I0409 22:47:54.069082 4221 upgrade_proto.cpp:1082] Attempting to upgrade input file specified using deprecated 'solver_type' field (enum)': /mnt/bigdisk/DIGITS-MAN-2/digits/jobs/20210409-205237-935d/solver.prototxt I0409 22:47:54.069238 4221 upgrade_proto.cpp:1089] Successfully upgraded file specified using deprecated 'solver_type' field (enum) to 'type' field (string). W0409 22:47:54.069244 4221 upgrade_proto.cpp:1091] Note that future Caffe releases will only support 'type' field (string) for a solver's type. I0409 22:47:54.069301 4221 caffe.cpp:218] Using GPUs 3 I0409 22:47:54.081862 4221 caffe.cpp:223] GPU 3: GeForce GTX 1080 Ti I0409 22:47:54.346282 4221 solver.cpp:44] Initializing solver from parameters: test_iter: 51 test_interval: 102 base_lr: 0.01 display: 12 max_iter: 10200 lr_policy: "exp" gamma: 0.99980193 momentum: 0.9 weight_decay: 0.0001 snapshot: 102 snapshot_prefix: "snapshot" solver_mode: GPU device_id: 3 net: "train_val.prototxt" train_state { level: 0 stage: "" } type: "SGD" I0409 22:47:54.347378 4221 solver.cpp:87] Creating training net from net file: train_val.prototxt I0409 22:47:54.348271 4221 net.cpp:294] The NetState phase (0) differed from the phase (1) specified by a rule in layer val-data I0409 22:47:54.348287 4221 net.cpp:294] The NetState phase (0) differed from the phase (1) specified by a rule in layer accuracy I0409 22:47:54.348414 4221 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: 1024 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: "fc8" type: "InnerProduct" bottom: "fc6" 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 22:47:54.348496 4221 layer_factory.hpp:77] Creating layer train-data I0409 22:47:54.351049 4221 db_lmdb.cpp:35] Opened lmdb /mnt/bigdisk/DIGITS-MAN-2/digits/jobs/20210407-214532-d396/train_db I0409 22:47:54.351462 4221 net.cpp:84] Creating Layer train-data I0409 22:47:54.351475 4221 net.cpp:380] train-data -> data I0409 22:47:54.351493 4221 net.cpp:380] train-data -> label I0409 22:47:54.351505 4221 data_transformer.cpp:25] Loading mean file from: /mnt/bigdisk/DIGITS-MAN-2/digits/jobs/20210407-214532-d396/mean.binaryproto I0409 22:47:54.356187 4221 data_layer.cpp:45] output data size: 128,3,227,227 I0409 22:47:54.476730 4221 net.cpp:122] Setting up train-data I0409 22:47:54.476753 4221 net.cpp:129] Top shape: 128 3 227 227 (19787136) I0409 22:47:54.476758 4221 net.cpp:129] Top shape: 128 (128) I0409 22:47:54.476763 4221 net.cpp:137] Memory required for data: 79149056 I0409 22:47:54.476773 4221 layer_factory.hpp:77] Creating layer conv1 I0409 22:47:54.476794 4221 net.cpp:84] Creating Layer conv1 I0409 22:47:54.476799 4221 net.cpp:406] conv1 <- data I0409 22:47:54.476811 4221 net.cpp:380] conv1 -> conv1 I0409 22:47:55.046655 4221 net.cpp:122] Setting up conv1 I0409 22:47:55.046677 4221 net.cpp:129] Top shape: 128 96 55 55 (37171200) I0409 22:47:55.046681 4221 net.cpp:137] Memory required for data: 227833856 I0409 22:47:55.046701 4221 layer_factory.hpp:77] Creating layer relu1 I0409 22:47:55.046712 4221 net.cpp:84] Creating Layer relu1 I0409 22:47:55.046716 4221 net.cpp:406] relu1 <- conv1 I0409 22:47:55.046722 4221 net.cpp:367] relu1 -> conv1 (in-place) I0409 22:47:55.047016 4221 net.cpp:122] Setting up relu1 I0409 22:47:55.047024 4221 net.cpp:129] Top shape: 128 96 55 55 (37171200) I0409 22:47:55.047029 4221 net.cpp:137] Memory required for data: 376518656 I0409 22:47:55.047031 4221 layer_factory.hpp:77] Creating layer norm1 I0409 22:47:55.047041 4221 net.cpp:84] Creating Layer norm1 I0409 22:47:55.047044 4221 net.cpp:406] norm1 <- conv1 I0409 22:47:55.047050 4221 net.cpp:380] norm1 -> norm1 I0409 22:47:55.047489 4221 net.cpp:122] Setting up norm1 I0409 22:47:55.047499 4221 net.cpp:129] Top shape: 128 96 55 55 (37171200) I0409 22:47:55.047503 4221 net.cpp:137] Memory required for data: 525203456 I0409 22:47:55.047508 4221 layer_factory.hpp:77] Creating layer pool1 I0409 22:47:55.047515 4221 net.cpp:84] Creating Layer pool1 I0409 22:47:55.047518 4221 net.cpp:406] pool1 <- norm1 I0409 22:47:55.047523 4221 net.cpp:380] pool1 -> pool1 I0409 22:47:55.047581 4221 net.cpp:122] Setting up pool1 I0409 22:47:55.047587 4221 net.cpp:129] Top shape: 128 96 27 27 (8957952) I0409 22:47:55.047591 4221 net.cpp:137] Memory required for data: 561035264 I0409 22:47:55.047595 4221 layer_factory.hpp:77] Creating layer conv2 I0409 22:47:55.047605 4221 net.cpp:84] Creating Layer conv2 I0409 22:47:55.047608 4221 net.cpp:406] conv2 <- pool1 I0409 22:47:55.047613 4221 net.cpp:380] conv2 -> conv2 I0409 22:47:55.054240 4221 net.cpp:122] Setting up conv2 I0409 22:47:55.054256 4221 net.cpp:129] Top shape: 128 256 27 27 (23887872) I0409 22:47:55.054260 4221 net.cpp:137] Memory required for data: 656586752 I0409 22:47:55.054271 4221 layer_factory.hpp:77] Creating layer relu2 I0409 22:47:55.054280 4221 net.cpp:84] Creating Layer relu2 I0409 22:47:55.054283 4221 net.cpp:406] relu2 <- conv2 I0409 22:47:55.054289 4221 net.cpp:367] relu2 -> conv2 (in-place) I0409 22:47:55.054711 4221 net.cpp:122] Setting up relu2 I0409 22:47:55.054721 4221 net.cpp:129] Top shape: 128 256 27 27 (23887872) I0409 22:47:55.054725 4221 net.cpp:137] Memory required for data: 752138240 I0409 22:47:55.054728 4221 layer_factory.hpp:77] Creating layer norm2 I0409 22:47:55.054736 4221 net.cpp:84] Creating Layer norm2 I0409 22:47:55.054740 4221 net.cpp:406] norm2 <- conv2 I0409 22:47:55.054745 4221 net.cpp:380] norm2 -> norm2 I0409 22:47:55.055034 4221 net.cpp:122] Setting up norm2 I0409 22:47:55.055042 4221 net.cpp:129] Top shape: 128 256 27 27 (23887872) I0409 22:47:55.055047 4221 net.cpp:137] Memory required for data: 847689728 I0409 22:47:55.055049 4221 layer_factory.hpp:77] Creating layer pool2 I0409 22:47:55.055058 4221 net.cpp:84] Creating Layer pool2 I0409 22:47:55.055061 4221 net.cpp:406] pool2 <- norm2 I0409 22:47:55.055066 4221 net.cpp:380] pool2 -> pool2 I0409 22:47:55.055094 4221 net.cpp:122] Setting up pool2 I0409 22:47:55.055099 4221 net.cpp:129] Top shape: 128 256 13 13 (5537792) I0409 22:47:55.055102 4221 net.cpp:137] Memory required for data: 869840896 I0409 22:47:55.055105 4221 layer_factory.hpp:77] Creating layer conv3 I0409 22:47:55.055115 4221 net.cpp:84] Creating Layer conv3 I0409 22:47:55.055119 4221 net.cpp:406] conv3 <- pool2 I0409 22:47:55.055124 4221 net.cpp:380] conv3 -> conv3 I0409 22:47:55.064981 4221 net.cpp:122] Setting up conv3 I0409 22:47:55.064999 4221 net.cpp:129] Top shape: 128 384 13 13 (8306688) I0409 22:47:55.065002 4221 net.cpp:137] Memory required for data: 903067648 I0409 22:47:55.065014 4221 layer_factory.hpp:77] Creating layer relu3 I0409 22:47:55.065022 4221 net.cpp:84] Creating Layer relu3 I0409 22:47:55.065026 4221 net.cpp:406] relu3 <- conv3 I0409 22:47:55.065032 4221 net.cpp:367] relu3 -> conv3 (in-place) I0409 22:47:55.065454 4221 net.cpp:122] Setting up relu3 I0409 22:47:55.065464 4221 net.cpp:129] Top shape: 128 384 13 13 (8306688) I0409 22:47:55.065467 4221 net.cpp:137] Memory required for data: 936294400 I0409 22:47:55.065471 4221 layer_factory.hpp:77] Creating layer conv4 I0409 22:47:55.065481 4221 net.cpp:84] Creating Layer conv4 I0409 22:47:55.065485 4221 net.cpp:406] conv4 <- conv3 I0409 22:47:55.065490 4221 net.cpp:380] conv4 -> conv4 I0409 22:47:55.075414 4221 net.cpp:122] Setting up conv4 I0409 22:47:55.075433 4221 net.cpp:129] Top shape: 128 384 13 13 (8306688) I0409 22:47:55.075435 4221 net.cpp:137] Memory required for data: 969521152 I0409 22:47:55.075444 4221 layer_factory.hpp:77] Creating layer relu4 I0409 22:47:55.075451 4221 net.cpp:84] Creating Layer relu4 I0409 22:47:55.075456 4221 net.cpp:406] relu4 <- conv4 I0409 22:47:55.075461 4221 net.cpp:367] relu4 -> conv4 (in-place) I0409 22:47:55.075737 4221 net.cpp:122] Setting up relu4 I0409 22:47:55.075745 4221 net.cpp:129] Top shape: 