.865893 32718 solver.cpp:218] Iteration 3180 (2.44472 iter/s, 4.90853s/12 iters), loss = 1.25424
I0407 22:51:13.865936 32718 solver.cpp:237]     Train net output #0: loss = 1.25424 (* 1 = 1.25424 loss)
I0407 22:51:13.865943 32718 sgd_solver.cpp:105] Iteration 3180, lr = 0.00719337
I0407 22:51:18.830358 32718 solver.cpp:218] Iteration 3192 (2.41721 iter/s, 4.96439s/12 iters), loss = 1.47882
I0407 22:51:18.830394 32718 solver.cpp:237]     Train net output #0: loss = 1.47882 (* 1 = 1.47882 loss)
I0407 22:51:18.830401 32718 sgd_solver.cpp:105] Iteration 3192, lr = 0.00718148
I0407 22:51:23.765771 32718 solver.cpp:218] Iteration 3204 (2.43144 iter/s, 4.93535s/12 iters), loss = 1.61445
I0407 22:51:23.765810 32718 solver.cpp:237]     Train net output #0: loss = 1.61445 (* 1 = 1.61445 loss)
I0407 22:51:23.765820 32718 sgd_solver.cpp:105] Iteration 3204, lr = 0.00716956
I0407 22:51:28.730113 32718 solver.cpp:218] Iteration 3216 (2.41728 iter/s, 4.96427s/12 iters), loss = 1.11018
I0407 22:51:28.730268 32718 solver.cpp:237]     Train net output #0: loss = 1.11018 (* 1 = 1.11018 loss)
I0407 22:51:28.730278 32718 sgd_solver.cpp:105] Iteration 3216, lr = 0.00715761
I0407 22:51:33.652433 32718 solver.cpp:218] Iteration 3228 (2.43797 iter/s, 4.92213s/12 iters), loss = 1.17618
I0407 22:51:33.652491 32718 solver.cpp:237]     Train net output #0: loss = 1.17618 (* 1 = 1.17618 loss)
I0407 22:51:33.652504 32718 sgd_solver.cpp:105] Iteration 3228, lr = 0.00714562
I0407 22:51:36.877039 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:51:38.608722 32718 solver.cpp:218] Iteration 3240 (2.42121 iter/s, 4.9562s/12 iters), loss = 1.32637
I0407 22:51:38.608763 32718 solver.cpp:237]     Train net output #0: loss = 1.32637 (* 1 = 1.32637 loss)
I0407 22:51:38.608772 32718 sgd_solver.cpp:105] Iteration 3240, lr = 0.00713361
I0407 22:51:43.541678 32718 solver.cpp:218] Iteration 3252 (2.43265 iter/s, 4.93289s/12 iters), loss = 1.21438
I0407 22:51:43.541714 32718 solver.cpp:237]     Train net output #0: loss = 1.21438 (* 1 = 1.21438 loss)
I0407 22:51:43.541721 32718 sgd_solver.cpp:105] Iteration 3252, lr = 0.00712157
I0407 22:51:48.051538 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_3264.caffemodel
I0407 22:51:51.376585 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_3264.solverstate
I0407 22:51:53.788971 32718 solver.cpp:330] Iteration 3264, Testing net (#0)
I0407 22:51:53.788990 32718 net.cpp:676] Ignoring source layer train-data
I0407 22:51:57.123473 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:51:58.473462 32718 solver.cpp:397]     Test net output #0: accuracy = 0.352941
I0407 22:51:58.473510 32718 solver.cpp:397]     Test net output #1: loss = 2.90155 (* 1 = 2.90155 loss)
I0407 22:51:58.568980 32718 solver.cpp:218] Iteration 3264 (0.798552 iter/s, 15.0272s/12 iters), loss = 1.30138
I0407 22:51:58.569028 32718 solver.cpp:237]     Train net output #0: loss = 1.30138 (* 1 = 1.30138 loss)
I0407 22:51:58.569036 32718 sgd_solver.cpp:105] Iteration 3264, lr = 0.00710949
I0407 22:52:02.700711 32718 solver.cpp:218] Iteration 3276 (2.9044 iter/s, 4.13166s/12 iters), loss = 1.31484
I0407 22:52:02.700826 32718 solver.cpp:237]     Train net output #0: loss = 1.31484 (* 1 = 1.31484 loss)
I0407 22:52:02.700835 32718 sgd_solver.cpp:105] Iteration 3276, lr = 0.00709739
I0407 22:52:07.694125 32718 solver.cpp:218] Iteration 3288 (2.40323 iter/s, 4.99327s/12 iters), loss = 1.10653
I0407 22:52:07.694164 32718 solver.cpp:237]     Train net output #0: loss = 1.10653 (* 1 = 1.10653 loss)
I0407 22:52:07.694172 32718 sgd_solver.cpp:105] Iteration 3288, lr = 0.00708526
I0407 22:52:12.679921 32718 solver.cpp:218] Iteration 3300 (2.40687 iter/s, 4.98573s/12 iters), loss = 1.35511
I0407 22:52:12.679957 32718 solver.cpp:237]     Train net output #0: loss = 1.35511 (* 1 = 1.35511 loss)
I0407 22:52:12.679965 32718 sgd_solver.cpp:105] Iteration 3300, lr = 0.0070731
I0407 22:52:17.620538 32718 solver.cpp:218] Iteration 3312 (2.42888 iter/s, 4.94055s/12 iters), loss = 1.49553
I0407 22:52:17.620575 32718 solver.cpp:237]     Train net output #0: loss = 1.49553 (* 1 = 1.49553 loss)
I0407 22:52:17.620584 32718 sgd_solver.cpp:105] Iteration 3312, lr = 0.0070609
I0407 22:52:22.566428 32718 solver.cpp:218] Iteration 3324 (2.42629 iter/s, 4.94582s/12 iters), loss = 1.49163
I0407 22:52:22.566466 32718 solver.cpp:237]     Train net output #0: loss = 1.49163 (* 1 = 1.49163 loss)
I0407 22:52:22.566473 32718 sgd_solver.cpp:105] Iteration 3324, lr = 0.00704868
I0407 22:52:27.508591 32718 solver.cpp:218] Iteration 3336 (2.42812 iter/s, 4.94209s/12 iters), loss = 1.20858
I0407 22:52:27.508633 32718 solver.cpp:237]     Train net output #0: loss = 1.20858 (* 1 = 1.20858 loss)
I0407 22:52:27.508642 32718 sgd_solver.cpp:105] Iteration 3336, lr = 0.00703643
I0407 22:52:27.957657 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:52:32.443492 32718 solver.cpp:218] Iteration 3348 (2.4317 iter/s, 4.93483s/12 iters), loss = 1.54842
I0407 22:52:32.443534 32718 solver.cpp:237]     Train net output #0: loss = 1.54842 (* 1 = 1.54842 loss)
I0407 22:52:32.443543 32718 sgd_solver.cpp:105] Iteration 3348, lr = 0.00702415
I0407 22:52:37.413424 32718 solver.cpp:218] Iteration 3360 (2.41455 iter/s, 4.96986s/12 iters), loss = 1.38588
I0407 22:52:37.413532 32718 solver.cpp:237]     Train net output #0: loss = 1.38588 (* 1 = 1.38588 loss)
I0407 22:52:37.413540 32718 sgd_solver.cpp:105] Iteration 3360, lr = 0.00701184
I0407 22:52:39.414008 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_3366.caffemodel
I0407 22:52:43.833631 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_3366.solverstate
I0407 22:52:47.050057 32718 solver.cpp:330] Iteration 3366, Testing net (#0)
I0407 22:52:47.050078 32718 net.cpp:676] Ignoring source layer train-data
I0407 22:52:50.160955 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:52:51.515159 32718 solver.cpp:397]     Test net output #0: accuracy = 0.340074
I0407 22:52:51.515233 32718 solver.cpp:397]     Test net output #1: loss = 2.85935 (* 1 = 2.85935 loss)
I0407 22:52:53.302055 32718 solver.cpp:218] Iteration 3372 (0.755265 iter/s, 15.8885s/12 iters), loss = 1.26435
I0407 22:52:53.302088 32718 solver.cpp:237]     Train net output #0: loss = 1.26435 (* 1 = 1.26435 loss)
I0407 22:52:53.302095 32718 sgd_solver.cpp:105] Iteration 3372, lr = 0.0069995
I0407 22:52:58.265169 32718 solver.cpp:218] Iteration 3384 (2.41787 iter/s, 4.96305s/12 iters), loss = 1.23419
I0407 22:52:58.265202 32718 solver.cpp:237]     Train net output #0: loss = 1.23419 (* 1 = 1.23419 loss)
I0407 22:52:58.265209 32718 sgd_solver.cpp:105] Iteration 3384, lr = 0.00698713
I0407 22:53:03.216053 32718 solver.cpp:218] Iteration 3396 (2.42384 iter/s, 4.95082s/12 iters), loss = 1.3397
I0407 22:53:03.216082 32718 solver.cpp:237]     Train net output #0: loss = 1.3397 (* 1 = 1.3397 loss)
I0407 22:53:03.216089 32718 sgd_solver.cpp:105] Iteration 3396, lr = 0.00697473
I0407 22:53:08.140911 32718 solver.cpp:218] Iteration 3408 (2.43665 iter/s, 4.9248s/12 iters), loss = 1.24972
I0407 22:53:08.141018 32718 solver.cpp:237]     Train net output #0: loss = 1.24972 (* 1 = 1.24972 loss)
I0407 22:53:08.141026 32718 sgd_solver.cpp:105] Iteration 3408, lr = 0.00696231
I0407 22:53:13.093631 32718 solver.cpp:218] Iteration 3420 (2.42298 iter/s, 4.95258s/12 iters), loss = 1.22369
I0407 22:53:13.093667 32718 solver.cpp:237]     Train net output #0: loss = 1.22369 (* 1 = 1.22369 loss)
I0407 22:53:13.093674 32718 sgd_solver.cpp:105] Iteration 3420, lr = 0.00694985
I0407 22:53:18.011979 32718 solver.cpp:218] Iteration 3432 (2.43987 iter/s, 4.91829s/12 iters), loss = 1.31111
I0407 22:53:18.012013 32718 solver.cpp:237]     Train net output #0: loss = 1.31111 (* 1 = 1.31111 loss)
I0407 22:53:18.012019 32718 sgd_solver.cpp:105] Iteration 3432, lr = 0.00693737
I0407 22:53:20.563860 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:53:22.960076 32718 solver.cpp:218] Iteration 3444 (2.42521 iter/s, 4.94803s/12 iters), loss = 1.27195
I0407 22:53:22.960114 32718 solver.cpp:237]     Train net output #0: loss = 1.27195 (* 1 = 1.27195 loss)
I0407 22:53:22.960120 32718 sgd_solver.cpp:105] Iteration 3444, lr = 0.00692485
I0407 22:53:27.889295 32718 solver.cpp:218] Iteration 3456 (2.4345 iter/s, 4.92915s/12 iters), loss = 1.34395
I0407 22:53:27.889336 32718 solver.cpp:237]     Train net output #0: loss = 1.34395 (* 1 = 1.34395 loss)
I0407 22:53:27.889344 32718 sgd_solver.cpp:105] Iteration 3456, lr = 0.00691231
I0407 22:53:32.386034 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_3468.caffemodel
I0407 22:53:37.078589 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_3468.solverstate
I0407 22:53:40.596930 32718 solver.cpp:330] Iteration 3468, Testing net (#0)
I0407 22:53:40.596987 32718 net.cpp:676] Ignoring source layer train-data
I0407 22:53:41.013741 32718 blocking_queue.cpp:49] Waiting for data
I0407 22:53:43.955284 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:53:45.505234 32718 solver.cpp:397]     Test net output #0: accuracy = 0.362745
I0407 22:53:45.505280 32718 solver.cpp:397]     Test net output #1: loss = 2.78083 (* 1 = 2.78083 loss)
I0407 22:53:45.601858 32718 solver.cpp:218] Iteration 3468 (0.677489 iter/s, 17.7125s/12 iters), loss = 1.22972
I0407 22:53:45.601899 32718 solver.cpp:237]     Train net output #0: loss = 1.22972 (* 1 = 1.22972 loss)
I0407 22:53:45.601907 32718 sgd_solver.cpp:105] Iteration 3468, lr = 0.00689974
I0407 22:53:49.661754 32718 solver.cpp:218] Iteration 3480 (2.95579 iter/s, 4.05982s/12 iters), loss = 1.16343
I0407 22:53:49.661798 32718 solver.cpp:237]     Train net output #0: loss = 1.16343 (* 1 = 1.16343 loss)
I0407 22:53:49.661806 32718 sgd_solver.cpp:105] Iteration 3480, lr = 0.00688715
I0407 22:53:54.599104 32718 solver.cpp:218] Iteration 3492 (2.43049 iter/s, 4.93728s/12 iters), loss = 0.962498
I0407 22:53:54.599141 32718 solver.cpp:237]     Train net output #0: loss = 0.962498 (* 1 = 0.962498 loss)
I0407 22:53:54.599148 32718 sgd_solver.cpp:105] Iteration 3492, lr = 0.00687452
I0407 22:53:59.590759 32718 solver.cpp:218] Iteration 3504 (2.40404 iter/s, 4.99159s/12 iters), loss = 1.44779
I0407 22:53:59.590801 32718 solver.cpp:237]     Train net output #0: loss = 1.44779 (* 1 = 1.44779 loss)
I0407 22:53:59.590811 32718 sgd_solver.cpp:105] Iteration 3504, lr = 0.00686187
I0407 22:54:04.505568 32718 solver.cpp:218] Iteration 3516 (2.44164 iter/s, 4.91474s/12 iters), loss = 1.18196
I0407 22:54:04.505602 32718 solver.cpp:237]     Train net output #0: loss = 1.18196 (* 1 = 1.18196 loss)
I0407 22:54:04.505609 32718 sgd_solver.cpp:105] Iteration 3516, lr = 0.00684919
I0407 22:54:09.491995 32718 solver.cpp:218] Iteration 3528 (2.40656 iter/s, 4.98636s/12 iters), loss = 1.08436
I0407 22:54:09.492031 32718 solver.cpp:237]     Train net output #0: loss = 1.08436 (* 1 = 1.08436 loss)
I0407 22:54:09.492039 32718 sgd_solver.cpp:105] Iteration 3528, lr = 0.00683648
I0407 22:54:14.080852 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:54:14.337935 32718 solver.cpp:218] Iteration 3540 (2.47633 iter/s, 4.84587s/12 iters), loss = 0.946487
I0407 22:54:14.337970 32718 solver.cpp:237]     Train net output #0: loss = 0.946487 (* 1 = 0.946487 loss)
I0407 22:54:14.337977 32718 sgd_solver.cpp:105] Iteration 3540, lr = 0.00682375
I0407 22:54:19.303233 32718 solver.cpp:218] Iteration 3552 (2.41681 iter/s, 4.96523s/12 iters), loss = 1.0789
I0407 22:54:19.303265 32718 solver.cpp:237]     Train net output #0: loss = 1.0789 (* 1 = 1.0789 loss)
I0407 22:54:19.303272 32718 sgd_solver.cpp:105] Iteration 3552, lr = 0.00681098
I0407 22:54:24.240447 32718 solver.cpp:218] Iteration 3564 (2.43055 iter/s, 4.93716s/12 iters), loss = 0.889279
I0407 22:54:24.240481 32718 solver.cpp:237]     Train net output #0: loss = 0.889279 (* 1 = 0.889279 loss)
I0407 22:54:24.240489 32718 sgd_solver.cpp:105] Iteration 3564, lr = 0.00679819
I0407 22:54:26.266434 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_3570.caffemodel
I0407 22:54:29.358181 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_3570.solverstate
I0407 22:54:32.322702 32718 solver.cpp:330] Iteration 3570, Testing net (#0)
I0407 22:54:32.322719 32718 net.cpp:676] Ignoring source layer train-data
I0407 22:54:35.307337 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:54:36.798521 32718 solver.cpp:397]     Test net output #0: accuracy = 0.353554
I0407 22:54:36.798566 32718 solver.cpp:397]     Test net output #1: loss = 3.00005 (* 1 = 3.00005 loss)
I0407 22:54:38.582823 32718 solver.cpp:218] Iteration 3576 (0.836687 iter/s, 14.3423s/12 iters), loss = 1.22438
I0407 22:54:38.582865 32718 solver.cpp:237]     Train net output #0: loss = 1.22438 (* 1 = 1.22438 loss)
I0407 22:54:38.582872 32718 sgd_solver.cpp:105] Iteration 3576, lr = 0.00678538
I0407 22:54:43.511123 32718 solver.cpp:218] Iteration 3588 (2.43495 iter/s, 4.92823s/12 iters), loss = 1.06235
I0407 22:54:43.511160 32718 solver.cpp:237]     Train net output #0: loss = 1.06235 (* 1 = 1.06235 loss)
I0407 22:54:43.511168 32718 sgd_solver.cpp:105] Iteration 3588, lr = 0.00677253
I0407 22:54:48.482667 32718 solver.cpp:218] Iteration 3600 (2.41377 iter/s, 4.97148s/12 iters), loss = 1.41386
I0407 22:54:48.482818 32718 solver.cpp:237]     Train net output #0: loss = 1.41386 (* 1 = 1.41386 loss)
I0407 22:54:48.482827 32718 sgd_solver.cpp:105] Iteration 3600, lr = 0.00675966
I0407 22:54:53.517355 32718 solver.cpp:218] Iteration 3612 (2.38355 iter/s, 5.03451s/12 iters), loss = 1.2517
I0407 22:54:53.517393 32718 solver.cpp:237]     Train net output #0: loss = 1.2517 (* 1 = 1.2517 loss)
I0407 22:54:53.517401 32718 sgd_solver.cpp:105] Iteration 3612, lr = 0.00674676
I0407 22:54:58.510754 32718 solver.cpp:218] Iteration 3624 (2.4032 iter/s, 4.99333s/12 iters), loss = 0.942894
I0407 22:54:58.510790 32718 solver.cpp:237]     Train net output #0: loss = 0.942894 (* 1 = 0.942894 loss)
I0407 22:54:58.510798 32718 sgd_solver.cpp:105] Iteration 3624, lr = 0.00673384
I0407 22:55:03.459013 32718 solver.cpp:218] Iteration 3636 (2.42513 iter/s, 4.94819s/12 iters), loss = 0.923541
I0407 22:55:03.459051 32718 solver.cpp:237]     Train net output #0: loss = 0.923541 (* 1 = 0.923541 loss)
I0407 22:55:03.459060 32718 sgd_solver.cpp:105] Iteration 3636, lr = 0.00672089
I0407 22:55:05.343448 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:55:08.449609 32718 solver.cpp:218] Iteration 3648 (2.40455 iter/s, 4.99053s/12 iters), loss = 1.21268
I0407 22:55:08.449636 32718 solver.cpp:237]     Train net output #0: loss = 1.21268 (* 1 = 1.21268 loss)
I0407 22:55:08.449642 32718 sgd_solver.cpp:105] Iteration 3648, lr = 0.00670791
I0407 22:55:13.479907 32718 solver.cpp:218] Iteration 3660 (2.38558 iter/s, 5.03023s/12 iters), loss = 1.03576
I0407 22:55:13.479951 32718 solver.cpp:237]     Train net output #0: loss = 1.03576 (* 1 = 1.03576 loss)
I0407 22:55:13.479959 32718 sgd_solver.cpp:105] Iteration 3660, lr = 0.00669491
I0407 22:55:17.977111 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_3672.caffemodel
I0407 22:55:21.080324 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_3672.solverstate
I0407 22:55:23.475968 32718 solver.cpp:330] Iteration 3672, Testing net (#0)
I0407 22:55:23.475986 32718 net.cpp:676] Ignoring source layer train-data
I0407 22:55:26.619868 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:55:28.110383 32718 solver.cpp:397]     Test net output #0: accuracy = 0.382966
I0407 22:55:28.110430 32718 solver.cpp:397]     Test net output #1: loss = 2.79243 (* 1 = 2.79243 loss)
I0407 22:55:28.207072 32718 solver.cpp:218] Iteration 3672 (0.814827 iter/s, 14.7271s/12 iters), loss = 1.0235
I0407 22:55:28.207130 32718 solver.cpp:237]     Train net output #0: loss = 1.0235 (* 1 = 1.0235 loss)
I0407 22:55:28.207142 32718 sgd_solver.cpp:105] Iteration 3672, lr = 0.00668188
I0407 22:55:32.320886 32718 solver.cpp:218] Iteration 3684 (2.91706 iter/s, 4.11374s/12 iters), loss = 1.02783
I0407 22:55:32.320921 32718 solver.cpp:237]     Train net output #0: loss = 1.02783 (* 1 = 1.02783 loss)
I0407 22:55:32.320930 32718 sgd_solver.cpp:105] Iteration 3684, lr = 0.00666882
I0407 22:55:37.253809 32718 solver.cpp:218] Iteration 3696 (2.43267 iter/s, 4.93286s/12 iters), loss = 0.99319
I0407 22:55:37.253849 32718 solver.cpp:237]     Train net output #0: loss = 0.99319 (* 1 = 0.99319 loss)
I0407 22:55:37.253855 32718 sgd_solver.cpp:105] Iteration 3696, lr = 0.00665574
I0407 22:55:42.422322 32718 solver.cpp:218] Iteration 3708 (2.32178 iter/s, 5.16844s/12 iters), loss = 1.07129
I0407 22:55:42.422363 32718 solver.cpp:237]     Train net output #0: loss = 1.07129 (* 1 = 1.07129 loss)
I0407 22:55:42.422371 32718 sgd_solver.cpp:105] Iteration 3708, lr = 0.00664264
I0407 22:55:47.430461 32718 solver.cpp:218] Iteration 3720 (2.39613 iter/s, 5.00807s/12 iters), loss = 0.816455
I0407 22:55:47.430507 32718 solver.cpp:237]     Train net output #0: loss = 0.816455 (* 1 = 0.816455 loss)
I0407 22:55:47.430516 32718 sgd_solver.cpp:105] Iteration 3720, lr = 0.00662951
I0407 22:55:52.389318 32718 solver.cpp:218] Iteration 3732 (2.41995 iter/s, 4.95878s/12 iters), loss = 1.08382
I0407 22:55:52.389467 32718 solver.cpp:237]     Train net output #0: loss = 1.08382 (* 1 = 1.08382 loss)
I0407 22:55:52.389477 32718 sgd_solver.cpp:105] Iteration 3732, lr = 0.00661635
I0407 22:55:56.350759 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:55:57.318233 32718 solver.cpp:218] Iteration 3744 (2.4347 iter/s, 4.92874s/12 iters), loss = 1.15315
I0407 22:55:57.318276 32718 solver.cpp:237]     Train net output #0: loss = 1.15315 (* 1 = 1.15315 loss)
I0407 22:55:57.318285 32718 sgd_solver.cpp:105] Iteration 3744, lr = 0.00660317
I0407 22:56:02.297087 32718 solver.cpp:218] Iteration 3756 (2.41023 iter/s, 4.97878s/12 iters), loss = 0.993454
I0407 22:56:02.297123 32718 solver.cpp:237]     Train net output #0: loss = 0.993454 (* 1 = 0.993454 loss)
I0407 22:56:02.297132 32718 sgd_solver.cpp:105] Iteration 3756, lr = 0.00658996
I0407 22:56:07.273684 32718 solver.cpp:218] Iteration 3768 (2.41132 iter/s, 4.97653s/12 iters), loss = 1.13791
I0407 22:56:07.273721 32718 solver.cpp:237]     Train net output #0: loss = 1.13791 (* 1 = 1.13791 loss)
I0407 22:56:07.273730 32718 sgd_solver.cpp:105] Iteration 3768, lr = 0.00657673
I0407 22:56:09.293299 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_3774.caffemodel
I0407 22:56:12.397639 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_3774.solverstate
I0407 22:56:14.799185 32718 solver.cpp:330] Iteration 3774, Testing net (#0)
I0407 22:56:14.799207 32718 net.cpp:676] Ignoring source layer train-data
I0407 22:56:17.710839 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:56:19.233510 32718 solver.cpp:397]     Test net output #0: accuracy = 0.34375
I0407 22:56:19.233557 32718 solver.cpp:397]     Test net output #1: loss = 2.9181 (* 1 = 2.9181 loss)
I0407 22:56:21.045267 32718 solver.cpp:218] Iteration 3780 (0.871365 iter/s, 13.7715s/12 iters), loss = 1.31191
I0407 22:56:21.045307 32718 solver.cpp:237]     Train net output #0: loss = 1.31191 (* 1 = 1.31191 loss)
I0407 22:56:21.045315 32718 sgd_solver.cpp:105] Iteration 3780, lr = 0.00656347
I0407 22:56:26.021593 32718 solver.cpp:218] Iteration 3792 (2.41145 iter/s, 4.97625s/12 iters), loss = 0.899677
I0407 22:56:26.021687 32718 solver.cpp:237]     Train net output #0: loss = 0.899677 (* 1 = 0.899677 loss)
I0407 22:56:26.021695 32718 sgd_solver.cpp:105] Iteration 3792, lr = 0.00655019
I0407 22:56:31.061904 32718 solver.cpp:218] Iteration 3804 (2.38086 iter/s, 5.04019s/12 iters), loss = 0.903151
I0407 22:56:31.061942 32718 solver.cpp:237]     Train net output #0: loss = 0.903151 (* 1 = 0.903151 loss)
I0407 22:56:31.061950 32718 sgd_solver.cpp:105] Iteration 3804, lr = 0.00653689
I0407 22:56:35.997985 32718 solver.cpp:218] Iteration 3816 (2.43111 iter/s, 4.93601s/12 iters), loss = 1.02435
I0407 22:56:35.998026 32718 solver.cpp:237]     Train net output #0: loss = 1.02435 (* 1 = 1.02435 loss)
I0407 22:56:35.998034 32718 sgd_solver.cpp:105] Iteration 3816, lr = 0.00652356
I0407 22:56:40.977126 32718 solver.cpp:218] Iteration 3828 (2.41009 iter/s, 4.97907s/12 iters), loss = 0.857411
I0407 22:56:40.977169 32718 solver.cpp:237]     Train net output #0: loss = 0.857411 (* 1 = 0.857411 loss)
I0407 22:56:40.977177 32718 sgd_solver.cpp:105] Iteration 3828, lr = 0.00651021
I0407 22:56:45.948088 32718 solver.cpp:218] Iteration 3840 (2.41406 iter/s, 4.97089s/12 iters), loss = 0.889354
I0407 22:56:45.948133 32718 solver.cpp:237]     Train net output #0: loss = 0.889354 (* 1 = 0.889354 loss)
I0407 22:56:45.948143 32718 sgd_solver.cpp:105] Iteration 3840, lr = 0.00649683
I0407 22:56:47.054481 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:56:50.849339 32718 solver.cpp:218] Iteration 3852 (2.44839 iter/s, 4.90117s/12 iters), loss = 1.06307
I0407 22:56:50.849385 32718 solver.cpp:237]     Train net output #0: loss = 1.06307 (* 1 = 1.06307 loss)
I0407 22:56:50.849392 32718 sgd_solver.cpp:105] Iteration 3852, lr = 0.00648343
I0407 22:56:55.796479 32718 solver.cpp:218] Iteration 3864 (2.42568 iter/s, 4.94707s/12 iters), loss = 0.775202
I0407 22:56:55.796519 32718 solver.cpp:237]     Train net output #0: loss = 0.775202 (* 1 = 0.775202 loss)
I0407 22:56:55.796526 32718 sgd_solver.cpp:105] Iteration 3864, lr = 0.00647001
I0407 22:57:00.257714 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_3876.caffemodel
I0407 22:57:03.327404 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_3876.solverstate
I0407 22:57:05.697715 32718 solver.cpp:330] Iteration 3876, Testing net (#0)
I0407 22:57:05.697732 32718 net.cpp:676] Ignoring source layer train-data
I0407 22:57:08.759795 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:57:10.310288 32718 solver.cpp:397]     Test net output #0: accuracy = 0.36826
I0407 22:57:10.310334 32718 solver.cpp:397]     Test net output #1: loss = 2.94577 (* 1 = 2.94577 loss)
I0407 22:57:10.406980 32718 solver.cpp:218] Iteration 3876 (0.821332 iter/s, 14.6104s/12 iters), loss = 1.09111
I0407 22:57:10.407024 32718 solver.cpp:237]     Train net output #0: loss = 1.09111 (* 1 = 1.09111 loss)
I0407 22:57:10.407033 32718 sgd_solver.cpp:105] Iteration 3876, lr = 0.00645656
I0407 22:57:14.519975 32718 solver.cpp:218] Iteration 3888 (2.91763 iter/s, 4.11292s/12 iters), loss = 0.703704
I0407 22:57:14.520020 32718 solver.cpp:237]     Train net output #0: loss = 0.703704 (* 1 = 0.703704 loss)
I0407 22:57:14.520027 32718 sgd_solver.cpp:105] Iteration 3888, lr = 0.00644309
I0407 22:57:19.443481 32718 solver.cpp:218] Iteration 3900 (2.43732 iter/s, 4.92343s/12 iters), loss = 0.842711
I0407 22:57:19.443527 32718 solver.cpp:237]     Train net output #0: loss = 0.842711 (* 1 = 0.842711 loss)
I0407 22:57:19.443536 32718 sgd_solver.cpp:105] Iteration 3900, lr = 0.0064296
I0407 22:57:24.411145 32718 solver.cpp:218] Iteration 3912 (2.41566 iter/s, 4.96759s/12 iters), loss = 1.09325
I0407 22:57:24.411185 32718 solver.cpp:237]     Train net output #0: loss = 1.09325 (* 1 = 1.09325 loss)
I0407 22:57:24.411192 32718 sgd_solver.cpp:105] Iteration 3912, lr = 0.00641609
I0407 22:57:29.317859 32718 solver.cpp:218] Iteration 3924 (2.44566 iter/s, 4.90664s/12 iters), loss = 0.913379
I0407 22:57:29.317903 32718 solver.cpp:237]     Train net output #0: loss = 0.913379 (* 1 = 0.913379 loss)
I0407 22:57:29.317910 32718 sgd_solver.cpp:105] Iteration 3924, lr = 0.00640255
I0407 22:57:34.265446 32718 solver.cpp:218] Iteration 3936 (2.42546 iter/s, 4.94752s/12 iters), loss = 0.838419
I0407 22:57:34.265554 32718 solver.cpp:237]     Train net output #0: loss = 0.838419 (* 1 = 0.838419 loss)
I0407 22:57:34.265563 32718 sgd_solver.cpp:105] Iteration 3936, lr = 0.00638899
I0407 22:57:37.569576 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:57:39.169837 32718 solver.cpp:218] Iteration 3948 (2.44685 iter/s, 4.90426s/12 iters), loss = 0.937886
I0407 22:57:39.169874 32718 solver.cpp:237]     Train net output #0: loss = 0.937886 (* 1 = 0.937886 loss)
I0407 22:57:39.169883 32718 sgd_solver.cpp:105] Iteration 3948, lr = 0.00637541
I0407 22:57:44.129997 32718 solver.cpp:218] Iteration 3960 (2.41931 iter/s, 4.9601s/12 iters), loss = 1.05682
I0407 22:57:44.130033 32718 solver.cpp:237]     Train net output #0: loss = 1.05682 (* 1 = 1.05682 loss)
I0407 22:57:44.130041 32718 sgd_solver.cpp:105] Iteration 3960, lr = 0.0063618
I0407 22:57:49.078214 32718 solver.cpp:218] Iteration 3972 (2.42515 iter/s, 4.94816s/12 iters), loss = 0.911508
I0407 22:57:49.078254 32718 solver.cpp:237]     Train net output #0: loss = 0.911508 (* 1 = 0.911508 loss)
I0407 22:57:49.078263 32718 sgd_solver.cpp:105] Iteration 3972, lr = 0.00634818
I0407 22:57:51.071239 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_3978.caffemodel
I0407 22:57:54.171285 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_3978.solverstate
I0407 22:57:56.548487 32718 solver.cpp:330] Iteration 3978, Testing net (#0)
I0407 22:57:56.548506 32718 net.cpp:676] Ignoring source layer train-data
I0407 22:57:59.401404 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:58:01.060564 32718 solver.cpp:397]     Test net output #0: accuracy = 0.407476
I0407 22:58:01.060607 32718 solver.cpp:397]     Test net output #1: loss = 2.77958 (* 1 = 2.77958 loss)
I0407 22:58:02.853452 32718 solver.cpp:218] Iteration 3984 (0.871134 iter/s, 13.7751s/12 iters), loss = 0.874666
I0407 22:58:02.853500 32718 solver.cpp:237]     Train net output #0: loss = 0.874666 (* 1 = 0.874666 loss)
I0407 22:58:02.853509 32718 sgd_solver.cpp:105] Iteration 3984, lr = 0.00633453
I0407 22:58:07.810863 32718 solver.cpp:218] Iteration 3996 (2.42065 iter/s, 4.95734s/12 iters), loss = 0.90057
I0407 22:58:07.811015 32718 solver.cpp:237]     Train net output #0: loss = 0.90057 (* 1 = 0.90057 loss)
I0407 22:58:07.811024 32718 sgd_solver.cpp:105] Iteration 3996, lr = 0.00632086
I0407 22:58:12.730525 32718 solver.cpp:218] Iteration 4008 (2.43928 iter/s, 4.91949s/12 iters), loss = 0.759469
I0407 22:58:12.730562 32718 solver.cpp:237]     Train net output #0: loss = 0.759469 (* 1 = 0.759469 loss)
I0407 22:58:12.730571 32718 sgd_solver.cpp:105] Iteration 4008, lr = 0.00630717
I0407 22:58:17.689056 32718 solver.cpp:218] Iteration 4020 (2.4201 iter/s, 4.95847s/12 iters), loss = 0.932925
I0407 22:58:17.689097 32718 solver.cpp:237]     Train net output #0: loss = 0.932925 (* 1 = 0.932925 loss)
I0407 22:58:17.689105 32718 sgd_solver.cpp:105] Iteration 4020, lr = 0.00629346
I0407 22:58:22.645659 32718 solver.cpp:218] Iteration 4032 (2.42105 iter/s, 4.95653s/12 iters), loss = 0.827329
I0407 22:58:22.645705 32718 solver.cpp:237]     Train net output #0: loss = 0.827329 (* 1 = 0.827329 loss)
I0407 22:58:22.645715 32718 sgd_solver.cpp:105] Iteration 4032, lr = 0.00627973
I0407 22:58:27.645452 32718 solver.cpp:218] Iteration 4044 (2.40013 iter/s, 4.99972s/12 iters), loss = 0.766334
I0407 22:58:27.645493 32718 solver.cpp:237]     Train net output #0: loss = 0.766334 (* 1 = 0.766334 loss)
I0407 22:58:27.645503 32718 sgd_solver.cpp:105] Iteration 4044, lr = 0.00626597
I0407 22:58:28.122980 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:58:32.596467 32718 solver.cpp:218] Iteration 4056 (2.42378 iter/s, 4.95094s/12 iters), loss = 1.38067
I0407 22:58:32.596510 32718 solver.cpp:237]     Train net output #0: loss = 1.38067 (* 1 = 1.38067 loss)
I0407 22:58:32.596519 32718 sgd_solver.cpp:105] Iteration 4056, lr = 0.0062522
I0407 22:58:37.556883 32718 solver.cpp:218] Iteration 4068 (2.41919 iter/s, 4.96035s/12 iters), loss = 0.668496
I0407 22:58:37.556919 32718 solver.cpp:237]     Train net output #0: loss = 0.668496 (* 1 = 0.668496 loss)
I0407 22:58:37.556927 32718 sgd_solver.cpp:105] Iteration 4068, lr = 0.00623841
I0407 22:58:41.886654 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_4080.caffemodel
