I0409 23:04:49.787638 6396 upgrade_proto.cpp:1082] Attempting to upgrade input file specified using deprecated 'solver_type' field (enum)': /mnt/bigdisk/DIGITS-MAN-2/digits/jobs/20210409-210951-7f10/solver.prototxt I0409 23:04:49.787843 6396 upgrade_proto.cpp:1089] Successfully upgraded file specified using deprecated 'solver_type' field (enum) to 'type' field (string). W0409 23:04:49.787851 6396 upgrade_proto.cpp:1091] Note that future Caffe releases will only support 'type' field (string) for a solver's type. I0409 23:04:49.787930 6396 caffe.cpp:218] Using GPUs 1 I0409 23:04:49.846495 6396 caffe.cpp:223] GPU 1: GeForce GTX 1080 Ti I0409 23:04:50.139474 6396 solver.cpp:44] Initializing solver from parameters: test_iter: 51 test_interval: 102 base_lr: 0.01 display: 12 max_iter: 10200 lr_policy: "exp" gamma: 0.99980193 momentum: 0.9 weight_decay: 0.0001 snapshot: 102 snapshot_prefix: "snapshot" solver_mode: GPU device_id: 1 net: "train_val.prototxt" train_state { level: 0 stage: "" } type: "SGD" I0409 23:04:50.218997 6396 solver.cpp:87] Creating training net from net file: train_val.prototxt I0409 23:04:50.270009 6396 net.cpp:294] The NetState phase (0) differed from the phase (1) specified by a rule in layer val-data I0409 23:04:50.270038 6396 net.cpp:294] The NetState phase (0) differed from the phase (1) specified by a rule in layer accuracy I0409 23:04:50.270239 6396 net.cpp:51] Initializing net from parameters: state { phase: TRAIN level: 0 stage: "" } layer { name: "train-data" type: "Data" top: "data" top: "label" include { phase: TRAIN } transform_param { mirror: true crop_size: 227 mean_file: "/mnt/bigdisk/DIGITS-MAN-2/digits/jobs/20210407-214532-d396/mean.binaryproto" } data_param { source: "/mnt/bigdisk/DIGITS-MAN-2/digits/jobs/20210407-214532-d396/train_db" batch_size: 128 backend: LMDB } } layer { name: "conv1" type: "Convolution" bottom: "data" top: "conv1" param { lr_mult: 1 decay_mult: 1 } param { lr_mult: 2 decay_mult: 0 } convolution_param { num_output: 96 kernel_size: 11 stride: 4 weight_filler { type: "gaussian" std: 0.01 } bias_filler { type: "constant" value: 0 } } } layer { name: "relu1" type: "ReLU" bottom: "conv1" top: "conv1" } layer { name: "norm1" type: "LRN" bottom: "conv1" top: "norm1" lrn_param { local_size: 5 alpha: 0.0001 beta: 0.75 } } layer { name: "pool1" type: "Pooling" bottom: "norm1" top: "pool1" pooling_param { pool: MAX kernel_size: 3 stride: 2 } } layer { name: "conv2" type: "Convolution" bottom: "pool1" top: "conv2" param { lr_mult: 1 decay_mult: 1 } param { lr_mult: 2 decay_mult: 0 } convolution_param { num_output: 256 pad: 2 kernel_size: 5 group: 2 weight_filler { type: "gaussian" std: 0.01 } bias_filler { type: "constant" value: 0.1 } } } layer { name: "relu2" type: "ReLU" bottom: "conv2" top: "conv2" } layer { name: "norm2" type: "LRN" bottom: "conv2" top: "norm2" lrn_param { local_size: 5 alpha: 0.0001 beta: 0.75 } } layer { name: "pool2" type: "Pooling" bottom: "norm2" top: "pool2" pooling_param { pool: MAX kernel_size: 3 stride: 2 } } layer { name: "conv3" type: "Convolution" bottom: "pool2" top: "conv3" param { lr_mult: 1 decay_mult: 1 } param { lr_mult: 2 decay_mult: 0 } convolution_param { num_output: 384 pad: 1 kernel_size: 3 weight_filler { type: "gaussian" std: 0.01 } bias_filler { type: "constant" value: 0 } } } layer { name: "relu3" type: "ReLU" bottom: "conv3" top: "conv3" } layer { name: "conv4" type: "Convolution" bottom: "conv3" top: "conv4" param { lr_mult: 1 decay_mult: 1 } param { lr_mult: 2 decay_mult: 0 } convolution_param { num_output: 384 pad: 1 kernel_size: 3 group: 2 weight_filler { type: "gaussian" std: 0.01 } bias_filler { type: "constant" value: 0.1 } } } layer { name: "relu4" type: "ReLU" bottom: "conv4" top: "conv4" } layer { name: "conv5" type: "Convolution" bottom: "conv4" top: "conv5" param { lr_mult: 1 decay_mult: 1 } param { lr_mult: 2 decay_mult: 0 } convolution_param { num_output: 256 pad: 1 kernel_size: 3 group: 2 weight_filler { type: "gaussian" std: 0.01 } bias_filler { type: "constant" value: 0.1 } } } layer { name: "relu5" type: "ReLU" bottom: "conv5" top: "conv5" } layer { name: "pool5" type: "Pooling" bottom: "conv5" top: "pool5" pooling_param { pool: MAX kernel_size: 3 stride: 2 } } layer { name: "fc6" type: "InnerProduct" bottom: "pool5" top: "fc6" param { lr_mult: 1 decay_mult: 1 } param { lr_mult: 2 decay_mult: 0 } inner_product_param { num_output: 512 weight_filler { type: "gaussian" std: 0.005 } bias_filler { type: "constant" value: 0.1 } } } layer { name: "relu6" type: "ReLU" bottom: "fc6" top: "fc6" } layer { name: "drop6" type: "Dropout" bottom: "fc6" top: "fc6" dropout_param { dropout_ratio: 0.5 } } layer { name: "fc7" type: "InnerProduct" bottom: "fc6" top: "fc7" param { lr_mult: 1 decay_mult: 1 } param { lr_mult: 2 decay_mult: 0 } inner_product_param { num_output: 512 weight_filler { type: "gaussian" std: 0.005 } bias_filler { type: "constant" value: 0.1 } } } layer { name: "relu7" type: "ReLU" bottom: "fc7" top: "fc7" } layer { name: "drop7" type: "Dropout" bottom: "fc7" top: "fc7" dropout_param { dropout_ratio: 0.5 } } layer { name: "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" } I0409 23:04:50.270352 6396 layer_factory.hpp:77] Creating layer train-data I0409 23:04:50.516469 6396 db_lmdb.cpp:35] Opened lmdb /mnt/bigdisk/DIGITS-MAN-2/digits/jobs/20210407-214532-d396/train_db I0409 23:04:50.517014 6396 net.cpp:84] Creating Layer train-data I0409 23:04:50.517041 6396 net.cpp:380] train-data -> data I0409 23:04:50.517081 6396 net.cpp:380] train-data -> label I0409 23:04:50.517105 6396 data_transformer.cpp:25] Loading mean file from: /mnt/bigdisk/DIGITS-MAN-2/digits/jobs/20210407-214532-d396/mean.binaryproto I0409 23:04:50.590185 6396 data_layer.cpp:45] output data size: 128,3,227,227 I0409 23:04:50.759325 6396 net.cpp:122] Setting up train-data I0409 23:04:50.759351 6396 net.cpp:129] Top shape: 128 3 227 227 (19787136) I0409 23:04:50.759356 6396 net.cpp:129] Top shape: 128 (128) I0409 23:04:50.759359 6396 net.cpp:137] Memory required for data: 79149056 I0409 23:04:50.759371 6396 layer_factory.hpp:77] Creating layer conv1 I0409 23:04:50.759392 6396 net.cpp:84] Creating Layer conv1 I0409 23:04:50.759399 6396 net.cpp:406] conv1 <- data I0409 23:04:50.759413 6396 net.cpp:380] conv1 -> conv1 I0409 23:04:51.343858 6396 net.cpp:122] Setting up conv1 I0409 23:04:51.343883 6396 net.cpp:129] Top shape: 128 96 55 55 (37171200) I0409 23:04:51.343888 6396 net.cpp:137] Memory required for data: 227833856 I0409 23:04:51.343909 6396 layer_factory.hpp:77] Creating layer relu1 I0409 23:04:51.343919 6396 net.cpp:84] Creating Layer relu1 I0409 23:04:51.343924 6396 net.cpp:406] relu1 <- conv1 I0409 23:04:51.343931 6396 net.cpp:367] relu1 -> conv1 (in-place) I0409 23:04:51.344242 6396 net.cpp:122] Setting up relu1 I0409 23:04:51.344252 6396 net.cpp:129] Top shape: 128 96 55 55 (37171200) I0409 23:04:51.344256 6396 net.cpp:137] Memory required for data: 376518656 I0409 23:04:51.344260 6396 layer_factory.hpp:77] Creating layer norm1 I0409 23:04:51.344269 6396 net.cpp:84] Creating Layer norm1 I0409 23:04:51.344274 6396 net.cpp:406] norm1 <- conv1 I0409 23:04:51.344305 6396 net.cpp:380] norm1 -> norm1 I0409 23:04:51.344785 6396 net.cpp:122] Setting up norm1 I0409 23:04:51.344797 6396 net.cpp:129] Top shape: 128 96 55 55 (37171200) I0409 23:04:51.344802 6396 net.cpp:137] Memory required for data: 525203456 I0409 23:04:51.344808 6396 layer_factory.hpp:77] Creating layer pool1 I0409 23:04:51.344816 6396 net.cpp:84] Creating Layer pool1 I0409 23:04:51.344821 6396 net.cpp:406] pool1 <- norm1 I0409 23:04:51.344828 6396 net.cpp:380] pool1 -> pool1 I0409 23:04:51.344867 6396 net.cpp:122] Setting up pool1 I0409 23:04:51.344875 6396 net.cpp:129] Top shape: 128 96 27 27 (8957952) I0409 23:04:51.344879 6396 net.cpp:137] Memory required for data: 561035264 I0409 23:04:51.344883 6396 layer_factory.hpp:77] Creating layer conv2 I0409 23:04:51.344894 6396 net.cpp:84] Creating Layer conv2 I0409 23:04:51.344897 6396 net.cpp:406] conv2 <- pool1 I0409 23:04:51.344903 6396 net.cpp:380] conv2 -> conv2 I0409 23:04:51.351981 6396 net.cpp:122] Setting up conv2 I0409 23:04:51.351996 6396 net.cpp:129] Top shape: 128 256 27 27 (23887872) I0409 23:04:51.352000 6396 net.cpp:137] Memory required for data: 656586752 I0409 23:04:51.352010 6396 layer_factory.hpp:77] Creating layer relu2 I0409 23:04:51.352017 6396 net.cpp:84] Creating Layer relu2 I0409 23:04:51.352021 6396 net.cpp:406] relu2 <- conv2 I0409 23:04:51.352028 6396 net.cpp:367] relu2 -> conv2 (in-place) I0409 23:04:51.352473 6396 net.cpp:122] Setting up relu2 I0409 23:04:51.352484 6396 net.cpp:129] Top shape: 128 256 27 27 (23887872) I0409 23:04:51.352488 6396 net.cpp:137] Memory required for data: 752138240 I0409 23:04:51.352494 6396 layer_factory.hpp:77] Creating layer norm2 I0409 23:04:51.352501 6396 net.cpp:84] Creating Layer norm2 I0409 23:04:51.352506 6396 net.cpp:406] norm2 <- conv2 I0409 23:04:51.352514 6396 net.cpp:380] norm2 -> norm2 I0409 23:04:51.352819 6396 net.cpp:122] Setting up norm2 I0409 23:04:51.352829 6396 net.cpp:129] Top shape: 128 256 27 27 (23887872) I0409 23:04:51.352833 6396 net.cpp:137] Memory required for data: 847689728 I0409 23:04:51.352838 6396 layer_factory.hpp:77] Creating layer pool2 I0409 23:04:51.352845 6396 net.cpp:84] Creating Layer pool2 I0409 23:04:51.352849 6396 net.cpp:406] pool2 <- norm2 I0409 23:04:51.352854 6396 net.cpp:380] pool2 -> pool2 I0409 23:04:51.352882 6396 net.cpp:122] Setting up pool2 I0409 23:04:51.352888 6396 net.cpp:129] Top shape: 128 256 13 13 (5537792) I0409 23:04:51.352891 6396 net.cpp:137] Memory required for data: 869840896 I0409 23:04:51.352895 6396 layer_factory.hpp:77] Creating layer conv3 I0409 23:04:51.352903 6396 net.cpp:84] Creating Layer conv3 I0409 23:04:51.352908 6396 net.cpp:406] conv3 <- pool2 I0409 23:04:51.352914 6396 net.cpp:380] conv3 -> conv3 I0409 23:04:51.363785 6396 net.cpp:122] Setting up conv3 I0409 23:04:51.363798 6396 net.cpp:129] Top shape: 128 384 13 13 (8306688) I0409 23:04:51.363802 6396 net.cpp:137] Memory required for data: 903067648 I0409 23:04:51.363812 6396 layer_factory.hpp:77] Creating layer relu3 I0409 23:04:51.363821 6396 net.cpp:84] Creating Layer relu3 I0409 23:04:51.363826 6396 net.cpp:406] relu3 <- conv3 I0409 23:04:51.363832 6396 net.cpp:367] relu3 -> conv3 (in-place) I0409 23:04:51.364352 6396 net.cpp:122] Setting up relu3 I0409 23:04:51.364362 6396 net.cpp:129] Top shape: 128 384 13 13 (8306688) I0409 23:04:51.364365 6396 net.cpp:137] Memory required for data: 936294400 I0409 23:04:51.364369 6396 layer_factory.hpp:77] Creating layer conv4 I0409 23:04:51.364380 6396 net.cpp:84] Creating Layer conv4 I0409 23:04:51.364384 6396 net.cpp:406] conv4 <- conv3 I0409 23:04:51.364392 6396 net.cpp:380] conv4 -> conv4 I0409 23:04:51.375677 6396 net.cpp:122] Setting up conv4 I0409 23:04:51.375695 6396 net.cpp:129] Top shape: 128 384 13 13 (8306688) I0409 23:04:51.375699 6396 net.cpp:137] Memory required for data: 969521152 I0409 23:04:51.375708 6396 layer_factory.hpp:77] Creating layer relu4 I0409 23:04:51.375716 6396 net.cpp:84] Creating Layer relu4 I0409 23:04:51.375739 6396 net.cpp:406] relu4 <- conv4 I0409 23:04:51.375747 6396 net.cpp:367] relu4 -> conv4 (in-place) I0409 23:04:51.376114 6396 net.cpp:122] Setting up relu4 I0409 23:04:51.376124 6396 net.cpp:129] Top shape: 128 384 13 13 (8306688) I0409 23:04:51.376128 6396 net.cpp:137] Memory required for data: 1002747904 I0409 23:04:51.376132 6396 layer_factory.hpp:77] Creating layer conv5 I0409 23:04:51.376143 6396 net.cpp:84] Creating Layer conv5 I0409 23:04:51.376147 6396 net.cpp:406] conv5 <- conv4 I0409 23:04:51.376157 6396 net.cpp:380] conv5 -> conv5 I0409 23:04:51.390148 6396 net.cpp:122] Setting up conv5 I0409 23:04:51.390172 6396 net.cpp:129] Top shape: 128 256 13 13 (5537792) I0409 23:04:51.390177 6396 net.cpp:137] Memory required for data: 1024899072 I0409 23:04:51.390206 6396 layer_factory.hpp:77] Creating layer relu5 I0409 23:04:51.390223 6396 net.cpp:84] Creating Layer relu5 I0409 23:04:51.390233 6396 net.cpp:406] relu5 <- conv5 I0409 23:04:51.390249 6396 net.cpp:367] relu5 -> conv5 (in-place) I0409 23:04:51.391422 6396 net.cpp:122] Setting up relu5 I0409 23:04:51.391443 6396 net.cpp:129] Top shape: 128 256 13 13 (5537792) I0409 23:04:51.391450 6396 net.cpp:137] Memory required for data: 1047050240 I0409 23:04:51.391458 6396 layer_factory.hpp:77] Creating layer pool5 I0409 23:04:51.391475 6396 net.cpp:84] Creating Layer pool5 I0409 23:04:51.391484 6396 net.cpp:406] pool5 <- conv5 I0409 23:04:51.391496 6396 net.cpp:380] pool5 -> pool5 I0409 23:04:51.391587 6396 net.cpp:122] Setting up pool5 I0409 23:04:51.391599 6396 net.cpp:129] Top shape: 128 256 6 6 (1179648) I0409 23:04:51.391606 6396 net.cpp:137] Memory required for data: 1051768832 I0409 23:04:51.391613 6396 layer_factory.hpp:77] Creating layer fc6 I0409 23:04:51.391633 6396 net.cpp:84] Creating Layer fc6 I0409 23:04:51.391640 6396 net.cpp:406] fc6 <- pool5 I0409 23:04:51.391652 6396 net.cpp:380] fc6 -> fc6 I0409 23:04:51.486677 6396 net.cpp:122] Setting up fc6 I0409 23:04:51.486703 6396 net.cpp:129] Top shape: 128 512 (65536) I0409 23:04:51.486708 6396 net.cpp:137] Memory required for data: 1052030976 I0409 23:04:51.486721 6396 layer_factory.hpp:77] Creating layer relu6 I0409 23:04:51.486733 6396 net.cpp:84] Creating Layer relu6 I0409 23:04:51.486739 6396 net.cpp:406] relu6 <- fc6 I0409 23:04:51.486748 6396 net.cpp:367] relu6 -> fc6 (in-place) I0409 23:04:51.487648 6396 net.cpp:122] Setting up relu6 I0409 23:04:51.487660 6396 net.cpp:129] Top shape: 128 512 (65536) I0409 23:04:51.487666 6396 net.cpp:137] Memory required for data: 1052293120 I0409 23:04:51.487671 6396 layer_factory.hpp:77] Creating layer drop6 I0409 23:04:51.487681 6396 net.cpp:84] Creating Layer drop6 I0409 23:04:51.487686 6396 net.cpp:406] drop6 <- fc6 I0409 23:04:51.487696 6396 net.cpp:367] drop6 -> fc6 (in-place) I0409 23:04:51.487736 6396 net.cpp:122] Setting up drop6 I0409 23:04:51.487743 6396 net.cpp:129] Top shape: 128 512 (65536) I0409 23:04:51.487748 6396 net.cpp:137] Memory required for data: 1052555264 I0409 23:04:51.487753 6396 layer_factory.hpp:77] Creating layer fc7 I0409 23:04:51.487766 6396 net.cpp:84] Creating Layer fc7 I0409 23:04:51.487771 6396 net.cpp:406] fc7 <- fc6 I0409 23:04:51.487778 6396 net.cpp:380] fc7 -> fc7 I0409 23:04:51.491689 6396 net.cpp:122] Setting up fc7 I0409 23:04:51.491699 6396 net.cpp:129] Top shape: 128 512 (65536) I0409 23:04:51.491704 6396 net.cpp:137] Memory required for data: 1052817408 I0409 23:04:51.491714 6396 layer_factory.hpp:77] Creating layer relu7 I0409 23:04:51.491721 6396 net.cpp:84] Creating Layer relu7 I0409 23:04:51.491725 6396 net.cpp:406] relu7 <- fc7 I0409 23:04:51.491732 6396 net.cpp:367] relu7 -> fc7 (in-place) I0409 23:04:51.493413 6396 net.cpp:122] Setting up relu7 I0409 23:04:51.493425 6396 net.cpp:129] Top shape: 128 512 (65536) I0409 23:04:51.493430 6396 net.cpp:137] Memory required for data: 1053079552 I0409 23:04:51.493435 6396 layer_factory.hpp:77] Creating layer drop7 I0409 23:04:51.493443 6396 net.cpp:84] Creating Layer drop7 I0409 23:04:51.493448 6396 net.cpp:406] drop7 <- fc7 I0409 23:04:51.493481 6396 net.cpp:367] drop7 -> fc7 (in-place) I0409 23:04:51.493516 6396 net.cpp:122] Setting up drop7 I0409 23:04:51.493523 6396 net.cpp:129] Top shape: 128 512 (65536) I0409 23:04:51.493527 6396 net.cpp:137] Memory required for data: 1053341696 I0409 23:04:51.493532 6396 layer_factory.hpp:77] Creating layer fc8 I0409 23:04:51.493542 6396 net.cpp:84] Creating Layer fc8 I0409 23:04:51.493547 6396 net.cpp:406] fc8 <- fc7 I0409 23:04:51.493556 6396 net.cpp:380] fc8 -> fc8 I0409 23:04:51.495100 6396 net.cpp:122] Setting up fc8 I0409 23:04:51.495108 6396 net.cpp:129] Top shape: 128 196 (25088) I0409 23:04:51.495113 6396 net.cpp:137] Memory required for data: 1053442048 I0409 23:04:51.495121 6396 layer_factory.hpp:77] Creating layer loss I0409 23:04:51.495129 6396 net.cpp:84] Creating Layer loss I0409 23:04:51.495134 6396 net.cpp:406] loss <- fc8 I0409 23:04:51.495141 6396 net.cpp:406] loss <- label I0409 23:04:51.495148 6396 net.cpp:380] loss -> loss I0409 23:04:51.495159 6396 layer_factory.hpp:77] Creating layer loss I0409 23:04:51.496053 6396 net.cpp:122] Setting up loss I0409 23:04:51.496068 6396 net.cpp:129] Top shape: (1) I0409 23:04:51.496073 6396 net.cpp:132] with loss weight 1 I0409 23:04:51.496094 6396 net.cpp:137] Memory required for data: 1053442052 I0409 23:04:51.496100 6396 net.cpp:198] loss needs backward computation. I0409 23:04:51.496110 6396 net.cpp:198] fc8 needs backward computation. I0409 23:04:51.496115 6396 net.cpp:198] drop7 needs backward computation. I0409 23:04:51.496120 6396 net.cpp:198] relu7 needs backward computation. I0409 23:04:51.496125 6396 net.cpp:198] fc7 needs backward computation. I0409 23:04:51.496130 6396 net.cpp:198] drop6 needs backward computation. I0409 23:04:51.496134 6396 net.cpp:198] relu6 needs backward computation. I0409 23:04:51.496140 6396 net.cpp:198] fc6 needs backward computation. I0409 23:04:51.496145 6396 net.cpp:198] pool5 needs backward computation. I0409 23:04:51.496150 6396 net.cpp:198] relu5 needs backward computation. I0409 23:04:51.496155 6396 net.cpp:198] conv5 needs backward computation. I0409 23:04:51.496160 6396 net.cpp:198] relu4 needs backward computation. I0409 23:04:51.496165 6396 net.cpp:198] conv4 needs backward computation. I0409 23:04:51.496170 6396 net.cpp:198] relu3 needs backward computation. I0409 23:04:51.496176 6396 net.cpp:198] conv3 needs backward computation. I0409 23:04:51.496181 6396 net.cpp:198] pool2 needs backward computation. I0409 23:04:51.496186 6396 net.cpp:198] norm2 needs backward computation. I0409 23:04:51.496191 6396 net.cpp:198] relu2 needs backward computation. I0409 23:04:51.496196 6396 net.cpp:198] conv2 needs backward computation. I0409 23:04:51.496201 6396 net.cpp:198] pool1 needs backward computation. I0409 23:04:51.496206 6396 net.cpp:198] norm1 needs backward computation. I0409 23:04:51.496210 6396 net.cpp:198] relu1 needs backward computation. I0409 23:04:51.496215 6396 net.cpp:198] conv1 needs backward computation. I0409 23:04:51.496222 6396 net.cpp:200] train-data does not need backward computation. I0409 23:04:51.496227 6396 net.cpp:242] This network produces output loss I0409 23:04:51.496246 6396 net.cpp:255] Network initialization done. I0409 23:04:51.526417 6396 solver.cpp:172] Creating test net (#0) specified by net file: train_val.prototxt I0409 23:04:51.526487 6396 net.cpp:294] The NetState phase (1) differed from the phase (0) specified by a rule in layer train-data I0409 23:04:51.526746 6396 net.cpp:51] Initializing net from parameters: state { phase: TEST } layer { name: "val-data" type: "Data" top: "data" top: "label" include { phase: TEST } transform_param { crop_size: 227 mean_file: "/mnt/bigdisk/DIGITS-MAN-2/digits/jobs/20210407-214532-d396/mean.binaryproto" } data_param { source: "/mnt/bigdisk/DIGITS-MAN-2/digits/jobs/20210407-214532-d396/val_db" batch_size: 32 backend: LMDB } } layer { name: "conv1" type: "Convolution" bottom: "data" top: "conv1" param { lr_mult: 1 decay_mult: 1 } param { lr_mult: 2 decay_mult: 0 } convolution_param { num_output: 96 kernel_size: 11 stride: 4 weight_filler { type: "gaussian" std: 0.01 } bias_filler { type: "constant" value: 0 } } } layer { name: "relu1" type: "ReLU" bottom: "conv1" top: "conv1" } layer { name: "norm1" type: "LRN" bottom: "conv1" top: "norm1" lrn_param { local_size: 5 alpha: 0.0001 beta: 0.75 } } layer { name: "pool1" type: "Pooling" bottom: "norm1" top: "pool1" pooling_param { pool: MAX kernel_size: 3 stride: 2 } } layer { name: "conv2" type: "Convolution" bottom: "pool1" top: "conv2" param { lr_mult: 1 decay_mult: 1 } param { lr_mult: 2 decay_mult: 0 } convolution_param { num_output: 256 pad: 2 kernel_size: 5 group: 2 weight_filler { type: "gaussian" std: 0.01 } bias_filler { type: "constant" value: 0.1 } } } layer { name: "relu2" type: "ReLU" bottom: "conv2" top: "conv2" } layer { name: "norm2" type: "LRN" bottom: "conv2" top: "norm2" lrn_param { local_size: 5 alpha: 0.0001 beta: 0.75 } } layer { name: "pool2" type: "Pooling" bottom: "norm2" top: "pool2" pooling_param { pool: MAX kernel_size: 3 stride: 2 } } layer { name: "conv3" type: "Convolution" bottom: "pool2" top: "conv3" param { lr_mult: 1 decay_mult: 1 } param { lr_mult: 2 decay_mult: 0 } convolution_param { num_output: 384 pad: 1 kernel_size: 3 weight_filler { type: "gaussian" std: 0.01 } bias_filler { type: "constant" value: 0 } } } layer { name: "relu3" type: "ReLU" bottom: "conv3" top: "conv3" } layer { name: "conv4" type: "Convolution" bottom: "conv3" top: "conv4" param { lr_mult: 1 decay_mult: 1 } param { lr_mult: 2 decay_mult: 0 } convolution_param { num_output: 384 pad: 1 kernel_size: 3 group: 2 weight_filler { type: "gaussian" std: 0.01 } bias_filler { type: "constant" value: 0.1 } } } layer { name: "relu4" type: "ReLU" bottom: "conv4" top: "conv4" } layer { name: "conv5" type: "Convolution" bottom: "conv4" top: "conv5" param { lr_mult: 1 decay_mult: 1 } param { lr_mult: 2 decay_mult: 0 } convolution_param { num_output: 256 pad: 1 kernel_size: 3 group: 2 weight_filler { type: "gaussian" std: 0.01 } bias_filler { type: "constant" value: 0.1 } } } layer { name: "relu5" type: "ReLU" bottom: "conv5" top: "conv5" } layer { name: "pool5" type: "Pooling" bottom: "conv5" top: "pool5" pooling_param { pool: MAX kernel_size: 3 stride: 2 } } layer { name: "fc6" type: "InnerProduct" bottom: "pool5" top: "fc6" param { lr_mult: 1 decay_mult: 1 } param { lr_mult: 2 decay_mult: 0 } inner_product_param { num_output: 512 weight_filler { type: "gaussian" std: 0.005 } bias_filler { type: "constant" value: 0.1 } } } layer { name: "relu6" type: "ReLU" bottom: "fc6" top: "fc6" } layer { name: "drop6" type: "Dropout" bottom: "fc6" top: "fc6" dropout_param { dropout_ratio: 0.5 } } layer { name: "fc7" type: "InnerProduct" bottom: "fc6" top: "fc7" param { lr_mult: 1 decay_mult: 1 } param { lr_mult: 2 decay_mult: 0 } inner_product_param { num_output: 512 weight_filler { type: "gaussian" std: 0.005 } bias_filler { type: "constant" value: 0.1 } } } layer { name: "relu7" type: "ReLU" bottom: "fc7" top: "fc7" } layer { name: "drop7" type: "Dropout" bottom: "fc7" top: "fc7" dropout_param { dropout_ratio: 0.5 } } layer { name: "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" } I0409 23:04:51.526924 6396 layer_factory.hpp:77] Creating layer val-data I0409 23:04:51.908569 6396 db_lmdb.cpp:35] Opened lmdb /mnt/bigdisk/DIGITS-MAN-2/digits/jobs/20210407-214532-d396/val_db I0409 23:04:51.909112 6396 net.cpp:84] Creating Layer val-data I0409 23:04:51.909137 6396 net.cpp:380] val-data -> data I0409 23:04:51.909159 6396 net.cpp:380] val-data -> label I0409 23:04:51.909174 6396 data_transformer.cpp:25] Loading mean file from: /mnt/bigdisk/DIGITS-MAN-2/digits/jobs/20210407-214532-d396/mean.binaryproto I0409 23:04:51.960507 6396 data_layer.cpp:45] output data size: 32,3,227,227 I0409 23:04:52.004259 6396 net.cpp:122] Setting up val-data I0409 23:04:52.004284 6396 net.cpp:129] Top shape: 32 3 227 227 (4946784) I0409 23:04:52.004292 6396 net.cpp:129] Top shape: 32 (32) I0409 23:04:52.004295 6396 net.cpp:137] Memory required for data: 19787264 I0409 23:04:52.004303 6396 layer_factory.hpp:77] Creating layer label_val-data_1_split I0409 23:04:52.004318 6396 net.cpp:84] Creating Layer label_val-data_1_split I0409 23:04:52.004324 6396 net.cpp:406] label_val-data_1_split <- label I0409 23:04:52.004333 6396 net.cpp:380] label_val-data_1_split -> label_val-data_1_split_0 I0409 23:04:52.004346 6396 net.cpp:380] label_val-data_1_split -> label_val-data_1_split_1 I0409 23:04:52.004489 6396 net.cpp:122] Setting up label_val-data_1_split I0409 23:04:52.004498 6396 net.cpp:129] Top shape: 32 (32) I0409 23:04:52.004503 6396 net.cpp:129] Top shape: 32 (32) I0409 23:04:52.004508 6396 net.cpp:137] Memory required for data: 19787520 I0409 23:04:52.004513 6396 layer_factory.hpp:77] Creating layer conv1 I0409 23:04:52.004528 6396 net.cpp:84] Creating Layer conv1 I0409 23:04:52.004532 6396 net.cpp:406] conv1 <- data I0409 23:04:52.004541 6396 net.cpp:380] conv1 -> conv1 I0409 23:04:52.007539 6396 net.cpp:122] Setting up conv1 I0409 23:04:52.007555 6396 net.cpp:129] Top shape: 32 96 55 55 (9292800) I0409 23:04:52.007560 6396 net.cpp:137] Memory required for data: 56958720 I0409 23:04:52.007573 6396 layer_factory.hpp:77] Creating layer relu1 I0409 23:04:52.007582 6396 net.cpp:84] Creating Layer relu1 I0409 23:04:52.007587 6396 net.cpp:406] relu1 <- conv1 I0409 23:04:52.007594 6396 net.cpp:367] relu1 -> conv1 (in-place) I0409 23:04:52.008026 6396 net.cpp:122] Setting up relu1 I0409 23:04:52.008038 6396 net.cpp:129] Top shape: 32 96 55 55 (9292800) I0409 23:04:52.008042 6396 net.cpp:137] Memory required for data: 94129920 I0409 23:04:52.008047 6396 layer_factory.hpp:77] Creating layer norm1 I0409 23:04:52.008057 6396 net.cpp:84] Creating Layer norm1 I0409 23:04:52.008062 6396 net.cpp:406] norm1 <- conv1 I0409 23:04:52.008069 6396 net.cpp:380] norm1 -> norm1 I0409 23:04:52.008869 6396 net.cpp:122] Setting up norm1 I0409 23:04:52.008883 6396 net.cpp:129] Top shape: 32 96 55 55 (9292800) I0409 23:04:52.008888 6396 net.cpp:137] Memory required for data: 131301120 I0409 23:04:52.008893 6396 layer_factory.hpp:77] Creating layer pool1 I0409 23:04:52.008903 6396 net.cpp:84] Creating Layer pool1 I0409 23:04:52.008908 6396 net.cpp:406] pool1 <- norm1 I0409 23:04:52.008914 6396 net.cpp:380] pool1 -> pool1 I0409 23:04:52.008957 6396 net.cpp:122] Setting up pool1 I0409 23:04:52.008965 6396 net.cpp:129] Top shape: 32 96 27 27 (2239488) I0409 23:04:52.008968 6396 net.cpp:137] Memory required for data: 140259072 I0409 23:04:52.008973 6396 layer_factory.hpp:77] Creating layer conv2 I0409 23:04:52.008985 6396 net.cpp:84] Creating Layer conv2 I0409 23:04:52.008988 6396 net.cpp:406] conv2 <- pool1 I0409 23:04:52.009021 6396 net.cpp:380] conv2 -> conv2 I0409 23:04:52.019079 6396 net.cpp:122] Setting up conv2 I0409 23:04:52.019102 6396 net.cpp:129] Top shape: 32 256 27 27 (5971968) I0409 23:04:52.019107 6396 net.cpp:137] Memory required for data: 164146944 I0409 23:04:52.019121 6396 layer_factory.hpp:77] Creating layer relu2 I0409 23:04:52.019132 6396 net.cpp:84] Creating Layer relu2 I0409 23:04:52.019137 6396 net.cpp:406] relu2 <- conv2 I0409 23:04:52.019145 6396 net.cpp:367] relu2 -> conv2 (in-place) I0409 23:04:52.019842 6396 net.cpp:122] Setting up relu2 I0409 23:04:52.019855 6396 net.cpp:129] Top shape: 32 256 27 27 (5971968) I0409 23:04:52.019860 6396 net.cpp:137] Memory required for data: 188034816 I0409 23:04:52.019865 6396 layer_factory.hpp:77] Creating layer norm2 I0409 23:04:52.019878 6396 net.cpp:84] Creating Layer norm2 I0409 23:04:52.019883 6396 net.cpp:406] norm2 <- conv2 I0409 23:04:52.019891 6396 net.cpp:380] norm2 -> norm2 I0409 23:04:52.020607 6396 net.cpp:122] Setting up norm2 I0409 23:04:52.020620 6396 net.cpp:129] Top shape: 32 256 27 27 (5971968) I0409 23:04:52.020625 6396 net.cpp:137] Memory required for data: 211922688 I0409 23:04:52.020630 6396 layer_factory.hpp:77] Creating layer pool2 I0409 23:04:52.020640 6396 net.cpp:84] Creating Layer pool2 I0409 23:04:52.020645 6396 net.cpp:406] pool2 <- norm2 I0409 23:04:52.020653 6396 net.cpp:380] pool2 -> pool2 I0409 23:04:52.020697 6396 net.cpp:122] Setting up pool2 I0409 23:04:52.020704 6396 net.cpp:129] Top shape: 32 256 13 13 (1384448) I0409 23:04:52.020709 6396 net.cpp:137] Memory required for data: 217460480 I0409 23:04:52.020714 6396 layer_factory.hpp:77] Creating layer conv3 I0409 23:04:52.020727 6396 net.cpp:84] Creating Layer conv3 I0409 23:04:52.020731 6396 net.cpp:406] conv3 <- pool2 I0409 23:04:52.020740 6396 net.cpp:380] conv3 -> conv3 I0409 23:04:52.037477 6396 net.cpp:122] Setting up conv3 I0409 23:04:52.037498 6396 net.cpp:129] Top shape: 32 384 13 13 (2076672) I0409 23:04:52.037503 6396 net.cpp:137] Memory required for data: 225767168 I0409 23:04:52.037518 6396 layer_factory.hpp:77] Creating layer relu3 I0409 23:04:52.037529 6396 net.cpp:84] Creating Layer relu3 I0409 23:04:52.037535 6396 net.cpp:406] relu3 <- conv3 I0409 23:04:52.037542 6396 net.cpp:367] relu3 -> conv3 (in-place) I0409 23:04:52.038245 6396 net.cpp:122] Setting up relu3 I0409 23:04:52.038259 6396 net.cpp:129] Top shape: 32 384 13 13 (2076672) I0409 23:04:52.038262 6396 net.cpp:137] Memory required for data: 234073856 I0409 23:04:52.038267 6396 layer_factory.hpp:77] Creating layer conv4 I0409 23:04:52.038282 6396 net.cpp:84] Creating Layer conv4 I0409 23:04:52.038287 6396 net.cpp:406] conv4 <- conv3 I0409 23:04:52.038297 6396 net.cpp:380] conv4 -> conv4 I0409 23:04:52.051004 6396 net.cpp:122] Setting up conv4 I0409 23:04:52.051024 6396 net.cpp:129] Top shape: 32 384 13 13 (2076672) I0409 23:04:52.051028 6396 net.cpp:137] Memory required for data: 242380544 I0409 23:04:52.051038 6396 layer_factory.hpp:77] Creating layer relu4 I0409 23:04:52.051048 6396 net.cpp:84] Creating Layer relu4 I0409 23:04:52.051054 6396 net.cpp:406] relu4 <- conv4 I0409 23:04:52.051060 6396 net.cpp:367] relu4 -> conv4 (in-place) I0409 23:04:52.051496 6396 net.cpp:122] Setting up relu4 I0409 23:04:52.051506 6396 net.cpp:129] Top shape: 32 384 13 13 (2076672) I0409 23:04:52.051509 6396 net.cpp:137] Memory required for data: 250687232 I0409 23:04:52.051514 6396 layer_factory.hpp:77] Creating layer conv5 I0409 23:04:52.051527 6396 net.cpp:84] Creating Layer conv5 I0409 23:04:52.051530 6396 net.cpp:406] conv5 <- conv4 I0409 23:04:52.051539 6396 net.cpp:380] conv5 -> conv5 I0409 23:04:52.062309 6396 net.cpp:122] Setting up conv5 I0409 23:04:52.062328 6396 net.cpp:129] Top shape: 32 256 13 13 (1384448) I0409 23:04:52.062332 6396 net.cpp:137] Memory required for data: 256225024 I0409 23:04:52.062350 6396 layer_factory.hpp:77] Creating layer relu5 I0409 23:04:52.062361 6396 net.cpp:84] Creating Layer relu5 I0409 23:04:52.062366 6396 net.cpp:406] relu5 <- conv5 I0409 23:04:52.062392 6396 net.cpp:367] relu5 -> conv5 (in-place) I0409 23:04:52.063001 6396 net.cpp:122] Setting up relu5 I0409 23:04:52.063014 6396 net.cpp:129] Top shape: 32 256 13 13 (1384448) I0409 23:04:52.063017 6396 net.cpp:137] Memory required for data: 261762816 I0409 23:04:52.063022 6396 layer_factory.hpp:77] Creating layer pool5 I0409 23:04:52.063033 6396 net.cpp:84] Creating Layer pool5 I0409 23:04:52.063037 6396 net.cpp:406] pool5 <- conv5 I0409 23:04:52.063045 6396 net.cpp:380] pool5 -> pool5 I0409 23:04:52.063089 6396 net.cpp:122] Setting up pool5 I0409 23:04:52.063097 6396 net.cpp:129] Top shape: 32 256 6 6 (294912) I0409 23:04:52.063099 6396 net.cpp:137] Memory required for data: 262942464 I0409 23:04:52.063103 6396 layer_factory.hpp:77] Creating layer fc6 I0409 23:04:52.063112 6396 net.cpp:84] Creating Layer fc6 I0409 23:04:52.063117 6396 net.cpp:406] fc6 <- pool5 I0409 23:04:52.063123 6396 net.cpp:380] fc6 -> fc6 I0409 23:04:52.115468 6396 net.cpp:122] Setting up fc6 I0409 23:04:52.115487 6396 net.cpp:129] Top shape: 32 512 (16384) I0409 23:04:52.115491 6396 net.cpp:137] Memory required for data: 263008000 I0409 23:04:52.115502 6396 layer_factory.hpp:77] Creating layer relu6 I0409 23:04:52.115512 6396 net.cpp:84] Creating Layer relu6 I0409 23:04:52.115517 6396 net.cpp:406] relu6 <- fc6 I0409 23:04:52.115526 6396 net.cpp:367] relu6 -> fc6 (in-place) I0409 23:04:52.116503 6396 net.cpp:122] Setting up relu6 I0409 23:04:52.116513 6396 net.cpp:129] Top shape: 32 512 (16384) I0409 23:04:52.116518 6396 net.cpp:137] Memory required for data: 263073536 I0409 23:04:52.116521 6396 layer_factory.hpp:77] Creating layer drop6 I0409 23:04:52.116530 6396 net.cpp:84] Creating Layer drop6 I0409 23:04:52.116534 6396 net.cpp:406] drop6 <- fc6 I0409 23:04:52.116544 6396 net.cpp:367] drop6 -> fc6 (in-place) I0409 23:04:52.116573 6396 net.cpp:122] Setting up drop6 I0409 23:04:52.116580 6396 net.cpp:129] Top shape: 32 512 (16384) I0409 23:04:52.116585 6396 net.cpp:137] Memory required for data: 263139072 I0409 23:04:52.116588 6396 layer_factory.hpp:77] Creating layer fc7 I0409 23:04:52.116596 6396 net.cpp:84] Creating Layer fc7 I0409 23:04:52.116601 6396 net.cpp:406] fc7 <- fc6 I0409 23:04:52.116611 6396 net.cpp:380] fc7 -> fc7 I0409 23:04:52.119923 6396 net.cpp:122] Setting up fc7 I0409 23:04:52.119935 6396 net.cpp:129] Top shape: 32 512 (16384) I0409 23:04:52.119938 6396 net.cpp:137] Memory required for data: 263204608 I0409 23:04:52.119946 6396 layer_factory.hpp:77] Creating layer relu7 I0409 23:04:52.119954 6396 net.cpp:84] Creating Layer relu7 I0409 23:04:52.119959 6396 net.cpp:406] relu7 <- fc7 I0409 23:04:52.119966 6396 net.cpp:367] relu7 -> fc7 (in-place) I0409 23:04:52.120349 6396 net.cpp:122] Setting up relu7 I0409 23:04:52.120358 6396 net.cpp:129] Top shape: 32 512 (16384) I0409 23:04:52.120363 6396 net.cpp:137] Memory required for data: 263270144 I0409 23:04:52.120368 6396 layer_factory.hpp:77] Creating layer drop7 I0409 23:04:52.120374 6396 net.cpp:84] Creating Layer drop7 I0409 23:04:52.120378 6396 net.cpp:406] drop7 <- fc7 I0409 23:04:52.120386 6396 net.cpp:367] drop7 -> fc7 (in-place) I0409 23:04:52.120411 6396 net.cpp:122] Setting up drop7 I0409 23:04:52.120417 6396 net.cpp:129] Top shape: 32 512 (16384) I0409 23:04:52.120422 6396 net.cpp:137] Memory required for data: 263335680 I0409 23:04:52.120427 6396 layer_factory.hpp:77] Creating layer fc8 I0409 23:04:52.120434 6396 net.cpp:84] Creating Layer fc8 I0409 23:04:52.120440 6396 net.cpp:406] fc8 <- fc7 I0409 23:04:52.120447 6396 net.cpp:380] fc8 -> fc8 I0409 23:04:52.121482 6396 net.cpp:122] Setting up fc8 I0409 23:04:52.121490 6396 net.cpp:129] Top shape: 32 196 (6272) I0409 23:04:52.121495 6396 net.cpp:137] Memory required for data: 263360768 I0409 23:04:52.121503 6396 layer_factory.hpp:77] Creating layer fc8_fc8_0_split I0409 23:04:52.121511 6396 net.cpp:84] Creating Layer fc8_fc8_0_split I0409 23:04:52.121516 6396 net.cpp:406] fc8_fc8_0_split <- fc8 I0409 23:04:52.121522 6396 net.cpp:380] fc8_fc8_0_split -> fc8_fc8_0_split_0 I0409 23:04:52.121547 6396 net.cpp:380] fc8_fc8_0_split -> fc8_fc8_0_split_1 I0409 23:04:52.121580 6396 net.cpp:122] Setting up fc8_fc8_0_split I0409 23:04:52.121585 6396 net.cpp:129] Top shape: 32 196 (6272) I0409 23:04:52.121592 6396 net.cpp:129] Top shape: 32 196 (6272) I0409 23:04:52.121595 6396 net.cpp:137] Memory required for data: 263410944 I0409 23:04:52.121600 6396 layer_factory.hpp:77] Creating layer accuracy I0409 23:04:52.121608 6396 net.cpp:84] Creating Layer accuracy I0409 23:04:52.121613 6396 net.cpp:406] accuracy <- fc8_fc8_0_split_0 I0409 23:04:52.121618 6396 net.cpp:406] accuracy <- label_val-data_1_split_0 I0409 23:04:52.121624 6396 net.cpp:380] accuracy -> accuracy I0409 23:04:52.121632 6396 net.cpp:122] Setting up accuracy I0409 23:04:52.121636 6396 net.cpp:129] Top shape: (1) I0409 23:04:52.121640 6396 net.cpp:137] Memory required for data: 263410948 I0409 23:04:52.121645 6396 layer_factory.hpp:77] Creating layer loss I0409 23:04:52.121654 6396 net.cpp:84] Creating Layer loss I0409 23:04:52.121657 6396 net.cpp:406] loss <- fc8_fc8_0_split_1 I0409 23:04:52.121663 6396 net.cpp:406] loss <- label_val-data_1_split_1 I0409 23:04:52.121670 6396 net.cpp:380] loss -> loss I0409 23:04:52.121678 6396 layer_factory.hpp:77] Creating layer loss I0409 23:04:52.122332 6396 net.cpp:122] Setting up loss I0409 23:04:52.122342 6396 net.cpp:129] Top shape: (1) I0409 23:04:52.122347 6396 net.cpp:132] with loss weight 1 I0409 23:04:52.122359 6396 net.cpp:137] Memory required for data: 263410952 I0409 23:04:52.122364 6396 net.cpp:198] loss needs backward computation. I0409 23:04:52.122370 6396 net.cpp:200] accuracy does not need backward computation. I0409 23:04:52.122375 6396 net.cpp:198] fc8_fc8_0_split needs backward computation. I0409 23:04:52.122380 6396 net.cpp:198] fc8 needs backward computation. I0409 23:04:52.122385 6396 net.cpp:198] drop7 needs backward computation. I0409 23:04:52.122390 6396 net.cpp:198] relu7 needs backward computation. I0409 23:04:52.122393 6396 net.cpp:198] fc7 needs backward computation. I0409 23:04:52.122400 6396 net.cpp:198] drop6 needs backward computation. I0409 23:04:52.122404 6396 net.cpp:198] relu6 needs backward computation. I0409 23:04:52.122408 6396 net.cpp:198] fc6 needs backward computation. I0409 23:04:52.122413 6396 net.cpp:198] pool5 needs backward computation. I0409 23:04:52.122417 6396 net.cpp:198] relu5 needs backward computation. I0409 23:04:52.122422 6396 net.cpp:198] conv5 needs backward computation. I0409 23:04:52.122426 6396 net.cpp:198] relu4 needs backward computation. I0409 23:04:52.122431 6396 net.cpp:198] conv4 needs backward computation. I0409 23:04:52.122437 6396 net.cpp:198] relu3 needs backward computation. I0409 23:04:52.122442 6396 net.cpp:198] conv3 needs backward computation. I0409 23:04:52.122447 6396 net.cpp:198] pool2 needs backward computation. I0409 23:04:52.122452 6396 net.cpp:198] norm2 needs backward computation. I0409 23:04:52.122455 6396 net.cpp:198] relu2 needs backward computation. I0409 23:04:52.122460 6396 net.cpp:198] conv2 needs backward computation. I0409 23:04:52.122465 6396 net.cpp:198] pool1 needs backward computation. I0409 23:04:52.122469 6396 net.cpp:198] norm1 needs backward computation. I0409 23:04:52.122474 6396 net.cpp:198] relu1 needs backward computation. I0409 23:04:52.122479 6396 net.cpp:198] conv1 needs backward computation. I0409 23:04:52.122484 6396 net.cpp:200] label_val-data_1_split does not need backward computation. I0409 23:04:52.122489 6396 net.cpp:200] val-data does not need backward computation. I0409 23:04:52.122494 6396 net.cpp:242] This network produces output accuracy I0409 23:04:52.122499 6396 net.cpp:242] This network produces output loss I0409 23:04:52.122517 6396 net.cpp:255] Network initialization done. I0409 23:04:52.122592 6396 solver.cpp:56] Solver scaffolding done. I0409 23:04:52.123062 6396 caffe.cpp:248] Starting Optimization I0409 23:04:52.123072 6396 solver.cpp:272] Solving I0409 23:04:52.123085 6396 solver.cpp:273] Learning Rate Policy: exp I0409 23:04:52.124389 6396 solver.cpp:330] Iteration 0, Testing net (#0) I0409 23:04:52.124400 6396 net.cpp:676] Ignoring source layer train-data I0409 23:04:52.128414 6396 blocking_queue.cpp:49] Waiting for data I0409 23:04:56.686985 6401 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:04:56.731577 6396 solver.cpp:397] Test net output #0: accuracy = 0.00612745 I0409 23:04:56.731622 6396 solver.cpp:397] Test net output #1: loss = 5.27807 (* 1 = 5.27807 loss) I0409 23:04:56.818626 6396 solver.cpp:218] Iteration 0 (-2.81661e-43 iter/s, 4.69537s/12 iters), loss = 5.27455 I0409 23:04:56.818673 6396 solver.cpp:237] Train net output #0: loss = 5.27455 (* 1 = 5.27455 loss) I0409 23:04:56.818701 6396 sgd_solver.cpp:105] Iteration 0, lr = 0.01 I0409 23:05:00.818203 6396 solver.cpp:218] Iteration 12 (3.00045 iter/s, 3.99939s/12 iters), loss = 5.27929 I0409 23:05:00.818249 6396 solver.cpp:237] Train net output #0: loss = 5.27929 (* 1 = 5.27929 loss) I0409 23:05:00.818260 6396 sgd_solver.cpp:105] Iteration 12, lr = 0.00997626 I0409 23:05:05.753098 6396 solver.cpp:218] Iteration 24 (2.43176 iter/s, 4.93469s/12 iters), loss = 5.2787 I0409 23:05:05.753139 6396 solver.cpp:237] Train net output #0: loss = 5.2787 (* 1 = 5.2787 loss) I0409 23:05:05.753149 6396 sgd_solver.cpp:105] Iteration 24, lr = 0.00995257 I0409 23:05:10.625360 6396 solver.cpp:218] Iteration 36 (2.46302 iter/s, 4.87206s/12 iters), loss = 5.28537 I0409 23:05:10.625419 6396 solver.cpp:237] Train net output #0: loss = 5.28537 (* 1 = 5.28537 loss) I0409 23:05:10.625432 6396 sgd_solver.cpp:105] Iteration 36, lr = 0.00992894 I0409 23:05:15.579774 6396 solver.cpp:218] Iteration 48 (2.42219 iter/s, 4.9542s/12 iters), loss = 5.28018 I0409 23:05:15.579818 6396 solver.cpp:237] Train net output #0: loss = 5.28018 (* 1 = 5.28018 loss) I0409 23:05:15.579828 6396 sgd_solver.cpp:105] Iteration 48, lr = 0.00990537 I0409 23:05:20.499451 6396 solver.cpp:218] Iteration 60 (2.43928 iter/s, 4.91947s/12 iters), loss = 5.27536 I0409 23:05:20.499606 6396 solver.cpp:237] Train net output #0: loss = 5.27536 (* 1 = 5.27536 loss) I0409 23:05:20.499622 6396 sgd_solver.cpp:105] Iteration 60, lr = 0.00988185 I0409 23:05:25.575270 6396 solver.cpp:218] Iteration 72 (2.3643 iter/s, 5.07551s/12 iters), loss = 5.27605 I0409 23:05:25.575312 6396 solver.cpp:237] Train net output #0: loss = 5.27605 (* 1 = 5.27605 loss) I0409 23:05:25.575321 6396 sgd_solver.cpp:105] Iteration 72, lr = 0.00985839 I0409 23:05:30.478696 6396 solver.cpp:218] Iteration 84 (2.44737 iter/s, 4.90322s/12 iters), loss = 5.28184 I0409 23:05:30.478751 6396 solver.cpp:237] Train net output #0: loss = 5.28184 (* 1 = 5.28184 loss) I0409 23:05:30.478763 6396 sgd_solver.cpp:105] Iteration 84, lr = 0.00983498 I0409 23:05:35.387403 6396 solver.cpp:218] Iteration 96 (2.44474 iter/s, 4.90849s/12 iters), loss = 5.2859 I0409 23:05:35.387456 6396 solver.cpp:237] Train net output #0: loss = 5.2859 (* 1 = 5.2859 loss) I0409 23:05:35.387467 6396 sgd_solver.cpp:105] Iteration 96, lr = 0.00981163 I0409 23:05:37.057448 6400 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:05:37.364966 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_102.caffemodel I0409 23:05:37.954406 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_102.solverstate I0409 23:05:40.057876 6396 solver.cpp:330] Iteration 102, Testing net (#0) I0409 23:05:40.057898 6396 net.cpp:676] Ignoring source layer train-data I0409 23:05:44.445679 6401 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:05:44.521896 6396 solver.cpp:397] Test net output #0: accuracy = 0.00551471 I0409 23:05:44.521947 6396 solver.cpp:397] Test net output #1: loss = 5.27944 (* 1 = 5.27944 loss) I0409 23:05:46.403074 6396 solver.cpp:218] Iteration 108 (1.08939 iter/s, 11.0153s/12 iters), loss = 5.28126 I0409 23:05:46.403115 6396 solver.cpp:237] Train net output #0: loss = 5.28126 (* 1 = 5.28126 loss) I0409 23:05:46.403123 6396 sgd_solver.cpp:105] Iteration 108, lr = 0.00978834 I0409 23:05:51.560257 6396 solver.cpp:218] Iteration 120 (2.32695 iter/s, 5.15697s/12 iters), loss = 5.27769 I0409 23:05:51.560406 6396 solver.cpp:237] Train net output #0: loss = 5.27769 (* 1 = 5.27769 loss) I0409 23:05:51.560416 6396 sgd_solver.cpp:105] Iteration 120, lr = 0.0097651 I0409 23:05:56.549398 6396 solver.cpp:218] Iteration 132 (2.40537 iter/s, 4.98884s/12 iters), loss = 5.25276 I0409 23:05:56.549441 6396 solver.cpp:237] Train net output #0: loss = 5.25276 (* 1 = 5.25276 loss) I0409 23:05:56.549451 6396 sgd_solver.cpp:105] Iteration 132, lr = 0.00974192 I0409 23:06:01.398185 6396 solver.cpp:218] Iteration 144 (2.47495 iter/s, 4.84859s/12 iters), loss = 5.28589 I0409 23:06:01.398245 6396 solver.cpp:237] Train net output #0: loss = 5.28589 (* 1 = 5.28589 loss) I0409 23:06:01.398257 6396 sgd_solver.cpp:105] Iteration 144, lr = 0.00971879 I0409 23:06:06.469995 6396 solver.cpp:218] Iteration 156 (2.36612 iter/s, 5.07159s/12 iters), loss = 5.26399 I0409 23:06:06.470037 6396 solver.cpp:237] Train net output #0: loss = 5.26399 (* 1 = 5.26399 loss) I0409 23:06:06.470047 6396 sgd_solver.cpp:105] Iteration 156, lr = 0.00969571 I0409 23:06:11.510557 6396 solver.cpp:218] Iteration 168 (2.38078 iter/s, 5.04036s/12 iters), loss = 5.27704 I0409 23:06:11.510607 6396 solver.cpp:237] Train net output #0: loss = 5.27704 (* 1 = 5.27704 loss) I0409 23:06:11.510618 6396 sgd_solver.cpp:105] Iteration 168, lr = 0.00967269 I0409 23:06:16.442725 6396 solver.cpp:218] Iteration 180 (2.43311 iter/s, 4.93196s/12 iters), loss = 5.2723 I0409 23:06:16.442778 6396 solver.cpp:237] Train net output #0: loss = 5.2723 (* 1 = 5.2723 loss) I0409 23:06:16.442790 6396 sgd_solver.cpp:105] Iteration 180, lr = 0.00964973 I0409 23:06:21.474241 6396 solver.cpp:218] Iteration 192 (2.38507 iter/s, 5.0313s/12 iters), loss = 5.2814 I0409 