128 384 13 13 (8306688) I0409 22:47:55.075749 4221 net.cpp:137] Memory required for data: 1002747904 I0409 22:47:55.075752 4221 layer_factory.hpp:77] Creating layer conv5 I0409 22:47:55.075762 4221 net.cpp:84] Creating Layer conv5 I0409 22:47:55.075765 4221 net.cpp:406] conv5 <- conv4 I0409 22:47:55.075790 4221 net.cpp:380] conv5 -> conv5 I0409 22:47:55.087088 4221 net.cpp:122] Setting up conv5 I0409 22:47:55.087107 4221 net.cpp:129] Top shape: 128 256 13 13 (5537792) I0409 22:47:55.087110 4221 net.cpp:137] Memory required for data: 1024899072 I0409 22:47:55.087123 4221 layer_factory.hpp:77] Creating layer relu5 I0409 22:47:55.087134 4221 net.cpp:84] Creating Layer relu5 I0409 22:47:55.087138 4221 net.cpp:406] relu5 <- conv5 I0409 22:47:55.087144 4221 net.cpp:367] relu5 -> conv5 (in-place) I0409 22:47:55.087630 4221 net.cpp:122] Setting up relu5 I0409 22:47:55.087641 4221 net.cpp:129] Top shape: 128 256 13 13 (5537792) I0409 22:47:55.087644 4221 net.cpp:137] Memory required for data: 1047050240 I0409 22:47:55.087648 4221 layer_factory.hpp:77] Creating layer pool5 I0409 22:47:55.087656 4221 net.cpp:84] Creating Layer pool5 I0409 22:47:55.087659 4221 net.cpp:406] pool5 <- conv5 I0409 22:47:55.087664 4221 net.cpp:380] pool5 -> pool5 I0409 22:47:55.087703 4221 net.cpp:122] Setting up pool5 I0409 22:47:55.087709 4221 net.cpp:129] Top shape: 128 256 6 6 (1179648) I0409 22:47:55.087713 4221 net.cpp:137] Memory required for data: 1051768832 I0409 22:47:55.087715 4221 layer_factory.hpp:77] Creating layer fc6 I0409 22:47:55.087726 4221 net.cpp:84] Creating Layer fc6 I0409 22:47:55.087729 4221 net.cpp:406] fc6 <- pool5 I0409 22:47:55.087734 4221 net.cpp:380] fc6 -> fc6 I0409 22:47:55.176427 4221 net.cpp:122] Setting up fc6 I0409 22:47:55.176447 4221 net.cpp:129] Top shape: 128 1024 (131072) I0409 22:47:55.176451 4221 net.cpp:137] Memory required for data: 1052293120 I0409 22:47:55.176461 4221 layer_factory.hpp:77] Creating layer relu6 I0409 22:47:55.176470 4221 net.cpp:84] Creating Layer relu6 I0409 22:47:55.176476 4221 net.cpp:406] relu6 <- fc6 I0409 22:47:55.176482 4221 net.cpp:367] relu6 -> fc6 (in-place) I0409 22:47:55.180379 4221 net.cpp:122] Setting up relu6 I0409 22:47:55.180389 4221 net.cpp:129] Top shape: 128 1024 (131072) I0409 22:47:55.180393 4221 net.cpp:137] Memory required for data: 1052817408 I0409 22:47:55.180397 4221 layer_factory.hpp:77] Creating layer drop6 I0409 22:47:55.180404 4221 net.cpp:84] Creating Layer drop6 I0409 22:47:55.180408 4221 net.cpp:406] drop6 <- fc6 I0409 22:47:55.180415 4221 net.cpp:367] drop6 -> fc6 (in-place) I0409 22:47:55.180444 4221 net.cpp:122] Setting up drop6 I0409 22:47:55.180450 4221 net.cpp:129] Top shape: 128 1024 (131072) I0409 22:47:55.180454 4221 net.cpp:137] Memory required for data: 1053341696 I0409 22:47:55.180457 4221 layer_factory.hpp:77] Creating layer fc8 I0409 22:47:55.180464 4221 net.cpp:84] Creating Layer fc8 I0409 22:47:55.180469 4221 net.cpp:406] fc8 <- fc6 I0409 22:47:55.180474 4221 net.cpp:380] fc8 -> fc8 I0409 22:47:55.182288 4221 net.cpp:122] Setting up fc8 I0409 22:47:55.182296 4221 net.cpp:129] Top shape: 128 196 (25088) I0409 22:47:55.182298 4221 net.cpp:137] Memory required for data: 1053442048 I0409 22:47:55.182305 4221 layer_factory.hpp:77] Creating layer loss I0409 22:47:55.182313 4221 net.cpp:84] Creating Layer loss I0409 22:47:55.182317 4221 net.cpp:406] loss <- fc8 I0409 22:47:55.182320 4221 net.cpp:406] loss <- label I0409 22:47:55.182327 4221 net.cpp:380] loss -> loss I0409 22:47:55.182334 4221 layer_factory.hpp:77] Creating layer loss I0409 22:47:55.182917 4221 net.cpp:122] Setting up loss I0409 22:47:55.182927 4221 net.cpp:129] Top shape: (1) I0409 22:47:55.182930 4221 net.cpp:132] with loss weight 1 I0409 22:47:55.182950 4221 net.cpp:137] Memory required for data: 1053442052 I0409 22:47:55.182953 4221 net.cpp:198] loss needs backward computation. I0409 22:47:55.182960 4221 net.cpp:198] fc8 needs backward computation. I0409 22:47:55.182965 4221 net.cpp:198] drop6 needs backward computation. I0409 22:47:55.182967 4221 net.cpp:198] relu6 needs backward computation. I0409 22:47:55.182971 4221 net.cpp:198] fc6 needs backward computation. I0409 22:47:55.182974 4221 net.cpp:198] pool5 needs backward computation. I0409 22:47:55.182978 4221 net.cpp:198] relu5 needs backward computation. I0409 22:47:55.182999 4221 net.cpp:198] conv5 needs backward computation. I0409 22:47:55.183003 4221 net.cpp:198] relu4 needs backward computation. I0409 22:47:55.183007 4221 net.cpp:198] conv4 needs backward computation. I0409 22:47:55.183010 4221 net.cpp:198] relu3 needs backward computation. I0409 22:47:55.183014 4221 net.cpp:198] conv3 needs backward computation. I0409 22:47:55.183018 4221 net.cpp:198] pool2 needs backward computation. I0409 22:47:55.183022 4221 net.cpp:198] norm2 needs backward computation. I0409 22:47:55.183025 4221 net.cpp:198] relu2 needs backward computation. I0409 22:47:55.183028 4221 net.cpp:198] conv2 needs backward computation. I0409 22:47:55.183033 4221 net.cpp:198] pool1 needs backward computation. I0409 22:47:55.183037 4221 net.cpp:198] norm1 needs backward computation. I0409 22:47:55.183040 4221 net.cpp:198] relu1 needs backward computation. I0409 22:47:55.183044 4221 net.cpp:198] conv1 needs backward computation. I0409 22:47:55.183048 4221 net.cpp:200] train-data does not need backward computation. I0409 22:47:55.183053 4221 net.cpp:242] This network produces output loss I0409 22:47:55.183065 4221 net.cpp:255] Network initialization done. I0409 22:47:55.183568 4221 solver.cpp:172] Creating test net (#0) specified by net file: train_val.prototxt I0409 22:47:55.183596 4221 net.cpp:294] The NetState phase (1) differed from the phase (0) specified by a rule in layer train-data I0409 22:47:55.183724 4221 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: 1024 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: "fc8" type: "InnerProduct" bottom: "fc6" 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 22:47:55.183813 4221 layer_factory.hpp:77] Creating layer val-data I0409 22:47:55.185420 4221 db_lmdb.cpp:35] Opened lmdb /mnt/bigdisk/DIGITS-MAN-2/digits/jobs/20210407-214532-d396/val_db I0409 22:47:55.185622 4221 net.cpp:84] Creating Layer val-data I0409 22:47:55.185633 4221 net.cpp:380] val-data -> data I0409 22:47:55.185642 4221 net.cpp:380] val-data -> label I0409 22:47:55.185647 4221 data_transformer.cpp:25] Loading mean file from: /mnt/bigdisk/DIGITS-MAN-2/digits/jobs/20210407-214532-d396/mean.binaryproto I0409 22:47:55.189513 4221 data_layer.cpp:45] output data size: 32,3,227,227 I0409 22:47:55.229689 4221 net.cpp:122] Setting up val-data I0409 22:47:55.229712 4221 net.cpp:129] Top shape: 32 3 227 227 (4946784) I0409 22:47:55.229717 4221 net.cpp:129] Top shape: 32 (32) I0409 22:47:55.229720 4221 net.cpp:137] Memory required for data: 19787264 I0409 22:47:55.229727 4221 layer_factory.hpp:77] Creating layer label_val-data_1_split I0409 22:47:55.229740 4221 net.cpp:84] Creating Layer label_val-data_1_split I0409 22:47:55.229744 4221 net.cpp:406] label_val-data_1_split <- label I0409 22:47:55.229753 4221 net.cpp:380] label_val-data_1_split -> label_val-data_1_split_0 I0409 22:47:55.229761 4221 net.cpp:380] label_val-data_1_split -> label_val-data_1_split_1 I0409 22:47:55.229861 4221 net.cpp:122] Setting up label_val-data_1_split I0409 22:47:55.229867 4221 net.cpp:129] Top shape: 32 (32) I0409 22:47:55.229871 4221 net.cpp:129] Top shape: 32 (32) I0409 22:47:55.229874 4221 net.cpp:137] Memory required for data: 19787520 I0409 22:47:55.229877 4221 layer_factory.hpp:77] Creating layer conv1 I0409 22:47:55.229890 4221 net.cpp:84] Creating Layer conv1 I0409 22:47:55.229895 4221 net.cpp:406] conv1 <- data I0409 22:47:55.229900 4221 net.cpp:380] conv1 -> conv1 I0409 22:47:55.233845 