I0407 22:58:44.973259 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_4080.solverstate
I0407 22:58:47.386766 32718 solver.cpp:330] Iteration 4080, Testing net (#0)
I0407 22:58:47.386783 32718 net.cpp:676] Ignoring source layer train-data
I0407 22:58:50.407755 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:58:52.182080 32718 solver.cpp:397]     Test net output #0: accuracy = 0.389706
I0407 22:58:52.182124 32718 solver.cpp:397]     Test net output #1: loss = 2.79132 (* 1 = 2.79132 loss)
I0407 22:58:52.278614 32718 solver.cpp:218] Iteration 4080 (0.815126 iter/s, 14.7216s/12 iters), loss = 0.790554
I0407 22:58:52.278656 32718 solver.cpp:237]     Train net output #0: loss = 0.790554 (* 1 = 0.790554 loss)
I0407 22:58:52.278664 32718 sgd_solver.cpp:105] Iteration 4080, lr = 0.00622459
I0407 22:58:56.447135 32718 solver.cpp:218] Iteration 4092 (2.87877 iter/s, 4.16845s/12 iters), loss = 0.772398
I0407 22:58:56.447180 32718 solver.cpp:237]     Train net output #0: loss = 0.772398 (* 1 = 0.772398 loss)
I0407 22:58:56.447188 32718 sgd_solver.cpp:105] Iteration 4092, lr = 0.00621076
I0407 22:59:01.408865 32718 solver.cpp:218] Iteration 4104 (2.41855 iter/s, 4.96165s/12 iters), loss = 0.895351
I0407 22:59:01.408910 32718 solver.cpp:237]     Train net output #0: loss = 0.895351 (* 1 = 0.895351 loss)
I0407 22:59:01.408918 32718 sgd_solver.cpp:105] Iteration 4104, lr = 0.00619691
I0407 22:59:06.416368 32718 solver.cpp:218] Iteration 4116 (2.39644 iter/s, 5.00744s/12 iters), loss = 0.930109
I0407 22:59:06.416404 32718 solver.cpp:237]     Train net output #0: loss = 0.930109 (* 1 = 0.930109 loss)
I0407 22:59:06.416410 32718 sgd_solver.cpp:105] Iteration 4116, lr = 0.00618303
I0407 22:59:11.425076 32718 solver.cpp:218] Iteration 4128 (2.39586 iter/s, 5.00864s/12 iters), loss = 0.828333
I0407 22:59:11.425115 32718 solver.cpp:237]     Train net output #0: loss = 0.828333 (* 1 = 0.828333 loss)
I0407 22:59:11.425123 32718 sgd_solver.cpp:105] Iteration 4128, lr = 0.00616914
I0407 22:59:16.396162 32718 solver.cpp:218] Iteration 4140 (2.41399 iter/s, 4.97102s/12 iters), loss = 0.717254
I0407 22:59:16.396288 32718 solver.cpp:237]     Train net output #0: loss = 0.717254 (* 1 = 0.717254 loss)
I0407 22:59:16.396298 32718 sgd_solver.cpp:105] Iteration 4140, lr = 0.00615523
I0407 22:59:18.979099 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:59:21.298804 32718 solver.cpp:218] Iteration 4152 (2.44774 iter/s, 4.90249s/12 iters), loss = 0.589708
I0407 22:59:21.298848 32718 solver.cpp:237]     Train net output #0: loss = 0.589708 (* 1 = 0.589708 loss)
I0407 22:59:21.298856 32718 sgd_solver.cpp:105] Iteration 4152, lr = 0.0061413
I0407 22:59:22.879806 32718 blocking_queue.cpp:49] Waiting for data
I0407 22:59:26.259426 32718 solver.cpp:218] Iteration 4164 (2.41908 iter/s, 4.96055s/12 iters), loss = 0.813596
I0407 22:59:26.259462 32718 solver.cpp:237]     Train net output #0: loss = 0.813596 (* 1 = 0.813596 loss)
I0407 22:59:26.259470 32718 sgd_solver.cpp:105] Iteration 4164, lr = 0.00612735
I0407 22:59:31.168186 32718 solver.cpp:218] Iteration 4176 (2.44464 iter/s, 4.9087s/12 iters), loss = 0.762031
I0407 22:59:31.168223 32718 solver.cpp:237]     Train net output #0: loss = 0.762031 (* 1 = 0.762031 loss)
I0407 22:59:31.168231 32718 sgd_solver.cpp:105] Iteration 4176, lr = 0.00611338
I0407 22:59:33.154845 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_4182.caffemodel
I0407 22:59:36.246282 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_4182.solverstate
I0407 22:59:38.665925 32718 solver.cpp:330] Iteration 4182, Testing net (#0)
I0407 22:59:38.665942 32718 net.cpp:676] Ignoring source layer train-data
I0407 22:59:41.713333 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 22:59:43.471762 32718 solver.cpp:397]     Test net output #0: accuracy = 0.408088
I0407 22:59:43.471807 32718 solver.cpp:397]     Test net output #1: loss = 2.74274 (* 1 = 2.74274 loss)
I0407 22:59:45.302588 32718 solver.cpp:218] Iteration 4188 (0.848998 iter/s, 14.1343s/12 iters), loss = 0.863449
I0407 22:59:45.302626 32718 solver.cpp:237]     Train net output #0: loss = 0.863449 (* 1 = 0.863449 loss)
I0407 22:59:45.302634 32718 sgd_solver.cpp:105] Iteration 4188, lr = 0.0060994
I0407 22:59:50.294224 32718 solver.cpp:218] Iteration 4200 (2.40405 iter/s, 4.99157s/12 iters), loss = 0.815781
I0407 22:59:50.294365 32718 solver.cpp:237]     Train net output #0: loss = 0.815781 (* 1 = 0.815781 loss)
I0407 22:59:50.294374 32718 sgd_solver.cpp:105] Iteration 4200, lr = 0.00608539
I0407 22:59:55.295661 32718 solver.cpp:218] Iteration 4212 (2.39939 iter/s, 5.00128s/12 iters), loss = 0.703529
I0407 22:59:55.295696 32718 solver.cpp:237]     Train net output #0: loss = 0.703529 (* 1 = 0.703529 loss)
I0407 22:59:55.295704 32718 sgd_solver.cpp:105] Iteration 4212, lr = 0.00607137
I0407 23:00:00.130764 32718 solver.cpp:218] Iteration 4224 (2.48188 iter/s, 4.83504s/12 iters), loss = 1.201
I0407 23:00:00.130812 32718 solver.cpp:237]     Train net output #0: loss = 1.201 (* 1 = 1.201 loss)
I0407 23:00:00.130820 32718 sgd_solver.cpp:105] Iteration 4224, lr = 0.00605733
I0407 23:00:05.104816 32718 solver.cpp:218] Iteration 4236 (2.41256 iter/s, 4.97398s/12 iters), loss = 0.931851
I0407 23:00:05.104857 32718 solver.cpp:237]     Train net output #0: loss = 0.931851 (* 1 = 0.931851 loss)
I0407 23:00:05.104866 32718 sgd_solver.cpp:105] Iteration 4236, lr = 0.00604327
I0407 23:00:09.773797 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:00:10.000805 32718 solver.cpp:218] Iteration 4248 (2.45102 iter/s, 4.89592s/12 iters), loss = 0.908264
I0407 23:00:10.000850 32718 solver.cpp:237]     Train net output #0: loss = 0.908264 (* 1 = 0.908264 loss)
I0407 23:00:10.000859 32718 sgd_solver.cpp:105] Iteration 4248, lr = 0.0060292
I0407 23:00:14.977360 32718 solver.cpp:218] Iteration 4260 (2.41135 iter/s, 4.97647s/12 iters), loss = 0.847157
I0407 23:00:14.977416 32718 solver.cpp:237]     Train net output #0: loss = 0.847157 (* 1 = 0.847157 loss)
I0407 23:00:14.977429 32718 sgd_solver.cpp:105] Iteration 4260, lr = 0.00601511
I0407 23:00:19.932128 32718 solver.cpp:218] Iteration 4272 (2.42195 iter/s, 4.95469s/12 iters), loss = 0.730753
I0407 23:00:19.932165 32718 solver.cpp:237]     Train net output #0: loss = 0.730753 (* 1 = 0.730753 loss)
I0407 23:00:19.932173 32718 sgd_solver.cpp:105] Iteration 4272, lr = 0.006001
I0407 23:00:24.420094 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_4284.caffemodel
I0407 23:00:28.620337 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_4284.solverstate
I0407 23:00:31.011293 32718 solver.cpp:330] Iteration 4284, Testing net (#0)
I0407 23:00:31.011312 32718 net.cpp:676] Ignoring source layer train-data
I0407 23:00:33.930657 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:00:35.785959 32718 solver.cpp:397]     Test net output #0: accuracy = 0.410539
I0407 23:00:35.786006 32718 solver.cpp:397]     Test net output #1: loss = 2.73907 (* 1 = 2.73907 loss)
I0407 23:00:35.882639 32718 solver.cpp:218] Iteration 4284 (0.752331 iter/s, 15.9504s/12 iters), loss = 0.644884
I0407 23:00:35.882684 32718 solver.cpp:237]     Train net output #0: loss = 0.644884 (* 1 = 0.644884 loss)
I0407 23:00:35.882694 32718 sgd_solver.cpp:105] Iteration 4284, lr = 0.00598688
I0407 23:00:40.104425 32718 solver.cpp:218] Iteration 4296 (2.84244 iter/s, 4.22172s/12 iters), loss = 0.434566
I0407 23:00:40.104462 32718 solver.cpp:237]     Train net output #0: loss = 0.434566 (* 1 = 0.434566 loss)
I0407 23:00:40.104470 32718 sgd_solver.cpp:105] Iteration 4296, lr = 0.00597274
I0407 23:00:45.109764 32718 solver.cpp:218] Iteration 4308 (2.39747 iter/s, 5.00527s/12 iters), loss = 0.694076
I0407 23:00:45.109804 32718 solver.cpp:237]     Train net output #0: loss = 0.694076 (* 1 = 0.694076 loss)
I0407 23:00:45.109814 32718 sgd_solver.cpp:105] Iteration 4308, lr = 0.00595858
I0407 23:00:50.037951 32718 solver.cpp:218] Iteration 4320 (2.43501 iter/s, 4.92812s/12 iters), loss = 0.752912
I0407 23:00:50.037992 32718 solver.cpp:237]     Train net output #0: loss = 0.752912 (* 1 = 0.752912 loss)
I0407 23:00:50.037999 32718 sgd_solver.cpp:105] Iteration 4320, lr = 0.0059444
I0407 23:00:55.023200 32718 solver.cpp:218] Iteration 4332 (2.40714 iter/s, 4.98517s/12 iters), loss = 0.653205
I0407 23:00:55.023346 32718 solver.cpp:237]     Train net output #0: loss = 0.653205 (* 1 = 0.653205 loss)
I0407 23:00:55.023356 32718 sgd_solver.cpp:105] Iteration 4332, lr = 0.00593022
I0407 23:00:59.997934 32718 solver.cpp:218] Iteration 4344 (2.41227 iter/s, 4.97456s/12 iters), loss = 0.69018
I0407 23:00:59.997979 32718 solver.cpp:237]     Train net output #0: loss = 0.69018 (* 1 = 0.69018 loss)
I0407 23:00:59.997987 32718 sgd_solver.cpp:105] Iteration 4344, lr = 0.00591601
I0407 23:01:01.859086 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:01:04.912873 32718 solver.cpp:218] Iteration 4356 (2.44157 iter/s, 4.91486s/12 iters), loss = 0.616321
I0407 23:01:04.912916 32718 solver.cpp:237]     Train net output #0: loss = 0.616321 (* 1 = 0.616321 loss)
I0407 23:01:04.912925 32718 sgd_solver.cpp:105] Iteration 4356, lr = 0.00590179
I0407 23:01:09.817785 32718 solver.cpp:218] Iteration 4368 (2.44656 iter/s, 4.90484s/12 iters), loss = 0.800664
I0407 23:01:09.817832 32718 solver.cpp:237]     Train net output #0: loss = 0.800664 (* 1 = 0.800664 loss)
I0407 23:01:09.817842 32718 sgd_solver.cpp:105] Iteration 4368, lr = 0.00588756
I0407 23:01:14.855509 32718 solver.cpp:218] Iteration 4380 (2.38206 iter/s, 5.03765s/12 iters), loss = 0.723156
I0407 23:01:14.855546 32718 solver.cpp:237]     Train net output #0: loss = 0.723156 (* 1 = 0.723156 loss)
I0407 23:01:14.855554 32718 sgd_solver.cpp:105] Iteration 4380, lr = 0.00587331
I0407 23:01:16.915756 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_4386.caffemodel
I0407 23:01:21.259976 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_4386.solverstate
I0407 23:01:26.594055 32718 solver.cpp:330] Iteration 4386, Testing net (#0)
I0407 23:01:26.594163 32718 net.cpp:676] Ignoring source layer train-data
I0407 23:01:29.356676 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:01:31.120225 32718 solver.cpp:397]     Test net output #0: accuracy = 0.422181
I0407 23:01:31.120250 32718 solver.cpp:397]     Test net output #1: loss = 2.79237 (* 1 = 2.79237 loss)
I0407 23:01:32.925869 32718 solver.cpp:218] Iteration 4392 (0.664075 iter/s, 18.0703s/12 iters), loss = 0.539267
I0407 23:01:32.925915 32718 solver.cpp:237]     Train net output #0: loss = 0.539267 (* 1 = 0.539267 loss)
I0407 23:01:32.925925 32718 sgd_solver.cpp:105] Iteration 4392, lr = 0.00585904
I0407 23:01:37.875988 32718 solver.cpp:218] Iteration 4404 (2.42422 iter/s, 4.95005s/12 iters), loss = 0.9484
I0407 23:01:37.876026 32718 solver.cpp:237]     Train net output #0: loss = 0.9484 (* 1 = 0.9484 loss)
I0407 23:01:37.876034 32718 sgd_solver.cpp:105] Iteration 4404, lr = 0.00584476
I0407 23:01:42.843683 32718 solver.cpp:218] Iteration 4416 (2.41564 iter/s, 4.96763s/12 iters), loss = 0.566697
I0407 23:01:42.843726 32718 solver.cpp:237]     Train net output #0: loss = 0.566697 (* 1 = 0.566697 loss)
I0407 23:01:42.843735 32718 sgd_solver.cpp:105] Iteration 4416, lr = 0.00583047
I0407 23:01:47.753899 32718 solver.cpp:218] Iteration 4428 (2.44392 iter/s, 4.91014s/12 iters), loss = 0.639526
I0407 23:01:47.753939 32718 solver.cpp:237]     Train net output #0: loss = 0.639526 (* 1 = 0.639526 loss)
I0407 23:01:47.753947 32718 sgd_solver.cpp:105] Iteration 4428, lr = 0.00581616
I0407 23:01:52.736768 32718 solver.cpp:218] Iteration 4440 (2.40829 iter/s, 4.9828s/12 iters), loss = 0.573495
I0407 23:01:52.736809 32718 solver.cpp:237]     Train net output #0: loss = 0.573495 (* 1 = 0.573495 loss)
I0407 23:01:52.736816 32718 sgd_solver.cpp:105] Iteration 4440, lr = 0.00580184
I0407 23:01:56.702853 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:01:57.655265 32718 solver.cpp:218] Iteration 4452 (2.4398 iter/s, 4.91843s/12 iters), loss = 0.500764
I0407 23:01:57.655305 32718 solver.cpp:237]     Train net output #0: loss = 0.500764 (* 1 = 0.500764 loss)
I0407 23:01:57.655313 32718 sgd_solver.cpp:105] Iteration 4452, lr = 0.00578751
I0407 23:02:02.644146 32718 solver.cpp:218] Iteration 4464 (2.40538 iter/s, 4.98882s/12 iters), loss = 0.650251
I0407 23:02:02.644181 32718 solver.cpp:237]     Train net output #0: loss = 0.650251 (* 1 = 0.650251 loss)
I0407 23:02:02.644188 32718 sgd_solver.cpp:105] Iteration 4464, lr = 0.00577316
I0407 23:02:07.613387 32718 solver.cpp:218] Iteration 4476 (2.41489 iter/s, 4.96918s/12 iters), loss = 0.619571
I0407 23:02:07.613430 32718 solver.cpp:237]     Train net output #0: loss = 0.619571 (* 1 = 0.619571 loss)
I0407 23:02:07.613438 32718 sgd_solver.cpp:105] Iteration 4476, lr = 0.0057588
I0407 23:02:12.074224 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_4488.caffemodel
I0407 23:02:15.175438 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_4488.solverstate
I0407 23:02:17.970979 32718 solver.cpp:330] Iteration 4488, Testing net (#0)
I0407 23:02:17.970999 32718 net.cpp:676] Ignoring source layer train-data
I0407 23:02:20.814456 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:02:22.767851 32718 solver.cpp:397]     Test net output #0: accuracy = 0.417279
I0407 23:02:22.767894 32718 solver.cpp:397]     Test net output #1: loss = 2.79697 (* 1 = 2.79697 loss)
I0407 23:02:22.864405 32718 solver.cpp:218] Iteration 4488 (0.786838 iter/s, 15.2509s/12 iters), loss = 0.639104
I0407 23:02:22.864452 32718 solver.cpp:237]     Train net output #0: loss = 0.639104 (* 1 = 0.639104 loss)
I0407 23:02:22.864459 32718 sgd_solver.cpp:105] Iteration 4488, lr = 0.00574443
I0407 23:02:26.995247 32718 solver.cpp:218] Iteration 4500 (2.90503 iter/s, 4.13077s/12 iters), loss = 0.435451
I0407 23:02:26.995370 32718 solver.cpp:237]     Train net output #0: loss = 0.435451 (* 1 = 0.435451 loss)
I0407 23:02:26.995381 32718 sgd_solver.cpp:105] Iteration 4500, lr = 0.00573004
I0407 23:02:31.949501 32718 solver.cpp:218] Iteration 4512 (2.42223 iter/s, 4.95411s/12 iters), loss = 0.708946
I0407 23:02:31.949538 32718 solver.cpp:237]     Train net output #0: loss = 0.708946 (* 1 = 0.708946 loss)
I0407 23:02:31.949546 32718 sgd_solver.cpp:105] Iteration 4512, lr = 0.00571564
I0407 23:02:36.875913 32718 solver.cpp:218] Iteration 4524 (2.43588 iter/s, 4.92635s/12 iters), loss = 0.600295
I0407 23:02:36.875958 32718 solver.cpp:237]     Train net output #0: loss = 0.600295 (* 1 = 0.600295 loss)
I0407 23:02:36.875967 32718 sgd_solver.cpp:105] Iteration 4524, lr = 0.00570123
I0407 23:02:41.861909 32718 solver.cpp:218] Iteration 4536 (2.40678 iter/s, 4.98592s/12 iters), loss = 0.966789
I0407 23:02:41.861949 32718 solver.cpp:237]     Train net output #0: loss = 0.966789 (* 1 = 0.966789 loss)
I0407 23:02:41.861958 32718 sgd_solver.cpp:105] Iteration 4536, lr = 0.00568681
I0407 23:02:46.830744 32718 solver.cpp:218] Iteration 4548 (2.41509 iter/s, 4.96877s/12 iters), loss = 0.534585
I0407 23:02:46.830782 32718 solver.cpp:237]     Train net output #0: loss = 0.534585 (* 1 = 0.534585 loss)
I0407 23:02:46.830791 32718 sgd_solver.cpp:105] Iteration 4548, lr = 0.00567237
I0407 23:02:48.055864 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:02:51.728072 32718 solver.cpp:218] Iteration 4560 (2.45035 iter/s, 4.89726s/12 iters), loss = 0.482049
I0407 23:02:51.728122 32718 solver.cpp:237]     Train net output #0: loss = 0.482049 (* 1 = 0.482049 loss)
I0407 23:02:51.728130 32718 sgd_solver.cpp:105] Iteration 4560, lr = 0.00565793
I0407 23:02:56.703907 32718 solver.cpp:218] Iteration 4572 (2.41169 iter/s, 4.97576s/12 iters), loss = 0.567612
I0407 23:02:56.703944 32718 solver.cpp:237]     Train net output #0: loss = 0.567611 (* 1 = 0.567611 loss)
I0407 23:02:56.703953 32718 sgd_solver.cpp:105] Iteration 4572, lr = 0.00564347
I0407 23:03:01.606660 32718 solver.cpp:218] Iteration 4584 (2.44764 iter/s, 4.90269s/12 iters), loss = 0.442943
I0407 23:03:01.606818 32718 solver.cpp:237]     Train net output #0: loss = 0.442943 (* 1 = 0.442943 loss)
I0407 23:03:01.606828 32718 sgd_solver.cpp:105] Iteration 4584, lr = 0.005629
I0407 23:03:03.605697 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_4590.caffemodel
I0407 23:03:07.125408 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_4590.solverstate
I0407 23:03:10.836084 32718 solver.cpp:330] Iteration 4590, Testing net (#0)
I0407 23:03:10.836103 32718 net.cpp:676] Ignoring source layer train-data
I0407 23:03:13.623032 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:03:15.589386 32718 solver.cpp:397]     Test net output #0: accuracy = 0.408088
I0407 23:03:15.589423 32718 solver.cpp:397]     Test net output #1: loss = 2.83106 (* 1 = 2.83106 loss)
I0407 23:03:17.390148 32718 solver.cpp:218] Iteration 4596 (0.760299 iter/s, 15.7833s/12 iters), loss = 0.398185
I0407 23:03:17.390197 32718 solver.cpp:237]     Train net output #0: loss = 0.398185 (* 1 = 0.398185 loss)
I0407 23:03:17.390204 32718 sgd_solver.cpp:105] Iteration 4596, lr = 0.00561452
I0407 23:03:22.318971 32718 solver.cpp:218] Iteration 4608 (2.4347 iter/s, 4.92875s/12 iters), loss = 0.518338
I0407 23:03:22.319010 32718 solver.cpp:237]     Train net output #0: loss = 0.518338 (* 1 = 0.518338 loss)
I0407 23:03:22.319017 32718 sgd_solver.cpp:105] Iteration 4608, lr = 0.00560004
I0407 23:03:27.283866 32718 solver.cpp:218] Iteration 4620 (2.417 iter/s, 4.96482s/12 iters), loss = 0.712403
I0407 23:03:27.283907 32718 solver.cpp:237]     Train net output #0: loss = 0.712403 (* 1 = 0.712403 loss)
I0407 23:03:27.283915 32718 sgd_solver.cpp:105] Iteration 4620, lr = 0.00558554
I0407 23:03:32.209993 32718 solver.cpp:218] Iteration 4632 (2.43602 iter/s, 4.92606s/12 iters), loss = 0.29218
I0407 23:03:32.210111 32718 solver.cpp:237]     Train net output #0: loss = 0.29218 (* 1 = 0.29218 loss)
I0407 23:03:32.210119 32718 sgd_solver.cpp:105] Iteration 4632, lr = 0.00557103
I0407 23:03:37.158493 32718 solver.cpp:218] Iteration 4644 (2.42505 iter/s, 4.94836s/12 iters), loss = 0.497633
I0407 23:03:37.158531 32718 solver.cpp:237]     Train net output #0: loss = 0.497633 (* 1 = 0.497633 loss)
I0407 23:03:37.158540 32718 sgd_solver.cpp:105] Iteration 4644, lr = 0.00555651
I0407 23:03:40.493163 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:03:42.099344 32718 solver.cpp:218] Iteration 4656 (2.42877 iter/s, 4.94078s/12 iters), loss = 0.366723
I0407 23:03:42.099380 32718 solver.cpp:237]     Train net output #0: loss = 0.366723 (* 1 = 0.366723 loss)
I0407 23:03:42.099387 32718 sgd_solver.cpp:105] Iteration 4656, lr = 0.00554198
I0407 23:03:47.038570 32718 solver.cpp:218] Iteration 4668 (2.42956 iter/s, 4.93916s/12 iters), loss = 0.430308
I0407 23:03:47.038612 32718 solver.cpp:237]     Train net output #0: loss = 0.430308 (* 1 = 0.430308 loss)
I0407 23:03:47.038620 32718 sgd_solver.cpp:105] Iteration 4668, lr = 0.00552744
I0407 23:03:52.000350 32718 solver.cpp:218] Iteration 4680 (2.41852 iter/s, 4.96171s/12 iters), loss = 0.452207
I0407 23:03:52.000392 32718 solver.cpp:237]     Train net output #0: loss = 0.452207 (* 1 = 0.452207 loss)
I0407 23:03:52.000399 32718 sgd_solver.cpp:105] Iteration 4680, lr = 0.0055129
I0407 23:03:56.458542 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_4692.caffemodel
I0407 23:03:59.556593 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_4692.solverstate
I0407 23:04:01.919319 32718 solver.cpp:330] Iteration 4692, Testing net (#0)
I0407 23:04:01.919338 32718 net.cpp:676] Ignoring source layer train-data
I0407 23:04:04.639719 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:04:06.527236 32718 solver.cpp:397]     Test net output #0: accuracy = 0.433824
I0407 23:04:06.527281 32718 solver.cpp:397]     Test net output #1: loss = 2.7845 (* 1 = 2.7845 loss)
I0407 23:04:06.624274 32718 solver.cpp:218] Iteration 4692 (0.820579 iter/s, 14.6238s/12 iters), loss = 0.56411
I0407 23:04:06.624318 32718 solver.cpp:237]     Train net output #0: loss = 0.56411 (* 1 = 0.56411 loss)
I0407 23:04:06.624327 32718 sgd_solver.cpp:105] Iteration 4692, lr = 0.00549834
I0407 23:04:10.766512 32718 solver.cpp:218] Iteration 4704 (2.89703 iter/s, 4.14217s/12 iters), loss = 0.517035
I0407 23:04:10.766551 32718 solver.cpp:237]     Train net output #0: loss = 0.517035 (* 1 = 0.517035 loss)
I0407 23:04:10.766559 32718 sgd_solver.cpp:105] Iteration 4704, lr = 0.00548378
I0407 23:04:15.643218 32718 solver.cpp:218] Iteration 4716 (2.46071 iter/s, 4.87664s/12 iters), loss = 0.327994
I0407 23:04:15.643257 32718 solver.cpp:237]     Train net output #0: loss = 0.327994 (* 1 = 0.327994 loss)
I0407 23:04:15.643265 32718 sgd_solver.cpp:105] Iteration 4716, lr = 0.0054692
I0407 23:04:20.541070 32718 solver.cpp:218] Iteration 4728 (2.45009 iter/s, 4.89778s/12 iters), loss = 0.425193
I0407 23:04:20.541111 32718 solver.cpp:237]     Train net output #0: loss = 0.425193 (* 1 = 0.425193 loss)
I0407 23:04:20.541119 32718 sgd_solver.cpp:105] Iteration 4728, lr = 0.00545462
I0407 23:04:25.471937 32718 solver.cpp:218] Iteration 4740 (2.43369 iter/s, 4.93079s/12 iters), loss = 0.519157
I0407 23:04:25.471982 32718 solver.cpp:237]     Train net output #0: loss = 0.519157 (* 1 = 0.519157 loss)
I0407 23:04:25.471990 32718 sgd_solver.cpp:105] Iteration 4740, lr = 0.00544003
I0407 23:04:30.441917 32718 solver.cpp:218] Iteration 4752 (2.41453 iter/s, 4.9699s/12 iters), loss = 0.628195
I0407 23:04:30.441962 32718 solver.cpp:237]     Train net output #0: loss = 0.628194 (* 1 = 0.628194 loss)
I0407 23:04:30.441969 32718 sgd_solver.cpp:105] Iteration 4752, lr = 0.00542544
I0407 23:04:30.949649 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:04:35.296705 32718 solver.cpp:218] Iteration 4764 (2.47183 iter/s, 4.85471s/12 iters), loss = 0.506479
I0407 23:04:35.296859 32718 solver.cpp:237]     Train net output #0: loss = 0.506479 (* 1 = 0.506479 loss)
I0407 23:04:35.296869 32718 sgd_solver.cpp:105] Iteration 4764, lr = 0.00541084
I0407 23:04:40.283408 32718 solver.cpp:218] Iteration 4776 (2.40649 iter/s, 4.98652s/12 iters), loss = 0.498028
I0407 23:04:40.283457 32718 solver.cpp:237]     Train net output #0: loss = 0.498027 (* 1 = 0.498027 loss)
I0407 23:04:40.283464 32718 sgd_solver.cpp:105] Iteration 4776, lr = 0.00539623
I0407 23:04:45.202618 32718 solver.cpp:218] Iteration 4788 (2.43945 iter/s, 4.91914s/12 iters), loss = 0.584043
I0407 23:04:45.202657 32718 solver.cpp:237]     Train net output #0: loss = 0.584043 (* 1 = 0.584043 loss)
I0407 23:04:45.202666 32718 sgd_solver.cpp:105] Iteration 4788, lr = 0.00538161
I0407 23:04:47.204723 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_4794.caffemodel
I0407 23:04:50.288926 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_4794.solverstate
I0407 23:04:53.779008 32718 solver.cpp:330] Iteration 4794, Testing net (#0)
I0407 23:04:53.779024 32718 net.cpp:676] Ignoring source layer train-data
I0407 23:04:56.564221 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:04:58.719349 32718 solver.cpp:397]     Test net output #0: accuracy = 0.415441
I0407 23:04:58.719390 32718 solver.cpp:397]     Test net output #1: loss = 2.8976 (* 1 = 2.8976 loss)
I0407 23:05:00.505964 32718 solver.cpp:218] Iteration 4800 (0.784147 iter/s, 15.3032s/12 iters), loss = 0.443823
I0407 23:05:00.506009 32718 solver.cpp:237]     Train net output #0: loss = 0.443823 (* 1 = 0.443823 loss)
I0407 23:05:00.506017 32718 sgd_solver.cpp:105] Iteration 4800, lr = 0.00536699
I0407 23:05:05.472918 32718 solver.cpp:218] Iteration 4812 (2.416 iter/s, 4.96688s/12 iters), loss = 0.598662
I0407 23:05:05.473043 32718 solver.cpp:237]     Train net output #0: loss = 0.598662 (* 1 = 0.598662 loss)
I0407 23:05:05.473052 32718 sgd_solver.cpp:105] Iteration 4812, lr = 0.00535236
I0407 23:05:10.432941 32718 solver.cpp:218] Iteration 4824 (2.41942 iter/s, 4.95987s/12 iters), loss = 0.72441
I0407 23:05:10.432981 32718 solver.cpp:237]     Train net output #0: loss = 0.72441 (* 1 = 0.72441 loss)
I0407 23:05:10.432989 32718 sgd_solver.cpp:105] Iteration 4824, lr = 0.00533772
I0407 23:05:15.363165 32718 solver.cpp:218] Iteration 4836 (2.434 iter/s, 4.93015s/12 iters), loss = 0.400541
I0407 23:05:15.363214 32718 solver.cpp:237]     Train net output #0: loss = 0.400541 (* 1 = 0.400541 loss)
I0407 23:05:15.363221 32718 sgd_solver.cpp:105] Iteration 4836, lr = 0.00532308
I0407 23:05:17.382731 32718 blocking_queue.cpp:49] Waiting for data
I0407 23:05:20.327533 32718 solver.cpp:218] Iteration 4848 (2.41726 iter/s, 4.96429s/12 iters), loss = 0.511023
I0407 23:05:20.327577 32718 solver.cpp:237]     Train net output #0: loss = 0.511022 (* 1 = 0.511022 loss)
I0407 23:05:20.327585 32718 sgd_solver.cpp:105] Iteration 4848, lr = 0.00530843
I0407 23:05:22.946800 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:05:25.240224 32718 solver.cpp:218] Iteration 4860 (2.44269 iter/s, 4.91262s/12 iters), loss = 0.370205
I0407 23:05:25.240267 32718 solver.cpp:237]     Train net output #0: loss = 0.370205 (* 1 = 0.370205 loss)
I0407 23:05:25.240276 32718 sgd_solver.cpp:105] Iteration 4860, lr = 0.00529378
I0407 23:05:30.218160 32718 solver.cpp:218] Iteration 4872 (2.41067 iter/s, 4.97786s/12 iters), loss = 0.438118
I0407 23:05:30.218207 32718 solver.cpp:237]     Train net output #0: loss = 0.438117 (* 1 = 0.438117 loss)
I0407 23:05:30.218215 32718 sgd_solver.cpp:105] Iteration 4872, lr = 0.00527912
I0407 23:05:35.198772 32718 solver.cpp:218] Iteration 4884 (2.40938 iter/s, 4.98054s/12 iters), loss = 0.654821
I0407 23:05:35.198809 32718 solver.cpp:237]     Train net output #0: loss = 0.654821 (* 1 = 0.654821 loss)
I0407 23:05:35.198817 32718 sgd_solver.cpp:105] Iteration 4884, lr = 0.00526446
I0407 23:05:39.656437 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_4896.caffemodel
I0407 23:05:43.221417 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_4896.solverstate
I0407 23:05:46.158545 32718 solver.cpp:330] Iteration 4896, Testing net (#0)
I0407 23:05:46.158561 32718 net.cpp:676] Ignoring source layer train-data
I0407 23:05:48.760529 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:05:50.732287 32718 solver.cpp:397]     Test net output #0: accuracy = 0.428922
I0407 23:05:50.732334 32718 solver.cpp:397]     Test net output #1: loss = 2.91033 (* 1 = 2.91033 loss)
I0407 23:05:50.828876 32718 solver.cpp:218] Iteration 4896 (0.767754 iter/s, 15.63s/12 iters), loss = 0.60927
I0407 23:05:50.828918 32718 solver.cpp:237]     Train net output #0: loss = 0.609269 (* 1 = 0.609269 loss)
I0407 23:05:50.828927 32718 sgd_solver.cpp:105] Iteration 4896, lr = 0.00524979
I0407 23:05:54.946517 32718 solver.cpp:218] Iteration 4908 (2.91434 iter/s, 4.11757s/12 iters), loss = 0.41203
I0407 23:05:54.946555 32718 solver.cpp:237]     Train net output #0: loss = 0.41203 (* 1 = 0.41203 loss)
I0407 23:05:54.946563 32718 sgd_solver.cpp:105] Iteration 4908, lr = 0.00523512
I0407 23:05:59.941251 32718 solver.cpp:218] Iteration 4920 (2.40256 iter/s, 4.99467s/12 iters), loss = 0.700749
I0407 23:05:59.941294 32718 solver.cpp:237]     Train net output #0: loss = 0.700749 (* 1 = 0.700749 loss)
I0407 23:05:59.941303 32718 sgd_solver.cpp:105] Iteration 4920, lr = 0.00522045
I0407 23:06:04.867993 32718 solver.cpp:218] Iteration 4932 (2.43572 iter/s, 4.92667s/12 iters), loss = 0.374582
I0407 23:06:04.868037 32718 solver.cpp:237]     Train net output #0: loss = 0.374582 (* 1 = 0.374582 loss)
I0407 23:06:04.868046 32718 sgd_solver.cpp:105] Iteration 4932, lr = 0.00520577
I0407 23:06:09.868149 32718 solver.cpp:218] Iteration 4944 (2.39996 iter/s, 5.00008s/12 iters), loss = 0.638677
I0407 23:06:09.868275 32718 solver.cpp:237]     Train net output #0: loss = 0.638677 (* 1 = 0.638677 loss)
I0407 23:06:09.868284 32718 sgd_solver.cpp:105] Iteration 4944, lr = 0.00519108