23:06:21.474280 6396 solver.cpp:237] Train net output #0: loss = 5.2814 (* 1 = 5.2814 loss) I0409 23:06:21.474288 6396 sgd_solver.cpp:105] Iteration 192, lr = 0.00962682 I0409 23:06:25.326884 6400 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:06:25.984145 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_204.caffemodel I0409 23:06:27.344069 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_204.solverstate I0409 23:06:27.665225 6396 solver.cpp:330] Iteration 204, Testing net (#0) I0409 23:06:27.665244 6396 net.cpp:676] Ignoring source layer train-data I0409 23:06:31.994709 6401 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:06:32.116238 6396 solver.cpp:397] Test net output #0: accuracy = 0.00551471 I0409 23:06:32.116293 6396 solver.cpp:397] Test net output #1: loss = 5.28107 (* 1 = 5.28107 loss) I0409 23:06:32.199648 6396 solver.cpp:218] Iteration 204 (1.11888 iter/s, 10.725s/12 iters), loss = 5.27436 I0409 23:06:32.199698 6396 solver.cpp:237] Train net output #0: loss = 5.27436 (* 1 = 5.27436 loss) I0409 23:06:32.199708 6396 sgd_solver.cpp:105] Iteration 204, lr = 0.00960396 I0409 23:06:36.417697 6396 solver.cpp:218] Iteration 216 (2.84504 iter/s, 4.21786s/12 iters), loss = 5.2777 I0409 23:06:36.417743 6396 solver.cpp:237] Train net output #0: loss = 5.2777 (* 1 = 5.2777 loss) I0409 23:06:36.417753 6396 sgd_solver.cpp:105] Iteration 216, lr = 0.00958116 I0409 23:06:41.320812 6396 solver.cpp:218] Iteration 228 (2.44753 iter/s, 4.90291s/12 iters), loss = 5.25923 I0409 23:06:41.320868 6396 solver.cpp:237] Train net output #0: loss = 5.25923 (* 1 = 5.25923 loss) I0409 23:06:41.320881 6396 sgd_solver.cpp:105] Iteration 228, lr = 0.00955841 I0409 23:06:46.224095 6396 solver.cpp:218] Iteration 240 (2.44745 iter/s, 4.90307s/12 iters), loss = 5.28509 I0409 23:06:46.224144 6396 solver.cpp:237] Train net output #0: loss = 5.28509 (* 1 = 5.28509 loss) I0409 23:06:46.224153 6396 sgd_solver.cpp:105] Iteration 240, lr = 0.00953572 I0409 23:06:51.241735 6396 solver.cpp:218] Iteration 252 (2.39166 iter/s, 5.01743s/12 iters), loss = 5.27584 I0409 23:06:51.241789 6396 solver.cpp:237] Train net output #0: loss = 5.27584 (* 1 = 5.27584 loss) I0409 23:06:51.241801 6396 sgd_solver.cpp:105] Iteration 252, lr = 0.00951308 I0409 23:06:56.286557 6396 solver.cpp:218] Iteration 264 (2.37878 iter/s, 5.04461s/12 iters), loss = 5.27004 I0409 23:06:56.286700 6396 solver.cpp:237] Train net output #0: loss = 5.27004 (* 1 = 5.27004 loss) I0409 23:06:56.286711 6396 sgd_solver.cpp:105] Iteration 264, lr = 0.00949049 I0409 23:07:01.266630 6396 solver.cpp:218] Iteration 276 (2.40975 iter/s, 4.97977s/12 iters), loss = 5.29071 I0409 23:07:01.266686 6396 solver.cpp:237] Train net output #0: loss = 5.29071 (* 1 = 5.29071 loss) I0409 23:07:01.266696 6396 sgd_solver.cpp:105] Iteration 276, lr = 0.00946796 I0409 23:07:06.287289 6396 solver.cpp:218] Iteration 288 (2.39022 iter/s, 5.02045s/12 iters), loss = 5.27801 I0409 23:07:06.287329 6396 solver.cpp:237] Train net output #0: loss = 5.27801 (* 1 = 5.27801 loss) I0409 23:07:06.287338 6396 sgd_solver.cpp:105] Iteration 288, lr = 0.00944548 I0409 23:07:11.268934 6396 solver.cpp:218] Iteration 300 (2.40894 iter/s, 4.98144s/12 iters), loss = 5.2789 I0409 23:07:11.268992 6396 solver.cpp:237] Train net output #0: loss = 5.2789 (* 1 = 5.2789 loss) I0409 23:07:11.269003 6396 sgd_solver.cpp:105] Iteration 300, lr = 0.00942305 I0409 23:07:12.225855 6400 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:07:13.268409 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_306.caffemodel I0409 23:07:13.719501 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_306.solverstate I0409 23:07:14.044073 6396 solver.cpp:330] Iteration 306, Testing net (#0) I0409 23:07:14.044092 6396 net.cpp:676] Ignoring source layer train-data I0409 23:07:18.424403 6401 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:07:18.581279 6396 solver.cpp:397] Test net output #0: accuracy = 0.00551471 I0409 23:07:18.581328 6396 solver.cpp:397] Test net output #1: loss = 5.28201 (* 1 = 5.28201 loss) I0409 23:07:20.459380 6396 solver.cpp:218] Iteration 312 (1.30575 iter/s, 9.19011s/12 iters), loss = 5.28612 I0409 23:07:20.459432 6396 solver.cpp:237] Train net output #0: loss = 5.28612 (* 1 = 5.28612 loss) I0409 23:07:20.459444 6396 sgd_solver.cpp:105] Iteration 312, lr = 0.00940068 I0409 23:07:25.387271 6396 solver.cpp:218] Iteration 324 (2.43522 iter/s, 4.92769s/12 iters), loss = 5.2522 I0409 23:07:25.387318 6396 solver.cpp:237] Train net output #0: loss = 5.2522 (* 1 = 5.2522 loss) I0409 23:07:25.387328 6396 sgd_solver.cpp:105] Iteration 324, lr = 0.00937836 I0409 23:07:30.326529 6396 solver.cpp:218] Iteration 336 (2.42962 iter/s, 4.93905s/12 iters), loss = 5.26202 I0409 23:07:30.326648 6396 solver.cpp:237] Train net output #0: loss = 5.26202 (* 1 = 5.26202 loss) I0409 23:07:30.326663 6396 sgd_solver.cpp:105] Iteration 336, lr = 0.0093561 I0409 23:07:35.248317 6396 solver.cpp:218] Iteration 348 (2.43827 iter/s, 4.92151s/12 iters), loss = 5.26278 I0409 23:07:35.248368 6396 solver.cpp:237] Train net output #0: loss = 5.26278 (* 1 = 5.26278 loss) I0409 23:07:35.248380 6396 sgd_solver.cpp:105] Iteration 348, lr = 0.00933388 I0409 23:07:40.159725 6396 solver.cpp:218] Iteration 360 (2.44339 iter/s, 4.9112s/12 iters), loss = 5.28856 I0409 23:07:40.159772 6396 solver.cpp:237] Train net output #0: loss = 5.28856 (* 1 = 5.28856 loss) I0409 23:07:40.159782 6396 sgd_solver.cpp:105] Iteration 360, lr = 0.00931172 I0409 23:07:45.100924 6396 solver.cpp:218] Iteration 372 (2.42866 iter/s, 4.941s/12 iters), loss = 5.27144 I0409 23:07:45.100980 6396 solver.cpp:237] Train net output #0: loss = 5.27144 (* 1 = 5.27144 loss) I0409 23:07:45.100993 6396 sgd_solver.cpp:105] Iteration 372, lr = 0.00928961 I0409 23:07:50.172209 6396 solver.cpp:218] Iteration 384 (2.36637 iter/s, 5.07107s/12 iters), loss = 5.27793 I0409 23:07:50.172259 6396 solver.cpp:237] Train net output #0: loss = 5.27793 (* 1 = 5.27793 loss) I0409 23:07:50.172269 6396 sgd_solver.cpp:105] Iteration 384, lr = 0.00926756 I0409 23:07:55.102274 6396 solver.cpp:218] Iteration 396 (2.43415 iter/s, 4.92986s/12 iters), loss = 5.25999 I0409 23:07:55.102330 6396 solver.cpp:237] Train net output #0: loss = 5.25999 (* 1 = 5.25999 loss) I0409 23:07:55.102341 6396 sgd_solver.cpp:105] Iteration 396, lr = 0.00924556 I0409 23:07:58.232841 6400 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:07:59.639489 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_408.caffemodel I0409 23:08:00.208518 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_408.solverstate I0409 23:08:00.517051 6396 solver.cpp:330] Iteration 408, Testing net (#0) I0409 23:08:00.517158 6396 net.cpp:676] Ignoring source layer train-data I0409 23:08:04.915843 6401 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:08:05.117702 6396 solver.cpp:397] Test net output #0: accuracy = 0.00551471 I0409 23:08:05.117748 6396 solver.cpp:397] Test net output #1: loss = 5.27655 (* 1 = 5.27655 loss) I0409 23:08:05.201004 6396 solver.cpp:218] Iteration 408 (1.18831 iter/s, 10.0984s/12 iters), loss = 5.27337 I0409 23:08:05.201056 6396 solver.cpp:237] Train net output #0: loss = 5.27337 (* 1 = 5.27337 loss) I0409 23:08:05.201066 6396 sgd_solver.cpp:105] Iteration 408, lr = 0.00922361 I0409 23:08:09.453930 6396 solver.cpp:218] Iteration 420 (2.82171 iter/s, 4.25274s/12 iters), loss = 5.27275 I0409 23:08:09.453987 6396 solver.cpp:237] Train net output #0: loss = 5.27275 (* 1 = 5.27275 loss) I0409 23:08:09.453999 6396 sgd_solver.cpp:105] Iteration 420, lr = 0.00920171 I0409 23:08:14.579460 6396 solver.cpp:218] Iteration 432 (2.34132 iter/s, 5.12531s/12 iters), loss = 5.21435 I0409 23:08:14.579506 6396 solver.cpp:237] Train net output #0: loss = 5.21435 (* 1 = 5.21435 loss) I0409 23:08:14.579516 6396 sgd_solver.cpp:105] Iteration 432, lr = 0.00917986 I0409 23:08:19.450215 6396 solver.cpp:218] Iteration 444 (2.46378 iter/s, 4.87056s/12 iters), loss = 5.20044 I0409 23:08:19.450260 6396 solver.cpp:237] Train net output #0: loss = 5.20044 (* 1 = 5.20044 loss) I0409 23:08:19.450268 6396 sgd_solver.cpp:105] Iteration 444, lr = 0.00915807 I0409 23:08:24.400693 6396 solver.cpp:218] Iteration 456 (2.42411 iter/s, 4.95027s/12 iters), loss = 5.24697 I0409 23:08:24.400746 6396 solver.cpp:237] Train net output #0: loss = 5.24697 (* 1 = 5.24697 loss) I0409 23:08:24.400758 6396 sgd_solver.cpp:105] Iteration 456, lr = 0.00913632 I0409 23:08:29.296018 6396 solver.cpp:218] Iteration 468 (2.45142 iter/s, 4.89512s/12 iters), loss = 5.16572 I0409 23:08:29.296068 6396 solver.cpp:237] Train net output #0: loss = 5.16572 (* 1 = 5.16572 loss) I0409 23:08:29.296080 6396 sgd_solver.cpp:105] Iteration 468, lr = 0.00911463 I0409 23:08:34.246402 6396 solver.cpp:218] Iteration 480 (2.42416 iter/s, 4.95017s/12 iters), loss = 5.17617 I0409 23:08:34.246495 6396 solver.cpp:237] Train net output #0: loss = 5.17617 (* 1 = 5.17617 loss) I0409 23:08:34.246506 6396 sgd_solver.cpp:105] Iteration 480, lr = 0.00909299 I0409 23:08:39.216249 6396 solver.cpp:218] Iteration 492 (2.41468 iter/s, 4.9696s/12 iters), loss = 5.19176 I0409 23:08:39.216305 6396 solver.cpp:237] Train net output #0: loss = 5.19176 (* 1 = 5.19176 loss) I0409 23:08:39.216316 6396 sgd_solver.cpp:105] Iteration 492, lr = 0.0090714 I0409 23:08:44.172780 6396 solver.cpp:218] Iteration 504 (2.42115 iter/s, 4.95632s/12 iters), loss = 5.18939 I0409 23:08:44.172818 6396 solver.cpp:237] Train net output #0: loss = 5.18939 (* 1 = 5.18939 loss) I0409 23:08:44.172828 6396 sgd_solver.cpp:105] Iteration 504, lr = 0.00904986 I0409 23:08:44.419914 6400 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:08:46.180029 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_510.caffemodel I0409 23:08:46.675359 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_510.solverstate I0409 23:08:46.993366 6396 solver.cpp:330] Iteration 510, Testing net (#0) I0409 23:08:46.993386 6396 net.cpp:676] Ignoring source layer train-data I0409 23:08:51.263931 6401 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:08:51.500875 6396 solver.cpp:397] Test net output #0: accuracy = 0.00857843 I0409 23:08:51.500921 6396 solver.cpp:397] Test net output #1: loss = 5.17309 (* 1 = 5.17309 loss) I0409 23:08:53.453568 6396 solver.cpp:218] Iteration 516 (1.29304 iter/s, 9.28047s/12 iters), loss = 5.15812 I0409 23:08:53.453613 6396 solver.cpp:237] Train net output #0: loss = 5.15812 (* 1 = 5.15812 loss) I0409 23:08:53.453621 6396 sgd_solver.cpp:105] Iteration 516, lr = 0.00902838 I0409 23:08:58.383664 6396 solver.cpp:218] Iteration 528 (2.43413 iter/s, 4.92989s/12 iters), loss = 5.21556 I0409 23:08:58.383710 6396 solver.cpp:237] Train net output #0: loss = 5.21556 (* 1 = 5.21556 loss) I0409 23:08:58.383719 6396 sgd_solver.cpp:105] Iteration 528, lr = 0.00900694 I0409 23:09:03.438783 6396 solver.cpp:218] Iteration 540 (2.37393 iter/s, 5.05491s/12 iters), loss = 5.16067 I0409 23:09:03.438830 6396 solver.cpp:237] Train net output #0: loss = 5.16067 (* 1 = 5.16067 loss) I0409 23:09:03.438840 6396 sgd_solver.cpp:105] Iteration 540, lr = 0.00898556 I0409 23:09:08.430047 6396 solver.cpp:218] Iteration 552 (2.4043 iter/s, 4.99106s/12 iters), loss = 5.10233 I0409 23:09:08.430193 6396 solver.cpp:237] Train net output #0: loss = 5.10233 (* 1 = 5.10233 loss) I0409 23:09:08.430207 6396 sgd_solver.cpp:105] Iteration 552, lr = 0.00896423 I0409 23:09:13.356006 6396 solver.cpp:218] Iteration 564 (2.43622 iter/s, 4.92566s/12 iters), loss = 5.18275 I0409 23:09:13.356046 6396 solver.cpp:237] Train net output #0: loss = 5.18275 (* 1 = 5.18275 loss) I0409 23:09:13.356055 6396 sgd_solver.cpp:105] Iteration 564, lr = 0.00894294 I0409 23:09:18.412827 6396 solver.cpp:218] Iteration 576 (2.37312 iter/s, 5.05662s/12 iters), loss = 5.14567 I0409 23:09:18.412869 6396 solver.cpp:237] Train net output #0: loss = 5.14567 (* 1 = 5.14567 loss) I0409 23:09:18.412879 6396 sgd_solver.cpp:105] Iteration 576, lr = 0.00892171 I0409 23:09:23.542287 6396 solver.cpp:218] Iteration 588 (2.33953 iter/s, 5.12924s/12 iters), loss = 5.15818 I0409 23:09:23.542341 6396 solver.cpp:237] Train net output #0: loss = 5.15818 (* 1 = 5.15818 loss) I0409 23:09:23.542353 6396 sgd_solver.cpp:105] Iteration 588, lr = 0.00890053 I0409 23:09:28.672533 6396 solver.cpp:218] Iteration 600 (2.33917 iter/s, 5.13003s/12 iters), loss = 5.17018 I0409 23:09:28.672587 6396 solver.cpp:237] Train net output #0: loss = 5.17018 (* 1 = 5.17018 loss) I0409 23:09:28.672600 6396 sgd_solver.cpp:105] Iteration 600, lr = 0.0088794 I0409 23:09:31.050797 6400 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:09:33.187615 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_612.caffemodel I0409 23:09:33.726255 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_612.solverstate I0409 23:09:34.047101 6396 solver.cpp:330] Iteration 612, Testing net (#0) I0409 23:09:34.047128 6396 net.cpp:676] Ignoring source layer train-data I0409 23:09:38.380204 6401 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:09:38.681798 6396 solver.cpp:397] Test net output #0: accuracy = 0.00796569 I0409 23:09:38.681874 6396 solver.cpp:397] Test net output #1: loss = 5.15308 (* 1 = 5.15308 loss) I0409 23:09:38.765374 6396 solver.cpp:218] Iteration 612 (1.189 iter/s, 10.0925s/12 iters), loss = 5.19614 I0409 23:09:38.765420 6396 solver.cpp:237] Train net output #0: loss = 5.19614 (* 1 = 5.19614 loss) I0409 23:09:38.765430 6396 sgd_solver.cpp:105] Iteration 612, lr = 0.00885831 I0409 23:09:43.021528 6396 solver.cpp:218] Iteration 624 (2.81957 iter/s, 4.25597s/12 iters), loss = 5.17875 I0409 23:09:43.021580 6396 solver.cpp:237] Train net output #0: loss = 5.17875 (* 1 = 5.17875 loss) I0409 23:09:43.021593 6396 sgd_solver.cpp:105] Iteration 624, lr = 0.00883728 I0409 23:09:48.006983 6396 solver.cpp:218] Iteration 636 (2.4071 iter/s, 4.98525s/12 iters), loss = 5.06564 I0409 23:09:48.007014 6396 solver.cpp:237] Train net output #0: loss = 5.06564 (* 1 = 5.06564 loss) I0409 23:09:48.007021 6396 sgd_solver.cpp:105] Iteration 636, lr = 0.0088163 I0409 23:09:53.000921 6396 solver.cpp:218] Iteration 648 (2.403 iter/s, 4.99375s/12 iters), loss = 5.18829 I0409 23:09:53.000957 6396 solver.cpp:237] Train net output #0: loss = 5.18829 (* 1 = 5.18829 loss) I0409 23:09:53.000967 6396 sgd_solver.cpp:105] Iteration 648, lr = 0.00879537 I0409 23:09:58.187526 6396 solver.cpp:218] Iteration 660 (2.31374 iter/s, 5.1864s/12 iters), loss = 5.16724 I0409 23:09:58.187574 6396 solver.cpp:237] Train net output #0: loss = 5.16724 (* 1 = 5.16724 loss) I0409 23:09:58.187583 6396 sgd_solver.cpp:105] Iteration 660, lr = 0.00877449 I0409 23:10:03.135460 6396 solver.cpp:218] Iteration 672 (2.42535 iter/s, 4.94773s/12 iters), loss = 5.12689 I0409 23:10:03.135499 6396 solver.cpp:237] Train net output #0: loss = 5.12689 (* 1 = 5.12689 loss) I0409 23:10:03.135507 6396 sgd_solver.cpp:105] Iteration 672, lr = 0.00875366 I0409 23:10:07.096614 6396 blocking_queue.cpp:49] Waiting for data I0409 23:10:07.961032 6396 solver.cpp:218] Iteration 684 (2.48685 iter/s, 4.82538s/12 iters), loss = 5.00176 I0409 23:10:07.961087 6396 solver.cpp:237] Train net output #0: loss = 5.00176 (* 1 = 5.00176 loss) I0409 23:10:07.961099 6396 sgd_solver.cpp:105] Iteration 684, lr = 0.00873287 I0409 23:10:12.847134 6396 solver.cpp:218] Iteration 696 (2.45605 iter/s, 4.88589s/12 iters), loss = 5.16521 I0409 23:10:12.847290 6396 solver.cpp:237] Train net output #0: loss = 5.16521 (* 1 = 5.16521 loss) I0409 23:10:12.847302 6396 sgd_solver.cpp:105] Iteration 696, lr = 0.00871214 I0409 23:10:17.381686 6400 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:10:17.760285 6396 solver.cpp:218] Iteration 708 (2.44257 iter/s, 4.91285s/12 iters), loss = 5.16187 I0409 23:10:17.760337 6396 solver.cpp:237] Train net output #0: loss = 5.16187 (* 1 = 5.16187 loss) I0409 23:10:17.760352 6396 sgd_solver.cpp:105] Iteration 708, lr = 0.00869145 I0409 23:10:19.738740 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_714.caffemodel I0409 23:10:21.973515 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_714.solverstate I0409 23:10:22.824321 6396 solver.cpp:330] Iteration 714, Testing net (#0) I0409 23:10:22.824352 6396 net.cpp:676] Ignoring source layer train-data I0409 23:10:27.192025 6401 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:10:27.509991 6396 solver.cpp:397] Test net output #0: accuracy = 0.00796569 I0409 23:10:27.510022 6396 solver.cpp:397] Test net output #1: loss = 5.14474 (* 1 = 5.14474 loss) I0409 23:10:29.433158 6396 solver.cpp:218] Iteration 720 (1.02806 iter/s, 11.6725s/12 iters), loss = 5.17338 I0409 23:10:29.433210 6396 solver.cpp:237] Train net output #0: loss = 5.17338 (* 1 = 5.17338 loss) I0409 23:10:29.433223 6396 sgd_solver.cpp:105] Iteration 720, lr = 0.00867082 I0409 23:10:34.451440 6396 solver.cpp:218] Iteration 732 (2.39136 iter/s, 5.01807s/12 iters), loss = 5.11556 I0409 23:10:34.451488 6396 solver.cpp:237] Train net output #0: loss = 5.11556 (* 1 = 5.11556 loss) I0409 23:10:34.451498 6396 sgd_solver.cpp:105] Iteration 732, lr = 0.00865023 I0409 23:10:39.334790 6396 solver.cpp:218] Iteration 744 (2.45743 iter/s, 4.88315s/12 iters), loss = 5.11764 I0409 23:10:39.334836 6396 solver.cpp:237] Train net output #0: loss = 5.11764 (* 1 = 5.11764 loss) I0409 23:10:39.334847 6396 sgd_solver.cpp:105] Iteration 744, lr = 0.0086297 I0409 23:10:44.261577 6396 solver.cpp:218] Iteration 756 (2.43577 iter/s, 4.92658s/12 iters), loss = 5.11053 I0409 23:10:44.263350 6396 solver.cpp:237] Train net output #0: loss = 5.11053 (* 1 = 5.11053 loss) I0409 23:10:44.263368 6396 sgd_solver.cpp:105] Iteration 756, lr = 0.00860921 I0409 23:10:49.212472 6396 solver.cpp:218] Iteration 768 (2.42474 iter/s, 4.94898s/12 iters), loss = 5.14857 I0409 23:10:49.212522 6396 solver.cpp:237] Train net output #0: loss = 5.14857 (* 1 = 5.14857 loss) I0409 23:10:49.212533 6396 sgd_solver.cpp:105] Iteration 768, lr = 0.00858877 I0409 23:10:54.105814 6396 solver.cpp:218] Iteration 780 (2.45241 iter/s, 4.89315s/12 iters), loss = 5.18963 I0409 23:10:54.105854 6396 solver.cpp:237] Train net output #0: loss = 5.18963 (* 1 = 5.18963 loss) I0409 23:10:54.105863 6396 sgd_solver.cpp:105] Iteration 780, lr = 0.00856838 I0409 23:10:59.018446 6396 solver.cpp:218] Iteration 792 (2.44278 iter/s, 4.91243s/12 iters), loss = 5.0572 I0409 23:10:59.018498 6396 solver.cpp:237] Train net output #0: loss = 5.0572 (* 1 = 5.0572 loss) I0409 23:10:59.018512 6396 sgd_solver.cpp:105] Iteration 792, lr = 0.00854803 I0409 23:11:04.005082 6396 solver.cpp:218] Iteration 804 (2.40653 iter/s, 4.98643s/12 iters), loss = 5.12295 I0409 23:11:04.005131 6396 solver.cpp:237] Train net output #0: loss = 5.12295 (* 1 = 5.12295 loss) I0409 23:11:04.005143 6396 sgd_solver.cpp:105] Iteration 804, lr = 0.00852774 I0409 23:11:05.717293 6400 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:11:08.525118 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_816.caffemodel I0409 23:11:09.007797 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_816.solverstate I0409 23:11:09.326476 6396 solver.cpp:330] Iteration 816, Testing net (#0) I0409 23:11:09.326499 6396 net.cpp:676] Ignoring source layer train-data I0409 23:11:13.609239 6401 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:11:13.991477 6396 solver.cpp:397] Test net output #0: accuracy = 0.0104167 I0409 23:11:13.991508 6396 solver.cpp:397] Test net output #1: loss = 5.11371 (* 1 = 5.11371 loss) I0409 23:11:14.074678 6396 solver.cpp:218] Iteration 816 (1.19175 iter/s, 10.0692s/12 iters), loss = 5.12455 I0409 23:11:14.074723 6396 solver.cpp:237] Train net output #0: loss = 5.12455 (* 1 = 5.12455 loss) I0409 23:11:14.074733 6396 sgd_solver.cpp:105] Iteration 816, lr = 0.00850749 I0409 23:11:18.372727 6396 solver.cpp:218] Iteration 828 (2.79208 iter/s, 4.29787s/12 iters), loss = 5.15723 I0409 23:11:18.374475 6396 solver.cpp:237] Train net output #0: loss = 5.15723 (* 1 = 5.15723 loss) I0409 23:11:18.374488 6396 sgd_solver.cpp:105] Iteration 828, lr = 0.00848729 I0409 23:11:23.330358 6396 solver.cpp:218] Iteration 840 (2.42144 iter/s, 4.95572s/12 iters), loss = 5.06996 I0409 23:11:23.330410 6396 solver.cpp:237] Train net output #0: loss = 5.06996 (* 1 = 5.06996 loss) I0409 23:11:23.330422 6396 sgd_solver.cpp:105] Iteration 840, lr = 0.00846714 I0409 23:11:28.588757 6396 solver.cpp:218] Iteration 852 (2.28216 iter/s, 5.25818s/12 iters), loss = 5.07888 I0409 23:11:28.588805 6396 solver.cpp:237] Train net output #0: loss = 5.07888 (* 1 = 5.07888 loss) I0409 23:11:28.588815 6396 sgd_solver.cpp:105] Iteration 852, lr = 0.00844704 I0409 23:11:33.706398 6396 solver.cpp:218] Iteration 864 (2.34493 iter/s, 5.11743s/12 iters), loss = 5.10176 I0409 23:11:33.706449 6396 solver.cpp:237] Train net output #0: loss = 5.10176 (* 1 = 5.10176 loss) I0409 23:11:33.706462 6396 sgd_solver.cpp:105] Iteration 864, lr = 0.00842698 I0409 23:11:39.106531 6396 solver.cpp:218] Iteration 876 (2.22226 iter/s, 5.39991s/12 iters), loss = 5.12121 I0409 23:11:39.106580 6396 solver.cpp:237] Train net output #0: loss = 5.12121 (* 1 = 5.12121 loss) I0409 23:11:39.106591 6396 sgd_solver.cpp:105] Iteration 876, lr = 0.00840698 I0409 23:11:44.202370 6396 solver.cpp:218] Iteration 888 (2.35496 iter/s, 5.09563s/12 iters), loss = 4.97436 I0409 23:11:44.202423 6396 solver.cpp:237] Train net output #0: loss = 4.97436 (* 1 = 4.97436 loss) I0409 23:11:44.202435 6396 sgd_solver.cpp:105] Iteration 888, lr = 0.00838702 I0409 23:11:49.439241 6396 solver.cpp:218] Iteration 900 (2.29154 iter/s, 5.23665s/12 iters), loss = 5.14885 I0409 23:11:49.439404 6396 solver.cpp:237] Train net output #0: loss = 5.14885 (* 1 = 5.14885 loss) I0409 23:11:49.439419 6396 sgd_solver.cpp:105] Iteration 900, lr = 0.0083671 I0409 23:11:53.412437 6400 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:11:54.582702 6396 solver.cpp:218] Iteration 912 (2.3332 iter/s, 5.14314s/12 iters), loss = 4.94136 I0409 23:11:54.582744 6396 solver.cpp:237] Train net output #0: loss = 4.94136 (* 1 = 4.94136 loss) I0409 23:11:54.582754 6396 sgd_solver.cpp:105] Iteration 912, lr = 0.00834724 I0409 23:11:56.544286 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_918.caffemodel I0409 23:11:57.030730 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_918.solverstate I0409 23:11:57.354992 6396 solver.cpp:330] Iteration 918, Testing net (#0) I0409 23:11:57.355023 6396 net.cpp:676] Ignoring source layer train-data I0409 23:12:01.379563 6401 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:12:01.778394 6396 solver.cpp:397] Test net output #0: accuracy = 0.0153186 I0409 23:12:01.778435 6396 solver.cpp:397] Test net output #1: loss = 5.04566 (* 1 = 5.04566 loss) I0409 23:12:03.647375 6396 solver.cpp:218] Iteration 924 (1.32387 iter/s, 9.06435s/12 iters), loss = 5.09454 I0409 23:12:03.647431 6396 solver.cpp:237] Train net output #0: loss = 5.09454 (* 1 = 5.09454 loss) I0409 23:12:03.647444 6396 sgd_solver.cpp:105] Iteration 924, lr = 0.00832742 I0409 23:12:08.624416 6396 solver.cpp:218] Iteration 936 (2.41118 iter/s, 4.97682s/12 iters), loss = 5.15762 I0409 23:12:08.624471 6396 solver.cpp:237] Train net output #0: loss = 5.15762 (* 1 = 5.15762 loss) I0409 23:12:08.624485 6396 sgd_solver.cpp:105] Iteration 936, lr = 0.00830765 I0409 23:12:13.545991 6396 solver.cpp:218] Iteration 948 (2.43835 iter/s, 4.92135s/12 iters), loss = 5.04396 I0409 23:12:13.546041 6396 solver.cpp:237] Train net output #0: loss = 5.04396 (* 1 = 5.04396 loss) I0409 23:12:13.546051 6396 sgd_solver.cpp:105] Iteration 948, lr = 0.00828793 I0409 23:12:18.511624 6396 solver.cpp:218] Iteration 960 (2.41671 iter/s, 4.96543s/12 iters), loss = 5.00934 I0409 23:12:18.511672 6396 solver.cpp:237] Train net output #0: loss = 5.00934 (* 1 = 5.00934 loss) I0409 23:12:18.511679 6396 sgd_solver.cpp:105] Iteration 960, lr = 0.00826825 I0409 23:12:23.546980 6396 solver.cpp:218] Iteration 972 (2.38325 iter/s, 5.03515s/12 iters), loss = 5.08225 I0409 23:12:23.547080 6396 solver.cpp:237] Train net output #0: loss = 5.08225 (* 1 = 5.08225 loss) I0409 23:12:23.547091 6396 sgd_solver.cpp:105] Iteration 972, lr = 0.00824862 I0409 23:12:28.684842 6396 solver.cpp:218] Iteration 984 (2.33572 iter/s, 5.1376s/12 iters), loss = 5.06988 I0409 23:12:28.684892 6396 solver.cpp:237] Train net output #0: loss = 5.06988 (* 1 = 5.06988 loss) I0409 23:12:28.684904 6396 sgd_solver.cpp:105] Iteration 984, lr = 0.00822903 I0409 23:12:33.641265 6396 solver.cpp:218] Iteration 996 (2.4212 iter/s, 4.95622s/12 iters), loss = 4.92916 I0409 23:12:33.641319 6396 solver.cpp:237] Train net output #0: loss = 4.92916 (* 1 = 4.92916 loss) I0409 23:12:33.641330 6396 sgd_solver.cpp:105] Iteration 996, lr = 0.0082095 I0409 23:12:38.553884 6396 solver.cpp:218] Iteration 1008 (2.44279 iter/s, 4.91241s/12 iters), loss = 5.05439 I0409 23:12:38.553938 6396 solver.cpp:237] Train net output #0: loss = 5.05439 (* 1 = 5.05439 loss) I0409 23:12:38.553949 6396 sgd_solver.cpp:105] Iteration 1008, lr = 0.00819001 I0409 23:12:39.550246 6400 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:12:43.027499 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_1020.caffemodel I0409 23:12:43.875685 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_1020.solverstate I0409 23:12:44.568985 6396 solver.cpp:330] Iteration 1020, Testing net (#0) I0409 23:12:44.569018 6396 net.cpp:676] Ignoring source layer train-data I0409 23:12:48.846451 6401 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:12:49.283015 6396 solver.cpp:397] Test net output #0: accuracy = 0.0153186 I0409 23:12:49.283066 6396 solver.cpp:397] Test net output #1: loss = 5.01896 (* 1 = 5.01896 loss) I0409 23:12:49.366451 6396 solver.cpp:218] Iteration 1020 (1.10986 iter/s, 10.8122s/12 iters), loss = 5.00835 I0409 23:12:49.366510 6396 solver.cpp:237] Train net output #0: loss = 5.00835 (* 1 = 5.00835 loss) I0409 23:12:49.366521 6396 sgd_solver.cpp:105] Iteration 1020, lr = 0.00817056 I0409 23:12:53.671056 6396 solver.cpp:218] Iteration 1032 (2.78784 iter/s, 4.30441s/12 iters), loss = 5.03018 I0409 23:12:53.671223 6396 solver.cpp:237] Train net output #0: loss = 5.03018 (* 1 = 5.03018 loss) I0409 23:12:53.671239 6396 sgd_solver.cpp:105] Iteration 1032, lr = 0.00815116 I0409 23:12:58.637876 6396 solver.cpp:218] Iteration 1044 (2.41619 iter/s, 4.9665s/12 iters), loss = 5.04159 I0409 23:12:58.637925 6396 solver.cpp:237] Train net output #0: loss = 5.04159 (* 1 = 5.04159 loss) I0409 23:12:58.637938 6396 sgd_solver.cpp:105] Iteration 1044, lr = 0.00813181 I0409 23:13:03.517397 6396 solver.cpp:218] Iteration 1056 (2.45936 iter/s, 4.87931s/12 iters), loss = 5.01864 I0409 23:13:03.517454 6396 solver.cpp:237] Train net output #0: loss = 5.01864 (* 1 = 5.01864 loss) I0409 23:13:03.517467 6396 sgd_solver.cpp:105] Iteration 1056, lr = 0.0081125 I0409 23:13:08.506633 6396 solver.cpp:218] Iteration 1068 (2.40528 iter/s, 4.98902s/12 iters), loss = 5.04865 I0409 23:13:08.506681 6396 solver.cpp:237] Train net output #0: loss = 5.04865 (* 1 = 5.04865 loss) I0409 23:13:08.506693 6396 sgd_solver.cpp:105] Iteration 1068, lr = 0.00809324 I0409 23:13:13.349357 6396 solver.cpp:218] Iteration 1080 (2.47804 iter/s, 4.84253s/12 iters), loss = 5.00438 I0409 23:13:13.349402 6396 solver.cpp:237] Train net output #0: loss = 5.00438 (* 1 = 5.00438 loss) I0409 23:13:13.349412 6396 sgd_solver.cpp:105] Iteration 1080, lr = 0.00807403 I0409 23:13:18.298717 6396 solver.cpp:218] Iteration 1092 (2.42466 iter/s, 4.94915s/12 iters), loss = 5.01233 I0409 23:13:18.298774 6396 solver.cpp:237] Train net output #0: loss = 5.01233 (* 1 = 5.01233 loss) I0409 23:13:18.298786 6396 sgd_solver.cpp:105] Iteration 1092, lr = 0.00805486 I0409 23:13:23.220844 6396 solver.cpp:218] Iteration 1104 (2.43807 iter/s, 4.92192s/12 iters), loss = 4.95184 I0409 23:13:23.220892 6396 solver.cpp:237] Train net output #0: loss = 4.95184 (* 1 = 4.95184 loss) I0409 23:13:23.220903 6396 sgd_solver.cpp:105] Iteration 1104, lr = 0.00803573 I0409 23:13:26.307430 6400 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:13:28.129155 6396 solver.cpp:218] Iteration 1116 (2.44493 iter/s, 4.90811s/12 iters), loss = 5.06477 I0409 23:13:28.129197 6396 solver.cpp:237] Train net output #0: loss = 5.06477 (* 1 = 5.06477 loss) I0409 23:13:28.129205 6396 sgd_solver.cpp:105] Iteration 1116, lr = 0.00801666 I0409 23:13:30.140933 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_1122.caffemodel I0409 23:13:31.180573 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_1122.solverstate I0409 23:13:32.329027 6396 solver.cpp:330] Iteration 1122, Testing net (#0) I0409 23:13:32.329052 6396 net.cpp:676] Ignoring source layer train-data I0409 23:13:36.353238 6401 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:13:36.827795 6396 solver.cpp:397] Test net output #0: accuracy = 0.0208333 I0409 23:13:36.827836 6396 solver.cpp:397] Test net output #1: loss = 4.98487 (* 1 = 4.98487 loss) I0409 23:13:38.527520 6396 solver.cpp:218] Iteration 1128 (1.15407 iter/s, 10.398s/12 iters), loss = 5.09501 I0409 23:13:38.527575 6396 solver.cpp:237] Train net output #0: loss = 5.09501 (* 1 = 5.09501 loss) I0409 23:13:38.527586 6396 sgd_solver.cpp:105] Iteration 1128, lr = 0.00799762 I0409 23:13:43.488430 6396 solver.cpp:218] Iteration 1140 (2.41901 iter/s, 4.9607s/12 iters), loss = 5.00849 I0409 23:13:43.488476 6396 solver.cpp:237] Train net output #0: loss = 5.00849 (* 1 = 5.00849 loss) I0409 23:13:43.488487 6396 sgd_solver.cpp:105] Iteration 1140, lr = 0.00797863 I0409 23:13:48.396610 6396 solver.cpp:218] Iteration 1152 (2.445 iter/s, 4.90798s/12 iters), loss = 4.87203 I0409 23:13:48.396656 6396 solver.cpp:237] Train net output #0: loss = 4.87203 (* 1 = 4.87203 loss) I0409 23:13:48.396667 6396 sgd_solver.cpp:105] Iteration 1152, lr = 0.00795969 I0409 23:13:53.322038 6396 solver.cpp:218] Iteration 1164 (2.43643 iter/s, 4.92523s/12 iters), loss = 4.93016 I0409 23:13:53.322091 6396 solver.cpp:237] Train net output #0: loss = 4.93016 (* 1 = 4.93016 loss) I0409 23:13:53.322105 6396 sgd_solver.cpp:105] Iteration 1164, lr = 0.00794079 I0409 23:13:58.448644 6396 solver.cpp:218] Iteration 1176 (2.34083 iter/s, 5.12639s/12 iters), loss = 4.95692 I0409 23:13:58.448763 6396 solver.cpp:237] Train net output #0: loss = 4.95692 (* 1 = 4.95692 loss) I0409 23:13:58.448776 6396 sgd_solver.cpp:105] Iteration 1176, lr = 0.00792194 I0409 23:14:03.308151 6396 solver.cpp:218] Iteration 1188 (2.46952 iter/s, 4.85924s/12 iters), loss = 4.98052 I0409 23:14:03.308194 6396 solver.cpp:237] Train net output #0: loss = 4.98052 (* 1 = 4.98052 loss) I0409 23:14:03.308203 6396 sgd_solver.cpp:105] Iteration 1188, lr = 0.00790313 I0409 23:14:08.438724 6396 solver.cpp:218] Iteration 1200 (2.33901 iter/s, 5.13037s/12 iters), loss = 5.02275 I0409 23:14:08.438771 6396 solver.cpp:237] Train net output #0: loss = 5.02275 (* 1 = 5.02275 loss) I0409 23:14:08.438782 6396 sgd_solver.cpp:105] Iteration 1200, lr = 0.00788437 I0409 23:14:13.384367 6396 solver.cpp:218] Iteration 1212 (2.42648 iter/s, 4.94544s/12 iters), loss = 5.00519 I0409 23:14:13.384414 6396 solver.cpp:237] Train net output #0: loss = 5.00519 (* 1 = 5.00519 loss) I0409 23:14:13.384428 6396 sgd_solver.cpp:105] Iteration 1212, lr = 0.00786565 I0409 23:14:13.662586 6400 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:14:17.849836 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_1224.caffemodel I0409 23:14:18.282027 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_1224.solverstate I0409 23:14:18.608780 6396 solver.cpp:330] Iteration 1224, Testing net (#0) I0409 23:14:18.608805 6396 net.cpp:676] Ignoring source layer train-data I0409 23:14:22.569582 6401 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:14:23.173488 6396 solver.cpp:397] Test net output #0: accuracy = 0.0300245 I0409 23:14:23.173525 6396 solver.cpp:397] Test net output #1: loss = 4.9107 (* 1 = 4.9107 loss) I0409 23:14:23.256559 6396 solver.cpp:218] Iteration 1224 (1.21558 iter/s, 9.87185s/12 iters), loss = 4.87699 I0409 23:14:23.256619 6396 solver.cpp:237] Train net output #0: loss = 4.87699 (* 1 = 4.87699 loss) I0409 23:14:23.256631 6396 sgd_solver.cpp:105] Iteration 1224, lr = 0.00784697 I0409 23:14:27.579262 6396 solver.cpp:218] Iteration 1236 (2.77617 iter/s, 4.32251s/12 iters), loss = 5.04432 I0409 23:14:27.579309 6396 solver.cpp:237] Train net output #0: loss = 5.04432 (* 1 = 5.04432 loss) I0409 23:14:27.579319 6396 sgd_solver.cpp:105] Iteration 1236, lr = 0.00782834 I0409 23:14:32.533716 6396 solver.cpp:218] Iteration 1248 (2.42216 iter/s, 4.95425s/12 iters), loss = 4.90206 I0409 23:14:32.533808 6396 solver.cpp:237] Train net output #0: loss = 4.90206 (* 1 = 4.90206 loss) I0409 23:14:32.533823 6396 sgd_solver.cpp:105] Iteration 1248, lr = 0.00780976 I0409 23:14:37.476972 6396 solver.cpp:218] Iteration 1260 (2.42767 iter/s, 4.94301s/12 iters), loss = 4.87208 I0409 23:14:37.477025 6396 solver.cpp:237] Train net output #0: loss = 4.87208 (* 1 = 4.87208 loss) I0409 23:14:37.477037 6396 sgd_solver.cpp:105] Iteration 1260, lr = 0.00779122 I0409 23:14:42.509850 6396 solver.cpp:218] Iteration 1272 (2.38442 iter/s, 5.03266s/12 iters), loss = 4.81342 I0409 23:14:42.509898 6396 solver.cpp:237] Train net output #0: loss = 4.81342 (* 1 = 4.81342 loss) I0409 23:14:42.509907 6396 sgd_solver.cpp:105] Iteration 1272, lr = 0.00777272 I0409 23:14:47.617373 6396 solver.cpp:218] Iteration 1284 (2.34957 iter/s, 5.10731s/12 iters), loss = 4.91452 I0409 23:14:47.617432 6396 solver.cpp:237] Train net output #0: loss = 4.91452 (* 1 = 4.91452 loss) I0409 23:14:47.617444 6396 sgd_solver.cpp:105] Iteration 1284, lr = 0.00775426 I0409 23:14:52.583127 6396 solver.cpp:218] Iteration 1296 (2.41666 iter/s, 4.96554s/12 iters), loss = 4.71023 I0409 23:14:52.583181 6396 solver.cpp:237] Train net output #0: loss = 4.71023 (* 1 = 4.71023 loss) I0409 23:14:52.583194 6396 sgd_solver.cpp:105] Iteration 1296, lr = 0.00773585 I0409 23:14:57.575850 6396 solver.cpp:218] Iteration 1308 (2.4036 iter/s, 4.99251s/12 iters), loss = 4.85882 I0409 23:14:57.575912 6396 solver.cpp:237] Train net output #0: loss = 4.85882 (* 1 = 4.85882 loss) I0409 23:14:57.575928 6396 sgd_solver.cpp:105] Iteration 1308, lr = 0.00771749 I0409 23:15:00.062047 6400 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:15:02.561372 6396 solver.cpp:218] Iteration 1320 (2.40707 iter/s, 4.98531s/12 iters), loss = 4.83453 I0409 23:15:02.561504 6396 solver.cpp:237] Train net output #0: loss = 4.83453 (* 1 = 4.83453 loss) I0409 23:15:02.561518 6396 sgd_solver.cpp:105] Iteration 1320, lr = 0.00769916 I0409 23:15:04.726630 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_1326.caffemodel I0409 23:15:05.181049 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_1326.solverstate I0409 23:15:05.497941 6396 solver.cpp:330] Iteration 1326, Testing net (#0) I0409 23:15:05.497992 6396 net.cpp:676] Ignoring source layer train-data I0409 23:15:09.406265 6401 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:15:09.960430 6396 solver.cpp:397] Test net output #0: accuracy = 0.0300245 I0409 23:15:09.960480 6396 solver.cpp:397] Test net output #1: loss = 4.83378 (* 1 = 4.83378 loss) I0409 23:15:11.845858 6396 solver.cpp:218] Iteration 1332 (1.29254 iter/s, 9.28408s/12 iters), loss = 4.80834 I0409 23:15:11.845906 6396 solver.cpp:237] Train net output #0: loss = 4.80834 (* 1 = 4.80834 loss) I0409 23:15:11.845918 6396 sgd_solver.cpp:105] Iteration 1332, lr = 0.00768088 I0409 23:15:16.891418 6396 solver.cpp:218] Iteration 1344 (2.37843 iter/s, 5.04535s/12 iters), loss = 4.66146 I0409 23:15:16.891471 6396 solver.cpp:237] Train net output #0: loss = 4.66146 (* 1 = 4.66146 loss) I0409 23:15:16.891482 6396 sgd_solver.cpp:105] Iteration 1344, lr = 0.00766265 I0409 23:15:22.133783 6396 solver.cpp:218] Iteration 1356 (2.28914 iter/s, 5.24215s/12 iters), loss = 4.87689 I0409 23:15:22.133833 6396 solver.cpp:237] Train net output #0: loss = 4.87689 (* 1 = 4.87689 loss) I0409 23:15:22.133846 6396 sgd_solver.cpp:105] Iteration 1356, lr = 0.00764446 I0409 23:15:26.868096 6396 blocking_queue.cpp:49] Waiting for data I0409 23:15:27.318863 6396 solver.cpp:218] Iteration 1368 (2.31443 iter/s, 5.18486s/12 iters), loss = 4.85403 I0409 23:15:27.318933 6396 solver.cpp:237] Train net output #0: loss = 4.85403 (* 1 = 4.85403 loss) I0409 23:15:27.318949 6396 sgd_solver.cpp:105] Iteration 1368, lr = 0.00762631 I0409 23:15:32.232381 6396 solver.cpp:218] Iteration 1380 (2.44235 iter/s, 4.9133s/12 iters), loss = 4.70112 I0409 23:15:32.232421 6396 solver.cpp:237] Train net output #0: loss = 4.70112 (* 1 = 4.70112 loss) I0409 23:15:32.232429 6396 sgd_solver.cpp:105] Iteration 1380, lr = 0.0076082 I0409 23:15:37.124405 6396 solver.cpp:218] Iteration 1392 (2.45307 iter/s, 4.89183s/12 iters), loss = 4.56327 I0409 23:15:37.124519 6396 solver.cpp:237] Train net output #0: loss = 4.56327 (* 1 = 4.56327 loss) I0409 23:15:37.124532 6396 sgd_solver.cpp:105] Iteration 1392, lr = 0.00759014 I0409 23:15:42.072276 6396 solver.cpp:218] Iteration 1404 (2.42542 iter/s, 4.94761s/12 iters), loss = 4.78942 I0409 23:15:42.072324 6396 solver.cpp:237] Train net output #0: loss = 4.78942 (* 1 = 4.78942 loss) I0409 23:15:42.072335 6396 sgd_solver.cpp:105] Iteration 1404, lr = 0.00757212 I0409 23:15:46.891553 6400 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:15:47.245496 6396 solver.cpp:218] Iteration 1416 (2.31974 iter/s, 5.173s/12 iters), loss = 4.87584 I0409 23:15:47.245548 6396 solver.cpp:237] Train net output #0: loss = 4.87584 (* 1 = 4.87584 loss) I0409 23:15:47.245559 6396 sgd_solver.cpp:105] Iteration 1416, lr = 0.00755414 I0409 23:15:51.779784 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_1428.caffemodel I0409 23:15:52.660035 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_1428.solverstate I0409 23:15:53.435796 6396 solver.cpp:330] Iteration 1428, Testing net (#0) I0409 23:15:53.435815 6396 net.cpp:676] Ignoring source layer train-data I0409 23:15:57.448776 6401 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:15:58.037214 6396 solver.cpp:397] Test net output #0: accuracy = 0.0392157 I0409 23:15:58.037264 6396 solver.cpp:397] Test net output #1: loss = 4.80582 (* 1 = 4.80582 loss) I0409 23:15:58.120535 6396 solver.cpp:218] Iteration 1428 (1.10348 iter/s, 10.8747s/12 iters), loss = 4.89245 I0409 23:15:58.120594 6396 solver.cpp:237] Train net output #0: loss = 4.89245 (* 1 = 4.89245 loss) I0409 23:15:58.120605 6396 sgd_solver.cpp:105] Iteration 1428, lr = 0.0075362 I0409 23:16:02.433589 6396 solver.cpp:218] Iteration 1440 (2.78238 iter/s, 4.31286s/12 iters), loss = 4.58012 I0409 23:16:02.433632 6396 solver.cpp:237] Train net output #0: loss = 4.58012 (* 1 = 4.58012 loss) I0409 23:16:02.433645 6396 sgd_solver.cpp:105] Iteration 1440, lr = 0.00751831 I0409 23:16:07.535856 6396 solver.cpp:218] Iteration 1452 (2.35199 iter/s, 5.10206s/12 iters), loss = 4.75584 I0409 23:16:07.536021 6396 solver.cpp:237] Train net output #0: loss = 4.75584 (* 1 = 4.75584 loss) I0409 23:16:07.536041 6396 sgd_solver.cpp:105] Iteration 1452, lr = 0.00750046 I0409 23:16:12.538585 6396 solver.cpp:218] Iteration 1464 (2.39884 iter/s, 5.00242s/12 iters), loss = 4.79226 I0409 23:16:12.538640 6396 solver.cpp:237] Train net output #0: loss = 4.79226 (* 1 = 4.79226 loss) I0409 23:16:12.538651 6396 sgd_solver.cpp:105] Iteration 1464, lr = 0.00748265 I0409 23:16:17.593446 6396 solver.cpp:218] Iteration 1476 (2.37406 iter/s, 5.05463s/12 iters), loss = 4.75938 I0409 23:16:17.593504 6396 solver.cpp:237] Train net output #0: loss = 4.75938 (* 1 = 4.75938 loss) I0409 23:16:17.593516 6396 sgd_solver.cpp:105] Iteration 1476, lr = 0.00746489 I0409 23:16:22.557855 6396 solver.cpp:218] Iteration 1488 (2.41731 iter/s, 4.96419s/12 iters), loss = 4.78902 I0409 23:16:22.557906 6396 solver.cpp:237] Train net output #0: loss = 4.78902 (* 1 = 4.78902 loss) I0409 23:16:22.557917 6396 sgd_solver.cpp:105] Iteration 1488, lr = 0.00744716 I0409 23:16:27.535130 6396 solver.cpp:218] Iteration 1500 (2.41106 iter/s, 4.97706s/12 iters), loss = 4.50856 I0409 23:16:27.535187 6396 solver.cpp:237] Train net output #0: loss = 4.50856 (* 1 = 4.50856 loss) I0409 23:16:27.535198 6396 sgd_solver.cpp:105] Iteration 1500, lr = 0.00742948 I0409 23:16:32.480310 6396 solver.cpp:218] Iteration 1512 (2.42671 iter/s, 4.94496s/12 iters), loss = 4.75476 I0409 23:16:32.480355 6396 solver.cpp:237] Train net output #0: loss = 4.75476 (* 1 = 4.75476 loss) I0409 23:16:32.480362 6396 sgd_solver.cpp:105] Iteration 1512, lr = 0.00741184 I0409 23:16:34.188673 6400 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:16:37.365240 6396 solver.cpp:218] Iteration 1524 (2.45664 iter/s, 4.88473s/12 iters), loss = 4.76191 I0409 23:16:37.365296 6396 solver.cpp:237] Train net output #0: loss = 4.76191 (* 1 = 4.76191 loss) I0409 23:16:37.365309 6396 sgd_solver.cpp:105] Iteration 1524, lr = 0.00739425 I0409 23:16:39.387161 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_1530.caffemodel I0409 23:16:39.856946 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_1530.solverstate I0409 23:16:40.254982 6396 solver.cpp:330] Iteration 1530, Testing net (#0) I0409 23:16:40.255002 6396 net.cpp:676] Ignoring source layer train-data I0409 23:16:44.056459 6401 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:16:44.710235 6396 solver.cpp:397] Test net output #0: accuracy = 0.0453431 I0409 23:16:44.710273 6396 solver.cpp:397] Test net output #1: loss = 4.63122 (* 1 = 4.63122 loss) I0409 23:16:46.478646 6396 solver.cpp:218] Iteration 1536 (1.31679 iter/s, 9.11307s/12 iters), loss = 4.8042 I0409 23:16:46.478700 6396 solver.cpp:237] Train net output #0: loss = 4.8042 (* 1 = 4.8042 loss) I0409 23:16:46.478711 6396 sgd_solver.cpp:105] Iteration 1536, lr = 0.00737669 I0409 23:16:51.465226 6396 solver.cpp:218] Iteration 1548 (2.40656 iter/s, 4.98637s/12 iters), loss = 4.39548 I0409 23:16:51.465270 6396 solver.cpp:237] Train net output #0: loss = 4.39548 (* 1 = 4.39548 loss) I0409 23:16:51.465279 6396 sgd_solver.cpp:105] Iteration 1548, lr = 0.00735918 I0409 23:16:56.490134 6396 solver.cpp:218] Iteration 1560 (2.3882 iter/s, 5.0247s/12 iters), loss = 4.57998 I0409 23:16:56.490178 6396 solver.cpp:237] Train net output #0: loss = 4.57998 (* 1 = 4.57998 loss) I0409 23:16:56.490188 6396 sgd_solver.cpp:105] Iteration 1560, lr = 0.00734171 I0409 23:17:01.540081 6396 solver.cpp:218] Iteration 1572 (2.37636 iter/s, 5.04975s/12 iters), loss = 4.46439 I0409 23:17:01.540122 6396 solver.cpp:237] Train net output #0: loss = 4.46439 (* 1 = 4.46439 loss) I0409 23:17:01.540132 6396 sgd_solver.cpp:105] Iteration 1572, lr = 0.00732427 I0409 23:17:06.425592 6396 solver.cpp:218] Iteration 1584 (2.45634 iter/s, 4.88531s/12 iters), loss = 4.69496 I0409 23:17:06.425643 6396 solver.cpp:237] Train net output #0: loss = 4.69496 (* 1 = 4.69496 loss) I0409 23:17:06.425657 6396 sgd_solver.cpp:105] Iteration 1584, lr = 0.00730688 I0409 23:17:11.337610 6396 solver.cpp:218] Iteration 1596 (2.44309 iter/s, 4.91181s/12 iters), loss = 4.51098 I0409 23:17:11.337690 6396 solver.cpp:237] Train net output #0: loss = 4.51098 (* 1 = 4.51098 loss) I0409 23:17:11.337699 6396 sgd_solver.cpp:105] Iteration 1596, lr = 0.00728954 I0409 23:17:16.311223 6396 solver.cpp:218] Iteration 1608 (2.41285 iter/s, 4.97337s/12 iters), loss = 4.68196 I0409 23:17:16.311281 6396 solver.cpp:237] Train net output #0: loss = 4.68196 (* 1 = 4.68196 loss) I0409 23:17:16.311292 6396 sgd_solver.cpp:105] Iteration 1608, lr = 0.00727223 I0409 23:17:20.102902 6400 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:17:21.165529 6396 solver.cpp:218] Iteration 1620 (2.47214 iter/s, 4.85409s/12 iters), loss = 4.5233 I0409 23:17:21.165578 6396 solver.cpp:237] Train net output #0: loss = 4.5233 (* 1 = 4.5233 loss) I0409 23:17:21.165587 6396 sgd_solver.cpp:105] Iteration 1620, lr = 0.00725496 I0409 23:17:25.586316 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_1632.caffemodel I0409 23:17:26.024292 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_1632.solverstate