4221 net.cpp:122] Setting up conv1 I0409 22:47:55.233857 4221 net.cpp:129] Top shape: 32 96 55 55 (9292800) I0409 22:47:55.233860 4221 net.cpp:137] Memory required for data: 56958720 I0409 22:47:55.233871 4221 layer_factory.hpp:77] Creating layer relu1 I0409 22:47:55.233877 4221 net.cpp:84] Creating Layer relu1 I0409 22:47:55.233898 4221 net.cpp:406] relu1 <- conv1 I0409 22:47:55.233904 4221 net.cpp:367] relu1 -> conv1 (in-place) I0409 22:47:55.234375 4221 net.cpp:122] Setting up relu1 I0409 22:47:55.234385 4221 net.cpp:129] Top shape: 32 96 55 55 (9292800) I0409 22:47:55.234387 4221 net.cpp:137] Memory required for data: 94129920 I0409 22:47:55.234391 4221 layer_factory.hpp:77] Creating layer norm1 I0409 22:47:55.234400 4221 net.cpp:84] Creating Layer norm1 I0409 22:47:55.234403 4221 net.cpp:406] norm1 <- conv1 I0409 22:47:55.234409 4221 net.cpp:380] norm1 -> norm1 I0409 22:47:55.234705 4221 net.cpp:122] Setting up norm1 I0409 22:47:55.234714 4221 net.cpp:129] Top shape: 32 96 55 55 (9292800) I0409 22:47:55.234719 4221 net.cpp:137] Memory required for data: 131301120 I0409 22:47:55.234721 4221 layer_factory.hpp:77] Creating layer pool1 I0409 22:47:55.234728 4221 net.cpp:84] Creating Layer pool1 I0409 22:47:55.234731 4221 net.cpp:406] pool1 <- norm1 I0409 22:47:55.234736 4221 net.cpp:380] pool1 -> pool1 I0409 22:47:55.234762 4221 net.cpp:122] Setting up pool1 I0409 22:47:55.234767 4221 net.cpp:129] Top shape: 32 96 27 27 (2239488) I0409 22:47:55.234771 4221 net.cpp:137] Memory required for data: 140259072 I0409 22:47:55.234774 4221 layer_factory.hpp:77] Creating layer conv2 I0409 22:47:55.234782 4221 net.cpp:84] Creating Layer conv2 I0409 22:47:55.234786 4221 net.cpp:406] conv2 <- pool1 I0409 22:47:55.234791 4221 net.cpp:380] conv2 -> conv2 I0409 22:47:55.243584 4221 net.cpp:122] Setting up conv2 I0409 22:47:55.243602 4221 net.cpp:129] Top shape: 32 256 27 27 (5971968) I0409 22:47:55.243605 4221 net.cpp:137] Memory required for data: 164146944 I0409 22:47:55.243616 4221 layer_factory.hpp:77] Creating layer relu2 I0409 22:47:55.243624 4221 net.cpp:84] Creating Layer relu2 I0409 22:47:55.243628 4221 net.cpp:406] relu2 <- conv2 I0409 22:47:55.243634 4221 net.cpp:367] relu2 -> conv2 (in-place) I0409 22:47:55.244136 4221 net.cpp:122] Setting up relu2 I0409 22:47:55.244148 4221 net.cpp:129] Top shape: 32 256 27 27 (5971968) I0409 22:47:55.244151 4221 net.cpp:137] Memory required for data: 188034816 I0409 22:47:55.244155 4221 layer_factory.hpp:77] Creating layer norm2 I0409 22:47:55.244165 4221 net.cpp:84] Creating Layer norm2 I0409 22:47:55.244169 4221 net.cpp:406] norm2 <- conv2 I0409 22:47:55.244175 4221 net.cpp:380] norm2 -> norm2 I0409 22:47:55.244697 4221 net.cpp:122] Setting up norm2 I0409 22:47:55.244706 4221 net.cpp:129] Top shape: 32 256 27 27 (5971968) I0409 22:47:55.244710 4221 net.cpp:137] Memory required for data: 211922688 I0409 22:47:55.244714 4221 layer_factory.hpp:77] Creating layer pool2 I0409 22:47:55.244722 4221 net.cpp:84] Creating Layer pool2 I0409 22:47:55.244725 4221 net.cpp:406] pool2 <- norm2 I0409 22:47:55.244730 4221 net.cpp:380] pool2 -> pool2 I0409 22:47:55.244761 4221 net.cpp:122] Setting up pool2 I0409 22:47:55.244767 4221 net.cpp:129] Top shape: 32 256 13 13 (1384448) I0409 22:47:55.244771 4221 net.cpp:137] Memory required for data: 217460480 I0409 22:47:55.244773 4221 layer_factory.hpp:77] Creating layer conv3 I0409 22:47:55.244784 4221 net.cpp:84] Creating Layer conv3 I0409 22:47:55.244787 4221 net.cpp:406] conv3 <- pool2 I0409 22:47:55.244794 4221 net.cpp:380] conv3 -> conv3 I0409 22:47:55.261601 4221 net.cpp:122] Setting up conv3 I0409 22:47:55.261621 4221 net.cpp:129] Top shape: 32 384 13 13 (2076672) I0409 22:47:55.261626 4221 net.cpp:137] Memory required for data: 225767168 I0409 22:47:55.261637 4221 layer_factory.hpp:77] Creating layer relu3 I0409 22:47:55.261646 4221 net.cpp:84] Creating Layer relu3 I0409 22:47:55.261651 4221 net.cpp:406] relu3 <- conv3 I0409 22:47:55.261658 4221 net.cpp:367] relu3 -> conv3 (in-place) I0409 22:47:55.263229 4221 net.cpp:122] Setting up relu3 I0409 22:47:55.263242 4221 net.cpp:129] Top shape: 32 384 13 13 (2076672) I0409 22:47:55.263247 4221 net.cpp:137] Memory required for data: 234073856 I0409 22:47:55.263250 4221 layer_factory.hpp:77] Creating layer conv4 I0409 22:47:55.263279 4221 net.cpp:84] Creating Layer conv4 I0409 22:47:55.263284 4221 net.cpp:406] conv4 <- conv3 I0409 22:47:55.263290 4221 net.cpp:380] conv4 -> conv4 I0409 22:47:55.274873 4221 net.cpp:122] Setting up conv4 I0409 22:47:55.274888 4221 net.cpp:129] Top shape: 32 384 13 13 (2076672) I0409 22:47:55.274891 4221 net.cpp:137] Memory required for data: 242380544 I0409 22:47:55.274901 4221 layer_factory.hpp:77] Creating layer relu4 I0409 22:47:55.274909 4221 net.cpp:84] Creating Layer relu4 I0409 22:47:55.274914 4221 net.cpp:406] relu4 <- conv4 I0409 22:47:55.274921 4221 net.cpp:367] relu4 -> conv4 (in-place) I0409 22:47:55.275416 4221 net.cpp:122] Setting up relu4 I0409 22:47:55.275425 4221 net.cpp:129] Top shape: 32 384 13 13 (2076672) I0409 22:47:55.275430 4221 net.cpp:137] Memory required for data: 250687232 I0409 22:47:55.275434 4221 layer_factory.hpp:77] Creating layer conv5 I0409 22:47:55.275446 4221 net.cpp:84] Creating Layer conv5 I0409 22:47:55.275451 4221 net.cpp:406] conv5 <- conv4 I0409 22:47:55.275456 4221 net.cpp:380] conv5 -> conv5 I0409 22:47:55.283867 4221 net.cpp:122] Setting up conv5 I0409 22:47:55.283885 4221 net.cpp:129] Top shape: 32 256 13 13 (1384448) I0409 22:47:55.283890 4221 net.cpp:137] Memory required for data: 256225024 I0409 22:47:55.283901 4221 layer_factory.hpp:77] Creating layer relu5 I0409 22:47:55.283910 4221 net.cpp:84] Creating Layer relu5 I0409 22:47:55.283913 4221 net.cpp:406] relu5 <- conv5 I0409 22:47:55.283921 4221 net.cpp:367] relu5 -> conv5 (in-place) I0409 22:47:55.284420 4221 net.cpp:122] Setting up relu5 I0409 22:47:55.284430 4221 net.cpp:129] Top shape: 32 256 13 13 (1384448) I0409 22:47:55.284433 4221 net.cpp:137] Memory required for data: 261762816 I0409 22:47:55.284437 4221 layer_factory.hpp:77] Creating layer pool5 I0409 22:47:55.284448 4221 net.cpp:84] Creating Layer pool5 I0409 22:47:55.284453 4221 net.cpp:406] pool5 <- conv5 I0409 22:47:55.284459 4221 net.cpp:380] pool5 -> pool5 I0409 22:47:55.284498 4221 net.cpp:122] Setting up pool5 I0409 22:47:55.284505 4221 net.cpp:129] Top shape: 32 256 6 6 (294912) I0409 22:47:55.284509 4221 net.cpp:137] Memory required for data: 262942464 I0409 22:47:55.284512 4221 layer_factory.hpp:77] Creating layer fc6 I0409 22:47:55.284520 4221 net.cpp:84] Creating Layer fc6 I0409 22:47:55.284523 4221 net.cpp:406] fc6 <- pool5 I0409 22:47:55.284529 4221 net.cpp:380] fc6 -> fc6 I0409 22:47:55.374289 4221 net.cpp:122] Setting up fc6 I0409 22:47:55.374308 4221 net.cpp:129] Top shape: 32 1024 (32768) I0409 22:47:55.374312 4221 net.cpp:137] Memory required for data: 263073536 I0409 22:47:55.374321 4221 layer_factory.hpp:77] Creating layer relu6 I0409 22:47:55.374331 4221 net.cpp:84] Creating Layer relu6 I0409 22:47:55.374336 4221 net.cpp:406] relu6 <- fc6 I0409 22:47:55.374341 4221 net.cpp:367] relu6 -> fc6 (in-place) I0409 22:47:55.374960 4221 net.cpp:122] Setting up relu6 I0409 22:47:55.374969 4221 net.cpp:129] Top shape: 32 1024 (32768) I0409 22:47:55.374972 4221 net.cpp:137] Memory required for data: 263204608 I0409 22:47:55.374976 4221 layer_factory.hpp:77] Creating layer drop6 I0409 22:47:55.374984 4221 net.cpp:84] Creating Layer drop6 I0409 22:47:55.374989 4221 net.cpp:406] drop6 <- fc6 I0409 22:47:55.374994 4221 net.cpp:367] drop6 -> fc6 (in-place) I0409 22:47:55.375018 4221 net.cpp:122] Setting up drop6 I0409 22:47:55.375023 4221 net.cpp:129] Top shape: 32 1024 (32768) I0409 22:47:55.375026 4221 net.cpp:137] Memory required for