I0407 23:06:14.613795 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:06:14.810921 32718 solver.cpp:218] Iteration 4956 (2.42787 iter/s, 4.94261s/12 iters), loss = 0.297619
I0407 23:06:14.810966 32718 solver.cpp:237]     Train net output #0: loss = 0.297619 (* 1 = 0.297619 loss)
I0407 23:06:14.810973 32718 sgd_solver.cpp:105] Iteration 4956, lr = 0.0051764
I0407 23:06:19.724983 32718 solver.cpp:218] Iteration 4968 (2.44201 iter/s, 4.91399s/12 iters), loss = 0.284668
I0407 23:06:19.725028 32718 solver.cpp:237]     Train net output #0: loss = 0.284668 (* 1 = 0.284668 loss)
I0407 23:06:19.725035 32718 sgd_solver.cpp:105] Iteration 4968, lr = 0.00516171
I0407 23:06:24.686628 32718 solver.cpp:218] Iteration 4980 (2.41859 iter/s, 4.96157s/12 iters), loss = 0.341253
I0407 23:06:24.686666 32718 solver.cpp:237]     Train net output #0: loss = 0.341253 (* 1 = 0.341253 loss)
I0407 23:06:24.686676 32718 sgd_solver.cpp:105] Iteration 4980, lr = 0.00514702
I0407 23:06:29.631749 32718 solver.cpp:218] Iteration 4992 (2.42667 iter/s, 4.94505s/12 iters), loss = 0.480901
I0407 23:06:29.631803 32718 solver.cpp:237]     Train net output #0: loss = 0.480901 (* 1 = 0.480901 loss)
I0407 23:06:29.631814 32718 sgd_solver.cpp:105] Iteration 4992, lr = 0.00513232
I0407 23:06:31.653298 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_4998.caffemodel
I0407 23:06:35.957798 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_4998.solverstate
I0407 23:06:39.281777 32718 solver.cpp:330] Iteration 4998, Testing net (#0)
I0407 23:06:39.281797 32718 net.cpp:676] Ignoring source layer train-data
I0407 23:06:41.937072 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:06:44.005645 32718 solver.cpp:397]     Test net output #0: accuracy = 0.41973
I0407 23:06:44.005693 32718 solver.cpp:397]     Test net output #1: loss = 2.95927 (* 1 = 2.95927 loss)
I0407 23:06:45.821683 32718 solver.cpp:218] Iteration 5004 (0.741206 iter/s, 16.1898s/12 iters), loss = 0.487982
I0407 23:06:45.821722 32718 solver.cpp:237]     Train net output #0: loss = 0.487982 (* 1 = 0.487982 loss)
I0407 23:06:45.821729 32718 sgd_solver.cpp:105] Iteration 5004, lr = 0.00511763
I0407 23:06:50.739399 32718 solver.cpp:218] Iteration 5016 (2.44019 iter/s, 4.91765s/12 iters), loss = 0.548959
I0407 23:06:50.739441 32718 solver.cpp:237]     Train net output #0: loss = 0.548959 (* 1 = 0.548959 loss)
I0407 23:06:50.739450 32718 sgd_solver.cpp:105] Iteration 5016, lr = 0.00510293
I0407 23:06:55.697069 32718 solver.cpp:218] Iteration 5028 (2.42053 iter/s, 4.95759s/12 iters), loss = 0.653746
I0407 23:06:55.697109 32718 solver.cpp:237]     Train net output #0: loss = 0.653746 (* 1 = 0.653746 loss)
I0407 23:06:55.697118 32718 sgd_solver.cpp:105] Iteration 5028, lr = 0.00508823
I0407 23:07:00.642675 32718 solver.cpp:218] Iteration 5040 (2.42643 iter/s, 4.94553s/12 iters), loss = 0.375206
I0407 23:07:00.642719 32718 solver.cpp:237]     Train net output #0: loss = 0.375206 (* 1 = 0.375206 loss)
I0407 23:07:00.642727 32718 sgd_solver.cpp:105] Iteration 5040, lr = 0.00507352
I0407 23:07:05.591671 32718 solver.cpp:218] Iteration 5052 (2.42477 iter/s, 4.94892s/12 iters), loss = 0.382983
I0407 23:07:05.591714 32718 solver.cpp:237]     Train net output #0: loss = 0.382983 (* 1 = 0.382983 loss)
I0407 23:07:05.591722 32718 sgd_solver.cpp:105] Iteration 5052, lr = 0.00505882
I0407 23:07:07.485702 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:07:10.546347 32718 solver.cpp:218] Iteration 5064 (2.42199 iter/s, 4.9546s/12 iters), loss = 0.390391
I0407 23:07:10.546389 32718 solver.cpp:237]     Train net output #0: loss = 0.390391 (* 1 = 0.390391 loss)
I0407 23:07:10.546397 32718 sgd_solver.cpp:105] Iteration 5064, lr = 0.00504412
I0407 23:07:15.509038 32718 solver.cpp:218] Iteration 5076 (2.41808 iter/s, 4.96262s/12 iters), loss = 0.328596
I0407 23:07:15.509148 32718 solver.cpp:237]     Train net output #0: loss = 0.328596 (* 1 = 0.328596 loss)
I0407 23:07:15.509157 32718 sgd_solver.cpp:105] Iteration 5076, lr = 0.00502941
I0407 23:07:20.398095 32718 solver.cpp:218] Iteration 5088 (2.45453 iter/s, 4.88892s/12 iters), loss = 0.409389
I0407 23:07:20.398137 32718 solver.cpp:237]     Train net output #0: loss = 0.409389 (* 1 = 0.409389 loss)
I0407 23:07:20.398145 32718 sgd_solver.cpp:105] Iteration 5088, lr = 0.00501471
I0407 23:07:24.890956 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_5100.caffemodel
I0407 23:07:28.048857 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_5100.solverstate
I0407 23:07:30.412568 32718 solver.cpp:330] Iteration 5100, Testing net (#0)
I0407 23:07:30.412586 32718 net.cpp:676] Ignoring source layer train-data
I0407 23:07:32.849118 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:07:34.928006 32718 solver.cpp:397]     Test net output #0: accuracy = 0.43076
I0407 23:07:34.928050 32718 solver.cpp:397]     Test net output #1: loss = 2.93126 (* 1 = 2.93126 loss)
I0407 23:07:35.024629 32718 solver.cpp:218] Iteration 5100 (0.820433 iter/s, 14.6264s/12 iters), loss = 0.425887
I0407 23:07:35.024677 32718 solver.cpp:237]     Train net output #0: loss = 0.425887 (* 1 = 0.425887 loss)
I0407 23:07:35.024685 32718 sgd_solver.cpp:105] Iteration 5100, lr = 0.005
I0407 23:07:39.146726 32718 solver.cpp:218] Iteration 5112 (2.91119 iter/s, 4.12203s/12 iters), loss = 0.254823
I0407 23:07:39.146762 32718 solver.cpp:237]     Train net output #0: loss = 0.254823 (* 1 = 0.254823 loss)
I0407 23:07:39.146770 32718 sgd_solver.cpp:105] Iteration 5112, lr = 0.00498529
I0407 23:07:44.164055 32718 solver.cpp:218] Iteration 5124 (2.39174 iter/s, 5.01727s/12 iters), loss = 0.297174
I0407 23:07:44.164091 32718 solver.cpp:237]     Train net output #0: loss = 0.297174 (* 1 = 0.297174 loss)
I0407 23:07:44.164098 32718 sgd_solver.cpp:105] Iteration 5124, lr = 0.00497059
I0407 23:07:49.090473 32718 solver.cpp:218] Iteration 5136 (2.43588 iter/s, 4.92636s/12 iters), loss = 0.373092
I0407 23:07:49.090610 32718 solver.cpp:237]     Train net output #0: loss = 0.373092 (* 1 = 0.373092 loss)
I0407 23:07:49.090620 32718 sgd_solver.cpp:105] Iteration 5136, lr = 0.00495588
I0407 23:07:54.058779 32718 solver.cpp:218] Iteration 5148 (2.41539 iter/s, 4.96814s/12 iters), loss = 0.292399
I0407 23:07:54.058818 32718 solver.cpp:237]     Train net output #0: loss = 0.292399 (* 1 = 0.292399 loss)
I0407 23:07:54.058825 32718 sgd_solver.cpp:105] Iteration 5148, lr = 0.00494118
I0407 23:07:58.043524 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:07:58.992844 32718 solver.cpp:218] Iteration 5160 (2.4321 iter/s, 4.934s/12 iters), loss = 0.405414
I0407 23:07:58.992879 32718 solver.cpp:237]     Train net output #0: loss = 0.405414 (* 1 = 0.405414 loss)
I0407 23:07:58.992887 32718 sgd_solver.cpp:105] Iteration 5160, lr = 0.00492648
I0407 23:08:03.951501 32718 solver.cpp:218] Iteration 5172 (2.42004 iter/s, 4.9586s/12 iters), loss = 0.363661
I0407 23:08:03.951536 32718 solver.cpp:237]     Train net output #0: loss = 0.363661 (* 1 = 0.363661 loss)
I0407 23:08:03.951543 32718 sgd_solver.cpp:105] Iteration 5172, lr = 0.00491177
I0407 23:08:08.912097 32718 solver.cpp:218] Iteration 5184 (2.41909 iter/s, 4.96054s/12 iters), loss = 0.29121
I0407 23:08:08.912132 32718 solver.cpp:237]     Train net output #0: loss = 0.29121 (* 1 = 0.29121 loss)
I0407 23:08:08.912138 32718 sgd_solver.cpp:105] Iteration 5184, lr = 0.00489707
I0407 23:08:13.829212 32718 solver.cpp:218] Iteration 5196 (2.44049 iter/s, 4.91705s/12 iters), loss = 0.463122
I0407 23:08:13.829252 32718 solver.cpp:237]     Train net output #0: loss = 0.463122 (* 1 = 0.463122 loss)
I0407 23:08:13.829259 32718 sgd_solver.cpp:105] Iteration 5196, lr = 0.00488237
I0407 23:08:15.816865 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_5202.caffemodel
I0407 23:08:22.020715 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_5202.solverstate
I0407 23:08:25.170593 32718 solver.cpp:330] Iteration 5202, Testing net (#0)
I0407 23:08:25.170612 32718 net.cpp:676] Ignoring source layer train-data
I0407 23:08:27.693085 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:08:29.939458 32718 solver.cpp:397]     Test net output #0: accuracy = 0.436887
I0407 23:08:29.939491 32718 solver.cpp:397]     Test net output #1: loss = 2.78722 (* 1 = 2.78722 loss)
I0407 23:08:31.737849 32718 solver.cpp:218] Iteration 5208 (0.670072 iter/s, 17.9085s/12 iters), loss = 0.345822
I0407 23:08:31.737888 32718 solver.cpp:237]     Train net output #0: loss = 0.345822 (* 1 = 0.345822 loss)
I0407 23:08:31.737896 32718 sgd_solver.cpp:105] Iteration 5208, lr = 0.00486768
I0407 23:08:36.681128 32718 solver.cpp:218] Iteration 5220 (2.42757 iter/s, 4.94321s/12 iters), loss = 0.286507
I0407 23:08:36.681169 32718 solver.cpp:237]     Train net output #0: loss = 0.286507 (* 1 = 0.286507 loss)
I0407 23:08:36.681176 32718 sgd_solver.cpp:105] Iteration 5220, lr = 0.00485298
I0407 23:08:41.628785 32718 solver.cpp:218] Iteration 5232 (2.42543 iter/s, 4.94759s/12 iters), loss = 0.292695
I0407 23:08:41.628823 32718 solver.cpp:237]     Train net output #0: loss = 0.292695 (* 1 = 0.292695 loss)
I0407 23:08:41.628829 32718 sgd_solver.cpp:105] Iteration 5232, lr = 0.00483829
I0407 23:08:46.558115 32718 solver.cpp:218] Iteration 5244 (2.43444 iter/s, 4.92926s/12 iters), loss = 0.444686
I0407 23:08:46.558156 32718 solver.cpp:237]     Train net output #0: loss = 0.444686 (* 1 = 0.444686 loss)
I0407 23:08:46.558164 32718 sgd_solver.cpp:105] Iteration 5244, lr = 0.0048236
I0407 23:08:51.514271 32718 solver.cpp:218] Iteration 5256 (2.42127 iter/s, 4.95609s/12 iters), loss = 0.32136
I0407 23:08:51.514312 32718 solver.cpp:237]     Train net output #0: loss = 0.32136 (* 1 = 0.32136 loss)
I0407 23:08:51.514320 32718 sgd_solver.cpp:105] Iteration 5256, lr = 0.00480892
I0407 23:08:52.770236 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:08:56.407001 32718 solver.cpp:218] Iteration 5268 (2.45265 iter/s, 4.89266s/12 iters), loss = 0.241507
I0407 23:08:56.407038 32718 solver.cpp:237]     Train net output #0: loss = 0.241507 (* 1 = 0.241507 loss)
I0407 23:08:56.407045 32718 sgd_solver.cpp:105] Iteration 5268, lr = 0.00479423
I0407 23:09:01.381428 32718 solver.cpp:218] Iteration 5280 (2.41237 iter/s, 4.97436s/12 iters), loss = 0.373735
I0407 23:09:01.381464 32718 solver.cpp:237]     Train net output #0: loss = 0.373735 (* 1 = 0.373735 loss)
I0407 23:09:01.381471 32718 sgd_solver.cpp:105] Iteration 5280, lr = 0.00477955
I0407 23:09:06.282075 32718 solver.cpp:218] Iteration 5292 (2.44869 iter/s, 4.90058s/12 iters), loss = 0.403954
I0407 23:09:06.282121 32718 solver.cpp:237]     Train net output #0: loss = 0.403954 (* 1 = 0.403954 loss)
I0407 23:09:06.282128 32718 sgd_solver.cpp:105] Iteration 5292, lr = 0.00476488
I0407 23:09:10.746160 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_5304.caffemodel
I0407 23:09:14.793373 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_5304.solverstate
I0407 23:09:17.443181 32718 solver.cpp:330] Iteration 5304, Testing net (#0)
I0407 23:09:17.443207 32718 net.cpp:676] Ignoring source layer train-data
I0407 23:09:19.768565 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:09:21.897176 32718 solver.cpp:397]     Test net output #0: accuracy = 0.439951
I0407 23:09:21.897223 32718 solver.cpp:397]     Test net output #1: loss = 2.78545 (* 1 = 2.78545 loss)
I0407 23:09:21.992007 32718 solver.cpp:218] Iteration 5304 (0.763853 iter/s, 15.7098s/12 iters), loss = 0.304116
I0407 23:09:21.992050 32718 solver.cpp:237]     Train net output #0: loss = 0.304116 (* 1 = 0.304116 loss)
I0407 23:09:21.992059 32718 sgd_solver.cpp:105] Iteration 5304, lr = 0.00475021
I0407 23:09:26.118074 32718 solver.cpp:218] Iteration 5316 (2.90839 iter/s, 4.12599s/12 iters), loss = 0.434034
I0407 23:09:26.118202 32718 solver.cpp:237]     Train net output #0: loss = 0.434034 (* 1 = 0.434034 loss)
I0407 23:09:26.118211 32718 sgd_solver.cpp:105] Iteration 5316, lr = 0.00473554
I0407 23:09:31.048571 32718 solver.cpp:218] Iteration 5328 (2.43391 iter/s, 4.93034s/12 iters), loss = 0.568243
I0407 23:09:31.048616 32718 solver.cpp:237]     Train net output #0: loss = 0.568243 (* 1 = 0.568243 loss)
I0407 23:09:31.048624 32718 sgd_solver.cpp:105] Iteration 5328, lr = 0.00472088
I0407 23:09:36.006762 32718 solver.cpp:218] Iteration 5340 (2.42027 iter/s, 4.95811s/12 iters), loss = 0.420622
I0407 23:09:36.006806 32718 solver.cpp:237]     Train net output #0: loss = 0.420622 (* 1 = 0.420622 loss)
I0407 23:09:36.006814 32718 sgd_solver.cpp:105] Iteration 5340, lr = 0.00470622
I0407 23:09:40.921124 32718 solver.cpp:218] Iteration 5352 (2.44186 iter/s, 4.91429s/12 iters), loss = 0.349743
I0407 23:09:40.921169 32718 solver.cpp:237]     Train net output #0: loss = 0.349743 (* 1 = 0.349743 loss)
I0407 23:09:40.921176 32718 sgd_solver.cpp:105] Iteration 5352, lr = 0.00469157
I0407 23:09:44.324905 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:09:45.861876 32718 solver.cpp:218] Iteration 5364 (2.42882 iter/s, 4.94068s/12 iters), loss = 0.222427
I0407 23:09:45.861918 32718 solver.cpp:237]     Train net output #0: loss = 0.222427 (* 1 = 0.222427 loss)
I0407 23:09:45.861927 32718 sgd_solver.cpp:105] Iteration 5364, lr = 0.00467692
I0407 23:09:50.819523 32718 solver.cpp:218] Iteration 5376 (2.42054 iter/s, 4.95758s/12 iters), loss = 0.326973
I0407 23:09:50.819564 32718 solver.cpp:237]     Train net output #0: loss = 0.326973 (* 1 = 0.326973 loss)
I0407 23:09:50.819572 32718 sgd_solver.cpp:105] Iteration 5376, lr = 0.00466228
I0407 23:09:55.762467 32718 solver.cpp:218] Iteration 5388 (2.42774 iter/s, 4.94287s/12 iters), loss = 0.44894
I0407 23:09:55.762509 32718 solver.cpp:237]     Train net output #0: loss = 0.44894 (* 1 = 0.44894 loss)
I0407 23:09:55.762517 32718 sgd_solver.cpp:105] Iteration 5388, lr = 0.00464764
I0407 23:10:00.679491 32718 solver.cpp:218] Iteration 5400 (2.44053 iter/s, 4.91696s/12 iters), loss = 0.238558
I0407 23:10:00.679677 32718 solver.cpp:237]     Train net output #0: loss = 0.238558 (* 1 = 0.238558 loss)
I0407 23:10:00.679687 32718 sgd_solver.cpp:105] Iteration 5400, lr = 0.00463301
I0407 23:10:02.707530 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_5406.caffemodel
I0407 23:10:06.600695 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_5406.solverstate
I0407 23:10:09.414106 32718 solver.cpp:330] Iteration 5406, Testing net (#0)
I0407 23:10:09.414129 32718 net.cpp:676] Ignoring source layer train-data
I0407 23:10:11.673431 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:10:13.838254 32718 solver.cpp:397]     Test net output #0: accuracy = 0.443627
I0407 23:10:13.838300 32718 solver.cpp:397]     Test net output #1: loss = 2.79847 (* 1 = 2.79847 loss)
I0407 23:10:15.643445 32718 solver.cpp:218] Iteration 5412 (0.80194 iter/s, 14.9637s/12 iters), loss = 0.241674
I0407 23:10:15.643491 32718 solver.cpp:237]     Train net output #0: loss = 0.241674 (* 1 = 0.241674 loss)
I0407 23:10:15.643497 32718 sgd_solver.cpp:105] Iteration 5412, lr = 0.00461839
I0407 23:10:20.568766 32718 solver.cpp:218] Iteration 5424 (2.43643 iter/s, 4.92524s/12 iters), loss = 0.262982
I0407 23:10:20.568801 32718 solver.cpp:237]     Train net output #0: loss = 0.262982 (* 1 = 0.262982 loss)
I0407 23:10:20.568809 32718 sgd_solver.cpp:105] Iteration 5424, lr = 0.00460377
I0407 23:10:25.457125 32718 solver.cpp:218] Iteration 5436 (2.45485 iter/s, 4.88829s/12 iters), loss = 0.221008
I0407 23:10:25.457183 32718 solver.cpp:237]     Train net output #0: loss = 0.221008 (* 1 = 0.221008 loss)
I0407 23:10:25.457199 32718 sgd_solver.cpp:105] Iteration 5436, lr = 0.00458916
I0407 23:10:30.413357 32718 solver.cpp:218] Iteration 5448 (2.42123 iter/s, 4.95615s/12 iters), loss = 0.298621
I0407 23:10:30.413386 32718 solver.cpp:237]     Train net output #0: loss = 0.298621 (* 1 = 0.298621 loss)
I0407 23:10:30.413393 32718 sgd_solver.cpp:105] Iteration 5448, lr = 0.00457456
I0407 23:10:35.387322 32718 solver.cpp:218] Iteration 5460 (2.41259 iter/s, 4.9739s/12 iters), loss = 0.364903
I0407 23:10:35.387426 32718 solver.cpp:237]     Train net output #0: loss = 0.364903 (* 1 = 0.364903 loss)
I0407 23:10:35.387435 32718 sgd_solver.cpp:105] Iteration 5460, lr = 0.00455996
I0407 23:10:35.923907 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:10:40.241354 32718 solver.cpp:218] Iteration 5472 (2.47224 iter/s, 4.85391s/12 iters), loss = 0.387594
I0407 23:10:40.241389 32718 solver.cpp:237]     Train net output #0: loss = 0.387594 (* 1 = 0.387594 loss)
I0407 23:10:40.241397 32718 sgd_solver.cpp:105] Iteration 5472, lr = 0.00454538
I0407 23:10:45.204176 32718 solver.cpp:218] Iteration 5484 (2.41801 iter/s, 4.96276s/12 iters), loss = 0.525348
I0407 23:10:45.204213 32718 solver.cpp:237]     Train net output #0: loss = 0.525348 (* 1 = 0.525348 loss)
I0407 23:10:45.204221 32718 sgd_solver.cpp:105] Iteration 5484, lr = 0.0045308
I0407 23:10:50.121773 32718 solver.cpp:218] Iteration 5496 (2.44025 iter/s, 4.91753s/12 iters), loss = 0.260814
I0407 23:10:50.121814 32718 solver.cpp:237]     Train net output #0: loss = 0.260814 (* 1 = 0.260814 loss)
I0407 23:10:50.121822 32718 sgd_solver.cpp:105] Iteration 5496, lr = 0.00451622
I0407 23:10:54.631959 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_5508.caffemodel
I0407 23:10:57.711582 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_5508.solverstate
I0407 23:11:00.134660 32718 solver.cpp:330] Iteration 5508, Testing net (#0)
I0407 23:11:00.134678 32718 net.cpp:676] Ignoring source layer train-data
I0407 23:11:02.532910 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:11:04.780833 32718 solver.cpp:397]     Test net output #0: accuracy = 0.427696
I0407 23:11:04.780879 32718 solver.cpp:397]     Test net output #1: loss = 2.8752 (* 1 = 2.8752 loss)
I0407 23:11:04.877516 32718 solver.cpp:218] Iteration 5508 (0.813248 iter/s, 14.7556s/12 iters), loss = 0.332867
I0407 23:11:04.877558 32718 solver.cpp:237]     Train net output #0: loss = 0.332867 (* 1 = 0.332867 loss)
I0407 23:11:04.877568 32718 sgd_solver.cpp:105] Iteration 5508, lr = 0.00450166
I0407 23:11:09.000705 32718 solver.cpp:218] Iteration 5520 (2.91041 iter/s, 4.12312s/12 iters), loss = 0.170182
I0407 23:11:09.001951 32718 solver.cpp:237]     Train net output #0: loss = 0.170182 (* 1 = 0.170182 loss)
I0407 23:11:09.001961 32718 sgd_solver.cpp:105] Iteration 5520, lr = 0.0044871
I0407 23:11:11.418459 32718 blocking_queue.cpp:49] Waiting for data
I0407 23:11:13.965103 32718 solver.cpp:218] Iteration 5532 (2.41783 iter/s, 4.96312s/12 iters), loss = 0.446527
I0407 23:11:13.965148 32718 solver.cpp:237]     Train net output #0: loss = 0.446527 (* 1 = 0.446527 loss)
I0407 23:11:13.965157 32718 sgd_solver.cpp:105] Iteration 5532, lr = 0.00447256
I0407 23:11:18.882711 32718 solver.cpp:218] Iteration 5544 (2.44025 iter/s, 4.91753s/12 iters), loss = 0.234371
I0407 23:11:18.882756 32718 solver.cpp:237]     Train net output #0: loss = 0.234371 (* 1 = 0.234371 loss)
I0407 23:11:18.882764 32718 sgd_solver.cpp:105] Iteration 5544, lr = 0.00445802
I0407 23:11:23.840610 32718 solver.cpp:218] Iteration 5556 (2.42042 iter/s, 4.95782s/12 iters), loss = 0.343967
I0407 23:11:23.840653 32718 solver.cpp:237]     Train net output #0: loss = 0.343967 (* 1 = 0.343967 loss)
I0407 23:11:23.840662 32718 sgd_solver.cpp:105] Iteration 5556, lr = 0.00444349
I0407 23:11:26.482266 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:11:28.744062 32718 solver.cpp:218] Iteration 5568 (2.44729 iter/s, 4.90338s/12 iters), loss = 0.263888
I0407 23:11:28.744104 32718 solver.cpp:237]     Train net output #0: loss = 0.263888 (* 1 = 0.263888 loss)
I0407 23:11:28.744112 32718 sgd_solver.cpp:105] Iteration 5568, lr = 0.00442897
I0407 23:11:33.660988 32718 solver.cpp:218] Iteration 5580 (2.44059 iter/s, 4.91685s/12 iters), loss = 0.448642
I0407 23:11:33.661033 32718 solver.cpp:237]     Train net output #0: loss = 0.448642 (* 1 = 0.448642 loss)
I0407 23:11:33.661042 32718 sgd_solver.cpp:105] Iteration 5580, lr = 0.00441446
I0407 23:11:38.521337 32718 solver.cpp:218] Iteration 5592 (2.469 iter/s, 4.86027s/12 iters), loss = 0.379519
I0407 23:11:38.521378 32718 solver.cpp:237]     Train net output #0: loss = 0.379518 (* 1 = 0.379518 loss)
I0407 23:11:38.521386 32718 sgd_solver.cpp:105] Iteration 5592, lr = 0.00439996
I0407 23:11:43.376650 32718 solver.cpp:218] Iteration 5604 (2.47155 iter/s, 4.85525s/12 iters), loss = 0.271301
I0407 23:11:43.376801 32718 solver.cpp:237]     Train net output #0: loss = 0.271301 (* 1 = 0.271301 loss)
I0407 23:11:43.376811 32718 sgd_solver.cpp:105] Iteration 5604, lr = 0.00438548
I0407 23:11:45.383113 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_5610.caffemodel
I0407 23:11:48.457963 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_5610.solverstate
I0407 23:11:51.990648 32718 solver.cpp:330] Iteration 5610, Testing net (#0)
I0407 23:11:51.990675 32718 net.cpp:676] Ignoring source layer train-data
I0407 23:11:54.304101 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:11:56.553200 32718 solver.cpp:397]     Test net output #0: accuracy = 0.443015
I0407 23:11:56.553246 32718 solver.cpp:397]     Test net output #1: loss = 2.88567 (* 1 = 2.88567 loss)
I0407 23:11:58.358603 32718 solver.cpp:218] Iteration 5616 (0.800975 iter/s, 14.9817s/12 iters), loss = 0.173353
I0407 23:11:58.358650 32718 solver.cpp:237]     Train net output #0: loss = 0.173353 (* 1 = 0.173353 loss)
I0407 23:11:58.358659 32718 sgd_solver.cpp:105] Iteration 5616, lr = 0.004371
I0407 23:12:03.290107 32718 solver.cpp:218] Iteration 5628 (2.43337 iter/s, 4.93143s/12 iters), loss = 0.285058
I0407 23:12:03.290140 32718 solver.cpp:237]     Train net output #0: loss = 0.285058 (* 1 = 0.285058 loss)
I0407 23:12:03.290148 32718 sgd_solver.cpp:105] Iteration 5628, lr = 0.00435653
I0407 23:12:08.233584 32718 solver.cpp:218] Iteration 5640 (2.42747 iter/s, 4.94341s/12 iters), loss = 0.464369
I0407 23:12:08.233621 32718 solver.cpp:237]     Train net output #0: loss = 0.464369 (* 1 = 0.464369 loss)
I0407 23:12:08.233629 32718 sgd_solver.cpp:105] Iteration 5640, lr = 0.00434207
I0407 23:12:13.186333 32718 solver.cpp:218] Iteration 5652 (2.42293 iter/s, 4.95268s/12 iters), loss = 0.389239
I0407 23:12:13.186374 32718 solver.cpp:237]     Train net output #0: loss = 0.389239 (* 1 = 0.389239 loss)
I0407 23:12:13.186383 32718 sgd_solver.cpp:105] Iteration 5652, lr = 0.00432763
I0407 23:12:17.964275 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:12:18.132974 32718 solver.cpp:218] Iteration 5664 (2.42592 iter/s, 4.94657s/12 iters), loss = 0.162345
I0407 23:12:18.133016 32718 solver.cpp:237]     Train net output #0: loss = 0.162345 (* 1 = 0.162345 loss)
I0407 23:12:18.133025 32718 sgd_solver.cpp:105] Iteration 5664, lr = 0.00431319
I0407 23:12:23.066098 32718 solver.cpp:218] Iteration 5676 (2.43257 iter/s, 4.93305s/12 iters), loss = 0.335337
I0407 23:12:23.066143 32718 solver.cpp:237]     Train net output #0: loss = 0.335337 (* 1 = 0.335337 loss)
I0407 23:12:23.066151 32718 sgd_solver.cpp:105] Iteration 5676, lr = 0.00429877
I0407 23:12:28.067782 32718 solver.cpp:218] Iteration 5688 (2.39923 iter/s, 5.00161s/12 iters), loss = 0.333068
I0407 23:12:28.067817 32718 solver.cpp:237]     Train net output #0: loss = 0.333068 (* 1 = 0.333068 loss)
I0407 23:12:28.067824 32718 sgd_solver.cpp:105] Iteration 5688, lr = 0.00428436
I0407 23:12:33.112319 32718 solver.cpp:218] Iteration 5700 (2.37884 iter/s, 5.04447s/12 iters), loss = 0.260753
I0407 23:12:33.112360 32718 solver.cpp:237]     Train net output #0: loss = 0.260753 (* 1 = 0.260753 loss)
I0407 23:12:33.112367 32718 sgd_solver.cpp:105] Iteration 5700, lr = 0.00426996
I0407 23:12:37.631028 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_5712.caffemodel
I0407 23:12:40.682340 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_5712.solverstate
I0407 23:12:43.045485 32718 solver.cpp:330] Iteration 5712, Testing net (#0)
I0407 23:12:43.045502 32718 net.cpp:676] Ignoring source layer train-data
I0407 23:12:45.240324 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:12:47.510076 32718 solver.cpp:397]     Test net output #0: accuracy = 0.45098
I0407 23:12:47.510109 32718 solver.cpp:397]     Test net output #1: loss = 2.82659 (* 1 = 2.82659 loss)
I0407 23:12:47.604308 32718 solver.cpp:218] Iteration 5712 (0.82805 iter/s, 14.4919s/12 iters), loss = 0.210606
I0407 23:12:47.604383 32718 solver.cpp:237]     Train net output #0: loss = 0.210606 (* 1 = 0.210606 loss)
I0407 23:12:47.604399 32718 sgd_solver.cpp:105] Iteration 5712, lr = 0.00425557
I0407 23:12:51.745806 32718 solver.cpp:218] Iteration 5724 (2.89756 iter/s, 4.14141s/12 iters), loss = 0.29118
I0407 23:12:51.745931 32718 solver.cpp:237]     Train net output #0: loss = 0.29118 (* 1 = 0.29118 loss)
I0407 23:12:51.745940 32718 sgd_solver.cpp:105] Iteration 5724, lr = 0.0042412
I0407 23:12:56.670980 32718 solver.cpp:218] Iteration 5736 (2.43653 iter/s, 4.92503s/12 iters), loss = 0.279082
I0407 23:12:56.671016 32718 solver.cpp:237]     Train net output #0: loss = 0.279082 (* 1 = 0.279082 loss)
I0407 23:12:56.671025 32718 sgd_solver.cpp:105] Iteration 5736, lr = 0.00422684
I0407 23:13:01.636137 32718 solver.cpp:218] Iteration 5748 (2.41687 iter/s, 4.96509s/12 iters), loss = 0.266268
I0407 23:13:01.636173 32718 solver.cpp:237]     Train net output #0: loss = 0.266268 (* 1 = 0.266268 loss)
I0407 23:13:01.636180 32718 sgd_solver.cpp:105] Iteration 5748, lr = 0.00421249
I0407 23:13:06.575747 32718 solver.cpp:218] Iteration 5760 (2.42937 iter/s, 4.93955s/12 iters), loss = 0.274327
I0407 23:13:06.575783 32718 solver.cpp:237]     Train net output #0: loss = 0.274327 (* 1 = 0.274327 loss)
I0407 23:13:06.575790 32718 sgd_solver.cpp:105] Iteration 5760, lr = 0.00419816
I0407 23:13:08.540534 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:13:11.538125 32718 solver.cpp:218] Iteration 5772 (2.41823 iter/s, 4.96231s/12 iters), loss = 0.243624
I0407 23:13:11.538173 32718 solver.cpp:237]     Train net output #0: loss = 0.243624 (* 1 = 0.243624 loss)
I0407 23:13:11.538182 32718 sgd_solver.cpp:105] Iteration 5772, lr = 0.00418384
I0407 23:13:16.494931 32718 solver.cpp:218] Iteration 5784 (2.42095 iter/s, 4.95673s/12 iters), loss = 0.254016
I0407 23:13:16.494976 32718 solver.cpp:237]     Train net output #0: loss = 0.254016 (* 1 = 0.254016 loss)
I0407 23:13:16.494984 32718 sgd_solver.cpp:105] Iteration 5784, lr = 0.00416953
I0407 23:13:21.415496 32718 solver.cpp:218] Iteration 5796 (2.43878 iter/s, 4.92049s/12 iters), loss = 0.18623
I0407 23:13:21.415534 32718 solver.cpp:237]     Train net output #0: loss = 0.18623 (* 1 = 0.18623 loss)
I0407 23:13:21.415541 32718 sgd_solver.cpp:105] Iteration 5796, lr = 0.00415524
I0407 23:13:26.363279 32718 solver.cpp:218] Iteration 5808 (2.42536 iter/s, 4.94771s/12 iters), loss = 0.238115
I0407 23:13:26.363384 32718 solver.cpp:237]     Train net output #0: loss = 0.238114 (* 1 = 0.238114 loss)
I0407 23:13:26.363394 32718 sgd_solver.cpp:105] Iteration 5808, lr = 0.00414096
I0407 23:13:28.375028 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_5814.caffemodel
I0407 23:13:31.536496 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_5814.solverstate
I0407 23:13:34.931814 32718 solver.cpp:330] Iteration 5814, Testing net (#0)
I0407 23:13:34.931833 32718 net.cpp:676] Ignoring source layer train-data
I0407 23:13:37.187229 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:13:39.699745 32718 solver.cpp:397]     Test net output #0: accuracy = 0.456495
I0407 23:13:39.699771 32718 solver.cpp:397]     Test net output #1: loss = 2.89816 (* 1 = 2.89816 loss)
I0407 23:13:41.424921 32718 solver.cpp:218] Iteration 5820 (0.796734 iter/s, 15.0615s/12 iters), loss = 0.327538
I0407 23:13:41.424962 32718 solver.cpp:237]     Train net output #0: loss = 0.327538 (* 1 = 0.327538 loss)