I0409 23:17:26.333432 6396 solver.cpp:330] Iteration 1632, Testing net (#0) I0409 23:17:26.333465 6396 net.cpp:676] Ignoring source layer train-data I0409 23:17:30.169979 6401 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:17:30.840970 6396 solver.cpp:397] Test net output #0: accuracy = 0.0453431 I0409 23:17:30.841017 6396 solver.cpp:397] Test net output #1: loss = 4.57721 (* 1 = 4.57721 loss) I0409 23:17:30.924233 6396 solver.cpp:218] Iteration 1632 (1.22972 iter/s, 9.75836s/12 iters), loss = 4.66223 I0409 23:17:30.924291 6396 solver.cpp:237] Train net output #0: loss = 4.66223 (* 1 = 4.66223 loss) I0409 23:17:30.924302 6396 sgd_solver.cpp:105] Iteration 1632, lr = 0.00723774 I0409 23:17:35.202394 6396 solver.cpp:218] Iteration 1644 (2.80507 iter/s, 4.27797s/12 iters), loss = 4.53743 I0409 23:17:35.202446 6396 solver.cpp:237] Train net output #0: loss = 4.53743 (* 1 = 4.53743 loss) I0409 23:17:35.202456 6396 sgd_solver.cpp:105] Iteration 1644, lr = 0.00722056 I0409 23:17:40.153878 6396 solver.cpp:218] Iteration 1656 (2.42362 iter/s, 4.95127s/12 iters), loss = 4.50593 I0409 23:17:40.153936 6396 solver.cpp:237] Train net output #0: loss = 4.50593 (* 1 = 4.50593 loss) I0409 23:17:40.153949 6396 sgd_solver.cpp:105] Iteration 1656, lr = 0.00720341 I0409 23:17:45.333623 6396 solver.cpp:218] Iteration 1668 (2.31682 iter/s, 5.17952s/12 iters), loss = 4.33459 I0409 23:17:45.333791 6396 solver.cpp:237] Train net output #0: loss = 4.33459 (* 1 = 4.33459 loss) I0409 23:17:45.333804 6396 sgd_solver.cpp:105] Iteration 1668, lr = 0.00718631 I0409 23:17:50.468796 6396 solver.cpp:218] Iteration 1680 (2.33697 iter/s, 5.13485s/12 iters), loss = 4.4101 I0409 23:17:50.468837 6396 solver.cpp:237] Train net output #0: loss = 4.4101 (* 1 = 4.4101 loss) I0409 23:17:50.468845 6396 sgd_solver.cpp:105] Iteration 1680, lr = 0.00716925 I0409 23:17:55.340991 6396 solver.cpp:218] Iteration 1692 (2.46305 iter/s, 4.872s/12 iters), loss = 4.40928 I0409 23:17:55.341045 6396 solver.cpp:237] Train net output #0: loss = 4.40928 (* 1 = 4.40928 loss) I0409 23:17:55.341056 6396 sgd_solver.cpp:105] Iteration 1692, lr = 0.00715223 I0409 23:18:00.218047 6396 solver.cpp:218] Iteration 1704 (2.46061 iter/s, 4.87685s/12 iters), loss = 4.41133 I0409 23:18:00.218108 6396 solver.cpp:237] Train net output #0: loss = 4.41133 (* 1 = 4.41133 loss) I0409 23:18:00.218124 6396 sgd_solver.cpp:105] Iteration 1704, lr = 0.00713525 I0409 23:18:05.122891 6396 solver.cpp:218] Iteration 1716 (2.44667 iter/s, 4.90463s/12 iters), loss = 4.46178 I0409 23:18:05.122948 6396 solver.cpp:237] Train net output #0: loss = 4.46178 (* 1 = 4.46178 loss) I0409 23:18:05.122961 6396 sgd_solver.cpp:105] Iteration 1716, lr = 0.00711831 I0409 23:18:06.142616 6400 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:18:10.014753 6396 solver.cpp:218] Iteration 1728 (2.45316 iter/s, 4.89165s/12 iters), loss = 4.41805 I0409 23:18:10.014816 6396 solver.cpp:237] Train net output #0: loss = 4.41805 (* 1 = 4.41805 loss) I0409 23:18:10.014827 6396 sgd_solver.cpp:105] Iteration 1728, lr = 0.00710141 I0409 23:18:12.078881 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_1734.caffemodel I0409 23:18:13.353004 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_1734.solverstate I0409 23:18:16.555765 6396 solver.cpp:330] Iteration 1734, Testing net (#0) I0409 23:18:16.555863 6396 net.cpp:676] Ignoring source layer train-data I0409 23:18:20.484387 6401 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:18:21.186722 6396 solver.cpp:397] Test net output #0: accuracy = 0.0508578 I0409 23:18:21.186787 6396 solver.cpp:397] Test net output #1: loss = 4.45494 (* 1 = 4.45494 loss) I0409 23:18:23.079223 6396 solver.cpp:218] Iteration 1740 (0.918554 iter/s, 13.064s/12 iters), loss = 4.44893 I0409 23:18:23.079311 6396 solver.cpp:237] Train net output #0: loss = 4.44893 (* 1 = 4.44893 loss) I0409 23:18:23.079322 6396 sgd_solver.cpp:105] Iteration 1740, lr = 0.00708455 I0409 23:18:28.020570 6396 solver.cpp:218] Iteration 1752 (2.42861 iter/s, 4.9411s/12 iters), loss = 4.30604 I0409 23:18:28.020615 6396 solver.cpp:237] Train net output #0: loss = 4.30604 (* 1 = 4.30604 loss) I0409 23:18:28.020625 6396 sgd_solver.cpp:105] Iteration 1752, lr = 0.00706773 I0409 23:18:33.023504 6396 solver.cpp:218] Iteration 1764 (2.39869 iter/s, 5.00273s/12 iters), loss = 4.40201 I0409 23:18:33.023550 6396 solver.cpp:237] Train net output #0: loss = 4.40201 (* 1 = 4.40201 loss) I0409 23:18:33.023558 6396 sgd_solver.cpp:105] Iteration 1764, lr = 0.00705094 I0409 23:18:38.031085 6396 solver.cpp:218] Iteration 1776 (2.39647 iter/s, 5.00737s/12 iters), loss = 4.40017 I0409 23:18:38.031128 6396 solver.cpp:237] Train net output #0: loss = 4.40017 (* 1 = 4.40017 loss) I0409 23:18:38.031138 6396 sgd_solver.cpp:105] Iteration 1776, lr = 0.0070342 I0409 23:18:42.891469 6396 solver.cpp:218] Iteration 1788 (2.46904 iter/s, 4.86018s/12 iters), loss = 4.40695 I0409 23:18:42.891516 6396 solver.cpp:237] Train net output #0: loss = 4.40695 (* 1 = 4.40695 loss) I0409 23:18:42.891525 6396 sgd_solver.cpp:105] Iteration 1788, lr = 0.0070175 I0409 23:18:47.910559 6396 solver.cpp:218] Iteration 1800 (2.39097 iter/s, 5.01888s/12 iters), loss = 4.18079 I0409 23:18:47.910701 6396 solver.cpp:237] Train net output #0: loss = 4.18079 (* 1 = 4.18079 loss) I0409 23:18:47.910713 6396 sgd_solver.cpp:105] Iteration 1800, lr = 0.00700084 I0409 23:18:52.937618 6396 solver.cpp:218] Iteration 1812 (2.38722 iter/s, 5.02676s/12 iters), loss = 4.29196 I0409 23:18:52.937667 6396 solver.cpp:237] Train net output #0: loss = 4.29196 (* 1 = 4.29196 loss) I0409 23:18:52.937678 6396 sgd_solver.cpp:105] Iteration 1812, lr = 0.00698422 I0409 23:18:56.095976 6400 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:18:57.864013 6396 solver.cpp:218] Iteration 1824 (2.43596 iter/s, 4.92619s/12 iters), loss = 4.3334 I0409 23:18:57.864076 6396 solver.cpp:237] Train net output #0: loss = 4.3334 (* 1 = 4.3334 loss) I0409 23:18:57.864089 6396 sgd_solver.cpp:105] Iteration 1824, lr = 0.00696764 I0409 23:19:02.511440 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_1836.caffemodel I0409 23:19:02.943543 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_1836.solverstate I0409 23:19:03.251252 6396 solver.cpp:330] Iteration 1836, Testing net (#0) I0409 23:19:03.251277 6396 net.cpp:676] Ignoring source layer train-data I0409 23:19:07.145830 6401 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:19:07.899140 6396 solver.cpp:397] Test net output #0: accuracy = 0.0784314 I0409 23:19:07.899170 6396 solver.cpp:397] Test net output #1: loss = 4.29694 (* 1 = 4.29694 loss) I0409 23:19:07.983600 6396 solver.cpp:218] Iteration 1836 (1.18586 iter/s, 10.1192s/12 iters), loss = 4.26144 I0409 23:19:07.983660 6396 solver.cpp:237] Train net output #0: loss = 4.26144 (* 1 = 4.26144 loss) I0409 23:19:07.983672 6396 sgd_solver.cpp:105] Iteration 1836, lr = 0.0069511 I0409 23:19:12.062814 6396 solver.cpp:218] Iteration 1848 (2.94189 iter/s, 4.07902s/12 iters), loss = 4.34542 I0409 23:19:12.062877 6396 solver.cpp:237] Train net output #0: loss = 4.34542 (* 1 = 4.34542 loss) I0409 23:19:12.062891 6396 sgd_solver.cpp:105] Iteration 1848, lr = 0.00693459 I0409 23:19:17.033352 6396 solver.cpp:218] Iteration 1860 (2.41433 iter/s, 4.97032s/12 iters), loss = 4.3279 I0409 23:19:17.033401 6396 solver.cpp:237] Train net output #0: loss = 4.3279 (* 1 = 4.3279 loss) I0409 23:19:17.033409 6396 sgd_solver.cpp:105] Iteration 1860, lr = 0.00691813 I0409 23:19:22.134953 6396 solver.cpp:218] Iteration 1872 (2.3523 iter/s, 5.10139s/12 iters), loss = 4.3064 I0409 23:19:22.135071 6396 solver.cpp:237] Train net output #0: loss = 4.3064 (* 1 = 4.3064 loss) I0409 23:19:22.135084 6396 sgd_solver.cpp:105] Iteration 1872, lr = 0.0069017 I0409 23:19:27.372979 6396 solver.cpp:218] Iteration 1884 (2.29106 iter/s, 5.23775s/12 iters), loss = 4.34952 I0409 23:19:27.373029 6396 solver.cpp:237] Train net output #0: loss = 4.34952 (* 1 = 4.34952 loss) I0409 23:19:27.373041 6396 sgd_solver.cpp:105] Iteration 1884, lr = 0.00688532 I0409 23:19:32.239475 6396 solver.cpp:218] Iteration 1896 (2.46595 iter/s, 4.86629s/12 iters), loss = 4.2163 I0409 23:19:32.239521 6396 solver.cpp:237] Train net output #0: loss = 4.2163 (* 1 = 4.2163 loss) I0409 23:19:32.239529 6396 sgd_solver.cpp:105] Iteration 1896, lr = 0.00686897 I0409 23:19:37.303911 6396 solver.cpp:218] Iteration 1908 (2.36956 iter/s, 5.06423s/12 iters), loss = 4.28224 I0409 23:19:37.303963 6396 solver.cpp:237] Train net output #0: loss = 4.28224 (* 1 = 4.28224 loss) I0409 23:19:37.303977 6396 sgd_solver.cpp:105] Iteration 1908, lr = 0.00685266 I0409 23:19:42.793198 6396 solver.cpp:218] Iteration 1920 (2.18617 iter/s, 5.48906s/12 iters), loss = 4.35912 I0409 23:19:42.793252 6396 solver.cpp:237] Train net output #0: loss = 4.35912 (* 1 = 4.35912 loss) I0409 23:19:42.793263 6396 sgd_solver.cpp:105] Iteration 1920, lr = 0.00683639 I0409 23:19:43.124256 6400 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:19:47.816746 6396 solver.cpp:218] Iteration 1932 (2.38886 iter/s, 5.02332s/12 iters), loss = 4.23774 I0409 23:19:47.816804 6396 solver.cpp:237] Train net output #0: loss = 4.23774 (* 1 = 4.23774 loss) I0409 23:19:47.816815 6396 sgd_solver.cpp:105] Iteration 1932, lr = 0.00682016 I0409 23:19:49.792270 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_1938.caffemodel I0409 23:19:50.235394 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_1938.solverstate I0409 23:19:50.533365 6396 solver.cpp:330] Iteration 1938, Testing net (#0) I0409 23:19:50.533382 6396 net.cpp:676] Ignoring source layer train-data I0409 23:19:54.174852 6401 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:19:54.959771 6396 solver.cpp:397] Test net output #0: accuracy = 0.0765931 I0409 23:19:54.959813 6396 solver.cpp:397] Test net output #1: loss = 4.15033 (* 1 = 4.15033 loss) I0409 23:19:56.875648 6396 solver.cpp:218] Iteration 1944 (1.32471 iter/s, 9.05857s/12 iters), loss = 4.26458 I0409 23:19:56.875694 6396 solver.cpp:237] Train net output #0: loss = 4.26458 (* 1 = 4.26458 loss) I0409 23:19:56.875702 6396 sgd_solver.cpp:105] Iteration 1944, lr = 0.00680397 I0409 23:20:01.825706 6396 solver.cpp:218] Iteration 1956 (2.42432 iter/s, 4.94985s/12 iters), loss = 4.32529 I0409 23:20:01.825760 6396 solver.cpp:237] Train net output #0: loss = 4.32529 (* 1 = 4.32529 loss) I0409 23:20:01.825770 6396 sgd_solver.cpp:105] Iteration 1956, lr = 0.00678782 I0409 23:20:06.776881 6396 solver.cpp:218] Iteration 1968 (2.42377 iter/s, 4.95096s/12 iters), loss = 4.2211 I0409 23:20:06.776930 6396 solver.cpp:237] Train net output #0: loss = 4.2211 (* 1 = 4.2211 loss) I0409 23:20:06.776939 6396 sgd_solver.cpp:105] Iteration 1968, lr = 0.0067717 I0409 23:20:11.752470 6396 solver.cpp:218] Iteration 1980 (2.41188 iter/s, 4.97538s/12 iters), loss = 4.0977 I0409 23:20:11.752511 6396 solver.cpp:237] Train net output #0: loss = 4.0977 (* 1 = 4.0977 loss) I0409 23:20:11.752519 6396 sgd_solver.cpp:105] Iteration 1980, lr = 0.00675562 I0409 23:20:16.666509 6396 solver.cpp:218] Iteration 1992 (2.44208 iter/s, 4.91384s/12 iters), loss = 4.17903 I0409 23:20:16.666550 6396 solver.cpp:237] Train net output #0: loss = 4.17903 (* 1 = 4.17903 loss) I0409 23:20:16.666558 6396 sgd_solver.cpp:105] Iteration 1992, lr = 0.00673958 I0409 23:20:21.528373 6396 solver.cpp:218] Iteration 2004 (2.46829 iter/s, 4.86167s/12 iters), loss = 4.06815 I0409 23:20:21.528424 6396 solver.cpp:237] Train net output #0: loss = 4.06815 (* 1 = 4.06815 loss) I0409 23:20:21.528434 6396 sgd_solver.cpp:105] Iteration 2004, lr = 0.00672358 I0409 23:20:26.554777 6396 solver.cpp:218] Iteration 2016 (2.38749 iter/s, 5.0262s/12 iters), loss = 4.24403 I0409 23:20:26.554877 6396 solver.cpp:237] Train net output #0: loss = 4.24403 (* 1 = 4.24403 loss) I0409 23:20:26.554886 6396 sgd_solver.cpp:105] Iteration 2016, lr = 0.00670762 I0409 23:20:29.084246 6400 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:20:31.509711 6396 solver.cpp:218] Iteration 2028 (2.42195 iter/s, 4.95468s/12 iters), loss = 4.00201 I0409 23:20:31.509757 6396 solver.cpp:237] Train net output #0: loss = 4.00201 (* 1 = 4.00201 loss) I0409 23:20:31.509765 6396 sgd_solver.cpp:105] Iteration 2028, lr = 0.00669169 I0409 23:20:36.060308 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_2040.caffemodel I0409 23:20:36.495018 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_2040.solverstate I0409 23:20:36.793994 6396 solver.cpp:330] Iteration 2040, Testing net (#0) I0409 23:20:36.794023 6396 net.cpp:676] Ignoring source layer train-data I0409 23:20:40.373848 6401 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:20:41.218719 6396 solver.cpp:397] Test net output #0: accuracy = 0.098652 I0409 23:20:41.218752 6396 solver.cpp:397] Test net output #1: loss = 3.96935 (* 1 = 3.96935 loss) I0409 23:20:41.302090 6396 solver.cpp:218] Iteration 2040 (1.22549 iter/s, 9.79203s/12 iters), loss = 3.91138 I0409 23:20:41.302137 6396 solver.cpp:237] Train net output #0: loss = 3.91138 (* 1 = 3.91138 loss) I0409 23:20:41.302146 6396 sgd_solver.cpp:105] Iteration 2040, lr = 0.00667581 I0409 23:20:45.418421 6396 solver.cpp:218] Iteration 2052 (2.91534 iter/s, 4.11615s/12 iters), loss = 4.02643 I0409 23:20:45.418464 6396 solver.cpp:237] Train net output #0: loss = 4.02643 (* 1 = 4.02643 loss) I0409 23:20:45.418473 6396 sgd_solver.cpp:105] Iteration 2052, lr = 0.00665996 I0409 23:20:45.418679 6396 blocking_queue.cpp:49] Waiting for data I0409 23:20:50.340124 6396 solver.cpp:218] Iteration 2064 (2.43828 iter/s, 4.92149s/12 iters), loss = 4.17505 I0409 23:20:50.340178 6396 solver.cpp:237] Train net output #0: loss = 4.17505 (* 1 = 4.17505 loss) I0409 23:20:50.340188 6396 sgd_solver.cpp:105] Iteration 2064, lr = 0.00664414 I0409 23:20:55.379323 6396 solver.cpp:218] Iteration 2076 (2.38143 iter/s, 5.03898s/12 iters), loss = 4.13792 I0409 23:20:55.379374 6396 solver.cpp:237] Train net output #0: loss = 4.13792 (* 1 = 4.13792 loss) I0409 23:20:55.379382 6396 sgd_solver.cpp:105] Iteration 2076, lr = 0.00662837 I0409 23:21:00.320042 6396 solver.cpp:218] Iteration 2088 (2.4289 iter/s, 4.94051s/12 iters), loss = 3.94799 I0409 23:21:00.321516 6396 solver.cpp:237] Train net output #0: loss = 3.94799 (* 1 = 3.94799 loss) I0409 23:21:00.321532 6396 sgd_solver.cpp:105] Iteration 2088, lr = 0.00661263 I0409 23:21:05.237515 6396 solver.cpp:218] Iteration 2100 (2.44109 iter/s, 4.91585s/12 iters), loss = 4.05609 I0409 23:21:05.237582 6396 solver.cpp:237] Train net output #0: loss = 4.05609 (* 1 = 4.05609 loss) I0409 23:21:05.237597 6396 sgd_solver.cpp:105] Iteration 2100, lr = 0.00659693 I0409 23:21:10.191763 6396 solver.cpp:218] Iteration 2112 (2.42227 iter/s, 4.95402s/12 iters), loss = 4.06933 I0409 23:21:10.191807 6396 solver.cpp:237] Train net output #0: loss = 4.06933 (* 1 = 4.06933 loss) I0409 23:21:10.191817 6396 sgd_solver.cpp:105] Iteration 2112, lr = 0.00658127 I0409 23:21:14.953332 6400 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:21:15.274793 6396 solver.cpp:218] Iteration 2124 (2.36089 iter/s, 5.08282s/12 iters), loss = 3.64441 I0409 23:21:15.274849 6396 solver.cpp:237] Train net output #0: loss = 3.64441 (* 1 = 3.64441 loss) I0409 23:21:15.274859 6396 sgd_solver.cpp:105] Iteration 2124, lr = 0.00656564 I0409 23:21:20.281240 6396 solver.cpp:218] Iteration 2136 (2.39701 iter/s, 5.00623s/12 iters), loss = 4.07856 I0409 23:21:20.281286 6396 solver.cpp:237] Train net output #0: loss = 4.07856 (* 1 = 4.07856 loss) I0409 23:21:20.281294 6396 sgd_solver.cpp:105] Iteration 2136, lr = 0.00655006 I0409 23:21:22.336412 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_2142.caffemodel I0409 23:21:23.170626 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_2142.solverstate I0409 23:21:23.494549 6396 solver.cpp:330] Iteration 2142, Testing net (#0) I0409 23:21:23.494580 6396 net.cpp:676] Ignoring source layer train-data I0409 23:21:27.345259 6401 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:21:28.216073 6396 solver.cpp:397] Test net output #0: accuracy = 0.0980392 I0409 23:21:28.216112 6396 solver.cpp:397] Test net output #1: loss = 3.90421 (* 1 = 3.90421 loss) I0409 23:21:30.015677 6396 solver.cpp:218] Iteration 2148 (1.23278 iter/s, 9.73409s/12 iters), loss = 3.90262 I0409 23:21:30.015723 6396 solver.cpp:237] Train net output #0: loss = 3.90262 (* 1 = 3.90262 loss) I0409 23:21:30.015733 6396 sgd_solver.cpp:105] Iteration 2148, lr = 0.00653451 I0409 23:21:35.018576 6396 solver.cpp:218] Iteration 2160 (2.39871 iter/s, 5.00269s/12 iters), loss = 4.22384 I0409 23:21:35.018719 6396 solver.cpp:237] Train net output #0: loss = 4.22384 (* 1 = 4.22384 loss) I0409 23:21:35.018729 6396 sgd_solver.cpp:105] Iteration 2160, lr = 0.00651899 I0409 23:21:40.031054 6396 solver.cpp:218] Iteration 2172 (2.39417 iter/s, 5.01217s/12 iters), loss = 3.83919 I0409 23:21:40.031107 6396 solver.cpp:237] Train net output #0: loss = 3.83919 (* 1 = 3.83919 loss) I0409 23:21:40.031117 6396 sgd_solver.cpp:105] Iteration 2172, lr = 0.00650351 I0409 23:21:45.012954 6396 solver.cpp:218] Iteration 2184 (2.40882 iter/s, 4.98169s/12 iters), loss = 3.73305 I0409 23:21:45.013000 6396 solver.cpp:237] Train net output #0: loss = 3.73305 (* 1 = 3.73305 loss) I0409 23:21:45.013010 6396 sgd_solver.cpp:105] Iteration 2184, lr = 0.00648807 I0409 23:21:50.341827 6396 solver.cpp:218] Iteration 2196 (2.25198 iter/s, 5.32865s/12 iters), loss = 3.83489 I0409 23:21:50.341883 6396 solver.cpp:237] Train net output #0: loss = 3.83489 (* 1 = 3.83489 loss) I0409 23:21:50.341895 6396 sgd_solver.cpp:105] Iteration 2196, lr = 0.00647267 I0409 23:21:55.305423 6396 solver.cpp:218] Iteration 2208 (2.41771 iter/s, 4.96338s/12 iters), loss = 3.84688 I0409 23:21:55.305477 6396 solver.cpp:237] Train net output #0: loss = 3.84688 (* 1 = 3.84688 loss) I0409 23:21:55.305488 6396 sgd_solver.cpp:105] Iteration 2208, lr = 0.0064573 I0409 23:22:00.267741 6396 solver.cpp:218] Iteration 2220 (2.41833 iter/s, 4.96211s/12 iters), loss = 3.8275 I0409 23:22:00.267781 6396 solver.cpp:237] Train net output #0: loss = 3.8275 (* 1 = 3.8275 loss) I0409 23:22:00.267788 6396 sgd_solver.cpp:105] Iteration 2220, lr = 0.00644197 I0409 23:22:02.064924 6400 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:22:05.241541 6396 solver.cpp:218] Iteration 2232 (2.41274 iter/s, 4.9736s/12 iters), loss = 3.90747 I0409 23:22:05.241639 6396 solver.cpp:237] Train net output #0: loss = 3.90747 (* 1 = 3.90747 loss) I0409 23:22:05.241648 6396 sgd_solver.cpp:105] Iteration 2232, lr = 0.00642668 I0409 23:22:09.818936 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_2244.caffemodel I0409 23:22:10.283102 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_2244.solverstate I0409 23:22:10.577862 6396 solver.cpp:330] Iteration 2244, Testing net (#0) I0409 23:22:10.577880 6396 net.cpp:676] Ignoring source layer train-data I0409 23:22:14.330479 6401 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:22:15.258924 6396 solver.cpp:397] Test net output #0: accuracy = 0.115809 I0409 23:22:15.258952 6396 solver.cpp:397] Test net output #1: loss = 3.85089 (* 1 = 3.85089 loss) I0409 23:22:15.343456 6396 solver.cpp:218] Iteration 2244 (1.18794 iter/s, 10.1015s/12 iters), loss = 4.07175 I0409 23:22:15.343499 6396 solver.cpp:237] Train net output #0: loss = 4.07175 (* 1 = 4.07175 loss) I0409 23:22:15.343508 6396 sgd_solver.cpp:105] Iteration 2244, lr = 0.00641142 I0409 23:22:19.926609 6396 solver.cpp:218] Iteration 2256 (2.6184 iter/s, 4.58296s/12 iters), loss = 3.71838 I0409 23:22:19.926649 6396 solver.cpp:237] Train net output #0: loss = 3.71838 (* 1 = 3.71838 loss) I0409 23:22:19.926657 6396 sgd_solver.cpp:105] Iteration 2256, lr = 0.0063962 I0409 23:22:24.922219 6396 solver.cpp:218] Iteration 2268 (2.4022 iter/s, 4.99541s/12 iters), loss = 3.7817 I0409 23:22:24.922259 6396 solver.cpp:237] Train net output #0: loss = 3.7817 (* 1 = 3.7817 loss) I0409 23:22:24.922268 6396 sgd_solver.cpp:105] Iteration 2268, lr = 0.00638101 I0409 23:22:29.810324 6396 solver.cpp:218] Iteration 2280 (2.45504 iter/s, 4.8879s/12 iters), loss = 3.65404 I0409 23:22:29.810372 6396 solver.cpp:237] Train net output #0: loss = 3.65404 (* 1 = 3.65404 loss) I0409 23:22:29.810380 6396 sgd_solver.cpp:105] Iteration 2280, lr = 0.00636586 I0409 23:22:34.993151 6396 solver.cpp:218] Iteration 2292 (2.31544 iter/s, 5.18261s/12 iters), loss = 3.71967 I0409 23:22:34.993206 6396 solver.cpp:237] Train net output #0: loss = 3.71967 (* 1 = 3.71967 loss) I0409 23:22:34.993217 6396 sgd_solver.cpp:105] Iteration 2292, lr = 0.00635075 I0409 23:22:40.059214 6396 solver.cpp:218] Iteration 2304 (2.3688 iter/s, 5.06585s/12 iters), loss = 3.95429 I0409 23:22:40.059348 6396 solver.cpp:237] Train net output #0: loss = 3.95429 (* 1 = 3.95429 loss) I0409 23:22:40.059358 6396 sgd_solver.cpp:105] Iteration 2304, lr = 0.00633567 I0409 23:22:44.995191 6396 solver.cpp:218] Iteration 2316 (2.43127 iter/s, 4.93568s/12 iters), loss = 3.85445 I0409 23:22:44.995236 6396 solver.cpp:237] Train net output #0: loss = 3.85445 (* 1 = 3.85445 loss) I0409 23:22:44.995245 6396 sgd_solver.cpp:105] Iteration 2316, lr = 0.00632063 I0409 23:22:48.855512 6400 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:22:49.899006 6396 solver.cpp:218] Iteration 2328 (2.44718 iter/s, 4.90361s/12 iters), loss = 3.53099 I0409 23:22:49.899051 6396 solver.cpp:237] Train net output #0: loss = 3.53099 (* 1 = 3.53099 loss) I0409 23:22:49.899060 6396 sgd_solver.cpp:105] Iteration 2328, lr = 0.00630562 I0409 23:22:54.804775 6396 solver.cpp:218] Iteration 2340 (2.4462 iter/s, 4.90557s/12 iters), loss = 3.75378 I0409 23:22:54.804814 6396 solver.cpp:237] Train net output #0: loss = 3.75378 (* 1 = 3.75378 loss) I0409 23:22:54.804822 6396 sgd_solver.cpp:105] Iteration 2340, lr = 0.00629065 I0409 23:22:56.784858 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_2346.caffemodel I0409 23:22:57.234802 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_2346.solverstate I0409 23:22:57.541805 6396 solver.cpp:330] Iteration 2346, Testing net (#0) I0409 23:22:57.541828 6396 net.cpp:676] Ignoring source layer train-data I0409 23:23:00.981339 6401 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:23:01.939874 6396 solver.cpp:397] Test net output #0: accuracy = 0.139093 I0409 23:23:01.939937 6396 solver.cpp:397] Test net output #1: loss = 3.68796 (* 1 = 3.68796 loss) I0409 23:23:04.070427 6396 solver.cpp:218] Iteration 2352 (1.29515 iter/s, 9.26533s/12 iters), loss = 3.63336 I0409 23:23:04.070474 6396 solver.cpp:237] Train net output #0: loss = 3.63336 (* 1 = 3.63336 loss) I0409 23:23:04.070483 6396 sgd_solver.cpp:105] Iteration 2352, lr = 0.00627571 I0409 23:23:09.050709 6396 solver.cpp:218] Iteration 2364 (2.40961 iter/s, 4.98007s/12 iters), loss = 3.70532 I0409 23:23:09.050774 6396 solver.cpp:237] Train net output #0: loss = 3.70532 (* 1 = 3.70532 loss) I0409 23:23:09.050787 6396 sgd_solver.cpp:105] Iteration 2364, lr = 0.00626081 I0409 23:23:14.010890 6396 solver.cpp:218] Iteration 2376 (2.41937 iter/s, 4.95996s/12 iters), loss = 3.5374 I0409 23:23:14.010998 6396 solver.cpp:237] Train net output #0: loss = 3.5374 (* 1 = 3.5374 loss) I0409 23:23:14.011008 6396 sgd_solver.cpp:105] Iteration 2376, lr = 0.00624595 I0409 23:23:19.013919 6396 solver.cpp:218] Iteration 2388 (2.39868 iter/s, 5.00276s/12 iters), loss = 3.70028 I0409 23:23:19.014000 6396 solver.cpp:237] Train net output #0: loss = 3.70028 (* 1 = 3.70028 loss) I0409 23:23:19.014012 6396 sgd_solver.cpp:105] Iteration 2388, lr = 0.00623112 I0409 23:23:23.914187 6396 solver.cpp:218] Iteration 2400 (2.44896 iter/s, 4.90003s/12 iters), loss = 3.683 I0409 23:23:23.914233 6396 solver.cpp:237] Train net output #0: loss = 3.683 (* 1 = 3.683 loss) I0409 23:23:23.914242 6396 sgd_solver.cpp:105] Iteration 2400, lr = 0.00621633 I0409 23:23:28.912034 6396 solver.cpp:218] Iteration 2412 (2.40114 iter/s, 4.99763s/12 iters), loss = 3.45108 I0409 23:23:28.912094 6396 solver.cpp:237] Train net output #0: loss = 3.45108 (* 1 = 3.45108 loss) I0409 23:23:28.912106 6396 sgd_solver.cpp:105] Iteration 2412, lr = 0.00620157 I0409 23:23:34.319797 6396 solver.cpp:218] Iteration 2424 (2.21913 iter/s, 5.40753s/12 iters), loss = 3.68421 I0409 23:23:34.319856 6396 solver.cpp:237] Train net output #0: loss = 3.68421 (* 1 = 3.68421 loss) I0409 23:23:34.319870 6396 sgd_solver.cpp:105] Iteration 2424, lr = 0.00618684 I0409 23:23:35.409070 6400 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:23:39.307972 6396 solver.cpp:218] Iteration 2436 (2.4058 iter/s, 4.98795s/12 iters), loss = 3.7046 I0409 23:23:39.308029 6396 solver.cpp:237] Train net output #0: loss = 3.7046 (* 1 = 3.7046 loss) I0409 23:23:39.308041 6396 sgd_solver.cpp:105] Iteration 2436, lr = 0.00617215 I0409 23:23:43.936532 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_2448.caffemodel I0409 23:23:44.969053 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_2448.solverstate I0409 23:23:45.883288 6396 solver.cpp:330] Iteration 2448, Testing net (#0) I0409 23:23:45.883319 6396 net.cpp:676] Ignoring source layer train-data I0409 23:23:49.501157 6401 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:23:50.917886 6396 solver.cpp:397] Test net output #0: accuracy = 0.164828 I0409 23:23:50.917933 6396 solver.cpp:397] Test net output #1: loss = 3.63631 (* 1 = 3.63631 loss) I0409 23:23:51.001318 6396 solver.cpp:218] Iteration 2448 (1.02626 iter/s, 11.6929s/12 iters), loss = 3.77965 I0409 23:23:51.001371 6396 solver.cpp:237] Train net output #0: loss = 3.77965 (* 1 = 3.77965 loss) I0409 23:23:51.001382 6396 sgd_solver.cpp:105] Iteration 2448, lr = 0.0061575 I0409 23:23:55.114737 6396 solver.cpp:218] Iteration 2460 (2.91741 iter/s, 4.11323s/12 iters), loss = 3.51167 I0409 23:23:55.114784 6396 solver.cpp:237] Train net output #0: loss = 3.51167 (* 1 = 3.51167 loss) I0409 23:23:55.114792 6396 sgd_solver.cpp:105] Iteration 2460, lr = 0.00614288 I0409 23:24:00.724881 6396 solver.cpp:218] Iteration 2472 (2.13907 iter/s, 5.60992s/12 iters), loss = 3.67142 I0409 23:24:00.724949 6396 solver.cpp:237] Train net output #0: loss = 3.67142 (* 1 = 3.67142 loss) I0409 23:24:00.724964 6396 sgd_solver.cpp:105] Iteration 2472, lr = 0.0061283 I0409 23:24:05.690819 6396 solver.cpp:218] Iteration 2484 (2.41657 iter/s, 4.96571s/12 iters), loss = 3.55337 I0409 23:24:05.690874 6396 solver.cpp:237] Train net output #0: loss = 3.55337 (* 1 = 3.55337 loss) I0409 23:24:05.690886 6396 sgd_solver.cpp:105] Iteration 2484, lr = 0.00611375 I0409 23:24:10.606384 6396 solver.cpp:218] Iteration 2496 (2.44133 iter/s, 4.91535s/12 iters), loss = 3.60504 I0409 23:24:10.606438 6396 solver.cpp:237] Train net output #0: loss = 3.60504 (* 1 = 3.60504 loss) I0409 23:24:10.606449 6396 sgd_solver.cpp:105] Iteration 2496, lr = 0.00609923 I0409 23:24:15.459794 6396 solver.cpp:218] Iteration 2508 (2.47259 iter/s, 4.85321s/12 iters), loss = 3.64453 I0409 23:24:15.459894 6396 solver.cpp:237] Train net output #0: loss = 3.64453 (* 1 = 3.64453 loss) I0409 23:24:15.459903 6396 sgd_solver.cpp:105] Iteration 2508, lr = 0.00608475 I0409 23:24:20.497493 6396 solver.cpp:218] Iteration 2520 (2.38216 iter/s, 5.03744s/12 iters), loss = 3.60964 I0409 23:24:20.497545 6396 solver.cpp:237] Train net output #0: loss = 3.60964 (* 1 = 3.60964 loss) I0409 23:24:20.497555 6396 sgd_solver.cpp:105] Iteration 2520, lr = 0.0060703 I0409 23:24:23.690412 6400 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:24:25.470041 6396 solver.cpp:218] Iteration 2532 (2.41335 iter/s, 4.97234s/12 iters), loss = 3.60237 I0409 23:24:25.470088 6396 solver.cpp:237] Train net output #0: loss = 3.60237 (* 1 = 3.60237 loss) I0409 23:24:25.470095 6396 sgd_solver.cpp:105] Iteration 2532, lr = 0.00605589 I0409 23:24:30.504165 6396 solver.cpp:218] Iteration 2544 (2.38383 iter/s, 5.03391s/12 iters), loss = 3.45615 I0409 23:24:30.504233 6396 solver.cpp:237] Train net output #0: loss = 3.45615 (* 1 = 3.45615 loss) I0409 23:24:30.504246 6396 sgd_solver.cpp:105] Iteration 2544, lr = 0.00604151 I0409 23:24:32.519446 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_2550.caffemodel I0409 23:24:32.974117 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_2550.solverstate I0409 23:24:33.293267 6396 solver.cpp:330] Iteration 2550, Testing net (#0) I0409 23:24:33.293287 6396 net.cpp:676] Ignoring source layer train-data I0409 23:24:36.685676 6401 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:24:37.707994 6396 solver.cpp:397] Test net output #0: accuracy = 0.170956 I0409 23:24:37.708034 6396 solver.cpp:397] Test net output #1: loss = 3.48558 (* 1 = 3.48558 loss) I0409 23:24:40.051139 6396 solver.cpp:218] Iteration 2556 (1.25699 iter/s, 9.54662s/12 iters), loss = 3.52544 I0409 23:24:40.051182 6396 solver.cpp:237] Train net output #0: loss = 3.52544 (* 1 = 3.52544 loss) I0409 23:24:40.051192 6396 sgd_solver.cpp:105] Iteration 2556, lr = 0.00602717 I0409 23:24:45.365500 6396 solver.cpp:218] Iteration 2568 (2.25812 iter/s, 5.31415s/12 iters), loss = 3.27331 I0409 23:24:45.365562 6396 solver.cpp:237] Train net output #0: loss = 3.27331 (* 1 = 3.27331 loss) I0409 23:24:45.365574 6396 sgd_solver.cpp:105] Iteration 2568, lr = 0.00601286 I0409 23:24:50.459228 6396 solver.cpp:218] Iteration 2580 (2.35594 iter/s, 5.0935s/12 iters), loss = 3.36237 I0409 23:24:50.460436 6396 solver.cpp:237] Train net output #0: loss = 3.36237 (* 1 = 3.36237 loss) I0409 23:24:50.460448 6396 sgd_solver.cpp:105] Iteration 2580, lr = 0.00599858 I0409 23:24:55.455416 6396 solver.cpp:218] Iteration 2592 (2.40249 iter/s, 4.99482s/12 iters), loss = 3.75403 I0409 23:24:55.455462 6396 solver.cpp:237] Train net output #0: loss = 3.75403 (* 1 = 3.75403 loss) I0409 23:24:55.455471 6396 sgd_solver.cpp:105] Iteration 2592, lr = 0.00598434 I0409 23:25:00.575925 6396 solver.cpp:218] Iteration 2604 (2.34361 iter/s, 5.1203s/12 iters), loss = 3.45138 I0409 23:25:00.575981 6396 solver.cpp:237] Train net output #0: loss = 3.45138 (* 1 = 3.45138 loss) I0409 23:25:00.575994 6396 sgd_solver.cpp:105] Iteration 2604, lr = 0.00597013 I0409 23:25:05.525979 6396 solver.cpp:218] Iteration 2616 (2.42433 iter/s, 4.94982s/12 iters), loss = 3.58074 I0409 23:25:05.526033 6396 solver.cpp:237] Train net output #0: loss = 3.58074 (* 1 = 3.58074 loss) I0409 23:25:05.526043 6396 sgd_solver.cpp:105] Iteration 2616, lr = 0.00595596 I0409 23:25:10.392393 6396 solver.cpp:218] Iteration 2628 (2.46599 iter/s, 4.8662s/12 iters), loss = 3.30902 I0409 23:25:10.392441 6396 solver.cpp:237] Train net output #0: loss = 3.30902 (* 1 = 3.30902 loss) I0409 23:25:10.392450 6396 sgd_solver.cpp:105] Iteration 2628, lr = 0.00594182 I0409 23:25:10.841361 6400 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:25:15.330222 6396 solver.cpp:218] Iteration 2640 (2.43032 iter/s, 4.93762s/12 iters), loss = 3.58375 I0409 23:25:15.330271 6396 solver.cpp:237] Train net output #0: loss = 3.58375 (* 1 = 3.58375 loss) I0409 23:25:15.330278 6396 sgd_solver.cpp:105] Iteration 2640, lr = 0.00592771 I0409 23:25:20.089196 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_2652.caffemodel I0409 23:25:20.558179 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_2652.solverstate I0409 23:25:20.873579 6396 solver.cpp:330] Iteration 2652, Testing net (#0) I0409 23:25:20.873598 6396 net.cpp:676] Ignoring source layer train-data I0409 23:25:24.368548 6401 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:25:25.589336 6396 solver.cpp:397] Test net output #0: accuracy = 0.191789 I0409 23:25:25.589372 6396 solver.cpp:397] Test net output #1: loss = 3.37555 (* 1 = 3.37555 loss) I0409 23:25:25.676592 6396 solver.cpp:218] Iteration 2652 (1.15987 iter/s, 10.346s/12 iters), loss = 3.35528 I0409 23:25:25.676632 6396 solver.cpp:237] Train net output #0: loss = 3.35528 (* 1 = 3.35528 loss) I0409 23:25:25.676640 6396 sgd_solver.cpp:105] Iteration 2652, lr = 0.00591364 I0409 23:25:29.802512 6396 solver.cpp:218] Iteration 2664 (2.90857 iter/s, 4.12574s/12 iters), loss = 3.27167 I0409 23:25:29.802567 6396 solver.cpp:237] Train net output #0: loss = 3.27167 (* 1 = 3.27167 loss) I0409 23:25:29.802577 6396 sgd_solver.cpp:105] Iteration 2664, lr = 0.0058996 I0409 23:25:34.687947 6396 solver.cpp:218] Iteration 2676 (2.45639 iter/s, 4.88523s/12 iters), loss = 3.38674 I0409 23:25:34.687994 6396 solver.cpp:237] Train net output #0: loss = 3.38674 (* 1 = 3.38674 loss) I0409 23:25:34.688004 6396 sgd_solver.cpp:105] Iteration 2676, lr = 0.00588559 I0409 23:25:39.699800 6396 solver.cpp:218] Iteration 2688 (2.39442 iter/s, 5.01165s/12 iters), loss = 3.32528 I0409 23:25:39.699852 6396 solver.cpp:237] Train net output #0: loss = 3.32528 (* 1 = 3.32528 loss) I0409 23:25:39.699863 6396 sgd_solver.cpp:105] Iteration 2688, lr = 0.00587162 I0409 23:25:44.878245 6396 solver.cpp:218] Iteration 2700 (2.31739 iter/s, 5.17823s/12 iters), loss = 3.43758 I0409 23:25:44.878288 6396 solver.cpp:237] Train net output #0: loss = 3.43758 (* 1 = 3.43758 loss) I0409 23:25:44.878296 6396 sgd_solver.cpp:105] Iteration 2700, lr = 0.00585768 I0409 23:25:49.865252 6396 solver.cpp:218] Iteration 2712 (2.40635 iter/s, 4.98681s/12 iters), loss = 3.23327 I0409 23:25:49.865291 6396 solver.cpp:237] Train net output #0: loss = 3.23327 (* 1 = 3.23327 loss) I0409 23:25:49.865303 6396 sgd_solver.cpp:105] Iteration 2712, lr = 0.00584377 I0409 23:25:54.819701 6396 solver.cpp:218] Iteration 2724 (2.42216 iter/s, 4.95425s/12 iters), loss = 3.50207 I0409 23:25:54.819859 6396 solver.cpp:237] Train net output #0: loss = 3.50207 (* 1 = 3.50207 loss) I0409 23:25:54.819869 6396 sgd_solver.cpp:105] Iteration 2724, lr = 0.0058299 I0409 23:25:57.325637 6400 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:25:59.691696 6396 solver.cpp:218] Iteration 2736 (2.46321 iter/s, 4.87169s/12 iters), loss = 2.99152 I0409 23:25:59.691751 6396 solver.cpp:237] Train net output #0: loss = 2.99152 (* 1 = 2.99152 loss) I0409 23:25:59.691763 6396 sgd_solver.cpp:105] Iteration 2736, lr = 0.00581605 I0409 23:26:04.618518 6396 solver.cpp:218] Iteration 2748 (2.43575 iter/s, 4.92661s/12 iters), loss = 3.4643 I0409 23:26:04.618579 6396 solver.cpp:237] Train net output #0: loss = 3.4643 (* 1 = 3.4643 loss) I0409 23:26:04.618592 6396 sgd_solver.cpp:105] Iteration 2748, lr = 0.00580225 I0409 23:26:06.601256 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_2754.caffemodel I0409 23:26:07.361196 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_2754.solverstate I0409 23:26:08.325081 6396 solver.cpp:330] Iteration 2754, Testing net (#0) I0409 23:26:08.325103 6396 net.cpp:676] Ignoring source layer train-data I0409 23:26:10.986815 6396 blocking_queue.cpp:49] Waiting for data I0409 23:26:11.584913 6401 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:26:12.686039 6396 solver.cpp:397] Test net output #0: accuracy = 0.193627 I0409 23:26:12.686069 6396 solver.cpp:397] Test net output #1: loss = 3.33955 (* 1 = 3.33955 loss) I0409 23:26:14.559674 6396 solver.cpp:218] Iteration 2760 (1.20715 iter/s, 9.9408s/12 iters), loss = 3.24949 I0409 23:26:14.559722 6396 solver.cpp:237] Train net output #0: loss = 3.24949 (* 1 = 3.24949 loss) I0409 23:26:14.559731 6396 sgd_solver.cpp:105] Iteration 2760, lr = 0.00578847 I0409 23:26:19.541671 6396 solver.cpp:218] Iteration 2772 (2.40878 iter/s, 4.98179s/12 iters), loss = 3.2756 I0409 23:26:19.541733 6396 solver.cpp:237] Train net output #0: loss = 3.2756 (* 1 = 3.2756 loss) I0409 23:26:19.541745 6396 sgd_solver.cpp:105] Iteration 2772, lr = 0.00577473 I0409 23:26:24.447470 6396 solver.cpp:218] Iteration 2784 (2.44619 iter/s, 4.90558s/12 iters), loss = 3.45906 I0409 23:26:24.447535 6396 solver.cpp:237] Train net output #0: loss = 3.45906 (* 1 = 3.45906 loss) I0409 23:26:24.447547 6396 sgd_solver.cpp:105] Iteration 2784, lr = 0.00576102 I0409 23:26:29.563442 6396 solver.cpp:218] Iteration 2796 (2.3457 iter/s, 5.11575s/12 iters), loss = 3.27261 I0409 23:26:29.563556 6396 solver.cpp:237] Train net output #0: loss = 3.27261 (* 1 = 3.27261 loss) I0409 23:26:29.563565 6396 sgd_solver.cpp:105] Iteration 2796, lr = 0.00574734 I0409 23:26:34.715662 6396 solver.cpp:218] Iteration 2808 (2.32922 iter/s, 5.15194s/12 iters), loss = 3.3906 I0409 23:26:34.715711 6396 solver.cpp:237] Train net output #0: loss = 3.3906 (* 1 = 3.3906 loss) I0409 23:26:34.715721 6396 sgd_solver.cpp:105] Iteration 2808, lr = 0.00573369 I0409 23:26:39.686753 6396 solver.cpp:218] Iteration 2820 (2.41406 iter/s, 4.97088s/12 iters), loss = 3.20911 I0409 23:26:39.686810 6396 solver.cpp:237] Train net output #0: loss = 3.20911 (* 1 = 3.20911 loss) I0409 23:26:39.686821 6396 sgd_solver.cpp:105] Iteration 2820, lr = 0.00572008 I0409 23:26:44.497700 6400 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:26:44.786018 6396 solver.cpp:218] Iteration 2832 (2.35338 iter/s, 5.09905s/12 iters), loss = 3.21195 I0409 23:26:44.786074 6396 solver.cpp:237] Train net output #0: loss = 3.21195 (* 1 = 3.21195 loss) I0409 23:26:44.786087 6396 sgd_solver.cpp:105] Iteration 2832, lr = 0.0057065 I0409 23:26:49.698231 6396 solver.cpp:218] Iteration 2844 (2.443 iter/s, 4.912s/12 iters), loss = 3.2163 I0409 23:26:49.698277 6396 solver.cpp:237] Train net output #0: loss = 3.2163 (* 1 = 3.2163 loss) I0409 23:26:49.698285 6396 sgd_solver.cpp:105] Iteration 2844, lr = 0.00569295 I0409 23:26:54.140373 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_2856.caffemodel I0409 23:26:54.586123 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_2856.solverstate I0409 23:26:54.898814 6396 solver.cpp:330] Iteration 2856, Testing net (#0) I0409 23:26:54.898842 6396 net.cpp:676] Ignoring source layer train-data I0409 23:26:58.148963 6401 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:26:59.555423 6396 solver.cpp:397] Test net output #0: accuracy = 0.216912 I0409 23:26:59.555467 6396 solver.cpp:397] Test net output #1: loss = 3.22982 (* 1 = 3.22982 loss) I0409 23:26:59.638754 6396 solver.cpp:218] Iteration 2856 (1.20722 iter/s, 9.94018s/12 iters), loss = 2.8985 I0409 23:26:59.638860 6396 solver.cpp:237] Train net output #0: loss = 2.8985 (* 1 = 2.8985 loss) I0409 23:26:59.638871 6396 sgd_solver.cpp:105] Iteration 2856, lr = 0.00567944 I0409 23:27:03.787151 6396 solver.cpp:218] Iteration 2868 (2.89286 iter/s, 4.14815s/12 iters), loss = 3.20744 I0409 23:27:03.787211 6396 solver.cpp:237] Train net output #0: loss = 3.20744 (* 1 = 3.20744 loss) I0409 23:27:03.787225 6396 sgd_solver.cpp:105] Iteration 2868, lr = 0.00566595 I0409 23:27:08.683121 6396 solver.cpp:218] Iteration 2880 (2.4511 iter/s, 4.89575s/12 iters), loss = 3.0272 I0409 23:27:08.683182 6396 solver.cpp:237] Train net output #0: loss = 3.0272 (* 1 = 3.0272 loss) I0409 23:27:08.683194 6396 sgd_solver.cpp:105] Iteration 2880, lr = 0.0056525 I0409 23:27:13.774137 6396 solver.cpp:218] Iteration 2892 (2.3572 iter/s, 5.0908s/12 iters), loss = 3.21877 I0409 23:27:13.774181 6396 solver.cpp:237] Train net output #0: loss = 3.21877 (* 1 = 3.21877 loss) I0409 23:27:13.774190 6396 sgd_solver.cpp:105] Iteration 2892, lr = 0.00563908 I0409 23:27:18.803162 6396 solver.cpp:218] Iteration 2904 (2.38625 iter/s, 5.02882s/12 iters), loss = 3.06068 I0409 23:27:18.803215 6396 solver.cpp:237] Train net output #0: loss = 3.06068 (* 1 = 3.06068 loss) I0409 23:27:18.803226 6396 sgd_solver.cpp:105] Iteration 2904, lr = 0.00562569 I0409 23:27:23.741020 6396 solver.cpp:218] Iteration 2916 (2.43031 iter/s, 4.93765s/12 iters), loss = 3.12346 I0409 23:27:23.741067 6396 solver.cpp:237] Train net output #0: loss = 3.12346 (* 1 = 3.12346 loss) I0409 23:27:23.741076 6396 sgd_solver.cpp:105] Iteration 2916, lr = 0.00561233 I0409 23:27:28.874442 6396 solver.cpp:218] Iteration 2928 (2.33772 iter/s, 5.13321s/12 iters), loss = 3.1726 I0409 23:27:28.874487 6396 solver.cpp:237] Train net output #0: loss = 3.1726 (* 1 = 3.1726 loss) I0409 23:27:28.874495 6396 sgd_solver.cpp:105] Iteration 2928, lr = 0.00559901 I0409 23:27:30.658066 6400 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:27:33.745502 6396 solver.cpp:218] Iteration 2940 (2.46363 iter/s, 4.87085s/12 iters), loss = 3.0785 I0409 23:27:33.745561 6396 solver.cpp:237] Train net output #0: loss = 3.0785 (* 1 = 3.0785 loss) I0409 23:27:33.745573 6396 sgd_solver.cpp:105] Iteration 2940, lr = 0.00558572 I0409 23:27:38.689675 6396 solver.cpp:218] Iteration 2952 (2.42721 iter/s, 4.94395s/12 iters), loss = 3.00427 I0409 23:27:38.689726 6396 solver.cpp:237] Train net output #0: loss = 3.00427 (* 1 = 3.00427 loss) I0409 23:27:38.689736 6396 sgd_solver.cpp:105] Iteration 2952, lr = 0.00557245 I0409 23:27:40.731464 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_2958.caffemodel I0409 23:27:41.184720 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_2958.solverstate I0409 23:27:41.489234 6396 solver.cpp:330] Iteration 2958, Testing net (#0) I0409 23:27:41.489257 6396 net.cpp:676] Ignoring source layer train-data I0409 23:27:44.702224 6401 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:27:46.003178 6396 solver.cpp:397] Test net output #0: accuracy = 0.222426 I0409 23:27:46.003221 6396 solver.cpp:397] Test net output #1: loss = 3.16018 (* 1 = 3.16018 loss) I0409 23:27:47.746841 6396 solver.cpp:218] Iteration 2964 (1.32497 iter/s, 9.05684s/12 iters), loss = 3.01048 I0409 23:27:47.746891 6396 solver.cpp:237] Train net output #0: loss = 3.01048 (* 1 = 3.01048 loss) I0409 23:27:47.746899 6396 sgd_solver.cpp:105] Iteration 2964, lr = 0.00555922 I0409 23:27:52.697995 6396 solver.cpp:218] Iteration 2976 (2.42378 iter/s, 4.95095s/12 iters), loss = 2.93954 I0409 23:27:52.698037 6396 solver.cpp:237] Train net output #0: loss = 2.93954 (* 1 = 2.93954 loss) I0409 23:27:52.698045 6396 sgd_solver.cpp:105] Iteration 2976, lr = 0.00554603 I0409 23:27:57.701207 6396 solver.cpp:218] Iteration 2988 (2.39856 iter/s, 5.00301s/12 iters), loss = 3.25113 I0409 23:27:57.701265 6396 solver.cpp:237] Train net output #0: loss = 3.25113 (* 1 = 3.25113 loss) I0409 23:27:57.701277 6396 sgd_solver.cpp:105] Iteration 2988, lr = 0.00553286 I0409 23:28:02.607034 6396 solver.cpp:218] Iteration 3000 (2.44618 iter/s, 4.90561s/12 iters), loss = 3.2066 I0409 23:28:02.607129 6396 solver.cpp:237] Train net output #0: loss = 3.2066 (* 1 = 3.2066 loss) I0409 23:28:02.607142 6396 sgd_solver.cpp:105] Iteration 3000, lr = 0.00551972 I0409 23:28:07.559376 6396 solver.cpp:218] Iteration 3012 (2.42322 iter/s, 4.95209s/12 iters), loss = 3.2979 I0409 23:28:07.559423 6396 solver.cpp:237] Train net output #0: loss = 3.2979 (* 1 = 3.2979 loss) I0409 23:28:07.559432 6396 sgd_solver.cpp:105] Iteration 3012, lr = 0.00550662 I0409 23:28:12.454695 6396 solver.cpp:218] Iteration 3024 (2.45142 iter/s, 4.89511s/12 iters), loss = 2.95977 I0409 23:28:12.454743 6396 solver.cpp:237] Train net output #0: loss = 2.95977 (* 1 = 2.95977 loss) I0409 23:28:12.454753 6396 sgd_solver.cpp:105] Iteration 3024, lr = 0.00549354 I0409 23:28:16.456532 6400 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:28:17.469342 6396 solver.cpp:218] Iteration 3036 (2.39309 iter/s, 5.01444s/12 iters), loss = 2.6581 I0409 23:28:17.469405 6396 solver.cpp:237] Train net output #0: loss = 2.6581 (* 1 = 2.6581 loss) I0409 23:28:17.469417 6396 sgd_solver.cpp:105] Iteration 3036, lr = 0.0054805 I0409 23:28:22.386585 6396 solver.cpp:218] Iteration 3048 (2.4405 iter/s, 4.91702s/12 iters), loss = 2.90884 I0409 23:28:22.386637 6396 solver.cpp:237] Train net output #0: loss = 2.90884 (* 1 = 2.90884 loss) I0409 23:28:22.386648 6396 sgd_solver.cpp:105] Iteration 3048, lr = 0.00546749 I0409 23:28:26.904120 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_3060.caffemodel I0409 23:28:27.356971 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_3060.solverstate I0409 23:28:27.654367 6396 solver.cpp:330] Iteration 3060, Testing net (#0) I0409 23:28:27.654386 6396 net.cpp:676] Ignoring source layer train-data I0409 23:28:30.774912 6401 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:28:32.154649 6396 solver.cpp:397] Test net output #0: accuracy = 