data: 263335680 I0409 22:47:55.375030 4221 layer_factory.hpp:77] Creating layer fc8 I0409 22:47:55.375037 4221 net.cpp:84] Creating Layer fc8 I0409 22:47:55.375041 4221 net.cpp:406] fc8 <- fc6 I0409 22:47:55.375046 4221 net.cpp:380] fc8 -> fc8 I0409 22:47:55.376833 4221 net.cpp:122] Setting up fc8 I0409 22:47:55.376840 4221 net.cpp:129] Top shape: 32 196 (6272) I0409 22:47:55.376843 4221 net.cpp:137] Memory required for data: 263360768 I0409 22:47:55.376850 4221 layer_factory.hpp:77] Creating layer fc8_fc8_0_split I0409 22:47:55.376855 4221 net.cpp:84] Creating Layer fc8_fc8_0_split I0409 22:47:55.376876 4221 net.cpp:406] fc8_fc8_0_split <- fc8 I0409 22:47:55.376881 4221 net.cpp:380] fc8_fc8_0_split -> fc8_fc8_0_split_0 I0409 22:47:55.376893 4221 net.cpp:380] fc8_fc8_0_split -> fc8_fc8_0_split_1 I0409 22:47:55.376926 4221 net.cpp:122] Setting up fc8_fc8_0_split I0409 22:47:55.376932 4221 net.cpp:129] Top shape: 32 196 (6272) I0409 22:47:55.376935 4221 net.cpp:129] Top shape: 32 196 (6272) I0409 22:47:55.376938 4221 net.cpp:137] Memory required for data: 263410944 I0409 22:47:55.376942 4221 layer_factory.hpp:77] Creating layer accuracy I0409 22:47:55.376950 4221 net.cpp:84] Creating Layer accuracy I0409 22:47:55.376953 4221 net.cpp:406] accuracy <- fc8_fc8_0_split_0 I0409 22:47:55.376957 4221 net.cpp:406] accuracy <- label_val-data_1_split_0 I0409 22:47:55.376963 4221 net.cpp:380] accuracy -> accuracy I0409 22:47:55.376971 4221 net.cpp:122] Setting up accuracy I0409 22:47:55.376974 4221 net.cpp:129] Top shape: (1) I0409 22:47:55.376977 4221 net.cpp:137] Memory required for data: 263410948 I0409 22:47:55.376981 4221 layer_factory.hpp:77] Creating layer loss I0409 22:47:55.376986 4221 net.cpp:84] Creating Layer loss I0409 22:47:55.376989 4221 net.cpp:406] loss <- fc8_fc8_0_split_1 I0409 22:47:55.376993 4221 net.cpp:406] loss <- label_val-data_1_split_1 I0409 22:47:55.376998 4221 net.cpp:380] loss -> loss I0409 22:47:55.377005 4221 layer_factory.hpp:77] Creating layer loss I0409 22:47:55.377804 4221 net.cpp:122] Setting up loss I0409 22:47:55.377815 4221 net.cpp:129] Top shape: (1) I0409 22:47:55.377817 4221 net.cpp:132] with loss weight 1 I0409 22:47:55.377828 4221 net.cpp:137] Memory required for data: 263410952 I0409 22:47:55.377831 4221 net.cpp:198] loss needs backward computation. I0409 22:47:55.377836 4221 net.cpp:200] accuracy does not need backward computation. I0409 22:47:55.377840 4221 net.cpp:198] fc8_fc8_0_split needs backward computation. I0409 22:47:55.377844 4221 net.cpp:198] fc8 needs backward computation. I0409 22:47:55.377847 4221 net.cpp:198] drop6 needs backward computation. I0409 22:47:55.377851 4221 net.cpp:198] relu6 needs backward computation. I0409 22:47:55.377853 4221 net.cpp:198] fc6 needs backward computation. I0409 22:47:55.377857 4221 net.cpp:198] pool5 needs backward computation. I0409 22:47:55.377861 4221 net.cpp:198] relu5 needs backward computation. I0409 22:47:55.377864 4221 net.cpp:198] conv5 needs backward computation. I0409 22:47:55.377867 4221 net.cpp:198] relu4 needs backward computation. I0409 22:47:55.377871 4221 net.cpp:198] conv4 needs backward computation. I0409 22:47:55.377874 4221 net.cpp:198] relu3 needs backward computation. I0409 22:47:55.377877 4221 net.cpp:198] conv3 needs backward computation. I0409 22:47:55.377880 4221 net.cpp:198] pool2 needs backward computation. I0409 22:47:55.377884 4221 net.cpp:198] norm2 needs backward computation. I0409 22:47:55.377887 4221 net.cpp:198] relu2 needs backward computation. I0409 22:47:55.377892 4221 net.cpp:198] conv2 needs backward computation. I0409 22:47:55.377894 4221 net.cpp:198] pool1 needs backward computation. I0409 22:47:55.377897 4221 net.cpp:198] norm1 needs backward computation. I0409 22:47:55.377902 4221 net.cpp:198] relu1 needs backward computation. I0409 22:47:55.377904 4221 net.cpp:198] conv1 needs backward computation. I0409 22:47:55.377907 4221 net.cpp:200] label_val-data_1_split does not need backward computation. I0409 22:47:55.377912 4221 net.cpp:200] val-data does not need backward computation. I0409 22:47:55.377915 4221 net.cpp:242] This network produces output accuracy I0409 22:47:55.377918 4221 net.cpp:242] This network produces output loss I0409 22:47:55.377934 4221 net.cpp:255] Network initialization done. I0409 22:47:55.378016 4221 solver.cpp:56] Solver scaffolding done. I0409 22:47:55.378408 4221 caffe.cpp:248] Starting Optimization I0409 22:47:55.378417 4221 solver.cpp:272] Solving I0409 22:47:55.378420 4221 solver.cpp:273] Learning Rate Policy: exp I0409 22:47:55.379362 4221 solver.cpp:330] Iteration 0, Testing net (#0) I0409 22:47:55.379382 4221 net.cpp:676] Ignoring source layer train-data I0409 22:47:55.399120 4221 blocking_queue.cpp:49] Waiting for data I0409 22:47:59.913327 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 22:47:59.957465 4221 solver.cpp:397] Test net output #0: accuracy = 0.00612745 I0409 22:47:59.957507 4221 solver.cpp:397] Test net output #1: loss = 5.27887 (* 1 = 5.27887 loss) I0409 22:48:00.067438 4221 solver.cpp:218] Iteration 0 (0 iter/s, 4.68878s/12 iters), loss = 5.28039 I0409 22:48:00.068953 4221 solver.cpp:237] Train net output #0: loss = 5.28039 (* 1 = 5.28039 loss) I0409 22:48:00.068972 4221 sgd_solver.cpp:105] Iteration 0, lr = 0.01 I0409 22:48:03.994359 4221 solver.cpp:218] Iteration 12 (3.05715 iter/s, 3.92522s/12 iters), loss = 5.27368 I0409 22:48:03.994411 4221 solver.cpp:237] Train net output #0: loss = 5.27368 (* 1 = 5.27368 loss) I0409 22:48:03.994423 4221 sgd_solver.cpp:105] Iteration 12, lr = 0.00997626 I0409 22:48:08.830438 4221 solver.cpp:218] Iteration 24 (2.48149 iter/s, 4.83581s/12 iters), loss = 5.27272 I0409 22:48:08.830497 4221 solver.cpp:237] Train net output #0: loss = 5.27272 (* 1 = 5.27272 loss) I0409 22:48:08.830509 4221 sgd_solver.cpp:105] Iteration 24, lr = 0.00995257 I0409 22:48:13.701555 4221 solver.cpp:218] Iteration 36 (2.46364 iter/s, 4.87085s/12 iters), loss = 5.28562 I0409 22:48:13.701607 4221 solver.cpp:237] Train net output #0: loss = 5.28562 (* 1 = 5.28562 loss) I0409 22:48:13.701617 4221 sgd_solver.cpp:105] Iteration 36, lr = 0.00992894 I0409 22:48:18.514457 4221 solver.cpp:218] Iteration 48 (2.49343 iter/s, 4.81264s/12 iters), loss = 5.2824 I0409 22:48:18.514506 4221 solver.cpp:237] Train net output #0: loss = 5.2824 (* 1 = 5.2824 loss) I0409 22:48:18.514519 4221 sgd_solver.cpp:105] Iteration 48, lr = 0.00990537 I0409 22:48:23.326609 4221 solver.cpp:218] Iteration 60 (2.49382 iter/s, 4.81189s/12 iters), loss = 5.2873 I0409 22:48:23.326668 4221 solver.cpp:237] Train net output #0: loss = 5.2873 (* 1 = 5.2873 loss) I0409 22:48:23.326680 4221 sgd_solver.cpp:105] Iteration 60, lr = 0.00988185 I0409 22:48:28.136777 4221 solver.cpp:218] Iteration 72 (2.49485 iter/s, 4.8099s/12 iters), loss = 5.27805 I0409 22:48:28.136895 4221 solver.cpp:237] Train net output #0: loss = 5.27805 (* 1 = 5.27805 loss) I0409 22:48:28.136907 4221 sgd_solver.cpp:105] Iteration 72, lr = 0.00985839 I0409 22:48:32.941704 4221 solver.cpp:218] Iteration 84 (2.49761 iter/s, 4.8046s/12 iters), loss = 5.28506 I0409 22:48:32.941767 4221 solver.cpp:237] Train net output #0: loss = 5.28506 (* 1 = 5.28506 loss) I0409 22:48:32.941778 4221 sgd_solver.cpp:105] Iteration 84, lr = 0.00983498 I0409 22:48:37.753037 4221 solver.cpp:218] Iteration 96 (2.49425 iter/s, 4.81106s/12 iters), loss = 5.29272 I0409 22:48:37.753085 4221 solver.cpp:237] Train net output #0: loss = 5.29272 (* 1 = 5.29272 loss) I0409 22:48:37.753096 4221 sgd_solver.cpp:105] Iteration 96, lr = 0.00981163 I0409 22:48:39.398953 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 22:48:39.704267 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_102.caffemodel I0409 22:48:40.972460 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_102.solverstate