I0407 23:13:41.424970 32718 sgd_solver.cpp:105] Iteration 5820, lr = 0.00412669
I0407 23:13:46.373013 32718 solver.cpp:218] Iteration 5832 (2.42521 iter/s, 4.94802s/12 iters), loss = 0.220175
I0407 23:13:46.373052 32718 solver.cpp:237]     Train net output #0: loss = 0.220175 (* 1 = 0.220175 loss)
I0407 23:13:46.373060 32718 sgd_solver.cpp:105] Iteration 5832, lr = 0.00411244
I0407 23:13:51.339320 32718 solver.cpp:218] Iteration 5844 (2.41631 iter/s, 4.96624s/12 iters), loss = 0.316349
I0407 23:13:51.339356 32718 solver.cpp:237]     Train net output #0: loss = 0.316349 (* 1 = 0.316349 loss)
I0407 23:13:51.339363 32718 sgd_solver.cpp:105] Iteration 5844, lr = 0.00409821
I0407 23:13:56.343976 32718 solver.cpp:218] Iteration 5856 (2.3978 iter/s, 5.00459s/12 iters), loss = 0.412847
I0407 23:13:56.344022 32718 solver.cpp:237]     Train net output #0: loss = 0.412847 (* 1 = 0.412847 loss)
I0407 23:13:56.344029 32718 sgd_solver.cpp:105] Iteration 5856, lr = 0.00408399
I0407 23:14:00.474251 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:14:01.259037 32718 solver.cpp:218] Iteration 5868 (2.44152 iter/s, 4.91498s/12 iters), loss = 0.118479
I0407 23:14:01.259092 32718 solver.cpp:237]     Train net output #0: loss = 0.118479 (* 1 = 0.118479 loss)
I0407 23:14:01.259102 32718 sgd_solver.cpp:105] Iteration 5868, lr = 0.00406978
I0407 23:14:06.184521 32718 solver.cpp:218] Iteration 5880 (2.43635 iter/s, 4.9254s/12 iters), loss = 0.168854
I0407 23:14:06.184561 32718 solver.cpp:237]     Train net output #0: loss = 0.168854 (* 1 = 0.168854 loss)
I0407 23:14:06.184571 32718 sgd_solver.cpp:105] Iteration 5880, lr = 0.0040556
I0407 23:14:11.070025 32718 solver.cpp:218] Iteration 5892 (2.45628 iter/s, 4.88544s/12 iters), loss = 0.309822
I0407 23:14:11.070062 32718 solver.cpp:237]     Train net output #0: loss = 0.309822 (* 1 = 0.309822 loss)
I0407 23:14:11.070071 32718 sgd_solver.cpp:105] Iteration 5892, lr = 0.00404142
I0407 23:14:16.049135 32718 solver.cpp:218] Iteration 5904 (2.4101 iter/s, 4.97904s/12 iters), loss = 0.271018
I0407 23:14:16.049185 32718 solver.cpp:237]     Train net output #0: loss = 0.271018 (* 1 = 0.271018 loss)
I0407 23:14:16.049196 32718 sgd_solver.cpp:105] Iteration 5904, lr = 0.00402726
I0407 23:14:20.491441 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_5916.caffemodel
I0407 23:14:23.560115 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_5916.solverstate
I0407 23:14:25.929137 32718 solver.cpp:330] Iteration 5916, Testing net (#0)
I0407 23:14:25.929157 32718 net.cpp:676] Ignoring source layer train-data
I0407 23:14:28.143687 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:14:30.692936 32718 solver.cpp:397]     Test net output #0: accuracy = 0.441176
I0407 23:14:30.693089 32718 solver.cpp:397]     Test net output #1: loss = 2.88824 (* 1 = 2.88824 loss)
I0407 23:14:30.790127 32718 solver.cpp:218] Iteration 5916 (0.814062 iter/s, 14.7409s/12 iters), loss = 0.274278
I0407 23:14:30.790171 32718 solver.cpp:237]     Train net output #0: loss = 0.274278 (* 1 = 0.274278 loss)
I0407 23:14:30.790179 32718 sgd_solver.cpp:105] Iteration 5916, lr = 0.00401312
I0407 23:14:34.933090 32718 solver.cpp:218] Iteration 5928 (2.89653 iter/s, 4.14289s/12 iters), loss = 0.301839
I0407 23:14:34.933135 32718 solver.cpp:237]     Train net output #0: loss = 0.301839 (* 1 = 0.301839 loss)
I0407 23:14:34.933143 32718 sgd_solver.cpp:105] Iteration 5928, lr = 0.003999
I0407 23:14:39.845438 32718 solver.cpp:218] Iteration 5940 (2.44286 iter/s, 4.91227s/12 iters), loss = 0.313832
I0407 23:14:39.845481 32718 solver.cpp:237]     Train net output #0: loss = 0.313832 (* 1 = 0.313832 loss)
I0407 23:14:39.845490 32718 sgd_solver.cpp:105] Iteration 5940, lr = 0.00398489
I0407 23:14:44.808274 32718 solver.cpp:218] Iteration 5952 (2.41801 iter/s, 4.96276s/12 iters), loss = 0.208369
I0407 23:14:44.808322 32718 solver.cpp:237]     Train net output #0: loss = 0.208369 (* 1 = 0.208369 loss)
I0407 23:14:44.808331 32718 sgd_solver.cpp:105] Iteration 5952, lr = 0.0039708
I0407 23:14:49.728799 32718 solver.cpp:218] Iteration 5964 (2.4388 iter/s, 4.92045s/12 iters), loss = 0.103257
I0407 23:14:49.728842 32718 solver.cpp:237]     Train net output #0: loss = 0.103257 (* 1 = 0.103257 loss)
I0407 23:14:49.728852 32718 sgd_solver.cpp:105] Iteration 5964, lr = 0.00395672
I0407 23:14:51.017771 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:14:54.673429 32718 solver.cpp:218] Iteration 5976 (2.42691 iter/s, 4.94456s/12 iters), loss = 0.182879
I0407 23:14:54.673472 32718 solver.cpp:237]     Train net output #0: loss = 0.182878 (* 1 = 0.182878 loss)
I0407 23:14:54.673481 32718 sgd_solver.cpp:105] Iteration 5976, lr = 0.00394267
I0407 23:14:59.552677 32718 solver.cpp:218] Iteration 5988 (2.45943 iter/s, 4.87917s/12 iters), loss = 0.231374
I0407 23:14:59.552723 32718 solver.cpp:237]     Train net output #0: loss = 0.231374 (* 1 = 0.231374 loss)
I0407 23:14:59.552731 32718 sgd_solver.cpp:105] Iteration 5988, lr = 0.00392863
I0407 23:15:04.466414 32718 solver.cpp:218] Iteration 6000 (2.44217 iter/s, 4.91367s/12 iters), loss = 0.0500223
I0407 23:15:04.466567 32718 solver.cpp:237]     Train net output #0: loss = 0.0500223 (* 1 = 0.0500223 loss)
I0407 23:15:04.466579 32718 sgd_solver.cpp:105] Iteration 6000, lr = 0.00391461
I0407 23:15:09.438127 32718 solver.cpp:218] Iteration 6012 (2.41374 iter/s, 4.97153s/12 iters), loss = 0.406675
I0407 23:15:09.438167 32718 solver.cpp:237]     Train net output #0: loss = 0.406674 (* 1 = 0.406674 loss)
I0407 23:15:09.438176 32718 sgd_solver.cpp:105] Iteration 6012, lr = 0.0039006
I0407 23:15:11.447216 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_6018.caffemodel
I0407 23:15:14.912168 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_6018.solverstate
I0407 23:15:18.773993 32718 solver.cpp:330] Iteration 6018, Testing net (#0)
I0407 23:15:18.774011 32718 net.cpp:676] Ignoring source layer train-data
I0407 23:15:20.843447 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:15:23.254165 32718 solver.cpp:397]     Test net output #0: accuracy = 0.446078
I0407 23:15:23.254209 32718 solver.cpp:397]     Test net output #1: loss = 2.91095 (* 1 = 2.91095 loss)
I0407 23:15:25.055048 32718 solver.cpp:218] Iteration 6024 (0.768402 iter/s, 15.6168s/12 iters), loss = 0.191863
I0407 23:15:25.055092 32718 solver.cpp:237]     Train net output #0: loss = 0.191863 (* 1 = 0.191863 loss)
I0407 23:15:25.055100 32718 sgd_solver.cpp:105] Iteration 6024, lr = 0.00388662
I0407 23:15:29.987507 32718 solver.cpp:218] Iteration 6036 (2.4329 iter/s, 4.93239s/12 iters), loss = 0.159178
I0407 23:15:29.987547 32718 solver.cpp:237]     Train net output #0: loss = 0.159178 (* 1 = 0.159178 loss)
I0407 23:15:29.987556 32718 sgd_solver.cpp:105] Iteration 6036, lr = 0.00387265
I0407 23:15:34.941915 32718 solver.cpp:218] Iteration 6048 (2.42212 iter/s, 4.95433s/12 iters), loss = 0.299724
I0407 23:15:34.942035 32718 solver.cpp:237]     Train net output #0: loss = 0.299724 (* 1 = 0.299724 loss)
I0407 23:15:34.942044 32718 sgd_solver.cpp:105] Iteration 6048, lr = 0.0038587
I0407 23:15:39.889956 32718 solver.cpp:218] Iteration 6060 (2.42527 iter/s, 4.9479s/12 iters), loss = 0.28043
I0407 23:15:39.889992 32718 solver.cpp:237]     Train net output #0: loss = 0.28043 (* 1 = 0.28043 loss)
I0407 23:15:39.889999 32718 sgd_solver.cpp:105] Iteration 6060, lr = 0.00384477
I0407 23:15:43.231225 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:15:44.736285 32718 solver.cpp:218] Iteration 6072 (2.47613 iter/s, 4.84627s/12 iters), loss = 0.196045
I0407 23:15:44.736325 32718 solver.cpp:237]     Train net output #0: loss = 0.196045 (* 1 = 0.196045 loss)
I0407 23:15:44.736332 32718 sgd_solver.cpp:105] Iteration 6072, lr = 0.00383086
I0407 23:15:49.610525 32718 solver.cpp:218] Iteration 6084 (2.46196 iter/s, 4.87417s/12 iters), loss = 0.168103
I0407 23:15:49.610569 32718 solver.cpp:237]     Train net output #0: loss = 0.168103 (* 1 = 0.168103 loss)
I0407 23:15:49.610577 32718 sgd_solver.cpp:105] Iteration 6084, lr = 0.00381697
I0407 23:15:54.529460 32718 solver.cpp:218] Iteration 6096 (2.43959 iter/s, 4.91886s/12 iters), loss = 0.26628
I0407 23:15:54.529497 32718 solver.cpp:237]     Train net output #0: loss = 0.26628 (* 1 = 0.26628 loss)
I0407 23:15:54.529505 32718 sgd_solver.cpp:105] Iteration 6096, lr = 0.00380309
I0407 23:15:59.475039 32718 solver.cpp:218] Iteration 6108 (2.42644 iter/s, 4.94552s/12 iters), loss = 0.143022
I0407 23:15:59.475076 32718 solver.cpp:237]     Train net output #0: loss = 0.143022 (* 1 = 0.143022 loss)
I0407 23:15:59.475085 32718 sgd_solver.cpp:105] Iteration 6108, lr = 0.00378924
I0407 23:16:03.951313 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_6120.caffemodel
I0407 23:16:07.593971 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_6120.solverstate
I0407 23:16:09.990262 32718 solver.cpp:330] Iteration 6120, Testing net (#0)
I0407 23:16:09.990280 32718 net.cpp:676] Ignoring source layer train-data
I0407 23:16:12.122649 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:16:14.769440 32718 solver.cpp:397]     Test net output #0: accuracy = 0.460172
I0407 23:16:14.769476 32718 solver.cpp:397]     Test net output #1: loss = 2.84229 (* 1 = 2.84229 loss)
I0407 23:16:14.866214 32718 solver.cpp:218] Iteration 6120 (0.779673 iter/s, 15.3911s/12 iters), loss = 0.167414
I0407 23:16:14.866262 32718 solver.cpp:237]     Train net output #0: loss = 0.167414 (* 1 = 0.167414 loss)
I0407 23:16:14.866271 32718 sgd_solver.cpp:105] Iteration 6120, lr = 0.00377541
I0407 23:16:18.975874 32718 solver.cpp:218] Iteration 6132 (2.92 iter/s, 4.10959s/12 iters), loss = 0.112383
I0407 23:16:18.975922 32718 solver.cpp:237]     Train net output #0: loss = 0.112383 (* 1 = 0.112383 loss)
I0407 23:16:18.975934 32718 sgd_solver.cpp:105] Iteration 6132, lr = 0.00376159
I0407 23:16:23.887446 32718 solver.cpp:218] Iteration 6144 (2.44325 iter/s, 4.9115s/12 iters), loss = 0.197079
I0407 23:16:23.887488 32718 solver.cpp:237]     Train net output #0: loss = 0.197079 (* 1 = 0.197079 loss)
I0407 23:16:23.887496 32718 sgd_solver.cpp:105] Iteration 6144, lr = 0.0037478
I0407 23:16:28.862318 32718 solver.cpp:218] Iteration 6156 (2.41216 iter/s, 4.97479s/12 iters), loss = 0.143775
I0407 23:16:28.862362 32718 solver.cpp:237]     Train net output #0: loss = 0.143775 (* 1 = 0.143775 loss)
I0407 23:16:28.862371 32718 sgd_solver.cpp:105] Iteration 6156, lr = 0.00373403
I0407 23:16:33.781985 32718 solver.cpp:218] Iteration 6168 (2.43923 iter/s, 4.91959s/12 iters), loss = 0.156323
I0407 23:16:33.782027 32718 solver.cpp:237]     Train net output #0: loss = 0.156323 (* 1 = 0.156323 loss)
I0407 23:16:33.782034 32718 sgd_solver.cpp:105] Iteration 6168, lr = 0.00372027
I0407 23:16:34.359953 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:16:38.734321 32718 solver.cpp:218] Iteration 6180 (2.42314 iter/s, 4.95226s/12 iters), loss = 0.267027
I0407 23:16:38.734441 32718 solver.cpp:237]     Train net output #0: loss = 0.267027 (* 1 = 0.267027 loss)
I0407 23:16:38.734450 32718 sgd_solver.cpp:105] Iteration 6180, lr = 0.00370654
I0407 23:16:43.701599 32718 solver.cpp:218] Iteration 6192 (2.41588 iter/s, 4.96713s/12 iters), loss = 0.26648
I0407 23:16:43.701644 32718 solver.cpp:237]     Train net output #0: loss = 0.26648 (* 1 = 0.26648 loss)
I0407 23:16:43.701653 32718 sgd_solver.cpp:105] Iteration 6192, lr = 0.00369283
I0407 23:16:48.617102 32718 solver.cpp:218] Iteration 6204 (2.4413 iter/s, 4.91542s/12 iters), loss = 0.133198
I0407 23:16:48.617149 32718 solver.cpp:237]     Train net output #0: loss = 0.133198 (* 1 = 0.133198 loss)
I0407 23:16:48.617158 32718 sgd_solver.cpp:105] Iteration 6204, lr = 0.00367914
I0407 23:16:53.587695 32718 solver.cpp:218] Iteration 6216 (2.41424 iter/s, 4.97051s/12 iters), loss = 0.162873
I0407 23:16:53.587738 32718 solver.cpp:237]     Train net output #0: loss = 0.162873 (* 1 = 0.162873 loss)
I0407 23:16:53.587747 32718 sgd_solver.cpp:105] Iteration 6216, lr = 0.00366547
I0407 23:16:55.595649 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_6222.caffemodel
I0407 23:17:01.375563 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_6222.solverstate
I0407 23:17:05.785441 32718 solver.cpp:330] Iteration 6222, Testing net (#0)
I0407 23:17:05.785466 32718 net.cpp:676] Ignoring source layer train-data
I0407 23:17:07.948206 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:17:09.312353 32718 blocking_queue.cpp:49] Waiting for data
I0407 23:17:10.505215 32718 solver.cpp:397]     Test net output #0: accuracy = 0.459559
I0407 23:17:10.505261 32718 solver.cpp:397]     Test net output #1: loss = 2.90079 (* 1 = 2.90079 loss)
I0407 23:17:12.283193 32718 solver.cpp:218] Iteration 6228 (0.641869 iter/s, 18.6954s/12 iters), loss = 0.157468
I0407 23:17:12.283234 32718 solver.cpp:237]     Train net output #0: loss = 0.157468 (* 1 = 0.157468 loss)
I0407 23:17:12.283241 32718 sgd_solver.cpp:105] Iteration 6228, lr = 0.00365182
I0407 23:17:17.210988 32718 solver.cpp:218] Iteration 6240 (2.4352 iter/s, 4.92773s/12 iters), loss = 0.222618
I0407 23:17:17.211028 32718 solver.cpp:237]     Train net output #0: loss = 0.222617 (* 1 = 0.222617 loss)
I0407 23:17:17.211036 32718 sgd_solver.cpp:105] Iteration 6240, lr = 0.0036382
I0407 23:17:22.172529 32718 solver.cpp:218] Iteration 6252 (2.41864 iter/s, 4.96148s/12 iters), loss = 0.220498
I0407 23:17:22.172566 32718 solver.cpp:237]     Train net output #0: loss = 0.220498 (* 1 = 0.220498 loss)
I0407 23:17:22.172574 32718 sgd_solver.cpp:105] Iteration 6252, lr = 0.00362459
I0407 23:17:27.139919 32718 solver.cpp:218] Iteration 6264 (2.41579 iter/s, 4.96732s/12 iters), loss = 0.13747
I0407 23:17:27.139961 32718 solver.cpp:237]     Train net output #0: loss = 0.13747 (* 1 = 0.13747 loss)
I0407 23:17:27.139971 32718 sgd_solver.cpp:105] Iteration 6264, lr = 0.00361101
I0407 23:17:29.811056 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:17:32.041991 32718 solver.cpp:218] Iteration 6276 (2.44798 iter/s, 4.90201s/12 iters), loss = 0.146256
I0407 23:17:32.042028 32718 solver.cpp:237]     Train net output #0: loss = 0.146256 (* 1 = 0.146256 loss)
I0407 23:17:32.042037 32718 sgd_solver.cpp:105] Iteration 6276, lr = 0.00359745
I0407 23:17:37.000116 32718 solver.cpp:218] Iteration 6288 (2.4203 iter/s, 4.95806s/12 iters), loss = 0.266033
I0407 23:17:37.000162 32718 solver.cpp:237]     Train net output #0: loss = 0.266033 (* 1 = 0.266033 loss)
I0407 23:17:37.000171 32718 sgd_solver.cpp:105] Iteration 6288, lr = 0.00358391
I0407 23:17:41.924182 32718 solver.cpp:218] Iteration 6300 (2.43705 iter/s, 4.92399s/12 iters), loss = 0.06875
I0407 23:17:41.924294 32718 solver.cpp:237]     Train net output #0: loss = 0.0687499 (* 1 = 0.0687499 loss)
I0407 23:17:41.924304 32718 sgd_solver.cpp:105] Iteration 6300, lr = 0.0035704
I0407 23:17:46.875741 32718 solver.cpp:218] Iteration 6312 (2.42355 iter/s, 4.95142s/12 iters), loss = 0.260799
I0407 23:17:46.875777 32718 solver.cpp:237]     Train net output #0: loss = 0.260799 (* 1 = 0.260799 loss)
I0407 23:17:46.875785 32718 sgd_solver.cpp:105] Iteration 6312, lr = 0.00355691
I0407 23:17:51.358108 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_6324.caffemodel
I0407 23:17:55.069451 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_6324.solverstate
I0407 23:17:58.096220 32718 solver.cpp:330] Iteration 6324, Testing net (#0)
I0407 23:17:58.096241 32718 net.cpp:676] Ignoring source layer train-data
I0407 23:18:00.133919 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:18:02.864054 32718 solver.cpp:397]     Test net output #0: accuracy = 0.46875
I0407 23:18:02.864099 32718 solver.cpp:397]     Test net output #1: loss = 2.82694 (* 1 = 2.82694 loss)
I0407 23:18:02.960659 32718 solver.cpp:218] Iteration 6324 (0.746045 iter/s, 16.0848s/12 iters), loss = 0.144369
I0407 23:18:02.960705 32718 solver.cpp:237]     Train net output #0: loss = 0.144369 (* 1 = 0.144369 loss)
I0407 23:18:02.960712 32718 sgd_solver.cpp:105] Iteration 6324, lr = 0.00354344
I0407 23:18:07.078264 32718 solver.cpp:218] Iteration 6336 (2.91437 iter/s, 4.11753s/12 iters), loss = 0.295891
I0407 23:18:07.078303 32718 solver.cpp:237]     Train net output #0: loss = 0.295891 (* 1 = 0.295891 loss)
I0407 23:18:07.078312 32718 sgd_solver.cpp:105] Iteration 6336, lr = 0.00352999
I0407 23:18:11.976161 32718 solver.cpp:218] Iteration 6348 (2.45007 iter/s, 4.89782s/12 iters), loss = 0.280723
I0407 23:18:11.976296 32718 solver.cpp:237]     Train net output #0: loss = 0.280723 (* 1 = 0.280723 loss)
I0407 23:18:11.976306 32718 sgd_solver.cpp:105] Iteration 6348, lr = 0.00351657
I0407 23:18:16.919184 32718 solver.cpp:218] Iteration 6360 (2.42775 iter/s, 4.94285s/12 iters), loss = 0.297817
I0407 23:18:16.919248 32718 solver.cpp:237]     Train net output #0: loss = 0.297817 (* 1 = 0.297817 loss)
I0407 23:18:16.919260 32718 sgd_solver.cpp:105] Iteration 6360, lr = 0.00350317
I0407 23:18:21.730629 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:18:21.871112 32718 solver.cpp:218] Iteration 6372 (2.42334 iter/s, 4.95183s/12 iters), loss = 0.289159
I0407 23:18:21.871170 32718 solver.cpp:237]     Train net output #0: loss = 0.289159 (* 1 = 0.289159 loss)
I0407 23:18:21.871181 32718 sgd_solver.cpp:105] Iteration 6372, lr = 0.00348979
I0407 23:18:26.787182 32718 solver.cpp:218] Iteration 6384 (2.44102 iter/s, 4.91598s/12 iters), loss = 0.197895
I0407 23:18:26.787235 32718 solver.cpp:237]     Train net output #0: loss = 0.197895 (* 1 = 0.197895 loss)
I0407 23:18:26.787243 32718 sgd_solver.cpp:105] Iteration 6384, lr = 0.00347644
I0407 23:18:31.755342 32718 solver.cpp:218] Iteration 6396 (2.41542 iter/s, 4.96808s/12 iters), loss = 0.074853
I0407 23:18:31.755390 32718 solver.cpp:237]     Train net output #0: loss = 0.0748529 (* 1 = 0.0748529 loss)
I0407 23:18:31.755400 32718 sgd_solver.cpp:105] Iteration 6396, lr = 0.00346311
I0407 23:18:36.670437 32718 solver.cpp:218] Iteration 6408 (2.4415 iter/s, 4.91501s/12 iters), loss = 0.210529
I0407 23:18:36.670482 32718 solver.cpp:237]     Train net output #0: loss = 0.210529 (* 1 = 0.210529 loss)
I0407 23:18:36.670490 32718 sgd_solver.cpp:105] Iteration 6408, lr = 0.00344981
I0407 23:18:41.641657 32718 solver.cpp:218] Iteration 6420 (2.41393 iter/s, 4.97114s/12 iters), loss = 0.141338
I0407 23:18:41.641702 32718 solver.cpp:237]     Train net output #0: loss = 0.141338 (* 1 = 0.141338 loss)
I0407 23:18:41.641711 32718 sgd_solver.cpp:105] Iteration 6420, lr = 0.00343653
I0407 23:18:43.638267 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_6426.caffemodel
I0407 23:18:47.799854 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_6426.solverstate
I0407 23:18:52.114337 32718 solver.cpp:330] Iteration 6426, Testing net (#0)
I0407 23:18:52.114356 32718 net.cpp:676] Ignoring source layer train-data
I0407 23:18:54.101518 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:18:56.890308 32718 solver.cpp:397]     Test net output #0: accuracy = 0.474265
I0407 23:18:56.890357 32718 solver.cpp:397]     Test net output #1: loss = 2.83904 (* 1 = 2.83904 loss)
I0407 23:18:58.693482 32718 solver.cpp:218] Iteration 6432 (0.703741 iter/s, 17.0517s/12 iters), loss = 0.17615
I0407 23:18:58.693521 32718 solver.cpp:237]     Train net output #0: loss = 0.17615 (* 1 = 0.17615 loss)
I0407 23:18:58.693529 32718 sgd_solver.cpp:105] Iteration 6432, lr = 0.00342327
I0407 23:19:03.641363 32718 solver.cpp:218] Iteration 6444 (2.42532 iter/s, 4.94781s/12 iters), loss = 0.0698032
I0407 23:19:03.641409 32718 solver.cpp:237]     Train net output #0: loss = 0.0698032 (* 1 = 0.0698032 loss)
I0407 23:19:03.641418 32718 sgd_solver.cpp:105] Iteration 6444, lr = 0.00341004
I0407 23:19:08.582726 32718 solver.cpp:218] Iteration 6456 (2.42852 iter/s, 4.94128s/12 iters), loss = 0.132315
I0407 23:19:08.582769 32718 solver.cpp:237]     Train net output #0: loss = 0.132315 (* 1 = 0.132315 loss)
I0407 23:19:08.582778 32718 sgd_solver.cpp:105] Iteration 6456, lr = 0.00339683
I0407 23:19:13.541826 32718 solver.cpp:218] Iteration 6468 (2.41983 iter/s, 4.95902s/12 iters), loss = 0.114162
I0407 23:19:13.541868 32718 solver.cpp:237]     Train net output #0: loss = 0.114162 (* 1 = 0.114162 loss)
I0407 23:19:13.541877 32718 sgd_solver.cpp:105] Iteration 6468, lr = 0.00338365
I0407 23:19:15.488256 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:19:18.443581 32718 solver.cpp:218] Iteration 6480 (2.44814 iter/s, 4.90168s/12 iters), loss = 0.203358
I0407 23:19:18.443625 32718 solver.cpp:237]     Train net output #0: loss = 0.203358 (* 1 = 0.203358 loss)
I0407 23:19:18.443634 32718 sgd_solver.cpp:105] Iteration 6480, lr = 0.00337049
I0407 23:19:23.395584 32718 solver.cpp:218] Iteration 6492 (2.4233 iter/s, 4.95193s/12 iters), loss = 0.227772
I0407 23:19:23.395623 32718 solver.cpp:237]     Train net output #0: loss = 0.227772 (* 1 = 0.227772 loss)
I0407 23:19:23.395632 32718 sgd_solver.cpp:105] Iteration 6492, lr = 0.00335736
I0407 23:19:28.307530 32718 solver.cpp:218] Iteration 6504 (2.44306 iter/s, 4.91188s/12 iters), loss = 0.22279
I0407 23:19:28.307570 32718 solver.cpp:237]     Train net output #0: loss = 0.22279 (* 1 = 0.22279 loss)
I0407 23:19:28.307579 32718 sgd_solver.cpp:105] Iteration 6504, lr = 0.00334426
I0407 23:19:33.265372 32718 solver.cpp:218] Iteration 6516 (2.42044 iter/s, 4.95777s/12 iters), loss = 0.136096
I0407 23:19:33.265415 32718 solver.cpp:237]     Train net output #0: loss = 0.136096 (* 1 = 0.136096 loss)
I0407 23:19:33.265424 32718 sgd_solver.cpp:105] Iteration 6516, lr = 0.00333118
I0407 23:19:37.734728 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_6528.caffemodel
I0407 23:19:40.805294 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_6528.solverstate
I0407 23:19:43.175177 32718 solver.cpp:330] Iteration 6528, Testing net (#0)
I0407 23:19:43.175204 32718 net.cpp:676] Ignoring source layer train-data
I0407 23:19:45.058631 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:19:47.678129 32718 solver.cpp:397]     Test net output #0: accuracy = 0.466912
I0407 23:19:47.678287 32718 solver.cpp:397]     Test net output #1: loss = 2.79208 (* 1 = 2.79208 loss)
I0407 23:19:47.774708 32718 solver.cpp:218] Iteration 6528 (0.82706 iter/s, 14.5092s/12 iters), loss = 0.145186
I0407 23:19:47.774750 32718 solver.cpp:237]     Train net output #0: loss = 0.145186 (* 1 = 0.145186 loss)
I0407 23:19:47.774760 32718 sgd_solver.cpp:105] Iteration 6528, lr = 0.00331812
I0407 23:19:51.907079 32718 solver.cpp:218] Iteration 6540 (2.90395 iter/s, 4.1323s/12 iters), loss = 0.117267
I0407 23:19:51.907121 32718 solver.cpp:237]     Train net output #0: loss = 0.117267 (* 1 = 0.117267 loss)
I0407 23:19:51.907130 32718 sgd_solver.cpp:105] Iteration 6540, lr = 0.00330509
I0407 23:19:56.850056 32718 solver.cpp:218] Iteration 6552 (2.42772 iter/s, 4.94291s/12 iters), loss = 0.141742
I0407 23:19:56.850093 32718 solver.cpp:237]     Train net output #0: loss = 0.141742 (* 1 = 0.141742 loss)
I0407 23:19:56.850100 32718 sgd_solver.cpp:105] Iteration 6552, lr = 0.00329209
I0407 23:20:01.838418 32718 solver.cpp:218] Iteration 6564 (2.40563 iter/s, 4.9883s/12 iters), loss = 0.0873375
I0407 23:20:01.838455 32718 solver.cpp:237]     Train net output #0: loss = 0.0873375 (* 1 = 0.0873375 loss)
I0407 23:20:01.838464 32718 sgd_solver.cpp:105] Iteration 6564, lr = 0.00327911
I0407 23:20:05.982697 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:20:06.738075 32718 solver.cpp:218] Iteration 6576 (2.44919 iter/s, 4.89959s/12 iters), loss = 0.150235
I0407 23:20:06.738116 32718 solver.cpp:237]     Train net output #0: loss = 0.150235 (* 1 = 0.150235 loss)
I0407 23:20:06.738123 32718 sgd_solver.cpp:105] Iteration 6576, lr = 0.00326616
I0407 23:20:11.728776 32718 solver.cpp:218] Iteration 6588 (2.4045 iter/s, 4.99063s/12 iters), loss = 0.150185
I0407 23:20:11.728816 32718 solver.cpp:237]     Train net output #0: loss = 0.150185 (* 1 = 0.150185 loss)
I0407 23:20:11.728824 32718 sgd_solver.cpp:105] Iteration 6588, lr = 0.00325324
I0407 23:20:16.671105 32718 solver.cpp:218] Iteration 6600 (2.42804 iter/s, 4.94226s/12 iters), loss = 0.102571
I0407 23:20:16.671152 32718 solver.cpp:237]     Train net output #0: loss = 0.102571 (* 1 = 0.102571 loss)
I0407 23:20:16.671160 32718 sgd_solver.cpp:105] Iteration 6600, lr = 0.00324034
I0407 23:20:21.619982 32718 solver.cpp:218] Iteration 6612 (2.42483 iter/s, 4.9488s/12 iters), loss = 0.112732
I0407 23:20:21.620138 32718 solver.cpp:237]     Train net output #0: loss = 0.112732 (* 1 = 0.112732 loss)
I0407 23:20:21.620148 32718 sgd_solver.cpp:105] Iteration 6612, lr = 0.00322747
I0407 23:20:26.595379 32718 solver.cpp:218] Iteration 6624 (2.41196 iter/s, 4.9752s/12 iters), loss = 0.117559
I0407 23:20:26.595427 32718 solver.cpp:237]     Train net output #0: loss = 0.117559 (* 1 = 0.117559 loss)
I0407 23:20:26.595434 32718 sgd_solver.cpp:105] Iteration 6624, lr = 0.00321462
I0407 23:20:28.591533 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_6630.caffemodel
I0407 23:20:32.785874 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_6630.solverstate
I0407 23:20:35.998574 32718 solver.cpp:330] Iteration 6630, Testing net (#0)
I0407 23:20:35.998591 32718 net.cpp:676] Ignoring source layer train-data
I0407 23:20:37.923302 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:20:40.594611 32718 solver.cpp:397]     Test net output #0: accuracy = 0.481618
I0407 23:20:40.594655 32718 solver.cpp:397]     Test net output #1: loss = 2.86071 (* 1 = 2.86071 loss)
I0407 23:20:42.372614 32718 solver.cpp:218] Iteration 6636 (0.760594 iter/s, 15.7771s/12 iters), loss = 0.100096
I0407 23:20:42.372650 32718 solver.cpp:237]     Train net output #0: loss = 0.100096 (* 1 = 0.100096 loss)
I0407 23:20:42.372658 32718 sgd_solver.cpp:105] Iteration 6636, lr = 0.00320181
I0407 23:20:47.337990 32718 solver.cpp:218] Iteration 6648 (2.41677 iter/s, 4.96531s/12 iters), loss = 0.187368
I0407 23:20:47.338027 32718 solver.cpp:237]     Train net output #0: loss = 0.187368 (* 1 = 0.187368 loss)
I0407 23:20:47.338035 32718 sgd_solver.cpp:105] Iteration 6648, lr = 0.00318902
I0407 23:20:52.302371 32718 solver.cpp:218] Iteration 6660 (2.41725 iter/s, 4.96432s/12 iters), loss = 0.21011
I0407 23:20:52.302495 32718 solver.cpp:237]     Train net output #0: loss = 0.21011 (* 1 = 0.21011 loss)
I0407 23:20:52.302502 32718 sgd_solver.cpp:105] Iteration 6660, lr = 0.00317625
I0407 23:20:57.248811 32718 solver.cpp:218] Iteration 6672 (2.42606 iter/s, 4.94629s/12 iters), loss = 0.150732
I0407 23:20:57.248847 32718 solver.cpp:237]     Train net output #0: loss = 0.150732 (* 1 = 0.150732 loss)
I0407 23:20:57.248854 32718 sgd_solver.cpp:105] Iteration 6672, lr = 0.00316352
I0407 23:20:58.570346 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:21:02.223121 32718 solver.cpp:218] Iteration 6684 (2.41243 iter/s, 4.97425s/12 iters), loss = 0.109327
I0407 23:21:02.223158 32718 solver.cpp:237]     Train net output #0: loss = 0.109327 (* 1 = 0.109327 loss)
I0407 23:21:02.223166 32718 sgd_solver.cpp:105] Iteration 6684, lr = 0.00315081
I0407 23:21:07.222260 32718 solver.cpp:218] Iteration 6696 (2.40044 iter/s, 4.99907s/12 iters), loss = 0.181092
I0407 23:21:07.222297 32718 solver.cpp:237]     Train net output #0: loss = 0.181092 (* 1 = 0.181092 loss)
I0407 23:21:07.222306 32718 sgd_solver.cpp:105] Iteration 6696, lr = 0.00313813
I0407 23:21:12.125809 32718 solver.cpp:218] Iteration 6708 (2.44724 iter/s, 4.90348s/12 iters), loss = 0.0909493
I0407 23:21:12.125849 32718 solver.cpp:237]     Train net output #0: loss = 0.0909493 (* 1 = 0.0909493 loss)
I0407 23:21:12.125855 32718 sgd_solver.cpp:105] Iteration 6708, lr = 0.00312548
I0407 23:21:17.084897 32718 solver.cpp:218] Iteration 6720 (2.41983 iter/s, 4.95902s/12 iters), loss = 0.0625245