0.260417 I0409 23:28:32.154690 6396 solver.cpp:397] Test net output #1: loss = 3.02677 (* 1 = 3.02677 loss) I0409 23:28:32.237820 6396 solver.cpp:218] Iteration 3060 (1.21816 iter/s, 9.85089s/12 iters), loss = 3.02081 I0409 23:28:32.237871 6396 solver.cpp:237] Train net output #0: loss = 3.02081 (* 1 = 3.02081 loss) I0409 23:28:32.237882 6396 sgd_solver.cpp:105] Iteration 3060, lr = 0.00545451 I0409 23:28:36.406498 6396 solver.cpp:218] Iteration 3072 (2.87874 iter/s, 4.16849s/12 iters), loss = 2.93876 I0409 23:28:36.406646 6396 solver.cpp:237] Train net output #0: loss = 2.93876 (* 1 = 2.93876 loss) I0409 23:28:36.406659 6396 sgd_solver.cpp:105] Iteration 3072, lr = 0.00544156 I0409 23:28:41.356833 6396 solver.cpp:218] Iteration 3084 (2.42423 iter/s, 4.95003s/12 iters), loss = 2.94269 I0409 23:28:41.356882 6396 solver.cpp:237] Train net output #0: loss = 2.94269 (* 1 = 2.94269 loss) I0409 23:28:41.356891 6396 sgd_solver.cpp:105] Iteration 3084, lr = 0.00542864 I0409 23:28:46.265175 6396 solver.cpp:218] Iteration 3096 (2.44492 iter/s, 4.90814s/12 iters), loss = 3.09695 I0409 23:28:46.265218 6396 solver.cpp:237] Train net output #0: loss = 3.09695 (* 1 = 3.09695 loss) I0409 23:28:46.265228 6396 sgd_solver.cpp:105] Iteration 3096, lr = 0.00541575 I0409 23:28:51.196815 6396 solver.cpp:218] Iteration 3108 (2.43337 iter/s, 4.93144s/12 iters), loss = 2.78946 I0409 23:28:51.196866 6396 solver.cpp:237] Train net output #0: loss = 2.78946 (* 1 = 2.78946 loss) I0409 23:28:51.196877 6396 sgd_solver.cpp:105] Iteration 3108, lr = 0.00540289 I0409 23:28:56.214243 6396 solver.cpp:218] Iteration 3120 (2.39176 iter/s, 5.01722s/12 iters), loss = 2.83518 I0409 23:28:56.214296 6396 solver.cpp:237] Train net output #0: loss = 2.83518 (* 1 = 2.83518 loss) I0409 23:28:56.214305 6396 sgd_solver.cpp:105] Iteration 3120, lr = 0.00539006 I0409 23:29:01.196305 6396 solver.cpp:218] Iteration 3132 (2.40874 iter/s, 4.98185s/12 iters), loss = 2.79641 I0409 23:29:01.196353 6396 solver.cpp:237] Train net output #0: loss = 2.79641 (* 1 = 2.79641 loss) I0409 23:29:01.196364 6396 sgd_solver.cpp:105] Iteration 3132, lr = 0.00537727 I0409 23:29:02.293349 6400 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:29:06.171682 6396 solver.cpp:218] Iteration 3144 (2.41198 iter/s, 4.97517s/12 iters), loss = 2.54074 I0409 23:29:06.171738 6396 solver.cpp:237] Train net output #0: loss = 2.54074 (* 1 = 2.54074 loss) I0409 23:29:06.171751 6396 sgd_solver.cpp:105] Iteration 3144, lr = 0.0053645 I0409 23:29:11.147612 6396 solver.cpp:218] Iteration 3156 (2.41171 iter/s, 4.97572s/12 iters), loss = 2.86028 I0409 23:29:11.147686 6396 solver.cpp:237] Train net output #0: loss = 2.86028 (* 1 = 2.86028 loss) I0409 23:29:11.147696 6396 sgd_solver.cpp:105] Iteration 3156, lr = 0.00535176 I0409 23:29:13.179744 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_3162.caffemodel I0409 23:29:14.641804 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_3162.solverstate I0409 23:29:14.966953 6396 solver.cpp:330] Iteration 3162, Testing net (#0) I0409 23:29:14.966975 6396 net.cpp:676] Ignoring source layer train-data I0409 23:29:18.050451 6401 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:29:19.310766 6396 solver.cpp:397] Test net output #0: accuracy = 0.271446 I0409 23:29:19.310799 6396 solver.cpp:397] Test net output #1: loss = 3.00344 (* 1 = 3.00344 loss) I0409 23:29:21.102608 6396 solver.cpp:218] Iteration 3168 (1.20547 iter/s, 9.95462s/12 iters), loss = 2.58983 I0409 23:29:21.102658 6396 solver.cpp:237] Train net output #0: loss = 2.58983 (* 1 = 2.58983 loss) I0409 23:29:21.102666 6396 sgd_solver.cpp:105] Iteration 3168, lr = 0.00533906 I0409 23:29:25.949996 6396 solver.cpp:218] Iteration 3180 (2.47566 iter/s, 4.84719s/12 iters), loss = 2.87345 I0409 23:29:25.950037 6396 solver.cpp:237] Train net output #0: loss = 2.87345 (* 1 = 2.87345 loss) I0409 23:29:25.950044 6396 sgd_solver.cpp:105] Iteration 3180, lr = 0.00532638 I0409 23:29:30.868533 6396 solver.cpp:218] Iteration 3192 (2.43985 iter/s, 4.91834s/12 iters), loss = 2.80032 I0409 23:29:30.868579 6396 solver.cpp:237] Train net output #0: loss = 2.80032 (* 1 = 2.80032 loss) I0409 23:29:30.868589 6396 sgd_solver.cpp:105] Iteration 3192, lr = 0.00531374 I0409 23:29:35.794553 6396 solver.cpp:218] Iteration 3204 (2.43614 iter/s, 4.92582s/12 iters), loss = 2.74604 I0409 23:29:35.794595 6396 solver.cpp:237] Train net output #0: loss = 2.74604 (* 1 = 2.74604 loss) I0409 23:29:35.794603 6396 sgd_solver.cpp:105] Iteration 3204, lr = 0.00530112 I0409 23:29:40.940201 6396 solver.cpp:218] Iteration 3216 (2.33216 iter/s, 5.14544s/12 iters), loss = 2.85592 I0409 23:29:40.940258 6396 solver.cpp:237] Train net output #0: loss = 2.85592 (* 1 = 2.85592 loss) I0409 23:29:40.940270 6396 sgd_solver.cpp:105] Iteration 3216, lr = 0.00528853 I0409 23:29:45.858140 6396 solver.cpp:218] Iteration 3228 (2.44015 iter/s, 4.91773s/12 iters), loss = 2.66524 I0409 23:29:45.858264 6396 solver.cpp:237] Train net output #0: loss = 2.66524 (* 1 = 2.66524 loss) I0409 23:29:45.858274 6396 sgd_solver.cpp:105] Iteration 3228, lr = 0.00527598 I0409 23:29:49.013859 6400 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:29:50.774327 6396 solver.cpp:218] Iteration 3240 (2.44105 iter/s, 4.91591s/12 iters), loss = 2.98872 I0409 23:29:50.774379 6396 solver.cpp:237] Train net output #0: loss = 2.98872 (* 1 = 2.98872 loss) I0409 23:29:50.774387 6396 sgd_solver.cpp:105] Iteration 3240, lr = 0.00526345 I0409 23:29:55.653903 6396 solver.cpp:218] Iteration 3252 (2.45933 iter/s, 4.87937s/12 iters), loss = 2.7101 I0409 23:29:55.653975 6396 solver.cpp:237] Train net output #0: loss = 2.7101 (* 1 = 2.7101 loss) I0409 23:29:55.653988 6396 sgd_solver.cpp:105] Iteration 3252, lr = 0.00525095 I0409 23:30:00.125454 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_3264.caffemodel I0409 23:30:00.658059 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_3264.solverstate I0409 23:30:00.975849 6396 solver.cpp:330] Iteration 3264, Testing net (#0) I0409 23:30:00.975873 6396 net.cpp:676] Ignoring source layer train-data I0409 23:30:04.206970 6401 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:30:05.557492 6396 solver.cpp:397] Test net output #0: accuracy = 0.302696 I0409 23:30:05.557520 6396 solver.cpp:397] Test net output #1: loss = 2.75662 (* 1 = 2.75662 loss) I0409 23:30:05.640612 6396 solver.cpp:218] Iteration 3264 (1.20164 iter/s, 9.98636s/12 iters), loss = 2.75577 I0409 23:30:05.640650 6396 solver.cpp:237] Train net output #0: loss = 2.75577 (* 1 = 2.75577 loss) I0409 23:30:05.640658 6396 sgd_solver.cpp:105] Iteration 3264, lr = 0.00523849 I0409 23:30:09.916091 6396 solver.cpp:218] Iteration 3276 (2.80682 iter/s, 4.2753s/12 iters), loss = 2.75935 I0409 23:30:09.916134 6396 solver.cpp:237] Train net output #0: loss = 2.75935 (* 1 = 2.75935 loss) I0409 23:30:09.916144 6396 sgd_solver.cpp:105] Iteration 3276, lr = 0.00522605 I0409 23:30:14.884045 6396 solver.cpp:218] Iteration 3288 (2.41558 iter/s, 4.96775s/12 iters), loss = 2.78987 I0409 23:30:14.884095 6396 solver.cpp:237] Train net output #0: loss = 2.78987 (* 1 = 2.78987 loss) I0409 23:30:14.884107 6396 sgd_solver.cpp:105] Iteration 3288, lr = 0.00521364 I0409 23:30:19.841665 6396 solver.cpp:218] Iteration 3300 (2.42062 iter/s, 4.95741s/12 iters), loss = 2.75551 I0409 23:30:19.841776 6396 solver.cpp:237] Train net output #0: loss = 2.75551 (* 1 = 2.75551 loss) I0409 23:30:19.841789 6396 sgd_solver.cpp:105] Iteration 3300, lr = 0.00520126 I0409 23:30:24.797587 6396 solver.cpp:218] Iteration 3312 (2.42147 iter/s, 4.95566s/12 iters), loss = 2.65682 I0409 23:30:24.797626 6396 solver.cpp:237] Train net output #0: loss = 2.65682 (* 1 = 2.65682 loss) I0409 23:30:24.797636 6396 sgd_solver.cpp:105] Iteration 3312, lr = 0.00518892 I0409 23:30:29.693454 6396 solver.cpp:218] Iteration 3324 (2.45115 iter/s, 4.89567s/12 iters), loss = 2.47431 I0409 23:30:29.693497 6396 solver.cpp:237] Train net output #0: loss = 2.47431 (* 1 = 2.47431 loss) I0409 23:30:29.693506 6396 sgd_solver.cpp:105] Iteration 3324, lr = 0.0051766 I0409 23:30:34.591689 6396 solver.cpp:218] Iteration 3336 (2.44996 iter/s, 4.89804s/12 iters), loss = 2.61373 I0409 23:30:34.591738 6396 solver.cpp:237] Train net output #0: loss = 2.61373 (* 1 = 2.61373 loss) I0409 23:30:34.591749 6396 sgd_solver.cpp:105] Iteration 3336, lr = 0.00516431 I0409 23:30:35.052076 6400 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:30:39.500286 6396 solver.cpp:218] Iteration 3348 (2.44479 iter/s, 4.90839s/12 iters), loss = 2.75752 I0409 23:30:39.500337 6396 solver.cpp:237] Train net output #0: loss = 2.75752 (* 1 = 2.75752 loss) I0409 23:30:39.500349 6396 sgd_solver.cpp:105] Iteration 3348, lr = 0.00515204 I0409 23:30:44.442811 6396 solver.cpp:218] Iteration 3360 (2.42801 iter/s, 4.94232s/12 iters), loss = 2.66446 I0409 23:30:44.442858 6396 solver.cpp:237] Train net output #0: loss = 2.66446 (* 1 = 2.66446 loss) I0409 23:30:44.442869 6396 sgd_solver.cpp:105] Iteration 3360, lr = 0.00513981 I0409 23:30:46.650827 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_3366.caffemodel I0409 23:30:47.109643 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_3366.solverstate I0409 23:30:47.428175 6396 solver.cpp:330] Iteration 3366, Testing net (#0) I0409 23:30:47.428205 6396 net.cpp:676] Ignoring source layer train-data I0409 23:30:50.436151 6401 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:30:51.768546 6396 solver.cpp:397] Test net output #0: accuracy = 0.30576 I0409 23:30:51.768577 6396 solver.cpp:397] Test net output #1: loss = 2.75935 (* 1 = 2.75935 loss) I0409 23:30:53.490573 6396 solver.cpp:218] Iteration 3372 (1.32634 iter/s, 9.04744s/12 iters), loss = 2.64847 I0409 23:30:53.490625 6396 solver.cpp:237] Train net output #0: loss = 2.64847 (* 1 = 2.64847 loss) I0409 23:30:53.490636 6396 sgd_solver.cpp:105] Iteration 3372, lr = 0.00512761 I0409 23:30:58.364307 6396 solver.cpp:218] Iteration 3384 (2.46228 iter/s, 4.87353s/12 iters), loss = 2.69692 I0409 23:30:58.364353 6396 solver.cpp:237] Train net output #0: loss = 2.69692 (* 1 = 2.69692 loss) I0409 23:30:58.364362 6396 sgd_solver.cpp:105] Iteration 3384, lr = 0.00511544 I0409 23:31:03.281452 6396 solver.cpp:218] Iteration 3396 (2.44054 iter/s, 4.91694s/12 iters), loss = 2.59377 I0409 23:31:03.281494 6396 solver.cpp:237] Train net output #0: loss = 2.59377 (* 1 = 2.59377 loss) I0409 23:31:03.281503 6396 sgd_solver.cpp:105] Iteration 3396, lr = 0.00510329 I0409 23:31:08.185282 6396 solver.cpp:218] Iteration 3408 (2.44717 iter/s, 4.90363s/12 iters), loss = 2.82858 I0409 23:31:08.185321 6396 solver.cpp:237] Train net output #0: loss = 2.82858 (* 1 = 2.82858 loss) I0409 23:31:08.185330 6396 sgd_solver.cpp:105] Iteration 3408, lr = 0.00509117 I0409 23:31:13.120580 6396 solver.cpp:218] Iteration 3420 (2.43156 iter/s, 4.9351s/12 iters), loss = 2.46627 I0409 23:31:13.120633 6396 solver.cpp:237] Train net output #0: loss = 2.46627 (* 1 = 2.46627 loss) I0409 23:31:13.120646 6396 sgd_solver.cpp:105] Iteration 3420, lr = 0.00507909 I0409 23:31:18.112628 6396 solver.cpp:218] Iteration 3432 (2.40392 iter/s, 4.99184s/12 iters), loss = 2.75714 I0409 23:31:18.112685 6396 solver.cpp:237] Train net output #0: loss = 2.75714 (* 1 = 2.75714 loss) I0409 23:31:18.112699 6396 sgd_solver.cpp:105] Iteration 3432, lr = 0.00506703 I0409 23:31:20.758184 6400 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:31:23.142920 6396 solver.cpp:218] Iteration 3444 (2.38565 iter/s, 5.03008s/12 iters), loss = 2.31122 I0409 23:31:23.142961 6396 solver.cpp:237] Train net output #0: loss = 2.31122 (* 1 = 2.31122 loss) I0409 23:31:23.142971 6396 sgd_solver.cpp:105] Iteration 3444, lr = 0.005055 I0409 23:31:28.021097 6396 solver.cpp:218] Iteration 3456 (2.46003 iter/s, 4.87798s/12 iters), loss = 2.59902 I0409 23:31:28.021142 6396 solver.cpp:237] Train net output #0: loss = 2.59902 (* 1 = 2.59902 loss) I0409 23:31:28.021150 6396 sgd_solver.cpp:105] Iteration 3456, lr = 0.005043 I0409 23:31:32.543267 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_3468.caffemodel I0409 23:31:32.977324 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_3468.solverstate I0409 23:31:33.288199 6396 solver.cpp:330] Iteration 3468, Testing net (#0) I0409 23:31:33.288221 6396 net.cpp:676] Ignoring source layer train-data I0409 23:31:33.319831 6396 blocking_queue.cpp:49] Waiting for data I0409 23:31:36.250692 6401 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:31:37.656399 6396 solver.cpp:397] Test net output #0: accuracy = 0.33701 I0409 23:31:37.656430 6396 solver.cpp:397] Test net output #1: loss = 2.65153 (* 1 = 2.65153 loss) I0409 23:31:37.739708 6396 solver.cpp:218] Iteration 3468 (1.23479 iter/s, 9.71828s/12 iters), loss = 2.43406 I0409 23:31:37.739750 6396 solver.cpp:237] Train net output #0: loss = 2.43406 (* 1 = 2.43406 loss) I0409 23:31:37.739758 6396 sgd_solver.cpp:105] Iteration 3468, lr = 0.00503102 I0409 23:31:41.881083 6396 solver.cpp:218] Iteration 3480 (2.89771 iter/s, 4.1412s/12 iters), loss = 2.41365 I0409 23:31:41.881124 6396 solver.cpp:237] Train net output #0: loss = 2.41365 (* 1 = 2.41365 loss) I0409 23:31:41.881131 6396 sgd_solver.cpp:105] Iteration 3480, lr = 0.00501908 I0409 23:31:46.895648 6396 solver.cpp:218] Iteration 3492 (2.39312 iter/s, 5.01437s/12 iters), loss = 2.67123 I0409 23:31:46.895696 6396 solver.cpp:237] Train net output #0: loss = 2.67123 (* 1 = 2.67123 loss) I0409 23:31:46.895705 6396 sgd_solver.cpp:105] Iteration 3492, lr = 0.00500716 I0409 23:31:51.813032 6396 solver.cpp:218] Iteration 3504 (2.44042 iter/s, 4.91718s/12 iters), loss = 2.56306 I0409 23:31:51.813154 6396 solver.cpp:237] Train net output #0: loss = 2.56306 (* 1 = 2.56306 loss) I0409 23:31:51.813166 6396 sgd_solver.cpp:105] Iteration 3504, lr = 0.00499527 I0409 23:31:56.673264 6396 solver.cpp:218] Iteration 3516 (2.46916 iter/s, 4.85996s/12 iters), loss = 2.50654 I0409 23:31:56.673321 6396 solver.cpp:237] Train net output #0: loss = 2.50654 (* 1 = 2.50654 loss) I0409 23:31:56.673332 6396 sgd_solver.cpp:105] Iteration 3516, lr = 0.00498341 I0409 23:32:01.633610 6396 solver.cpp:218] Iteration 3528 (2.41929 iter/s, 4.96014s/12 iters), loss = 2.46962 I0409 23:32:01.633652 6396 solver.cpp:237] Train net output #0: loss = 2.46962 (* 1 = 2.46962 loss) I0409 23:32:01.633661 6396 sgd_solver.cpp:105] Iteration 3528, lr = 0.00497158 I0409 23:32:06.510089 6400 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:32:06.768647 6396 solver.cpp:218] Iteration 3540 (2.33698 iter/s, 5.13483s/12 iters), loss = 2.43491 I0409 23:32:06.768698 6396 solver.cpp:237] Train net output #0: loss = 2.43491 (* 1 = 2.43491 loss) I0409 23:32:06.768710 6396 sgd_solver.cpp:105] Iteration 3540, lr = 0.00495978 I0409 23:32:11.753778 6396 solver.cpp:218] Iteration 3552 (2.40726 iter/s, 4.98492s/12 iters), loss = 2.42762 I0409 23:32:11.753827 6396 solver.cpp:237] Train net output #0: loss = 2.42762 (* 1 = 2.42762 loss) I0409 23:32:11.753840 6396 sgd_solver.cpp:105] Iteration 3552, lr = 0.004948 I0409 23:32:16.710485 6396 solver.cpp:218] Iteration 3564 (2.42106 iter/s, 4.9565s/12 iters), loss = 2.22869 I0409 23:32:16.710539 6396 solver.cpp:237] Train net output #0: loss = 2.22869 (* 1 = 2.22869 loss) I0409 23:32:16.710551 6396 sgd_solver.cpp:105] Iteration 3564, lr = 0.00493626 I0409 23:32:18.883941 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_3570.caffemodel I0409 23:32:19.357728 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_3570.solverstate I0409 23:32:19.690624 6396 solver.cpp:330] Iteration 3570, Testing net (#0) I0409 23:32:19.690661 6396 net.cpp:676] Ignoring source layer train-data I0409 23:32:22.862548 6401 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:32:24.371155 6396 solver.cpp:397] Test net output #0: accuracy = 0.334559 I0409 23:32:24.371191 6396 solver.cpp:397] Test net output #1: loss = 2.58929 (* 1 = 2.58929 loss) I0409 23:32:26.236177 6396 solver.cpp:218] Iteration 3576 (1.2598 iter/s, 9.52535s/12 iters), loss = 2.65176 I0409 23:32:26.236222 6396 solver.cpp:237] Train net output #0: loss = 2.65176 (* 1 = 2.65176 loss) I0409 23:32:26.236230 6396 sgd_solver.cpp:105] Iteration 3576, lr = 0.00492454 I0409 23:32:31.104327 6396 solver.cpp:218] Iteration 3588 (2.4651 iter/s, 4.86795s/12 iters), loss = 2.38618 I0409 23:32:31.104377 6396 solver.cpp:237] Train net output #0: loss = 2.38618 (* 1 = 2.38618 loss) I0409 23:32:31.104390 6396 sgd_solver.cpp:105] Iteration 3588, lr = 0.00491284 I0409 23:32:36.097313 6396 solver.cpp:218] Iteration 3600 (2.40347 iter/s, 4.99278s/12 iters), loss = 2.53459 I0409 23:32:36.097369 6396 solver.cpp:237] Train net output #0: loss = 2.53459 (* 1 = 2.53459 loss) I0409 23:32:36.097381 6396 sgd_solver.cpp:105] Iteration 3600, lr = 0.00490118 I0409 23:32:40.941974 6396 solver.cpp:218] Iteration 3612 (2.47707 iter/s, 4.84443s/12 iters), loss = 2.50511 I0409 23:32:40.942040 6396 solver.cpp:237] Train net output #0: loss = 2.50511 (* 1 = 2.50511 loss) I0409 23:32:40.942051 6396 sgd_solver.cpp:105] Iteration 3612, lr = 0.00488954 I0409 23:32:45.949292 6396 solver.cpp:218] Iteration 3624 (2.3966 iter/s, 5.0071s/12 iters), loss = 2.5658 I0409 23:32:45.949333 6396 solver.cpp:237] Train net output #0: loss = 2.5658 (* 1 = 2.5658 loss) I0409 23:32:45.949342 6396 sgd_solver.cpp:105] Iteration 3624, lr = 0.00487793 I0409 23:32:50.937438 6396 solver.cpp:218] Iteration 3636 (2.4058 iter/s, 4.98794s/12 iters), loss = 2.36077 I0409 23:32:50.937489 6396 solver.cpp:237] Train net output #0: loss = 2.36077 (* 1 = 2.36077 loss) I0409 23:32:50.937500 6396 sgd_solver.cpp:105] Iteration 3636, lr = 0.00486635 I0409 23:32:52.790081 6400 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:32:55.905297 6396 solver.cpp:218] Iteration 3648 (2.41563 iter/s, 4.96765s/12 iters), loss = 2.04617 I0409 23:32:55.906711 6396 solver.cpp:237] Train net output #0: loss = 2.04617 (* 1 = 2.04617 loss) I0409 23:32:55.906725 6396 sgd_solver.cpp:105] Iteration 3648, lr = 0.0048548 I0409 23:33:01.010923 6396 solver.cpp:218] Iteration 3660 (2.35107 iter/s, 5.10405s/12 iters), loss = 2.37293 I0409 23:33:01.010979 6396 solver.cpp:237] Train net output #0: loss = 2.37293 (* 1 = 2.37293 loss) I0409 23:33:01.010990 6396 sgd_solver.cpp:105] Iteration 3660, lr = 0.00484327 I0409 23:33:05.510445 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_3672.caffemodel I0409 23:33:06.374545 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_3672.solverstate I0409 23:33:08.487803 6396 solver.cpp:330] Iteration 3672, Testing net (#0) I0409 23:33:08.487833 6396 net.cpp:676] Ignoring source layer train-data I0409 23:33:11.459666 6401 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:33:12.915874 6396 solver.cpp:397] Test net output #0: accuracy = 0.370098 I0409 23:33:12.915910 6396 solver.cpp:397] Test net output #1: loss = 2.42089 (* 1 = 2.42089 loss) I0409 23:33:12.998996 6396 solver.cpp:218] Iteration 3672 (1.00103 iter/s, 11.9877s/12 iters), loss = 1.98345 I0409 23:33:12.999044 6396 solver.cpp:237] Train net output #0: loss = 1.98345 (* 1 = 1.98345 loss) I0409 23:33:12.999058 6396 sgd_solver.cpp:105] Iteration 3672, lr = 0.00483177 I0409 23:33:17.035991 6396 solver.cpp:218] Iteration 3684 (2.97264 iter/s, 4.03682s/12 iters), loss = 2.34353 I0409 23:33:17.036038 6396 solver.cpp:237] Train net output #0: loss = 2.34353 (* 1 = 2.34353 loss) I0409 23:33:17.036048 6396 sgd_solver.cpp:105] Iteration 3684, lr = 0.0048203 I0409 23:33:22.042064 6396 solver.cpp:218] Iteration 3696 (2.39718 iter/s, 5.00587s/12 iters), loss = 1.84847 I0409 23:33:22.042098 6396 solver.cpp:237] Train net output #0: loss = 1.84847 (* 1 = 1.84847 loss) I0409 23:33:22.042106 6396 sgd_solver.cpp:105] Iteration 3696, lr = 0.00480886 I0409 23:33:27.036886 6396 solver.cpp:218] Iteration 3708 (2.40259 iter/s, 4.99462s/12 iters), loss = 2.31439 I0409 23:33:27.037037 6396 solver.cpp:237] Train net output #0: loss = 2.31439 (* 1 = 2.31439 loss) I0409 23:33:27.037051 6396 sgd_solver.cpp:105] Iteration 3708, lr = 0.00479744 I0409 23:33:32.025446 6396 solver.cpp:218] Iteration 3720 (2.40565 iter/s, 4.98826s/12 iters), loss = 2.33099 I0409 23:33:32.025488 6396 solver.cpp:237] Train net output #0: loss = 2.33099 (* 1 = 2.33099 loss) I0409 23:33:32.025497 6396 sgd_solver.cpp:105] Iteration 3720, lr = 0.00478605 I0409 23:33:36.899219 6396 solver.cpp:218] Iteration 3732 (2.46226 iter/s, 4.87357s/12 iters), loss = 2.22179 I0409 23:33:36.899267 6396 solver.cpp:237] Train net output #0: loss = 2.22179 (* 1 = 2.22179 loss) I0409 23:33:36.899276 6396 sgd_solver.cpp:105] Iteration 3732, lr = 0.00477469 I0409 23:33:40.969275 6400 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:33:42.038000 6396 solver.cpp:218] Iteration 3744 (2.33528 iter/s, 5.13856s/12 iters), loss = 2.24235 I0409 23:33:42.038056 6396 solver.cpp:237] Train net output #0: loss = 2.24235 (* 1 = 2.24235 loss) I0409 23:33:42.038069 6396 sgd_solver.cpp:105] Iteration 3744, lr = 0.00476335 I0409 23:33:47.058269 6396 solver.cpp:218] Iteration 3756 (2.39041 iter/s, 5.02006s/12 iters), loss = 2.43215 I0409 23:33:47.058313 6396 solver.cpp:237] Train net output #0: loss = 2.43215 (* 1 = 2.43215 loss) I0409 23:33:47.058322 6396 sgd_solver.cpp:105] Iteration 3756, lr = 0.00475204 I0409 23:33:52.111732 6396 solver.cpp:218] Iteration 3768 (2.37471 iter/s, 5.05326s/12 iters), loss = 2.34006 I0409 23:33:52.111774 6396 solver.cpp:237] Train net output #0: loss = 2.34006 (* 1 = 2.34006 loss) I0409 23:33:52.111783 6396 sgd_solver.cpp:105] Iteration 3768, lr = 0.00474076 I0409 23:33:54.132010 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_3774.caffemodel I0409 23:33:54.549702 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_3774.solverstate I0409 23:33:54.856447 6396 solver.cpp:330] Iteration 3774, Testing net (#0) I0409 23:33:54.856477 6396 net.cpp:676] Ignoring source layer train-data I0409 23:33:57.868743 6401 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:33:59.361579 6396 solver.cpp:397] Test net output #0: accuracy = 0.362745 I0409 23:33:59.361632 6396 solver.cpp:397] Test net output #1: loss = 2.48612 (* 1 = 2.48612 loss) I0409 23:34:01.060631 6396 solver.cpp:218] Iteration 3780 (1.34099 iter/s, 8.94859s/12 iters), loss = 2.24261 I0409 23:34:01.060678 6396 solver.cpp:237] Train net output #0: loss = 2.24261 (* 1 = 2.24261 loss) I0409 23:34:01.060689 6396 sgd_solver.cpp:105] Iteration 3780, lr = 0.00472951 I0409 23:34:05.971174 6396 solver.cpp:218] Iteration 3792 (2.44382 iter/s, 4.91035s/12 iters), loss = 2.19902 I0409 23:34:05.971218 6396 solver.cpp:237] Train net output #0: loss = 2.19902 (* 1 = 2.19902 loss) I0409 23:34:05.971227 6396 sgd_solver.cpp:105] Iteration 3792, lr = 0.00471828 I0409 23:34:10.907732 6396 solver.cpp:218] Iteration 3804 (2.43094 iter/s, 4.93636s/12 iters), loss = 2.32504 I0409 23:34:10.907768 6396 solver.cpp:237] Train net output #0: loss = 2.32504 (* 1 = 2.32504 loss) I0409 23:34:10.907775 6396 sgd_solver.cpp:105] Iteration 3804, lr = 0.00470707 I0409 23:34:15.892066 6396 solver.cpp:218] Iteration 3816 (2.40764 iter/s, 4.98413s/12 iters), loss = 2.13417 I0409 23:34:15.892117 6396 solver.cpp:237] Train net output #0: loss = 2.13417 (* 1 = 2.13417 loss) I0409 23:34:15.892128 6396 sgd_solver.cpp:105] Iteration 3816, lr = 0.0046959 I0409 23:34:20.848233 6396 solver.cpp:218] Iteration 3828 (2.42133 iter/s, 4.95596s/12 iters), loss = 2.05313 I0409 23:34:20.848289 6396 solver.cpp:237] Train net output #0: loss = 2.05313 (* 1 = 2.05313 loss) I0409 23:34:20.848300 6396 sgd_solver.cpp:105] Iteration 3828, lr = 0.00468475 I0409 23:34:25.807925 6396 solver.cpp:218] Iteration 3840 (2.41961 iter/s, 4.95948s/12 iters), loss = 2.15897 I0409 23:34:25.807974 6396 solver.cpp:237] Train net output #0: loss = 2.15897 (* 1 = 2.15897 loss) I0409 23:34:25.807984 6396 sgd_solver.cpp:105] Iteration 3840, lr = 0.00467363 I0409 23:34:26.915323 6400 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:34:30.741466 6396 solver.cpp:218] Iteration 3852 (2.43243 iter/s, 4.93333s/12 iters), loss = 2.07998 I0409 23:34:30.741614 6396 solver.cpp:237] Train net output #0: loss = 2.07998 (* 1 = 2.07998 loss) I0409 23:34:30.741627 6396 sgd_solver.cpp:105] Iteration 3852, lr = 0.00466253 I0409 23:34:35.674064 6396 solver.cpp:218] Iteration 3864 (2.43295 iter/s, 4.93229s/12 iters), loss = 2.29533 I0409 23:34:35.674120 6396 solver.cpp:237] Train net output #0: loss = 2.29533 (* 1 = 2.29533 loss) I0409 23:34:35.674132 6396 sgd_solver.cpp:105] Iteration 3864, lr = 0.00465146 I0409 23:34:40.158047 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_3876.caffemodel I0409 23:34:40.628317 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_3876.solverstate I0409 23:34:40.948771 6396 solver.cpp:330] Iteration 3876, Testing net (#0) I0409 23:34:40.948799 6396 net.cpp:676] Ignoring source layer train-data I0409 23:34:43.812628 6401 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:34:45.352830 6396 solver.cpp:397] Test net output #0: accuracy = 0.367034 I0409 23:34:45.352875 6396 solver.cpp:397] Test net output #1: loss = 2.52871 (* 1 = 2.52871 loss) I0409 23:34:45.436288 6396 solver.cpp:218] Iteration 3876 (1.22927 iter/s, 9.76187s/12 iters), loss = 2.20479 I0409 23:34:45.436358 6396 solver.cpp:237] Train net output #0: loss = 2.20479 (* 1 = 2.20479 loss) I0409 23:34:45.436374 6396 sgd_solver.cpp:105] Iteration 3876, lr = 0.00464042 I0409 23:34:49.551844 6396 solver.cpp:218] Iteration 3888 (2.91591 iter/s, 4.11535s/12 iters), loss = 2.12435 I0409 23:34:49.551905 6396 solver.cpp:237] Train net output #0: loss = 2.12435 (* 1 = 2.12435 loss) I0409 23:34:49.551918 6396 sgd_solver.cpp:105] Iteration 3888, lr = 0.0046294 I0409 23:34:54.484526 6396 solver.cpp:218] Iteration 3900 (2.43286 iter/s, 4.93246s/12 iters), loss = 2.13643 I0409 23:34:54.484573 6396 solver.cpp:237] Train net output #0: loss = 2.13643 (* 1 = 2.13643 loss) I0409 23:34:54.484583 6396 sgd_solver.cpp:105] Iteration 3900, lr = 0.00461841 I0409 23:34:59.421525 6396 solver.cpp:218] Iteration 3912 (2.43073 iter/s, 4.93679s/12 iters), loss = 1.96929 I0409 23:34:59.421571 6396 solver.cpp:237] Train net output #0: loss = 1.96929 (* 1 = 1.96929 loss) I0409 23:34:59.421581 6396 sgd_solver.cpp:105] Iteration 3912, lr = 0.00460744 I0409 23:35:04.347028 6396 solver.cpp:218] Iteration 3924 (2.4364 iter/s, 4.9253s/12 iters), loss = 2.30534 I0409 23:35:04.347158 6396 solver.cpp:237] Train net output #0: loss = 2.30534 (* 1 = 2.30534 loss) I0409 23:35:04.347173 6396 sgd_solver.cpp:105] Iteration 3924, lr = 0.0045965 I0409 23:35:09.307339 6396 solver.cpp:218] Iteration 3936 (2.41934 iter/s, 4.96003s/12 iters), loss = 2.10263 I0409 23:35:09.307390 6396 solver.cpp:237] Train net output #0: loss = 2.10263 (* 1 = 2.10263 loss) I0409 23:35:09.307402 6396 sgd_solver.cpp:105] Iteration 3936, lr = 0.00458559 I0409 23:35:12.663535 6400 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:35:14.285925 6396 solver.cpp:218] Iteration 3948 (2.41042 iter/s, 4.97838s/12 iters), loss = 1.994 I0409 23:35:14.285995 6396 solver.cpp:237] Train net output #0: loss = 1.994 (* 1 = 1.994 loss) I0409 23:35:14.286008 6396 sgd_solver.cpp:105] Iteration 3948, lr = 0.0045747 I0409 23:35:19.295078 6396 solver.cpp:218] Iteration 3960 (2.39572 iter/s, 5.00893s/12 iters), loss = 2.40154 I0409 23:35:19.295126 6396 solver.cpp:237] Train net output #0: loss = 2.40154 (* 1 = 2.40154 loss) I0409 23:35:19.295135 6396 sgd_solver.cpp:105] Iteration 3960, lr = 0.00456384 I0409 23:35:24.194074 6396 solver.cpp:218] Iteration 3972 (2.44958 iter/s, 4.89879s/12 iters), loss = 2.20827 I0409 23:35:24.194120 6396 solver.cpp:237] Train net output #0: loss = 2.20827 (* 1 = 2.20827 loss) I0409 23:35:24.194129 6396 sgd_solver.cpp:105] Iteration 3972, lr = 0.00455301 I0409 23:35:26.241912 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_3978.caffemodel I0409 23:35:26.665354 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_3978.solverstate I0409 23:35:26.973456 6396 solver.cpp:330] Iteration 3978, Testing net (#0) I0409 23:35:26.973479 6396 net.cpp:676] Ignoring source layer train-data I0409 23:35:29.895996 6401 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:35:31.671955 6396 solver.cpp:397] Test net output #0: accuracy = 0.387255 I0409 23:35:31.671998 6396 solver.cpp:397] Test net output #1: loss = 2.38403 (* 1 = 2.38403 loss) I0409 23:35:33.603036 6396 solver.cpp:218] Iteration 3984 (1.27542 iter/s, 9.40863s/12 iters), loss = 2.30946 I0409 23:35:33.603082 6396 solver.cpp:237] Train net output #0: loss = 2.30946 (* 1 = 2.30946 loss) I0409 23:35:33.603091 6396 sgd_solver.cpp:105] Iteration 3984, lr = 0.0045422 I0409 23:35:38.536051 6396 solver.cpp:218] Iteration 3996 (2.43269 iter/s, 4.93281s/12 iters), loss = 2.14429 I0409 23:35:38.536187 6396 solver.cpp:237] Train net output #0: loss = 2.14429 (* 1 = 2.14429 loss) I0409 23:35:38.536197 6396 sgd_solver.cpp:105] Iteration 3996, lr = 0.00453141 I0409 23:35:43.449183 6396 solver.cpp:218] Iteration 4008 (2.44258 iter/s, 4.91284s/12 iters), loss = 2.22945 I0409 23:35:43.449240 6396 solver.cpp:237] Train net output #0: loss = 2.22945 (* 1 = 2.22945 loss) I0409 23:35:43.449252 6396 sgd_solver.cpp:105] Iteration 4008, lr = 0.00452066 I0409 23:35:48.357161 6396 solver.cpp:218] Iteration 4020 (2.44511 iter/s, 4.90776s/12 iters), loss = 2.24465 I0409 23:35:48.357206 6396 solver.cpp:237] Train net output #0: loss = 2.24465 (* 1 = 2.24465 loss) I0409 23:35:48.357215 6396 sgd_solver.cpp:105] Iteration 4020, lr = 0.00450992 I0409 23:35:53.292090 6396 solver.cpp:218] Iteration 4032 (2.43175 iter/s, 4.93472s/12 iters), loss = 2.2244 I0409 23:35:53.292135 6396 solver.cpp:237] Train net output #0: loss = 2.2244 (* 1 = 2.2244 loss) I0409 23:35:53.292145 6396 sgd_solver.cpp:105] Iteration 4032, lr = 0.00449921 I0409 23:35:58.181742 6396 solver.cpp:218] Iteration 4044 (2.45426 iter/s, 4.88945s/12 iters), loss = 2.00603 I0409 23:35:58.181797 6396 solver.cpp:237] Train net output #0: loss = 2.00603 (* 1 = 2.00603 loss) I0409 23:35:58.181807 6396 sgd_solver.cpp:105] Iteration 4044, lr = 0.00448853 I0409 23:35:58.662981 6400 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:36:03.108047 6396 solver.cpp:218] Iteration 4056 (2.43601 iter/s, 4.92609s/12 iters), loss = 2.24287 I0409 23:36:03.108100 6396 solver.cpp:237] Train net output #0: loss = 2.24287 (* 1 = 2.24287 loss) I0409 23:36:03.108112 6396 sgd_solver.cpp:105] Iteration 4056, lr = 0.00447788 I0409 23:36:08.046268 6396 solver.cpp:218] Iteration 4068 (2.43013 iter/s, 4.93801s/12 iters), loss = 1.96323 I0409 23:36:08.046326 6396 solver.cpp:237] Train net output #0: loss = 1.96323 (* 1 = 1.96323 loss) I0409 23:36:08.046339 6396 sgd_solver.cpp:105] Iteration 4068, lr = 0.00446724 I0409 23:36:12.511226 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_4080.caffemodel I0409 23:36:13.382416 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_4080.solverstate I0409 23:36:13.778892 6396 solver.cpp:330] Iteration 4080, Testing net (#0) I0409 23:36:13.778913 6396 net.cpp:676] Ignoring source layer train-data I0409 23:36:16.554854 6401 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:36:18.185047 6396 solver.cpp:397] Test net output #0: accuracy = 0.382966 I0409 23:36:18.185091 6396 solver.cpp:397] Test net output #1: loss = 2.38865 (* 1 = 2.38865 loss) I0409 23:36:18.268390 6396 solver.cpp:218] Iteration 4080 (1.17397 iter/s, 10.2218s/12 iters), loss = 1.90834 I0409 23:36:18.268440 6396 solver.cpp:237] Train net output #0: loss = 1.90834 (* 1 = 1.90834 loss) I0409 23:36:18.268451 6396 sgd_solver.cpp:105] Iteration 4080, lr = 0.00445664 I0409 23:36:22.477504 6396 solver.cpp:218] Iteration 4092 (2.85108 iter/s, 4.20893s/12 iters), loss = 2.25098 I0409 23:36:22.477558 6396 solver.cpp:237] Train net output #0: loss = 2.25098 (* 1 = 2.25098 loss) I0409 23:36:22.477571 6396 sgd_solver.cpp:105] Iteration 4092, lr = 0.00444606 I0409 23:36:27.595609 6396 solver.cpp:218] Iteration 4104 (2.34472 iter/s, 5.11789s/12 iters), loss = 2.21235 I0409 23:36:27.595661 6396 solver.cpp:237] Train net output #0: loss = 2.21235 (* 1 = 2.21235 loss) I0409 23:36:27.595672 6396 sgd_solver.cpp:105] Iteration 4104, lr = 0.0044355 I0409 23:36:32.777334 6396 solver.cpp:218] Iteration 4116 (2.31593 iter/s, 5.1815s/12 iters), loss = 2.11744 I0409 23:36:32.777392 6396 solver.cpp:237] Train net output #0: loss = 2.11744 (* 1 = 2.11744 loss) I0409 23:36:32.777405 6396 sgd_solver.cpp:105] Iteration 4116, lr = 0.00442497 I0409 23:36:37.698987 6396 solver.cpp:218] Iteration 4128 (2.43831 iter/s, 4.92144s/12 iters), loss = 1.88276 I0409 23:36:37.699039 6396 solver.cpp:237] Train net output #0: loss = 1.88276 (* 1 = 1.88276 loss) I0409 23:36:37.699051 6396 sgd_solver.cpp:105] Iteration 4128, lr = 0.00441447 I0409 23:36:42.639921 6396 solver.cpp:218] Iteration 4140 (2.42879 iter/s, 4.94073s/12 iters), loss = 1.93207 I0409 23:36:42.640041 6396 solver.cpp:237] Train net output #0: loss = 1.93207 (* 1 = 1.93207 loss) I0409 23:36:42.640051 6396 sgd_solver.cpp:105] Iteration 4140, lr = 0.00440398 I0409 23:36:45.355258 6400 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:36:47.773627 6396 solver.cpp:218] Iteration 4152 (2.33762 iter/s, 5.13342s/12 iters), loss = 1.7884 I0409 23:36:47.773684 6396 solver.cpp:237] Train net output #0: loss = 1.7884 (* 1 = 1.7884 loss) I0409 23:36:47.773695 6396 sgd_solver.cpp:105] Iteration 4152, lr = 0.00439353 I0409 23:36:47.774013 6396 blocking_queue.cpp:49] Waiting for data I0409 23:36:52.672773 6396 solver.cpp:218] Iteration 4164 (2.44951 iter/s, 4.89893s/12 iters), loss = 2.11157 I0409 23:36:52.672830 6396 solver.cpp:237] Train net output #0: loss = 2.11157 (* 1 = 2.11157 loss) I0409 23:36:52.672843 6396 sgd_solver.cpp:105] Iteration 4164, lr = 0.0043831 I0409 23:36:57.580739 6396 solver.cpp:218] Iteration 4176 (2.44511 iter/s, 4.90776s/12 iters), loss = 1.95386 I0409 23:36:57.580790 6396 solver.cpp:237] Train net output #0: loss = 1.95386 (* 1 = 1.95386 loss) I0409 23:36:57.580801 6396 sgd_solver.cpp:105] Iteration 4176, lr = 0.00437269 I0409 23:36:59.584210 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_4182.caffemodel I0409 23:37:00.082202 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_4182.solverstate I0409 23:37:00.712510 6396 solver.cpp:330] Iteration 4182, Testing net (#0) I0409 23:37:00.712535 6396 net.cpp:676] Ignoring source layer train-data I0409 23:37:03.535848 6401 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:37:05.201402 6396 solver.cpp:397] Test net output #0: accuracy = 0.413603 I0409 23:37:05.201449 6396 solver.cpp:397] Test net output #1: loss = 2.29466 (* 1 = 2.29466 loss) I0409 23:37:07.077759 6396 solver.cpp:218] Iteration 4188 (1.2636 iter/s, 9.49668s/12 iters), loss = 1.84102 I0409 23:37:07.077811 6396 solver.cpp:237] Train net output #0: loss = 1.84102 (* 1 = 1.84102 loss) I0409 23:37:07.077822 6396 sgd_solver.cpp:105] Iteration 4188, lr = 0.00436231 I0409 23:37:12.088758 6396 solver.cpp:218] Iteration 4200 (2.39483 iter/s, 5.01079s/12 iters), loss = 1.92706 I0409 23:37:12.088809 6396 solver.cpp:237] Train net output #0: loss = 1.92706 (* 1 = 1.92706 loss) I0409 23:37:12.088819 6396 sgd_solver.cpp:105] Iteration 4200, lr = 0.00435195 I0409 23:37:17.070518 6396 solver.cpp:218] Iteration 4212 (2.40889 iter/s, 4.98155s/12 iters), loss = 1.87248 I0409 23:37:17.070683 6396 solver.cpp:237] Train net output #0: loss = 1.87248 (* 1 = 1.87248 loss) I0409 23:37:17.070698 6396 sgd_solver.cpp:105] Iteration 4212, lr = 0.00434162 I0409 23:37:22.173370 6396 solver.cpp:218] Iteration 4224 (2.35177 iter/s, 5.10254s/12 iters), loss = 1.98834 I0409 23:37:22.173411 6396 solver.cpp:237] Train net output #0: loss = 1.98834 (* 1 = 1.98834 loss) I0409 23:37:22.173420 6396 sgd_solver.cpp:105] Iteration 4224, lr = 0.00433131 I0409 23:37:27.071841 6396 solver.cpp:218] Iteration 4236 (2.44984 iter/s, 4.89828s/12 iters), loss = 1.98223 I0409 23:37:27.071884 6396 solver.cpp:237] Train net output #0: loss = 1.98223 (* 1 = 1.98223 loss) I0409 23:37:27.071893 6396 sgd_solver.cpp:105] Iteration 4236, lr = 0.00432103 I0409 23:37:31.731400 6400 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:37:31.959520 6396 solver.cpp:218] Iteration 4248 (2.45525 iter/s, 4.88748s/12 iters), loss = 1.96889 I0409 23:37:31.959575 6396 solver.cpp:237] Train net output #0: loss = 1.96889 (* 1 = 1.96889 loss) I0409 23:37:31.959587 6396 sgd_solver.cpp:105] Iteration 4248, lr = 0.00431077 I0409 23:37:36.869053 6396 solver.cpp:218] Iteration 4260 (2.44433 iter/s, 4.90932s/12 iters), loss = 1.84662 I0409 23:37:36.869110 6396 solver.cpp:237] Train net output #0: loss = 1.84662 (* 1 = 1.84662 loss) I0409 23:37:36.869122 6396 sgd_solver.cpp:105] Iteration 4260, lr = 0.00430053 I0409 23:37:41.787611 6396 solver.cpp:218] Iteration 4272 (2.43984 iter/s, 4.91835s/12 iters), loss = 1.88822 I0409 23:37:41.787659 6396 solver.cpp:237] Train net output #0: loss = 1.88822 (* 1 = 1.88822 loss) I0409 23:37:41.787670 6396 sgd_solver.cpp:105] Iteration 4272, lr = 0.00429032 I0409 23:37:46.326473 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_4284.caffemodel I0409 23:37:46.806380 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_4284.solverstate I0409 23:37:47.437104 6396 solver.cpp:330] Iteration 4284, Testing net (#0) I0409 23:37:47.437191 6396 net.cpp:676] Ignoring source layer train-data I0409 23:37:50.091257 6401 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:37:51.796758 6396 solver.cpp:397] Test net output #0: accuracy = 0.425245 I0409 23:37:51.796808 6396 solver.cpp:397] Test net output #1: loss = 2.22491 (* 1 = 2.22491 loss) I0409 23:37:51.880434 6396 solver.cpp:218] Iteration 4284 (1.189 iter/s, 10.0925s/12 iters), loss = 1.9019 I0409 23:37:51.880482 6396 solver.cpp:237] Train net output #0: loss = 1.9019 (* 1 = 1.9019 loss) I0409 23:37:51.880493 6396 sgd_solver.cpp:105] Iteration 4284, lr = 0.00428014 I0409 23:37:56.150620 6396 solver.cpp:218] Iteration 4296 (2.8103 iter/s, 4.27s/12 iters), loss = 1.88204 I0409 23:37:56.150665 6396 solver.cpp:237] Train net output #0: loss = 1.88204 (* 1 = 1.88204 loss) I0409 23:37:56.150673 6396 sgd_solver.cpp:105] Iteration 4296, lr = 0.00426998 I0409 23:38:01.225839 6396 solver.cpp:218] Iteration 4308 (2.36452 iter/s, 5.07502s/12 iters), loss = 1.6653 I0409 23:38:01.225886 6396 solver.cpp:237] Train net output #0: loss = 1.6653 (* 1 = 1.6653 loss) I0409 23:38:01.225898 6396 sgd_solver.cpp:105] Iteration 4308, lr = 0.00425984 I0409 23:38:06.203702 6396 solver.cpp:218] Iteration 4320 (2.41077 iter/s, 4.97766s/12 iters), loss = 2.0276 I0409 23:38:06.203755 6396 solver.cpp:237] Train net output #0: loss = 2.0276 (* 1 = 2.0276 loss) I0409 23:38:06.203768 6396 sgd_solver.cpp:105] Iteration 4320, lr = 0.00424972 I0409 23:38:11.212116 6396 solver.cpp:218] Iteration 4332 (2.39607 iter/s, 5.0082s/12 iters), loss = 1.92924 I0409 23:38:11.212170 6396 solver.cpp:237] Train net output #0: loss = 1.92924 (* 1 = 1.92924 loss) I0409 23:38:11.212182 6396 sgd_solver.cpp:105] Iteration 4332, lr = 0.00423964 I0409 23:38:16.242043 6396 solver.cpp:218] Iteration 4344 (2.38582 iter/s, 5.02972s/12 iters), loss = 1.80821 I0409 23:38:16.242107 6396 solver.cpp:237] Train net output #0: loss = 1.80821 (* 1 = 1.80821 loss) I0409 23:38:16.242120 6396 sgd_solver.cpp:105] Iteration 4344, lr = 0.00422957 I0409 23:38:18.142449 6400 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:38:21.203181 6396 solver.cpp:218] Iteration 4356 (2.4189 iter/s, 4.96093s/12 iters), loss = 1.73378 I0409 23:38:21.203230 6396 solver.cpp:237] Train net output #0: loss = 1.73378 (* 1 = 1.73378 loss) I0409 23:38:21.203241 6396 sgd_solver.cpp:105] Iteration 4356, lr = 0.00421953 I0409 23:38:26.260812 6396 solver.cpp:218] Iteration 4368 (2.37275 iter/s, 5.05742s/12 iters), loss = 1.75265 I0409 23:38:26.260867 6396 solver.cpp:237] Train net output #0: loss = 1.75265 (* 1 = 1.75265 loss) I0409 23:38:26.260879 6396 sgd_solver.cpp:105] Iteration 4368, lr = 0.00420951 I0409 23:38:31.276007 6396 solver.cpp:218] Iteration 4380 (2.39283 iter/s, 5.01498s/12 iters), loss = 1.73971 I0409 23:38:31.276059 6396 solver.cpp:237] Train net output #0: loss = 1.73971 (* 1 = 1.73971 loss) I0409 23:38:31.276072 6396 sgd_solver.cpp:105] Iteration 4380, lr = 0.00419952 I0409 23:38:33.281033 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_4386.caffemodel I0409 23:38:33.797456 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_4386.solverstate I0409 23:38:34.401669 6396 solver.cpp:330] Iteration 4386, Testing net (#0) I0409 23:38:34.401700 6396 net.cpp:676] Ignoring source layer train-data I0409 23:38:37.014873 6401 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:38:38.765575 6396 solver.cpp:397] Test net output #0: accuracy = 0.432598 I0409 23:38:38.765623 6396 solver.cpp:397] Test net output #1: loss = 2.1914 (* 1 = 2.1914 loss) I0409 23:38:40.697240 6396 solver.cpp:218] Iteration 4392 (1.27376 iter/s, 9.4209s/12 iters), loss = 1.89606 I0409 23:38:40.697288 6396 solver.cpp:237] Train net output #0: loss = 1.89606 (* 1 = 1.89606 loss) I0409 23:38:40.697297 6396 sgd_solver.cpp:105] Iteration 4392, lr = 0.00418954 I0409 23:38:45.646157 6396 solver.cpp:218] Iteration 4404 (2.42487 iter/s, 4.94871s/12 iters), loss = 1.55932 I0409 23:38:45.646209 6396 solver.cpp:237] Train net output #0: loss = 1.55932 (* 1 = 1.55932 loss) I0409 23:38:45.646220 6396 sgd_solver.cpp:105] Iteration 4404, lr = 0.0041796 I0409 23:38:50.668445 6396 solver.cpp:218] Iteration 4416 (2.38945 iter/s, 5.02208s/12 iters), loss = 1.82919 I0409 23:38:50.668553 6396 solver.cpp:237] Train net output #0: loss = 1.82919 (* 1 = 1.82919 loss) I0409 23:38:50.668565 6396 sgd_solver.cpp:105] Iteration 4416, lr = 0.00416967 I0409 23:38:55.885342 6396 solver.cpp:218] Iteration 4428 (2.30034 iter/s, 5.21663s/12 iters), loss = 1.91874 I0409 23:38:55.885398 6396 solver.cpp:237] Train net output #0: loss = 1.91874 (* 1 = 1.91874 loss) I0409 23:38:55.885411 6396 sgd_solver.cpp:105] Iteration 4428, lr = 0.00415977 I0409 23:39:00.888667 6396 solver.cpp:218] Iteration 4440 (2.39851 iter/s, 5.00312s/12 iters), loss = 1.5552 I0409 23:39:00.888715 6396 solver.cpp:237] Train net output #0: loss = 1.5552 (* 1 = 1.5552 loss) I0409 23:39:00.888727 6396 sgd_solver.cpp:105] Iteration 4440, lr = 0.0041499 I0409 23:39:04.841382 6400 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:39:05.809969 6396 solver.cpp:218] Iteration 4452 (2.43849 iter/s, 4.92108s/12 iters), loss = 1.66639 I0409 23:39:05.810025 6396 solver.cpp:237] Train net output #0: loss = 1.66639 (* 1 = 1.66639 loss) I0409 23:39:05.810037 6396 sgd_solver.cpp:105] Iteration 4452, lr = 0.00414005 I0409 23:39:10.718080 6396 solver.cpp:218] Iteration 4464 (2.44503 iter/s, 4.90791s/12 iters), loss = 1.98536 I0409 23:39:10.718137 6396 solver.cpp:237] Train net output #0: loss = 1.98536 (* 1 = 1.98536 loss) I0409 23:39:10.718148 6396 sgd_solver.cpp:105] Iteration 4464, lr = 0.00413022 I0409 23:39:15.597328 6396 solver.cpp:218] Iteration 4476 (2.4595 iter/s, 4.87904s/12 iters), loss = 1.75028 I0409 23:39:15.597383 6396 solver.cpp:237] Train net output #0: loss = 1.75028 (* 1 = 1.75028 loss) I0409 23:39:15.597394 6396 sgd_solver.cpp:105] Iteration 4476, lr = 0.00412041 I0409 23:39:20.077693 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_4488.caffemodel I0409 23:39:20.804855 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_4488.solverstate I0409 23:39:21.361821 6396 solver.cpp:330] Iteration 4488, Testing net (#0) I0409 23:39:21.361845 6396 net.cpp:676] Ignoring source layer train-data I0409 23:39:23.921548 6401 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:39:25.692267 6396 solver.cpp:397] Test net output #0: accuracy = 0.431985 I0409 23:39:25.692304 6396 solver.cpp:397] Test net output #1: loss = 2.23635 (* 1 = 2.23635 loss) I0409 23:39:25.775730 6396 solver.cpp:218] Iteration 4488 (1.17901 iter/s, 10.1781s/12 iters), loss = 1.76075 I0409 23:39:25.775784 6396 solver.cpp:237] Train net output #0: loss = 