I0409 22:48:42.088553 4221 solver.cpp:330] Iteration 102, Testing net (#0) I0409 22:48:42.088585 4221 net.cpp:676] Ignoring source layer train-data I0409 22:48:46.458106 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 22:48:46.534135 4221 solver.cpp:397] Test net output #0: accuracy = 0.00551471 I0409 22:48:46.534186 4221 solver.cpp:397] Test net output #1: loss = 5.27925 (* 1 = 5.27925 loss) I0409 22:48:48.338774 4221 solver.cpp:218] Iteration 108 (1.13365 iter/s, 10.5852s/12 iters), loss = 5.28059 I0409 22:48:48.338830 4221 solver.cpp:237] Train net output #0: loss = 5.28059 (* 1 = 5.28059 loss) I0409 22:48:48.338841 4221 sgd_solver.cpp:105] Iteration 108, lr = 0.00978834 I0409 22:48:53.232730 4221 solver.cpp:218] Iteration 120 (2.45214 iter/s, 4.89369s/12 iters), loss = 5.27139 I0409 22:48:53.232767 4221 solver.cpp:237] Train net output #0: loss = 5.27139 (* 1 = 5.27139 loss) I0409 22:48:53.232777 4221 sgd_solver.cpp:105] Iteration 120, lr = 0.0097651 I0409 22:48:58.156252 4221 solver.cpp:218] Iteration 132 (2.43741 iter/s, 4.92327s/12 iters), loss = 5.22872 I0409 22:48:58.156379 4221 solver.cpp:237] Train net output #0: loss = 5.22872 (* 1 = 5.22872 loss) I0409 22:48:58.156394 4221 sgd_solver.cpp:105] Iteration 132, lr = 0.00974192 I0409 22:49:02.966440 4221 solver.cpp:218] Iteration 144 (2.49488 iter/s, 4.80985s/12 iters), loss = 5.27045 I0409 22:49:02.966496 4221 solver.cpp:237] Train net output #0: loss = 5.27045 (* 1 = 5.27045 loss) I0409 22:49:02.966507 4221 sgd_solver.cpp:105] Iteration 144, lr = 0.00971879 I0409 22:49:07.781404 4221 solver.cpp:218] Iteration 156 (2.49237 iter/s, 4.8147s/12 iters), loss = 5.21511 I0409 22:49:07.781453 4221 solver.cpp:237] Train net output #0: loss = 5.21511 (* 1 = 5.21511 loss) I0409 22:49:07.781466 4221 sgd_solver.cpp:105] Iteration 156, lr = 0.00969571 I0409 22:49:12.858525 4221 solver.cpp:218] Iteration 168 (2.36367 iter/s, 5.07685s/12 iters), loss = 5.19318 I0409 22:49:12.858578 4221 solver.cpp:237] Train net output #0: loss = 5.19318 (* 1 = 5.19318 loss) I0409 22:49:12.858590 4221 sgd_solver.cpp:105] Iteration 168, lr = 0.00967269 I0409 22:49:17.721390 4221 solver.cpp:218] Iteration 180 (2.46782 iter/s, 4.8626s/12 iters), loss = 5.15652 I0409 22:49:17.721442 4221 solver.cpp:237] Train net output #0: loss = 5.15652 (* 1 = 5.15652 loss) I0409 22:49:17.721453 4221 sgd_solver.cpp:105] Iteration 180, lr = 0.00964973 I0409 22:49:22.847829 4221 solver.cpp:218] Iteration 192 (2.34093 iter/s, 5.12616s/12 iters), loss = 5.23807 I0409 22:49:22.847884 4221 solver.cpp:237] Train net output #0: loss = 5.23807 (* 1 = 5.23807 loss) I0409 22:49:22.847898 4221 sgd_solver.cpp:105] Iteration 192, lr = 0.00962682 I0409 22:49:26.595958 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 22:49:27.254276 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_204.caffemodel I0409 22:49:27.993767 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_204.solverstate I0409 22:49:28.489436 4221 solver.cpp:330] Iteration 204, Testing net (#0) I0409 22:49:28.489513 4221 net.cpp:676] Ignoring source layer train-data I0409 22:49:32.958518 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 22:49:33.094182 4221 solver.cpp:397] Test net output #0: accuracy = 0.00735294 I0409 22:49:33.094216 4221 solver.cpp:397] Test net output #1: loss = 5.1881 (* 1 = 5.1881 loss) I0409 22:49:33.177574 4221 solver.cpp:218] Iteration 204 (1.16175 iter/s, 10.3293s/12 iters), loss = 5.12571 I0409 22:49:33.177616 4221 solver.cpp:237] Train net output #0: loss = 5.12571 (* 1 = 5.12571 loss) I0409 22:49:33.177624 4221 sgd_solver.cpp:105] Iteration 204, lr = 0.00960396 I0409 22:49:37.340734 4221 solver.cpp:218] Iteration 216 (2.88258 iter/s, 4.16293s/12 iters), loss = 5.18359 I0409 22:49:37.340782 4221 solver.cpp:237] Train net output #0: loss = 5.18359 (* 1 = 5.18359 loss) I0409 22:49:37.340795 4221 sgd_solver.cpp:105] Iteration 216, lr = 0.00958116 I0409 22:49:42.234980 4221 solver.cpp:218] Iteration 228 (2.45199 iter/s, 4.89398s/12 iters), loss = 5.19932 I0409 22:49:42.235041 4221 solver.cpp:237] Train net output #0: loss = 5.19932 (* 1 = 5.19932 loss) I0409 22:49:42.235054 4221 sgd_solver.cpp:105] Iteration 228, lr = 0.00955841 I0409 22:49:47.032968 4221 solver.cpp:218] Iteration 240 (2.50119 iter/s, 4.79772s/12 iters), loss = 5.20539 I0409 22:49:47.033017 4221 solver.cpp:237] Train net output #0: loss = 5.20539 (* 1 = 5.20539 loss) I0409 22:49:47.033028 4221 sgd_solver.cpp:105] Iteration 240, lr = 0.00953572 I0409 22:49:51.853164 4221 solver.cpp:218] Iteration 252 (2.48966 iter/s, 4.81994s/12 iters), loss = 5.0947 I0409 22:49:51.853214 4221 solver.cpp:237] Train net output #0: loss = 5.0947 (* 1 = 5.0947 loss) I0409 22:49:51.853225 4221 sgd_solver.cpp:105] Iteration 252, lr = 0.00951308 I0409 22:49:56.688650 4221 solver.cpp:218] Iteration 264 (2.48179 iter/s, 4.83523s/12 iters), loss = 5.19976 I0409 22:49:56.688700 4221 solver.cpp:237] Train net output #0: loss = 5.19976 (* 1 = 5.19976 loss) I0409 22:49:56.688711 4221 sgd_solver.cpp:105] Iteration 264, lr = 0.00949049 I0409 22:50:01.533262 4221 solver.cpp:218] Iteration 276 (2.47711 iter/s, 4.84435s/12 iters), loss = 5.17812 I0409 22:50:01.533381 4221 solver.cpp:237] Train net output #0: loss = 5.17812 (* 1 = 5.17812 loss) I0409 22:50:01.533391 4221 sgd_solver.cpp:105] Iteration 276, lr = 0.00946796 I0409 22:50:06.363451 4221 solver.cpp:218] Iteration 288 (2.48454 iter/s, 4.82986s/12 iters), loss = 5.04067 I0409 22:50:06.363498 4221 solver.cpp:237] Train net output #0: loss = 5.04067 (* 1 = 5.04067 loss) I0409 22:50:06.363509 4221 sgd_solver.cpp:105] Iteration 288, lr = 0.00944548 I0409 22:50:11.171166 4221 solver.cpp:218] Iteration 300 (2.49612 iter/s, 4.80746s/12 iters), loss = 5.1351 I0409 22:50:11.171206 4221 solver.cpp:237] Train net output #0: loss = 5.1351 (* 1 = 5.1351 loss) I0409 22:50:11.171216 4221 sgd_solver.cpp:105] Iteration 300, lr = 0.00942305 I0409 22:50:12.114235 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 22:50:13.130764 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_306.caffemodel I0409 22:50:15.898541 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_306.solverstate I0409 22:50:17.295846 4221 solver.cpp:330] Iteration 306, Testing net (#0) I0409 22:50:17.295876 4221 net.cpp:676] Ignoring source layer train-data I0409 22:50:21.702610 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 22:50:21.859093 4221 solver.cpp:397] Test net output #0: accuracy = 0.0153186 I0409 22:50:21.859143 4221 solver.cpp:397] Test net output #1: loss = 5.10796 (* 1 = 5.10796 loss) I0409 22:50:23.764350 4221 solver.cpp:218] Iteration 312 (0.952939 iter/s, 12.5926s/12 iters), loss = 5.05809 I0409 22:50:23.764402 4221 solver.cpp:237] Train net output #0: loss = 5.05809 (* 1 = 5.05809 loss) I0409 22:50:23.764415 4221 sgd_solver.cpp:105] Iteration 312, lr = 0.00940068 I0409 22:50:28.679682 4221 solver.cpp:218] Iteration 324 (2.44147 iter/s, 4.91507s/12 iters), loss = 5.12274 I0409 22:50:28.679725 4221 solver.cpp:237] Train net output #0: loss = 5.12274 (* 1 = 5.12274 loss) I0409 22:50:28.679734 4221 sgd_solver.cpp:105] Iteration 324, lr = 0.00937836 I0409 22:50:33.548092 4221 solver.cpp:218] Iteration 336 (2.465 iter/s, 4.86815s/12 iters), loss = 5.07955 I0409 22:50:33.548171 4221 solver.cpp:237] Train net output #0: loss = 5.07955 (* 1 = 5.07955 loss) I0409 22:50:33.548182 4221 sgd_solver.cpp:105] Iteration 336, lr = 0.0093561 I0409 22:50:38.416586 4221 solver.cpp:218] Iteration 348 (2.46497 iter/s, 4.86821s/12 iters), loss = 5.03192 I0409 22:50:38.416635 4221 solver.cpp:237] Train net output #0: loss = 5.03192 (* 1 = 5.03192 loss) I0409 22:50:38.416646 4221 sgd_solver.cpp:105] Iteration 348, lr = 0.00933388 I0409 22:50:43.227648 4221 solver.cpp:218] Iteration 360 (2.49438 