I0407 23:21:17.084936 32718 solver.cpp:237]     Train net output #0: loss = 0.0625245 (* 1 = 0.0625245 loss)
I0407 23:21:17.084944 32718 sgd_solver.cpp:105] Iteration 6720, lr = 0.00311285
I0407 23:21:21.545794 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_6732.caffemodel
I0407 23:21:24.614598 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_6732.solverstate
I0407 23:21:26.981302 32718 solver.cpp:330] Iteration 6732, Testing net (#0)
I0407 23:21:26.981318 32718 net.cpp:676] Ignoring source layer train-data
I0407 23:21:28.750092 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:21:31.460608 32718 solver.cpp:397]     Test net output #0: accuracy = 0.473652
I0407 23:21:31.460655 32718 solver.cpp:397]     Test net output #1: loss = 2.84236 (* 1 = 2.84236 loss)
I0407 23:21:31.557365 32718 solver.cpp:218] Iteration 6732 (0.829166 iter/s, 14.4724s/12 iters), loss = 0.120756
I0407 23:21:31.557411 32718 solver.cpp:237]     Train net output #0: loss = 0.120756 (* 1 = 0.120756 loss)
I0407 23:21:31.557420 32718 sgd_solver.cpp:105] Iteration 6732, lr = 0.00310026
I0407 23:21:35.628823 32718 solver.cpp:218] Iteration 6744 (2.9474 iter/s, 4.07139s/12 iters), loss = 0.158553
I0407 23:21:35.628859 32718 solver.cpp:237]     Train net output #0: loss = 0.158553 (* 1 = 0.158553 loss)
I0407 23:21:35.628868 32718 sgd_solver.cpp:105] Iteration 6744, lr = 0.00308769
I0407 23:21:40.560073 32718 solver.cpp:218] Iteration 6756 (2.43349 iter/s, 4.93118s/12 iters), loss = 0.168803
I0407 23:21:40.560112 32718 solver.cpp:237]     Train net output #0: loss = 0.168803 (* 1 = 0.168803 loss)
I0407 23:21:40.560122 32718 sgd_solver.cpp:105] Iteration 6756, lr = 0.00307515
I0407 23:21:45.500576 32718 solver.cpp:218] Iteration 6768 (2.42894 iter/s, 4.94043s/12 iters), loss = 0.0999142
I0407 23:21:45.500620 32718 solver.cpp:237]     Train net output #0: loss = 0.0999142 (* 1 = 0.0999142 loss)
I0407 23:21:45.500628 32718 sgd_solver.cpp:105] Iteration 6768, lr = 0.00306263
I0407 23:21:48.936156 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:21:50.446470 32718 solver.cpp:218] Iteration 6780 (2.42629 iter/s, 4.94582s/12 iters), loss = 0.107978
I0407 23:21:50.446512 32718 solver.cpp:237]     Train net output #0: loss = 0.107978 (* 1 = 0.107978 loss)
I0407 23:21:50.446521 32718 sgd_solver.cpp:105] Iteration 6780, lr = 0.00305015
I0407 23:21:55.380384 32718 solver.cpp:218] Iteration 6792 (2.43218 iter/s, 4.93384s/12 iters), loss = 0.128956
I0407 23:21:55.380527 32718 solver.cpp:237]     Train net output #0: loss = 0.128956 (* 1 = 0.128956 loss)
I0407 23:21:55.380537 32718 sgd_solver.cpp:105] Iteration 6792, lr = 0.00303769
I0407 23:22:00.325978 32718 solver.cpp:218] Iteration 6804 (2.42648 iter/s, 4.94543s/12 iters), loss = 0.13304
I0407 23:22:00.326015 32718 solver.cpp:237]     Train net output #0: loss = 0.13304 (* 1 = 0.13304 loss)
I0407 23:22:00.326023 32718 sgd_solver.cpp:105] Iteration 6804, lr = 0.00302527
I0407 23:22:05.264951 32718 solver.cpp:218] Iteration 6816 (2.42969 iter/s, 4.93891s/12 iters), loss = 0.0618262
I0407 23:22:05.264984 32718 solver.cpp:237]     Train net output #0: loss = 0.0618262 (* 1 = 0.0618262 loss)
I0407 23:22:05.264991 32718 sgd_solver.cpp:105] Iteration 6816, lr = 0.00301287
I0407 23:22:10.241096 32718 solver.cpp:218] Iteration 6828 (2.41154 iter/s, 4.97607s/12 iters), loss = 0.0591927
I0407 23:22:10.241153 32718 solver.cpp:237]     Train net output #0: loss = 0.0591928 (* 1 = 0.0591928 loss)
I0407 23:22:10.241166 32718 sgd_solver.cpp:105] Iteration 6828, lr = 0.0030005
I0407 23:22:12.231320 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_6834.caffemodel
I0407 23:22:15.736647 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_6834.solverstate
I0407 23:22:19.130089 32718 solver.cpp:330] Iteration 6834, Testing net (#0)
I0407 23:22:19.130105 32718 net.cpp:676] Ignoring source layer train-data
I0407 23:22:20.965101 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:22:23.864514 32718 solver.cpp:397]     Test net output #0: accuracy = 0.484681
I0407 23:22:23.864560 32718 solver.cpp:397]     Test net output #1: loss = 2.85453 (* 1 = 2.85453 loss)
I0407 23:22:25.783458 32718 solver.cpp:218] Iteration 6840 (0.772089 iter/s, 15.5422s/12 iters), loss = 0.340952
I0407 23:22:25.783599 32718 solver.cpp:237]     Train net output #0: loss = 0.340952 (* 1 = 0.340952 loss)
I0407 23:22:25.783609 32718 sgd_solver.cpp:105] Iteration 6840, lr = 0.00298816
I0407 23:22:30.798733 32718 solver.cpp:218] Iteration 6852 (2.39277 iter/s, 5.01511s/12 iters), loss = 0.0890957
I0407 23:22:30.798771 32718 solver.cpp:237]     Train net output #0: loss = 0.0890958 (* 1 = 0.0890958 loss)
I0407 23:22:30.798780 32718 sgd_solver.cpp:105] Iteration 6852, lr = 0.00297585
I0407 23:22:35.896039 32718 solver.cpp:218] Iteration 6864 (2.35422 iter/s, 5.09723s/12 iters), loss = 0.0726074
I0407 23:22:35.896082 32718 solver.cpp:237]     Train net output #0: loss = 0.0726074 (* 1 = 0.0726074 loss)
I0407 23:22:35.896090 32718 sgd_solver.cpp:105] Iteration 6864, lr = 0.00296357
I0407 23:22:40.820291 32718 solver.cpp:218] Iteration 6876 (2.43696 iter/s, 4.92417s/12 iters), loss = 0.11508
I0407 23:22:40.820334 32718 solver.cpp:237]     Train net output #0: loss = 0.11508 (* 1 = 0.11508 loss)
I0407 23:22:40.820343 32718 sgd_solver.cpp:105] Iteration 6876, lr = 0.00295132
I0407 23:22:41.356704 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:22:45.663805 32718 solver.cpp:218] Iteration 6888 (2.47758 iter/s, 4.84344s/12 iters), loss = 0.0966519
I0407 23:22:45.663841 32718 solver.cpp:237]     Train net output #0: loss = 0.0966519 (* 1 = 0.0966519 loss)
I0407 23:22:45.663849 32718 sgd_solver.cpp:105] Iteration 6888, lr = 0.0029391
I0407 23:22:50.638397 32718 solver.cpp:218] Iteration 6900 (2.41229 iter/s, 4.97453s/12 iters), loss = 0.0803192
I0407 23:22:50.638439 32718 solver.cpp:237]     Train net output #0: loss = 0.0803192 (* 1 = 0.0803192 loss)
I0407 23:22:50.638447 32718 sgd_solver.cpp:105] Iteration 6900, lr = 0.0029269
I0407 23:22:55.562803 32718 solver.cpp:218] Iteration 6912 (2.43688 iter/s, 4.92434s/12 iters), loss = 0.108753
I0407 23:22:55.562836 32718 solver.cpp:237]     Train net output #0: loss = 0.108753 (* 1 = 0.108753 loss)
I0407 23:22:55.562844 32718 sgd_solver.cpp:105] Iteration 6912, lr = 0.00291474
I0407 23:23:00.538659 32718 solver.cpp:218] Iteration 6924 (2.41167 iter/s, 4.9758s/12 iters), loss = 0.067289
I0407 23:23:00.538811 32718 solver.cpp:237]     Train net output #0: loss = 0.067289 (* 1 = 0.067289 loss)
I0407 23:23:00.538822 32718 sgd_solver.cpp:105] Iteration 6924, lr = 0.00290261
I0407 23:23:05.048343 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_6936.caffemodel
I0407 23:23:08.113930 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_6936.solverstate
I0407 23:23:10.529723 32718 solver.cpp:330] Iteration 6936, Testing net (#0)
I0407 23:23:10.529740 32718 net.cpp:676] Ignoring source layer train-data
I0407 23:23:11.143926 32718 blocking_queue.cpp:49] Waiting for data
I0407 23:23:12.257228 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:23:15.045269 32718 solver.cpp:397]     Test net output #0: accuracy = 0.485294
I0407 23:23:15.045315 32718 solver.cpp:397]     Test net output #1: loss = 2.85075 (* 1 = 2.85075 loss)
I0407 23:23:15.141888 32718 solver.cpp:218] Iteration 6936 (0.821748 iter/s, 14.603s/12 iters), loss = 0.201052
I0407 23:23:15.141933 32718 solver.cpp:237]     Train net output #0: loss = 0.201052 (* 1 = 0.201052 loss)
I0407 23:23:15.141942 32718 sgd_solver.cpp:105] Iteration 6936, lr = 0.0028905
I0407 23:23:19.228102 32718 solver.cpp:218] Iteration 6948 (2.93676 iter/s, 4.08614s/12 iters), loss = 0.0758336
I0407 23:23:19.228144 32718 solver.cpp:237]     Train net output #0: loss = 0.0758336 (* 1 = 0.0758336 loss)
I0407 23:23:19.228152 32718 sgd_solver.cpp:105] Iteration 6948, lr = 0.00287843
I0407 23:23:24.195787 32718 solver.cpp:218] Iteration 6960 (2.41565 iter/s, 4.9676s/12 iters), loss = 0.117501
I0407 23:23:24.195854 32718 solver.cpp:237]     Train net output #0: loss = 0.117501 (* 1 = 0.117501 loss)
I0407 23:23:24.195866 32718 sgd_solver.cpp:105] Iteration 6960, lr = 0.00286639
I0407 23:23:29.119216 32718 solver.cpp:218] Iteration 6972 (2.43737 iter/s, 4.92333s/12 iters), loss = 0.119741
I0407 23:23:29.119262 32718 solver.cpp:237]     Train net output #0: loss = 0.119741 (* 1 = 0.119741 loss)
I0407 23:23:29.119271 32718 sgd_solver.cpp:105] Iteration 6972, lr = 0.00285438
I0407 23:23:31.832007 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:23:34.069370 32718 solver.cpp:218] Iteration 6984 (2.42421 iter/s, 4.95007s/12 iters), loss = 0.0629897
I0407 23:23:34.069415 32718 solver.cpp:237]     Train net output #0: loss = 0.0629897 (* 1 = 0.0629897 loss)
I0407 23:23:34.069424 32718 sgd_solver.cpp:105] Iteration 6984, lr = 0.00284239
I0407 23:23:38.998080 32718 solver.cpp:218] Iteration 6996 (2.43475 iter/s, 4.92864s/12 iters), loss = 0.133893
I0407 23:23:38.998119 32718 solver.cpp:237]     Train net output #0: loss = 0.133893 (* 1 = 0.133893 loss)
I0407 23:23:38.998127 32718 sgd_solver.cpp:105] Iteration 6996, lr = 0.00283044
I0407 23:23:43.943845 32718 solver.cpp:218] Iteration 7008 (2.42635 iter/s, 4.9457s/12 iters), loss = 0.0614158
I0407 23:23:43.943883 32718 solver.cpp:237]     Train net output #0: loss = 0.0614158 (* 1 = 0.0614158 loss)
I0407 23:23:43.943892 32718 sgd_solver.cpp:105] Iteration 7008, lr = 0.00281852
I0407 23:23:48.847092 32718 solver.cpp:218] Iteration 7020 (2.44739 iter/s, 4.90318s/12 iters), loss = 0.114955
I0407 23:23:48.847142 32718 solver.cpp:237]     Train net output #0: loss = 0.114955 (* 1 = 0.114955 loss)
I0407 23:23:48.847151 32718 sgd_solver.cpp:105] Iteration 7020, lr = 0.00280663
I0407 23:23:53.808979 32718 solver.cpp:218] Iteration 7032 (2.41847 iter/s, 4.96181s/12 iters), loss = 0.131383
I0407 23:23:53.809022 32718 solver.cpp:237]     Train net output #0: loss = 0.131383 (* 1 = 0.131383 loss)
I0407 23:23:53.809031 32718 sgd_solver.cpp:105] Iteration 7032, lr = 0.00279477
I0407 23:23:55.874150 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_7038.caffemodel
I0407 23:24:00.121731 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_7038.solverstate
I0407 23:24:03.466851 32718 solver.cpp:330] Iteration 7038, Testing net (#0)
I0407 23:24:03.466948 32718 net.cpp:676] Ignoring source layer train-data
I0407 23:24:05.207357 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:24:08.223698 32718 solver.cpp:397]     Test net output #0: accuracy = 0.489583
I0407 23:24:08.223739 32718 solver.cpp:397]     Test net output #1: loss = 2.88694 (* 1 = 2.88694 loss)
I0407 23:24:10.035498 32718 solver.cpp:218] Iteration 7044 (0.739535 iter/s, 16.2264s/12 iters), loss = 0.113657
I0407 23:24:10.035540 32718 solver.cpp:237]     Train net output #0: loss = 0.113657 (* 1 = 0.113657 loss)
I0407 23:24:10.035548 32718 sgd_solver.cpp:105] Iteration 7044, lr = 0.00278294
I0407 23:24:14.949920 32718 solver.cpp:218] Iteration 7056 (2.44183 iter/s, 4.91435s/12 iters), loss = 0.182145
I0407 23:24:14.949960 32718 solver.cpp:237]     Train net output #0: loss = 0.182145 (* 1 = 0.182145 loss)
I0407 23:24:14.949970 32718 sgd_solver.cpp:105] Iteration 7056, lr = 0.00277114
I0407 23:24:19.941880 32718 solver.cpp:218] Iteration 7068 (2.4039 iter/s, 4.99189s/12 iters), loss = 0.108546
I0407 23:24:19.941926 32718 solver.cpp:237]     Train net output #0: loss = 0.108546 (* 1 = 0.108546 loss)
I0407 23:24:19.941934 32718 sgd_solver.cpp:105] Iteration 7068, lr = 0.00275937
I0407 23:24:24.754539 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:24:24.865255 32718 solver.cpp:218] Iteration 7080 (2.43739 iter/s, 4.9233s/12 iters), loss = 0.270629
I0407 23:24:24.865296 32718 solver.cpp:237]     Train net output #0: loss = 0.270629 (* 1 = 0.270629 loss)
I0407 23:24:24.865305 32718 sgd_solver.cpp:105] Iteration 7080, lr = 0.00274763
I0407 23:24:29.842873 32718 solver.cpp:218] Iteration 7092 (2.41083 iter/s, 4.97755s/12 iters), loss = 0.106521
I0407 23:24:29.842916 32718 solver.cpp:237]     Train net output #0: loss = 0.10652 (* 1 = 0.10652 loss)
I0407 23:24:29.842923 32718 sgd_solver.cpp:105] Iteration 7092, lr = 0.00273593
I0407 23:24:34.837683 32718 solver.cpp:218] Iteration 7104 (2.40253 iter/s, 4.99474s/12 iters), loss = 0.210721
I0407 23:24:34.837810 32718 solver.cpp:237]     Train net output #0: loss = 0.210721 (* 1 = 0.210721 loss)
I0407 23:24:34.837818 32718 sgd_solver.cpp:105] Iteration 7104, lr = 0.00272425
I0407 23:24:39.752970 32718 solver.cpp:218] Iteration 7116 (2.44144 iter/s, 4.91513s/12 iters), loss = 0.137267
I0407 23:24:39.753013 32718 solver.cpp:237]     Train net output #0: loss = 0.137267 (* 1 = 0.137267 loss)
I0407 23:24:39.753023 32718 sgd_solver.cpp:105] Iteration 7116, lr = 0.00271261
I0407 23:24:44.735502 32718 solver.cpp:218] Iteration 7128 (2.40845 iter/s, 4.98245s/12 iters), loss = 0.104648
I0407 23:24:44.735539 32718 solver.cpp:237]     Train net output #0: loss = 0.104648 (* 1 = 0.104648 loss)
I0407 23:24:44.735548 32718 sgd_solver.cpp:105] Iteration 7128, lr = 0.002701
I0407 23:24:49.267112 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_7140.caffemodel
I0407 23:24:52.378823 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_7140.solverstate
I0407 23:24:54.787854 32718 solver.cpp:330] Iteration 7140, Testing net (#0)
I0407 23:24:54.787871 32718 net.cpp:676] Ignoring source layer train-data
I0407 23:24:56.418367 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:24:59.238323 32718 solver.cpp:397]     Test net output #0: accuracy = 0.500613
I0407 23:24:59.238370 32718 solver.cpp:397]     Test net output #1: loss = 2.79921 (* 1 = 2.79921 loss)
I0407 23:24:59.334833 32718 solver.cpp:218] Iteration 7140 (0.821961 iter/s, 14.5992s/12 iters), loss = 0.134791
I0407 23:24:59.334872 32718 solver.cpp:237]     Train net output #0: loss = 0.13479 (* 1 = 0.13479 loss)
I0407 23:24:59.334879 32718 sgd_solver.cpp:105] Iteration 7140, lr = 0.00268941
I0407 23:25:03.460888 32718 solver.cpp:218] Iteration 7152 (2.90839 iter/s, 4.12599s/12 iters), loss = 0.0898209
I0407 23:25:03.460927 32718 solver.cpp:237]     Train net output #0: loss = 0.0898209 (* 1 = 0.0898209 loss)
I0407 23:25:03.460935 32718 sgd_solver.cpp:105] Iteration 7152, lr = 0.00267786
I0407 23:25:08.415722 32718 solver.cpp:218] Iteration 7164 (2.42191 iter/s, 4.95477s/12 iters), loss = 0.11443
I0407 23:25:08.415858 32718 solver.cpp:237]     Train net output #0: loss = 0.11443 (* 1 = 0.11443 loss)
I0407 23:25:08.415866 32718 sgd_solver.cpp:105] Iteration 7164, lr = 0.00266635
I0407 23:25:13.383999 32718 solver.cpp:218] Iteration 7176 (2.41541 iter/s, 4.96811s/12 iters), loss = 0.0832127
I0407 23:25:13.384044 32718 solver.cpp:237]     Train net output #0: loss = 0.0832127 (* 1 = 0.0832127 loss)
I0407 23:25:13.384052 32718 sgd_solver.cpp:105] Iteration 7176, lr = 0.00265486
I0407 23:25:15.500808 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:25:18.378773 32718 solver.cpp:218] Iteration 7188 (2.40255 iter/s, 4.9947s/12 iters), loss = 0.13745
I0407 23:25:18.378818 32718 solver.cpp:237]     Train net output #0: loss = 0.13745 (* 1 = 0.13745 loss)
I0407 23:25:18.378826 32718 sgd_solver.cpp:105] Iteration 7188, lr = 0.0026434
I0407 23:25:23.361120 32718 solver.cpp:218] Iteration 7200 (2.40854 iter/s, 4.98228s/12 iters), loss = 0.153436
I0407 23:25:23.361158 32718 solver.cpp:237]     Train net output #0: loss = 0.153436 (* 1 = 0.153436 loss)
I0407 23:25:23.361166 32718 sgd_solver.cpp:105] Iteration 7200, lr = 0.00263198
I0407 23:25:28.316924 32718 solver.cpp:218] Iteration 7212 (2.42144 iter/s, 4.95574s/12 iters), loss = 0.122278
I0407 23:25:28.316964 32718 solver.cpp:237]     Train net output #0: loss = 0.122278 (* 1 = 0.122278 loss)
I0407 23:25:28.316972 32718 sgd_solver.cpp:105] Iteration 7212, lr = 0.00262059
I0407 23:25:33.221151 32718 solver.cpp:218] Iteration 7224 (2.4469 iter/s, 4.90416s/12 iters), loss = 0.0875321
I0407 23:25:33.221187 32718 solver.cpp:237]     Train net output #0: loss = 0.0875322 (* 1 = 0.0875322 loss)
I0407 23:25:33.221194 32718 sgd_solver.cpp:105] Iteration 7224, lr = 0.00260923
I0407 23:25:38.182601 32718 solver.cpp:218] Iteration 7236 (2.41868 iter/s, 4.96139s/12 iters), loss = 0.047358
I0407 23:25:38.182639 32718 solver.cpp:237]     Train net output #0: loss = 0.047358 (* 1 = 0.047358 loss)
I0407 23:25:38.182646 32718 sgd_solver.cpp:105] Iteration 7236, lr = 0.0025979
I0407 23:25:40.184844 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_7242.caffemodel
I0407 23:25:43.866883 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_7242.solverstate
I0407 23:25:46.901350 32718 solver.cpp:330] Iteration 7242, Testing net (#0)
I0407 23:25:46.901368 32718 net.cpp:676] Ignoring source layer train-data
I0407 23:25:48.418316 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:25:51.270036 32718 solver.cpp:397]     Test net output #0: accuracy = 0.503676
I0407 23:25:51.270079 32718 solver.cpp:397]     Test net output #1: loss = 2.80472 (* 1 = 2.80472 loss)
I0407 23:25:53.076879 32718 solver.cpp:218] Iteration 7248 (0.805684 iter/s, 14.8942s/12 iters), loss = 0.0971339
I0407 23:25:53.076918 32718 solver.cpp:237]     Train net output #0: loss = 0.0971339 (* 1 = 0.0971339 loss)
I0407 23:25:53.076926 32718 sgd_solver.cpp:105] Iteration 7248, lr = 0.00258661
I0407 23:25:57.998442 32718 solver.cpp:218] Iteration 7260 (2.43829 iter/s, 4.92149s/12 iters), loss = 0.106712
I0407 23:25:57.998486 32718 solver.cpp:237]     Train net output #0: loss = 0.106712 (* 1 = 0.106712 loss)
I0407 23:25:57.998495 32718 sgd_solver.cpp:105] Iteration 7260, lr = 0.00257534
I0407 23:26:02.965097 32718 solver.cpp:218] Iteration 7272 (2.41615 iter/s, 4.96658s/12 iters), loss = 0.134101
I0407 23:26:02.965142 32718 solver.cpp:237]     Train net output #0: loss = 0.134101 (* 1 = 0.134101 loss)
I0407 23:26:02.965152 32718 sgd_solver.cpp:105] Iteration 7272, lr = 0.00256411
I0407 23:26:07.129315 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:26:07.868561 32718 solver.cpp:218] Iteration 7284 (2.44729 iter/s, 4.90338s/12 iters), loss = 0.186677
I0407 23:26:07.868611 32718 solver.cpp:237]     Train net output #0: loss = 0.186677 (* 1 = 0.186677 loss)
I0407 23:26:07.868620 32718 sgd_solver.cpp:105] Iteration 7284, lr = 0.00255291
I0407 23:26:12.838802 32718 solver.cpp:218] Iteration 7296 (2.41441 iter/s, 4.97016s/12 iters), loss = 0.0552554
I0407 23:26:12.838924 32718 solver.cpp:237]     Train net output #0: loss = 0.0552554 (* 1 = 0.0552554 loss)
I0407 23:26:12.838933 32718 sgd_solver.cpp:105] Iteration 7296, lr = 0.00254174
I0407 23:26:17.793177 32718 solver.cpp:218] Iteration 7308 (2.42218 iter/s, 4.95422s/12 iters), loss = 0.121566
I0407 23:26:17.793220 32718 solver.cpp:237]     Train net output #0: loss = 0.121566 (* 1 = 0.121566 loss)
I0407 23:26:17.793228 32718 sgd_solver.cpp:105] Iteration 7308, lr = 0.00253061
I0407 23:26:22.697803 32718 solver.cpp:218] Iteration 7320 (2.44671 iter/s, 4.90455s/12 iters), loss = 0.040907
I0407 23:26:22.697845 32718 solver.cpp:237]     Train net output #0: loss = 0.040907 (* 1 = 0.040907 loss)
I0407 23:26:22.697854 32718 sgd_solver.cpp:105] Iteration 7320, lr = 0.00251951
I0407 23:26:27.648636 32718 solver.cpp:218] Iteration 7332 (2.42387 iter/s, 4.95076s/12 iters), loss = 0.0392373
I0407 23:26:27.648681 32718 solver.cpp:237]     Train net output #0: loss = 0.0392373 (* 1 = 0.0392373 loss)
I0407 23:26:27.648690 32718 sgd_solver.cpp:105] Iteration 7332, lr = 0.00250844
I0407 23:26:32.110550 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_7344.caffemodel
I0407 23:26:36.499876 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_7344.solverstate
I0407 23:26:38.873983 32718 solver.cpp:330] Iteration 7344, Testing net (#0)
I0407 23:26:38.874004 32718 net.cpp:676] Ignoring source layer train-data
I0407 23:26:40.502009 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:26:43.665222 32718 solver.cpp:397]     Test net output #0: accuracy = 0.495711
I0407 23:26:43.665395 32718 solver.cpp:397]     Test net output #1: loss = 2.84463 (* 1 = 2.84463 loss)
I0407 23:26:43.761756 32718 solver.cpp:218] Iteration 7344 (0.74474 iter/s, 16.113s/12 iters), loss = 0.223308
I0407 23:26:43.761799 32718 solver.cpp:237]     Train net output #0: loss = 0.223308 (* 1 = 0.223308 loss)
I0407 23:26:43.761808 32718 sgd_solver.cpp:105] Iteration 7344, lr = 0.0024974
I0407 23:26:47.876760 32718 solver.cpp:218] Iteration 7356 (2.91621 iter/s, 4.11493s/12 iters), loss = 0.178581
I0407 23:26:47.876798 32718 solver.cpp:237]     Train net output #0: loss = 0.178581 (* 1 = 0.178581 loss)
I0407 23:26:47.876806 32718 sgd_solver.cpp:105] Iteration 7356, lr = 0.00248639
I0407 23:26:52.851467 32718 solver.cpp:218] Iteration 7368 (2.41223 iter/s, 4.97464s/12 iters), loss = 0.0528688
I0407 23:26:52.851505 32718 solver.cpp:237]     Train net output #0: loss = 0.0528688 (* 1 = 0.0528688 loss)
I0407 23:26:52.851511 32718 sgd_solver.cpp:105] Iteration 7368, lr = 0.00247542
I0407 23:26:57.793015 32718 solver.cpp:218] Iteration 7380 (2.42842 iter/s, 4.94149s/12 iters), loss = 0.115768
I0407 23:26:57.793057 32718 solver.cpp:237]     Train net output #0: loss = 0.115768 (* 1 = 0.115768 loss)
I0407 23:26:57.793064 32718 sgd_solver.cpp:105] Iteration 7380, lr = 0.00246448
I0407 23:26:59.148586 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:27:02.709689 32718 solver.cpp:218] Iteration 7392 (2.44071 iter/s, 4.9166s/12 iters), loss = 0.0857693
I0407 23:27:02.709729 32718 solver.cpp:237]     Train net output #0: loss = 0.0857694 (* 1 = 0.0857694 loss)
I0407 23:27:02.709736 32718 sgd_solver.cpp:105] Iteration 7392, lr = 0.00245357
I0407 23:27:07.769726 32718 solver.cpp:218] Iteration 7404 (2.37156 iter/s, 5.05997s/12 iters), loss = 0.102597
I0407 23:27:07.769767 32718 solver.cpp:237]     Train net output #0: loss = 0.102597 (* 1 = 0.102597 loss)
I0407 23:27:07.769773 32718 sgd_solver.cpp:105] Iteration 7404, lr = 0.0024427
I0407 23:27:12.716094 32718 solver.cpp:218] Iteration 7416 (2.42606 iter/s, 4.9463s/12 iters), loss = 0.0567761
I0407 23:27:12.716126 32718 solver.cpp:237]     Train net output #0: loss = 0.0567761 (* 1 = 0.0567761 loss)
I0407 23:27:12.716135 32718 sgd_solver.cpp:105] Iteration 7416, lr = 0.00243185
I0407 23:27:17.631129 32718 solver.cpp:218] Iteration 7428 (2.44152 iter/s, 4.91496s/12 iters), loss = 0.0970025
I0407 23:27:17.631294 32718 solver.cpp:237]     Train net output #0: loss = 0.0970025 (* 1 = 0.0970025 loss)
I0407 23:27:17.631304 32718 sgd_solver.cpp:105] Iteration 7428, lr = 0.00242104
I0407 23:27:22.602633 32718 solver.cpp:218] Iteration 7440 (2.41385 iter/s, 4.97132s/12 iters), loss = 0.0537865
I0407 23:27:22.602676 32718 solver.cpp:237]     Train net output #0: loss = 0.0537865 (* 1 = 0.0537865 loss)
I0407 23:27:22.602685 32718 sgd_solver.cpp:105] Iteration 7440, lr = 0.00241027
I0407 23:27:24.601308 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_7446.caffemodel
I0407 23:27:30.291127 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_7446.solverstate
I0407 23:27:38.422639 32718 solver.cpp:330] Iteration 7446, Testing net (#0)
I0407 23:27:38.422659 32718 net.cpp:676] Ignoring source layer train-data
I0407 23:27:39.990347 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:27:43.207396 32718 solver.cpp:397]     Test net output #0: accuracy = 0.501838
I0407 23:27:43.207443 32718 solver.cpp:397]     Test net output #1: loss = 2.85095 (* 1 = 2.85095 loss)
I0407 23:27:44.961833 32718 solver.cpp:218] Iteration 7452 (0.536695 iter/s, 22.3591s/12 iters), loss = 0.0709224
I0407 23:27:44.961877 32718 solver.cpp:237]     Train net output #0: loss = 0.0709224 (* 1 = 0.0709224 loss)
I0407 23:27:44.961885 32718 sgd_solver.cpp:105] Iteration 7452, lr = 0.00239952
I0407 23:27:49.821792 32718 solver.cpp:218] Iteration 7464 (2.46919 iter/s, 4.85988s/12 iters), loss = 0.0165189
I0407 23:27:49.821913 32718 solver.cpp:237]     Train net output #0: loss = 0.0165189 (* 1 = 0.0165189 loss)
I0407 23:27:49.821921 32718 sgd_solver.cpp:105] Iteration 7464, lr = 0.00238881
I0407 23:27:54.814180 32718 solver.cpp:218] Iteration 7476 (2.40373 iter/s, 4.99224s/12 iters), loss = 0.0916946
I0407 23:27:54.814222 32718 solver.cpp:237]     Train net output #0: loss = 0.0916947 (* 1 = 0.0916947 loss)
I0407 23:27:54.814231 32718 sgd_solver.cpp:105] Iteration 7476, lr = 0.00237813
I0407 23:27:58.311432 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:27:59.764204 32718 solver.cpp:218] Iteration 7488 (2.42427 iter/s, 4.94995s/12 iters), loss = 0.118324
I0407 23:27:59.764251 32718 solver.cpp:237]     Train net output #0: loss = 0.118324 (* 1 = 0.118324 loss)
I0407 23:27:59.764261 32718 sgd_solver.cpp:105] Iteration 7488, lr = 0.00236749
I0407 23:28:04.728385 32718 solver.cpp:218] Iteration 7500 (2.41736 iter/s, 4.9641s/12 iters), loss = 0.0415497
I0407 23:28:04.728446 32718 solver.cpp:237]     Train net output #0: loss = 0.0415497 (* 1 = 0.0415497 loss)
I0407 23:28:04.728456 32718 sgd_solver.cpp:105] Iteration 7500, lr = 0.00235687
I0407 23:28:09.701154 32718 solver.cpp:218] Iteration 7512 (2.41319 iter/s, 4.97267s/12 iters), loss = 0.105863
I0407 23:28:09.701221 32718 solver.cpp:237]     Train net output #0: loss = 0.105863 (* 1 = 0.105863 loss)
I0407 23:28:09.701236 32718 sgd_solver.cpp:105] Iteration 7512, lr = 0.00234629
I0407 23:28:14.659559 32718 solver.cpp:218] Iteration 7524 (2.42018 iter/s, 4.95832s/12 iters), loss = 0.058018
I0407 23:28:14.659596 32718 solver.cpp:237]     Train net output #0: loss = 0.058018 (* 1 = 0.058018 loss)
I0407 23:28:14.659605 32718 sgd_solver.cpp:105] Iteration 7524, lr = 0.00233575
I0407 23:28:19.576568 32718 solver.cpp:218] Iteration 7536 (2.44054 iter/s, 4.91694s/12 iters), loss = 0.0480961
I0407 23:28:19.576613 32718 solver.cpp:237]     Train net output #0: loss = 0.0480961 (* 1 = 0.0480961 loss)
I0407 23:28:19.576622 32718 sgd_solver.cpp:105] Iteration 7536, lr = 0.00232523
I0407 23:28:24.062072 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_7548.caffemodel
I0407 23:28:27.140717 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_7548.solverstate
I0407 23:28:31.570619 32718 solver.cpp:330] Iteration 7548, Testing net (#0)
I0407 23:28:31.570638 32718 net.cpp:676] Ignoring source layer train-data
I0407 23:28:33.096066 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:28:36.341048 32718 solver.cpp:397]     Test net output #0: accuracy = 0.494485
I0407 23:28:36.341094 32718 solver.cpp:397]     Test net output #1: loss = 2.80866 (* 1 = 2.80866 loss)
I0407 23:28:36.437490 32718 solver.cpp:218] Iteration 7548 (0.711709 iter/s, 16.8608s/12 iters), loss = 0.0705691
I0407 23:28:36.437534 32718 solver.cpp:237]     Train net output #0: loss = 0.0705692 (* 1 = 0.0705692 loss)
I0407 23:28:36.437543 32718 sgd_solver.cpp:105] Iteration 7548, lr = 0.00231475
I0407 23:28:40.690182 32718 solver.cpp:218] Iteration 7560 (2.82179 iter/s, 4.25262s/12 iters), loss = 0.0629695
I0407 23:28:40.690217 32718 solver.cpp:237]     Train net output #0: loss = 0.0629695 (* 1 = 0.0629695 loss)
I0407 23:28:40.690224 32718 sgd_solver.cpp:105] Iteration 7560, lr = 0.0023043
I0407 23:28:45.604382 32718 solver.cpp:218] Iteration 7572 (2.44194 iter/s, 4.91413s/12 iters), loss = 0.101887
I0407 23:28:45.604421 32718 solver.cpp:237]     Train net output #0: loss = 0.101887 (* 1 = 0.101887 loss)
I0407 23:28:45.604429 32718 sgd_solver.cpp:105] Iteration 7572, lr = 0.00229389
I0407 23:28:50.485589 32718 solver.cpp:218] Iteration 7584 (2.45844 iter/s, 4.88114s/12 iters), loss = 0.0507761
I0407 23:28:50.485625 32718 solver.cpp:237]     Train net output #0: loss = 0.0507762 (* 1 = 0.0507762 loss)
I0407 23:28:50.485631 32718 sgd_solver.cpp:105] Iteration 7584, lr = 0.00228351