1.76075 (* 1 = 1.76075 loss) I0409 23:39:25.775794 6396 sgd_solver.cpp:105] Iteration 4488, lr = 0.00411063 I0409 23:39:30.086803 6396 solver.cpp:218] Iteration 4500 (2.78365 iter/s, 4.31088s/12 iters), loss = 1.67328 I0409 23:39:30.086854 6396 solver.cpp:237] Train net output #0: loss = 1.67328 (* 1 = 1.67328 loss) I0409 23:39:30.086866 6396 sgd_solver.cpp:105] Iteration 4500, lr = 0.00410087 I0409 23:39:35.025619 6396 solver.cpp:218] Iteration 4512 (2.42983 iter/s, 4.93861s/12 iters), loss = 1.77123 I0409 23:39:35.025665 6396 solver.cpp:237] Train net output #0: loss = 1.77123 (* 1 = 1.77123 loss) I0409 23:39:35.025674 6396 sgd_solver.cpp:105] Iteration 4512, lr = 0.00409113 I0409 23:39:40.032374 6396 solver.cpp:218] Iteration 4524 (2.39686 iter/s, 5.00655s/12 iters), loss = 1.73755 I0409 23:39:40.032416 6396 solver.cpp:237] Train net output #0: loss = 1.73755 (* 1 = 1.73755 loss) I0409 23:39:40.032424 6396 sgd_solver.cpp:105] Iteration 4524, lr = 0.00408142 I0409 23:39:45.096984 6396 solver.cpp:218] Iteration 4536 (2.36948 iter/s, 5.06441s/12 iters), loss = 1.6224 I0409 23:39:45.097029 6396 solver.cpp:237] Train net output #0: loss = 1.6224 (* 1 = 1.6224 loss) I0409 23:39:45.097040 6396 sgd_solver.cpp:105] Iteration 4536, lr = 0.00407173 I0409 23:39:50.041045 6396 solver.cpp:218] Iteration 4548 (2.42725 iter/s, 4.94386s/12 iters), loss = 1.82898 I0409 23:39:50.041090 6396 solver.cpp:237] Train net output #0: loss = 1.82898 (* 1 = 1.82898 loss) I0409 23:39:50.041098 6396 sgd_solver.cpp:105] Iteration 4548, lr = 0.00406206 I0409 23:39:51.277822 6400 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:39:54.954428 6396 solver.cpp:218] Iteration 4560 (2.44241 iter/s, 4.91318s/12 iters), loss = 1.54523 I0409 23:39:54.954473 6396 solver.cpp:237] Train net output #0: loss = 1.54523 (* 1 = 1.54523 loss) I0409 23:39:54.954481 6396 sgd_solver.cpp:105] Iteration 4560, lr = 0.00405242 I0409 23:39:59.893108 6396 solver.cpp:218] Iteration 4572 (2.4299 iter/s, 4.93848s/12 iters), loss = 1.62023 I0409 23:39:59.893153 6396 solver.cpp:237] Train net output #0: loss = 1.62023 (* 1 = 1.62023 loss) I0409 23:39:59.893164 6396 sgd_solver.cpp:105] Iteration 4572, lr = 0.0040428 I0409 23:40:04.789886 6396 solver.cpp:218] Iteration 4584 (2.45069 iter/s, 4.89658s/12 iters), loss = 1.50383 I0409 23:40:04.789932 6396 solver.cpp:237] Train net output #0: loss = 1.50383 (* 1 = 1.50383 loss) I0409 23:40:04.789939 6396 sgd_solver.cpp:105] Iteration 4584, lr = 0.0040332 I0409 23:40:06.771414 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_4590.caffemodel I0409 23:40:09.112443 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_4590.solverstate I0409 23:40:09.833427 6396 solver.cpp:330] Iteration 4590, Testing net (#0) I0409 23:40:09.833447 6396 net.cpp:676] Ignoring source layer train-data I0409 23:40:12.534955 6401 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:40:14.351265 6396 solver.cpp:397] Test net output #0: accuracy = 0.463848 I0409 23:40:14.351295 6396 solver.cpp:397] Test net output #1: loss = 2.06363 (* 1 = 2.06363 loss) I0409 23:40:16.234881 6396 solver.cpp:218] Iteration 4596 (1.04853 iter/s, 11.4446s/12 iters), loss = 1.53011 I0409 23:40:16.234943 6396 solver.cpp:237] Train net output #0: loss = 1.53011 (* 1 = 1.53011 loss) I0409 23:40:16.234956 6396 sgd_solver.cpp:105] Iteration 4596, lr = 0.00402362 I0409 23:40:21.181660 6396 solver.cpp:218] Iteration 4608 (2.42593 iter/s, 4.94657s/12 iters), loss = 1.35859 I0409 23:40:21.181704 6396 solver.cpp:237] Train net output #0: loss = 1.35859 (* 1 = 1.35859 loss) I0409 23:40:21.181715 6396 sgd_solver.cpp:105] Iteration 4608, lr = 0.00401407 I0409 23:40:26.070713 6396 solver.cpp:218] Iteration 4620 (2.45456 iter/s, 4.88886s/12 iters), loss = 1.31442 I0409 23:40:26.070869 6396 solver.cpp:237] Train net output #0: loss = 1.31442 (* 1 = 1.31442 loss) I0409 23:40:26.070885 6396 sgd_solver.cpp:105] Iteration 4620, lr = 0.00400454 I0409 23:40:31.068348 6396 solver.cpp:218] Iteration 4632 (2.40128 iter/s, 4.99733s/12 iters), loss = 1.47417 I0409 23:40:31.068395 6396 solver.cpp:237] Train net output #0: loss = 1.47417 (* 1 = 1.47417 loss) I0409 23:40:31.068404 6396 sgd_solver.cpp:105] Iteration 4632, lr = 0.00399503 I0409 23:40:36.127948 6396 solver.cpp:218] Iteration 4644 (2.37182 iter/s, 5.0594s/12 iters), loss = 1.68549 I0409 23:40:36.127995 6396 solver.cpp:237] Train net output #0: loss = 1.68549 (* 1 = 1.68549 loss) I0409 23:40:36.128005 6396 sgd_solver.cpp:105] Iteration 4644, lr = 0.00398555 I0409 23:40:39.623687 6400 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:40:41.197274 6396 solver.cpp:218] Iteration 4656 (2.36727 iter/s, 5.06912s/12 iters), loss = 1.57728 I0409 23:40:41.197324 6396 solver.cpp:237] Train net output #0: loss = 1.57728 (* 1 = 1.57728 loss) I0409 23:40:41.197335 6396 sgd_solver.cpp:105] Iteration 4656, lr = 0.00397608 I0409 23:40:46.215263 6396 solver.cpp:218] Iteration 4668 (2.3915 iter/s, 5.01778s/12 iters), loss = 1.64207 I0409 23:40:46.215322 6396 solver.cpp:237] Train net output #0: loss = 1.64207 (* 1 = 1.64207 loss) I0409 23:40:46.215335 6396 sgd_solver.cpp:105] Iteration 4668, lr = 0.00396664 I0409 23:40:51.153918 6396 solver.cpp:218] Iteration 4680 (2.42992 iter/s, 4.93844s/12 iters), loss = 1.68641 I0409 23:40:51.153995 6396 solver.cpp:237] Train net output #0: loss = 1.68641 (* 1 = 1.68641 loss) I0409 23:40:51.154009 6396 sgd_solver.cpp:105] Iteration 4680, lr = 0.00395723 I0409 23:40:55.716061 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_4692.caffemodel I0409 23:40:56.130509 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_4692.solverstate I0409 23:40:56.432812 6396 solver.cpp:330] Iteration 4692, Testing net (#0) I0409 23:40:56.432837 6396 net.cpp:676] Ignoring source layer train-data I0409 23:40:58.938733 6401 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:41:00.802918 6396 solver.cpp:397] Test net output #0: accuracy = 0.443627 I0409 23:41:00.802966 6396 solver.cpp:397] Test net output #1: loss = 2.18758 (* 1 = 2.18758 loss) I0409 23:41:00.886438 6396 solver.cpp:218] Iteration 4692 (1.23303 iter/s, 9.73214s/12 iters), loss = 1.42562 I0409 23:41:00.886490 6396 solver.cpp:237] Train net output #0: loss = 1.42562 (* 1 = 1.42562 loss) I0409 23:41:00.886500 6396 sgd_solver.cpp:105] Iteration 4692, lr = 0.00394783 I0409 23:41:05.200695 6396 solver.cpp:218] Iteration 4704 (2.7816 iter/s, 4.31406s/12 iters), loss = 1.72659 I0409 23:41:05.200754 6396 solver.cpp:237] Train net output #0: loss = 1.72659 (* 1 = 1.72659 loss) I0409 23:41:05.200767 6396 sgd_solver.cpp:105] Iteration 4704, lr = 0.00393846 I0409 23:41:10.069742 6396 solver.cpp:218] Iteration 4716 (2.46466 iter/s, 4.86883s/12 iters), loss = 1.69849 I0409 23:41:10.069795 6396 solver.cpp:237] Train net output #0: loss = 1.69849 (* 1 = 1.69849 loss) I0409 23:41:10.069808 6396 sgd_solver.cpp:105] Iteration 4716, lr = 0.00392911 I0409 23:41:15.009306 6396 solver.cpp:218] Iteration 4728 (2.42946 iter/s, 4.93937s/12 iters), loss = 1.59687 I0409 23:41:15.009338 6396 solver.cpp:237] Train net output #0: loss = 1.59687 (* 1 = 1.59687 loss) I0409 23:41:15.009347 6396 sgd_solver.cpp:105] Iteration 4728, lr = 0.00391978 I0409 23:41:19.939828 6396 solver.cpp:218] Iteration 4740 (2.43391 iter/s, 4.93033s/12 iters), loss = 1.42146 I0409 23:41:19.939870 6396 solver.cpp:237] Train net output #0: loss = 1.42146 (* 1 = 1.42146 loss) I0409 23:41:19.939879 6396 sgd_solver.cpp:105] Iteration 4740, lr = 0.00391047 I0409 23:41:24.881698 6396 solver.cpp:218] Iteration 4752 (2.42833 iter/s, 4.94167s/12 iters), loss = 1.66931 I0409 23:41:24.881752 6396 solver.cpp:237] Train net output #0: loss = 1.66931 (* 1 = 1.66931 loss) I0409 23:41:24.881764 6396 sgd_solver.cpp:105] Iteration 4752, lr = 0.00390119 I0409 23:41:25.402940 6400 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:41:29.797991 6396 solver.cpp:218] Iteration 4764 (2.44097 iter/s, 4.91608s/12 iters), loss = 1.6675 I0409 23:41:29.798152 6396 solver.cpp:237] Train net output #0: loss = 1.6675 (* 1 = 1.6675 loss) I0409 23:41:29.798166 6396 sgd_solver.cpp:105] Iteration 4764, lr = 0.00389193 I0409 23:41:34.663902 6396 solver.cpp:218] Iteration 4776 (2.4663 iter/s, 4.8656s/12 iters), loss = 1.55913 I0409 23:41:34.663956 6396 solver.cpp:237] Train net output #0: loss = 1.55913 (* 1 = 1.55913 loss) I0409 23:41:34.663969 6396 sgd_solver.cpp:105] Iteration 4776, lr = 0.00388269 I0409 23:41:39.595480 6396 solver.cpp:218] Iteration 4788 (2.4334 iter/s, 4.93137s/12 iters), loss = 1.28777 I0409 23:41:39.595535 6396 solver.cpp:237] Train net output #0: loss = 1.28777 (* 1 = 1.28777 loss) I0409 23:41:39.595546 6396 sgd_solver.cpp:105] Iteration 4788, lr = 0.00387347 I0409 23:41:41.590755 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_4794.caffemodel I0409 23:41:42.069672 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_4794.solverstate I0409 23:41:42.383643 6396 solver.cpp:330] Iteration 4794, Testing net (#0) I0409 23:41:42.383661 6396 net.cpp:676] Ignoring source layer train-data I0409 23:41:44.944722 6401 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:41:46.838264 6396 solver.cpp:397] Test net output #0: accuracy = 0.452819 I0409 23:41:46.838299 6396 solver.cpp:397] Test net output #1: loss = 2.09064 (* 1 = 2.09064 loss) I0409 23:41:48.720752 6396 solver.cpp:218] Iteration 4800 (1.31508 iter/s, 9.12495s/12 iters), loss = 1.44423 I0409 23:41:48.720803 6396 solver.cpp:237] Train net output #0: loss = 1.44423 (* 1 = 1.44423 loss) I0409 23:41:48.720815 6396 sgd_solver.cpp:105] Iteration 4800, lr = 0.00386427 I0409 23:41:53.698823 6396 solver.cpp:218] Iteration 4812 (2.41067 iter/s, 4.97787s/12 iters), loss = 1.78568 I0409 23:41:53.698873 6396 solver.cpp:237] Train net output #0: loss = 1.78568 (* 1 = 1.78568 loss) I0409 23:41:53.698885 6396 sgd_solver.cpp:105] Iteration 4812, lr = 0.0038551 I0409 23:41:58.674353 6396 solver.cpp:218] Iteration 4824 (2.4119 iter/s, 4.97533s/12 iters), loss = 1.41975 I0409 23:41:58.674407 6396 solver.cpp:237] Train net output #0: loss = 1.41975 (* 1 = 1.41975 loss) I0409 23:41:58.674419 6396 sgd_solver.cpp:105] Iteration 4824, lr = 0.00384594 I0409 23:42:03.643152 6396 solver.cpp:218] Iteration 4836 (2.41517 iter/s, 4.96859s/12 iters), loss = 1.45016 I0409 23:42:03.643257 6396 solver.cpp:237] Train net output #0: loss = 1.45016 (* 1 = 1.45016 loss) I0409 23:42:03.643266 6396 sgd_solver.cpp:105] Iteration 4836, lr = 0.00383681 I0409 23:42:04.036335 6396 blocking_queue.cpp:49] Waiting for data I0409 23:42:08.697185 6396 solver.cpp:218] Iteration 4848 (2.37447 iter/s, 5.05377s/12 iters), loss = 1.7338 I0409 23:42:08.697237 6396 solver.cpp:237] Train net output #0: loss = 1.7338 (* 1 = 1.7338 loss) I0409 23:42:08.697249 6396 sgd_solver.cpp:105] Iteration 4848, lr = 0.0038277 I0409 23:42:11.335813 6400 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:42:13.595362 6396 solver.cpp:218] Iteration 4860 (2.44999 iter/s, 4.89797s/12 iters), loss = 1.31022 I0409 23:42:13.595418 6396 solver.cpp:237] Train net output #0: loss = 1.31022 (* 1 = 1.31022 loss) I0409 23:42:13.595430 6396 sgd_solver.cpp:105] Iteration 4860, lr = 0.00381862 I0409 23:42:18.477286 6396 solver.cpp:218] Iteration 4872 (2.45815 iter/s, 4.88172s/12 iters), loss = 1.44428 I0409 23:42:18.477349 6396 solver.cpp:237] Train net output #0: loss = 1.44428 (* 1 = 1.44428 loss) I0409 23:42:18.477361 6396 sgd_solver.cpp:105] Iteration 4872, lr = 0.00380955 I0409 23:42:23.345371 6396 solver.cpp:218] Iteration 4884 (2.46514 iter/s, 4.86787s/12 iters), loss = 1.58386 I0409 23:42:23.345427 6396 solver.cpp:237] Train net output #0: loss = 1.58386 (* 1 = 1.58386 loss) I0409 23:42:23.345438 6396 sgd_solver.cpp:105] Iteration 4884, lr = 0.0038005 I0409 23:42:27.754204 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_4896.caffemodel I0409 23:42:28.212955 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_4896.solverstate I0409 23:42:28.516018 6396 solver.cpp:330] Iteration 4896, Testing net (#0) I0409 23:42:28.516042 6396 net.cpp:676] Ignoring source layer train-data I0409 23:42:30.900669 6401 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:42:32.934656 6396 solver.cpp:397] Test net output #0: accuracy = 0.45098 I0409 23:42:32.934700 6396 solver.cpp:397] Test net output #1: loss = 2.0654 (* 1 = 2.0654 loss) I0409 23:42:33.017901 6396 solver.cpp:218] Iteration 4896 (1.24067 iter/s, 9.67219s/12 iters), loss = 1.38006 I0409 23:42:33.017951 6396 solver.cpp:237] Train net output #0: loss = 1.38006 (* 1 = 1.38006 loss) I0409 23:42:33.017979 6396 sgd_solver.cpp:105] Iteration 4896, lr = 0.00379148 I0409 23:42:37.368402 6396 solver.cpp:218] Iteration 4908 (2.75842 iter/s, 4.35031s/12 iters), loss = 1.37288 I0409 23:42:37.368503 6396 solver.cpp:237] Train net output #0: loss = 1.37288 (* 1 = 1.37288 loss) I0409 23:42:37.368513 6396 sgd_solver.cpp:105] Iteration 4908, lr = 0.00378248 I0409 23:42:42.252970 6396 solver.cpp:218] Iteration 4920 (2.45684 iter/s, 4.88432s/12 iters), loss = 1.28975 I0409 23:42:42.253006 6396 solver.cpp:237] Train net output #0: loss = 1.28975 (* 1 = 1.28975 loss) I0409 23:42:42.253015 6396 sgd_solver.cpp:105] Iteration 4920, lr = 0.0037735 I0409 23:42:47.607815 6396 solver.cpp:218] Iteration 4932 (2.24105 iter/s, 5.35464s/12 iters), loss = 1.46738 I0409 23:42:47.607856 6396 solver.cpp:237] Train net output #0: loss = 1.46738 (* 1 = 1.46738 loss) I0409 23:42:47.607864 6396 sgd_solver.cpp:105] Iteration 4932, lr = 0.00376454 I0409 23:42:52.722395 6396 solver.cpp:218] Iteration 4944 (2.34633 iter/s, 5.11438s/12 iters), loss = 1.33787 I0409 23:42:52.722451 6396 solver.cpp:237] Train net output #0: loss = 1.33787 (* 1 = 1.33787 loss) I0409 23:42:52.722465 6396 sgd_solver.cpp:105] Iteration 4944, lr = 0.0037556 I0409 23:42:57.392863 6400 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:42:57.589288 6396 solver.cpp:218] Iteration 4956 (2.46574 iter/s, 4.86668s/12 iters), loss = 1.51169 I0409 23:42:57.589339 6396 solver.cpp:237] Train net output #0: loss = 1.51169 (* 1 = 1.51169 loss) I0409 23:42:57.589350 6396 sgd_solver.cpp:105] Iteration 4956, lr = 0.00374669 I0409 23:43:02.575052 6396 solver.cpp:218] Iteration 4968 (2.40696 iter/s, 4.98555s/12 iters), loss = 1.63893 I0409 23:43:02.575103 6396 solver.cpp:237] Train net output #0: loss = 1.63893 (* 1 = 1.63893 loss) I0409 23:43:02.575114 6396 sgd_solver.cpp:105] Iteration 4968, lr = 0.00373779 I0409 23:43:07.564990 6396 solver.cpp:218] Iteration 4980 (2.40494 iter/s, 4.98973s/12 iters), loss = 1.35185 I0409 23:43:07.565172 6396 solver.cpp:237] Train net output #0: loss = 1.35185 (* 1 = 1.35185 loss) I0409 23:43:07.565186 6396 sgd_solver.cpp:105] Iteration 4980, lr = 0.00372892 I0409 23:43:12.579942 6396 solver.cpp:218] Iteration 4992 (2.393 iter/s, 5.01462s/12 iters), loss = 1.57119 I0409 23:43:12.579993 6396 solver.cpp:237] Train net output #0: loss = 1.57119 (* 1 = 1.57119 loss) I0409 23:43:12.580005 6396 sgd_solver.cpp:105] Iteration 4992, lr = 0.00372006 I0409 23:43:14.616120 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_4998.caffemodel I0409 23:43:16.171270 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_4998.solverstate I0409 23:43:19.097430 6396 solver.cpp:330] Iteration 4998, Testing net (#0) I0409 23:43:19.097457 6396 net.cpp:676] Ignoring source layer train-data I0409 23:43:21.507513 6401 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:43:23.549865 6396 solver.cpp:397] Test net output #0: accuracy = 0.449142 I0409 23:43:23.549912 6396 solver.cpp:397] Test net output #1: loss = 2.1465 (* 1 = 2.1465 loss) I0409 23:43:25.414203 6396 solver.cpp:218] Iteration 5004 (0.935028 iter/s, 12.8338s/12 iters), loss = 1.34208 I0409 23:43:25.414252 6396 solver.cpp:237] Train net output #0: loss = 1.34208 (* 1 = 1.34208 loss) I0409 23:43:25.414261 6396 sgd_solver.cpp:105] Iteration 5004, lr = 0.00371123 I0409 23:43:30.357372 6396 solver.cpp:218] Iteration 5016 (2.42769 iter/s, 4.94297s/12 iters), loss = 1.54273 I0409 23:43:30.357419 6396 solver.cpp:237] Train net output #0: loss = 1.54273 (* 1 = 1.54273 loss) I0409 23:43:30.357430 6396 sgd_solver.cpp:105] Iteration 5016, lr = 0.00370242 I0409 23:43:35.259367 6396 solver.cpp:218] Iteration 5028 (2.44808 iter/s, 4.9018s/12 iters), loss = 1.33465 I0409 23:43:35.259408 6396 solver.cpp:237] Train net output #0: loss = 1.33465 (* 1 = 1.33465 loss) I0409 23:43:35.259418 6396 sgd_solver.cpp:105] Iteration 5028, lr = 0.00369363 I0409 23:43:40.732924 6396 solver.cpp:218] Iteration 5040 (2.19244 iter/s, 5.47335s/12 iters), loss = 1.53542 I0409 23:43:40.733029 6396 solver.cpp:237] Train net output #0: loss = 1.53542 (* 1 = 1.53542 loss) I0409 23:43:40.733040 6396 sgd_solver.cpp:105] Iteration 5040, lr = 0.00368486 I0409 23:43:45.752785 6396 solver.cpp:218] Iteration 5052 (2.39063 iter/s, 5.0196s/12 iters), loss = 1.33773 I0409 23:43:45.752830 6396 solver.cpp:237] Train net output #0: loss = 1.33773 (* 1 = 1.33773 loss) I0409 23:43:45.752840 6396 sgd_solver.cpp:105] Iteration 5052, lr = 0.00367611 I0409 23:43:47.668295 6400 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:43:50.727340 6396 solver.cpp:218] Iteration 5064 (2.41237 iter/s, 4.97435s/12 iters), loss = 1.32446 I0409 23:43:50.727393 6396 solver.cpp:237] Train net output #0: loss = 1.32446 (* 1 = 1.32446 loss) I0409 23:43:50.727406 6396 sgd_solver.cpp:105] Iteration 5064, lr = 0.00366738 I0409 23:43:55.576844 6396 solver.cpp:218] Iteration 5076 (2.47458 iter/s, 4.8493s/12 iters), loss = 1.52258 I0409 23:43:55.576886 6396 solver.cpp:237] Train net output #0: loss = 1.52258 (* 1 = 1.52258 loss) I0409 23:43:55.576896 6396 sgd_solver.cpp:105] Iteration 5076, lr = 0.00365868 I0409 23:44:00.483932 6396 solver.cpp:218] Iteration 5088 (2.44554 iter/s, 4.90689s/12 iters), loss = 1.21961 I0409 23:44:00.483987 6396 solver.cpp:237] Train net output #0: loss = 1.21961 (* 1 = 1.21961 loss) I0409 23:44:00.484000 6396 sgd_solver.cpp:105] Iteration 5088, lr = 0.00364999 I0409 23:44:04.981174 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_5100.caffemodel I0409 23:44:05.427546 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_5100.solverstate I0409 23:44:05.758401 6396 solver.cpp:330] Iteration 5100, Testing net (#0) I0409 23:44:05.758431 6396 net.cpp:676] Ignoring source layer train-data I0409 23:44:08.302244 6401 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:44:10.338086 6396 solver.cpp:397] Test net output #0: accuracy = 0.476716 I0409 23:44:10.338115 6396 solver.cpp:397] Test net output #1: loss = 2.08713 (* 1 = 2.08713 loss) I0409 23:44:10.421272 6396 solver.cpp:218] Iteration 5100 (1.20761 iter/s, 9.937s/12 iters), loss = 1.40138 I0409 23:44:10.421311 6396 solver.cpp:237] Train net output #0: loss = 1.40138 (* 1 = 1.40138 loss) I0409 23:44:10.421320 6396 sgd_solver.cpp:105] Iteration 5100, lr = 0.00364132 I0409 23:44:14.656049 6396 solver.cpp:218] Iteration 5112 (2.8338 iter/s, 4.2346s/12 iters), loss = 1.21988 I0409 23:44:14.656150 6396 solver.cpp:237] Train net output #0: loss = 1.21988 (* 1 = 1.21988 loss) I0409 23:44:14.656160 6396 sgd_solver.cpp:105] Iteration 5112, lr = 0.00363268 I0409 23:44:19.583092 6396 solver.cpp:218] Iteration 5124 (2.43567 iter/s, 4.92678s/12 iters), loss = 1.28602 I0409 23:44:19.583145 6396 solver.cpp:237] Train net output #0: loss = 1.28602 (* 1 = 1.28602 loss) I0409 23:44:19.583156 6396 sgd_solver.cpp:105] Iteration 5124, lr = 0.00362405 I0409 23:44:24.470885 6396 solver.cpp:218] Iteration 5136 (2.4552 iter/s, 4.88759s/12 iters), loss = 1.20332 I0409 23:44:24.470937 6396 solver.cpp:237] Train net output #0: loss = 1.20332 (* 1 = 1.20332 loss) I0409 23:44:24.470949 6396 sgd_solver.cpp:105] Iteration 5136, lr = 0.00361545 I0409 23:44:29.477636 6396 solver.cpp:218] Iteration 5148 (2.39686 iter/s, 5.00655s/12 iters), loss = 1.56055 I0409 23:44:29.477676 6396 solver.cpp:237] Train net output #0: loss = 1.56055 (* 1 = 1.56055 loss) I0409 23:44:29.477684 6396 sgd_solver.cpp:105] Iteration 5148, lr = 0.00360687 I0409 23:44:33.552160 6400 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:44:34.488935 6396 solver.cpp:218] Iteration 5160 (2.39468 iter/s, 5.0111s/12 iters), loss = 1.29191 I0409 23:44:34.488982 6396 solver.cpp:237] Train net output #0: loss = 1.29191 (* 1 = 1.29191 loss) I0409 23:44:34.488993 6396 sgd_solver.cpp:105] Iteration 5160, lr = 0.0035983 I0409 23:44:39.463025 6396 solver.cpp:218] Iteration 5172 (2.4126 iter/s, 4.97389s/12 iters), loss = 1.26019 I0409 23:44:39.463078 6396 solver.cpp:237] Train net output #0: loss = 1.26019 (* 1 = 1.26019 loss) I0409 23:44:39.463089 6396 sgd_solver.cpp:105] Iteration 5172, lr = 0.00358976 I0409 23:44:44.385620 6396 solver.cpp:218] Iteration 5184 (2.43784 iter/s, 4.92238s/12 iters), loss = 1.35202 I0409 23:44:44.385675 6396 solver.cpp:237] Train net output #0: loss = 1.35202 (* 1 = 1.35202 loss) I0409 23:44:44.385687 6396 sgd_solver.cpp:105] Iteration 5184, lr = 0.00358124 I0409 23:44:49.452234 6396 solver.cpp:218] Iteration 5196 (2.36854 iter/s, 5.0664s/12 iters), loss = 1.21079 I0409 23:44:49.452304 6396 solver.cpp:237] Train net output #0: loss = 1.21079 (* 1 = 1.21079 loss) I0409 23:44:49.452314 6396 sgd_solver.cpp:105] Iteration 5196, lr = 0.00357273 I0409 23:44:51.447501 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_5202.caffemodel I0409 23:44:52.300827 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_5202.solverstate I0409 23:44:53.107239 6396 solver.cpp:330] Iteration 5202, Testing net (#0) I0409 23:44:53.107270 6396 net.cpp:676] Ignoring source layer train-data I0409 23:44:55.518594 6401 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:44:57.566684 6396 solver.cpp:397] Test net output #0: accuracy = 0.481618 I0409 23:44:57.566735 6396 solver.cpp:397] Test net output #1: loss = 2.10685 (* 1 = 2.10685 loss) I0409 23:44:59.327812 6396 solver.cpp:218] Iteration 5208 (1.21516 iter/s, 9.87522s/12 iters), loss = 1.28189 I0409 23:44:59.327857 6396 solver.cpp:237] Train net output #0: loss = 1.28189 (* 1 = 1.28189 loss) I0409 23:44:59.327867 6396 sgd_solver.cpp:105] Iteration 5208, lr = 0.00356425 I0409 23:45:04.574476 6396 solver.cpp:218] Iteration 5220 (2.28726 iter/s, 5.24645s/12 iters), loss = 1.31913 I0409 23:45:04.574530 6396 solver.cpp:237] Train net output #0: loss = 1.31913 (* 1 = 1.31913 loss) I0409 23:45:04.574542 6396 sgd_solver.cpp:105] Iteration 5220, lr = 0.00355579 I0409 23:45:09.535621 6396 solver.cpp:218] Iteration 5232 (2.4189 iter/s, 4.96094s/12 iters), loss = 1.04105 I0409 23:45:09.535678 6396 solver.cpp:237] Train net output #0: loss = 1.04105 (* 1 = 1.04105 loss) I0409 23:45:09.535691 6396 sgd_solver.cpp:105] Iteration 5232, lr = 0.00354735 I0409 23:45:14.467449 6396 solver.cpp:218] Iteration 5244 (2.43328 iter/s, 4.93162s/12 iters), loss = 1.25135 I0409 23:45:14.467490 6396 solver.cpp:237] Train net output #0: loss = 1.25135 (* 1 = 1.25135 loss) I0409 23:45:14.467500 6396 sgd_solver.cpp:105] Iteration 5244, lr = 0.00353892 I0409 23:45:19.696308 6396 solver.cpp:218] Iteration 5256 (2.29504 iter/s, 5.22866s/12 iters), loss = 1.07518 I0409 23:45:19.696451 6396 solver.cpp:237] Train net output #0: loss = 1.07518 (* 1 = 1.07518 loss) I0409 23:45:19.696465 6396 sgd_solver.cpp:105] Iteration 5256, lr = 0.00353052 I0409 23:45:21.092583 6400 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:45:24.878989 6396 solver.cpp:218] Iteration 5268 (2.31554 iter/s, 5.18238s/12 iters), loss = 1.29942 I0409 23:45:24.879045 6396 solver.cpp:237] Train net output #0: loss = 1.29942 (* 1 = 1.29942 loss) I0409 23:45:24.879056 6396 sgd_solver.cpp:105] Iteration 5268, lr = 0.00352214 I0409 23:45:29.824676 6396 solver.cpp:218] Iteration 5280 (2.42646 iter/s, 4.94548s/12 iters), loss = 1.27317 I0409 23:45:29.824718 6396 solver.cpp:237] Train net output #0: loss = 1.27317 (* 1 = 1.27317 loss) I0409 23:45:29.824728 6396 sgd_solver.cpp:105] Iteration 5280, lr = 0.00351378 I0409 23:45:34.726204 6396 solver.cpp:218] Iteration 5292 (2.44831 iter/s, 4.90134s/12 iters), loss = 1.16052 I0409 23:45:34.726246 6396 solver.cpp:237] Train net output #0: loss = 1.16052 (* 1 = 1.16052 loss) I0409 23:45:34.726254 6396 sgd_solver.cpp:105] Iteration 5292, lr = 0.00350544 I0409 23:45:39.175698 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_5304.caffemodel I0409 23:45:39.619940 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_5304.solverstate I0409 23:45:39.936585 6396 solver.cpp:330] Iteration 5304, Testing net (#0) I0409 23:45:39.936602 6396 net.cpp:676] Ignoring source layer train-data I0409 23:45:42.501847 6401 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:45:44.600886 6396 solver.cpp:397] Test net output #0: accuracy = 0.469363 I0409 23:45:44.600934 6396 solver.cpp:397] Test net output #1: loss = 2.06457 (* 1 = 2.06457 loss) I0409 23:45:44.684166 6396 solver.cpp:218] Iteration 5304 (1.20511 iter/s, 9.95762s/12 iters), loss = 1.23886 I0409 23:45:44.684218 6396 solver.cpp:237] Train net output #0: loss = 1.23886 (* 1 = 1.23886 loss) I0409 23:45:44.684229 6396 sgd_solver.cpp:105] Iteration 5304, lr = 0.00349711 I0409 23:45:49.103423 6396 solver.cpp:218] Iteration 5316 (2.71551 iter/s, 4.41906s/12 iters), loss = 1.07565 I0409 23:45:49.103469 6396 solver.cpp:237] Train net output #0: loss = 1.07565 (* 1 = 1.07565 loss) I0409 23:45:49.103480 6396 sgd_solver.cpp:105] Iteration 5316, lr = 0.00348881 I0409 23:45:53.974779 6396 solver.cpp:218] Iteration 5328 (2.46348 iter/s, 4.87116s/12 iters), loss = 1.12109 I0409 23:45:53.974854 6396 solver.cpp:237] Train net output #0: loss = 1.12109 (* 1 = 1.12109 loss) I0409 23:45:53.974864 6396 sgd_solver.cpp:105] Iteration 5328, lr = 0.00348053 I0409 23:45:59.008924 6396 solver.cpp:218] Iteration 5340 (2.38383 iter/s, 5.03391s/12 iters), loss = 1.23459 I0409 23:45:59.008981 6396 solver.cpp:237] Train net output #0: loss = 1.23459 (* 1 = 1.23459 loss) I0409 23:45:59.008992 6396 sgd_solver.cpp:105] Iteration 5340, lr = 0.00347226 I0409 23:46:03.996592 6396 solver.cpp:218] Iteration 5352 (2.40604 iter/s, 4.98745s/12 iters), loss = 1.26893 I0409 23:46:03.996634 6396 solver.cpp:237] Train net output #0: loss = 1.26893 (* 1 = 1.26893 loss) I0409 23:46:03.996644 6396 sgd_solver.cpp:105] Iteration 5352, lr = 0.00346402 I0409 23:46:07.326485 6400 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:46:08.877097 6396 solver.cpp:218] Iteration 5364 (2.45886 iter/s, 4.88031s/12 iters), loss = 1.20722 I0409 23:46:08.877154 6396 solver.cpp:237] Train net output #0: loss = 1.20722 (* 1 = 1.20722 loss) I0409 23:46:08.877166 6396 sgd_solver.cpp:105] Iteration 5364, lr = 0.0034558 I0409 23:46:13.848398 6396 solver.cpp:218] Iteration 5376 (2.41396 iter/s, 4.97109s/12 iters), loss = 1.32018 I0409 23:46:13.848454 6396 solver.cpp:237] Train net output #0: loss = 1.32018 (* 1 = 1.32018 loss) I0409 23:46:13.848465 6396 sgd_solver.cpp:105] Iteration 5376, lr = 0.00344759 I0409 23:46:18.771906 6396 solver.cpp:218] Iteration 5388 (2.43739 iter/s, 4.9233s/12 iters), loss = 1.36778 I0409 23:46:18.771965 6396 solver.cpp:237] Train net output #0: loss = 1.36778 (* 1 = 1.36778 loss) I0409 23:46:18.771976 6396 sgd_solver.cpp:105] Iteration 5388, lr = 0.00343941 I0409 23:46:23.697522 6396 solver.cpp:218] Iteration 5400 (2.43635 iter/s, 4.9254s/12 iters), loss = 1.30054 I0409 23:46:23.697580 6396 solver.cpp:237] Train net output #0: loss = 1.30054 (* 1 = 1.30054 loss) I0409 23:46:23.697593 6396 sgd_solver.cpp:105] Iteration 5400, lr = 0.00343124 I0409 23:46:25.657006 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_5406.caffemodel I0409 23:46:26.107805 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_5406.solverstate I0409 23:46:26.424870 6396 solver.cpp:330] Iteration 5406, Testing net (#0) I0409 23:46:26.424890 6396 net.cpp:676] Ignoring source layer train-data I0409 23:46:28.684685 6401 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:46:30.814730 6396 solver.cpp:397] Test net output #0: accuracy = 0.491422 I0409 23:46:30.814779 6396 solver.cpp:397] Test net output #1: loss = 2.02388 (* 1 = 2.02388 loss) I0409 23:46:32.611711 6396 solver.cpp:218] Iteration 5412 (1.34622 iter/s, 8.91386s/12 iters), loss = 1.37871 I0409 23:46:32.611764 6396 solver.cpp:237] Train net output #0: loss = 1.37871 (* 1 = 1.37871 loss) I0409 23:46:32.611775 6396 sgd_solver.cpp:105] Iteration 5412, lr = 0.00342309 I0409 23:46:37.597203 6396 solver.cpp:218] Iteration 5424 (2.40708 iter/s, 4.98529s/12 iters), loss = 1.28707 I0409 23:46:37.597249 6396 solver.cpp:237] Train net output #0: loss = 1.28707 (* 1 = 1.28707 loss) I0409 23:46:37.597259 6396 sgd_solver.cpp:105] Iteration 5424, lr = 0.00341497 I0409 23:46:42.512411 6396 solver.cpp:218] Iteration 5436 (2.4415 iter/s, 4.91501s/12 iters), loss = 1.14722 I0409 23:46:42.512470 6396 solver.cpp:237] Train net output #0: loss = 1.14722 (* 1 = 1.14722 loss) I0409 23:46:42.512482 6396 sgd_solver.cpp:105] Iteration 5436, lr = 0.00340686 I0409 23:46:47.459014 6396 solver.cpp:218] Iteration 5448 (2.42601 iter/s, 4.94639s/12 iters), loss = 0.98418 I0409 23:46:47.459064 6396 solver.cpp:237] Train net output #0: loss = 0.98418 (* 1 = 0.98418 loss) I0409 23:46:47.459075 6396 sgd_solver.cpp:105] Iteration 5448, lr = 0.00339877 I0409 23:46:52.345592 6396 solver.cpp:218] Iteration 5460 (2.45581 iter/s, 4.88637s/12 iters), loss = 1.20199 I0409 23:46:52.345644 6396 solver.cpp:237] Train net output #0: loss = 1.20199 (* 1 = 1.20199 loss) I0409 23:46:52.345657 6396 sgd_solver.cpp:105] Iteration 5460, lr = 0.0033907 I0409 23:46:52.916646 6400 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:46:57.399863 6396 solver.cpp:218] Iteration 5472 (2.37433 iter/s, 5.05406s/12 iters), loss = 1.22487 I0409 23:46:57.399982 6396 solver.cpp:237] Train net output #0: loss = 1.22487 (* 1 = 1.22487 loss) I0409 23:46:57.399996 6396 sgd_solver.cpp:105] Iteration 5472, lr = 0.00338265 I0409 23:47:02.351845 6396 solver.cpp:218] Iteration 5484 (2.42341 iter/s, 4.95171s/12 iters), loss = 1.11418 I0409 23:47:02.351903 6396 solver.cpp:237] Train net output #0: loss = 1.11418 (* 1 = 1.11418 loss) I0409 23:47:02.351917 6396 sgd_solver.cpp:105] Iteration 5484, lr = 0.00337462 I0409 23:47:07.425994 6396 solver.cpp:218] Iteration 5496 (2.36503 iter/s, 5.07394s/12 iters), loss = 1.14154 I0409 23:47:07.426044 6396 solver.cpp:237] Train net output #0: loss = 1.14154 (* 1 = 1.14154 loss) I0409 23:47:07.426057 6396 sgd_solver.cpp:105] Iteration 5496, lr = 0.00336661 I0409 23:47:11.954000 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_5508.caffemodel I0409 23:47:13.431457 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_5508.solverstate I0409 23:47:14.557116 6396 solver.cpp:330] Iteration 5508, Testing net (#0) I0409 23:47:14.557140 6396 net.cpp:676] Ignoring source layer train-data I0409 23:47:16.841966 6401 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:47:19.031122 6396 solver.cpp:397] Test net output #0: accuracy = 0.496324 I0409 23:47:19.031162 6396 solver.cpp:397] Test net output #1: loss = 2.02636 (* 1 = 2.02636 loss) I0409 23:47:19.119766 6396 solver.cpp:218] Iteration 5508 (1.02622 iter/s, 11.6934s/12 iters), loss = 1.35926 I0409 23:47:19.119812 6396 solver.cpp:237] Train net output #0: loss = 1.35926 (* 1 = 1.35926 loss) I0409 23:47:19.119822 6396 sgd_solver.cpp:105] Iteration 5508, lr = 0.00335861 I0409 23:47:23.412274 6396 solver.cpp:218] Iteration 5520 (2.79569 iter/s, 4.29232s/12 iters), loss = 1.01343 I0409 23:47:23.412330 6396 solver.cpp:237] Train net output #0: loss = 1.01343 (* 1 = 1.01343 loss) I0409 23:47:23.412341 6396 sgd_solver.cpp:105] Iteration 5520, lr = 0.00335064 I0409 23:47:24.192956 6396 blocking_queue.cpp:49] Waiting for data I0409 23:47:28.387045 6396 solver.cpp:218] Iteration 5532 (2.41227 iter/s, 4.97456s/12 iters), loss = 1.29008 I0409 23:47:28.387166 6396 solver.cpp:237] Train net output #0: loss = 1.29008 (* 1 = 1.29008 loss) I0409 23:47:28.387177 6396 sgd_solver.cpp:105] Iteration 5532, lr = 0.00334268 I0409 23:47:33.422134 6396 solver.cpp:218] Iteration 5544 (2.38341 iter/s, 5.03481s/12 iters), loss = 1.06908 I0409 23:47:33.422186 6396 solver.cpp:237] Train net output #0: loss = 1.06908 (* 1 = 1.06908 loss) I0409 23:47:33.422197 6396 sgd_solver.cpp:105] Iteration 5544, lr = 0.00333475 I0409 23:47:38.452695 6396 solver.cpp:218] Iteration 5556 (2.38552 iter/s, 5.03035s/12 iters), loss = 1.23074 I0409 23:47:38.452740 6396 solver.cpp:237] Train net output #0: loss = 1.23074 (* 1 = 1.23074 loss) I0409 23:47:38.452749 6396 sgd_solver.cpp:105] Iteration 5556, lr = 0.00332683 I0409 23:47:41.079183 6400 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:47:43.369825 6396 solver.cpp:218] Iteration 5568 (2.44055 iter/s, 4.91693s/12 iters), loss = 1.16131 I0409 23:47:43.369875 6396 solver.cpp:237] Train net output #0: loss = 1.16131 (* 1 = 1.16131 loss) I0409 23:47:43.369889 6396 sgd_solver.cpp:105] Iteration 5568, lr = 0.00331893 I0409 23:47:48.279623 6396 solver.cpp:218] Iteration 5580 (2.44419 iter/s, 4.90959s/12 iters), loss = 1.21144 I0409 23:47:48.279673 6396 solver.cpp:237] Train net output #0: loss = 1.21144 (* 1 = 1.21144 loss) I0409 23:47:48.279686 6396 sgd_solver.cpp:105] Iteration 5580, lr = 0.00331105 I0409 23:47:53.315661 6396 solver.cpp:218] Iteration 5592 (2.38292 iter/s, 5.03584s/12 iters), loss = 1.1609 I0409 23:47:53.315711 6396 solver.cpp:237] Train net output #0: loss = 1.1609 (* 1 = 1.1609 loss) I0409 23:47:53.315719 6396 sgd_solver.cpp:105] Iteration 5592, lr = 0.00330319 I0409 23:47:58.428105 6396 solver.cpp:218] Iteration 5604 (2.34731 iter/s, 5.11223s/12 iters), loss = 0.886791 I0409 23:47:58.428217 6396 solver.cpp:237] Train net output #0: loss = 0.886791 (* 1 = 0.886791 loss) I0409 23:47:58.428229 6396 sgd_solver.cpp:105] Iteration 5604, lr = 0.00329535 I0409 23:48:00.437288 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_5610.caffemodel I0409 23:48:00.860831 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_5610.solverstate I0409 23:48:01.158358 6396 solver.cpp:330] Iteration 5610, Testing net (#0) I0409 23:48:01.158378 6396 net.cpp:676] Ignoring source layer train-data I0409 23:48:03.341715 6401 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:48:05.783078 6396 solver.cpp:397] Test net output #0: accuracy = 0.498774 I0409 23:48:05.783125 6396 solver.cpp:397] Test net output #1: loss = 2.09805 (* 1 = 2.09805 loss) I0409 23:48:07.664916 6396 solver.cpp:218] Iteration 5616 (1.2992 iter/s, 9.23643s/12 iters), loss = 1.28246 I0409 23:48:07.664964 6396 solver.cpp:237] Train net output #0: loss = 1.28246 (* 1 = 1.28246 loss) I0409 23:48:07.664975 6396 sgd_solver.cpp:105] Iteration 5616, lr = 0.00328752 I0409 23:48:12.650821 6396 solver.cpp:218] Iteration 5628 (2.40688 iter/s, 4.9857s/12 iters), loss = 1.07345 I0409 23:48:12.650869 6396 solver.cpp:237] Train net output #0: loss = 1.07345 (* 1 = 1.07345 loss) I0409 23:48:12.650882 6396 sgd_solver.cpp:105] Iteration 5628, lr = 0.00327972 I0409 23:48:17.554878 6396 solver.cpp:218] Iteration 5640 (2.44706 iter/s, 4.90385s/12 iters), loss = 1.0793 I0409 23:48:17.554935 6396 solver.cpp:237] Train net output #0: loss = 1.0793 (* 1 = 1.0793 loss) I0409 23:48:17.554947 6396 sgd_solver.cpp:105] Iteration 5640, lr = 0.00327193 I0409 23:48:22.588385 6396 solver.cpp:218] Iteration 5652 (2.38413 iter/s, 5.03329s/12 iters), loss = 1.1851 I0409 23:48:22.588443 6396 solver.cpp:237] Train net output #0: loss = 1.1851 (* 1 = 1.1851 loss) I0409 23:48:22.588454 6396 sgd_solver.cpp:105] Iteration 5652, lr = 0.00326416 I0409 23:48:27.389619 6400 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:48:27.551592 6396 solver.cpp:218] Iteration 5664 (2.41789 iter/s, 4.963s/12 iters), loss = 1.03357 I0409 23:48:27.551628 6396 solver.cpp:237] Train net output #0: loss = 1.03357 (* 1 = 1.03357 loss) I0409 23:48:27.551636 6396 sgd_solver.cpp:105] Iteration 5664, lr = 0.00325641 I0409 23:48:32.482208 6396 solver.cpp:218] Iteration 5676 (2.43387 iter/s, 4.93042s/12 iters), loss = 0.994841 I0409 23:48:32.482375 6396 solver.cpp:237] Train net output #0: loss = 0.994841 (* 1 = 0.994841 loss) I0409 23:48:32.482390 6396 sgd_solver.cpp:105] Iteration 5676, lr = 0.00324868 I0409 23:48:37.394071 6396 solver.cpp:218] Iteration 5688 (2.44322 iter/s, 4.91154s/12 iters), loss = 0.784141 I0409 23:48:37.394121 6396 solver.cpp:237] Train net output #0: loss = 0.784141 (* 1 = 0.784141 loss) I0409 23:48:37.394131 6396 sgd_solver.cpp:105] Iteration 5688, lr = 0.00324097 I0409 23:48:42.299687 6396 solver.cpp:218] Iteration 5700 (2.44628 iter/s, 4.90541s/12 iters), loss = 1.12322 I0409 23:48:42.299744 6396 solver.cpp:237] Train net output #0: loss = 1.12322 (* 1 = 1.12322 loss) I0409 23:48:42.299757 6396 sgd_solver.cpp:105] Iteration 5700, lr = 0.00323328 I0409 23:48:46.773326 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_5712.caffemodel I0409 23:48:47.196101 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_5712.solverstate I0409 23:48:47.499562 6396 solver.cpp:330] Iteration 5712, Testing net (#0) I0409 23:48:47.499581 6396 net.cpp:676] Ignoring source layer train-data I0409 23:48:49.726727 6401 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:48:51.985849 6396 solver.cpp:397] Test net output #0: accuracy = 0.51348 I0409 23:48:51.985893 6396 solver.cpp:397] Test net output #1: loss = 1.95575 (* 1 = 1.95575 loss) I0409 23:48:52.069000 6396 solver.cpp:218] Iteration 5712 (1.22838 iter/s, 9.76897s/12 iters), loss = 1.13294 I0409 23:48:52.069046 6396 solver.cpp:237] Train net output #0: loss = 1.13294 (* 1 = 1.13294 loss) I0409 23:48:52.069058 6396 sgd_solver.cpp:105] Iteration 5712, lr = 0.0032256 I0409 23:48:56.283720 6396 solver.cpp:218] Iteration 5724 (2.84729 iter/s, 4.21454s/12 iters), loss = 1.25827 I0409 23:48:56.283772 6396 solver.cpp:237] Train net output #0: loss = 1.25827 (* 1 = 1.25827 loss) I0409 23:48:56.283785 6396 sgd_solver.cpp:105] Iteration 5724, lr = 0.00321794 I0409 23:49:01.385385 6396 solver.cpp:218] Iteration 5736 (2.35227 iter/s, 5.10145s/12 iters), loss = 0.772492 I0409 23:49:01.385437 6396 solver.cpp:237] Train net output #0: loss = 0.772492 (* 1 = 0.772492 loss) I0409 23:49:01.385448 6396 sgd_solver.cpp:105] Iteration 5736, lr = 0.0032103 I0409 23:49:06.243407 6396 solver.cpp:218] Iteration 5748 (2.47025 iter/s, 4.85782s/12 iters), loss = 0.972716 I0409 23:49:06.243566 6396 solver.cpp:237] Train net output #0: loss = 0.972716 (* 1 = 0.972716 loss) I0409 23:49:06.243582 6396 sgd_solver.cpp:105] Iteration 5748, lr = 0.00320268 I0409 23:49:11.193011 6396 solver.cpp:218] Iteration 5760 (2.42459 iter/s, 4.94929s/12 iters), loss = 0.890603 I0409 23:49:11.193070 6396 solver.cpp:237] Train net output #0: loss = 0.890603 (* 1 = 0.890603 loss) I0409 23:49:11.193081 6396 sgd_solver.cpp:105] Iteration 5760, lr = 0.00319508 I0409 23:49:13.158334 6400 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:49:16.112597 6396 solver.cpp:218] Iteration 5772 (2.43934 iter/s, 4.91937s/12 iters), loss = 1.27382 I0409 23:49:16.112648 6396 solver.cpp:237] Train net output #0: loss = 1.27382 (* 1 = 1.27382 loss) I0409 23:49:16.112658 6396 sgd_solver.cpp:105] Iteration 5772, lr = 0.00318749 I0409 23:49:21.039188 6396 solver.cpp:218] Iteration 5784 (2.43587 iter/s, 4.92638s/12 iters), loss = 1.22907 I0409 23:49:21.039239 6396 solver.cpp:237] Train net output #0: loss = 1.22907 (* 1 = 1.22907 loss) I0409 23:49:21.039250 6396 sgd_solver.cpp:105] Iteration 5784, lr = 0.00317992 I0409 23:49:25.941267 6396 solver.cpp:218] Iteration 5796 (2.44804 iter/s, 4.90187s/12 iters), loss = 1.02648 I0409 23:49:25.941313 6396 solver.cpp:237] Train net output #0: loss = 1.02648 (* 1 = 1.02648 loss) I0409 23:49:25.941321 6396 sgd_solver.cpp:105] Iteration 5796, lr = 0.00317237 I0409 23:49:30.853169 6396 solver.cpp:218] Iteration 5808 (2.44314 iter/s, 4.91171s/12 iters), loss = 1.10264 I0409 23:49:30.853212 6396 solver.cpp:237] Train net output #0: loss = 1.10264 (* 1 = 1.10264 loss) I0409 23:49:30.853222 6396 sgd_solver.cpp:105] Iteration 5808, lr = 0.00316484 I0409 23:49:32.843935 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_5814.caffemodel I0409 23:49:33.818534 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_5814.solverstate I0409 23:49:34.947930 6396 solver.cpp:330] Iteration 5814, Testing net (#0) I0409 23:49:34.947955 6396 net.cpp:676] Ignoring source layer train-data I0409 23:49:37.093533 6401 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:49:39.379472 6396 solver.cpp:397] Test net output #0: accuracy = 0.51777 I0409 23:49:39.379520 6396 solver.cpp:397] Test net output #1: loss = 1.99054 (* 1 = 1.99054 loss) I0409 23:49:41.232180 6396 solver.cpp:218] Iteration 5820 (1.15622 iter/s, 10.3787s/12 iters), loss = 0.974285 I0409 23:49:41.232234 6396 solver.cpp:237] Train net output #0: loss = 0.974285 (* 1 = 0.974285 loss) I0409 23:49:41.232244 6396 sgd_solver.cpp:105] Iteration 5820, lr = 0.00315733 I0409 23:49:46.160537 6396 solver.cpp:218] Iteration 5832 (2.43499 iter/s, 4.92815s/12 iters), loss = 1.21351 I0409 23:49:46.160591 6396 solver.cpp:237] Train net output #0: loss = 1.21351 (* 1 = 1.21351 loss) I0409 23:49:46.160604 6396 sgd_solver.cpp:105] Iteration 5832, lr = 0.00314983 I0409 23:49:51.136642 6396 solver.cpp:218] Iteration 5844 (2.41162 iter/s, 4.9759s/12 iters), loss = 1.14345 I0409 23:49:51.136679 6396 solver.cpp:237] Train net output #0: loss = 1.14345 (* 1 = 1.14345 loss) I0409 23:49:51.136687 6396 sgd_solver.cpp:105] Iteration 5844, lr = 0.00314235 I0409 23:49:56.078155 6396 solver.cpp:218] Iteration 5856 (2.4285 iter/s, 4.94132s/12 iters), loss = 0.906926 I0409 23:49:56.078205 6396 solver.cpp:237] Train net output #0: loss = 0.906926 (* 1 = 0.906926 loss) I0409 23:49:56.078214 6396 sgd_solver.cpp:105] Iteration 5856, lr = 0.00313489 I0409 23:50:00.223799 6400 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:50:01.035733 6396 solver.cpp:218] Iteration 5868 (2.42064 iter/s, 4.95737s/12 iters), loss = 0.882607 I0409 23:50:01.035779 6396 solver.cpp:237] Train net output #0: loss = 0.882607 (* 1 = 0.882607 loss) I0409 23:50:01.035789 6396 sgd_solver.cpp:105] Iteration 5868, lr = 0.00312745 I0409 23:50:06.083715 6396 solver.cpp:218] Iteration 5880 (2.37728 iter/s, 5.04778s/12 iters), loss = 1.10167 I0409 23:50:06.083765 6396 solver.cpp:237] Train net output #0: loss = 1.10167 (* 1 = 1.10167 loss) I0409 23:50:06.083776 6396 sgd_solver.cpp:105] Iteration 5880, lr = 0.00312002 I0409 23:50:10.988432 6396 solver.cpp:218] Iteration 5892 (2.44672 iter/s, 4.90452s/12 iters), loss = 1.17707 I0409 23:50:10.988557 6396 solver.cpp:237] Train net output #0: loss = 1.17707 (* 1 = 1.17707 loss) I0409 23:50:10.988567 6396 sgd_solver.cpp:105] Iteration 5892, lr = 0.00311262 I0409 23:50:15.901196 6396 solver.cpp:218] Iteration 5904 (2.44276 iter/s, 4.91248s/12 iters), loss = 0.919068 I0409 23:50:15.901248 6396 solver.cpp:237] Train net output #0: loss = 0.919068 (* 1 = 0.919068 loss) I0409 23:50:15.901258 6396 sgd_solver.cpp:105] Iteration 5904, lr = 0.00310523 I0409 23:50:20.398375 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_5916.caffemodel I0409 23:50:20.855705 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_5916.solverstate I0409 23:50:21.175101 6396 solver.cpp:330] Iteration 5916, Testing net (#0) I0409 23:50:21.175130 6396 net.cpp:676] Ignoring source layer train-data I0409 23:50:23.406951 6401 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:50:25.804229 6396 solver.cpp:397] Test net output #0: accuracy = 0.518995 I0409 23:50:25.804273 6396 solver.cpp:397] Test net output #1: loss = 2.02677 (* 1 = 2.02677 loss) I0409 23:50:25.887398 6396 solver.cpp:218] Iteration 5916 (1.2017 iter/s, 9.98585s/12 iters), loss = 1.19185 I0409 23:50:25.887444 6396 solver.cpp:237] Train net output #0: loss = 1.19185 (* 1 = 1.19185 loss) I0409 23:50:25.887454 6396 sgd_solver.cpp:105] Iteration 5916, lr = 0.00309785 I0409 23:50:30.184517 6396 solver.cpp:218] Iteration 5928 (2.79269 iter/s, 4.29694s/12 iters), loss = 1.00369 I0409 23:50:30.184561 6396 