iter/s, 4.81081s/12 iters), loss = 5.06121 I0409 22:50:43.227692 4221 solver.cpp:237] Train net output #0: loss = 5.06121 (* 1 = 5.06121 loss) I0409 22:50:43.227701 4221 sgd_solver.cpp:105] Iteration 360, lr = 0.00931172 I0409 22:50:48.065443 4221 solver.cpp:218] Iteration 372 (2.4806 iter/s, 4.83754s/12 iters), loss = 4.98512 I0409 22:50:48.065490 4221 solver.cpp:237] Train net output #0: loss = 4.98512 (* 1 = 4.98512 loss) I0409 22:50:48.065497 4221 sgd_solver.cpp:105] Iteration 372, lr = 0.00928961 I0409 22:50:52.914916 4221 solver.cpp:218] Iteration 384 (2.47463 iter/s, 4.84921s/12 iters), loss = 5.10166 I0409 22:50:52.914968 4221 solver.cpp:237] Train net output #0: loss = 5.10166 (* 1 = 5.10166 loss) I0409 22:50:52.914978 4221 sgd_solver.cpp:105] Iteration 384, lr = 0.00926756 I0409 22:50:57.769318 4221 solver.cpp:218] Iteration 396 (2.47212 iter/s, 4.85414s/12 iters), loss = 5.02094 I0409 22:50:57.769369 4221 solver.cpp:237] Train net output #0: loss = 5.02094 (* 1 = 5.02094 loss) I0409 22:50:57.769380 4221 sgd_solver.cpp:105] Iteration 396, lr = 0.00924556 I0409 22:51:00.792229 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 22:51:02.162448 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_408.caffemodel I0409 22:51:03.979949 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_408.solverstate I0409 22:51:06.312731 4221 solver.cpp:330] Iteration 408, Testing net (#0) I0409 22:51:06.312749 4221 net.cpp:676] Ignoring source layer train-data I0409 22:51:10.634052 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 22:51:10.844585 4221 solver.cpp:397] Test net output #0: accuracy = 0.0220588 I0409 22:51:10.844635 4221 solver.cpp:397] Test net output #1: loss = 5.03741 (* 1 = 5.03741 loss) I0409 22:51:10.926493 4221 solver.cpp:218] Iteration 408 (0.912092 iter/s, 13.1566s/12 iters), loss = 5.06879 I0409 22:51:10.926548 4221 solver.cpp:237] Train net output #0: loss = 5.06879 (* 1 = 5.06879 loss) I0409 22:51:10.926558 4221 sgd_solver.cpp:105] Iteration 408, lr = 0.00922361 I0409 22:51:15.141006 4221 solver.cpp:218] Iteration 420 (2.84747 iter/s, 4.21427s/12 iters), loss = 5.09318 I0409 22:51:15.141050 4221 solver.cpp:237] Train net output #0: loss = 5.09318 (* 1 = 5.09318 loss) I0409 22:51:15.141058 4221 sgd_solver.cpp:105] Iteration 420, lr = 0.00920171 I0409 22:51:19.965104 4221 solver.cpp:218] Iteration 432 (2.48764 iter/s, 4.82384s/12 iters), loss = 5.04506 I0409 22:51:19.965155 4221 solver.cpp:237] Train net output #0: loss = 5.04506 (* 1 = 5.04506 loss) I0409 22:51:19.965167 4221 sgd_solver.cpp:105] Iteration 432, lr = 0.00917986 I0409 22:51:24.791097 4221 solver.cpp:218] Iteration 444 (2.48667 iter/s, 4.82573s/12 iters), loss = 4.94556 I0409 22:51:24.791147 4221 solver.cpp:237] Train net output #0: loss = 4.94556 (* 1 = 4.94556 loss) I0409 22:51:24.791159 4221 sgd_solver.cpp:105] Iteration 444, lr = 0.00915807 I0409 22:51:29.634987 4221 solver.cpp:218] Iteration 456 (2.47748 iter/s, 4.84363s/12 iters), loss = 4.98926 I0409 22:51:29.635041 4221 solver.cpp:237] Train net output #0: loss = 4.98926 (* 1 = 4.98926 loss) I0409 22:51:29.635053 4221 sgd_solver.cpp:105] Iteration 456, lr = 0.00913632 I0409 22:51:34.459417 4221 solver.cpp:218] Iteration 468 (2.48748 iter/s, 4.82417s/12 iters), loss = 4.99261 I0409 22:51:34.459507 4221 solver.cpp:237] Train net output #0: loss = 4.99261 (* 1 = 4.99261 loss) I0409 22:51:34.459517 4221 sgd_solver.cpp:105] Iteration 468, lr = 0.00911463 I0409 22:51:39.313997 4221 solver.cpp:218] Iteration 480 (2.47204 iter/s, 4.85428s/12 iters), loss = 4.98419 I0409 22:51:39.314038 4221 solver.cpp:237] Train net output #0: loss = 4.98419 (* 1 = 4.98419 loss) I0409 22:51:39.314047 4221 sgd_solver.cpp:105] Iteration 480, lr = 0.00909299 I0409 22:51:44.147225 4221 solver.cpp:218] Iteration 492 (2.48294 iter/s, 4.83297s/12 iters), loss = 5.00092 I0409 22:51:44.147280 4221 solver.cpp:237] Train net output #0: loss = 5.00092 (* 1 = 5.00092 loss) I0409 22:51:44.147294 4221 sgd_solver.cpp:105] Iteration 492, lr = 0.0090714 I0409 22:51:48.986286 4221 solver.cpp:218] Iteration 504 (2.47995 iter/s, 4.8388s/12 iters), loss = 5.08446 I0409 22:51:48.986330 4221 solver.cpp:237] Train net output #0: loss = 5.08446 (* 1 = 5.08446 loss) I0409 22:51:48.986341 4221 sgd_solver.cpp:105] Iteration 504, lr = 0.00904986 I0409 22:51:49.273102 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 22:51:51.010774 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_510.caffemodel I0409 22:51:51.676230 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_510.solverstate I0409 22:51:52.171375 4221 solver.cpp:330] Iteration 510, Testing net (#0) I0409 22:51:52.171403 4221 net.cpp:676] Ignoring source layer train-data I0409 22:51:56.610823 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 22:51:56.847576 4221 solver.cpp:397] Test net output #0: accuracy = 0.0245098 I0409 22:51:56.847617 4221 solver.cpp:397] Test net output #1: loss = 4.99314 (* 1 = 4.99314 loss) I0409 22:51:59.137660 4221 solver.cpp:218] Iteration 516 (1.18216 iter/s, 10.1509s/12 iters), loss = 4.93348 I0409 22:51:59.137713 4221 solver.cpp:237] Train net output #0: loss = 4.93348 (* 1 = 4.93348 loss) I0409 22:51:59.137723 4221 sgd_solver.cpp:105] Iteration 516, lr = 0.00902838 I0409 22:52:03.987942 4221 solver.cpp:218] Iteration 528 (2.47422 iter/s, 4.85001s/12 iters), loss = 4.97733 I0409 22:52:03.987994 4221 solver.cpp:237] Train net output #0: loss = 4.97733 (* 1 = 4.97733 loss) I0409 22:52:03.988005 4221 sgd_solver.cpp:105] Iteration 528, lr = 0.00900694 I0409 22:52:08.852067 4221 solver.cpp:218] Iteration 540 (2.46717 iter/s, 4.86386s/12 iters), loss = 4.9358 I0409 22:52:08.852193 4221 solver.cpp:237] Train net output #0: loss = 4.9358 (* 1 = 4.9358 loss) I0409 22:52:08.852205 4221 sgd_solver.cpp:105] Iteration 540, lr = 0.00898556 I0409 22:52:13.668972 4221 solver.cpp:218] Iteration 552 (2.4914 iter/s, 4.81657s/12 iters), loss = 5.01575 I0409 22:52:13.669021 4221 solver.cpp:237] Train net output #0: loss = 5.01575 (* 1 = 5.01575 loss) I0409 22:52:13.669030 4221 sgd_solver.cpp:105] Iteration 552, lr = 0.00896423 I0409 22:52:18.496017 4221 solver.cpp:218] Iteration 564 (2.48613 iter/s, 4.82678s/12 iters), loss = 4.89903 I0409 22:52:18.496078 4221 solver.cpp:237] Train net output #0: loss = 4.89903 (* 1 = 4.89903 loss) I0409 22:52:18.496091 4221 sgd_solver.cpp:105] Iteration 564, lr = 0.00894294 I0409 22:52:23.332466 4221 solver.cpp:218] Iteration 576 (2.4813 iter/s, 4.83617s/12 iters), loss = 4.95518 I0409 22:52:23.332526 4221 solver.cpp:237] Train net output #0: loss = 4.95518 (* 1 = 4.95518 loss) I0409 22:52:23.332540 4221 sgd_solver.cpp:105] Iteration 576, lr = 0.00892171 I0409 22:52:28.169344 4221 solver.cpp:218] Iteration 588 (2.48107 iter/s, 4.83661s/12 iters), loss = 4.75715 I0409 22:52:28.169390 4221 solver.cpp:237] Train net output #0: loss = 4.75715 (* 1 = 4.75715 loss) I0409 22:52:28.169401 4221 sgd_solver.cpp:105] Iteration 588, lr = 0.00890053 I0409 22:52:33.034236 4221 solver.cpp:218] Iteration 600 (2.46678 iter/s, 4.86464s/12 iters), loss = 4.92039 I0409 22:52:33.034269 4221 solver.cpp:237] Train net output #0: loss = 4.92039 (* 1 = 4.92039 loss) I0409 22:52:33.034278 4221 sgd_solver.cpp:105] Iteration 600, lr = 0.0088794 I0409 22:52:35.355270 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 22:52:37.422019 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_612.caffemodel I0409 22:52:39.738430 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_612.solverstate I0409 22:52:41.540009 4221 solver.cpp:330] Iteration 612, Testing net (#0) I0409 22:52:41.540031 4221 net.cpp:676] Ignoring source layer train-data I0409 22:52:45.889107 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 22:52:46.177991 4221 solver.cpp:397] Test net output #0: accuracy = 0.03125 I0409 22:52:46.178041 4221 solver.cpp:397] Test net output #1: loss = 4.92251 (* 1 = 4.92251 loss) I0409 22:52:46.261435 4221 solver.cpp:218] Iteration 612 (0.907262 iter/s, 13.2266s/12 iters), loss = 4.83588 I0409 22:52:46.261487 4221 solver.cpp:237] Train net output #0: loss = 4.83588 (* 1 = 4.83588 loss) I0409 22:52:46.261497 4221 sgd_solver.cpp:105] Iteration 612, lr = 0.00885831 I0409 22:52:50.409279 4221 solver.cpp:218] Iteration 624 (2.89323 iter/s, 4.14761s/12 iters), loss = 4.77282 I0409 22:52:50.409332 4221 solver.cpp:237] Train net output #0: loss = 4.77282 (* 1 = 4.77282 loss) I0409 22:52:50.409344 4221 sgd_solver.cpp:105] Iteration 624, lr = 0.00883728 I0409 22:52:55.247087 4221 solver.cpp:218] Iteration 636 (2.4806 iter/s, 4.83755s/12 iters), loss = 4.75799 I0409 22:52:55.247138 4221 solver.cpp:237] Train net output #0: loss = 4.75799 (* 1 = 4.75799 loss) I0409 22:52:55.247150 4221 sgd_solver.cpp:105] Iteration 636, lr = 0.0088163 I0409 22:53:00.069984 4221 solver.cpp:218] Iteration 648 (2.48827 iter/s, 4.82263s/12 iters), loss = 4.96856 I0409 22:53:00.070036 4221 solver.cpp:237] Train net output #0: loss = 4.96856 (* 1 = 4.96856 loss) I0409 22:53:00.070048 4221 sgd_solver.cpp:105] Iteration 648, lr = 0.00879537 I0409 22:53:04.914465 4221 solver.cpp:218] Iteration 660 (2.47718 iter/s, 4.84422s/12 iters), loss = 4.84366 I0409 22:53:04.914510 4221 solver.cpp:237] Train net output #0: loss = 4.84366 (* 1 = 4.84366 loss) I0409 22:53:04.914517 4221 sgd_solver.cpp:105] Iteration 660, lr = 0.00877449 I0409 22:53:09.741343 4221 solver.cpp:218] Iteration 672 (2.48621 iter/s, 4.82662s/12 iters), loss = 4.66519 I0409 22:53:09.741479 4221 solver.cpp:237] Train net output #0: loss = 4.66519 (* 1 = 4.66519 loss) I0409 22:53:09.741492 4221 sgd_solver.cpp:105] Iteration 672, lr = 0.00875366 I0409 22:53:14.109205 4221 blocking_queue.cpp:49] Waiting for data I0409 22:53:14.548789 4221 solver.cpp:218] Iteration 684 (2.49631 iter/s, 4.8071s/12 iters), loss = 4.67076 I0409 22:53:14.548846 4221 solver.cpp:237] Train net output #0: loss = 4.67076 (* 1 = 4.67076 loss) I0409 22:53:14.548858 4221 sgd_solver.cpp:105] Iteration 684, lr = 0.00873287 I0409 22:53:19.358399 4221 solver.cpp:218] Iteration 696 (2.49515 iter/s, 4.80934s/12 iters), loss = 4.82814 I0409 22:53:19.358464 4221 solver.cpp:237] Train net output #0: loss = 4.82814 (* 1 = 4.82814 loss) I0409 22:53:19.358476 4221 sgd_solver.cpp:105] Iteration 696, lr = 0.00871214 I0409 22:53:23.787799 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 22:53:24.155534 4221 solver.cpp:218] Iteration 708 (2.50164 iter/s, 4.79686s/12 iters), loss = 4.89735 I0409 22:53:24.155578 4221 solver.cpp:237] Train net output #0: loss = 4.89735 (* 1 = 4.89735 loss) I0409 22:53:24.155587 4221 sgd_solver.cpp:105] Iteration 708, lr = 0.00869145 I0409 22:53:26.139319 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_714.caffemodel I0409 22:53:27.325037 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_714.solverstate I0409 22:53:28.343668 4221 solver.cpp:330] Iteration 714, Testing net (#0) I0409 22:53:28.343691 4221 net.cpp:676] Ignoring source layer train-data I0409 22:53:32.510661 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 22:53:32.829365 4221 solver.cpp:397] Test net output #0: accuracy = 0.0435049 I0409 22:53:32.829411 4221 solver.cpp:397] Test net output #1: loss = 4.84813 (* 1 = 4.84813 loss) I0409 22:53:34.779381 4221 solver.cpp:218] Iteration 720 (1.12959 iter/s, 10.6234s/12 iters), loss = 4.78946 I0409 22:53:34.779440 4221 solver.cpp:237] Train net output #0: loss = 4.78946 (* 1 = 4.78946 loss) I0409 22:53:34.779453 4221 sgd_solver.cpp:105] Iteration 720, lr = 0.00867082 I0409 22:53:39.687414 4221 solver.cpp:218] Iteration 732 (2.44511 iter/s, 4.90776s/12 iters), loss = 4.63193 I0409 22:53:39.687465 4221 solver.cpp:237] Train net output #0: loss = 4.63193 (* 1 = 4.63193 loss) I0409 22:53:39.687475 4221 sgd_solver.cpp:105] Iteration 732, lr = 0.00865023 I0409 22:53:44.539405 4221 solver.cpp:218] Iteration 744 (2.47335 iter/s, 4.85173s/12 iters), loss = 4.82361 I0409 22:53:44.539517 4221 solver.cpp:237] Train net output #0: loss = 4.82361 (* 1 = 4.82361 loss) I0409 22:53:44.539526 4221 sgd_solver.cpp:105] Iteration 744, lr = 0.0086297 I0409 22:53:49.455185 4221 solver.cpp:218] Iteration 756 (2.44128 iter/s, 4.91545s/12 iters), loss = 4.80122 I0409 22:53:49.455233 4221 solver.cpp:237] Train net output #0: loss = 4.80122 (* 1 = 4.80122 loss) I0409 22:53:49.455243 4221 sgd_solver.cpp:105] Iteration 756, lr = 0.00860921 I0409 22:53:54.364554 4221 solver.cpp:218] Iteration 768 (2.44444 iter/s, 4.9091s/12 iters), loss = 4.68829 I0409 22:53:54.364612 4221 solver.cpp:237] Train net output #0: loss = 4.68829 (* 1 = 4.68829 loss) I0409 22:53:54.364624 4221 sgd_solver.cpp:105] Iteration 768, lr = 0.00858877 I0409 22:53:59.255071 4221 solver.cpp:218] Iteration 780 (2.45387 iter/s, 4.89024s/12 iters), loss = 4.7231 I0409 22:53:59.255138 4221 solver.cpp:237] Train net output #0: loss = 4.7231 (* 1 = 4.7231 loss) I0409 22:53:59.255149 4221 sgd_solver.cpp:105] Iteration 780, lr = 0.00856838 I0409 22:54:04.030709 4221 solver.cpp:218] Iteration 792 (2.5129 iter/s, 4.77536s/12 iters), loss = 4.47977 I0409 22:54:04.030764 4221 solver.cpp:237] Train net output #0: loss = 4.47977 (* 1 = 4.47977 loss) I0409 22:54:04.030776 4221 sgd_solver.cpp:105] Iteration 792, lr = 0.00854803 I0409 22:54:08.824741 4221 solver.cpp:218] Iteration 804 (2.50325 iter/s, 4.79377s/12 iters), loss = 4.64771 I0409 22:54:08.824800 4221 solver.cpp:237] Train net output #0: loss = 4.64771 (* 1 = 4.64771 loss) I0409 22:54:08.824811 4221 sgd_solver.cpp:105] Iteration 804, lr = 0.00852774 I0409 22:54:10.501020 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 22:54:13.175520 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_816.caffemodel I0409 22:54:14.706593 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_816.solverstate I0409 22:54:15.276017 4221 solver.cpp:330] Iteration 816, Testing net (#0) I0409 22:54:15.276041 4221 net.cpp:676] Ignoring source layer train-data I0409 22:54:19.250862 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 22:54:19.624311 4221 solver.cpp:397] Test net output #0: accuracy = 0.0484069 I0409 22:54:19.624351 4221 solver.cpp:397] Test net output #1: loss = 4.69287 (* 1 = 4.69287 loss) I0409 22:54:19.707929 4221 solver.cpp:218] Iteration 816 (1.10267 iter/s, 10.8827s/12 iters), loss = 4.66028 I0409 22:54:19.707983 4221 solver.cpp:237] Train net output #0: loss = 4.66028 (* 1 = 4.66028 loss) I0409 22:54:19.707993 4221 sgd_solver.cpp:105] Iteration 816, lr = 0.00850749 I0409 22:54:23.876225 4221 solver.cpp:218] Iteration 828 (2.87904 iter/s, 4.16806s/12 iters), loss = 4.94076 I0409 22:54:23.876272 4221 solver.cpp:237] Train net output #0: loss = 4.94076 (* 1 = 4.94076 loss) I0409 22:54:23.876281 4221 sgd_solver.cpp:105] Iteration 828, lr = 0.00848729 I0409 22:54:28.789570 4221 solver.cpp:218] Iteration 840 (2.44246 iter/s, 4.91308s/12 iters), loss = 4.26236 I0409 22:54:28.789619 4221 solver.cpp:237] Train net output #0: loss = 4.26236 (* 1 = 4.26236 loss) I0409 22:54:28.789628 4221 sgd_solver.cpp:105] Iteration 840, lr = 0.00846714 I0409 22:54:33.698683 4221 solver.cpp:218] Iteration 852 (2.44457 iter/s, 4.90885s/12 iters), loss = 4.58574 I0409 22:54:33.698751 4221 solver.cpp:237] Train net output #0: loss = 4.58574 (* 1 = 4.58574 loss) I0409 22:54:33.698763 4221 sgd_solver.cpp:105] Iteration 852, lr = 0.00844704 I0409 22:54:38.667444 4221 solver.cpp:218] Iteration 864 (2.41523 iter/s, 4.96847s/12 iters), loss = 4.62392 I0409 22:54:38.667520 4221 solver.cpp:237] Train net output #0: loss = 4.62392 (* 1 = 4.62392 loss) I0409 22:54:38.667537 4221 sgd_solver.cpp:105] Iteration 864, lr = 0.00842698 I0409 