I0407 23:28:51.110430 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:28:55.343086 32718 solver.cpp:218] Iteration 7596 (2.47044 iter/s, 4.85743s/12 iters), loss = 0.0418113
I0407 23:28:55.343225 32718 solver.cpp:237]     Train net output #0: loss = 0.0418113 (* 1 = 0.0418113 loss)
I0407 23:28:55.343235 32718 sgd_solver.cpp:105] Iteration 7596, lr = 0.00227316
I0407 23:29:00.306589 32718 solver.cpp:218] Iteration 7608 (2.41773 iter/s, 4.96334s/12 iters), loss = 0.0875035
I0407 23:29:00.306625 32718 solver.cpp:237]     Train net output #0: loss = 0.0875035 (* 1 = 0.0875035 loss)
I0407 23:29:00.306632 32718 sgd_solver.cpp:105] Iteration 7608, lr = 0.00226284
I0407 23:29:05.158555 32718 solver.cpp:218] Iteration 7620 (2.47326 iter/s, 4.8519s/12 iters), loss = 0.0716875
I0407 23:29:05.158597 32718 solver.cpp:237]     Train net output #0: loss = 0.0716875 (* 1 = 0.0716875 loss)
I0407 23:29:05.158605 32718 sgd_solver.cpp:105] Iteration 7620, lr = 0.00225256
I0407 23:29:07.459026 32718 blocking_queue.cpp:49] Waiting for data
I0407 23:29:09.986678 32718 solver.cpp:218] Iteration 7632 (2.48548 iter/s, 4.82805s/12 iters), loss = 0.201693
I0407 23:29:09.986719 32718 solver.cpp:237]     Train net output #0: loss = 0.201693 (* 1 = 0.201693 loss)
I0407 23:29:09.986728 32718 sgd_solver.cpp:105] Iteration 7632, lr = 0.00224231
I0407 23:29:14.917007 32718 solver.cpp:218] Iteration 7644 (2.43395 iter/s, 4.93026s/12 iters), loss = 0.0860038
I0407 23:29:14.917045 32718 solver.cpp:237]     Train net output #0: loss = 0.0860039 (* 1 = 0.0860039 loss)
I0407 23:29:14.917053 32718 sgd_solver.cpp:105] Iteration 7644, lr = 0.0022321
I0407 23:29:16.944692 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_7650.caffemodel
I0407 23:29:20.964367 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_7650.solverstate
I0407 23:29:23.341696 32718 solver.cpp:330] Iteration 7650, Testing net (#0)
I0407 23:29:23.341715 32718 net.cpp:676] Ignoring source layer train-data
I0407 23:29:24.812958 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:29:28.116366 32718 solver.cpp:397]     Test net output #0: accuracy = 0.518995
I0407 23:29:28.116559 32718 solver.cpp:397]     Test net output #1: loss = 2.8182 (* 1 = 2.8182 loss)
I0407 23:29:29.742660 32718 solver.cpp:218] Iteration 7656 (0.809413 iter/s, 14.8256s/12 iters), loss = 0.0483912
I0407 23:29:29.742700 32718 solver.cpp:237]     Train net output #0: loss = 0.0483913 (* 1 = 0.0483913 loss)
I0407 23:29:29.742708 32718 sgd_solver.cpp:105] Iteration 7656, lr = 0.00222191
I0407 23:29:34.696342 32718 solver.cpp:218] Iteration 7668 (2.42247 iter/s, 4.95362s/12 iters), loss = 0.0741889
I0407 23:29:34.696386 32718 solver.cpp:237]     Train net output #0: loss = 0.0741889 (* 1 = 0.0741889 loss)
I0407 23:29:34.696396 32718 sgd_solver.cpp:105] Iteration 7668, lr = 0.00221176
I0407 23:29:39.642192 32718 solver.cpp:218] Iteration 7680 (2.42631 iter/s, 4.94578s/12 iters), loss = 0.137655
I0407 23:29:39.642226 32718 solver.cpp:237]     Train net output #0: loss = 0.137655 (* 1 = 0.137655 loss)
I0407 23:29:39.642235 32718 sgd_solver.cpp:105] Iteration 7680, lr = 0.00220165
I0407 23:29:42.407531 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:29:44.579221 32718 solver.cpp:218] Iteration 7692 (2.43065 iter/s, 4.93696s/12 iters), loss = 0.0649138
I0407 23:29:44.579258 32718 solver.cpp:237]     Train net output #0: loss = 0.0649138 (* 1 = 0.0649138 loss)
I0407 23:29:44.579267 32718 sgd_solver.cpp:105] Iteration 7692, lr = 0.00219157
I0407 23:29:49.504514 32718 solver.cpp:218] Iteration 7704 (2.43644 iter/s, 4.92522s/12 iters), loss = 0.0865833
I0407 23:29:49.504551 32718 solver.cpp:237]     Train net output #0: loss = 0.0865833 (* 1 = 0.0865833 loss)
I0407 23:29:49.504559 32718 sgd_solver.cpp:105] Iteration 7704, lr = 0.00218152
I0407 23:29:54.412103 32718 solver.cpp:218] Iteration 7716 (2.44523 iter/s, 4.90752s/12 iters), loss = 0.124941
I0407 23:29:54.412144 32718 solver.cpp:237]     Train net output #0: loss = 0.124941 (* 1 = 0.124941 loss)
I0407 23:29:54.412151 32718 sgd_solver.cpp:105] Iteration 7716, lr = 0.0021715
I0407 23:29:59.310616 32718 solver.cpp:218] Iteration 7728 (2.44976 iter/s, 4.89844s/12 iters), loss = 0.0743926
I0407 23:29:59.310752 32718 solver.cpp:237]     Train net output #0: loss = 0.0743926 (* 1 = 0.0743926 loss)
I0407 23:29:59.310762 32718 sgd_solver.cpp:105] Iteration 7728, lr = 0.00216152
I0407 23:30:04.192813 32718 solver.cpp:218] Iteration 7740 (2.45799 iter/s, 4.88203s/12 iters), loss = 0.0718537
I0407 23:30:04.192855 32718 solver.cpp:237]     Train net output #0: loss = 0.0718537 (* 1 = 0.0718537 loss)
I0407 23:30:04.192863 32718 sgd_solver.cpp:105] Iteration 7740, lr = 0.00215157
I0407 23:30:08.699014 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_7752.caffemodel
I0407 23:30:11.778038 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_7752.solverstate
I0407 23:30:14.138590 32718 solver.cpp:330] Iteration 7752, Testing net (#0)
I0407 23:30:14.138609 32718 net.cpp:676] Ignoring source layer train-data
I0407 23:30:15.473364 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:30:18.589434 32718 solver.cpp:397]     Test net output #0: accuracy = 0.507353
I0407 23:30:18.589481 32718 solver.cpp:397]     Test net output #1: loss = 2.84532 (* 1 = 2.84532 loss)
I0407 23:30:18.686097 32718 solver.cpp:218] Iteration 7752 (0.827975 iter/s, 14.4932s/12 iters), loss = 0.113427
I0407 23:30:18.686144 32718 solver.cpp:237]     Train net output #0: loss = 0.113427 (* 1 = 0.113427 loss)
I0407 23:30:18.686154 32718 sgd_solver.cpp:105] Iteration 7752, lr = 0.00214165
I0407 23:30:22.859874 32718 solver.cpp:218] Iteration 7764 (2.87514 iter/s, 4.17371s/12 iters), loss = 0.0972985
I0407 23:30:22.859910 32718 solver.cpp:237]     Train net output #0: loss = 0.0972986 (* 1 = 0.0972986 loss)
I0407 23:30:22.859917 32718 sgd_solver.cpp:105] Iteration 7764, lr = 0.00213177
I0407 23:30:27.848109 32718 solver.cpp:218] Iteration 7776 (2.40569 iter/s, 4.98817s/12 iters), loss = 0.0598435
I0407 23:30:27.848148 32718 solver.cpp:237]     Train net output #0: loss = 0.0598435 (* 1 = 0.0598435 loss)
I0407 23:30:27.848156 32718 sgd_solver.cpp:105] Iteration 7776, lr = 0.00212192
I0407 23:30:32.792856 32718 solver.cpp:218] Iteration 7788 (2.42685 iter/s, 4.94468s/12 iters), loss = 0.10543
I0407 23:30:32.793018 32718 solver.cpp:237]     Train net output #0: loss = 0.10543 (* 1 = 0.10543 loss)
I0407 23:30:32.793028 32718 sgd_solver.cpp:105] Iteration 7788, lr = 0.0021121
I0407 23:30:32.799363 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:30:37.736706 32718 solver.cpp:218] Iteration 7800 (2.42735 iter/s, 4.94366s/12 iters), loss = 0.0466201
I0407 23:30:37.736747 32718 solver.cpp:237]     Train net output #0: loss = 0.0466202 (* 1 = 0.0466202 loss)
I0407 23:30:37.736757 32718 sgd_solver.cpp:105] Iteration 7800, lr = 0.00210232
I0407 23:30:42.737802 32718 solver.cpp:218] Iteration 7812 (2.39951 iter/s, 5.00103s/12 iters), loss = 0.0811311
I0407 23:30:42.737838 32718 solver.cpp:237]     Train net output #0: loss = 0.0811311 (* 1 = 0.0811311 loss)
I0407 23:30:42.737845 32718 sgd_solver.cpp:105] Iteration 7812, lr = 0.00209257
I0407 23:30:47.643776 32718 solver.cpp:218] Iteration 7824 (2.44603 iter/s, 4.90591s/12 iters), loss = 0.0237032
I0407 23:30:47.643819 32718 solver.cpp:237]     Train net output #0: loss = 0.0237032 (* 1 = 0.0237032 loss)
I0407 23:30:47.643827 32718 sgd_solver.cpp:105] Iteration 7824, lr = 0.00208285
I0407 23:30:52.620106 32718 solver.cpp:218] Iteration 7836 (2.41145 iter/s, 4.97626s/12 iters), loss = 0.094964
I0407 23:30:52.620146 32718 solver.cpp:237]     Train net output #0: loss = 0.094964 (* 1 = 0.094964 loss)
I0407 23:30:52.620154 32718 sgd_solver.cpp:105] Iteration 7836, lr = 0.00207317
I0407 23:30:57.579599 32718 solver.cpp:218] Iteration 7848 (2.41964 iter/s, 4.95942s/12 iters), loss = 0.0558626
I0407 23:30:57.579638 32718 solver.cpp:237]     Train net output #0: loss = 0.0558626 (* 1 = 0.0558626 loss)
I0407 23:30:57.579645 32718 sgd_solver.cpp:105] Iteration 7848, lr = 0.00206352
I0407 23:30:59.577098 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_7854.caffemodel
I0407 23:31:02.931596 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_7854.solverstate
I0407 23:31:05.290840 32718 solver.cpp:330] Iteration 7854, Testing net (#0)
I0407 23:31:05.290863 32718 net.cpp:676] Ignoring source layer train-data
I0407 23:31:06.685976 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:31:10.090133 32718 solver.cpp:397]     Test net output #0: accuracy = 0.505515
I0407 23:31:10.090179 32718 solver.cpp:397]     Test net output #1: loss = 2.79477 (* 1 = 2.79477 loss)
I0407 23:31:11.877815 32718 solver.cpp:218] Iteration 7860 (0.839271 iter/s, 14.2981s/12 iters), loss = 0.0776179
I0407 23:31:11.877853 32718 solver.cpp:237]     Train net output #0: loss = 0.0776179 (* 1 = 0.0776179 loss)
I0407 23:31:11.877861 32718 sgd_solver.cpp:105] Iteration 7860, lr = 0.0020539
I0407 23:31:16.742660 32718 solver.cpp:218] Iteration 7872 (2.46671 iter/s, 4.86478s/12 iters), loss = 0.0271532
I0407 23:31:16.742699 32718 solver.cpp:237]     Train net output #0: loss = 0.0271532 (* 1 = 0.0271532 loss)
I0407 23:31:16.742707 32718 sgd_solver.cpp:105] Iteration 7872, lr = 0.00204432
I0407 23:31:21.704392 32718 solver.cpp:218] Iteration 7884 (2.41854 iter/s, 4.96166s/12 iters), loss = 0.1019
I0407 23:31:21.704432 32718 solver.cpp:237]     Train net output #0: loss = 0.1019 (* 1 = 0.1019 loss)
I0407 23:31:21.704440 32718 sgd_solver.cpp:105] Iteration 7884, lr = 0.00203477
I0407 23:31:23.803752 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:31:26.600383 32718 solver.cpp:218] Iteration 7896 (2.45102 iter/s, 4.89592s/12 iters), loss = 0.00943052
I0407 23:31:26.600423 32718 solver.cpp:237]     Train net output #0: loss = 0.00943052 (* 1 = 0.00943052 loss)
I0407 23:31:26.600431 32718 sgd_solver.cpp:105] Iteration 7896, lr = 0.00202525
I0407 23:31:31.556986 32718 solver.cpp:218] Iteration 7908 (2.42105 iter/s, 4.95654s/12 iters), loss = 0.11285
I0407 23:31:31.557024 32718 solver.cpp:237]     Train net output #0: loss = 0.11285 (* 1 = 0.11285 loss)
I0407 23:31:31.557031 32718 sgd_solver.cpp:105] Iteration 7908, lr = 0.00201576
I0407 23:31:36.473986 32718 solver.cpp:218] Iteration 7920 (2.44055 iter/s, 4.91693s/12 iters), loss = 0.0438029
I0407 23:31:36.474140 32718 solver.cpp:237]     Train net output #0: loss = 0.0438029 (* 1 = 0.0438029 loss)
I0407 23:31:36.474149 32718 sgd_solver.cpp:105] Iteration 7920, lr = 0.00200631
I0407 23:31:41.404525 32718 solver.cpp:218] Iteration 7932 (2.4339 iter/s, 4.93036s/12 iters), loss = 0.147764
I0407 23:31:41.404567 32718 solver.cpp:237]     Train net output #0: loss = 0.147764 (* 1 = 0.147764 loss)
I0407 23:31:41.404577 32718 sgd_solver.cpp:105] Iteration 7932, lr = 0.0019969
I0407 23:31:46.347259 32718 solver.cpp:218] Iteration 7944 (2.42784 iter/s, 4.94266s/12 iters), loss = 0.0208504
I0407 23:31:46.347303 32718 solver.cpp:237]     Train net output #0: loss = 0.0208504 (* 1 = 0.0208504 loss)
I0407 23:31:46.347311 32718 sgd_solver.cpp:105] Iteration 7944, lr = 0.00198751
I0407 23:31:50.843536 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_7956.caffemodel
I0407 23:31:55.410605 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_7956.solverstate
I0407 23:31:57.819490 32718 solver.cpp:330] Iteration 7956, Testing net (#0)
I0407 23:31:57.819509 32718 net.cpp:676] Ignoring source layer train-data
I0407 23:31:59.169695 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:32:02.617864 32718 solver.cpp:397]     Test net output #0: accuracy = 0.507353
I0407 23:32:02.617910 32718 solver.cpp:397]     Test net output #1: loss = 2.84066 (* 1 = 2.84066 loss)
I0407 23:32:02.714484 32718 solver.cpp:218] Iteration 7956 (0.733177 iter/s, 16.3671s/12 iters), loss = 0.0710412
I0407 23:32:02.714529 32718 solver.cpp:237]     Train net output #0: loss = 0.0710412 (* 1 = 0.0710412 loss)
I0407 23:32:02.714537 32718 sgd_solver.cpp:105] Iteration 7956, lr = 0.00197816
I0407 23:32:06.774744 32718 solver.cpp:218] Iteration 7968 (2.95553 iter/s, 4.06019s/12 iters), loss = 0.0304347
I0407 23:32:06.774880 32718 solver.cpp:237]     Train net output #0: loss = 0.0304347 (* 1 = 0.0304347 loss)
I0407 23:32:06.774895 32718 sgd_solver.cpp:105] Iteration 7968, lr = 0.00196884
I0407 23:32:11.713455 32718 solver.cpp:218] Iteration 7980 (2.42986 iter/s, 4.93855s/12 iters), loss = 0.107225
I0407 23:32:11.713493 32718 solver.cpp:237]     Train net output #0: loss = 0.107225 (* 1 = 0.107225 loss)
I0407 23:32:11.713501 32718 sgd_solver.cpp:105] Iteration 7980, lr = 0.00195956
I0407 23:32:15.949095 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:32:16.595329 32718 solver.cpp:218] Iteration 7992 (2.45811 iter/s, 4.88181s/12 iters), loss = 0.102656
I0407 23:32:16.595377 32718 solver.cpp:237]     Train net output #0: loss = 0.102656 (* 1 = 0.102656 loss)
I0407 23:32:16.595389 32718 sgd_solver.cpp:105] Iteration 7992, lr = 0.00195031
I0407 23:32:21.561530 32718 solver.cpp:218] Iteration 8004 (2.41637 iter/s, 4.96613s/12 iters), loss = 0.0455987
I0407 23:32:21.561571 32718 solver.cpp:237]     Train net output #0: loss = 0.0455987 (* 1 = 0.0455987 loss)
I0407 23:32:21.561579 32718 sgd_solver.cpp:105] Iteration 8004, lr = 0.00194109
I0407 23:32:26.518941 32718 solver.cpp:218] Iteration 8016 (2.42065 iter/s, 4.95735s/12 iters), loss = 0.114773
I0407 23:32:26.518973 32718 solver.cpp:237]     Train net output #0: loss = 0.114773 (* 1 = 0.114773 loss)
I0407 23:32:26.518980 32718 sgd_solver.cpp:105] Iteration 8016, lr = 0.0019319
I0407 23:32:31.392602 32718 solver.cpp:218] Iteration 8028 (2.46224 iter/s, 4.8736s/12 iters), loss = 0.0809719
I0407 23:32:31.392642 32718 solver.cpp:237]     Train net output #0: loss = 0.0809719 (* 1 = 0.0809719 loss)
I0407 23:32:31.392649 32718 sgd_solver.cpp:105] Iteration 8028, lr = 0.00192275
I0407 23:32:36.333784 32718 solver.cpp:218] Iteration 8040 (2.4286 iter/s, 4.94112s/12 iters), loss = 0.0703065
I0407 23:32:36.333819 32718 solver.cpp:237]     Train net output #0: loss = 0.0703065 (* 1 = 0.0703065 loss)
I0407 23:32:36.333827 32718 sgd_solver.cpp:105] Iteration 8040, lr = 0.00191363
I0407 23:32:41.278009 32718 solver.cpp:218] Iteration 8052 (2.4271 iter/s, 4.94417s/12 iters), loss = 0.0635816
I0407 23:32:41.278146 32718 solver.cpp:237]     Train net output #0: loss = 0.0635816 (* 1 = 0.0635816 loss)
I0407 23:32:41.278154 32718 sgd_solver.cpp:105] Iteration 8052, lr = 0.00190455
I0407 23:32:43.294237 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_8058.caffemodel
I0407 23:32:46.367888 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_8058.solverstate
I0407 23:32:48.725754 32718 solver.cpp:330] Iteration 8058, Testing net (#0)
I0407 23:32:48.725771 32718 net.cpp:676] Ignoring source layer train-data
I0407 23:32:50.009716 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:32:53.231387 32718 solver.cpp:397]     Test net output #0: accuracy = 0.512868
I0407 23:32:53.231428 32718 solver.cpp:397]     Test net output #1: loss = 2.76861 (* 1 = 2.76861 loss)
I0407 23:32:55.016433 32718 solver.cpp:218] Iteration 8064 (0.873475 iter/s, 13.7382s/12 iters), loss = 0.112795
I0407 23:32:55.016476 32718 solver.cpp:237]     Train net output #0: loss = 0.112795 (* 1 = 0.112795 loss)
I0407 23:32:55.016484 32718 sgd_solver.cpp:105] Iteration 8064, lr = 0.00189549
I0407 23:32:59.971485 32718 solver.cpp:218] Iteration 8076 (2.42181 iter/s, 4.95497s/12 iters), loss = 0.0679882
I0407 23:32:59.971547 32718 solver.cpp:237]     Train net output #0: loss = 0.0679882 (* 1 = 0.0679882 loss)
I0407 23:32:59.971560 32718 sgd_solver.cpp:105] Iteration 8076, lr = 0.00188647
I0407 23:33:05.006857 32718 solver.cpp:218] Iteration 8088 (2.38318 iter/s, 5.03528s/12 iters), loss = 0.0525046
I0407 23:33:05.006894 32718 solver.cpp:237]     Train net output #0: loss = 0.0525046 (* 1 = 0.0525046 loss)
I0407 23:33:05.006903 32718 sgd_solver.cpp:105] Iteration 8088, lr = 0.00187749
I0407 23:33:06.384248 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:33:09.933284 32718 solver.cpp:218] Iteration 8100 (2.43588 iter/s, 4.92636s/12 iters), loss = 0.135005
I0407 23:33:09.933323 32718 solver.cpp:237]     Train net output #0: loss = 0.135005 (* 1 = 0.135005 loss)
I0407 23:33:09.933331 32718 sgd_solver.cpp:105] Iteration 8100, lr = 0.00186853
I0407 23:33:14.877136 32718 solver.cpp:218] Iteration 8112 (2.42729 iter/s, 4.94378s/12 iters), loss = 0.0279139
I0407 23:33:14.877261 32718 solver.cpp:237]     Train net output #0: loss = 0.0279139 (* 1 = 0.0279139 loss)
I0407 23:33:14.877271 32718 sgd_solver.cpp:105] Iteration 8112, lr = 0.00185961
I0407 23:33:19.743958 32718 solver.cpp:218] Iteration 8124 (2.46575 iter/s, 4.86667s/12 iters), loss = 0.0432566
I0407 23:33:19.743996 32718 solver.cpp:237]     Train net output #0: loss = 0.0432566 (* 1 = 0.0432566 loss)
I0407 23:33:19.744004 32718 sgd_solver.cpp:105] Iteration 8124, lr = 0.00185072
I0407 23:33:24.706105 32718 solver.cpp:218] Iteration 8136 (2.41834 iter/s, 4.96207s/12 iters), loss = 0.150013
I0407 23:33:24.706148 32718 solver.cpp:237]     Train net output #0: loss = 0.150013 (* 1 = 0.150013 loss)
I0407 23:33:24.706157 32718 sgd_solver.cpp:105] Iteration 8136, lr = 0.00184187
I0407 23:33:29.659191 32718 solver.cpp:218] Iteration 8148 (2.42277 iter/s, 4.95301s/12 iters), loss = 0.0107595
I0407 23:33:29.659240 32718 solver.cpp:237]     Train net output #0: loss = 0.0107596 (* 1 = 0.0107596 loss)
I0407 23:33:29.659248 32718 sgd_solver.cpp:105] Iteration 8148, lr = 0.00183304
I0407 23:33:34.139111 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_8160.caffemodel
I0407 23:33:37.256503 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_8160.solverstate
I0407 23:33:39.619148 32718 solver.cpp:330] Iteration 8160, Testing net (#0)
I0407 23:33:39.619166 32718 net.cpp:676] Ignoring source layer train-data
I0407 23:33:40.817606 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:33:44.069480 32718 solver.cpp:397]     Test net output #0: accuracy = 0.51348
I0407 23:33:44.069511 32718 solver.cpp:397]     Test net output #1: loss = 2.81277 (* 1 = 2.81277 loss)
I0407 23:33:44.165827 32718 solver.cpp:218] Iteration 8160 (0.827214 iter/s, 14.5065s/12 iters), loss = 0.0985616
I0407 23:33:44.165876 32718 solver.cpp:237]     Train net output #0: loss = 0.0985617 (* 1 = 0.0985617 loss)
I0407 23:33:44.165884 32718 sgd_solver.cpp:105] Iteration 8160, lr = 0.00182425
I0407 23:33:48.278046 32718 solver.cpp:218] Iteration 8172 (2.91819 iter/s, 4.11214s/12 iters), loss = 0.0317634
I0407 23:33:48.278203 32718 solver.cpp:237]     Train net output #0: loss = 0.0317634 (* 1 = 0.0317634 loss)
I0407 23:33:48.278213 32718 sgd_solver.cpp:105] Iteration 8172, lr = 0.0018155
I0407 23:33:53.245088 32718 solver.cpp:218] Iteration 8184 (2.41601 iter/s, 4.96686s/12 iters), loss = 0.104974
I0407 23:33:53.245128 32718 solver.cpp:237]     Train net output #0: loss = 0.104974 (* 1 = 0.104974 loss)
I0407 23:33:53.245136 32718 sgd_solver.cpp:105] Iteration 8184, lr = 0.00180677
I0407 23:33:56.680718 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:33:58.094657 32718 solver.cpp:218] Iteration 8196 (2.47448 iter/s, 4.84949s/12 iters), loss = 0.0346849
I0407 23:33:58.094703 32718 solver.cpp:237]     Train net output #0: loss = 0.0346849 (* 1 = 0.0346849 loss)
I0407 23:33:58.094712 32718 sgd_solver.cpp:105] Iteration 8196, lr = 0.00179808
I0407 23:34:03.068048 32718 solver.cpp:218] Iteration 8208 (2.41288 iter/s, 4.97331s/12 iters), loss = 0.104151
I0407 23:34:03.068090 32718 solver.cpp:237]     Train net output #0: loss = 0.104151 (* 1 = 0.104151 loss)
I0407 23:34:03.068099 32718 sgd_solver.cpp:105] Iteration 8208, lr = 0.00178942
I0407 23:34:08.013190 32718 solver.cpp:218] Iteration 8220 (2.42666 iter/s, 4.94507s/12 iters), loss = 0.014001
I0407 23:34:08.013233 32718 solver.cpp:237]     Train net output #0: loss = 0.014001 (* 1 = 0.014001 loss)
I0407 23:34:08.013242 32718 sgd_solver.cpp:105] Iteration 8220, lr = 0.0017808
I0407 23:34:12.960093 32718 solver.cpp:218] Iteration 8232 (2.4258 iter/s, 4.94683s/12 iters), loss = 0.0792821
I0407 23:34:12.960137 32718 solver.cpp:237]     Train net output #0: loss = 0.0792821 (* 1 = 0.0792821 loss)
I0407 23:34:12.960146 32718 sgd_solver.cpp:105] Iteration 8232, lr = 0.0017722
I0407 23:34:17.891216 32718 solver.cpp:218] Iteration 8244 (2.43356 iter/s, 4.93105s/12 iters), loss = 0.0224014
I0407 23:34:17.891261 32718 solver.cpp:237]     Train net output #0: loss = 0.0224014 (* 1 = 0.0224014 loss)
I0407 23:34:17.891269 32718 sgd_solver.cpp:105] Iteration 8244, lr = 0.00176364
I0407 23:34:22.857656 32718 solver.cpp:218] Iteration 8256 (2.41626 iter/s, 4.96636s/12 iters), loss = 0.0610395
I0407 23:34:22.857806 32718 solver.cpp:237]     Train net output #0: loss = 0.0610395 (* 1 = 0.0610395 loss)
I0407 23:34:22.857815 32718 sgd_solver.cpp:105] Iteration 8256, lr = 0.00175511
I0407 23:34:24.867961 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_8262.caffemodel
I0407 23:34:27.936203 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_8262.solverstate
I0407 23:34:30.311393 32718 solver.cpp:330] Iteration 8262, Testing net (#0)
I0407 23:34:30.311409 32718 net.cpp:676] Ignoring source layer train-data
I0407 23:34:31.484871 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:34:34.804189 32718 solver.cpp:397]     Test net output #0: accuracy = 0.523284
I0407 23:34:34.804234 32718 solver.cpp:397]     Test net output #1: loss = 2.82128 (* 1 = 2.82128 loss)
I0407 23:34:36.605682 32718 solver.cpp:218] Iteration 8268 (0.872865 iter/s, 13.7478s/12 iters), loss = 0.0567808
I0407 23:34:36.605718 32718 solver.cpp:237]     Train net output #0: loss = 0.0567808 (* 1 = 0.0567808 loss)
I0407 23:34:36.605726 32718 sgd_solver.cpp:105] Iteration 8268, lr = 0.00174662
I0407 23:34:41.596124 32718 solver.cpp:218] Iteration 8280 (2.40463 iter/s, 4.99038s/12 iters), loss = 0.0803485
I0407 23:34:41.596170 32718 solver.cpp:237]     Train net output #0: loss = 0.0803485 (* 1 = 0.0803485 loss)
I0407 23:34:41.596179 32718 sgd_solver.cpp:105] Iteration 8280, lr = 0.00173816
I0407 23:34:46.496624 32718 solver.cpp:218] Iteration 8292 (2.44877 iter/s, 4.90042s/12 iters), loss = 0.148516
I0407 23:34:46.496675 32718 solver.cpp:237]     Train net output #0: loss = 0.148516 (* 1 = 0.148516 loss)
I0407 23:34:46.496685 32718 sgd_solver.cpp:105] Iteration 8292, lr = 0.00172972
I0407 23:34:47.093199 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:34:51.399471 32718 solver.cpp:218] Iteration 8304 (2.4476 iter/s, 4.90277s/12 iters), loss = 0.0943353
I0407 23:34:51.399503 32718 solver.cpp:237]     Train net output #0: loss = 0.0943353 (* 1 = 0.0943353 loss)
I0407 23:34:51.399513 32718 sgd_solver.cpp:105] Iteration 8304, lr = 0.00172133
I0407 23:34:54.233000 32718 blocking_queue.cpp:49] Waiting for data
I0407 23:34:56.369274 32718 solver.cpp:218] Iteration 8316 (2.41461 iter/s, 4.96974s/12 iters), loss = 0.0180675
I0407 23:34:56.369313 32718 solver.cpp:237]     Train net output #0: loss = 0.0180675 (* 1 = 0.0180675 loss)
I0407 23:34:56.369321 32718 sgd_solver.cpp:105] Iteration 8316, lr = 0.00171296
I0407 23:35:01.255679 32718 solver.cpp:218] Iteration 8328 (2.45583 iter/s, 4.88633s/12 iters), loss = 0.0904267
I0407 23:35:01.255719 32718 solver.cpp:237]     Train net output #0: loss = 0.0904268 (* 1 = 0.0904268 loss)
I0407 23:35:01.255728 32718 sgd_solver.cpp:105] Iteration 8328, lr = 0.00170462
I0407 23:35:06.194988 32718 solver.cpp:218] Iteration 8340 (2.42952 iter/s, 4.93924s/12 iters), loss = 0.0874588
I0407 23:35:06.195042 32718 solver.cpp:237]     Train net output #0: loss = 0.0874588 (* 1 = 0.0874588 loss)
I0407 23:35:06.195055 32718 sgd_solver.cpp:105] Iteration 8340, lr = 0.00169632
I0407 23:35:11.162539 32718 solver.cpp:218] Iteration 8352 (2.41572 iter/s, 4.96747s/12 iters), loss = 0.0708697
I0407 23:35:11.162582 32718 solver.cpp:237]     Train net output #0: loss = 0.0708697 (* 1 = 0.0708697 loss)
I0407 23:35:11.162591 32718 sgd_solver.cpp:105] Iteration 8352, lr = 0.00168805
I0407 23:35:15.676337 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_8364.caffemodel
I0407 23:35:18.739502 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_8364.solverstate
I0407 23:35:21.102090 32718 solver.cpp:330] Iteration 8364, Testing net (#0)
I0407 23:35:21.102106 32718 net.cpp:676] Ignoring source layer train-data
I0407 23:35:22.275034 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:35:25.708307 32718 solver.cpp:397]     Test net output #0: accuracy = 0.515319
I0407 23:35:25.708483 32718 solver.cpp:397]     Test net output #1: loss = 2.75555 (* 1 = 2.75555 loss)
I0407 23:35:25.805297 32718 solver.cpp:218] Iteration 8364 (0.819524 iter/s, 14.6427s/12 iters), loss = 0.033332
I0407 23:35:25.805342 32718 solver.cpp:237]     Train net output #0: loss = 0.033332 (* 1 = 0.033332 loss)
I0407 23:35:25.805351 32718 sgd_solver.cpp:105] Iteration 8364, lr = 0.00167982
I0407 23:35:29.895164 32718 solver.cpp:218] Iteration 8376 (2.93413 iter/s, 4.08979s/12 iters), loss = 0.132225
I0407 23:35:29.895212 32718 solver.cpp:237]     Train net output #0: loss = 0.132225 (* 1 = 0.132225 loss)
I0407 23:35:29.895221 32718 sgd_solver.cpp:105] Iteration 8376, lr = 0.00167161
I0407 23:35:34.868137 32718 solver.cpp:218] Iteration 8388 (2.41308 iter/s, 4.97289s/12 iters), loss = 0.0358579
I0407 23:35:34.868181 32718 solver.cpp:237]     Train net output #0: loss = 0.035858 (* 1 = 0.035858 loss)
I0407 23:35:34.868189 32718 sgd_solver.cpp:105] Iteration 8388, lr = 0.00166344
I0407 23:35:37.651650 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:35:39.819361 32718 solver.cpp:218] Iteration 8400 (2.42368 iter/s, 4.95115s/12 iters), loss = 0.0718623
I0407 23:35:39.819406 32718 solver.cpp:237]     Train net output #0: loss = 0.0718623 (* 1 = 0.0718623 loss)
I0407 23:35:39.819414 32718 sgd_solver.cpp:105] Iteration 8400, lr = 0.0016553
I0407 23:35:44.782763 32718 solver.cpp:218] Iteration 8412 (2.41773 iter/s, 4.96333s/12 iters), loss = 0.0449645
I0407 23:35:44.782799 32718 solver.cpp:237]     Train net output #0: loss = 0.0449645 (* 1 = 0.0449645 loss)
I0407 23:35:44.782806 32718 sgd_solver.cpp:105] Iteration 8412, lr = 0.00164719
I0407 23:35:49.769816 32718 solver.cpp:218] Iteration 8424 (2.40626 iter/s, 4.98699s/12 iters), loss = 0.0347374
I0407 23:35:49.769855 32718 solver.cpp:237]     Train net output #0: loss = 0.0347374 (* 1 = 0.0347374 loss)
I0407 23:35:49.769861 32718 sgd_solver.cpp:105] Iteration 8424, lr = 0.00163911
I0407 23:35:54.746101 32718 solver.cpp:218] Iteration 8436 (2.41147 iter/s, 4.97622s/12 iters), loss = 0.133465
I0407 23:35:54.746140 32718 solver.cpp:237]     Train net output #0: loss = 0.133465 (* 1 = 0.133465 loss)
I0407 23:35:54.746146 32718 sgd_solver.cpp:105] Iteration 8436, lr = 0.00163106
I0407 23:35:59.681851 32718 solver.cpp:218] Iteration 8448 (2.43127 iter/s, 4.93569s/12 iters), loss = 0.0754896
I0407 23:35:59.681959 32718 solver.cpp:237]     Train net output #0: loss = 0.0754896 (* 1 = 0.0754896 loss)
I0407 23:35:59.681967 32718 sgd_solver.cpp:105] Iteration 8448, lr = 0.00162305
I0407 23:36:04.713295 32718 solver.cpp:218] Iteration 8460 (2.38506 iter/s, 5.03131s/12 iters), loss = 0.0994054
I0407 23:36:04.713335 32718 solver.cpp:237]     Train net output #0: loss = 0.0994054 (* 1 = 0.0994054 loss)
I0407 23:36:04.713342 32718 sgd_solver.cpp:105] Iteration 8460, lr = 0.00161507
I0407 23:36:06.725461 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_8466.caffemodel
I0407 23:36:09.872121 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_8466.solverstate
I0407 23:36:12.748955 32718 solver.cpp:330] Iteration 8466, Testing net (#0)