solver.cpp:237] Train net output #0: loss = 1.00369 (* 1 = 1.00369 loss) I0409 23:50:30.184569 6396 sgd_solver.cpp:105] Iteration 5928, lr = 0.0030905 I0409 23:50:35.126665 6396 solver.cpp:218] Iteration 5940 (2.42819 iter/s, 4.94195s/12 iters), loss = 1.35095 I0409 23:50:35.126710 6396 solver.cpp:237] Train net output #0: loss = 1.35095 (* 1 = 1.35095 loss) I0409 23:50:35.126719 6396 sgd_solver.cpp:105] Iteration 5940, lr = 0.00308316 I0409 23:50:40.094236 6396 solver.cpp:218] Iteration 5952 (2.41577 iter/s, 4.96736s/12 iters), loss = 0.970139 I0409 23:50:40.094291 6396 solver.cpp:237] Train net output #0: loss = 0.970139 (* 1 = 0.970139 loss) I0409 23:50:40.094301 6396 sgd_solver.cpp:105] Iteration 5952, lr = 0.00307584 I0409 23:50:44.994009 6396 solver.cpp:218] Iteration 5964 (2.4492 iter/s, 4.89956s/12 iters), loss = 1.02891 I0409 23:50:44.994127 6396 solver.cpp:237] Train net output #0: loss = 1.02891 (* 1 = 1.02891 loss) I0409 23:50:44.994140 6396 sgd_solver.cpp:105] Iteration 5964, lr = 0.00306854 I0409 23:50:46.274405 6400 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:50:49.918916 6396 solver.cpp:218] Iteration 5976 (2.43673 iter/s, 4.92463s/12 iters), loss = 0.776308 I0409 23:50:49.918967 6396 solver.cpp:237] Train net output #0: loss = 0.776308 (* 1 = 0.776308 loss) I0409 23:50:49.918975 6396 sgd_solver.cpp:105] Iteration 5976, lr = 0.00306125 I0409 23:50:54.817873 6396 solver.cpp:218] Iteration 5988 (2.4496 iter/s, 4.89875s/12 iters), loss = 1.19161 I0409 23:50:54.817925 6396 solver.cpp:237] Train net output #0: loss = 1.19161 (* 1 = 1.19161 loss) I0409 23:50:54.817937 6396 sgd_solver.cpp:105] Iteration 5988, lr = 0.00305398 I0409 23:50:59.790978 6396 solver.cpp:218] Iteration 6000 (2.41308 iter/s, 4.9729s/12 iters), loss = 0.833682 I0409 23:50:59.791024 6396 solver.cpp:237] Train net output #0: loss = 0.833682 (* 1 = 0.833682 loss) I0409 23:50:59.791033 6396 sgd_solver.cpp:105] Iteration 6000, lr = 0.00304673 I0409 23:51:04.653532 6396 solver.cpp:218] Iteration 6012 (2.46794 iter/s, 4.86235s/12 iters), loss = 0.790592 I0409 23:51:04.653594 6396 solver.cpp:237] Train net output #0: loss = 0.790592 (* 1 = 0.790592 loss) I0409 23:51:04.653607 6396 sgd_solver.cpp:105] Iteration 6012, lr = 0.0030395 I0409 23:51:06.660596 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_6018.caffemodel I0409 23:51:07.366806 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_6018.solverstate I0409 23:51:07.685896 6396 solver.cpp:330] Iteration 6018, Testing net (#0) I0409 23:51:07.685926 6396 net.cpp:676] Ignoring source layer train-data I0409 23:51:09.681866 6401 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:51:12.085785 6396 solver.cpp:397] Test net output #0: accuracy = 0.51348 I0409 23:51:12.085824 6396 solver.cpp:397] Test net output #1: loss = 1.98972 (* 1 = 1.98972 loss) I0409 23:51:13.949035 6396 solver.cpp:218] Iteration 6024 (1.29099 iter/s, 9.29516s/12 iters), loss = 0.88475 I0409 23:51:13.949090 6396 solver.cpp:237] Train net output #0: loss = 0.88475 (* 1 = 0.88475 loss) I0409 23:51:13.949101 6396 sgd_solver.cpp:105] Iteration 6024, lr = 0.00303228 I0409 23:51:18.800797 6396 solver.cpp:218] Iteration 6036 (2.47343 iter/s, 4.85155s/12 iters), loss = 0.99033 I0409 23:51:18.800945 6396 solver.cpp:237] Train net output #0: loss = 0.99033 (* 1 = 0.99033 loss) I0409 23:51:18.800959 6396 sgd_solver.cpp:105] Iteration 6036, lr = 0.00302508 I0409 23:51:23.710374 6396 solver.cpp:218] Iteration 6048 (2.44435 iter/s, 4.90928s/12 iters), loss = 1.1571 I0409 23:51:23.710427 6396 solver.cpp:237] Train net output #0: loss = 1.1571 (* 1 = 1.1571 loss) I0409 23:51:23.710438 6396 sgd_solver.cpp:105] Iteration 6048, lr = 0.0030179 I0409 23:51:28.632125 6396 solver.cpp:218] Iteration 6060 (2.43826 iter/s, 4.92154s/12 iters), loss = 1.04913 I0409 23:51:28.632174 6396 solver.cpp:237] Train net output #0: loss = 1.04913 (* 1 = 1.04913 loss) I0409 23:51:28.632184 6396 sgd_solver.cpp:105] Iteration 6060, lr = 0.00301074 I0409 23:51:32.038890 6400 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:51:33.572576 6396 solver.cpp:218] Iteration 6072 (2.42903 iter/s, 4.94025s/12 iters), loss = 1.15423 I0409 23:51:33.572625 6396 solver.cpp:237] Train net output #0: loss = 1.15423 (* 1 = 1.15423 loss) I0409 23:51:33.572638 6396 sgd_solver.cpp:105] Iteration 6072, lr = 0.00300359 I0409 23:51:38.471980 6396 solver.cpp:218] Iteration 6084 (2.44938 iter/s, 4.8992s/12 iters), loss = 1.16375 I0409 23:51:38.472023 6396 solver.cpp:237] Train net output #0: loss = 1.16375 (* 1 = 1.16375 loss) I0409 23:51:38.472033 6396 sgd_solver.cpp:105] Iteration 6084, lr = 0.00299646 I0409 23:51:43.376546 6396 solver.cpp:218] Iteration 6096 (2.4468 iter/s, 4.90437s/12 iters), loss = 1.05462 I0409 23:51:43.376591 6396 solver.cpp:237] Train net output #0: loss = 1.05462 (* 1 = 1.05462 loss) I0409 23:51:43.376601 6396 sgd_solver.cpp:105] Iteration 6096, lr = 0.00298934 I0409 23:51:48.320433 6396 solver.cpp:218] Iteration 6108 (2.42734 iter/s, 4.94368s/12 iters), loss = 0.767485 I0409 23:51:48.320490 6396 solver.cpp:237] Train net output #0: loss = 0.767485 (* 1 = 0.767485 loss) I0409 23:51:48.320503 6396 sgd_solver.cpp:105] Iteration 6108, lr = 0.00298225 I0409 23:51:52.787204 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_6120.caffemodel I0409 23:51:53.218420 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_6120.solverstate I0409 23:51:53.517920 6396 solver.cpp:330] Iteration 6120, Testing net (#0) I0409 23:51:53.517940 6396 net.cpp:676] Ignoring source layer train-data I0409 23:51:55.580144 6401 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:51:57.975822 6396 solver.cpp:397] Test net output #0: accuracy = 0.525735 I0409 23:51:57.975860 6396 solver.cpp:397] Test net output #1: loss = 1.96531 (* 1 = 1.96531 loss) I0409 23:51:58.059106 6396 solver.cpp:218] Iteration 6120 (1.23224 iter/s, 9.73833s/12 iters), loss = 0.848845 I0409 23:51:58.059170 6396 solver.cpp:237] Train net output #0: loss = 0.848845 (* 1 = 0.848845 loss) I0409 23:51:58.059181 6396 sgd_solver.cpp:105] Iteration 6120, lr = 0.00297517 I0409 23:52:02.161110 6396 solver.cpp:218] Iteration 6132 (2.92553 iter/s, 4.10182s/12 iters), loss = 0.963003 I0409 23:52:02.161149 6396 solver.cpp:237] Train net output #0: loss = 0.963003 (* 1 = 0.963003 loss) I0409 23:52:02.161159 6396 sgd_solver.cpp:105] Iteration 6132, lr = 0.0029681 I0409 23:52:07.150951 6396 solver.cpp:218] Iteration 6144 (2.40498 iter/s, 4.98965s/12 iters), loss = 0.995381 I0409 23:52:07.150998 6396 solver.cpp:237] Train net output #0: loss = 0.995381 (* 1 = 0.995381 loss) I0409 23:52:07.151008 6396 sgd_solver.cpp:105] Iteration 6144, lr = 0.00296105 I0409 23:52:12.079670 6396 solver.cpp:218] Iteration 6156 (2.43481 iter/s, 4.92852s/12 iters), loss = 0.873192 I0409 23:52:12.079716 6396 solver.cpp:237] Train net output #0: loss = 0.873192 (* 1 = 0.873192 loss) I0409 23:52:12.079728 6396 sgd_solver.cpp:105] Iteration 6156, lr = 0.00295402 I0409 23:52:16.996079 6396 solver.cpp:218] Iteration 6168 (2.44091 iter/s, 4.91621s/12 iters), loss = 0.818559 I0409 23:52:16.996136 6396 solver.cpp:237] Train net output #0: loss = 0.818559 (* 1 = 0.818559 loss) I0409 23:52:16.996150 6396 sgd_solver.cpp:105] Iteration 6168, lr = 0.00294701 I0409 23:52:17.582381 6400 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:52:21.880861 6396 solver.cpp:218] Iteration 6180 (2.45672 iter/s, 4.88457s/12 iters), loss = 0.839055 I0409 23:52:21.880918 6396 solver.cpp:237] Train net output #0: loss = 0.839055 (* 1 = 0.839055 loss) I0409 23:52:21.880930 6396 sgd_solver.cpp:105] Iteration 6180, lr = 0.00294001 I0409 23:52:26.748980 6396 solver.cpp:218] Iteration 6192 (2.46512 iter/s, 4.86791s/12 iters), loss = 0.984422 I0409 23:52:26.749123 6396 solver.cpp:237] Train net output #0: loss = 0.984422 (* 1 = 0.984422 loss) I0409 23:52:26.749136 6396 sgd_solver.cpp:105] Iteration 6192, lr = 0.00293303 I0409 23:52:31.647132 6396 solver.cpp:218] Iteration 6204 (2.45005 iter/s, 4.89786s/12 iters), loss = 0.901569 I0409 23:52:31.647179 6396 solver.cpp:237] Train net output #0: loss = 0.901569 (* 1 = 0.901569 loss) I0409 23:52:31.647190 6396 sgd_solver.cpp:105] Iteration 6204, lr = 0.00292607 I0409 23:52:36.516243 6396 solver.cpp:218] Iteration 6216 (2.46462 iter/s, 4.86891s/12 iters), loss = 1.2934 I0409 23:52:36.516294 6396 solver.cpp:237] Train net output #0: loss = 1.2934 (* 1 = 1.2934 loss) I0409 23:52:36.516306 6396 sgd_solver.cpp:105] Iteration 6216, lr = 0.00291912 I0409 23:52:38.505594 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_6222.caffemodel I0409 23:52:38.981986 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_6222.solverstate I0409 23:52:39.299414 6396 solver.cpp:330] Iteration 6222, Testing net (#0) I0409 23:52:39.299443 6396 net.cpp:676] Ignoring source layer train-data I0409 23:52:41.433678 6401 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:52:42.345520 6396 blocking_queue.cpp:49] Waiting for data I0409 23:52:43.875663 6396 solver.cpp:397] Test net output #0: accuracy = 0.505515 I0409 23:52:43.875727 6396 solver.cpp:397] Test net output #1: loss = 2.04526 (* 1 = 2.04526 loss) I0409 23:52:45.815692 6396 solver.cpp:218] Iteration 6228 (1.29044 iter/s, 9.29912s/12 iters), loss = 0.994679 I0409 23:52:45.815737 6396 solver.cpp:237] Train net output #0: loss = 0.994679 (* 1 = 0.994679 loss) I0409 23:52:45.815747 6396 sgd_solver.cpp:105] Iteration 6228, lr = 0.00291219 I0409 23:52:50.781051 6396 solver.cpp:218] Iteration 6240 (2.41684 iter/s, 4.96516s/12 iters), loss = 0.782719 I0409 23:52:50.781093 6396 solver.cpp:237] Train net output #0: loss = 0.782719 (* 1 = 0.782719 loss) I0409 23:52:50.781101 6396 sgd_solver.cpp:105] Iteration 6240, lr = 0.00290528 I0409 23:52:55.677510 6396 solver.cpp:218] Iteration 6252 (2.45085 iter/s, 4.89625s/12 iters), loss = 0.99411 I0409 23:52:55.677572 6396 solver.cpp:237] Train net output #0: loss = 0.99411 (* 1 = 0.99411 loss) I0409 23:52:55.677584 6396 sgd_solver.cpp:105] Iteration 6252, lr = 0.00289838 I0409 23:53:00.640974 6396 solver.cpp:218] Iteration 6264 (2.41777 iter/s, 4.96325s/12 iters), loss = 1.02187 I0409 23:53:00.641093 6396 solver.cpp:237] Train net output #0: loss = 1.02187 (* 1 = 1.02187 loss) I0409 23:53:00.641103 6396 sgd_solver.cpp:105] Iteration 6264, lr = 0.0028915 I0409 23:53:03.321924 6400 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:53:05.507732 6396 solver.cpp:218] Iteration 6276 (2.46585 iter/s, 4.86649s/12 iters), loss = 0.847722 I0409 23:53:05.507779 6396 solver.cpp:237] Train net output #0: loss = 0.847722 (* 1 = 0.847722 loss) I0409 23:53:05.507788 6396 sgd_solver.cpp:105] Iteration 6276, lr = 0.00288463 I0409 23:53:10.447453 6396 solver.cpp:218] Iteration 6288 (2.42939 iter/s, 4.93951s/12 iters), loss = 0.962475 I0409 23:53:10.447510 6396 solver.cpp:237] Train net output #0: loss = 0.962475 (* 1 = 0.962475 loss) I0409 23:53:10.447523 6396 sgd_solver.cpp:105] Iteration 6288, lr = 0.00287779 I0409 23:53:15.453933 6396 solver.cpp:218] Iteration 6300 (2.397 iter/s, 5.00627s/12 iters), loss = 0.988435 I0409 23:53:15.454005 6396 solver.cpp:237] Train net output #0: loss = 0.988435 (* 1 = 0.988435 loss) I0409 23:53:15.454016 6396 sgd_solver.cpp:105] Iteration 6300, lr = 0.00287095 I0409 23:53:20.420783 6396 solver.cpp:218] Iteration 6312 (2.41613 iter/s, 4.96662s/12 iters), loss = 0.871215 I0409 23:53:20.420841 6396 solver.cpp:237] Train net output #0: loss = 0.871215 (* 1 = 0.871215 loss) I0409 23:53:20.420853 6396 sgd_solver.cpp:105] Iteration 6312, lr = 0.00286414 I0409 23:53:24.909654 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_6324.caffemodel I0409 23:53:25.929073 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_6324.solverstate I0409 23:53:26.244889 6396 solver.cpp:330] Iteration 6324, Testing net (#0) I0409 23:53:26.244907 6396 net.cpp:676] Ignoring source layer train-data I0409 23:53:28.332283 6401 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:53:30.807737 6396 solver.cpp:397] Test net output #0: accuracy = 0.52451 I0409 23:53:30.807858 6396 solver.cpp:397] Test net output #1: loss = 2.01769 (* 1 = 2.01769 loss) I0409 23:53:30.891368 6396 solver.cpp:218] Iteration 6324 (1.14611 iter/s, 10.4702s/12 iters), loss = 0.992378 I0409 23:53:30.891417 6396 solver.cpp:237] Train net output #0: loss = 0.992378 (* 1 = 0.992378 loss) I0409 23:53:30.891428 6396 sgd_solver.cpp:105] Iteration 6324, lr = 0.00285734 I0409 23:53:35.112056 6396 solver.cpp:218] Iteration 6336 (2.84326 iter/s, 4.22051s/12 iters), loss = 0.842865 I0409 23:53:35.112102 6396 solver.cpp:237] Train net output #0: loss = 0.842865 (* 1 = 0.842865 loss) I0409 23:53:35.112110 6396 sgd_solver.cpp:105] Iteration 6336, lr = 0.00285055 I0409 23:53:40.079713 6396 solver.cpp:218] Iteration 6348 (2.41572 iter/s, 4.96746s/12 iters), loss = 0.923089 I0409 23:53:40.079761 6396 solver.cpp:237] Train net output #0: loss = 0.923089 (* 1 = 0.923089 loss) I0409 23:53:40.079774 6396 sgd_solver.cpp:105] Iteration 6348, lr = 0.00284379 I0409 23:53:45.018323 6396 solver.cpp:218] Iteration 6360 (2.42994 iter/s, 4.9384s/12 iters), loss = 0.843535 I0409 23:53:45.018381 6396 solver.cpp:237] Train net output #0: loss = 0.843535 (* 1 = 0.843535 loss) I0409 23:53:45.018393 6396 sgd_solver.cpp:105] Iteration 6360, lr = 0.00283703 I0409 23:53:49.791493 6400 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:53:49.929862 6396 solver.cpp:218] Iteration 6372 (2.44333 iter/s, 4.91133s/12 iters), loss = 0.754288 I0409 23:53:49.929908 6396 solver.cpp:237] Train net output #0: loss = 0.754288 (* 1 = 0.754288 loss) I0409 23:53:49.929916 6396 sgd_solver.cpp:105] Iteration 6372, lr = 0.0028303 I0409 23:53:54.816990 6396 solver.cpp:218] Iteration 6384 (2.45553 iter/s, 4.88693s/12 iters), loss = 0.760821 I0409 23:53:54.817039 6396 solver.cpp:237] Train net output #0: loss = 0.760821 (* 1 = 0.760821 loss) I0409 23:53:54.817050 6396 sgd_solver.cpp:105] Iteration 6384, lr = 0.00282358 I0409 23:53:59.908077 6396 solver.cpp:218] Iteration 6396 (2.35716 iter/s, 5.09088s/12 iters), loss = 0.701372 I0409 23:53:59.908131 6396 solver.cpp:237] Train net output #0: loss = 0.701372 (* 1 = 0.701372 loss) I0409 23:53:59.908143 6396 sgd_solver.cpp:105] Iteration 6396, lr = 0.00281687 I0409 23:54:04.824442 6396 solver.cpp:218] Iteration 6408 (2.44093 iter/s, 4.91616s/12 iters), loss = 0.934794 I0409 23:54:04.824591 6396 solver.cpp:237] Train net output #0: loss = 0.934794 (* 1 = 0.934794 loss) I0409 23:54:04.824604 6396 sgd_solver.cpp:105] Iteration 6408, lr = 0.00281019 I0409 23:54:09.760883 6396 solver.cpp:218] Iteration 6420 (2.43105 iter/s, 4.93614s/12 iters), loss = 0.979845 I0409 23:54:09.760937 6396 solver.cpp:237] Train net output #0: loss = 0.979845 (* 1 = 0.979845 loss) I0409 23:54:09.760951 6396 sgd_solver.cpp:105] Iteration 6420, lr = 0.00280351 I0409 23:54:11.771391 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_6426.caffemodel I0409 23:54:12.211131 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_6426.solverstate I0409 23:54:12.513878 6396 solver.cpp:330] Iteration 6426, Testing net (#0) I0409 23:54:12.513903 6396 net.cpp:676] Ignoring source layer train-data I0409 23:54:14.374596 6401 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:54:16.894865 6396 solver.cpp:397] Test net output #0: accuracy = 0.529412 I0409 23:54:16.894907 6396 solver.cpp:397] Test net output #1: loss = 1.93795 (* 1 = 1.93795 loss) I0409 23:54:18.840032 6396 solver.cpp:218] Iteration 6432 (1.32176 iter/s, 9.07882s/12 iters), loss = 0.681316 I0409 23:54:18.840090 6396 solver.cpp:237] Train net output #0: loss = 0.681316 (* 1 = 0.681316 loss) I0409 23:54:18.840101 6396 sgd_solver.cpp:105] Iteration 6432, lr = 0.00279686 I0409 23:54:23.816783 6396 solver.cpp:218] Iteration 6444 (2.41132 iter/s, 4.97653s/12 iters), loss = 0.650329 I0409 23:54:23.816843 6396 solver.cpp:237] Train net output #0: loss = 0.650329 (* 1 = 0.650329 loss) I0409 23:54:23.816854 6396 sgd_solver.cpp:105] Iteration 6444, lr = 0.00279022 I0409 23:54:28.756048 6396 solver.cpp:218] Iteration 6456 (2.42962 iter/s, 4.93905s/12 iters), loss = 0.978873 I0409 23:54:28.756100 6396 solver.cpp:237] Train net output #0: loss = 0.978873 (* 1 = 0.978873 loss) I0409 23:54:28.756114 6396 sgd_solver.cpp:105] Iteration 6456, lr = 0.00278359 I0409 23:54:33.713376 6396 solver.cpp:218] Iteration 6468 (2.42076 iter/s, 4.95712s/12 iters), loss = 0.655662 I0409 23:54:33.713428 6396 solver.cpp:237] Train net output #0: loss = 0.655662 (* 1 = 0.655662 loss) I0409 23:54:33.713438 6396 sgd_solver.cpp:105] Iteration 6468, lr = 0.00277698 I0409 23:54:35.711524 6400 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:54:38.635504 6396 solver.cpp:218] Iteration 6480 (2.43807 iter/s, 4.92192s/12 iters), loss = 0.950461 I0409 23:54:38.635560 6396 solver.cpp:237] Train net output #0: loss = 0.950461 (* 1 = 0.950461 loss) I0409 23:54:38.635573 6396 sgd_solver.cpp:105] Iteration 6480, lr = 0.00277039 I0409 23:54:43.567827 6396 solver.cpp:218] Iteration 6492 (2.43303 iter/s, 4.93212s/12 iters), loss = 0.947059 I0409 23:54:43.567879 6396 solver.cpp:237] Train net output #0: loss = 0.947059 (* 1 = 0.947059 loss) I0409 23:54:43.567891 6396 sgd_solver.cpp:105] Iteration 6492, lr = 0.00276381 I0409 23:54:48.501179 6396 solver.cpp:218] Iteration 6504 (2.43252 iter/s, 4.93315s/12 iters), loss = 0.801746 I0409 23:54:48.501224 6396 solver.cpp:237] Train net output #0: loss = 0.801746 (* 1 = 0.801746 loss) I0409 23:54:48.501232 6396 sgd_solver.cpp:105] Iteration 6504, lr = 0.00275725 I0409 23:54:53.384951 6396 solver.cpp:218] Iteration 6516 (2.45722 iter/s, 4.88357s/12 iters), loss = 0.780661 I0409 23:54:53.384997 6396 solver.cpp:237] Train net output #0: loss = 0.780661 (* 1 = 0.780661 loss) I0409 23:54:53.385007 6396 sgd_solver.cpp:105] Iteration 6516, lr = 0.00275071 I0409 23:54:57.834748 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_6528.caffemodel I0409 23:54:58.282397 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_6528.solverstate I0409 23:54:58.598738 6396 solver.cpp:330] Iteration 6528, Testing net (#0) I0409 23:54:58.598762 6396 net.cpp:676] Ignoring source layer train-data I0409 23:55:00.505369 6401 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:55:03.095752 6396 solver.cpp:397] Test net output #0: accuracy = 0.519608 I0409 23:55:03.095795 6396 solver.cpp:397] Test net output #1: loss = 1.97383 (* 1 = 1.97383 loss) I0409 23:55:03.179100 6396 solver.cpp:218] Iteration 6528 (1.22526 iter/s, 9.79381s/12 iters), loss = 0.793367 I0409 23:55:03.179149 6396 solver.cpp:237] Train net output #0: loss = 0.793367 (* 1 = 0.793367 loss) I0409 23:55:03.179160 6396 sgd_solver.cpp:105] Iteration 6528, lr = 0.00274418 I0409 23:55:07.276386 6396 solver.cpp:218] Iteration 6540 (2.9289 iter/s, 4.0971s/12 iters), loss = 0.764667 I0409 23:55:07.276559 6396 solver.cpp:237] Train net output #0: loss = 0.764667 (* 1 = 0.764667 loss) I0409 23:55:07.276573 6396 sgd_solver.cpp:105] Iteration 6540, lr = 0.00273766 I0409 23:55:12.265012 6396 solver.cpp:218] Iteration 6552 (2.40564 iter/s, 4.98829s/12 iters), loss = 0.799266 I0409 23:55:12.265084 6396 solver.cpp:237] Train net output #0: loss = 0.799266 (* 1 = 0.799266 loss) I0409 23:55:12.265100 6396 sgd_solver.cpp:105] Iteration 6552, lr = 0.00273116 I0409 23:55:17.215070 6396 solver.cpp:218] Iteration 6564 (2.42432 iter/s, 4.94984s/12 iters), loss = 1.01906 I0409 23:55:17.215112 6396 solver.cpp:237] Train net output #0: loss = 1.01906 (* 1 = 1.01906 loss) I0409 23:55:17.215121 6396 sgd_solver.cpp:105] Iteration 6564, lr = 0.00272468 I0409 23:55:21.368147 6400 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:55:22.139032 6396 solver.cpp:218] Iteration 6576 (2.43716 iter/s, 4.92376s/12 iters), loss = 0.87882 I0409 23:55:22.139083 6396 solver.cpp:237] Train net output #0: loss = 0.87882 (* 1 = 0.87882 loss) I0409 23:55:22.139094 6396 sgd_solver.cpp:105] Iteration 6576, lr = 0.00271821 I0409 23:55:27.149272 6396 solver.cpp:218] Iteration 6588 (2.3952 iter/s, 5.01003s/12 iters), loss = 0.942534 I0409 23:55:27.149333 6396 solver.cpp:237] Train net output #0: loss = 0.942534 (* 1 = 0.942534 loss) I0409 23:55:27.149344 6396 sgd_solver.cpp:105] Iteration 6588, lr = 0.00271175 I0409 23:55:32.089977 6396 solver.cpp:218] Iteration 6600 (2.42892 iter/s, 4.94046s/12 iters), loss = 0.828873 I0409 23:55:32.090039 6396 solver.cpp:237] Train net output #0: loss = 0.828873 (* 1 = 0.828873 loss) I0409 23:55:32.090051 6396 sgd_solver.cpp:105] Iteration 6600, lr = 0.00270532 I0409 23:55:37.032243 6396 solver.cpp:218] Iteration 6612 (2.42814 iter/s, 4.94205s/12 iters), loss = 0.681193 I0409 23:55:37.032299 6396 solver.cpp:237] Train net output #0: loss = 0.681193 (* 1 = 0.681193 loss) I0409 23:55:37.032310 6396 sgd_solver.cpp:105] Iteration 6612, lr = 0.00269889 I0409 23:55:42.005010 6396 solver.cpp:218] Iteration 6624 (2.41325 iter/s, 4.97255s/12 iters), loss = 0.806195 I0409 23:55:42.005115 6396 solver.cpp:237] Train net output #0: loss = 0.806195 (* 1 = 0.806195 loss) I0409 23:55:42.005125 6396 sgd_solver.cpp:105] Iteration 6624, lr = 0.00269248 I0409 23:55:44.013720 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_6630.caffemodel I0409 23:55:44.464143 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_6630.solverstate I0409 23:55:44.789587 6396 solver.cpp:330] Iteration 6630, Testing net (#0) I0409 23:55:44.789606 6396 net.cpp:676] Ignoring source layer train-data I0409 23:55:46.675779 6401 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:55:49.274042 6396 solver.cpp:397] Test net output #0: accuracy = 0.529412 I0409 23:55:49.274091 6396 solver.cpp:397] Test net output #1: loss = 1.97565 (* 1 = 1.97565 loss) I0409 23:55:51.060660 6396 solver.cpp:218] Iteration 6636 (1.3252 iter/s, 9.05527s/12 iters), loss = 0.761862 I0409 23:55:51.060719 6396 solver.cpp:237] Train net output #0: loss = 0.761862 (* 1 = 0.761862 loss) I0409 23:55:51.060730 6396 sgd_solver.cpp:105] Iteration 6636, lr = 0.00268609 I0409 23:55:55.985746 6396 solver.cpp:218] Iteration 6648 (2.43661 iter/s, 4.92487s/12 iters), loss = 0.619172 I0409 23:55:55.985805 6396 solver.cpp:237] Train net output #0: loss = 0.619172 (* 1 = 0.619172 loss) I0409 23:55:55.985817 6396 sgd_solver.cpp:105] Iteration 6648, lr = 0.00267971 I0409 23:56:00.922060 6396 solver.cpp:218] Iteration 6660 (2.43107 iter/s, 4.9361s/12 iters), loss = 0.960879 I0409 23:56:00.922122 6396 solver.cpp:237] Train net output #0: loss = 0.960879 (* 1 = 0.960879 loss) I0409 23:56:00.922134 6396 sgd_solver.cpp:105] Iteration 6660, lr = 0.00267335 I0409 23:56:05.886056 6396 solver.cpp:218] Iteration 6672 (2.41751 iter/s, 4.96378s/12 iters), loss = 0.88338 I0409 23:56:05.886117 6396 solver.cpp:237] Train net output #0: loss = 0.88338 (* 1 = 0.88338 loss) I0409 23:56:05.886129 6396 sgd_solver.cpp:105] Iteration 6672, lr = 0.00266701 I0409 23:56:07.239257 6400 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:56:10.841478 6396 solver.cpp:218] Iteration 6684 (2.4217 iter/s, 4.9552s/12 iters), loss = 0.597468 I0409 23:56:10.841536 6396 solver.cpp:237] Train net output #0: loss = 0.597468 (* 1 = 0.597468 loss) I0409 23:56:10.841547 6396 sgd_solver.cpp:105] Iteration 6684, lr = 0.00266067 I0409 23:56:15.772181 6396 solver.cpp:218] Iteration 6696 (2.43384 iter/s, 4.93049s/12 iters), loss = 0.768611 I0409 23:56:15.772337 6396 solver.cpp:237] Train net output #0: loss = 0.768611 (* 1 = 0.768611 loss) I0409 23:56:15.772351 6396 sgd_solver.cpp:105] Iteration 6696, lr = 0.00265436 I0409 23:56:20.685463 6396 solver.cpp:218] Iteration 6708 (2.44251 iter/s, 4.91297s/12 iters), loss = 0.788428 I0409 23:56:20.685513 6396 solver.cpp:237] Train net output #0: loss = 0.788428 (* 1 = 0.788428 loss) I0409 23:56:20.685523 6396 sgd_solver.cpp:105] Iteration 6708, lr = 0.00264805 I0409 23:56:25.596266 6396 solver.cpp:218] Iteration 6720 (2.4437 iter/s, 4.9106s/12 iters), loss = 0.67188 I0409 23:56:25.596313 6396 solver.cpp:237] Train net output #0: loss = 0.67188 (* 1 = 0.67188 loss) I0409 23:56:25.596326 6396 sgd_solver.cpp:105] Iteration 6720, lr = 0.00264177 I0409 23:56:30.042732 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_6732.caffemodel I0409 23:56:30.699683 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_6732.solverstate I0409 23:56:33.099964 6396 solver.cpp:330] Iteration 6732, Testing net (#0) I0409 23:56:33.099985 6396 net.cpp:676] Ignoring source layer train-data I0409 23:56:34.778334 6401 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:56:37.409770 6396 solver.cpp:397] Test net output #0: accuracy = 0.518995 I0409 23:56:37.409809 6396 solver.cpp:397] Test net output #1: loss = 1.95163 (* 1 = 1.95163 loss) I0409 23:56:37.492944 6396 solver.cpp:218] Iteration 6732 (1.00872 iter/s, 11.8963s/12 iters), loss = 0.675438 I0409 23:56:37.492990 6396 solver.cpp:237] Train net output #0: loss = 0.675438 (* 1 = 0.675438 loss) I0409 23:56:37.493000 6396 sgd_solver.cpp:105] Iteration 6732, lr = 0.0026355 I0409 23:56:41.703135 6396 solver.cpp:218] Iteration 6744 (2.85035 iter/s, 4.21001s/12 iters), loss = 0.640104 I0409 23:56:41.703191 6396 solver.cpp:237] Train net output #0: loss = 0.640104 (* 1 = 0.640104 loss) I0409 23:56:41.703202 6396 sgd_solver.cpp:105] Iteration 6744, lr = 0.00262924 I0409 23:56:46.645893 6396 solver.cpp:218] Iteration 6756 (2.4279 iter/s, 4.94254s/12 iters), loss = 0.923736 I0409 23:56:46.646041 6396 solver.cpp:237] Train net output #0: loss = 0.923736 (* 1 = 0.923736 loss) I0409 23:56:46.646054 6396 sgd_solver.cpp:105] Iteration 6756, lr = 0.002623 I0409 23:56:51.933611 6396 solver.cpp:218] Iteration 6768 (2.26955 iter/s, 5.2874s/12 iters), loss = 0.66228 I0409 23:56:51.933666 6396 solver.cpp:237] Train net output #0: loss = 0.66228 (* 1 = 0.66228 loss) I0409 23:56:51.933678 6396 sgd_solver.cpp:105] Iteration 6768, lr = 0.00261677 I0409 23:56:55.403510 6400 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:56:56.896687 6396 solver.cpp:218] Iteration 6780 (2.41796 iter/s, 4.96287s/12 iters), loss = 0.870623 I0409 23:56:56.896739 6396 solver.cpp:237] Train net output #0: loss = 0.870623 (* 1 = 0.870623 loss) I0409 23:56:56.896751 6396 sgd_solver.cpp:105] Iteration 6780, lr = 0.00261056 I0409 23:57:01.876158 6396 solver.cpp:218] Iteration 6792 (2.40999 iter/s, 4.97927s/12 iters), loss = 0.919935 I0409 23:57:01.876202 6396 solver.cpp:237] Train net output #0: loss = 0.919935 (* 1 = 0.919935 loss) I0409 23:57:01.876211 6396 sgd_solver.cpp:105] Iteration 6792, lr = 0.00260436 I0409 23:57:06.739792 6396 solver.cpp:218] Iteration 6804 (2.46739 iter/s, 4.86344s/12 iters), loss = 0.888565 I0409 23:57:06.739841 6396 solver.cpp:237] Train net output #0: loss = 0.888565 (* 1 = 0.888565 loss) I0409 23:57:06.739851 6396 sgd_solver.cpp:105] Iteration 6804, lr = 0.00259817 I0409 23:57:11.638533 6396 solver.cpp:218] Iteration 6816 (2.44971 iter/s, 4.89854s/12 iters), loss = 0.687512 I0409 23:57:11.638566 6396 solver.cpp:237] Train net output #0: loss = 0.687512 (* 1 = 0.687512 loss) I0409 23:57:11.638573 6396 sgd_solver.cpp:105] Iteration 6816, lr = 0.00259201 I0409 23:57:16.573742 6396 solver.cpp:218] Iteration 6828 (2.4316 iter/s, 4.93502s/12 iters), loss = 0.619806 I0409 23:57:16.573801 6396 solver.cpp:237] Train net output #0: loss = 0.619806 (* 1 = 0.619806 loss) I0409 23:57:16.573814 6396 sgd_solver.cpp:105] Iteration 6828, lr = 0.00258585 I0409 23:57:18.600248 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_6834.caffemodel I0409 23:57:19.182363 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_6834.solverstate I0409 23:57:19.513450 6396 solver.cpp:330] Iteration 6834, Testing net (#0) I0409 23:57:19.513478 6396 net.cpp:676] Ignoring source layer train-data I0409 23:57:21.391649 6401 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:57:24.102596 6396 solver.cpp:397] Test net output #0: accuracy = 0.545956 I0409 23:57:24.102638 6396 solver.cpp:397] Test net output #1: loss = 1.93661 (* 1 = 1.93661 loss) I0409 23:57:25.861629 6396 solver.cpp:218] Iteration 6840 (1.29205 iter/s, 9.28755s/12 iters), loss = 0.478748 I0409 23:57:25.861680 6396 solver.cpp:237] Train net output #0: loss = 0.478748 (* 1 = 0.478748 loss) I0409 23:57:25.861692 6396 sgd_solver.cpp:105] Iteration 6840, lr = 0.00257971 I0409 23:57:30.708218 6396 solver.cpp:218] Iteration 6852 (2.47607 iter/s, 4.84638s/12 iters), loss = 0.710092 I0409 23:57:30.708276 6396 solver.cpp:237] Train net output #0: loss = 0.710092 (* 1 = 0.710092 loss) I0409 23:57:30.708289 6396 sgd_solver.cpp:105] Iteration 6852, lr = 0.00257359 I0409 23:57:35.667367 6396 solver.cpp:218] Iteration 6864 (2.41987 iter/s, 4.95894s/12 iters), loss = 0.929079 I0409 23:57:35.667412 6396 solver.cpp:237] Train net output #0: loss = 0.929079 (* 1 = 0.929079 loss) I0409 23:57:35.667421 6396 sgd_solver.cpp:105] Iteration 6864, lr = 0.00256748 I0409 23:57:40.536710 6396 solver.cpp:218] Iteration 6876 (2.4645 iter/s, 4.86914s/12 iters), loss = 0.812541 I0409 23:57:40.536756 6396 solver.cpp:237] Train net output #0: loss = 0.812541 (* 1 = 0.812541 loss) I0409 23:57:40.536765 6396 sgd_solver.cpp:105] Iteration 6876, lr = 0.00256138 I0409 23:57:41.168723 6400 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:57:45.449782 6396 solver.cpp:218] Iteration 6888 (2.44257 iter/s, 4.91287s/12 iters), loss = 0.797272 I0409 23:57:45.449841 6396 solver.cpp:237] Train net output #0: loss = 0.797272 (* 1 = 0.797272 loss) I0409 23:57:45.449854 6396 sgd_solver.cpp:105] Iteration 6888, lr = 0.0025553 I0409 23:57:50.370096 6396 solver.cpp:218] Iteration 6900 (2.43898 iter/s, 4.92009s/12 iters), loss = 0.743687 I0409 23:57:50.370270 6396 solver.cpp:237] Train net output #0: loss = 0.743687 (* 1 = 0.743687 loss) I0409 23:57:50.370285 6396 sgd_solver.cpp:105] Iteration 6900, lr = 0.00254923 I0409 23:57:55.252480 6396 solver.cpp:218] Iteration 6912 (2.45798 iter/s, 4.88206s/12 iters), loss = 0.857599 I0409 23:57:55.252535 6396 solver.cpp:237] Train net output #0: loss = 0.857599 (* 1 = 0.857599 loss) I0409 23:57:55.252549 6396 sgd_solver.cpp:105] Iteration 6912, lr = 0.00254318 I0409 23:58:00.196331 6396 solver.cpp:218] Iteration 6924 (2.42736 iter/s, 4.94364s/12 iters), loss = 0.994493 I0409 23:58:00.196382 6396 solver.cpp:237] Train net output #0: loss = 0.994493 (* 1 = 0.994493 loss) I0409 23:58:00.196393 6396 sgd_solver.cpp:105] Iteration 6924, lr = 0.00253714 I0409 23:58:04.668382 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_6936.caffemodel I0409 23:58:05.094630 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_6936.solverstate I0409 23:58:05.397547 6396 solver.cpp:330] Iteration 6936, Testing net (#0) I0409 23:58:05.397570 6396 net.cpp:676] Ignoring source layer train-data I0409 23:58:05.680248 6396 blocking_queue.cpp:49] Waiting for data I0409 23:58:07.164506 6401 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:58:09.967836 6396 solver.cpp:397] Test net output #0: accuracy = 0.511029 I0409 23:58:09.967881 6396 solver.cpp:397] Test net output #1: loss = 2.01742 (* 1 = 2.01742 loss) I0409 23:58:10.051323 6396 solver.cpp:218] Iteration 6936 (1.2177 iter/s, 9.85465s/12 iters), loss = 0.613284 I0409 23:58:10.051371 6396 solver.cpp:237] Train net output #0: loss = 0.613284 (* 1 = 0.613284 loss) I0409 23:58:10.051381 6396 sgd_solver.cpp:105] Iteration 6936, lr = 0.00253112 I0409 23:58:14.234899 6396 solver.cpp:218] Iteration 6948 (2.86848 iter/s, 4.18339s/12 iters), loss = 0.859434 I0409 23:58:14.234946 6396 solver.cpp:237] Train net output #0: loss = 0.859434 (* 1 = 0.859434 loss) I0409 23:58:14.234953 6396 sgd_solver.cpp:105] Iteration 6948, lr = 0.00252511 I0409 23:58:19.222745 6396 solver.cpp:218] Iteration 6960 (2.40595 iter/s, 4.98764s/12 iters), loss = 0.713329 I0409 23:58:19.222796 6396 solver.cpp:237] Train net output #0: loss = 0.713329 (* 1 = 0.713329 loss) I0409 23:58:19.222807 6396 sgd_solver.cpp:105] Iteration 6960, lr = 0.00251911 I0409 23:58:24.277379 6396 solver.cpp:218] Iteration 6972 (2.37416 iter/s, 5.05442s/12 iters), loss = 0.711089 I0409 23:58:24.277467 6396 solver.cpp:237] Train net output #0: loss = 0.711089 (* 1 = 0.711089 loss) I0409 23:58:24.277475 6396 sgd_solver.cpp:105] Iteration 6972, lr = 0.00251313 I0409 23:58:27.032757 6400 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:58:29.254009 6396 solver.cpp:218] Iteration 6984 (2.41139 iter/s, 4.97639s/12 iters), loss = 0.913944 I0409 23:58:29.254052 6396 solver.cpp:237] Train net output #0: loss = 0.913944 (* 1 = 0.913944 loss) I0409 23:58:29.254063 6396 sgd_solver.cpp:105] Iteration 6984, lr = 0.00250717 I0409 23:58:34.236649 6396 solver.cpp:218] Iteration 6996 (2.40846 iter/s, 4.98244s/12 iters), loss = 0.787754 I0409 23:58:34.236690 6396 solver.cpp:237] Train net output #0: loss = 0.787754 (* 1 = 0.787754 loss) I0409 23:58:34.236698 6396 sgd_solver.cpp:105] Iteration 6996, lr = 0.00250121 I0409 23:58:39.159011 6396 solver.cpp:218] Iteration 7008 (2.43795 iter/s, 4.92217s/12 iters), loss = 0.660303 I0409 23:58:39.159057 6396 solver.cpp:237] Train net output #0: loss = 0.660303 (* 1 = 0.660303 loss) I0409 23:58:39.159067 6396 sgd_solver.cpp:105] Iteration 7008, lr = 0.00249528 I0409 23:58:44.185220 6396 solver.cpp:218] Iteration 7020 (2.38759 iter/s, 5.026s/12 iters), loss = 0.73754 I0409 23:58:44.185274 6396 solver.cpp:237] Train net output #0: loss = 0.73754 (* 1 = 0.73754 loss) I0409 23:58:44.185286 6396 sgd_solver.cpp:105] Iteration 7020, lr = 0.00248935 I0409 23:58:49.138160 6396 solver.cpp:218] Iteration 7032 (2.42291 iter/s, 4.95273s/12 iters), loss = 0.831974 I0409 23:58:49.138207 6396 solver.cpp:237] Train net output #0: loss = 0.831974 (* 1 = 0.831974 loss) I0409 23:58:49.138216 6396 sgd_solver.cpp:105] Iteration 7032, lr = 0.00248344 I0409 23:58:51.122125 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_7038.caffemodel I0409 23:58:51.576687 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_7038.solverstate I0409 23:58:51.893054 6396 solver.cpp:330] Iteration 7038, Testing net (#0) I0409 23:58:51.893080 6396 net.cpp:676] Ignoring source layer train-data I0409 23:58:53.590219 6401 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:58:56.334398 6396 solver.cpp:397] Test net output #0: accuracy = 0.531863 I0409 23:58:56.334568 6396 solver.cpp:397] Test net output #1: loss = 1.96661 (* 1 = 1.96661 loss) I0409 23:58:58.180212 6396 solver.cpp:218] Iteration 7044 (1.32718 iter/s, 9.04173s/12 iters), loss = 0.788345 I0409 23:58:58.180263 6396 solver.cpp:237] Train net output #0: loss = 0.788345 (* 1 = 0.788345 loss) I0409 23:58:58.180274 6396 sgd_solver.cpp:105] Iteration 7044, lr = 0.00247755 I0409 23:59:03.297711 6396 solver.cpp:218] Iteration 7056 (2.34499 iter/s, 5.11729s/12 iters), loss = 0.504124 I0409 23:59:03.297763 6396 solver.cpp:237] Train net output #0: loss = 0.504124 (* 1 = 0.504124 loss) I0409 23:59:03.297775 6396 sgd_solver.cpp:105] Iteration 7056, lr = 0.00247166 I0409 23:59:08.273528 6396 solver.cpp:218] Iteration 7068 (2.41177 iter/s, 4.9756s/12 iters), loss = 0.74639 I0409 23:59:08.273581 6396 solver.cpp:237] Train net output #0: loss = 0.74639 (* 1 = 0.74639 loss) I0409 23:59:08.273593 6396 sgd_solver.cpp:105] Iteration 7068, lr = 0.0024658 I0409 23:59:13.082933 6400 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:59:13.192281 6396 solver.cpp:218] Iteration 7080 (2.43975 iter/s, 4.91855s/12 iters), loss = 0.682843 I0409 23:59:13.192332 6396 solver.cpp:237] Train net output #0: loss = 0.682843 (* 1 = 0.682843 loss) I0409 23:59:13.192342 6396 sgd_solver.cpp:105] Iteration 7080, lr = 0.00245994 I0409 23:59:18.157563 6396 solver.cpp:218] Iteration 7092 (2.41689 iter/s, 4.96507s/12 iters), loss = 0.581838 I0409 23:59:18.157613 6396 solver.cpp:237] Train net output #0: loss = 0.581838 (* 1 = 0.581838 loss) I0409 23:59:18.157622 6396 sgd_solver.cpp:105] Iteration 7092, lr = 0.0024541 I0409 23:59:23.102349 6396 solver.cpp:218] Iteration 7104 (2.4269 iter/s, 4.94458s/12 iters), loss = 0.470003 I0409 23:59:23.102393 6396 solver.cpp:237] Train net output #0: loss = 0.470003 (* 1 = 0.470003 loss) I0409 23:59:23.102402 6396 sgd_solver.cpp:105] Iteration 7104, lr = 0.00244827 I0409 23:59:28.128913 6396 solver.cpp:218] Iteration 7116 (2.38741 iter/s, 5.02636s/12 iters), loss = 0.6785 I0409 23:59:28.129032 6396 solver.cpp:237] Train net output #0: loss = 0.6785 (* 1 = 0.6785 loss) I0409 23:59:28.129047 6396 sgd_solver.cpp:105] Iteration 7116, lr = 0.00244246 I0409 23:59:33.179520 6396 solver.cpp:218] Iteration 7128 (2.37608 iter/s, 5.05033s/12 iters), loss = 0.577561 I0409 23:59:33.179566 6396 solver.cpp:237] Train net output #0: loss = 0.577561 (* 1 = 0.577561 loss) I0409 23:59:33.179577 6396 sgd_solver.cpp:105] Iteration 7128, lr = 0.00243666 I0409 23:59:37.686298 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_7140.caffemodel I0409 23:59:38.160656 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_7140.solverstate I0409 23:59:38.482975 6396 solver.cpp:330] Iteration 7140, Testing net (#0) I0409 23:59:38.483002 6396 net.cpp:676] Ignoring source layer train-data I0409 23:59:40.059579 6401 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:59:42.869243 6396 solver.cpp:397] Test net output #0: accuracy = 0.541054 I0409 23:59:42.869292 6396 solver.cpp:397] Test net output #1: loss = 2.03287 (* 1 = 2.03287 loss) I0409 23:59:42.952859 6396 solver.cpp:218] Iteration 7140 (1.22787 iter/s, 9.773s/12 iters), loss = 0.626334 I0409 23:59:42.952913 6396 solver.cpp:237] Train net output #0: loss = 0.626334 (* 1 = 0.626334 loss) I0409 23:59:42.952925 6396 sgd_solver.cpp:105] Iteration 7140, lr = 0.00243088 I0409 23:59:47.135026 6396 solver.cpp:218] Iteration 7152 (2.86946 iter/s, 4.18197s/12 iters), loss = 0.454517 I0409 23:59:47.135076 6396 solver.cpp:237] Train net output #0: loss = 0.454517 (* 1 = 0.454517 loss) I0409 23:59:47.135085 6396 sgd_solver.cpp:105] Iteration 7152, lr = 0.00242511 I0409 23:59:52.038527 6396 solver.cpp:218] Iteration 7164 (2.44733 iter/s, 4.9033s/12 iters), loss = 0.730396 I0409 23:59:52.038574 6396 solver.cpp:237] Train net output #0: loss = 0.730396 (* 1 = 0.730396 loss) I0409 23:59:52.038583 6396 sgd_solver.cpp:105] Iteration 7164, lr = 0.00241935 I0409 23:59:57.042497 6396 solver.cpp:218] Iteration 7176 (2.3982 iter/s, 5.00376s/12 iters), loss = 0.720915 I0409 23:59:57.042551 6396 solver.cpp:237] Train net output #0: loss = 0.720915 (* 1 = 0.720915 loss) I0409 23:59:57.042563 6396 sgd_solver.cpp:105] Iteration 7176, lr = 0.0024136 I0409 23:59:59.152545 6400 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:00:02.026098 6396 solver.cpp:218] Iteration 7188 (2.408 iter/s, 4.98339s/12 iters), loss = 0.818263 I0410 00:00:02.026146 6396 solver.cpp:237] Train net output #0: loss = 0.818263 (* 1 = 0.818263 loss) I0410 00:00:02.026155 6396 sgd_solver.cpp:105] Iteration 7188, lr = 0.00240787 I0410 00:00:06.981920 6396 solver.cpp:218] Iteration 7200 (2.4215 iter/s, 4.95562s/12 iters), loss = 0.60214 I0410 00:00:06.981995 6396 solver.cpp:237] Train net output #0: loss = 0.60214 (* 1 = 0.60214 loss) I0410 00:00:06.982005 6396 sgd_solver.cpp:105] Iteration 7200, lr = 0.00240216 I0410 00:00:11.925021 6396 solver.cpp:218] Iteration 7212 (2.42774 iter/s, 4.94287s/12 iters), loss = 0.571428 I0410 00:00:11.925078 6396 solver.cpp:237] Train net output #0: loss = 0.571428 (* 1 = 0.571428 loss) I0410 00:00:11.925091 6396 sgd_solver.cpp:105] Iteration 7212, lr = 0.00239645 I0410 00:00:16.928952 6396 solver.cpp:218] Iteration 7224 (2.39822 iter/s, 5.00371s/12 iters), loss = 0.64254 I0410 00:00:16.929003 6396 solver.cpp:237] Train net output #0: loss = 0.64254 (* 1 = 0.64254 loss) I0410 00:00:16.929013 6396 sgd_solver.cpp:105] Iteration 7224, lr = 0.00239076 I0410 00:00:21.927843 6396 solver.cpp:218] Iteration 7236 (2.40064 iter/s, 4.99868s/12 iters), loss = 0.437528 I0410 00:00:21.927912 6396 solver.cpp:237] Train net output #0: loss = 0.437528 (* 1 = 0.437528 loss) I0410 00:00:21.927929 6396 sgd_solver.cpp:105] Iteration 7236, lr = 0.00238509 I0410 00:00:23.960222 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_7242.caffemodel I0410 00:00:25.485577 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_7242.solverstate I0410 00:00:26.591641 6396 solver.cpp:330] Iteration 7242, Testing net (#0) I0410 00:00:26.591670 6396 net.cpp:676] Ignoring source layer train-data I0410 00:00:28.225821 6401 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:00:31.125572 6396 solver.cpp:397] Test net output #0: accuracy = 0.544118 I0410 00:00:31.125697 6396 solver.cpp:397] Test net output #1: loss = 1.9328 (* 1 = 1.9328 loss) I0410 00:00:33.054443 6396 solver.cpp:218] Iteration 7248 (1.07854 iter/s, 11.1262s/12 iters), loss = 0.643116 I0410 00:00:33.054503 6396 solver.cpp:237] Train net output #0: loss = 0.643116 (* 1 = 0.643116 loss) I0410 00:00:33.054515 6396 sgd_solver.cpp:105] Iteration 7248, lr = 0.00237942 I0410 00:00:37.993185 6396 solver.cpp:218] Iteration 7260 (2.42987 iter/s, 4.93853s/12 iters), loss = 0.731656 I0410 00:00:37.993233 6396 solver.cpp:237] Train net output #0: loss = 0.731656 (* 1 = 0.731656 loss) I0410 00:00:37.993244 6396 sgd_solver.cpp:105] Iteration 7260, lr = 0.00237378 I0410 00:00:42.927254 6396 solver.cpp:218] Iteration 7272 (2.43217 iter/s, 4.93386s/12 iters), loss = 0.664226 I0410 00:00:42.927311 6396 solver.cpp:237] Train net output #0: loss = 0.664226 (* 1 = 0.664226 loss) I0410 00:00:42.927325 6396 sgd_solver.cpp:105] Iteration 7272, lr = 0.00236814 I0410 00:00:47.139536 6400 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:00:47.875774 6396 solver.cpp:218] Iteration 7284 (2.42507 iter/s, 4.94831s/12 iters), loss = 0.720213 I0410 00:00:47.875833 6396 solver.cpp:237] Train net output #0: loss = 0.720213 (* 1 = 0.720213 loss) I0410 00:00:47.875845 6396 sgd_solver.cpp:105] Iteration 7284, lr = 0.00236252 I0410 00:00:52.848039 6396 solver.cpp:218] Iteration 7296 (2.41349 iter/s, 4.97205s/12 iters), loss = 0.608607 I0410 00:00:52.848085 6396 solver.cpp:237] Train net output #0: loss = 0.608607 (* 1 = 0.608607 loss) I0410 00:00:52.848096 6396 sgd_solver.cpp:105] Iteration 7296, lr = 0.00235691 I0410 00:00:57.758594 6396 solver.cpp:218] Iteration 7308 (2.44382 iter/s, 4.91035s/12 iters), loss = 0.613965 I0410 00:00:57.758646 6396 solver.cpp:237] Train net output #0: loss = 0.613965 (* 1 = 0.613965 loss) I0410 00:00:57.758657 6396 sgd_solver.cpp:105] Iteration 7308, lr = 0.00235131 I0410 00:01:02.656047 6396 solver.cpp:218] Iteration 7320 (2.45036 iter/s, 4.89725s/12 iters), loss = 0.444402 I0410 00:01:02.658304 6396 solver.cpp:237] Train net output #0: loss = 0.444402 (* 1 = 0.444402 loss) I0410 00:01:02.658316 6396 sgd_solver.cpp:105] Iteration 7320, lr = 0.00234573 I0410 00:01:07.562597 6396 solver.cpp:218] Iteration 7332 (2.44691 iter/s, 4.90414s/12 iters), loss = 0.577648 I0410 00:01:07.562646 6396 solver.cpp:237] Train net output #0: loss = 0.577648 (* 1 = 0.577648 loss) I0410 00:01:07.562656 6396 sgd_solver.cpp:105] Iteration 7332, lr = 0.00234016 I0410 00:01:12.017565 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_7344.caffemodel I0410 00:01:12.434422 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_7344.solverstate I0410 00:01:12.732647 6396 solver.cpp:330] Iteration 7344, Testing net (#0) I0410 00:01:12.732676 6396 net.cpp:676] Ignoring source layer train-data I0410 00:01:14.338770 6401 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:01:17.254411 6396 solver.cpp:397] Test net output #0: accuracy = 0.556373 I0410 00:01:17.254458 6396 solver.cpp:397] Test net output #1: loss = 1.89776 (* 1 = 1.89776 loss) I0410 00:01:17.337860 6396 solver.cpp:218] Iteration 7344 (1.22763 iter/s, 9.77491s/12 iters), loss = 0.580602 I0410 00:01:17.337911 6396 solver.cpp:237] Train net output #0: loss = 0.580602 (* 1 = 0.580602 loss) I0410 00:01:17.337924 6396 sgd_solver.cpp:105] Iteration 7344, lr = 0.0023346 I0410 00:01:21.482645 6396 solver.cpp:218] Iteration 7356 (2.89533 iter/s, 4.1446s/12 iters), loss = 0.397144 I0410 00:01:21.482686 6396 solver.cpp:237] Train net output #0: loss = 0.397144 (* 1 = 0.397144 loss) I0410 00:01:21.482694 6396 sgd_solver.cpp:105] Iteration 7356, lr = 0.00232906 I0410 00:01:26.370476 6396 solver.cpp:218] Iteration 7368 (2.45518 iter/s, 4.88762s/12 iters), loss = 0.594387 I0410 00:01:26.370545 6396 solver.cpp:237] Train net output #0: loss = 0.594387 (* 1 = 0.594387 loss) I0410 00:01:26.370563 6396 sgd_solver.cpp:105] Iteration 7368, lr = 0.00232353 I0410 00:01:31.328253 6396 solver.cpp:218] Iteration 7380 (2.42055 iter/s, 4.95756s/12 iters), loss = 0.511821 I0410 00:01:31.328294 6396 solver.cpp:237] Train net output #0: loss = 0.511821 (* 1 = 0.511821 loss) I0410 00:01:31.328303 6396 sgd_solver.cpp:105] Iteration 7380, lr = 0.00231802 I0410 00:01:32.701251 6400 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:01:36.244905 6396 solver.cpp:218] Iteration 7392 (2.44079 iter/s, 4.91645s/12 iters), loss = 0.744804 I0410 00:01:36.244966 6396 solver.cpp:237] Train net output #0: loss = 0.744804 (* 1 = 0.744804 loss) I0410 00:01:36.244979 6396 sgd_solver.cpp:105] Iteration 7392, lr = 0.00231251 I0410 00:01:41.177532 6396 solver.cpp:218] Iteration 7404 (2.43289 iter/s, 4.93241s/12 iters), loss = 0.617495 I0410 00:01:41.177589 6396 solver.cpp:237] Train net output #0: loss = 0.617495 (* 1 = 0.617495 loss) I0410 00:01:41.177603 6396 sgd_solver.cpp:105] Iteration 7404, lr = 0.00230702 I0410 00:01:46.136521 6396 solver.cpp:218] Iteration 7416 (2.41995 iter/s, 4.95878s/12 iters), loss = 0.490691 I0410 00:01:46.136579 6396 solver.cpp:237] Train net output #0: loss = 0.490691 (* 1 = 0.490691 loss) I0410 00:01:46.136590 6396 sgd_solver.cpp:105] Iteration 7416, lr = 0.00230154 I0410 00:01:51.026942 6396 solver.cpp:218] Iteration 7428 (2.45388 iter/s, 4.89021s/12 iters), loss = 0.589477 I0410 00:01:51.027002 6396 solver.cpp:237] Train net output #0: loss = 0.589477 (* 1 = 0.589477 loss) I0410 00:01:51.027014 6396 sgd_solver.cpp:105] Iteration 7428, lr = 0.00229608 I0410 00:01:55.949924 6396 solver.cpp:218] Iteration 7440 (2.43765 iter/s, 4.92276s/12 iters), loss = 0.531264 I0410 00:01:55.949981 6396 solver.cpp:237] Train net output #0: loss = 0.531264 (* 1 = 0.531264 loss) I0410 00:01:55.949990 6396 sgd_solver.cpp:105] Iteration 7440, lr = 0.00229063 I0410 00:01:57.905782 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_7446.caffemodel I0410 00:01:59.164742 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_7446.solverstate I0410 00:01:59.476202 6396 solver.cpp:330] Iteration 7446, Testing net (#0) I0410 00:01:59.476231 6396 net.cpp:676] Ignoring source layer train-data I0410 00:02:01.029978 6401 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:02:03.936115 6396 solver.cpp:397] Test net output #0: accuracy = 0.544118 I0410 00:02:03.936233 6396 solver.cpp:397] Test net output #1: loss = 2.00447 (* 1 = 2.00447 loss) I0410 00:02:05.830524 6396 solver.cpp:218] Iteration 7452 (1.21455 iter/s, 9.88024s/12 iters), loss = 0.521374 I0410 00:02:05.830576 6396 solver.cpp:237] Train net output #0: loss = 0.521374 (* 1 = 0.521374 loss) I0410 00:02:05.830587 6396 sgd_solver.cpp:105] Iteration 7452, lr = 0.00228519 I0410 00:02:10.713801 6396 solver.cpp:218] Iteration 7464 (2.45747 iter/s, 4.88306s/12 iters), loss = 0.658872 I0410 00:02:10.713857 6396 solver.cpp:237] Train net output #0: loss = 0.658872 (* 1 = 0.658872 loss) I0410 00:02:10.713871 6396 sgd_solver.cpp:105] Iteration 7464, lr = 0.00227976 I0410 00:02:15.672991 6396 solver.cpp:218] Iteration 7476 (2.41985 iter/s, 4.95898s/12 iters), loss = 0.695809 I0410 00:02:15.673048 6396 solver.cpp:237] Train net output #0: loss = 0.695809 (* 1 = 0.695809 loss) I0410 00:02:15.673063 6396 sgd_solver.cpp:105] Iteration 7476, lr = 0.00227435 I0410 00:02:19.125316 6400 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:02:20.612300 6396 solver.cpp:218] Iteration 7488 (2.4296 iter/s, 4.93909s/12 iters), loss = 0.69765 I0410 00:02:20.612354 6396 solver.cpp:237] Train net output #0: loss = 0.69765 (* 1 = 0.69765 loss) I0410 00:02:20.612366 6396 sgd_solver.cpp:105] Iteration 7488, lr = 0.00226895 I0410 00:02:25.686451 6396 solver.cpp:218] Iteration 7500 (2.36503 iter/s, 5.07394s/12 iters), loss = 0.589432 I0410 00:02:25.686504 6396 solver.cpp:237] Train net output #0: loss = 0.589432 (* 1 = 0.589432 loss) I0410 00:02:25.686514 6396 sgd_solver.cpp:105] Iteration 7500, lr = 0.00226357 I0410 00:02:30.567566 6396 solver.cpp:218] Iteration 7512 (2.45856 iter/s, 4.88091s/12 iters), loss = 0.600003 I0410 00:02:30.567616 6396 solver.cpp:237] Train net output #0: loss = 0.600003 (* 1 = 0.600003 loss) I0410 00:02:30.567625 6396 sgd_solver.cpp:105] Iteration 7512, lr = 0.00225819 I0410 00:02:35.446323 6396 solver.cpp:218] Iteration 7524 (2.45975 iter/s, 4.87855s/12 iters), loss = 0.601211 I0410 00:02:35.446461 6396 solver.cpp:237] Train net output #0: loss = 0.601211 (* 1 = 0.601211 loss) I0410 00:02:35.446470 6396 sgd_solver.cpp:105] Iteration 7524, lr = 0.00225283 I0410 00:02:40.363965 6396 solver.cpp:218] Iteration 7536 (2.44034 iter/s, 4.91735s/12 iters), loss = 0.608871 I0410 00:02:40.364013 6396 solver.cpp:237] Train net output #0: loss = 0.608871 (* 1 = 0.608871 loss) I0410 00:02:40.364022 6396 sgd_solver.cpp:105] Iteration 7536, lr = 0.00224748 I0410 00:02:44.834403 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_7548.caffemodel I0410 00:02:45.290529 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_7548.solverstate I0410 00:02:45.607193 6396 solver.cpp:330] Iteration 7548, Testing net (#0) I0410 00:02:45.607218 6396 net.cpp:676] Ignoring source layer train-data I0410 00:02:47.088316 6401 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:02:50.029294 6396 solver.cpp:397] Test net output #0: accuracy = 0.558824 I0410 00:02:50.029345 6396 solver.cpp:397] Test net output #1: loss = 1.94891 (* 1 = 1.94891 loss) I0410 00:02:50.112569 6396 solver.cpp:218] Iteration 7548 (1.23099 iter/s, 9.74826s/12 iters), loss = 0.48554 I0410 00:02:50.112623 6396 solver.cpp:237] Train net output #0: loss = 0.48554 (* 1 = 0.48554 loss) I0410 00:02:50.112634 6396 sgd_solver.cpp:105] Iteration 7548, lr = 0.00224215 I0410 00:02:54.304950 6396 solver.cpp:218] Iteration 7560 (2.86246 iter/s, 4.19219s/12 iters), loss = 0.718588 I0410 00:02:54.304993 6396 solver.cpp:237] Train net output #0: loss = 0.718588 (* 1 = 0.718588 loss) I0410 00:02:54.305003 6396 sgd_solver.cpp:105] Iteration 7560, lr = 0.00223682 I0410 00:02:59.254824 6396 solver.cpp:218] Iteration 7572 (2.4244 iter/s, 4.94967s/12 iters), loss = 0.631978 I0410 00:02:59.254878 6396 solver.cpp:237] Train net output #0: loss = 0.631978 (* 1 = 0.631978 loss) I0410 00:02:59.254887 6396 sgd_solver.cpp:105] Iteration 7572, lr = 0.00223151 I0410 00:03:04.170388 6396 solver.cpp:218] Iteration 7584 (2.44133 iter/s, 4.91535s/12 iters), loss = 0.682018 I0410 00:03:04.170445 6396 solver.cpp:237] Train net output #0: loss = 0.682018 (* 1 = 0.682018 loss) I0410 00:03:04.170459 6396 sgd_solver.cpp:105] Iteration 7584, lr = 0.00222621 I0410 00:03:04.813354 6400 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:03:09.225067 6396 solver.cpp:218] Iteration 7596 (2.37414 iter/s, 5.05446s/12 iters), loss = 0.479076 I0410 00:03:09.225188 6396 solver.cpp:237] Train net output #0: loss = 0.479076 (* 1 = 0.479076 loss) I0410 00:03:09.225200 6396 sgd_solver.cpp:105] Iteration 7596, lr = 0.00222093 I0410 00:03:14.219568 6396 solver.cpp:218] Iteration 7608 (2.40278 iter/s, 4.99422s/12 iters), loss = 0.70589 I0410 00:03:14.219624 6396 solver.cpp:237] Train net output #0: loss = 0.70589 (* 1 = 0.70589 loss) I0410 00:03:14.219637 6396 sgd_solver.cpp:105] Iteration 7608, lr = 0.00221565 I0410 00:03:19.211242 6396 solver.cpp:218] Iteration 7620 (2.4041 iter/s, 4.99147s/12 iters), loss = 0.573745 I0410 00:03:19.211288 6396 solver.cpp:237] Train net output #0: loss = 0.573745 (* 1 = 0.573745 loss) I0410 00:03:19.211298 6396 sgd_solver.cpp:105] Iteration 7620, lr = 0.00221039 I0410 00:03:19.964200 6396 blocking_queue.cpp:49] Waiting for data I0410 00:03:24.156847 6396 solver.cpp:218] Iteration 7632 (2.42649 iter/s, 4.94541s/12 iters), loss = 0.593114 I0410 00:03:24.156898 6396 solver.cpp:237] Train net output #0: loss = 0.593114 (* 1 = 0.593114 loss) I0410 00:03:24.156909 6396 sgd_solver.cpp:105] Iteration 7632, lr = 0.00220515 I0410 00:03:29.103682 6396 solver.cpp:218] Iteration 7644 (2.4259 iter/s, 4.94663s/12 iters), loss = 0.526375 I0410 00:03:29.103735 6396 solver.cpp:237] Train net output #0: loss = 0.526375 (* 1 = 0.526375 loss) I0410 00:03:29.103747 6396 sgd_solver.cpp:105] Iteration 7644, lr = 0.00219991 I0410 00:03:31.103133 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_7650.caffemodel I0410 00:03:31.577872 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_7650.solverstate I0410 00:03:31.886904 6396 solver.cpp:330] Iteration 7650, Testing net (#0) I0410 00:03:31.886920 6396 net.cpp:676] Ignoring source layer train-data I0410 00:03:33.354883 6401 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:03:36.343739 6396 solver.cpp:397] Test net output #0: accuracy = 0.552083 I0410 00:03:36.343791 6396 solver.cpp:397] Test net output #1: loss = 1.96015 (* 1 = 1.96015 loss) I0410 00:03:38.095127 6396 solver.cpp:218] Iteration 7656 (1.33465 iter/s, 8.99112s/12 iters), loss = 0.571128 I0410 00:03:38.095171 6396 solver.cpp:237] Train net output #0: loss = 0.571128 (* 1 = 0.571128 loss) I0410 00:03:38.095180 6396 sgd_solver.cpp:105] Iteration 7656, lr = 0.00219469 I0410 00:03:43.052511 6396 solver.cpp:218] Iteration 7668 (2.42073 iter/s, 4.95718s/12 iters), loss = 0.55561 I0410 00:03:43.052635 6396 solver.cpp:237] Train net output #0: loss = 0.55561 (* 1 = 0.55561 loss) I0410 00:03:43.052646 6396 sgd_solver.cpp:105] Iteration 7668, lr = 0.00218948 I0410 00:03:47.962146 6396 solver.cpp:218] Iteration 7680 (2.44431 iter/s, 4.90935s/12 iters), loss = 0.529483 I0410 00:03:47.962203 6396 solver.cpp:237] Train net output #0: loss = 0.529483 (* 1 = 0.529483 loss) I0410 00:03:47.962214 6396 sgd_solver.cpp:105] Iteration 7680, lr = 0.00218428 I0410 00:03:50.734066 6400 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:03:52.903115 6396 solver.cpp:218] Iteration 7692 (2.42878 iter/s, 4.94076s/12 iters), loss = 0.528502 I0410 00:03:52.903156 6396 solver.cpp:237] Train net output #0: loss = 0.528502 (* 1 = 0.528502 loss) I0410 00:03:52.903165 6396 sgd_solver.cpp:105] Iteration 7692, lr = 0.00217909 I0410 00:03:57.919350 6396 solver.cpp:218] Iteration 7704 (2.39233 iter/s, 5.01603s/12 iters), loss = 0.559512 I0410 00:03:57.919392 6396 solver.cpp:237] Train net output #0: loss = 0.559512 (* 1 = 0.559512 loss) I0410 00:03:57.919401 6396 sgd_solver.cpp:105] Iteration 7704, lr = 0.00217392 I0410 00:04:02.893317 6396 solver.cpp:218] Iteration 7716 (2.41266 iter/s, 4.97376s/12 iters), loss = 0.614838 I0410 00:04:02.893393 6396 solver.cpp:237] Train net output #0: loss = 0.614838 (* 1 = 0.614838 loss) I0410 00:04:02.893407 6396 sgd_solver.cpp:105] Iteration 7716, lr = 0.00216876 I0410 00:04:07.810540 6396 solver.cpp:218] Iteration 7728 (2.44051 iter/s, 4.917s/12 iters), loss = 0.397453 I0410 00:04:07.810586 6396 solver.cpp:237] Train net output #0: loss = 0.397453 (* 1 = 0.397453 loss) I0410 00:04:07.810595 6396 sgd_solver.cpp:105] Iteration 7728, lr = 0.00216361 I0410 00:04:12.722501 6396 solver.cpp:218] Iteration 7740 (2.44312 iter/s, 4.91175s/12 iters), loss = 0.530043 I0410 00:04:12.722559 6396 solver.cpp:237] Train net output #0: loss = 0.530043 (* 1 = 0.530043 loss) I0410 00:04:12.722571 6396 sgd_solver.cpp:105] Iteration 7740, lr = 0.00215847 I0410 00:04:17.179324 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_7752.caffemodel I0410 00:04:17.636726 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_7752.solverstate I0410 00:04:19.083039 6396 solver.cpp:330] Iteration 7752, Testing net (#0) I0410 00:04:19.083070 6396 net.cpp:676] Ignoring source layer train-data I0410 00:04:20.481665 6401 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:04:23.506044 6396 solver.cpp:397] Test net output #0: accuracy = 0.553309 I0410 00:04:23.506096 6396 solver.cpp:397] Test net output #1: loss = 1.9487 (* 1 = 1.9487 loss) I0410 00:04:23.589452 6396 solver.cpp:218] Iteration 7752 (1.1043 iter/s, 10.8666s/12 iters), loss = 0.492757 I0410 00:04:23.589507 6396 solver.cpp:237] Train net output #0: loss = 0.492757 (* 1 = 0.492757 loss) I0410 00:04:23.589519 6396 sgd_solver.cpp:105] Iteration 7752, lr = 0.00215335 I0410 00:04:27.675400 6396 solver.cpp:218] Iteration 7764 (2.93703 iter/s, 4.08576s/12 iters), loss = 0.564788 I0410 00:04:27.675468 6396 solver.cpp:237] Train net output #0: loss = 0.564788 (* 1 = 0.564788 loss) I0410 00:04:27.675480 6396 sgd_solver.cpp:105] Iteration 7764, lr = 0.00214823 I0410 00:04:32.545477 6396 solver.cpp:218] Iteration 7776 (2.46414 iter/s, 4.86985s/12 iters), loss = 0.437101 I0410 00:04:32.545537 6396 solver.cpp:237] Train net output #0: loss = 0.437101 (* 1 = 0.437101 loss) I0410 00:04:32.545548 6396 sgd_solver.cpp:105] Iteration 7776, lr = 0.00214313 I0410 00:04:37.562237 6396 solver.cpp:218] Iteration 7788 (2.39209 iter/s, 5.01654s/12 iters), loss = 0.500784 I0410 00:04:37.562297 6396 solver.cpp:237] Train net output #0: loss = 0.500784 (* 1 = 0.500784 loss) I0410 00:04:37.562310 6396 sgd_solver.cpp:105] Iteration 7788, lr = 0.00213805 I0410 00:04:37.574162 6400 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:04:42.576879 6396 solver.cpp:218] Iteration 7800 (2.3931 iter/s, 5.01442s/12 iters), loss = 0.463285 I0410 00:04:42.576928 6396 solver.cpp:237] Train net output #0: loss = 0.463285 (* 1 = 0.463285 loss) I0410 00:04:42.576938 6396 sgd_solver.cpp:105] Iteration 7800, lr = 0.00213297 I0410 00:04:47.552256 6396 solver.cpp:218] Iteration 7812 (2.41198 iter/s, 4.97517s/12 iters), loss = 0.358046 I0410 00:04:47.552397 6396 solver.cpp:237] Train net output #0: loss = 0.358046 (* 1 = 0.358046 loss) I0410 00:04:47.552409 6396 sgd_solver.cpp:105] Iteration 7812, lr = 0.00212791 I0410 00:04:52.453280 6396 solver.cpp:218] Iteration 7824 (2.44862 iter/s, 4.90073s/12 iters), loss = 0.438287 I0410 00:04:52.453328 6396 solver.cpp:237] Train net output #0: loss = 0.438287 (* 1 = 0.438287 loss) I0410 00:04:52.453338 6396 sgd_solver.cpp:105] Iteration 7824, lr = 0.00212285 I0410 00:04:57.433306 6396 solver.cpp:218] Iteration 7836 (2.40973 iter/s, 4.97982s/12 iters), loss = 0.568691 I0410 00:04:57.433351 6396 solver.cpp:237] Train net output #0: loss = 0.568691 (* 1 = 0.568691 loss) I0410 00:04:57.433359 6396 sgd_solver.cpp:105] Iteration 7836, lr = 0.00211781 I0410 00:05:02.341286 6396 solver.cpp:218] Iteration 7848 (2.4451 iter/s, 4.90777s/12 iters), loss = 0.386934 I0410 00:05:02.341364 6396 solver.cpp:237] Train net output #0: loss = 0.386934 (* 1 = 0.386934 loss) I0410 00:05:02.341382 6396 sgd_solver.cpp:105] Iteration 7848, lr = 0.00211279 I0410 00:05:04.336887 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_7854.caffemodel I0410 00:05:05.173714 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_7854.solverstate I0410 00:05:05.602349 6396 solver.cpp:330] Iteration 7854, Testing net (#0) I0410 00:05:05.602365 6396 net.cpp:676] Ignoring source layer train-data I0410 00:05:06.991987 6401 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:05:10.059276 6396 solver.cpp:397] Test net output #0: accuracy = 0.576593 I0410 00:05:10.059325 6396 solver.cpp:397] Test net output #1: loss = 1.87098 (* 1 = 1.87098 loss) I0410 00:05:12.008946 6396 solver.cpp:218] Iteration 7860 (1.2413 iter/s, 9.66729s/12 iters), loss = 0.517304 I0410 00:05:12.008998 6396 solver.cpp:237] Train net output #0: loss = 0.517304 (* 1 = 0.517304 loss) I0410 00:05:12.009009 6396 sgd_solver.cpp:105] Iteration 7860, lr = 0.00210777 I0410 00:05:16.989269 6396 solver.cpp:218] Iteration 7872 (2.40958 iter/s, 4.98011s/12 iters), loss = 0.407837 I0410 00:05:16.989320 6396 solver.cpp:237] Train net output #0: loss = 0.407837 (* 1 = 0.407837 loss) I0410 00:05:16.989329 6396 sgd_solver.cpp:105] Iteration 7872, lr = 0.00210277 I0410 00:05:21.990563 6396 solver.cpp:218] Iteration 7884 (2.39948 iter/s, 5.00109s/12 iters), loss = 0.397466 I0410 00:05:21.990684 6396 solver.cpp:237] Train net output #0: loss = 0.397466 (* 1 = 0.397466 loss) I0410 00:05:21.990697 6396 sgd_solver.cpp:105] Iteration 7884, lr = 0.00209777 I0410 00:05:24.092942 6400 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:05:26.887303 6396 solver.cpp:218] Iteration 7896 (2.45074 iter/s, 4.89647s/12 iters), loss = 0.531649 I0410 00:05:26.887343 6396 solver.cpp:237] Train net output #0: loss = 0.531649 (* 1 = 0.531649 loss) I0410 00:05:26.887351 6396 sgd_solver.cpp:105] Iteration 7896, lr = 0.00209279 I0410 00:05:31.806394 6396 solver.cpp:218] Iteration 7908 (2.43958 iter/s, 4.91889s/12 iters), loss = 0.461568 I0410 00:05:31.806452 6396 solver.cpp:237] Train net output #0: loss = 0.461568 (* 1 = 0.461568 loss) I0410 00:05:31.806464 6396 sgd_solver.cpp:105] Iteration 7908, lr = 0.00208782 I0410 00:05:36.749994 6396 solver.cpp:218] Iteration 7920 (2.4275 iter/s, 4.94336s/12 iters), loss = 0.601116 I0410 00:05:36.750047 6396 solver.cpp:237] Train net output #0: loss = 0.601116 (* 1 = 0.601116 loss) I0410 00:05:36.750057 6396 sgd_solver.cpp:105] Iteration 7920, lr = 0.00208287 I0410 00:05:41.777727 6396 solver.cpp:218] Iteration 7932 (2.38686 iter/s, 5.02752s/12 iters), loss = 0.607718 I0410 00:05:41.777774 6396 solver.cpp:237] Train net output #0: loss = 0.607718 (* 1 = 0.607718 loss) I0410 00:05:41.777786 6396 sgd_solver.cpp:105] Iteration 7932, lr = 0.00207792 I0410 00:05:46.912987 6396 solver.cpp:218] Iteration 7944 (2.33688 iter/s, 5.13505s/12 iters), loss = 0.471929 I0410 00:05:46.913033 6396 solver.cpp:237] Train net output #0: loss = 0.471929 (* 1 = 0.471929 loss) I0410 00:05:46.913043 6396 sgd_solver.cpp:105] Iteration 7944, lr = 0.00207299 I0410 00:05:51.336571 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_7956.caffemodel I0410 00:05:52.061283 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_7956.solverstate I0410 00:05:53.704035 6396 solver.cpp:330] Iteration 7956, Testing net (#0) I0410 00:05:53.704066 6396 net.cpp:676] Ignoring source layer train-data I0410 00:05:54.981454 6401 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:05:58.199651 6396 solver.cpp:397] Test net output #0: accuracy = 0.56924 I0410 00:05:58.199681 6396 solver.cpp:397] Test net output #1: loss = 1.91525 (* 1 = 1.91525 loss) I0410 00:05:58.282640 6396 solver.cpp:218] Iteration 7956 (1.05548 iter/s, 11.3693s/12 iters), loss = 0.396886 I0410 00:05:58.282687 6396 solver.cpp:237] Train net output #0: loss = 0.396886 (* 1 = 0.396886 loss) I0410 00:05:58.282696 6396 sgd_solver.cpp:105] Iteration 7956, lr = 0.00206807 I0410 00:06:02.651715 6396 solver.cpp:218] Iteration 7968 (2.74669 iter/s, 4.36889s/12 iters), loss = 0.460044 I0410 00:06:02.651757 6396 solver.cpp:237] Train net output #0: loss = 0.460044 (* 1 = 0.460044 loss) I0410 00:06:02.651764 6396 sgd_solver.cpp:105] Iteration 7968, lr = 0.00206316 I0410 00:06:07.609383 6396 solver.cpp:218] Iteration 7980 (2.42059 iter/s, 4.95747s/12 iters), loss = 0.480798 I0410 00:06:07.609441 6396 solver.cpp:237] Train net output #0: loss = 0.480798 (* 1 = 0.480798 loss) I0410 00:06:07.609454 6396 sgd_solver.cpp:105] Iteration 7980, lr = 0.00205826 I0410 00:06:11.849902 6400 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:06:12.553833 6396 solver.cpp:218] Iteration 7992 (2.42707 iter/s, 4.94424s/12 iters), loss = 0.388416 I0410 00:06:12.553879 6396 solver.cpp:237] Train net output #0: loss = 0.388416 (* 1 = 0.388416 loss) I0410 00:06:12.553889 6396 sgd_solver.cpp:105] Iteration 7992, lr = 0.00205337 I0410 00:06:17.627029 6396 solver.cpp:218] Iteration 8004 (2.36547 iter/s, 5.07299s/12 iters), loss = 0.677141 I0410 00:06:17.627084 6396 solver.cpp:237] Train net output #0: loss = 0.677141 (* 1 = 0.677141 loss) I0410 00:06:17.627095 6396 sgd_solver.cpp:105] Iteration 8004, lr = 0.0020485 I0410 00:06:22.614588 6396 solver.cpp:218] Iteration 8016 (2.40609 iter/s, 4.98735s/12 iters), loss = 0.582079 I0410 00:06:22.614729 6396 solver.cpp:237] Train net output #0: loss = 0.582079 (* 1 = 0.582079 loss) I0410 00:06:22.614737 6396 sgd_solver.cpp:105] Iteration 8016, lr = 0.00204363 I0410 00:06:27.546828 6396 solver.cpp:218] Iteration 8028 (2.43312 iter/s, 4.93194s/12 iters), loss = 0.365035 I0410 00:06:27.546882 6396 solver.cpp:237] Train net output #0: loss = 0.365035 (* 1 = 0.365035 loss) I0410 00:06:27.546895 6396 sgd_solver.cpp:105] Iteration 8028, lr = 0.00203878 I0410 00:06:32.891311 6396 solver.cpp:218] Iteration 8040 (2.2454 iter/s, 5.34426s/12 iters), loss = 0.678818 I0410 00:06:32.891363 6396 solver.cpp:237] Train net output #0: loss = 0.678818 (* 1 = 0.678818 loss) I0410 00:06:32.891376 6396 sgd_solver.cpp:105] Iteration 8040, lr = 0.00203394 I0410 00:06:38.097669 6396 solver.cpp:218] Iteration 8052 (2.30497 iter/s, 5.20614s/12 iters), loss = 0.393924 I0410 00:06:38.097726 6396 solver.cpp:237] Train net output #0: loss = 0.393924 (* 1 = 0.393924 loss) I0410 00:06:38.097739 6396 sgd_solver.cpp:105] Iteration 8052, lr = 0.00202911 I0410 00:06:40.126593 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_8058.caffemodel I0410 00:06:40.550215 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_8058.solverstate I0410 00:06:40.850068 6396 solver.cpp:330] Iteration 8058, Testing net (#0) I0410 00:06:40.850093 6396 net.cpp:676] Ignoring source layer train-data I0410 00:06:42.098682 6401 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:06:45.245569 6396 solver.cpp:397] Test net output #0: accuracy = 0.570466 I0410 00:06:45.245607 6396 solver.cpp:397] Test net output #1: loss = 1.95526 (* 1 = 1.95526 loss) I0410 00:06:46.986936 6396 solver.cpp:218] Iteration 8064 (1.34999 iter/s, 8.88894s/12 iters), loss = 0.530227 I0410 00:06:46.986984 6396 solver.cpp:237] Train net output #0: loss = 0.530227 (* 1 = 0.530227 loss) I0410 00:06:46.986994 6396 sgd_solver.cpp:105] Iteration 8064, lr = 0.00202429 I0410 00:06:51.831825 6396 solver.cpp:218] Iteration 8076 (2.47694 iter/s, 4.84469s/12 iters), loss = 0.349906 I0410 00:06:51.831863 6396 solver.cpp:237] Train net output #0: loss = 0.349906 (* 1 = 0.349906 loss) I0410 00:06:51.831872 6396 sgd_solver.cpp:105] Iteration 8076, lr = 0.00201949 I0410 00:06:56.723253 6396 solver.cpp:218] Iteration 8088 (2.45337 iter/s, 4.89123s/12 iters), loss = 0.387069 I0410 00:06:56.723352 6396 solver.cpp:237] Train net output #0: loss = 0.387069 (* 1 = 0.387069 loss) I0410 00:06:56.723362 6396 sgd_solver.cpp:105] Iteration 8088, lr = 0.00201469 I0410 00:06:58.095882 6400 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:07:01.628275 6396 solver.cpp:218] Iteration 8100 (2.4466 iter/s, 4.90477s/12 iters), loss = 0.52828 I0410 00:07:01.628326 6396 solver.cpp:237] Train net output #0: loss = 0.52828 (* 1 = 0.52828 loss) I0410 00:07:01.628338 6396 sgd_solver.cpp:105] Iteration 8100, lr = 0.00200991 I0410 00:07:06.515169 6396 solver.cpp:218] Iteration 8112 (2.45565 iter/s, 4.88668s/12 iters), loss = 0.42516 I0410 00:07:06.515228 6396 solver.cpp:237] Train net output #0: loss = 0.42516 (* 1 = 0.42516 loss) I0410 00:07:06.515239 6396 sgd_solver.cpp:105] Iteration 8112, lr = 0.00200514 I0410 00:07:11.500157 6396 solver.cpp:218] Iteration 8124 (2.40733 iter/s, 4.98477s/12 iters), loss = 0.299437 I0410 00:07:11.500216 6396 solver.cpp:237] Train net output #0: loss = 0.299437 (* 1 = 0.299437 loss) I0410 00:07:11.500227 6396 sgd_solver.cpp:105] Iteration 8124, lr = 0.00200038 I0410 00:07:16.427662 6396 solver.cpp:218] Iteration 8136 (2.43541 iter/s, 4.9273s/12 iters), loss = 0.40451 I0410 00:07:16.427706 6396 solver.cpp:237] Train net output #0: loss = 0.40451 (* 1 = 0.40451 loss) I0410 00:07:16.427716 6396 sgd_solver.cpp:105] Iteration 8136, lr = 0.00199563 I0410 00:07:21.508090 6396 solver.cpp:218] Iteration 8148 (2.3621 iter/s, 5.08022s/12 iters), loss = 0.402816 I0410 00:07:21.508142 6396 solver.cpp:237] Train net output #0: loss = 0.402816 (* 1 = 0.402816 loss) I0410 00:07:21.508152 6396 sgd_solver.cpp:105] Iteration 8148, lr = 0.00199089 I0410 00:07:26.063889 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_8160.caffemodel I0410 00:07:26.486358 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_8160.solverstate I0410 00:07:26.792093 6396 solver.cpp:330] Iteration 8160, Testing net (#0) I0410 00:07:26.792217 6396 net.cpp:676] Ignoring source layer train-data I0410 00:07:27.927893 6401 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:07:31.109035 6396 solver.cpp:397] Test net output #0: accuracy = 0.578431 I0410 00:07:31.109083 6396 solver.cpp:397] Test net output #1: loss = 1.93416 (* 1 = 1.93416 loss) I0410 00:07:31.192672 6396 solver.cpp:218] Iteration 8160 (1.23913 iter/s, 9.68423s/12 iters), loss = 0.41202 I0410 00:07:31.192744 6396 solver.cpp:237] Train net output #0: loss = 0.41202 (* 1 = 0.41202 loss) I0410 00:07:31.192759 6396 sgd_solver.cpp:105] Iteration 8160, lr = 0.00198616 I0410 00:07:35.306380 6396 solver.cpp:218] Iteration 8172 (2.91722 iter/s, 4.11351s/12 iters), loss = 0.569727 I0410 00:07:35.306433 6396 solver.cpp:237] Train net output #0: loss = 0.569727 (* 1 = 0.569727 loss) I0410 00:07:35.306447 6396 sgd_solver.cpp:105] Iteration 8172, lr = 0.00198145 I0410 00:07:40.173933 6396 solver.cpp:218] Iteration 8184 (2.46541 iter/s, 4.86734s/12 iters), loss = 0.415882 I0410 00:07:40.174013 6396 solver.cpp:237] Train net output #0: loss = 0.415882 (* 1 = 0.415882 loss) I0410 00:07:40.174026 6396 sgd_solver.cpp:105] Iteration 8184, lr = 0.00197674 I0410 00:07:43.628414 6400 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:07:45.047102 6396 solver.cpp:218] Iteration 8196 (2.46258 iter/s, 4.87293s/12 iters), loss = 0.475072 I0410 00:07:45.047168 6396 solver.cpp:237] Train net output #0: loss = 0.475072 (* 1 = 0.475072 loss) I0410 00:07:45.047179 6396 sgd_solver.cpp:105] Iteration 8196, lr = 0.00197205 I0410 00:07:49.932523 6396 solver.cpp:218] Iteration 8208 (2.4564 iter/s, 4.8852s/12 iters), loss = 0.542122 I0410 00:07:49.932581 6396 solver.cpp:237] Train net output #0: loss = 0.542122 (* 1 = 0.542122 loss) I0410 00:07:49.932595 6396 sgd_solver.cpp:105] Iteration 8208, lr = 0.00196737 I0410 00:07:54.753012 6396 solver.cpp:218] Iteration 8220 (2.48948 iter/s, 4.82028s/12 iters), loss = 0.58407 I0410 00:07:54.753070 6396 solver.cpp:237] Train net output #0: loss = 0.58407 (* 1 = 0.58407 loss) I0410 00:07:54.753082 6396 sgd_solver.cpp:105] Iteration 8220, lr = 0.0019627 I0410 00:07:59.602155 6396 solver.cpp:218] Iteration 8232 (2.47477 iter/s, 4.84893s/12 iters), loss = 0.432641 I0410 00:07:59.602241 6396 solver.cpp:237] Train net output #0: loss = 0.432641 (* 1 = 0.432641 loss) I0410 00:07:59.602252 6396 sgd_solver.cpp:105] Iteration 8232, lr = 0.00195804 I0410 00:08:04.655858 6396 solver.cpp:218] Iteration 8244 (2.37461 iter/s, 5.05345s/12 iters), loss = 0.564514 I0410 00:08:04.655915 6396 solver.cpp:237] Train net output #0: loss = 0.564514 (* 1 = 0.564514 loss) I0410 00:08:04.655927 6396 sgd_solver.cpp:105] Iteration 8244, lr = 0.00195339 I0410 00:08:09.681438 6396 solver.cpp:218] Iteration 8256 (2.38789 iter/s, 5.02536s/12 iters), loss = 0.337274 I0410 00:08:09.681495 6396 solver.cpp:237] Train net output #0: loss = 0.337274 (* 1 = 0.337274 loss) I0410 00:08:09.681509 6396 sgd_solver.cpp:105] Iteration 8256, lr = 0.00194875 I0410 00:08:11.673821 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_8262.caffemodel I0410 00:08:12.142056 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_8262.solverstate I0410 00:08:12.463647 6396 solver.cpp:330] Iteration 8262, Testing net (#0) I0410 00:08:12.463675 6396 net.cpp:676] Ignoring source layer train-data I0410 00:08:13.652036 6401 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:08:16.866932 6396 solver.cpp:397] Test net output #0: accuracy = 0.572917 I0410 00:08:16.866982 6396 solver.cpp:397] Test net output #1: loss = 1.99994 (* 1 = 1.99994 loss) I0410 00:08:18.730360 6396 solver.cpp:218] Iteration 8268 (1.32617 iter/s, 9.04859s/12 iters), loss = 0.423492 I0410 00:08:18.730407 6396 solver.cpp:237] Train net output #0: loss = 0.423492 (* 1 = 0.423492 loss) I0410 00:08:18.730417 6396 sgd_solver.cpp:105] Iteration 8268, lr = 0.00194412 I0410 00:08:23.620959 6396 solver.cpp:218] Iteration 8280 (2.45379 iter/s, 4.89039s/12 iters), loss = 0.605371 I0410 00:08:23.621019 6396 solver.cpp:237] Train net output #0: loss = 0.605371 (* 1 = 0.605371 loss) I0410 00:08:23.621031 6396 sgd_solver.cpp:105] Iteration 8280, lr = 0.00193951 I0410 00:08:28.536170 6396 solver.cpp:218] Iteration 8292 (2.44151 iter/s, 4.915s/12 iters), loss = 0.469845 I0410 00:08:28.536216 6396 solver.cpp:237] Train net output #0: loss = 0.469845 (* 1 = 0.469845 loss) I0410 00:08:28.536226 6396 sgd_solver.cpp:105] Iteration 8292, lr = 0.0019349 I0410 00:08:29.224299 6400 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:08:33.470839 6396 solver.cpp:218] Iteration 8304 (2.43188 iter/s, 4.93446s/12 iters), loss = 0.38628 I0410 00:08:33.470939 6396 solver.cpp:237] Train net output #0: loss = 0.38628 (* 1 = 0.38628 loss) I0410 00:08:33.470949 6396 sgd_solver.cpp:105] Iteration 8304, lr = 0.00193031 I0410 00:08:34.640234 6396 blocking_queue.cpp:49] Waiting for data I0410 00:08:38.357290 6396 solver.cpp:218] Iteration 8316 (2.4559 iter/s, 4.8862s/12 iters), loss = 0.365104 I0410 00:08:38.357334 6396 solver.cpp:237] Train net output #0: loss = 0.365104 (* 1 = 0.365104 loss) I0410 00:08:38.357347 6396 sgd_solver.cpp:105] Iteration 8316, lr = 0.00192573 I0410 00:08:43.296741 6396 solver.cpp:218] Iteration 8328 (2.42952 iter/s, 4.93925s/12 iters), loss = 0.637254 I0410 00:08:43.296790 6396 solver.cpp:237] Train net output #0: loss = 0.637254 (* 1 = 0.637254 loss) I0410 00:08:43.296799 6396 sgd_solver.cpp:105] Iteration 8328, lr = 0.00192115 I0410 00:08:48.216491 6396 solver.cpp:218] Iteration 8340 (2.43925 iter/s, 4.91954s/12 iters), loss = 0.428658 I0410 00:08:48.216545 6396 solver.cpp:237] Train net output #0: loss = 0.428658 (* 1 = 0.428658 loss) I0410 00:08:48.216558 6396 sgd_solver.cpp:105] Iteration 8340, lr = 0.00191659 I0410 00:08:53.117655 6396 solver.cpp:218] Iteration 8352 (2.4485 iter/s, 4.90096s/12 iters), loss = 0.450404 I0410 00:08:53.117712 6396 solver.cpp:237] Train net output #0: loss = 0.450404 (* 1 = 0.450404 loss) I0410 00:08:53.117724 6396 sgd_solver.cpp:105] Iteration 8352, lr = 0.00191204 I0410 00:08:57.579622 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_8364.caffemodel I0410 00:08:58.063048 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_8364.solverstate I0410 00:08:58.382337 6396 solver.cpp:330] Iteration 8364, Testing net (#0) I0410 00:08:58.382367 6396 net.cpp:676] Ignoring source layer train-data I0410 00:08:59.563184 6401 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:09:02.927919 6396 solver.cpp:397] Test net output #0: accuracy = 0.585172 I0410 00:09:02.927961 6396 solver.cpp:397] Test net output #1: loss = 1.90502 (* 1 = 1.90502 loss) I0410 00:09:03.011157 6396 solver.cpp:218] Iteration 8364 (1.21296 iter/s, 9.89315s/12 iters), loss = 0.521031 I0410 00:09:03.011206 6396 solver.cpp:237] Train net output #0: loss = 0.521031 (* 1 = 0.521031 loss) I0410 00:09:03.011216 6396 sgd_solver.cpp:105] Iteration 8364, lr = 0.0019075 I0410 00:09:07.050709 6396 solver.cpp:218] Iteration 8376 (2.97076 iter/s, 4.03937s/12 iters), loss = 0.335354 I0410 00:09:07.050810 6396 solver.cpp:237] Train net output #0: loss = 0.335354 (* 1 = 0.335354 loss) I0410 00:09:07.050819 6396 sgd_solver.cpp:105] Iteration 8376, lr = 0.00190297 I0410 00:09:11.979878 6396 solver.cpp:218] Iteration 8388 (2.43461 iter/s, 4.92891s/12 iters), loss = 0.356941 I0410 00:09:11.979928 6396 solver.cpp:237] Train net output #0: loss = 0.356941 (* 1 = 0.356941 loss) I0410 00:09:11.979938 6396 sgd_solver.cpp:105] Iteration 8388, lr = 0.00189846 I0410 00:09:14.718103 6400 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:09:16.896919 6396 solver.cpp:218] Iteration 8400 (2.44059 iter/s, 4.91684s/12 iters), loss = 0.364593 I0410 00:09:16.896970 6396 solver.cpp:237] Train net output #0: loss = 0.364593 (* 1 = 0.364593 loss) I0410 00:09:16.896981 6396 sgd_solver.cpp:105] Iteration 8400, lr = 0.00189395 I0410 00:09:22.045809 6396 solver.cpp:218] Iteration 8412 (2.33069 iter/s, 5.14868s/12 iters), loss = 0.316095 I0410 00:09:22.045855 6396 solver.cpp:237] Train net output #0: loss = 0.316095 (* 1 = 0.316095 loss) I0410 00:09:22.045867 6396 sgd_solver.cpp:105] Iteration 8412, lr = 0.00188945 I0410 00:09:26.958256 6396 solver.cpp:218] Iteration 8424 (2.44287 iter/s, 4.91224s/12 iters), loss = 0.470128 I0410 00:09:26.958309 6396 solver.cpp:237] Train net output #0: loss = 0.470128 (* 1 = 0.470128 loss) I0410 00:09:26.958321 6396 sgd_solver.cpp:105] Iteration 8424, lr = 0.00188497 I0410 00:09:31.806356 6396 solver.cpp:218] Iteration 8436 (2.4753 iter/s, 4.84789s/12 iters), loss = 0.463089 I0410 00:09:31.806414 6396 solver.cpp:237] Train net output #0: loss = 0.463089 (* 1 = 0.463089 loss) I0410 00:09:31.806427 6396 sgd_solver.cpp:105] Iteration 8436, lr = 0.00188049 I0410 00:09:36.698513 6396 solver.cpp:218] Iteration 8448 (2.45301 iter/s, 4.89194s/12 iters), loss = 0.660579 I0410 00:09:36.698567 6396 solver.cpp:237] Train net output #0: loss = 0.660579 (* 1 = 0.660579 loss) I0410 00:09:36.698580 6396 sgd_solver.cpp:105] Iteration 8448, lr = 0.00187603 I0410 00:09:41.638962 6396 solver.cpp:218] Iteration 8460 (2.42903 iter/s, 4.94024s/12 iters), loss = 0.374917 I0410 00:09:41.639122 6396 solver.cpp:237] Train net output #0: loss = 0.374917 (* 1 = 0.374917 loss) I0410 00:09:41.639135 6396 sgd_solver.cpp:105] Iteration 8460, lr = 0.00187157 I0410 00:09:43.655508 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_8466.caffemodel I0410 00:09:44.111865 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_8466.solverstate I0410 00:09:45.121467 6396 solver.cpp:330] Iteration 8466, Testing net (#0) I0410 00:09:45.121488 6396 net.cpp:676] Ignoring source layer train-data I0410 00:09:46.166056 6401 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:09:49.481554 6396 solver.cpp:397] Test net output #0: accuracy = 0.588848 I0410 00:09:49.481602 6396 solver.cpp:397] Test net output #1: loss = 1.90164 (* 1 = 1.90164 loss) I0410 00:09:51.409667 6396 solver.cpp:218] Iteration 8472 (1.22822 iter/s, 9.77025s/12 iters), loss = 0.308575 I0410 00:09:51.409723 6396 solver.cpp:237] Train net output #0: loss = 0.308575 (* 1 = 0.308575 loss) I0410 00:09:51.409735 6396 sgd_solver.cpp:105] Iteration 8472, lr = 0.00186713 I0410 00:09:56.413303 6396 solver.cpp:218] Iteration 8484 (2.39836 iter/s, 5.00342s/12 iters), loss = 0.445113 I0410 00:09:56.413354 6396 solver.cpp:237] Train net output #0: loss = 0.445113 (* 1 = 0.445113 loss) I0410 00:09:56.413365 6396 sgd_solver.cpp:105] Iteration 8484, lr = 0.0018627 I0410 00:10:01.316525 6396 solver.cpp:218] Iteration 8496 (2.44748 iter/s, 4.90301s/12 iters), loss = 0.485582 I0410 00:10:01.316584 6396 solver.cpp:237] Train net output #0: loss = 0.485582 (* 1 = 0.485582 loss) I0410 00:10:01.316596 6396 sgd_solver.cpp:105] Iteration 8496, lr = 0.00185827 I0410 00:10:01.364382 6400 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:10:06.241578 6396 solver.cpp:218] Iteration 8508 (2.43663 iter/s, 4.92484s/12 iters), loss = 0.371981 I0410 00:10:06.241634 6396 solver.cpp:237] Train net output #0: loss = 0.371981 (* 1 = 0.371981 loss) I0410 00:10:06.241647 6396 sgd_solver.cpp:105] Iteration 8508, lr = 0.00185386 I0410 00:10:11.158877 6396 solver.cpp:218] Iteration 8520 (2.44047 iter/s, 4.91709s/12 iters), loss = 0.349761 I0410 00:10:11.158929 6396 solver.cpp:237] Train net output #0: loss = 0.349761 (* 1 = 0.349761 loss) I0410 00:10:11.158941 6396 sgd_solver.cpp:105] Iteration 8520, lr = 0.00184946 I0410 00:10:16.084738 6396 solver.cpp:218] Iteration 8532 (2.43623 iter/s, 4.92565s/12 iters), loss = 0.386573 I0410 00:10:16.084949 6396 solver.cpp:237] Train net output #0: loss = 0.386573 (* 1 = 0.386573 loss) I0410 00:10:16.084962 6396 sgd_solver.cpp:105] Iteration 8532, lr = 0.00184507 I0410 00:10:20.992347 6396 solver.cpp:218] Iteration 8544 (2.44536 iter/s, 4.90725s/12 iters), loss = 0.493034 I0410 00:10:20.992385 6396 solver.cpp:237] Train net output #0: loss = 0.493034 (* 1 = 0.493034 loss) I0410 00:10:20.992394 6396 sgd_solver.cpp:105] Iteration 8544, lr = 0.00184069 I0410 00:10:25.903465 6396 solver.cpp:218] Iteration 8556 (2.44353 iter/s, 4.91093s/12 iters), loss = 0.317279 I0410 00:10:25.903502 6396 solver.cpp:237] Train net output #0: loss = 0.317279 (* 1 = 0.317279 loss) I0410 00:10:25.903510 6396 sgd_solver.cpp:105] Iteration 8556, lr = 0.00183632 I0410 00:10:30.378567 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_8568.caffemodel I0410 00:10:32.773603 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_8568.solverstate I0410 00:10:33.591302 6396 solver.cpp:330] Iteration 8568, Testing net (#0) I0410 00:10:33.591323 6396 net.cpp:676] Ignoring source layer train-data I0410 00:10:34.606254 6401 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:10:38.131099 6396 solver.cpp:397] Test net output #0: accuracy = 0.577819 I0410 00:10:38.131129 6396 solver.cpp:397] Test net output #1: loss = 1.97421 (* 1 = 1.97421 loss) I0410 00:10:38.214097 6396 solver.cpp:218] Iteration 8568 (0.974799 iter/s, 12.3102s/12 iters), loss = 0.369813 I0410 00:10:38.214143 6396 solver.cpp:237] Train net output #0: loss = 0.369813 (* 1 = 0.369813 loss) I0410 00:10:38.214152 6396 sgd_solver.cpp:105] Iteration 8568, lr = 0.00183196 I0410 00:10:42.288489 6396 solver.cpp:218] Iteration 8580 (2.94536 iter/s, 4.07421s/12 iters), loss = 0.315365 I0410 00:10:42.288547 6396 solver.cpp:237] Train net output #0: loss = 0.315365 (* 1 = 0.315365 loss) I0410 00:10:42.288560 6396 sgd_solver.cpp:105] Iteration 8580, lr = 0.00182761 I0410 00:10:47.401667 6396 solver.cpp:218] Iteration 8592 (2.34698 iter/s, 5.11296s/12 iters), loss = 0.342833 I0410 00:10:47.401762 6396 solver.cpp:237] Train net output #0: loss = 0.342833 (* 1 = 0.342833 loss) I0410 00:10:47.401773 6396 sgd_solver.cpp:105] Iteration 8592, lr = 0.00182327 I0410 00:10:49.580457 6400 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:10:52.403015 6396 solver.cpp:218] Iteration 8604 (2.39947 iter/s, 5.0011s/12 iters), loss = 0.491314 I0410 00:10:52.403055 6396 solver.cpp:237] Train net output #0: loss = 0.491314 (* 1 = 0.491314 loss) I0410 00:10:52.403064 6396 sgd_solver.cpp:105] Iteration 8604, lr = 0.00181894 I0410 00:10:57.495164 6396 solver.cpp:218] Iteration 8616 (2.35666 iter/s, 5.09194s/12 iters), loss = 0.332876 I0410 00:10:57.495216 6396 solver.cpp:237] Train net output #0: loss = 0.332876 (* 1 = 0.332876 loss) I0410 00:10:57.495227 6396 sgd_solver.cpp:105] Iteration 8616, lr = 0.00181462 I0410 00:11:02.376142 6396 solver.cpp:218] Iteration 8628 (2.45863 iter/s, 4.88077s/12 iters), loss = 0.430373 I0410 00:11:02.376183 6396 solver.cpp:237] Train net output #0: loss = 0.430373 (* 1 = 0.430373 loss) I0410 00:11:02.376190 6396 sgd_solver.cpp:105] Iteration 8628, lr = 0.00181031 I0410 00:11:07.421723 6396 solver.cpp:218] Iteration 8640 (2.37841 iter/s, 5.04538s/12 iters), loss = 0.384442 I0410 00:11:07.421770 6396 solver.cpp:237] Train net output #0: loss = 0.384442 (* 1 = 0.384442 loss) I0410 00:11:07.421780 6396 sgd_solver.cpp:105] Iteration 8640, lr = 0.00180602 I0410 00:11:12.428439 6396 solver.cpp:218] Iteration 8652 (2.39688 iter/s, 5.00651s/12 iters), loss = 0.344907 I0410 00:11:12.428494 6396 solver.cpp:237] Train net output #0: loss = 0.344907 (* 1 = 0.344907 loss) I0410 00:11:12.428508 6396 sgd_solver.cpp:105] Iteration 8652, lr = 0.00180173 I0410 00:11:17.351442 6396 solver.cpp:218] Iteration 8664 (2.43764 iter/s, 4.92279s/12 iters), loss = 0.272319 I0410 00:11:17.351491 6396 solver.cpp:237] Train net output #0: loss = 0.272319 (* 1 = 0.272319 loss) I0410 00:11:17.351502 6396 sgd_solver.cpp:105] Iteration 8664, lr = 0.00179745 I0410 00:11:19.367138 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_8670.caffemodel I0410 00:11:21.060982 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_8670.solverstate I0410 00:11:21.382299 6396 solver.cpp:330] Iteration 8670, Testing net (#0) I0410 00:11:21.382328 6396 net.cpp:676] Ignoring source layer train-data I0410 00:11:22.369428 6401 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:11:26.016113 6396 solver.cpp:397] Test net output #0: accuracy = 0.568627 I0410 00:11:26.016147 6396 solver.cpp:397] Test net output #1: loss = 2.03448 (* 1 = 2.03448 loss) I0410 00:11:27.864353 6396 solver.cpp:218] Iteration 8676 (1.14149 iter/s, 10.5125s/12 iters), loss = 0.451089 I0410 00:11:27.864406 6396 solver.cpp:237] Train net output #0: loss = 0.451089 (* 1 = 0.451089 loss) I0410 00:11:27.864418 6396 sgd_solver.cpp:105] Iteration 8676, lr = 0.00179318 I0410 00:11:32.870273 6396 solver.cpp:218] Iteration 8688 (2.39726 iter/s, 5.00572s/12 iters), loss = 0.595131 I0410 00:11:32.870321 6396 solver.cpp:237] Train net output #0: loss = 0.595131 (* 1 = 0.595131 loss) I0410 00:11:32.870332 6396 sgd_solver.cpp:105] Iteration 8688, lr = 0.00178893 I0410 00:11:37.148720 6400 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:11:37.816661 6396 solver.cpp:218] Iteration 8700 (2.42611 iter/s, 4.94618s/12 iters), loss = 0.42975 I0410 00:11:37.816718 6396 solver.cpp:237] Train net output #0: loss = 0.42975 (* 1 = 0.42975 loss) I0410 00:11:37.816731 6396 sgd_solver.cpp:105] Iteration 8700, lr = 0.00178468 I0410 00:11:42.723575 6396 solver.cpp:218] Iteration 8712 (2.44563 iter/s, 4.90671s/12 iters), loss = 0.596835 I0410 00:11:42.723621 6396 solver.cpp:237] Train net output #0: loss = 0.596835 (* 1 = 0.596835 loss) I0410 00:11:42.723634 6396 sgd_solver.cpp:105] Iteration 8712, lr = 0.00178044 I0410 00:11:47.685669 6396 solver.cpp:218] Iteration 8724 (2.41843 iter/s, 4.96189s/12 iters), loss = 0.445356 I0410 00:11:47.685724 6396 solver.cpp:237] Train net output #0: loss = 0.445356 (* 1 = 0.445356 loss) I0410 00:11:47.685736 6396 sgd_solver.cpp:105] Iteration 8724, lr = 0.00177621 I0410 00:11:52.632117 6396 solver.cpp:218] Iteration 8736 (2.42609 iter/s, 4.94623s/12 iters), loss = 0.380434 I0410 00:11:52.632239 6396 solver.cpp:237] Train net output #0: loss = 0.380434 (* 1 = 0.380434 loss) I0410 00:11:52.632252 6396 sgd_solver.cpp:105] Iteration 8736, lr = 0.001772 I0410 00:11:57.626664 6396 solver.cpp:218] Iteration 8748 (2.40275 iter/s, 4.99427s/12 iters), loss = 0.335815 I0410 00:11:57.626718 6396 solver.cpp:237] Train net output #0: loss = 0.335815 (* 1 = 0.335815 loss) I0410 00:11:57.626729 6396 sgd_solver.cpp:105] Iteration 8748, lr = 0.00176779 I0410 00:12:02.565513 6396 solver.cpp:218] Iteration 8760 (2.42982 iter/s, 4.93864s/12 iters), loss = 0.382096 I0410 00:12:02.565570 6396 solver.cpp:237] Train net output #0: loss = 0.382096 (* 1 = 0.382096 loss) I0410 00:12:02.565580 6396 sgd_solver.cpp:105] Iteration 8760, lr = 0.00176359 I0410 00:12:07.055663 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_8772.caffemodel I0410 00:12:07.892357 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_8772.solverstate I0410 00:12:08.303040 6396 solver.cpp:330] Iteration 8772, Testing net (#0) I0410 00:12:08.303057 6396 net.cpp:676] Ignoring source layer train-data I0410 00:12:09.230808 6401 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:12:12.810276 6396 solver.cpp:397] Test net output #0: accuracy = 0.587623 I0410 00:12:12.810304 6396 solver.cpp:397] Test net output #1: loss = 1.90743 (* 1 = 1.90743 loss) I0410 00:12:12.893365 6396 solver.cpp:218] Iteration 8772 (1.16195 iter/s, 10.3275s/12 iters), loss = 0.346937 I0410 00:12:12.893406 6396 solver.cpp:237] Train net output #0: loss = 0.346937 (* 1 = 0.346937 loss) I0410 00:12:12.893415 6396 sgd_solver.cpp:105] Iteration 8772, lr = 0.00175941 I0410 00:12:17.031539 6396 solver.cpp:218] Iteration 