I0407 23:36:12.748972 32718 net.cpp:676] Ignoring source layer train-data
I0407 23:36:13.876993 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:36:17.248435 32718 solver.cpp:397]     Test net output #0: accuracy = 0.508578
I0407 23:36:17.248466 32718 solver.cpp:397]     Test net output #1: loss = 2.81187 (* 1 = 2.81187 loss)
I0407 23:36:19.030795 32718 solver.cpp:218] Iteration 8472 (0.83814 iter/s, 14.3174s/12 iters), loss = 0.0183983
I0407 23:36:19.030833 32718 solver.cpp:237]     Train net output #0: loss = 0.0183983 (* 1 = 0.0183983 loss)
I0407 23:36:19.030840 32718 sgd_solver.cpp:105] Iteration 8472, lr = 0.00160712
I0407 23:36:24.000020 32718 solver.cpp:218] Iteration 8484 (2.4149 iter/s, 4.96916s/12 iters), loss = 0.0611635
I0407 23:36:24.000059 32718 solver.cpp:237]     Train net output #0: loss = 0.0611635 (* 1 = 0.0611635 loss)
I0407 23:36:24.000067 32718 sgd_solver.cpp:105] Iteration 8484, lr = 0.0015992
I0407 23:36:28.946864 32718 solver.cpp:218] Iteration 8496 (2.42582 iter/s, 4.94677s/12 iters), loss = 0.0346841
I0407 23:36:28.946900 32718 solver.cpp:237]     Train net output #0: loss = 0.0346841 (* 1 = 0.0346841 loss)
I0407 23:36:28.946908 32718 sgd_solver.cpp:105] Iteration 8496, lr = 0.00159131
I0407 23:36:28.983279 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:36:33.909817 32718 solver.cpp:218] Iteration 8508 (2.41795 iter/s, 4.96289s/12 iters), loss = 0.0663393
I0407 23:36:33.909958 32718 solver.cpp:237]     Train net output #0: loss = 0.0663393 (* 1 = 0.0663393 loss)
I0407 23:36:33.909968 32718 sgd_solver.cpp:105] Iteration 8508, lr = 0.00158346
I0407 23:36:38.954478 32718 solver.cpp:218] Iteration 8520 (2.37883 iter/s, 5.04449s/12 iters), loss = 0.0824151
I0407 23:36:38.954519 32718 solver.cpp:237]     Train net output #0: loss = 0.0824151 (* 1 = 0.0824151 loss)
I0407 23:36:38.954526 32718 sgd_solver.cpp:105] Iteration 8520, lr = 0.00157563
I0407 23:36:43.931820 32718 solver.cpp:218] Iteration 8532 (2.41096 iter/s, 4.97727s/12 iters), loss = 0.0836602
I0407 23:36:43.931864 32718 solver.cpp:237]     Train net output #0: loss = 0.0836602 (* 1 = 0.0836602 loss)
I0407 23:36:43.931872 32718 sgd_solver.cpp:105] Iteration 8532, lr = 0.00156784
I0407 23:36:48.911784 32718 solver.cpp:218] Iteration 8544 (2.40969 iter/s, 4.97989s/12 iters), loss = 0.0315724
I0407 23:36:48.911825 32718 solver.cpp:237]     Train net output #0: loss = 0.0315724 (* 1 = 0.0315724 loss)
I0407 23:36:48.911834 32718 sgd_solver.cpp:105] Iteration 8544, lr = 0.00156008
I0407 23:36:53.832823 32718 solver.cpp:218] Iteration 8556 (2.43855 iter/s, 4.92096s/12 iters), loss = 0.0556315
I0407 23:36:53.832868 32718 solver.cpp:237]     Train net output #0: loss = 0.0556316 (* 1 = 0.0556316 loss)
I0407 23:36:53.832876 32718 sgd_solver.cpp:105] Iteration 8556, lr = 0.00155235
I0407 23:36:58.277418 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_8568.caffemodel
I0407 23:37:01.352035 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_8568.solverstate
I0407 23:37:03.773746 32718 solver.cpp:330] Iteration 8568, Testing net (#0)
I0407 23:37:03.773763 32718 net.cpp:676] Ignoring source layer train-data
I0407 23:37:04.858407 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:37:08.354027 32718 solver.cpp:397]     Test net output #0: accuracy = 0.516544
I0407 23:37:08.354074 32718 solver.cpp:397]     Test net output #1: loss = 2.79796 (* 1 = 2.79796 loss)
I0407 23:37:08.450420 32718 solver.cpp:218] Iteration 8568 (0.820934 iter/s, 14.6175s/12 iters), loss = 0.0609294
I0407 23:37:08.450461 32718 solver.cpp:237]     Train net output #0: loss = 0.0609294 (* 1 = 0.0609294 loss)
I0407 23:37:08.450469 32718 sgd_solver.cpp:105] Iteration 8568, lr = 0.00154465
I0407 23:37:12.585425 32718 solver.cpp:218] Iteration 8580 (2.9021 iter/s, 4.13494s/12 iters), loss = 0.058905
I0407 23:37:12.585465 32718 solver.cpp:237]     Train net output #0: loss = 0.058905 (* 1 = 0.058905 loss)
I0407 23:37:12.585474 32718 sgd_solver.cpp:105] Iteration 8580, lr = 0.00153699
I0407 23:37:17.566839 32718 solver.cpp:218] Iteration 8592 (2.40901 iter/s, 4.9813s/12 iters), loss = 0.0319104
I0407 23:37:17.566877 32718 solver.cpp:237]     Train net output #0: loss = 0.0319104 (* 1 = 0.0319104 loss)
I0407 23:37:17.566885 32718 sgd_solver.cpp:105] Iteration 8592, lr = 0.00152935
I0407 23:37:19.736083 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:37:22.504155 32718 solver.cpp:218] Iteration 8604 (2.4305 iter/s, 4.93725s/12 iters), loss = 0.0561022
I0407 23:37:22.504197 32718 solver.cpp:237]     Train net output #0: loss = 0.0561022 (* 1 = 0.0561022 loss)
I0407 23:37:22.504205 32718 sgd_solver.cpp:105] Iteration 8604, lr = 0.00152174
I0407 23:37:27.429584 32718 solver.cpp:218] Iteration 8616 (2.43637 iter/s, 4.92536s/12 iters), loss = 0.0601061
I0407 23:37:27.429625 32718 solver.cpp:237]     Train net output #0: loss = 0.0601061 (* 1 = 0.0601061 loss)
I0407 23:37:27.429633 32718 sgd_solver.cpp:105] Iteration 8616, lr = 0.00151417
I0407 23:37:32.373746 32718 solver.cpp:218] Iteration 8628 (2.42714 iter/s, 4.94409s/12 iters), loss = 0.0105092
I0407 23:37:32.373793 32718 solver.cpp:237]     Train net output #0: loss = 0.0105092 (* 1 = 0.0105092 loss)
I0407 23:37:32.373803 32718 sgd_solver.cpp:105] Iteration 8628, lr = 0.00150663
I0407 23:37:37.278228 32718 solver.cpp:218] Iteration 8640 (2.44678 iter/s, 4.90441s/12 iters), loss = 0.0920459
I0407 23:37:37.278383 32718 solver.cpp:237]     Train net output #0: loss = 0.0920459 (* 1 = 0.0920459 loss)
I0407 23:37:37.278393 32718 sgd_solver.cpp:105] Iteration 8640, lr = 0.00149912
I0407 23:37:42.233026 32718 solver.cpp:218] Iteration 8652 (2.42198 iter/s, 4.95462s/12 iters), loss = 0.0173894
I0407 23:37:42.233060 32718 solver.cpp:237]     Train net output #0: loss = 0.0173894 (* 1 = 0.0173894 loss)
I0407 23:37:42.233069 32718 sgd_solver.cpp:105] Iteration 8652, lr = 0.00149164
I0407 23:37:47.166700 32718 solver.cpp:218] Iteration 8664 (2.4323 iter/s, 4.93361s/12 iters), loss = 0.0192104
I0407 23:37:47.166736 32718 solver.cpp:237]     Train net output #0: loss = 0.0192104 (* 1 = 0.0192104 loss)
I0407 23:37:47.166743 32718 sgd_solver.cpp:105] Iteration 8664, lr = 0.00148419
I0407 23:37:49.194670 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_8670.caffemodel
I0407 23:37:52.292817 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_8670.solverstate
I0407 23:37:54.663678 32718 solver.cpp:330] Iteration 8670, Testing net (#0)
I0407 23:37:54.663697 32718 net.cpp:676] Ignoring source layer train-data
I0407 23:37:55.728580 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:37:59.422531 32718 solver.cpp:397]     Test net output #0: accuracy = 0.518995
I0407 23:37:59.422576 32718 solver.cpp:397]     Test net output #1: loss = 2.79232 (* 1 = 2.79232 loss)
I0407 23:38:01.220362 32718 solver.cpp:218] Iteration 8676 (0.853875 iter/s, 14.0536s/12 iters), loss = 0.0465351
I0407 23:38:01.220402 32718 solver.cpp:237]     Train net output #0: loss = 0.0465352 (* 1 = 0.0465352 loss)
I0407 23:38:01.220410 32718 sgd_solver.cpp:105] Iteration 8676, lr = 0.00147677
I0407 23:38:06.184197 32718 solver.cpp:218] Iteration 8688 (2.41752 iter/s, 4.96377s/12 iters), loss = 0.0372439
I0407 23:38:06.184238 32718 solver.cpp:237]     Train net output #0: loss = 0.0372439 (* 1 = 0.0372439 loss)
I0407 23:38:06.184247 32718 sgd_solver.cpp:105] Iteration 8688, lr = 0.00146938
I0407 23:38:10.449720 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:38:11.074203 32718 solver.cpp:218] Iteration 8700 (2.45402 iter/s, 4.88993s/12 iters), loss = 0.119389
I0407 23:38:11.074245 32718 solver.cpp:237]     Train net output #0: loss = 0.119389 (* 1 = 0.119389 loss)
I0407 23:38:11.074254 32718 sgd_solver.cpp:105] Iteration 8700, lr = 0.00146202
I0407 23:38:16.028534 32718 solver.cpp:218] Iteration 8712 (2.42216 iter/s, 4.95425s/12 iters), loss = 0.0215071
I0407 23:38:16.028595 32718 solver.cpp:237]     Train net output #0: loss = 0.0215071 (* 1 = 0.0215071 loss)
I0407 23:38:16.028607 32718 sgd_solver.cpp:105] Iteration 8712, lr = 0.00145469
I0407 23:38:20.957219 32718 solver.cpp:218] Iteration 8724 (2.43477 iter/s, 4.9286s/12 iters), loss = 0.0322
I0407 23:38:20.957262 32718 solver.cpp:237]     Train net output #0: loss = 0.0322 (* 1 = 0.0322 loss)
I0407 23:38:20.957271 32718 sgd_solver.cpp:105] Iteration 8724, lr = 0.00144739
I0407 23:38:25.890267 32718 solver.cpp:218] Iteration 8736 (2.43261 iter/s, 4.93298s/12 iters), loss = 0.0650152
I0407 23:38:25.890307 32718 solver.cpp:237]     Train net output #0: loss = 0.0650152 (* 1 = 0.0650152 loss)
I0407 23:38:25.890316 32718 sgd_solver.cpp:105] Iteration 8736, lr = 0.00144013
I0407 23:38:30.829855 32718 solver.cpp:218] Iteration 8748 (2.42939 iter/s, 4.93951s/12 iters), loss = 0.0915179
I0407 23:38:30.829916 32718 solver.cpp:237]     Train net output #0: loss = 0.0915179 (* 1 = 0.0915179 loss)
I0407 23:38:30.829928 32718 sgd_solver.cpp:105] Iteration 8748, lr = 0.00143289
I0407 23:38:35.769670 32718 solver.cpp:218] Iteration 8760 (2.42929 iter/s, 4.93972s/12 iters), loss = 0.0781322
I0407 23:38:35.769732 32718 solver.cpp:237]     Train net output #0: loss = 0.0781322 (* 1 = 0.0781322 loss)
I0407 23:38:35.769747 32718 sgd_solver.cpp:105] Iteration 8760, lr = 0.00142569
I0407 23:38:40.261694 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_8772.caffemodel
I0407 23:38:43.352938 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_8772.solverstate
I0407 23:38:45.782883 32718 solver.cpp:330] Iteration 8772, Testing net (#0)
I0407 23:38:45.782907 32718 net.cpp:676] Ignoring source layer train-data
I0407 23:38:46.778859 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:38:50.554893 32718 solver.cpp:397]     Test net output #0: accuracy = 0.514706
I0407 23:38:50.554937 32718 solver.cpp:397]     Test net output #1: loss = 2.83378 (* 1 = 2.83378 loss)
I0407 23:38:50.651532 32718 solver.cpp:218] Iteration 8772 (0.806357 iter/s, 14.8818s/12 iters), loss = 0.0507861
I0407 23:38:50.651568 32718 solver.cpp:237]     Train net output #0: loss = 0.0507861 (* 1 = 0.0507861 loss)
I0407 23:38:50.651576 32718 sgd_solver.cpp:105] Iteration 8772, lr = 0.00141851
I0407 23:38:54.804577 32718 solver.cpp:218] Iteration 8784 (2.88949 iter/s, 4.15299s/12 iters), loss = 0.0349711
I0407 23:38:54.804615 32718 solver.cpp:237]     Train net output #0: loss = 0.0349711 (* 1 = 0.0349711 loss)
I0407 23:38:54.804625 32718 sgd_solver.cpp:105] Iteration 8784, lr = 0.00141136
I0407 23:38:59.720044 32718 solver.cpp:218] Iteration 8796 (2.44131 iter/s, 4.9154s/12 iters), loss = 0.0364367
I0407 23:38:59.720083 32718 solver.cpp:237]     Train net output #0: loss = 0.0364366 (* 1 = 0.0364366 loss)
I0407 23:38:59.720090 32718 sgd_solver.cpp:105] Iteration 8796, lr = 0.00140425
I0407 23:39:01.127498 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:39:04.655663 32718 solver.cpp:218] Iteration 8808 (2.43134 iter/s, 4.93555s/12 iters), loss = 0.0669116
I0407 23:39:04.655707 32718 solver.cpp:237]     Train net output #0: loss = 0.0669116 (* 1 = 0.0669116 loss)
I0407 23:39:04.655716 32718 sgd_solver.cpp:105] Iteration 8808, lr = 0.00139716
I0407 23:39:09.482682 32718 solver.cpp:218] Iteration 8820 (2.48604 iter/s, 4.82695s/12 iters), loss = 0.0406679
I0407 23:39:09.482728 32718 solver.cpp:237]     Train net output #0: loss = 0.0406679 (* 1 = 0.0406679 loss)
I0407 23:39:09.482735 32718 sgd_solver.cpp:105] Iteration 8820, lr = 0.00139011
I0407 23:39:14.400549 32718 solver.cpp:218] Iteration 8832 (2.44012 iter/s, 4.91779s/12 iters), loss = 0.015536
I0407 23:39:14.400674 32718 solver.cpp:237]     Train net output #0: loss = 0.015536 (* 1 = 0.015536 loss)
I0407 23:39:14.400684 32718 sgd_solver.cpp:105] Iteration 8832, lr = 0.00138308
I0407 23:39:19.339470 32718 solver.cpp:218] Iteration 8844 (2.42976 iter/s, 4.93875s/12 iters), loss = 0.0663258
I0407 23:39:19.339543 32718 solver.cpp:237]     Train net output #0: loss = 0.0663258 (* 1 = 0.0663258 loss)
I0407 23:39:19.339560 32718 sgd_solver.cpp:105] Iteration 8844, lr = 0.00137609
I0407 23:39:24.270880 32718 solver.cpp:218] Iteration 8856 (2.43343 iter/s, 4.93132s/12 iters), loss = 0.0655622
I0407 23:39:24.270917 32718 solver.cpp:237]     Train net output #0: loss = 0.0655622 (* 1 = 0.0655622 loss)
I0407 23:39:24.270925 32718 sgd_solver.cpp:105] Iteration 8856, lr = 0.00136912
I0407 23:39:29.238554 32718 solver.cpp:218] Iteration 8868 (2.41565 iter/s, 4.96761s/12 iters), loss = 0.0869624
I0407 23:39:29.238596 32718 solver.cpp:237]     Train net output #0: loss = 0.0869624 (* 1 = 0.0869624 loss)
I0407 23:39:29.238605 32718 sgd_solver.cpp:105] Iteration 8868, lr = 0.00136219
I0407 23:39:31.219712 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_8874.caffemodel
I0407 23:39:34.352921 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_8874.solverstate
I0407 23:39:36.714447 32718 solver.cpp:330] Iteration 8874, Testing net (#0)
I0407 23:39:36.714468 32718 net.cpp:676] Ignoring source layer train-data
I0407 23:39:37.733788 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:39:41.514204 32718 solver.cpp:397]     Test net output #0: accuracy = 0.515931
I0407 23:39:41.514250 32718 solver.cpp:397]     Test net output #1: loss = 2.81598 (* 1 = 2.81598 loss)
I0407 23:39:43.340344 32718 solver.cpp:218] Iteration 8880 (0.850962 iter/s, 14.1017s/12 iters), loss = 0.0340876
I0407 23:39:43.340389 32718 solver.cpp:237]     Train net output #0: loss = 0.0340876 (* 1 = 0.0340876 loss)
I0407 23:39:43.340396 32718 sgd_solver.cpp:105] Iteration 8880, lr = 0.00135528
I0407 23:39:48.295449 32718 solver.cpp:218] Iteration 8892 (2.42178 iter/s, 4.95504s/12 iters), loss = 0.0359496
I0407 23:39:48.295603 32718 solver.cpp:237]     Train net output #0: loss = 0.0359496 (* 1 = 0.0359496 loss)
I0407 23:39:48.295612 32718 sgd_solver.cpp:105] Iteration 8892, lr = 0.0013484
I0407 23:39:51.814965 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:39:53.197837 32718 solver.cpp:218] Iteration 8904 (2.44788 iter/s, 4.90221s/12 iters), loss = 0.0281389
I0407 23:39:53.197877 32718 solver.cpp:237]     Train net output #0: loss = 0.0281389 (* 1 = 0.0281389 loss)
I0407 23:39:53.197886 32718 sgd_solver.cpp:105] Iteration 8904, lr = 0.00134155
I0407 23:39:58.082294 32718 solver.cpp:218] Iteration 8916 (2.45681 iter/s, 4.88439s/12 iters), loss = 0.0163322
I0407 23:39:58.082336 32718 solver.cpp:237]     Train net output #0: loss = 0.0163322 (* 1 = 0.0163322 loss)
I0407 23:39:58.082345 32718 sgd_solver.cpp:105] Iteration 8916, lr = 0.00133474
I0407 23:40:03.042747 32718 solver.cpp:218] Iteration 8928 (2.41917 iter/s, 4.96038s/12 iters), loss = 0.0217027
I0407 23:40:03.042793 32718 solver.cpp:237]     Train net output #0: loss = 0.0217027 (* 1 = 0.0217027 loss)
I0407 23:40:03.042800 32718 sgd_solver.cpp:105] Iteration 8928, lr = 0.00132795
I0407 23:40:07.836047 32718 solver.cpp:218] Iteration 8940 (2.50353 iter/s, 4.79322s/12 iters), loss = 0.0185085
I0407 23:40:07.836092 32718 solver.cpp:237]     Train net output #0: loss = 0.0185085 (* 1 = 0.0185085 loss)
I0407 23:40:07.836102 32718 sgd_solver.cpp:105] Iteration 8940, lr = 0.00132119
I0407 23:40:12.765722 32718 solver.cpp:218] Iteration 8952 (2.43428 iter/s, 4.92959s/12 iters), loss = 0.0124033
I0407 23:40:12.765767 32718 solver.cpp:237]     Train net output #0: loss = 0.0124033 (* 1 = 0.0124033 loss)
I0407 23:40:12.765775 32718 sgd_solver.cpp:105] Iteration 8952, lr = 0.00131446
I0407 23:40:17.720571 32718 solver.cpp:218] Iteration 8964 (2.42191 iter/s, 4.95477s/12 iters), loss = 0.0166354
I0407 23:40:17.720613 32718 solver.cpp:237]     Train net output #0: loss = 0.0166354 (* 1 = 0.0166354 loss)
I0407 23:40:17.720621 32718 sgd_solver.cpp:105] Iteration 8964, lr = 0.00130776
I0407 23:40:22.178328 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_8976.caffemodel
I0407 23:40:25.235649 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_8976.solverstate
I0407 23:40:27.630434 32718 solver.cpp:330] Iteration 8976, Testing net (#0)
I0407 23:40:27.630455 32718 net.cpp:676] Ignoring source layer train-data
I0407 23:40:28.567384 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:40:32.227674 32718 solver.cpp:397]     Test net output #0: accuracy = 0.51777
I0407 23:40:32.227720 32718 solver.cpp:397]     Test net output #1: loss = 2.8108 (* 1 = 2.8108 loss)
I0407 23:40:32.324296 32718 solver.cpp:218] Iteration 8976 (0.821714 iter/s, 14.6036s/12 iters), loss = 0.00865678
I0407 23:40:32.324334 32718 solver.cpp:237]     Train net output #0: loss = 0.00865678 (* 1 = 0.00865678 loss)
I0407 23:40:32.324342 32718 sgd_solver.cpp:105] Iteration 8976, lr = 0.00130108
I0407 23:40:36.464102 32718 solver.cpp:218] Iteration 8988 (2.89873 iter/s, 4.13974s/12 iters), loss = 0.10419
I0407 23:40:36.464138 32718 solver.cpp:237]     Train net output #0: loss = 0.10419 (* 1 = 0.10419 loss)
I0407 23:40:36.464148 32718 sgd_solver.cpp:105] Iteration 8988, lr = 0.00129444
I0407 23:40:39.684715 32718 blocking_queue.cpp:49] Waiting for data
I0407 23:40:41.423506 32718 solver.cpp:218] Iteration 9000 (2.41968 iter/s, 4.95934s/12 iters), loss = 0.0271623
I0407 23:40:41.423553 32718 solver.cpp:237]     Train net output #0: loss = 0.0271623 (* 1 = 0.0271623 loss)
I0407 23:40:41.423563 32718 sgd_solver.cpp:105] Iteration 9000, lr = 0.00128783
I0407 23:40:42.105378 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:40:46.321892 32718 solver.cpp:218] Iteration 9012 (2.44982 iter/s, 4.89831s/12 iters), loss = 0.0165763
I0407 23:40:46.321930 32718 solver.cpp:237]     Train net output #0: loss = 0.0165763 (* 1 = 0.0165763 loss)
I0407 23:40:46.321938 32718 sgd_solver.cpp:105] Iteration 9012, lr = 0.00128124
I0407 23:40:51.295207 32718 solver.cpp:218] Iteration 9024 (2.41291 iter/s, 4.97324s/12 iters), loss = 0.120441
I0407 23:40:51.295249 32718 solver.cpp:237]     Train net output #0: loss = 0.120441 (* 1 = 0.120441 loss)
I0407 23:40:51.295258 32718 sgd_solver.cpp:105] Iteration 9024, lr = 0.00127468
I0407 23:40:56.202802 32718 solver.cpp:218] Iteration 9036 (2.44523 iter/s, 4.90752s/12 iters), loss = 0.00395652
I0407 23:40:56.202951 32718 solver.cpp:237]     Train net output #0: loss = 0.00395651 (* 1 = 0.00395651 loss)
I0407 23:40:56.202960 32718 sgd_solver.cpp:105] Iteration 9036, lr = 0.00126816
I0407 23:41:01.169106 32718 solver.cpp:218] Iteration 9048 (2.41637 iter/s, 4.96613s/12 iters), loss = 0.115286
I0407 23:41:01.169142 32718 solver.cpp:237]     Train net output #0: loss = 0.115286 (* 1 = 0.115286 loss)
I0407 23:41:01.169149 32718 sgd_solver.cpp:105] Iteration 9048, lr = 0.00126166
I0407 23:41:06.093072 32718 solver.cpp:218] Iteration 9060 (2.43709 iter/s, 4.9239s/12 iters), loss = 0.0570471
I0407 23:41:06.093111 32718 solver.cpp:237]     Train net output #0: loss = 0.0570471 (* 1 = 0.0570471 loss)
I0407 23:41:06.093118 32718 sgd_solver.cpp:105] Iteration 9060, lr = 0.00125519
I0407 23:41:11.050027 32718 solver.cpp:218] Iteration 9072 (2.42087 iter/s, 4.95689s/12 iters), loss = 0.0518132
I0407 23:41:11.050065 32718 solver.cpp:237]     Train net output #0: loss = 0.0518132 (* 1 = 0.0518132 loss)
I0407 23:41:11.050074 32718 sgd_solver.cpp:105] Iteration 9072, lr = 0.00124874
I0407 23:41:13.031411 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_9078.caffemodel
I0407 23:41:16.125417 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_9078.solverstate
I0407 23:41:18.484460 32718 solver.cpp:330] Iteration 9078, Testing net (#0)
I0407 23:41:18.484483 32718 net.cpp:676] Ignoring source layer train-data
I0407 23:41:19.349367 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:41:23.248847 32718 solver.cpp:397]     Test net output #0: accuracy = 0.509804
I0407 23:41:23.248893 32718 solver.cpp:397]     Test net output #1: loss = 2.80493 (* 1 = 2.80493 loss)
I0407 23:41:25.082944 32718 solver.cpp:218] Iteration 9084 (0.855138 iter/s, 14.0328s/12 iters), loss = 0.0412173
I0407 23:41:25.082991 32718 solver.cpp:237]     Train net output #0: loss = 0.0412173 (* 1 = 0.0412173 loss)
I0407 23:41:25.082998 32718 sgd_solver.cpp:105] Iteration 9084, lr = 0.00124233
I0407 23:41:30.013756 32718 solver.cpp:218] Iteration 9096 (2.43371 iter/s, 4.93074s/12 iters), loss = 0.0353497
I0407 23:41:30.013887 32718 solver.cpp:237]     Train net output #0: loss = 0.0353497 (* 1 = 0.0353497 loss)
I0407 23:41:30.013896 32718 sgd_solver.cpp:105] Iteration 9096, lr = 0.00123594
I0407 23:41:32.925017 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:41:34.953727 32718 solver.cpp:218] Iteration 9108 (2.42924 iter/s, 4.93982s/12 iters), loss = 0.0233563
I0407 23:41:34.953765 32718 solver.cpp:237]     Train net output #0: loss = 0.0233563 (* 1 = 0.0233563 loss)
I0407 23:41:34.953773 32718 sgd_solver.cpp:105] Iteration 9108, lr = 0.00122959
I0407 23:41:39.802487 32718 solver.cpp:218] Iteration 9120 (2.4749 iter/s, 4.84869s/12 iters), loss = 0.0878657
I0407 23:41:39.802533 32718 solver.cpp:237]     Train net output #0: loss = 0.0878657 (* 1 = 0.0878657 loss)
I0407 23:41:39.802542 32718 sgd_solver.cpp:105] Iteration 9120, lr = 0.00122326
I0407 23:41:44.669121 32718 solver.cpp:218] Iteration 9132 (2.46581 iter/s, 4.86656s/12 iters), loss = 0.0417881
I0407 23:41:44.669163 32718 solver.cpp:237]     Train net output #0: loss = 0.041788 (* 1 = 0.041788 loss)
I0407 23:41:44.669171 32718 sgd_solver.cpp:105] Iteration 9132, lr = 0.00121696
I0407 23:41:49.536437 32718 solver.cpp:218] Iteration 9144 (2.46546 iter/s, 4.86724s/12 iters), loss = 0.0411171
I0407 23:41:49.536474 32718 solver.cpp:237]     Train net output #0: loss = 0.0411171 (* 1 = 0.0411171 loss)
I0407 23:41:49.536482 32718 sgd_solver.cpp:105] Iteration 9144, lr = 0.00121068
I0407 23:41:54.491355 32718 solver.cpp:218] Iteration 9156 (2.42187 iter/s, 4.95486s/12 iters), loss = 0.106168
I0407 23:41:54.491391 32718 solver.cpp:237]     Train net output #0: loss = 0.106168 (* 1 = 0.106168 loss)
I0407 23:41:54.491398 32718 sgd_solver.cpp:105] Iteration 9156, lr = 0.00120444
I0407 23:41:59.421644 32718 solver.cpp:218] Iteration 9168 (2.43397 iter/s, 4.93023s/12 iters), loss = 0.0113136
I0407 23:41:59.421679 32718 solver.cpp:237]     Train net output #0: loss = 0.0113135 (* 1 = 0.0113135 loss)
I0407 23:41:59.421687 32718 sgd_solver.cpp:105] Iteration 9168, lr = 0.00119822
I0407 23:42:03.885854 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_9180.caffemodel
I0407 23:42:07.002938 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_9180.solverstate
I0407 23:42:09.375034 32718 solver.cpp:330] Iteration 9180, Testing net (#0)
I0407 23:42:09.375051 32718 net.cpp:676] Ignoring source layer train-data
I0407 23:42:10.219146 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:42:14.036113 32718 solver.cpp:397]     Test net output #0: accuracy = 0.521446
I0407 23:42:14.036157 32718 solver.cpp:397]     Test net output #1: loss = 2.84495 (* 1 = 2.84495 loss)
I0407 23:42:14.132789 32718 solver.cpp:218] Iteration 9180 (0.815713 iter/s, 14.7111s/12 iters), loss = 0.0274543
I0407 23:42:14.132833 32718 solver.cpp:237]     Train net output #0: loss = 0.0274542 (* 1 = 0.0274542 loss)
I0407 23:42:14.132840 32718 sgd_solver.cpp:105] Iteration 9180, lr = 0.00119203
I0407 23:42:18.273212 32718 solver.cpp:218] Iteration 9192 (2.8983 iter/s, 4.14036s/12 iters), loss = 0.0339462
I0407 23:42:18.273253 32718 solver.cpp:237]     Train net output #0: loss = 0.0339462 (* 1 = 0.0339462 loss)
I0407 23:42:18.273262 32718 sgd_solver.cpp:105] Iteration 9192, lr = 0.00118587
I0407 23:42:23.219928 32718 solver.cpp:218] Iteration 9204 (2.42589 iter/s, 4.94665s/12 iters), loss = 0.107338
I0407 23:42:23.219964 32718 solver.cpp:237]     Train net output #0: loss = 0.107338 (* 1 = 0.107338 loss)
I0407 23:42:23.219971 32718 sgd_solver.cpp:105] Iteration 9204, lr = 0.00117973
I0407 23:42:23.283663 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:42:28.321540 32718 solver.cpp:218] Iteration 9216 (2.35223 iter/s, 5.10155s/12 iters), loss = 0.0490973
I0407 23:42:28.321575 32718 solver.cpp:237]     Train net output #0: loss = 0.0490973 (* 1 = 0.0490973 loss)
I0407 23:42:28.321583 32718 sgd_solver.cpp:105] Iteration 9216, lr = 0.00117362
I0407 23:42:33.292084 32718 solver.cpp:218] Iteration 9228 (2.41425 iter/s, 4.97048s/12 iters), loss = 0.0149973
I0407 23:42:33.292121 32718 solver.cpp:237]     Train net output #0: loss = 0.0149973 (* 1 = 0.0149973 loss)
I0407 23:42:33.292129 32718 sgd_solver.cpp:105] Iteration 9228, lr = 0.00116755
I0407 23:42:38.367978 32718 solver.cpp:218] Iteration 9240 (2.36415 iter/s, 5.07583s/12 iters), loss = 0.0299915
I0407 23:42:38.368098 32718 solver.cpp:237]     Train net output #0: loss = 0.0299914 (* 1 = 0.0299914 loss)
I0407 23:42:38.368106 32718 sgd_solver.cpp:105] Iteration 9240, lr = 0.00116149
I0407 23:42:43.383854 32718 solver.cpp:218] Iteration 9252 (2.39248 iter/s, 5.01572s/12 iters), loss = 0.0300962
I0407 23:42:43.383908 32718 solver.cpp:237]     Train net output #0: loss = 0.0300962 (* 1 = 0.0300962 loss)
I0407 23:42:43.383920 32718 sgd_solver.cpp:105] Iteration 9252, lr = 0.00115547
I0407 23:42:48.360719 32718 solver.cpp:218] Iteration 9264 (2.41119 iter/s, 4.97679s/12 iters), loss = 0.0489489
I0407 23:42:48.360759 32718 solver.cpp:237]     Train net output #0: loss = 0.0489488 (* 1 = 0.0489488 loss)
I0407 23:42:48.360767 32718 sgd_solver.cpp:105] Iteration 9264, lr = 0.00114947
I0407 23:42:53.291676 32718 solver.cpp:218] Iteration 9276 (2.43364 iter/s, 4.93089s/12 iters), loss = 0.0374232
I0407 23:42:53.291714 32718 solver.cpp:237]     Train net output #0: loss = 0.0374232 (* 1 = 0.0374232 loss)
I0407 23:42:53.291723 32718 sgd_solver.cpp:105] Iteration 9276, lr = 0.0011435
I0407 23:42:55.274129 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_9282.caffemodel
I0407 23:42:58.378386 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_9282.solverstate
I0407 23:43:02.747026 32718 solver.cpp:330] Iteration 9282, Testing net (#0)
I0407 23:43:02.747051 32718 net.cpp:676] Ignoring source layer train-data
I0407 23:43:03.554188 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:43:07.812903 32718 solver.cpp:397]     Test net output #0: accuracy = 0.521446
I0407 23:43:07.812948 32718 solver.cpp:397]     Test net output #1: loss = 2.80924 (* 1 = 2.80924 loss)
I0407 23:43:09.618925 32718 solver.cpp:218] Iteration 9288 (0.734972 iter/s, 16.3272s/12 iters), loss = 0.107651
I0407 23:43:09.619081 32718 solver.cpp:237]     Train net output #0: loss = 0.107651 (* 1 = 0.107651 loss)
I0407 23:43:09.619091 32718 sgd_solver.cpp:105] Iteration 9288, lr = 0.00113756
I0407 23:43:14.578058 32718 solver.cpp:218] Iteration 9300 (2.41986 iter/s, 4.95895s/12 iters), loss = 0.0827954
I0407 23:43:14.578095 32718 solver.cpp:237]     Train net output #0: loss = 0.0827954 (* 1 = 0.0827954 loss)
I0407 23:43:14.578102 32718 sgd_solver.cpp:105] Iteration 9300, lr = 0.00113164
I0407 23:43:16.740684 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:43:19.481873 32718 solver.cpp:218] Iteration 9312 (2.44711 iter/s, 4.90375s/12 iters), loss = 0.04508
I0407 23:43:19.481914 32718 solver.cpp:237]     Train net output #0: loss = 0.04508 (* 1 = 0.04508 loss)
I0407 23:43:19.481921 32718 sgd_solver.cpp:105] Iteration 9312, lr = 0.00112575
I0407 23:43:24.366003 32718 solver.cpp:218] Iteration 9324 (2.45697 iter/s, 4.88406s/12 iters), loss = 0.0169742
I0407 23:43:24.366040 32718 solver.cpp:237]     Train net output #0: loss = 0.0169741 (* 1 = 0.0169741 loss)
I0407 23:43:24.366047 32718 sgd_solver.cpp:105] Iteration 9324, lr = 0.00111989
I0407 23:43:29.280457 32718 solver.cpp:218] Iteration 9336 (2.44181 iter/s, 4.91439s/12 iters), loss = 0.0123409
I0407 23:43:29.280495 32718 solver.cpp:237]     Train net output #0: loss = 0.0123409 (* 1 = 0.0123409 loss)
I0407 23:43:29.280503 32718 sgd_solver.cpp:105] Iteration 9336, lr = 0.00111405
I0407 23:43:34.227820 32718 solver.cpp:218] Iteration 9348 (2.42557 iter/s, 4.9473s/12 iters), loss = 0.00830755
I0407 23:43:34.227861 32718 solver.cpp:237]     Train net output #0: loss = 0.00830751 (* 1 = 0.00830751 loss)
I0407 23:43:34.227870 32718 sgd_solver.cpp:105] Iteration 9348, lr = 0.00110824
I0407 23:43:39.158942 32718 solver.cpp:218] Iteration 9360 (2.43356 iter/s, 4.93105s/12 iters), loss = 0.069061
I0407 23:43:39.158978 32718 solver.cpp:237]     Train net output #0: loss = 0.069061 (* 1 = 0.069061 loss)
I0407 23:43:39.158985 32718 sgd_solver.cpp:105] Iteration 9360, lr = 0.00110246
I0407 23:43:44.143079 32718 solver.cpp:218] Iteration 9372 (2.40767 iter/s, 4.98407s/12 iters), loss = 0.0300276
I0407 23:43:44.143224 32718 solver.cpp:237]     Train net output #0: loss = 0.0300276 (* 1 = 0.0300276 loss)
I0407 23:43:44.143234 32718 sgd_solver.cpp:105] Iteration 9372, lr = 0.0010967
I0407 23:43:48.581149 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_9384.caffemodel
I0407 23:43:51.668951 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_9384.solverstate
I0407 23:43:54.417860 32718 solver.cpp:330] Iteration 9384, Testing net (#0)
I0407 23:43:54.417876 32718 net.cpp:676] Ignoring source layer train-data
I0407 23:43:55.226420 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:43:59.556788 32718 solver.cpp:397]     Test net output #0: accuracy = 0.521446
I0407 23:43:59.556834 32718 solver.cpp:397]     Test net output #1: loss = 2.82249 (* 1 = 2.82249 loss)
I0407 23:43:59.653492 32718 solver.cpp:218] Iteration 9384 (0.773684 iter/s, 15.5102s/12 iters), loss = 0.0118996
I0407 23:43:59.653537 32718 solver.cpp:237]     Train net output #0: loss = 0.0118996 (* 1 = 0.0118996 loss)
I0407 23:43:59.653545 32718 sgd_solver.cpp:105] Iteration 9384, lr = 0.00109097
I0407 23:44:03.867031 32718 solver.cpp:218] Iteration 9396 (2.84801 iter/s, 4.21346s/12 iters), loss = 0.0595221
I0407 23:44:03.867077 32718 solver.cpp:237]     Train net output #0: loss = 0.059522 (* 1 = 0.059522 loss)
I0407 23:44:03.867085 32718 sgd_solver.cpp:105] Iteration 9396, lr = 0.00108526
I0407 23:44:08.166199 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:44:08.810914 32718 solver.cpp:218] Iteration 9408 (2.42728 iter/s, 4.94381s/12 iters), loss = 0.0303462
I0407 23:44:08.810961 32718 solver.cpp:237]     Train net output #0: loss = 0.0303462 (* 1 = 0.0303462 loss)
I0407 23:44:08.810968 32718 sgd_solver.cpp:105] Iteration 9408, lr = 0.00107959
I0407 23:44:13.744577 32718 solver.cpp:218] Iteration 9420 (2.43231 iter/s, 4.93358s/12 iters), loss = 0.00565969
I0407 23:44:13.744626 32718 solver.cpp:237]     Train net output #0: loss = 0.00565967 (* 1 = 0.00565967 loss)
I0407 23:44:13.744635 32718 sgd_solver.cpp:105] Iteration 9420, lr = 0.00107393
I0407 23:44:18.712467 32718 solver.cpp:218] Iteration 9432 (2.41555 iter/s, 4.96781s/12 iters), loss = 0.0331987
I0407 23:44:18.712633 32718 solver.cpp:237]     Train net output #0: loss = 0.0331987 (* 1 = 0.0331987 loss)
I0407 23:44:18.712643 32718 sgd_solver.cpp:105] Iteration 9432, lr = 0.00106831
I0407 23:44:23.643535 32718 solver.cpp:218] Iteration 9444 (2.43365 iter/s, 4.93087s/12 iters), loss = 0.00688609
I0407 23:44:23.643580 32718 solver.cpp:237]     Train net output #0: loss = 0.0068861 (* 1 = 0.0068861 loss)
I0407 23:44:23.643589 32718 sgd_solver.cpp:105] Iteration 9444, lr = 0.00106271
I0407 23:44:28.623422 32718 solver.cpp:218] Iteration 9456 (2.40973 iter/s, 4.97981s/12 iters), loss = 0.0729163
I0407 23:44:28.623466 32718 solver.cpp:237]     Train net output #0: loss = 0.0729163 (* 1 = 0.0729163 loss)
I0407 23:44:28.623474 32718 sgd_solver.cpp:105] Iteration 9456, lr = 0.00105713
I0407 23:44:33.590502 32718 solver.cpp:218] Iteration 9468 (2.41594 iter/s, 4.96701s/12 iters), loss = 0.0553833
I0407 23:44:33.590545 32718 solver.cpp:237]     Train net output #0: loss = 0.0553833 (* 1 = 0.0553833 loss)
I0407 23:44:33.590553 32718 sgd_solver.cpp:105] Iteration 9468, lr = 0.00105159
I0407 23:44:38.531157 32718 solver.cpp:218] Iteration 9480 (2.42886 iter/s, 4.94059s/12 iters), loss = 0.0285038
I0407 23:44:38.531205 32718 solver.cpp:237]     Train net output #0: loss = 0.0285038 (* 1 = 0.0285038 loss)
I0407 23:44:38.531215 32718 sgd_solver.cpp:105] Iteration 9480, lr = 0.00104606
I0407 23:44:40.569171 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_9486.caffemodel
I0407 23:44:43.518057 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_9486.solverstate
I0407 23:44:46.636442 32718 solver.cpp:330] Iteration 9486, Testing net (#0)
I0407 23:44:46.636458 32718 net.cpp:676] Ignoring source layer train-data
I0407 23:44:47.393323 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:44:51.378134 32718 solver.cpp:397]     Test net output #0: accuracy = 0.521446
I0407 23:44:51.378279 32718 solver.cpp:397]     Test net output #1: loss = 2.79851 (* 1 = 2.79851 loss)
I0407 23:44:53.193543 32718 solver.cpp:218] Iteration 9492 (0.818427 iter/s, 14.6623s/12 iters), loss = 0.0783576
I0407 23:44:53.193594 32718 solver.cpp:237]     Train net output #0: loss = 0.0783576 (* 1 = 0.0783576 loss)
I0407 23:44:53.193603 32718 sgd_solver.cpp:105] Iteration 9492, lr = 0.00104057
I0407 23:44:58.140126 32718 solver.cpp:218] Iteration 9504 (2.42596 iter/s, 4.9465s/12 iters), loss = 0.0129336
I0407 23:44:58.140167 32718 solver.cpp:237]     Train net output #0: loss = 0.0129336 (* 1 = 0.0129336 loss)
I0407 23:44:58.140175 32718 sgd_solver.cpp:105] Iteration 9504, lr = 0.0010351
I0407 23:44:59.573988 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:45:03.034271 32718 solver.cpp:218] Iteration 9516 (2.45194 iter/s, 4.89408s/12 iters), loss = 0.0337759
I0407 23:45:03.034308 32718 solver.cpp:237]     Train net output #0: loss = 0.0337759 (* 1 = 0.0337759 loss)
I0407 23:45:03.034317 32718 sgd_solver.cpp:105] Iteration 9516, lr = 0.00102965
I0407 23:45:07.999087 32718 solver.cpp:218] Iteration 9528 (2.41704 iter/s, 4.96475s/12 iters), loss = 0.0438171
I0407 23:45:07.999125 32718 solver.cpp:237]     Train net output #0: loss = 0.0438171 (* 1 = 0.0438171 loss)
I0407 23:45:07.999131 32718 sgd_solver.cpp:105] Iteration 9528, lr = 0.00102423
I0407 23:45:12.924742 32718 solver.cpp:218] Iteration 9540 (2.43626 iter/s, 4.92558s/12 iters), loss = 0.0157008
I0407 23:45:12.924785 32718 solver.cpp:237]     Train net output #0: loss = 0.0157008 (* 1 = 0.0157008 loss)
I0407 23:45:12.924793 32718 sgd_solver.cpp:105] Iteration 9540, lr = 0.00101883
I0407 23:45:17.887499 32718 solver.cpp:218] Iteration 9552 (2.41805 iter/s, 4.96268s/12 iters), loss = 0.050874
I0407 23:45:17.887542 32718 solver.cpp:237]     Train net output #0: loss = 0.050874 (* 1 = 0.050874 loss)
I0407 23:45:17.887550 32718 sgd_solver.cpp:105] Iteration 9552, lr = 0.00101346
I0407 23:45:22.868276 32718 solver.cpp:218] Iteration 9564 (2.4093 iter/s, 4.98071s/12 iters), loss = 0.0219514
I0407 23:45:22.868424 32718 solver.cpp:237]     Train net output #0: loss = 0.0219514 (* 1 = 0.0219514 loss)
I0407 23:45:22.868433 32718 sgd_solver.cpp:105] Iteration 9564, lr = 0.00100812
I0407 23:45:27.868288 32718 solver.cpp:218] Iteration 9576 (2.40008 iter/s, 4.99984s/12 iters), loss = 0.0153725
I0407 23:45:27.868328 32718 solver.cpp:237]     Train net output #0: loss = 0.0153726 (* 1 = 0.0153726 loss)
I0407 23:45:27.868335 32718 sgd_solver.cpp:105] Iteration 9576, lr = 0.0010028
I0407 23:45:32.254173 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_9588.caffemodel
I0407 23:45:35.328761 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_9588.solverstate
I0407 23:45:38.083140 32718 solver.cpp:330] Iteration 9588, Testing net (#0)
I0407 23:45:38.083158 32718 net.cpp:676] Ignoring source layer train-data
I0407 23:45:38.687618 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:45:42.516052 32718 solver.cpp:397]     Test net output #0: accuracy = 0.519608
I0407 23:45:42.516099 32718 solver.cpp:397]     Test net output #1: loss = 2.84287 (* 1 = 2.84287 loss)
I0407 23:45:42.612843 32718 solver.cpp:218] Iteration 9588 (0.813865 iter/s, 14.7445s/12 iters), loss = 0.0210052
I0407 23:45:42.612888 32718 solver.cpp:237]     Train net output #0: loss = 0.0210053 (* 1 = 0.0210053 loss)
I0407 23:45:42.612897 32718 sgd_solver.cpp:105] Iteration 9588, lr = 0.000997505
I0407 23:45:46.764995 32718 solver.cpp:218] Iteration 9600 (2.89012 iter/s, 4.15208s/12 iters), loss = 0.015546
I0407 23:45:46.765041 32718 solver.cpp:237]     Train net output #0: loss = 0.015546 (* 1 = 0.015546 loss)
I0407 23:45:46.765049 32718 sgd_solver.cpp:105] Iteration 9600, lr = 0.000992235
I0407 23:45:50.310437 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:45:51.666844 32718 solver.cpp:218] Iteration 9612 (2.44809 iter/s, 4.90178s/12 iters), loss = 0.072204
I0407 23:45:51.666880 32718 solver.cpp:237]     Train net output #0: loss = 0.072204 (* 1 = 0.072204 loss)
I0407 23:45:51.666889 32718 sgd_solver.cpp:105] Iteration 9612, lr = 0.00098699
I0407 23:45:56.628681 32718 solver.cpp:218] Iteration 9624 (2.41849 iter/s, 4.96177s/12 iters), loss = 0.00457142
I0407 23:45:56.628845 32718 solver.cpp:237]     Train net output #0: loss = 0.00457145 (* 1 = 0.00457145 loss)
I0407 23:45:56.628855 32718 sgd_solver.cpp:105] Iteration 9624, lr = 0.000981769
I0407 23:46:01.545886 32718 solver.cpp:218] Iteration 9636 (2.44051 iter/s, 4.91701s/12 iters), loss = 0.00935571
I0407 23:46:01.545928 32718 solver.cpp:237]     Train net output #0: loss = 0.00935574 (* 1 = 0.00935574 loss)
I0407 23:46:01.545938 32718 sgd_solver.cpp:105] Iteration 9636, lr = 0.000976573
I0407 23:46:06.486418 32718 solver.cpp:218] Iteration 9648 (2.42893 iter/s, 4.94046s/12 iters), loss = 0.055027
I0407 23:46:06.486461 32718 solver.cpp:237]     Train net output #0: loss = 0.055027 (* 1 = 0.055027 loss)
I0407 23:46:06.486469 32718 sgd_solver.cpp:105] Iteration 9648, lr = 0.000971402
I0407 23:46:11.429988 32718 solver.cpp:218] Iteration 9660 (2.42743 iter/s, 4.9435s/12 iters), loss = 0.0393004
I0407 23:46:11.430033 32718 solver.cpp:237]     Train net output #0: loss = 0.0393004 (* 1 = 0.0393004 loss)
I0407 23:46:11.430042 32718 sgd_solver.cpp:105] Iteration 9660, lr = 0.000966255
I0407 23:46:16.382328 32718 solver.cpp:218] Iteration 9672 (2.42313 iter/s, 4.95226s/12 iters), loss = 0.0538814
I0407 23:46:16.382372 32718 solver.cpp:237]     Train net output #0: loss = 0.0538814 (* 1 = 0.0538814 loss)
I0407 23:46:16.382380 32718 sgd_solver.cpp:105] Iteration 9672, lr = 0.000961133
I0407 23:46:21.302594 32718 solver.cpp:218] Iteration 9684 (2.43893 iter/s, 4.92019s/12 iters), loss = 0.0142846
I0407 23:46:21.302640 32718 solver.cpp:237]     Train net output #0: loss = 0.0142846 (* 1 = 0.0142846 loss)
I0407 23:46:21.302649 32718 sgd_solver.cpp:105] Iteration 9684, lr = 0.000956035
I0407 23:46:23.331111 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_9690.caffemodel
I0407 23:46:26.426443 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_9690.solverstate
I0407 23:46:28.843072 32718 solver.cpp:330] Iteration 9690, Testing net (#0)
I0407 23:46:28.843165 32718 net.cpp:676] Ignoring source layer train-data
I0407 23:46:29.431917 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:46:32.327981 32718 blocking_queue.cpp:49] Waiting for data
I0407 23:46:33.417241 32718 solver.cpp:397]     Test net output #0: accuracy = 0.530637
I0407 23:46:33.417289 32718 solver.cpp:397]     Test net output #1: loss = 2.82594 (* 1 = 2.82594 loss)
I0407 23:46:35.253907 32718 solver.cpp:218] Iteration 9696 (0.86014 iter/s, 13.9512s/12 iters), loss = 0.0387199
I0407 23:46:35.253952 32718 solver.cpp:237]     Train net output #0: loss = 0.0387199 (* 1 = 0.0387199 loss)
I0407 23:46:35.253960 32718 sgd_solver.cpp:105] Iteration 9696, lr = 0.000950961
I0407 23:46:40.188733 32718 solver.cpp:218] Iteration 9708 (2.43173 iter/s, 4.93476s/12 iters), loss = 0.0589518
I0407 23:46:40.188769 32718 solver.cpp:237]     Train net output #0: loss = 0.0589518 (* 1 = 0.0589518 loss)
I0407 23:46:40.188777 32718 sgd_solver.cpp:105] Iteration 9708, lr = 0.000945911
I0407 23:46:40.901515 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:46:45.131048 32718 solver.cpp:218] Iteration 9720 (2.42804 iter/s, 4.94225s/12 iters), loss = 0.0709656
I0407 23:46:45.131089 32718 solver.cpp:237]     Train net output #0: loss = 0.0709656 (* 1 = 0.0709656 loss)
I0407 23:46:45.131098 32718 sgd_solver.cpp:105] Iteration 9720, lr = 0.000940885
I0407 23:46:50.061707 32718 solver.cpp:218] Iteration 9732 (2.43379 iter/s, 4.93059s/12 iters), loss = 0.00944918
I0407 23:46:50.061744 32718 solver.cpp:237]     Train net output #0: loss = 0.00944921 (* 1 = 0.00944921 loss)
I0407 23:46:50.061753 32718 sgd_solver.cpp:105] Iteration 9732, lr = 0.000935883
I0407 23:46:55.006196 32718 solver.cpp:218] Iteration 9744 (2.42698 iter/s, 4.94442s/12 iters), loss = 0.0036025
I0407 23:46:55.006234 32718 solver.cpp:237]     Train net output #0: loss = 0.00360253 (* 1 = 0.00360253 loss)
I0407 23:46:55.006242 32718 sgd_solver.cpp:105] Iteration 9744, lr = 0.000930905
I0407 23:46:59.925755 32718 solver.cpp:218] Iteration 9756 (2.43928 iter/s, 4.91949s/12 iters), loss = 0.0469583
I0407 23:46:59.925904 32718 solver.cpp:237]     Train net output #0: loss = 0.0469583 (* 1 = 0.0469583 loss)
I0407 23:46:59.925913 32718 sgd_solver.cpp:105] Iteration 9756, lr = 0.00092595
I0407 23:47:04.878888 32718 solver.cpp:218] Iteration 9768 (2.42279 iter/s, 4.95296s/12 iters), loss = 0.0298527
I0407 23:47:04.878927 32718 solver.cpp:237]     Train net output #0: loss = 0.0298528 (* 1 = 0.0298528 loss)
I0407 23:47:04.878937 32718 sgd_solver.cpp:105] Iteration 9768, lr = 0.00092102
I0407 23:47:09.822983 32718 solver.cpp:218] Iteration 9780 (2.42717 iter/s, 4.94402s/12 iters), loss = 0.00533896
I0407 23:47:09.823025 32718 solver.cpp:237]     Train net output #0: loss = 0.005339 (* 1 = 0.005339 loss)
I0407 23:47:09.823033 32718 sgd_solver.cpp:105] Iteration 9780, lr = 0.000916113
I0407 23:47:14.282982 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_9792.caffemodel
I0407 23:47:17.137629 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_9792.solverstate
I0407 23:47:19.504146 32718 solver.cpp:330] Iteration 9792, Testing net (#0)
I0407 23:47:19.504163 32718 net.cpp:676] Ignoring source layer train-data
I0407 23:47:20.044433 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:47:23.948315 32718 solver.cpp:397]     Test net output #0: accuracy = 0.525123
I0407 23:47:23.948359 32718 solver.cpp:397]     Test net output #1: loss = 2.80552 (* 1 = 2.80552 loss)
I0407 23:47:24.043119 32718 solver.cpp:218] Iteration 9792 (0.84388 iter/s, 14.22s/12 iters), loss = 0.0119279
I0407 23:47:24.043165 32718 solver.cpp:237]     Train net output #0: loss = 0.0119279 (* 1 = 0.0119279 loss)
I0407 23:47:24.043174 32718 sgd_solver.cpp:105] Iteration 9792, lr = 0.000911229
I0407 23:47:28.164856 32718 solver.cpp:218] Iteration 9804 (2.91144 iter/s, 4.12167s/12 iters), loss = 0.0275094
I0407 23:47:28.164892 32718 solver.cpp:237]     Train net output #0: loss = 0.0275095 (* 1 = 0.0275095 loss)
I0407 23:47:28.164901 32718 sgd_solver.cpp:105] Iteration 9804, lr = 0.000906369
I0407 23:47:31.089529 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:47:33.074146 32718 solver.cpp:218] Iteration 9816 (2.44438 iter/s, 4.90923s/12 iters), loss = 0.0195011
I0407 23:47:33.074184 32718 solver.cpp:237]     Train net output #0: loss = 0.0195011 (* 1 = 0.0195011 loss)
I0407 23:47:33.074193 32718 sgd_solver.cpp:105] Iteration 9816, lr = 0.000901533
I0407 23:47:38.032040 32718 solver.cpp:218] Iteration 9828 (2.42041 iter/s, 4.95783s/12 iters), loss = 0.0180341
I0407 23:47:38.032074 32718 solver.cpp:237]     Train net output #0: loss = 0.0180341 (* 1 = 0.0180341 loss)
I0407 23:47:38.032084 32718 sgd_solver.cpp:105] Iteration 9828, lr = 0.000896719
I0407 23:47:42.950951 32718 solver.cpp:218] Iteration 9840 (2.4396 iter/s, 4.91884s/12 iters), loss = 0.0168049
I0407 23:47:42.950989 32718 solver.cpp:237]     Train net output #0: loss = 0.0168049 (* 1 = 0.0168049 loss)
I0407 23:47:42.950997 32718 sgd_solver.cpp:105] Iteration 9840, lr = 0.000891929
I0407 23:47:47.902271 32718 solver.cpp:218] Iteration 9852 (2.42363 iter/s, 4.95125s/12 iters), loss = 0.0448971
I0407 23:47:47.902307 32718 solver.cpp:237]     Train net output #0: loss = 0.0448971 (* 1 = 0.0448971 loss)
I0407 23:47:47.902313 32718 sgd_solver.cpp:105] Iteration 9852, lr = 0.000887162
I0407 23:47:52.830657 32718 solver.cpp:218] Iteration 9864 (2.4349 iter/s, 4.92833s/12 iters), loss = 0.0377048
I0407 23:47:52.830693 32718 solver.cpp:237]     Train net output #0: loss = 0.0377049 (* 1 = 0.0377049 loss)
I0407 23:47:52.830699 32718 sgd_solver.cpp:105] Iteration 9864, lr = 0.000882418
I0407 23:47:57.837020 32718 solver.cpp:218] Iteration 9876 (2.39698 iter/s, 5.0063s/12 iters), loss = 0.00539715
I0407 23:47:57.837060 32718 solver.cpp:237]     Train net output #0: loss = 0.00539719 (* 1 = 0.00539719 loss)
I0407 23:47:57.837069 32718 sgd_solver.cpp:105] Iteration 9876, lr = 0.000877697
I0407 23:48:02.804754 32718 solver.cpp:218] Iteration 9888 (2.41562 iter/s, 4.96767s/12 iters), loss = 0.0555708
I0407 23:48:02.804905 32718 solver.cpp:237]     Train net output #0: loss = 0.0555709 (* 1 = 0.0555709 loss)
I0407 23:48:02.804915 32718 sgd_solver.cpp:105] Iteration 9888, lr = 0.000872998
I0407 23:48:04.765662 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_9894.caffemodel
I0407 23:48:07.896140 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_9894.solverstate
I0407 23:48:10.271648 32718 solver.cpp:330] Iteration 9894, Testing net (#0)
I0407 23:48:10.271667 32718 net.cpp:676] Ignoring source layer train-data
I0407 23:48:10.819119 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:48:15.090669 32718 solver.cpp:397]     Test net output #0: accuracy = 0.528799
I0407 23:48:15.090714 32718 solver.cpp:397]     Test net output #1: loss = 2.79589 (* 1 = 2.79589 loss)
I0407 23:48:16.901855 32718 solver.cpp:218] Iteration 9900 (0.851251 iter/s, 14.0969s/12 iters), loss = 0.0274475
I0407 23:48:16.901901 32718 solver.cpp:237]     Train net output #0: loss = 0.0274475 (* 1 = 0.0274475 loss)
I0407 23:48:16.901909 32718 sgd_solver.cpp:105] Iteration 9900, lr = 0.000868323
I0407 23:48:21.868470 32718 solver.cpp:218] Iteration 9912 (2.41617 iter/s, 4.96655s/12 iters), loss = 0.0954231
I0407 23:48:21.868506 32718 solver.cpp:237]     Train net output #0: loss = 0.0954232 (* 1 = 0.0954232 loss)
I0407 23:48:21.868513 32718 sgd_solver.cpp:105] Iteration 9912, lr = 0.00086367
I0407 23:48:21.960600 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:48:26.755844 32718 solver.cpp:218] Iteration 9924 (2.45534 iter/s, 4.88731s/12 iters), loss = 0.0261711
I0407 23:48:26.755887 32718 solver.cpp:237]     Train net output #0: loss = 0.0261711 (* 1 = 0.0261711 loss)
I0407 23:48:26.755897 32718 sgd_solver.cpp:105] Iteration 9924, lr = 0.000859039
I0407 23:48:31.749814 32718 solver.cpp:218] Iteration 9936 (2.40293 iter/s, 4.9939s/12 iters), loss = 0.0490789
I0407 23:48:31.749861 32718 solver.cpp:237]     Train net output #0: loss = 0.0490789 (* 1 = 0.0490789 loss)
I0407 23:48:31.749871 32718 sgd_solver.cpp:105] Iteration 9936, lr = 0.000854432
I0407 23:48:36.665983 32718 solver.cpp:218] Iteration 9948 (2.44096 iter/s, 4.91609s/12 iters), loss = 0.0320697
I0407 23:48:36.666121 32718 solver.cpp:237]     Train net output #0: loss = 0.0320697 (* 1 = 0.0320697 loss)
I0407 23:48:36.666131 32718 sgd_solver.cpp:105] Iteration 9948, lr = 0.000849846
I0407 23:48:41.550279 32718 solver.cpp:218] Iteration 9960 (2.45694 iter/s, 4.88413s/12 iters), loss = 0.0269977
I0407 23:48:41.550323 32718 solver.cpp:237]     Train net output #0: loss = 0.0269978 (* 1 = 0.0269978 loss)
I0407 23:48:41.550331 32718 sgd_solver.cpp:105] Iteration 9960, lr = 0.000845283
I0407 23:48:46.491263 32718 solver.cpp:218] Iteration 9972 (2.4287 iter/s, 4.94091s/12 iters), loss = 0.0114788
I0407 23:48:46.491307 32718 solver.cpp:237]     Train net output #0: loss = 0.0114788 (* 1 = 0.0114788 loss)
I0407 23:48:46.491315 32718 sgd_solver.cpp:105] Iteration 9972, lr = 0.000840742
I0407 23:48:51.460572 32718 solver.cpp:218] Iteration 9984 (2.41486 iter/s, 4.96923s/12 iters), loss = 0.0830278
I0407 23:48:51.460618 32718 solver.cpp:237]     Train net output #0: loss = 0.0830279 (* 1 = 0.0830279 loss)
I0407 23:48:51.460625 32718 sgd_solver.cpp:105] Iteration 9984, lr = 0.000836223
I0407 23:48:55.913259 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_9996.caffemodel
I0407 23:48:58.985496 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_9996.solverstate
I0407 23:49:01.354072 32718 solver.cpp:330] Iteration 9996, Testing net (#0)
I0407 23:49:01.354089 32718 net.cpp:676] Ignoring source layer train-data
I0407 23:49:01.829361 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:49:06.010347 32718 solver.cpp:397]     Test net output #0: accuracy = 0.530025
I0407 23:49:06.010392 32718 solver.cpp:397]     Test net output #1: loss = 2.8151 (* 1 = 2.8151 loss)
I0407 23:49:06.106715 32718 solver.cpp:218] Iteration 9996 (0.819334 iter/s, 14.646s/12 iters), loss = 0.030846
I0407 23:49:06.106756 32718 solver.cpp:237]     Train net output #0: loss = 0.0308461 (* 1 = 0.0308461 loss)
I0407 23:49:06.106765 32718 sgd_solver.cpp:105] Iteration 9996, lr = 0.000831727
I0407 23:49:10.155211 32718 solver.cpp:218] Iteration 10008 (2.96412 iter/s, 4.04842s/12 iters), loss = 0.0255459
I0407 23:49:10.155325 32718 solver.cpp:237]     Train net output #0: loss = 0.0255459 (* 1 = 0.0255459 loss)
I0407 23:49:10.155335 32718 sgd_solver.cpp:105] Iteration 10008, lr = 0.000827252
I0407 23:49:12.349050 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:49:15.072634 32718 solver.cpp:218] Iteration 10020 (2.44037 iter/s, 4.91728s/12 iters), loss = 0.00413825
I0407 23:49:15.072674 32718 solver.cpp:237]     Train net output #0: loss = 0.00413828 (* 1 = 0.00413828 loss)
I0407 23:49:15.072682 32718 sgd_solver.cpp:105] Iteration 10020, lr = 0.0008228
I0407 23:49:20.096818 32718 solver.cpp:218] Iteration 10032 (2.38848 iter/s, 5.02411s/12 iters), loss = 0.123191
I0407 23:49:20.096861 32718 solver.cpp:237]     Train net output #0: loss = 0.123191 (* 1 = 0.123191 loss)
I0407 23:49:20.096869 32718 sgd_solver.cpp:105] Iteration 10032, lr = 0.000818369
I0407 23:49:25.047973 32718 solver.cpp:218] Iteration 10044 (2.42372 iter/s, 4.95108s/12 iters), loss = 0.044477
I0407 23:49:25.048018 32718 solver.cpp:237]     Train net output #0: loss = 0.0444771 (* 1 = 0.0444771 loss)
I0407 23:49:25.048027 32718 sgd_solver.cpp:105] Iteration 10044, lr = 0.00081396
I0407 23:49:29.969295 32718 solver.cpp:218] Iteration 10056 (2.43841 iter/s, 4.92125s/12 iters), loss = 0.020237
I0407 23:49:29.969347 32718 solver.cpp:237]     Train net output #0: loss = 0.020237 (* 1 = 0.020237 loss)
I0407 23:49:29.969354 32718 sgd_solver.cpp:105] Iteration 10056, lr = 0.000809572
I0407 23:49:34.934468 32718 solver.cpp:218] Iteration 10068 (2.41687 iter/s, 4.96509s/12 iters), loss = 0.0674384
I0407 23:49:34.934509 32718 solver.cpp:237]     Train net output #0: loss = 0.0674384 (* 1 = 0.0674384 loss)
I0407 23:49:34.934518 32718 sgd_solver.cpp:105] Iteration 10068, lr = 0.000805206
I0407 23:49:39.868515 32718 solver.cpp:218] Iteration 10080 (2.43212 iter/s, 4.93398s/12 iters), loss = 0.0161841
I0407 23:49:39.868556 32718 solver.cpp:237]     Train net output #0: loss = 0.0161841 (* 1 = 0.0161841 loss)
I0407 23:49:39.868564 32718 sgd_solver.cpp:105] Iteration 10080, lr = 0.000800862
I0407 23:49:44.813542 32718 solver.cpp:218] Iteration 10092 (2.42672 iter/s, 4.94496s/12 iters), loss = 0.049747
I0407 23:49:44.813673 32718 solver.cpp:237]     Train net output #0: loss = 0.049747 (* 1 = 0.049747 loss)
I0407 23:49:44.813681 32718 sgd_solver.cpp:105] Iteration 10092, lr = 0.000796539
I0407 23:49:46.819715 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_10098.caffemodel
I0407 23:49:49.956627 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_10098.solverstate
I0407 23:49:52.354790 32718 solver.cpp:330] Iteration 10098, Testing net (#0)
I0407 23:49:52.354807 32718 net.cpp:676] Ignoring source layer train-data
I0407 23:49:52.759898 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:49:57.080415 32718 solver.cpp:397]     Test net output #0: accuracy = 0.534926
I0407 23:49:57.080459 32718 solver.cpp:397]     Test net output #1: loss = 2.7924 (* 1 = 2.7924 loss)
I0407 23:49:58.888387 32718 solver.cpp:218] Iteration 10104 (0.852596 iter/s, 14.0747s/12 iters), loss = 0.0350366
I0407 23:49:58.888435 32718 solver.cpp:237]     Train net output #0: loss = 0.0350367 (* 1 = 0.0350367 loss)
I0407 23:49:58.888443 32718 sgd_solver.cpp:105] Iteration 10104, lr = 0.000792237
I0407 23:50:03.278091 32725 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:50:03.890576 32718 solver.cpp:218] Iteration 10116 (2.39898 iter/s, 5.00212s/12 iters), loss = 0.0464235
I0407 23:50:03.890609 32718 solver.cpp:237]     Train net output #0: loss = 0.0464236 (* 1 = 0.0464236 loss)
I0407 23:50:03.890616 32718 sgd_solver.cpp:105] Iteration 10116, lr = 0.000787957
I0407 23:50:08.795544 32718 solver.cpp:218] Iteration 10128 (2.44653 iter/s, 4.90491s/12 iters), loss = 0.0182167
I0407 23:50:08.795580 32718 solver.cpp:237]     Train net output #0: loss = 0.0182168 (* 1 = 0.0182168 loss)
I0407 23:50:08.795588 32718 sgd_solver.cpp:105] Iteration 10128, lr = 0.000783698
I0407 23:50:13.628290 32718 solver.cpp:218] Iteration 10140 (2.48309 iter/s, 4.83269s/12 iters), loss = 0.0469299
I0407 23:50:13.628324 32718 solver.cpp:237]     Train net output #0: loss = 0.0469299 (* 1 = 0.0469299 loss)
I0407 23:50:13.628331 32718 sgd_solver.cpp:105] Iteration 10140, lr = 0.000779459
I0407 23:50:18.446642 32718 solver.cpp:218] Iteration 10152 (2.49051 iter/s, 4.81829s/12 iters), loss = 0.0719232
I0407 23:50:18.446786 32718 solver.cpp:237]     Train net output #0: loss = 0.0719232 (* 1 = 0.0719232 loss)
I0407 23:50:18.446795 32718 sgd_solver.cpp:105] Iteration 10152, lr = 0.000775242
I0407 23:50:23.279631 32718 solver.cpp:218] Iteration 10164 (2.48302 iter/s, 4.83282s/12 iters), loss = 0.0898683
I0407 23:50:23.279665 32718 solver.cpp:237]     Train net output #0: loss = 0.0898683 (* 1 = 0.0898683 loss)
I0407 23:50:23.279673 32718 sgd_solver.cpp:105] Iteration 10164, lr = 0.000771046
I0407 23:50:28.095840 32718 solver.cpp:218] Iteration 10176 (2.49162 iter/s, 4.81615s/12 iters), loss = 0.0536779
I0407 23:50:28.095877 32718 solver.cpp:237]     Train net output #0: loss = 0.053678 (* 1 = 0.053678 loss)
I0407 23:50:28.095885 32718 sgd_solver.cpp:105] Iteration 10176, lr = 0.00076687
I0407 23:50:32.928413 32718 solver.cpp:218] Iteration 10188 (2.48318 iter/s, 4.83251s/12 iters), loss = 0.0339025
I0407 23:50:32.928448 32718 solver.cpp:237]     Train net output #0: loss = 0.0339026 (* 1 = 0.0339026 loss)
I0407 23:50:32.928457 32718 sgd_solver.cpp:105] Iteration 10188, lr = 0.000762716
I0407 23:50:37.338142 32718 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_10200.caffemodel
I0407 23:50:40.418303 32718 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_10200.solverstate
I0407 23:50:42.814378 32718 solver.cpp:310] Iteration 10200, loss = 0.00940958
I0407 23:50:42.814400 32718 solver.cpp:330] Iteration 10200, Testing net (#0)
I0407 23:50:42.814405 32718 net.cpp:676] Ignoring source layer train-data
I0407 23:50:43.212003 32736 data_layer.cpp:73] Restarting data prefetching from start.
I0407 23:50:47.592864 32718 solver.cpp:397]     Test net output #0: accuracy = 0.529412
I0407 23:50:47.592909 32718 solver.cpp:397]     Test net output #1: loss = 2.79391 (* 1 = 2.79391 loss)
I0407 23:50:47.592921 32718 solver.cpp:315] Optimization Done.
I0407 23:50:47.592931 32718 caffe.cpp:259] Optimization Done.