8784 (2.89995 iter/s, 4.138s/12 iters), loss = 0.301242 I0410 00:12:17.031584 6396 solver.cpp:237] Train net output #0: loss = 0.301242 (* 1 = 0.301242 loss) I0410 00:12:17.031594 6396 sgd_solver.cpp:105] Iteration 8784, lr = 0.00175523 I0410 00:12:21.998991 6396 solver.cpp:218] Iteration 8796 (2.41582 iter/s, 4.96726s/12 iters), loss = 0.416966 I0410 00:12:21.999034 6396 solver.cpp:237] Train net output #0: loss = 0.416966 (* 1 = 0.416966 loss) I0410 00:12:21.999043 6396 sgd_solver.cpp:105] Iteration 8796, lr = 0.00175106 I0410 00:12:23.428459 6400 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:12:26.948529 6396 solver.cpp:218] Iteration 8808 (2.42457 iter/s, 4.94933s/12 iters), loss = 0.467832 I0410 00:12:26.948571 6396 solver.cpp:237] Train net output #0: loss = 0.467832 (* 1 = 0.467832 loss) I0410 00:12:26.948580 6396 sgd_solver.cpp:105] Iteration 8808, lr = 0.0017469 I0410 00:12:31.862856 6396 solver.cpp:218] Iteration 8820 (2.44194 iter/s, 4.91413s/12 iters), loss = 0.314849 I0410 00:12:31.862905 6396 solver.cpp:237] Train net output #0: loss = 0.314849 (* 1 = 0.314849 loss) I0410 00:12:31.862915 6396 sgd_solver.cpp:105] Iteration 8820, lr = 0.00174276 I0410 00:12:36.790035 6396 solver.cpp:218] Iteration 8832 (2.43557 iter/s, 4.92698s/12 iters), loss = 0.343688 I0410 00:12:36.790091 6396 solver.cpp:237] Train net output #0: loss = 0.343688 (* 1 = 0.343688 loss) I0410 00:12:36.790104 6396 sgd_solver.cpp:105] Iteration 8832, lr = 0.00173862 I0410 00:12:41.673601 6396 solver.cpp:218] Iteration 8844 (2.45733 iter/s, 4.88336s/12 iters), loss = 0.389762 I0410 00:12:41.673658 6396 solver.cpp:237] Train net output #0: loss = 0.389762 (* 1 = 0.389762 loss) I0410 00:12:41.673671 6396 sgd_solver.cpp:105] Iteration 8844, lr = 0.00173449 I0410 00:12:46.568892 6396 solver.cpp:218] Iteration 8856 (2.45144 iter/s, 4.89508s/12 iters), loss = 0.35646 I0410 00:12:46.568948 6396 solver.cpp:237] Train net output #0: loss = 0.35646 (* 1 = 0.35646 loss) I0410 00:12:46.568959 6396 sgd_solver.cpp:105] Iteration 8856, lr = 0.00173037 I0410 00:12:51.506934 6396 solver.cpp:218] Iteration 8868 (2.43022 iter/s, 4.93783s/12 iters), loss = 0.321884 I0410 00:12:51.506995 6396 solver.cpp:237] Train net output #0: loss = 0.321884 (* 1 = 0.321884 loss) I0410 00:12:51.507007 6396 sgd_solver.cpp:105] Iteration 8868, lr = 0.00172626 I0410 00:12:53.517149 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_8874.caffemodel I0410 00:12:54.885995 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_8874.solverstate I0410 00:12:55.590562 6396 solver.cpp:330] Iteration 8874, Testing net (#0) I0410 00:12:55.590591 6396 net.cpp:676] Ignoring source layer train-data I0410 00:12:56.642469 6401 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:13:00.119251 6396 solver.cpp:397] Test net output #0: accuracy = 0.57598 I0410 00:13:00.119302 6396 solver.cpp:397] Test net output #1: loss = 2.02685 (* 1 = 2.02685 loss) I0410 00:13:01.908854 6396 solver.cpp:218] Iteration 8880 (1.15367 iter/s, 10.4016s/12 iters), loss = 0.344603 I0410 00:13:01.908916 6396 solver.cpp:237] Train net output #0: loss = 0.344603 (* 1 = 0.344603 loss) I0410 00:13:01.908928 6396 sgd_solver.cpp:105] Iteration 8880, lr = 0.00172217 I0410 00:13:06.919478 6396 solver.cpp:218] Iteration 8892 (2.39501 iter/s, 5.01041s/12 iters), loss = 0.404381 I0410 00:13:06.919523 6396 solver.cpp:237] Train net output #0: loss = 0.404381 (* 1 = 0.404381 loss) I0410 00:13:06.919531 6396 sgd_solver.cpp:105] Iteration 8892, lr = 0.00171808 I0410 00:13:10.543962 6400 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:13:11.954643 6396 solver.cpp:218] Iteration 8904 (2.38334 iter/s, 5.03496s/12 iters), loss = 0.293729 I0410 00:13:11.954689 6396 solver.cpp:237] Train net output #0: loss = 0.293729 (* 1 = 0.293729 loss) I0410 00:13:11.954697 6396 sgd_solver.cpp:105] Iteration 8904, lr = 0.001714 I0410 00:13:16.922925 6396 solver.cpp:218] Iteration 8916 (2.41542 iter/s, 4.96808s/12 iters), loss = 0.352945 I0410 00:13:16.922976 6396 solver.cpp:237] Train net output #0: loss = 0.352945 (* 1 = 0.352945 loss) I0410 00:13:16.922987 6396 sgd_solver.cpp:105] Iteration 8916, lr = 0.00170993 I0410 00:13:21.851711 6396 solver.cpp:218] Iteration 8928 (2.43478 iter/s, 4.92858s/12 iters), loss = 0.331451 I0410 00:13:21.851769 6396 solver.cpp:237] Train net output #0: loss = 0.331451 (* 1 = 0.331451 loss) I0410 00:13:21.851781 6396 sgd_solver.cpp:105] Iteration 8928, lr = 0.00170587 I0410 00:13:26.756417 6396 solver.cpp:218] Iteration 8940 (2.44674 iter/s, 4.90449s/12 iters), loss = 0.471949 I0410 00:13:26.756558 6396 solver.cpp:237] Train net output #0: loss = 0.471949 (* 1 = 0.471949 loss) I0410 00:13:26.756569 6396 sgd_solver.cpp:105] Iteration 8940, lr = 0.00170182 I0410 00:13:31.672936 6396 solver.cpp:218] Iteration 8952 (2.4409 iter/s, 4.91622s/12 iters), loss = 0.556343 I0410 00:13:31.672996 6396 solver.cpp:237] Train net output #0: loss = 0.556343 (* 1 = 0.556343 loss) I0410 00:13:31.673007 6396 sgd_solver.cpp:105] Iteration 8952, lr = 0.00169778 I0410 00:13:36.544576 6396 solver.cpp:218] Iteration 8964 (2.46334 iter/s, 4.87143s/12 iters), loss = 0.367866 I0410 00:13:36.544631 6396 solver.cpp:237] Train net output #0: loss = 0.367866 (* 1 = 0.367866 loss) I0410 00:13:36.544643 6396 sgd_solver.cpp:105] Iteration 8964, lr = 0.00169375 I0410 00:13:41.015273 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_8976.caffemodel I0410 00:13:41.868012 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_8976.solverstate I0410 00:13:42.598122 6396 solver.cpp:330] Iteration 8976, Testing net (#0) I0410 00:13:42.598150 6396 net.cpp:676] Ignoring source layer train-data I0410 00:13:43.552790 6401 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:13:47.135288 6396 solver.cpp:397] Test net output #0: accuracy = 0.585784 I0410 00:13:47.135327 6396 solver.cpp:397] Test net output #1: loss = 1.91383 (* 1 = 1.91383 loss) I0410 00:13:47.218454 6396 solver.cpp:218] Iteration 8976 (1.12428 iter/s, 10.6735s/12 iters), loss = 0.342751 I0410 00:13:47.218514 6396 solver.cpp:237] Train net output #0: loss = 0.342751 (* 1 = 0.342751 loss) I0410 00:13:47.218526 6396 sgd_solver.cpp:105] Iteration 8976, lr = 0.00168973 I0410 00:13:51.514147 6396 solver.cpp:218] Iteration 8988 (2.79362 iter/s, 4.2955s/12 iters), loss = 0.242347 I0410 00:13:51.514199 6396 solver.cpp:237] Train net output #0: loss = 0.242347 (* 1 = 0.242347 loss) I0410 00:13:51.514209 6396 sgd_solver.cpp:105] Iteration 8988, lr = 0.00168571 I0410 00:13:53.123708 6396 blocking_queue.cpp:49] Waiting for data I0410 00:13:56.462961 6396 solver.cpp:218] Iteration 9000 (2.42493 iter/s, 4.94861s/12 iters), loss = 0.338435 I0410 00:13:56.463004 6396 solver.cpp:237] Train net output #0: loss = 0.338435 (* 1 = 0.338435 loss) I0410 00:13:56.463014 6396 sgd_solver.cpp:105] Iteration 9000, lr = 0.00168171 I0410 00:13:57.161679 6400 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:14:01.360677 6396 solver.cpp:218] Iteration 9012 (2.45022 iter/s, 4.89751s/12 iters), loss = 0.368831 I0410 00:14:01.360738 6396 solver.cpp:237] Train net output #0: loss = 0.368831 (* 1 = 0.368831 loss) I0410 00:14:01.360751 6396 sgd_solver.cpp:105] Iteration 9012, lr = 0.00167772 I0410 00:14:06.342826 6396 solver.cpp:218] Iteration 9024 (2.4087 iter/s, 4.98193s/12 iters), loss = 0.194407 I0410 00:14:06.342869 6396 solver.cpp:237] Train net output #0: loss = 0.194407 (* 1 = 0.194407 loss) I0410 00:14:06.342880 6396 sgd_solver.cpp:105] Iteration 9024, lr = 0.00167374 I0410 00:14:11.289824 6396 solver.cpp:218] Iteration 9036 (2.42581 iter/s, 4.94679s/12 iters), loss = 0.359875 I0410 00:14:11.289886 6396 solver.cpp:237] Train net output #0: loss = 0.359875 (* 1 = 0.359875 loss) I0410 00:14:11.289898 6396 sgd_solver.cpp:105] Iteration 9036, lr = 0.00166976 I0410 00:14:16.230299 6396 solver.cpp:218] Iteration 9048 (2.42902 iter/s, 4.94026s/12 iters), loss = 0.266714 I0410 00:14:16.230351 6396 solver.cpp:237] Train net output #0: loss = 0.266714 (* 1 = 0.266714 loss) I0410 00:14:16.230362 6396 sgd_solver.cpp:105] Iteration 9048, lr = 0.0016658 I0410 00:14:21.162751 6396 solver.cpp:218] Iteration 9060 (2.43297 iter/s, 4.93225s/12 iters), loss = 0.365277 I0410 00:14:21.162793 6396 solver.cpp:237] Train net output #0: loss = 0.365277 (* 1 = 0.365277 loss) I0410 00:14:21.162802 6396 sgd_solver.cpp:105] Iteration 9060, lr = 0.00166184 I0410 00:14:26.119132 6396 solver.cpp:218] Iteration 9072 (2.42122 iter/s, 4.95618s/12 iters), loss = 0.372175 I0410 00:14:26.119184 6396 solver.cpp:237] Train net output #0: loss = 0.372175 (* 1 = 0.372175 loss) I0410 00:14:26.119194 6396 sgd_solver.cpp:105] Iteration 9072, lr = 0.0016579 I0410 00:14:28.127213 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_9078.caffemodel I0410 00:14:28.547144 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_9078.solverstate I0410 00:14:28.851264 6396 solver.cpp:330] Iteration 9078, Testing net (#0) I0410 00:14:28.851291 6396 net.cpp:676] Ignoring source layer train-data I0410 00:14:29.658507 6401 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:14:33.374421 6396 solver.cpp:397] Test net output #0: accuracy = 0.591299 I0410 00:14:33.374462 6396 solver.cpp:397] Test net output #1: loss = 2.01067 (* 1 = 2.01067 loss) I0410 00:14:35.333117 6396 solver.cpp:218] Iteration 9084 (1.30241 iter/s, 9.21366s/12 iters), loss = 0.267445 I0410 00:14:35.333153 6396 solver.cpp:237] Train net output #0: loss = 0.267445 (* 1 = 0.267445 loss) I0410 00:14:35.333163 6396 sgd_solver.cpp:105] Iteration 9084, lr = 0.00165396 I0410 00:14:40.314560 6396 solver.cpp:218] Iteration 9096 (2.40903 iter/s, 4.98125s/12 iters), loss = 0.316134 I0410 00:14:40.314607 6396 solver.cpp:237] Train net output #0: loss = 0.316134 (* 1 = 0.316134 loss) I0410 00:14:40.314616 6396 sgd_solver.cpp:105] Iteration 9096, lr = 0.00165003 I0410 00:14:43.169747 6400 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:14:45.161088 6396 solver.cpp:218] Iteration 9108 (2.4761 iter/s, 4.84633s/12 iters), loss = 0.433043 I0410 00:14:45.161134 6396 solver.cpp:237] Train net output #0: loss = 0.433043 (* 1 = 0.433043 loss) I0410 00:14:45.161145 6396 sgd_solver.cpp:105] Iteration 9108, lr = 0.00164612 I0410 00:14:50.107589 6396 solver.cpp:218] Iteration 9120 (2.42605 iter/s, 4.9463s/12 iters), loss = 0.250071 I0410 00:14:50.107642 6396 solver.cpp:237] Train net output #0: loss = 0.250071 (* 1 = 0.250071 loss) I0410 00:14:50.107654 6396 sgd_solver.cpp:105] Iteration 9120, lr = 0.00164221 I0410 00:14:55.018559 6396 solver.cpp:218] Iteration 9132 (2.44361 iter/s, 4.91077s/12 iters), loss = 0.302604 I0410 00:14:55.018601 6396 solver.cpp:237] Train net output #0: loss = 0.302604 (* 1 = 0.302604 loss) I0410 00:14:55.018610 6396 sgd_solver.cpp:105] Iteration 9132, lr = 0.00163831 I0410 00:15:00.043408 6396 solver.cpp:218] Iteration 9144 (2.38822 iter/s, 5.02465s/12 iters), loss = 0.194635 I0410 00:15:00.043478 6396 solver.cpp:237] Train net output #0: loss = 0.194635 (* 1 = 0.194635 loss) I0410 00:15:00.043488 6396 sgd_solver.cpp:105] Iteration 9144, lr = 0.00163442 I0410 00:15:04.959128 6396 solver.cpp:218] Iteration 9156 (2.44126 iter/s, 4.91549s/12 iters), loss = 0.210477 I0410 00:15:04.959182 6396 solver.cpp:237] Train net output #0: loss = 0.210477 (* 1 = 0.210477 loss) I0410 00:15:04.959194 6396 sgd_solver.cpp:105] Iteration 9156, lr = 0.00163054 I0410 00:15:09.842293 6396 solver.cpp:218] Iteration 9168 (2.45753 iter/s, 4.88296s/12 iters), loss = 0.273947 I0410 00:15:09.842342 6396 solver.cpp:237] Train net output #0: loss = 0.273947 (* 1 = 0.273947 loss) I0410 00:15:09.842351 6396 sgd_solver.cpp:105] Iteration 9168, lr = 0.00162667 I0410 00:15:14.397244 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_9180.caffemodel I0410 00:15:15.402431 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_9180.solverstate I0410 00:15:16.000537 6396 solver.cpp:330] Iteration 9180, Testing net (#0) I0410 00:15:16.000561 6396 net.cpp:676] Ignoring source layer train-data I0410 00:15:16.811435 6401 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:15:20.419971 6396 solver.cpp:397] Test net output #0: accuracy = 0.582108 I0410 00:15:20.420011 6396 solver.cpp:397] Test net output #1: loss = 1.88264 (* 1 = 1.88264 loss) I0410 00:15:20.503207 6396 solver.cpp:218] Iteration 9180 (1.12565 iter/s, 10.6605s/12 iters), loss = 0.257854 I0410 00:15:20.503262 6396 solver.cpp:237] Train net output #0: loss = 0.257854 (* 1 = 0.257854 loss) I0410 00:15:20.503271 6396 sgd_solver.cpp:105] Iteration 9180, lr = 0.00162281 I0410 00:15:24.714336 6396 solver.cpp:218] Iteration 9192 (2.84972 iter/s, 4.21094s/12 iters), loss = 0.312886 I0410 00:15:24.714395 6396 solver.cpp:237] Train net output #0: loss = 0.312886 (* 1 = 0.312886 loss) I0410 00:15:24.714407 6396 sgd_solver.cpp:105] Iteration 9192, lr = 0.00161895 I0410 00:15:29.628618 6396 solver.cpp:218] Iteration 9204 (2.44197 iter/s, 4.91407s/12 iters), loss = 0.308744 I0410 00:15:29.628670 6396 solver.cpp:237] Train net output #0: loss = 0.308744 (* 1 = 0.308744 loss) I0410 00:15:29.628684 6396 sgd_solver.cpp:105] Iteration 9204, lr = 0.00161511 I0410 00:15:29.711163 6400 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:15:34.566426 6396 solver.cpp:218] Iteration 9216 (2.43033 iter/s, 4.9376s/12 iters), loss = 0.437619 I0410 00:15:34.566740 6396 solver.cpp:237] Train net output #0: loss = 0.437619 (* 1 = 0.437619 loss) I0410 00:15:34.566753 6396 sgd_solver.cpp:105] Iteration 9216, lr = 0.00161128 I0410 00:15:39.602350 6396 solver.cpp:218] Iteration 9228 (2.3831 iter/s, 5.03545s/12 iters), loss = 0.318563 I0410 00:15:39.602401 6396 solver.cpp:237] Train net output #0: loss = 0.318563 (* 1 = 0.318563 loss) I0410 00:15:39.602413 6396 sgd_solver.cpp:105] Iteration 9228, lr = 0.00160745 I0410 00:15:44.559433 6396 solver.cpp:218] Iteration 9240 (2.42088 iter/s, 4.95687s/12 iters), loss = 0.467073 I0410 00:15:44.559489 6396 solver.cpp:237] Train net output #0: loss = 0.467073 (* 1 = 0.467073 loss) I0410 00:15:44.559504 6396 sgd_solver.cpp:105] Iteration 9240, lr = 0.00160363 I0410 00:15:49.523311 6396 solver.cpp:218] Iteration 9252 (2.41757 iter/s, 4.96367s/12 iters), loss = 0.318748 I0410 00:15:49.523366 6396 solver.cpp:237] Train net output #0: loss = 0.318748 (* 1 = 0.318748 loss) I0410 00:15:49.523378 6396 sgd_solver.cpp:105] Iteration 9252, lr = 0.00159983 I0410 00:15:54.437443 6396 solver.cpp:218] Iteration 9264 (2.44204 iter/s, 4.91393s/12 iters), loss = 0.344712 I0410 00:15:54.437489 6396 solver.cpp:237] Train net output #0: loss = 0.344712 (* 1 = 0.344712 loss) I0410 00:15:54.437501 6396 sgd_solver.cpp:105] Iteration 9264, lr = 0.00159603 I0410 00:15:59.393448 6396 solver.cpp:218] Iteration 9276 (2.4214 iter/s, 4.9558s/12 iters), loss = 0.482375 I0410 00:15:59.393501 6396 solver.cpp:237] Train net output #0: loss = 0.482375 (* 1 = 0.482375 loss) I0410 00:15:59.393512 6396 sgd_solver.cpp:105] Iteration 9276, lr = 0.00159224 I0410 00:16:01.413583 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_9282.caffemodel I0410 00:16:01.867919 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_9282.solverstate I0410 00:16:02.183306 6396 solver.cpp:330] Iteration 9282, Testing net (#0) I0410 00:16:02.183331 6396 net.cpp:676] Ignoring source layer train-data I0410 00:16:02.902556 6401 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:16:06.624948 6396 solver.cpp:397] Test net output #0: accuracy = 0.590686 I0410 00:16:06.625185 6396 solver.cpp:397] Test net output #1: loss = 1.91906 (* 1 = 1.91906 loss) I0410 00:16:08.389098 6396 solver.cpp:218] Iteration 9288 (1.33403 iter/s, 8.99533s/12 iters), loss = 0.340156 I0410 00:16:08.389151 6396 solver.cpp:237] Train net output #0: loss = 0.340156 (* 1 = 0.340156 loss) I0410 00:16:08.389163 6396 sgd_solver.cpp:105] Iteration 9288, lr = 0.00158846 I0410 00:16:13.335496 6396 solver.cpp:218] Iteration 9300 (2.42611 iter/s, 4.94619s/12 iters), loss = 0.237741 I0410 00:16:13.335561 6396 solver.cpp:237] Train net output #0: loss = 0.237741 (* 1 = 0.237741 loss) I0410 00:16:13.335573 6396 sgd_solver.cpp:105] Iteration 9300, lr = 0.00158469 I0410 00:16:15.524364 6400 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:16:18.650324 6396 solver.cpp:218] Iteration 9312 (2.25793 iter/s, 5.3146s/12 iters), loss = 0.359171 I0410 00:16:18.650382 6396 solver.cpp:237] Train net output #0: loss = 0.359171 (* 1 = 0.359171 loss) I0410 00:16:18.650394 6396 sgd_solver.cpp:105] Iteration 9312, lr = 0.00158092 I0410 00:16:24.937219 6396 solver.cpp:218] Iteration 9324 (1.90881 iter/s, 6.28664s/12 iters), loss = 0.213887 I0410 00:16:24.937279 6396 solver.cpp:237] Train net output #0: loss = 0.213887 (* 1 = 0.213887 loss) I0410 00:16:24.937291 6396 sgd_solver.cpp:105] Iteration 9324, lr = 0.00157717 I0410 00:16:30.182960 6396 solver.cpp:218] Iteration 9336 (2.28767 iter/s, 5.24552s/12 iters), loss = 0.486827 I0410 00:16:30.183015 6396 solver.cpp:237] Train net output #0: loss = 0.486827 (* 1 = 0.486827 loss) I0410 00:16:30.183028 6396 sgd_solver.cpp:105] Iteration 9336, lr = 0.00157343 I0410 00:16:35.132249 6396 solver.cpp:218] Iteration 9348 (2.42469 iter/s, 4.94908s/12 iters), loss = 0.272722 I0410 00:16:35.132308 6396 solver.cpp:237] Train net output #0: loss = 0.272722 (* 1 = 0.272722 loss) I0410 00:16:35.132320 6396 sgd_solver.cpp:105] Iteration 9348, lr = 0.00156969 I0410 00:16:40.077342 6396 solver.cpp:218] Iteration 9360 (2.42675 iter/s, 4.94488s/12 iters), loss = 0.346792 I0410 00:16:40.077448 6396 solver.cpp:237] Train net output #0: loss = 0.346792 (* 1 = 0.346792 loss) I0410 00:16:40.077459 6396 sgd_solver.cpp:105] Iteration 9360, lr = 0.00156596 I0410 00:16:45.003368 6396 solver.cpp:218] Iteration 9372 (2.43617 iter/s, 4.92577s/12 iters), loss = 0.360427 I0410 00:16:45.003415 6396 solver.cpp:237] Train net output #0: loss = 0.360427 (* 1 = 0.360427 loss) I0410 00:16:45.003428 6396 sgd_solver.cpp:105] Iteration 9372, lr = 0.00156225 I0410 00:16:49.479207 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_9384.caffemodel I0410 00:16:50.104101 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_9384.solverstate I0410 00:16:50.824465 6396 solver.cpp:330] Iteration 9384, Testing net (#0) I0410 00:16:50.824483 6396 net.cpp:676] Ignoring source layer train-data I0410 00:16:51.572676 6401 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:16:55.244220 6396 solver.cpp:397] Test net output #0: accuracy = 0.584559 I0410 00:16:55.244264 6396 solver.cpp:397] Test net output #1: loss = 1.99377 (* 1 = 1.99377 loss) I0410 00:16:55.327518 6396 solver.cpp:218] Iteration 9384 (1.16236 iter/s, 10.3238s/12 iters), loss = 0.273328 I0410 00:16:55.327569 6396 solver.cpp:237] Train net output #0: loss = 0.273328 (* 1 = 0.273328 loss) I0410 00:16:55.327580 6396 sgd_solver.cpp:105] Iteration 9384, lr = 0.00155854 I0410 00:16:59.630846 6396 solver.cpp:218] Iteration 9396 (2.78866 iter/s, 4.30313s/12 iters), loss = 0.360437 I0410 00:16:59.630900 6396 solver.cpp:237] Train net output #0: loss = 0.360437 (* 1 = 0.360437 loss) I0410 00:16:59.630914 6396 sgd_solver.cpp:105] Iteration 9396, lr = 0.00155484 I0410 00:17:03.982820 6400 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:17:04.636047 6396 solver.cpp:218] Iteration 9408 (2.39761 iter/s, 5.00499s/12 iters), loss = 0.28216 I0410 00:17:04.636101 6396 solver.cpp:237] Train net output #0: loss = 0.28216 (* 1 = 0.28216 loss) I0410 00:17:04.636113 6396 sgd_solver.cpp:105] Iteration 9408, lr = 0.00155114 I0410 00:17:09.625209 6396 solver.cpp:218] Iteration 9420 (2.40531 iter/s, 4.98895s/12 iters), loss = 0.299242 I0410 00:17:09.625262 6396 solver.cpp:237] Train net output #0: loss = 0.299242 (* 1 = 0.299242 loss) I0410 00:17:09.625275 6396 sgd_solver.cpp:105] Iteration 9420, lr = 0.00154746 I0410 00:17:14.666551 6396 solver.cpp:218] Iteration 9432 (2.38042 iter/s, 5.04114s/12 iters), loss = 0.329691 I0410 00:17:14.666703 6396 solver.cpp:237] Train net output #0: loss = 0.329691 (* 1 = 0.329691 loss) I0410 00:17:14.666713 6396 sgd_solver.cpp:105] Iteration 9432, lr = 0.00154379 I0410 00:17:19.588968 6396 solver.cpp:218] Iteration 9444 (2.43798 iter/s, 4.92211s/12 iters), loss = 0.212146 I0410 00:17:19.589020 6396 solver.cpp:237] Train net output #0: loss = 0.212146 (* 1 = 0.212146 loss) I0410 00:17:19.589030 6396 sgd_solver.cpp:105] Iteration 9444, lr = 0.00154012 I0410 00:17:24.664875 6396 solver.cpp:218] Iteration 9456 (2.36421 iter/s, 5.0757s/12 iters), loss = 0.204075 I0410 00:17:24.664929 6396 solver.cpp:237] Train net output #0: loss = 0.204075 (* 1 = 0.204075 loss) I0410 00:17:24.664942 6396 sgd_solver.cpp:105] Iteration 9456, lr = 0.00153647 I0410 00:17:29.616412 6396 solver.cpp:218] Iteration 9468 (2.42359 iter/s, 4.95133s/12 iters), loss = 0.364914 I0410 00:17:29.616462 6396 solver.cpp:237] Train net output #0: loss = 0.364914 (* 1 = 0.364914 loss) I0410 00:17:29.616473 6396 sgd_solver.cpp:105] Iteration 9468, lr = 0.00153282 I0410 00:17:34.679483 6396 solver.cpp:218] Iteration 9480 (2.3702 iter/s, 5.06287s/12 iters), loss = 0.278094 I0410 00:17:34.679520 6396 solver.cpp:237] Train net output #0: loss = 0.278094 (* 1 = 0.278094 loss) I0410 00:17:34.679530 6396 sgd_solver.cpp:105] Iteration 9480, lr = 0.00152918 I0410 00:17:36.859290 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_9486.caffemodel I0410 00:17:37.727900 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_9486.solverstate I0410 00:17:38.412685 6396 solver.cpp:330] Iteration 9486, Testing net (#0) I0410 00:17:38.412710 6396 net.cpp:676] Ignoring source layer train-data I0410 00:17:39.182260 6401 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:17:43.119410 6396 solver.cpp:397] Test net output #0: accuracy = 0.602328 I0410 00:17:43.119457 6396 solver.cpp:397] Test net output #1: loss = 1.95104 (* 1 = 1.95104 loss) I0410 00:17:45.016870 6396 solver.cpp:218] Iteration 9492 (1.16087 iter/s, 10.337s/12 iters), loss = 0.498526 I0410 00:17:45.016952 6396 solver.cpp:237] Train net output #0: loss = 0.498526 (* 1 = 0.498526 loss) I0410 00:17:45.016964 6396 sgd_solver.cpp:105] Iteration 9492, lr = 0.00152555 I0410 00:17:49.908056 6396 solver.cpp:218] Iteration 9504 (2.45351 iter/s, 4.89095s/12 iters), loss = 0.254243 I0410 00:17:49.908099 6396 solver.cpp:237] Train net output #0: loss = 0.254243 (* 1 = 0.254243 loss) I0410 00:17:49.908108 6396 sgd_solver.cpp:105] Iteration 9504, lr = 0.00152193 I0410 00:17:51.392474 6400 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:17:54.925875 6396 solver.cpp:218] Iteration 9516 (2.39157 iter/s, 5.01762s/12 iters), loss = 0.269046 I0410 00:17:54.925923 6396 solver.cpp:237] Train net output #0: loss = 0.269046 (* 1 = 0.269046 loss) I0410 00:17:54.925935 6396 sgd_solver.cpp:105] Iteration 9516, lr = 0.00151831 I0410 00:17:59.950393 6396 solver.cpp:218] Iteration 9528 (2.38839 iter/s, 5.02431s/12 iters), loss = 0.397849 I0410 00:17:59.950438 6396 solver.cpp:237] Train net output #0: loss = 0.397849 (* 1 = 0.397849 loss) I0410 00:17:59.950446 6396 sgd_solver.cpp:105] Iteration 9528, lr = 0.00151471 I0410 00:18:04.936559 6396 solver.cpp:218] Iteration 9540 (2.40676 iter/s, 4.98596s/12 iters), loss = 0.287938 I0410 00:18:04.936617 6396 solver.cpp:237] Train net output #0: loss = 0.287938 (* 1 = 0.287938 loss) I0410 00:18:04.936630 6396 sgd_solver.cpp:105] Iteration 9540, lr = 0.00151111 I0410 00:18:09.840886 6396 solver.cpp:218] Iteration 9552 (2.44693 iter/s, 4.90411s/12 iters), loss = 0.236524 I0410 00:18:09.840942 6396 solver.cpp:237] Train net output #0: loss = 0.236524 (* 1 = 0.236524 loss) I0410 00:18:09.840956 6396 sgd_solver.cpp:105] Iteration 9552, lr = 0.00150752 I0410 00:18:14.774518 6396 solver.cpp:218] Iteration 9564 (2.43239 iter/s, 4.93342s/12 iters), loss = 0.369479 I0410 00:18:14.774566 6396 solver.cpp:237] Train net output #0: loss = 0.369479 (* 1 = 0.369479 loss) I0410 00:18:14.774577 6396 sgd_solver.cpp:105] Iteration 9564, lr = 0.00150395 I0410 00:18:19.771984 6396 solver.cpp:218] Iteration 9576 (2.40132 iter/s, 4.99726s/12 iters), loss = 0.233391 I0410 00:18:19.772119 6396 solver.cpp:237] Train net output #0: loss = 0.233391 (* 1 = 0.233391 loss) I0410 00:18:19.772131 6396 sgd_solver.cpp:105] Iteration 9576, lr = 0.00150037 I0410 00:18:24.300995 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_9588.caffemodel I0410 00:18:24.863373 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_9588.solverstate I0410 00:18:25.447408 6396 solver.cpp:330] Iteration 9588, Testing net (#0) I0410 00:18:25.447429 6396 net.cpp:676] Ignoring source layer train-data I0410 00:18:26.157671 6401 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:18:30.030586 6396 solver.cpp:397] Test net output #0: accuracy = 0.598652 I0410 00:18:30.030622 6396 solver.cpp:397] Test net output #1: loss = 1.82979 (* 1 = 1.82979 loss) I0410 00:18:30.113903 6396 solver.cpp:218] Iteration 9588 (1.16038 iter/s, 10.3415s/12 iters), loss = 0.235028 I0410 00:18:30.113952 6396 solver.cpp:237] Train net output #0: loss = 0.235028 (* 1 = 0.235028 loss) I0410 00:18:30.113977 6396 sgd_solver.cpp:105] Iteration 9588, lr = 0.00149681 I0410 00:18:34.318742 6396 solver.cpp:218] Iteration 9600 (2.85398 iter/s, 4.20466s/12 iters), loss = 0.23668 I0410 00:18:34.318786 6396 solver.cpp:237] Train net output #0: loss = 0.23668 (* 1 = 0.23668 loss) I0410 00:18:34.318799 6396 sgd_solver.cpp:105] Iteration 9600, lr = 0.00149326 I0410 00:18:37.883088 6400 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:18:39.272212 6396 solver.cpp:218] Iteration 9612 (2.42264 iter/s, 4.95327s/12 iters), loss = 0.168449 I0410 00:18:39.272260 6396 solver.cpp:237] Train net output #0: loss = 0.168449 (* 1 = 0.168449 loss) I0410 00:18:39.272271 6396 sgd_solver.cpp:105] Iteration 9612, lr = 0.00148971 I0410 00:18:44.378024 6396 solver.cpp:218] Iteration 9624 (2.35036 iter/s, 5.10561s/12 iters), loss = 0.196578 I0410 00:18:44.378075 6396 solver.cpp:237] Train net output #0: loss = 0.196578 (* 1 = 0.196578 loss) I0410 00:18:44.378087 6396 sgd_solver.cpp:105] Iteration 9624, lr = 0.00148618 I0410 00:18:49.325606 6396 solver.cpp:218] Iteration 9636 (2.42553 iter/s, 4.94738s/12 iters), loss = 0.191904 I0410 00:18:49.325655 6396 solver.cpp:237] Train net output #0: loss = 0.191904 (* 1 = 0.191904 loss) I0410 00:18:49.325667 6396 sgd_solver.cpp:105] Iteration 9636, lr = 0.00148265 I0410 00:18:54.191556 6396 solver.cpp:218] Iteration 9648 (2.46622 iter/s, 4.86574s/12 iters), loss = 0.434557 I0410 00:18:54.191648 6396 solver.cpp:237] Train net output #0: loss = 0.434557 (* 1 = 0.434557 loss) I0410 00:18:54.191658 6396 sgd_solver.cpp:105] Iteration 9648, lr = 0.00147913 I0410 00:18:59.151273 6396 solver.cpp:218] Iteration 9660 (2.41961 iter/s, 4.95947s/12 iters), loss = 0.402251 I0410 00:18:59.151326 6396 solver.cpp:237] Train net output #0: loss = 0.402251 (* 1 = 0.402251 loss) I0410 00:18:59.151340 6396 sgd_solver.cpp:105] Iteration 9660, lr = 0.00147562 I0410 00:19:04.135964 6396 solver.cpp:218] Iteration 9672 (2.40747 iter/s, 4.98448s/12 iters), loss = 0.268414 I0410 00:19:04.136018 6396 solver.cpp:237] Train net output #0: loss = 0.268414 (* 1 = 0.268414 loss) I0410 00:19:04.136030 6396 sgd_solver.cpp:105] Iteration 9672, lr = 0.00147211 I0410 00:19:09.149112 6396 solver.cpp:218] Iteration 9684 (2.39381 iter/s, 5.01294s/12 iters), loss = 0.132732 I0410 00:19:09.149168 6396 solver.cpp:237] Train net output #0: loss = 0.132732 (* 1 = 0.132732 loss) I0410 00:19:09.149178 6396 sgd_solver.cpp:105] Iteration 9684, lr = 0.00146862 I0410 00:19:11.203940 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_9690.caffemodel I0410 00:19:12.765493 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_9690.solverstate I0410 00:19:13.083618 6396 solver.cpp:330] Iteration 9690, Testing net (#0) I0410 00:19:13.083649 6396 net.cpp:676] Ignoring source layer train-data I0410 00:19:13.695060 6401 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:19:16.272696 6396 blocking_queue.cpp:49] Waiting for data I0410 00:19:17.612666 6396 solver.cpp:397] Test net output #0: accuracy = 0.593137 I0410 00:19:17.612704 6396 solver.cpp:397] Test net output #1: loss = 2.06587 (* 1 = 2.06587 loss) I0410 00:19:19.502679 6396 solver.cpp:218] Iteration 9696 (1.15906 iter/s, 10.3532s/12 iters), loss = 0.236343 I0410 00:19:19.502733 6396 solver.cpp:237] Train net output #0: loss = 0.236343 (* 1 = 0.236343 loss) I0410 00:19:19.502746 6396 sgd_solver.cpp:105] Iteration 9696, lr = 0.00146513 I0410 00:19:24.434227 6396 solver.cpp:218] Iteration 9708 (2.43341 iter/s, 4.93134s/12 iters), loss = 0.282385 I0410 00:19:24.434348 6396 solver.cpp:237] Train net output #0: loss = 0.282385 (* 1 = 0.282385 loss) I0410 00:19:24.434358 6396 sgd_solver.cpp:105] Iteration 9708, lr = 0.00146165 I0410 00:19:25.168294 6400 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:19:29.421656 6396 solver.cpp:218] Iteration 9720 (2.40618 iter/s, 4.98716s/12 iters), loss = 0.431054 I0410 00:19:29.421697 6396 solver.cpp:237] Train net output #0: loss = 0.431054 (* 1 = 0.431054 loss) I0410 00:19:29.421706 6396 sgd_solver.cpp:105] Iteration 9720, lr = 0.00145818 I0410 00:19:34.361536 6396 solver.cpp:218] Iteration 9732 (2.42931 iter/s, 4.93968s/12 iters), loss = 0.298937 I0410 00:19:34.361594 6396 solver.cpp:237] Train net output #0: loss = 0.298937 (* 1 = 0.298937 loss) I0410 00:19:34.361606 6396 sgd_solver.cpp:105] Iteration 9732, lr = 0.00145472 I0410 00:19:39.350464 6396 solver.cpp:218] Iteration 9744 (2.40543 iter/s, 4.98871s/12 iters), loss = 0.304827 I0410 00:19:39.350515 6396 solver.cpp:237] Train net output #0: loss = 0.304827 (* 1 = 0.304827 loss) I0410 00:19:39.350528 6396 sgd_solver.cpp:105] Iteration 9744, lr = 0.00145127 I0410 00:19:44.194856 6396 solver.cpp:218] Iteration 9756 (2.4772 iter/s, 4.84419s/12 iters), loss = 0.451593 I0410 00:19:44.194911 6396 solver.cpp:237] Train net output #0: loss = 0.451593 (* 1 = 0.451593 loss) I0410 00:19:44.194921 6396 sgd_solver.cpp:105] Iteration 9756, lr = 0.00144782 I0410 00:19:49.146050 6396 solver.cpp:218] Iteration 9768 (2.42376 iter/s, 4.95098s/12 iters), loss = 0.231429 I0410 00:19:49.146106 6396 solver.cpp:237] Train net output #0: loss = 0.231429 (* 1 = 0.231429 loss) I0410 00:19:49.146117 6396 sgd_solver.cpp:105] Iteration 9768, lr = 0.00144438 I0410 00:19:54.061326 6396 solver.cpp:218] Iteration 9780 (2.44147 iter/s, 4.91506s/12 iters), loss = 0.395176 I0410 00:19:54.061378 6396 solver.cpp:237] Train net output #0: loss = 0.395176 (* 1 = 0.395176 loss) I0410 00:19:54.061391 6396 sgd_solver.cpp:105] Iteration 9780, lr = 0.00144095 I0410 00:19:58.516482 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_9792.caffemodel I0410 00:19:58.968020 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_9792.solverstate I0410 00:19:59.295851 6396 solver.cpp:330] Iteration 9792, Testing net (#0) I0410 00:19:59.295877 6396 net.cpp:676] Ignoring source layer train-data I0410 00:19:59.900741 6401 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:20:03.786464 6396 solver.cpp:397] Test net output #0: accuracy = 0.599265 I0410 00:20:03.786511 6396 solver.cpp:397] Test net output #1: loss = 1.95589 (* 1 = 1.95589 loss) I0410 00:20:03.870143 6396 solver.cpp:218] Iteration 9792 (1.22343 iter/s, 9.80847s/12 iters), loss = 0.161826 I0410 00:20:03.870193 6396 solver.cpp:237] Train net output #0: loss = 0.161826 (* 1 = 0.161826 loss) I0410 00:20:03.870204 6396 sgd_solver.cpp:105] Iteration 9792, lr = 0.00143753 I0410 00:20:08.400173 6396 solver.cpp:218] Iteration 9804 (2.6491 iter/s, 4.52983s/12 iters), loss = 0.187464 I0410 00:20:08.400228 6396 solver.cpp:237] Train net output #0: loss = 0.187464 (* 1 = 0.187464 loss) I0410 00:20:08.400239 6396 sgd_solver.cpp:105] Iteration 9804, lr = 0.00143412 I0410 00:20:11.457254 6400 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:20:13.481290 6396 solver.cpp:218] Iteration 9816 (2.36179 iter/s, 5.0809s/12 iters), loss = 0.261944 I0410 00:20:13.481349 6396 solver.cpp:237] Train net output #0: loss = 0.261944 (* 1 = 0.261944 loss) I0410 00:20:13.481361 6396 sgd_solver.cpp:105] Iteration 9816, lr = 0.00143072 I0410 00:20:18.433984 6396 solver.cpp:218] Iteration 9828 (2.42303 iter/s, 4.95247s/12 iters), loss = 0.297139 I0410 00:20:18.434032 6396 solver.cpp:237] Train net output #0: loss = 0.297139 (* 1 = 0.297139 loss) I0410 00:20:18.434042 6396 sgd_solver.cpp:105] Iteration 9828, lr = 0.00142732 I0410 00:20:23.341804 6396 solver.cpp:218] Iteration 9840 (2.44518 iter/s, 4.90762s/12 iters), loss = 0.2925 I0410 00:20:23.341847 6396 solver.cpp:237] Train net output #0: loss = 0.2925 (* 1 = 0.2925 loss) I0410 00:20:23.341857 6396 sgd_solver.cpp:105] Iteration 9840, lr = 0.00142393 I0410 00:20:28.340561 6396 solver.cpp:218] Iteration 9852 (2.40069 iter/s, 4.99855s/12 iters), loss = 0.31005 I0410 00:20:28.340608 6396 solver.cpp:237] Train net output #0: loss = 0.31005 (* 1 = 0.31005 loss) I0410 00:20:28.340618 6396 sgd_solver.cpp:105] Iteration 9852, lr = 0.00142055 I0410 00:20:33.326498 6396 solver.cpp:218] Iteration 9864 (2.40687 iter/s, 4.98573s/12 iters), loss = 0.195311 I0410 00:20:33.326645 6396 solver.cpp:237] Train net output #0: loss = 0.195311 (* 1 = 0.195311 loss) I0410 00:20:33.326658 6396 sgd_solver.cpp:105] Iteration 9864, lr = 0.00141718 I0410 00:20:38.238315 6396 solver.cpp:218] Iteration 9876 (2.44324 iter/s, 4.91152s/12 iters), loss = 0.240621 I0410 00:20:38.238365 6396 solver.cpp:237] Train net output #0: loss = 0.240621 (* 1 = 0.240621 loss) I0410 00:20:38.238377 6396 sgd_solver.cpp:105] Iteration 9876, lr = 0.00141381 I0410 00:20:43.285228 6396 solver.cpp:218] Iteration 9888 (2.37779 iter/s, 5.04671s/12 iters), loss = 0.284928 I0410 00:20:43.285284 6396 solver.cpp:237] Train net output #0: loss = 0.284928 (* 1 = 0.284928 loss) I0410 00:20:43.285295 6396 sgd_solver.cpp:105] Iteration 9888, lr = 0.00141045 I0410 00:20:45.294121 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_9894.caffemodel I0410 00:20:46.172189 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_9894.solverstate I0410 00:20:46.900033 6396 solver.cpp:330] Iteration 9894, Testing net (#0) I0410 00:20:46.900063 6396 net.cpp:676] Ignoring source layer train-data I0410 00:20:47.468470 6401 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:20:51.342661 6396 solver.cpp:397] Test net output #0: accuracy = 0.599877 I0410 00:20:51.342701 6396 solver.cpp:397] Test net output #1: loss = 1.87582 (* 1 = 1.87582 loss) I0410 00:20:53.089365 6396 solver.cpp:218] Iteration 9900 (1.22402 iter/s, 9.80379s/12 iters), loss = 0.253212 I0410 00:20:53.089418 6396 solver.cpp:237] Train net output #0: loss = 0.253212 (* 1 = 0.253212 loss) I0410 00:20:53.089429 6396 sgd_solver.cpp:105] Iteration 9900, lr = 0.00140711 I0410 00:20:57.914072 6396 solver.cpp:218] Iteration 9912 (2.4873 iter/s, 4.8245s/12 iters), loss = 0.238367 I0410 00:20:57.914122 6396 solver.cpp:237] Train net output #0: loss = 0.238367 (* 1 = 0.238367 loss) I0410 00:20:57.914134 6396 sgd_solver.cpp:105] Iteration 9912, lr = 0.00140377 I0410 00:20:58.012249 6400 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:21:03.035797 6396 solver.cpp:218] Iteration 9924 (2.34306 iter/s, 5.12151s/12 iters), loss = 0.121326 I0410 00:21:03.035852 6396 solver.cpp:237] Train net output #0: loss = 0.121326 (* 1 = 0.121326 loss) I0410 00:21:03.035864 6396 sgd_solver.cpp:105] Iteration 9924, lr = 0.00140043 I0410 00:21:08.095734 6396 solver.cpp:218] Iteration 9936 (2.37167 iter/s, 5.05972s/12 iters), loss = 0.269581 I0410 00:21:08.095866 6396 solver.cpp:237] Train net output #0: loss = 0.269581 (* 1 = 0.269581 loss) I0410 00:21:08.095882 6396 sgd_solver.cpp:105] Iteration 9936, lr = 0.00139711 I0410 00:21:13.090895 6396 solver.cpp:218] Iteration 9948 (2.40246 iter/s, 4.99488s/12 iters), loss = 0.262196 I0410 00:21:13.090941 6396 solver.cpp:237] Train net output #0: loss = 0.262196 (* 1 = 0.262196 loss) I0410 00:21:13.090953 6396 sgd_solver.cpp:105] Iteration 9948, lr = 0.00139379 I0410 00:21:18.038066 6396 solver.cpp:218] Iteration 9960 (2.42573 iter/s, 4.94697s/12 iters), loss = 0.231219 I0410 00:21:18.038113 6396 solver.cpp:237] Train net output #0: loss = 0.231219 (* 1 = 0.231219 loss) I0410 00:21:18.038125 6396 sgd_solver.cpp:105] Iteration 9960, lr = 0.00139048 I0410 00:21:22.942950 6396 solver.cpp:218] Iteration 9972 (2.44664 iter/s, 4.90468s/12 iters), loss = 0.361502 I0410 00:21:22.942996 6396 solver.cpp:237] Train net output #0: loss = 0.361502 (* 1 = 0.361502 loss) I0410 00:21:22.943006 6396 sgd_solver.cpp:105] Iteration 9972, lr = 0.00138718 I0410 00:21:27.862638 6396 solver.cpp:218] Iteration 9984 (2.43928 iter/s, 4.91949s/12 iters), loss = 0.232692 I0410 00:21:27.862687 6396 solver.cpp:237] Train net output #0: loss = 0.232692 (* 1 = 0.232692 loss) I0410 00:21:27.862695 6396 sgd_solver.cpp:105] Iteration 9984, lr = 0.00138389 I0410 00:21:32.380268 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_9996.caffemodel I0410 00:21:33.394409 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_9996.solverstate I0410 00:21:33.717134 6396 solver.cpp:330] Iteration 9996, Testing net (#0) I0410 00:21:33.717162 6396 net.cpp:676] Ignoring source layer train-data I0410 00:21:34.295199 6401 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:21:38.235960 6396 solver.cpp:397] Test net output #0: accuracy = 0.598039 I0410 00:21:38.236095 6396 solver.cpp:397] Test net output #1: loss = 1.93357 (* 1 = 1.93357 loss) I0410 00:21:38.319384 6396 solver.cpp:218] Iteration 9996 (1.14762 iter/s, 10.4564s/12 iters), loss = 0.138446 I0410 00:21:38.319442 6396 solver.cpp:237] Train net output #0: loss = 0.138446 (* 1 = 0.138446 loss) I0410 00:21:38.319455 6396 sgd_solver.cpp:105] Iteration 9996, lr = 0.0013806 I0410 00:21:42.692276 6396 solver.cpp:218] Iteration 10008 (2.7443 iter/s, 4.3727s/12 iters), loss = 0.328357 I0410 00:21:42.692325 6396 solver.cpp:237] Train net output #0: loss = 0.328357 (* 1 = 0.328357 loss) I0410 00:21:42.692335 6396 sgd_solver.cpp:105] Iteration 10008, lr = 0.00137732 I0410 00:21:44.946401 6400 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:21:47.680586 6396 solver.cpp:218] Iteration 10020 (2.40572 iter/s, 4.9881s/12 iters), loss = 0.301588 I0410 00:21:47.680636 6396 solver.cpp:237] Train net output #0: loss = 0.301588 (* 1 = 0.301588 loss) I0410 00:21:47.680647 6396 sgd_solver.cpp:105] Iteration 10020, lr = 0.00137405 I0410 00:21:52.672595 6396 solver.cpp:218] Iteration 10032 (2.40394 iter/s, 4.9918s/12 iters), loss = 0.231714 I0410 00:21:52.672648 6396 solver.cpp:237] Train net output #0: loss = 0.231714 (* 1 = 0.231714 loss) I0410 00:21:52.672662 6396 sgd_solver.cpp:105] Iteration 10032, lr = 0.00137079 I0410 00:21:57.632632 6396 solver.cpp:218] Iteration 10044 (2.41944 iter/s, 4.95983s/12 iters), loss = 0.27375 I0410 00:21:57.632684 6396 solver.cpp:237] Train net output #0: loss = 0.27375 (* 1 = 0.27375 loss) I0410 00:21:57.632699 6396 sgd_solver.cpp:105] Iteration 10044, lr = 0.00136754 I0410 00:22:02.656038 6396 solver.cpp:218] Iteration 10056 (2.38892 iter/s, 5.02319s/12 iters), loss = 0.171432 I0410 00:22:02.656092 6396 solver.cpp:237] Train net output #0: loss = 0.171432 (* 1 = 0.171432 loss) I0410 00:22:02.656103 6396 sgd_solver.cpp:105] Iteration 10056, lr = 0.00136429 I0410 00:22:07.808368 6396 solver.cpp:218] Iteration 10068 (2.32914 iter/s, 5.15212s/12 iters), loss = 0.224659 I0410 00:22:07.808409 6396 solver.cpp:237] Train net output #0: loss = 0.224659 (* 1 = 0.224659 loss) I0410 00:22:07.808418 6396 sgd_solver.cpp:105] Iteration 10068, lr = 0.00136105 I0410 00:22:13.204452 6396 solver.cpp:218] Iteration 10080 (2.22392 iter/s, 5.39588s/12 iters), loss = 0.198015 I0410 00:22:13.218039 6396 solver.cpp:237] Train net output #0: loss = 0.198015 (* 1 = 0.198015 loss) I0410 00:22:13.218052 6396 sgd_solver.cpp:105] Iteration 10080, lr = 0.00135782 I0410 00:22:18.160141 6396 solver.cpp:218] Iteration 10092 (2.42819 iter/s, 4.94195s/12 iters), loss = 0.22793 I0410 00:22:18.160192 6396 solver.cpp:237] Train net output #0: loss = 0.22793 (* 1 = 0.22793 loss) I0410 00:22:18.160202 6396 sgd_solver.cpp:105] Iteration 10092, lr = 0.0013546 I0410 00:22:20.263829 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_10098.caffemodel I0410 00:22:22.395021 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_10098.solverstate I0410 00:22:22.951604 6396 solver.cpp:330] Iteration 10098, Testing net (#0) I0410 00:22:22.951623 6396 net.cpp:676] Ignoring source layer train-data I0410 00:22:23.378715 6401 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:22:27.364400 6396 solver.cpp:397] Test net output #0: accuracy = 0.595588 I0410 00:22:27.364449 6396 solver.cpp:397] Test net output #1: loss = 1.9279 (* 1 = 1.9279 loss) I0410 00:22:29.213155 6396 solver.cpp:218] Iteration 10104 (1.08571 iter/s, 11.0526s/12 iters), loss = 0.233035 I0410 00:22:29.213197 6396 solver.cpp:237] Train net output #0: loss = 0.233035 (* 1 = 0.233035 loss) I0410 00:22:29.213205 6396 sgd_solver.cpp:105] Iteration 10104, lr = 0.00135138 I0410 00:22:33.553717 6400 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:22:34.183845 6396 solver.cpp:218] Iteration 10116 (2.41425 iter/s, 4.97049s/12 iters), loss = 0.231064 I0410 00:22:34.183895 6396 solver.cpp:237] Train net output #0: loss = 0.231064 (* 1 = 0.231064 loss) I0410 00:22:34.183907 6396 sgd_solver.cpp:105] Iteration 10116, lr = 0.00134817 I0410 00:22:39.098832 6396 solver.cpp:218] Iteration 10128 (2.44162 iter/s, 4.91478s/12 iters), loss = 0.251912 I0410 00:22:39.098888 6396 solver.cpp:237] Train net output #0: loss = 0.251912 (* 1 = 0.251912 loss) I0410 00:22:39.098901 6396 sgd_solver.cpp:105] Iteration 10128, lr = 0.00134497 I0410 00:22:43.993654 6396 solver.cpp:218] Iteration 10140 (2.45167 iter/s, 4.89462s/12 iters), loss = 0.217957 I0410 00:22:43.993764 6396 solver.cpp:237] Train net output #0: loss = 0.217957 (* 1 = 0.217957 loss) I0410 00:22:43.993773 6396 sgd_solver.cpp:105] Iteration 10140, lr = 0.00134178 I0410 00:22:49.002866 6396 solver.cpp:218] Iteration 10152 (2.39571 iter/s, 5.00895s/12 iters), loss = 0.229969 I0410 00:22:49.002912 6396 solver.cpp:237] Train net output #0: loss = 0.229969 (* 1 = 0.229969 loss) I0410 00:22:49.002924 6396 sgd_solver.cpp:105] Iteration 10152, lr = 0.00133859 I0410 00:22:54.037717 6396 solver.cpp:218] Iteration 10164 (2.38349 iter/s, 5.03464s/12 iters), loss = 0.193928 I0410 00:22:54.037778 6396 solver.cpp:237] Train net output #0: loss = 0.193928 (* 1 = 0.193928 loss) I0410 00:22:54.037791 6396 sgd_solver.cpp:105] Iteration 10164, lr = 0.00133541 I0410 00:22:59.010346 6396 solver.cpp:218] Iteration 10176 (2.41332 iter/s, 4.97241s/12 iters), loss = 0.212104 I0410 00:22:59.010396 6396 solver.cpp:237] Train net output #0: loss = 0.212104 (* 1 = 0.212104 loss) I0410 00:22:59.010407 6396 sgd_solver.cpp:105] Iteration 10176, lr = 0.00133224 I0410 00:23:03.960711 6396 solver.cpp:218] Iteration 10188 (2.42416 iter/s, 4.95016s/12 iters), loss = 0.240575 I0410 00:23:03.960752 6396 solver.cpp:237] Train net output #0: loss = 0.240575 (* 1 = 0.240575 loss) I0410 00:23:03.960762 6396 sgd_solver.cpp:105] Iteration 10188, lr = 0.00132908 I0410 00:23:08.519349 6396 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_10200.caffemodel I0410 00:23:08.950157 6396 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_10200.solverstate I0410 00:23:09.287446 6396 solver.cpp:310] Iteration 10200, loss = 0.179988 I0410 00:23:09.287487 6396 solver.cpp:330] Iteration 10200, Testing net (#0) I0410 00:23:09.287497 6396 net.cpp:676] Ignoring source layer train-data I0410 00:23:09.712990 6401 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:23:13.919181 6396 solver.cpp:397] Test net output #0: accuracy = 0.602328 I0410 00:23:13.919232 6396 solver.cpp:397] Test net output #1: loss = 1.94631 (* 1 = 1.94631 loss) I0410 00:23:13.919245 6396 solver.cpp:315] Optimization Done. I0410 00:23:13.919252 6396 caffe.cpp:259] Optimization Done.