22:54:43.550221 4221 solver.cpp:218] Iteration 876 (2.45776 iter/s, 4.88249s/12 iters), loss = 4.4092 I0409 22:54:43.550282 4221 solver.cpp:237] Train net output #0: loss = 4.4092 (* 1 = 4.4092 loss) I0409 22:54:43.550294 4221 sgd_solver.cpp:105] Iteration 876, lr = 0.00840698 I0409 22:54:48.370714 4221 solver.cpp:218] Iteration 888 (2.48951 iter/s, 4.82022s/12 iters), loss = 4.488 I0409 22:54:48.370887 4221 solver.cpp:237] Train net output #0: loss = 4.488 (* 1 = 4.488 loss) I0409 22:54:48.370899 4221 sgd_solver.cpp:105] Iteration 888, lr = 0.00838702 I0409 22:54:53.173985 4221 solver.cpp:218] Iteration 900 (2.4985 iter/s, 4.80287s/12 iters), loss = 4.6424 I0409 22:54:53.174044 4221 solver.cpp:237] Train net output #0: loss = 4.6424 (* 1 = 4.6424 loss) I0409 22:54:53.174057 4221 sgd_solver.cpp:105] Iteration 900, lr = 0.0083671 I0409 22:54:56.885756 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 22:54:57.953279 4221 solver.cpp:218] Iteration 912 (2.51097 iter/s, 4.77902s/12 iters), loss = 4.37309 I0409 22:54:57.953341 4221 solver.cpp:237] Train net output #0: loss = 4.37309 (* 1 = 4.37309 loss) I0409 22:54:57.953351 4221 sgd_solver.cpp:105] Iteration 912, lr = 0.00834724 I0409 22:54:59.912039 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_918.caffemodel I0409 22:55:02.091667 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_918.solverstate I0409 22:55:04.285298 4221 solver.cpp:330] Iteration 918, Testing net (#0) I0409 22:55:04.285322 4221 net.cpp:676] Ignoring source layer train-data I0409 22:55:08.492120 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 22:55:08.902536 4221 solver.cpp:397] Test net output #0: accuracy = 0.057598 I0409 22:55:08.902586 4221 solver.cpp:397] Test net output #1: loss = 4.59414 (* 1 = 4.59414 loss) I0409 22:55:10.773571 4221 solver.cpp:218] Iteration 924 (0.93606 iter/s, 12.8197s/12 iters), loss = 4.34955 I0409 22:55:10.773628 4221 solver.cpp:237] Train net output #0: loss = 4.34955 (* 1 = 4.34955 loss) I0409 22:55:10.773639 4221 sgd_solver.cpp:105] Iteration 924, lr = 0.00832742 I0409 22:55:15.875422 4221 solver.cpp:218] Iteration 936 (2.35222 iter/s, 5.10157s/12 iters), loss = 4.40821 I0409 22:55:15.875468 4221 solver.cpp:237] Train net output #0: loss = 4.40821 (* 1 = 4.40821 loss) I0409 22:55:15.875476 4221 sgd_solver.cpp:105] Iteration 936, lr = 0.00830765 I0409 22:55:20.815543 4221 solver.cpp:218] Iteration 948 (2.42922 iter/s, 4.93985s/12 iters), loss = 4.60082 I0409 22:55:20.815662 4221 solver.cpp:237] Train net output #0: loss = 4.60082 (* 1 = 4.60082 loss) I0409 22:55:20.815675 4221 sgd_solver.cpp:105] Iteration 948, lr = 0.00828793 I0409 22:55:25.852164 4221 solver.cpp:218] Iteration 960 (2.38271 iter/s, 5.03629s/12 iters), loss = 4.374 I0409 22:55:25.852202 4221 solver.cpp:237] Train net output #0: loss = 4.374 (* 1 = 4.374 loss) I0409 22:55:25.852210 4221 sgd_solver.cpp:105] Iteration 960, lr = 0.00826825 I0409 22:55:30.711621 4221 solver.cpp:218] Iteration 972 (2.46954 iter/s, 4.8592s/12 iters), loss = 4.31737 I0409 22:55:30.711679 4221 solver.cpp:237] Train net output #0: loss = 4.31737 (* 1 = 4.31737 loss) I0409 22:55:30.711691 4221 sgd_solver.cpp:105] Iteration 972, lr = 0.00824862 I0409 22:55:36.140674 4221 solver.cpp:218] Iteration 984 (2.21045 iter/s, 5.42877s/12 iters), loss = 4.24516 I0409 22:55:36.140710 4221 solver.cpp:237] Train net output #0: loss = 4.24516 (* 1 = 4.24516 loss) I0409 22:55:36.140718 4221 sgd_solver.cpp:105] Iteration 984, lr = 0.00822903 I0409 22:55:41.250609 4221 solver.cpp:218] Iteration 996 (2.34849 iter/s, 5.10967s/12 iters), loss = 4.30323 I0409 22:55:41.250659 4221 solver.cpp:237] Train net output #0: loss = 4.30323 (* 1 = 4.30323 loss) I0409 22:55:41.250669 4221 sgd_solver.cpp:105] Iteration 996, lr = 0.0082095 I0409 22:55:46.160804 4221 solver.cpp:218] Iteration 1008 (2.44403 iter/s, 4.90993s/12 iters), loss = 4.35442 I0409 22:55:46.160854 4221 solver.cpp:237] Train net output #0: loss = 4.35442 (* 1 = 4.35442 loss) I0409 22:55:46.160863 4221 sgd_solver.cpp:105] Iteration 1008, lr = 0.00819001 I0409 22:55:47.163786 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 22:55:50.625869 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_1020.caffemodel I0409 22:55:51.338021 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_1020.solverstate I0409 22:55:52.157647 4221 solver.cpp:330] Iteration 1020, Testing net (#0) I0409 22:55:52.157675 4221 net.cpp:676] Ignoring source layer train-data I0409 22:55:56.231642 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 22:55:56.663017 4221 solver.cpp:397] Test net output #0: accuracy = 0.0741422 I0409 22:55:56.663059 4221 solver.cpp:397] Test net output #1: loss = 4.45789 (* 1 = 4.45789 loss) I0409 22:55:56.746482 4221 solver.cpp:218] Iteration 1020 (1.13366 iter/s, 10.5852s/12 iters), loss = 4.16957 I0409 22:55:56.746537 4221 solver.cpp:237] Train net output #0: loss = 4.16957 (* 1 = 4.16957 loss) I0409 22:55:56.746548 4221 sgd_solver.cpp:105] Iteration 1020, lr = 0.00817056 I0409 22:56:00.893090 4221 solver.cpp:218] Iteration 1032 (2.8941 iter/s, 4.14637s/12 iters), loss = 4.32383 I0409 22:56:00.893142 4221 solver.cpp:237] Train net output #0: loss = 4.32383 (* 1 = 4.32383 loss) I0409 22:56:00.893151 4221 sgd_solver.cpp:105] Iteration 1032, lr = 0.00815116 I0409 22:56:05.917999 4221 solver.cpp:218] Iteration 1044 (2.38824 iter/s, 5.02463s/12 iters), loss = 4.38298 I0409 22:56:05.918052 4221 solver.cpp:237] Train net output #0: loss = 4.38298 (* 1 = 4.38298 loss) I0409 22:56:05.918062 4221 sgd_solver.cpp:105] Iteration 1044, lr = 0.00813181 I0409 22:56:11.021534 4221 solver.cpp:218] Iteration 1056 (2.35144 iter/s, 5.10326s/12 iters), loss = 4.21331 I0409 22:56:11.021589 4221 solver.cpp:237] Train net output #0: loss = 4.21331 (* 1 = 4.21331 loss) I0409 22:56:11.021598 4221 sgd_solver.cpp:105] Iteration 1056, lr = 0.0081125 I0409 22:56:15.867918 4221 solver.cpp:218] Iteration 1068 (2.47621 iter/s, 4.84611s/12 iters), loss = 4.23018 I0409 22:56:15.867975 4221 solver.cpp:237] Train net output #0: loss = 4.23018 (* 1 = 4.23018 loss) I0409 22:56:15.867987 4221 sgd_solver.cpp:105] Iteration 1068, lr = 0.00809324 I0409 22:56:20.688589 4221 solver.cpp:218] Iteration 1080 (2.48942 iter/s, 4.8204s/12 iters), loss = 4.06049 I0409 22:56:20.688639 4221 solver.cpp:237] Train net output #0: loss = 4.06049 (* 1 = 4.06049 loss) I0409 22:56:20.688650 4221 sgd_solver.cpp:105] Iteration 1080, lr = 0.00807403 I0409 22:56:25.468267 4221 solver.cpp:218] Iteration 1092 (2.51076 iter/s, 4.77942s/12 iters), loss = 4.25938 I0409 22:56:25.468384 4221 solver.cpp:237] Train net output #0: loss = 4.25938 (* 1 = 4.25938 loss) I0409 22:56:25.468394 4221 sgd_solver.cpp:105] Iteration 1092, lr = 0.00805486 I0409 22:56:30.298019 4221 solver.cpp:218] Iteration 1104 (2.48477 iter/s, 4.82942s/12 iters), loss = 4.22342 I0409 22:56:30.298076 4221 solver.cpp:237] Train net output #0: loss = 4.22342 (* 1 = 4.22342 loss) I0409 22:56:30.298087 4221 sgd_solver.cpp:105] Iteration 1104, lr = 0.00803573 I0409 22:56:33.313747 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 22:56:35.119938 4221 solver.cpp:218] Iteration 1116 (2.48877 iter/s, 4.82166s/12 iters), loss = 4.19847 I0409 22:56:35.119980 4221 solver.cpp:237] Train net output #0: loss = 4.19847 (* 1 = 4.19847 loss) I0409 22:56:35.119990 4221 sgd_solver.cpp:105] Iteration 1116, lr = 0.00801666 I0409 22:56:37.078619 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_1122.caffemodel I0409 22:56:37.813922 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_1122.solverstate I0409 22:56:38.367789 4221 solver.cpp:330] Iteration 1122, Testing net (#0) I0409 22:56:38.367821 4221 net.cpp:676] Ignoring source layer train-data I0409 22:56:42.652226 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 22:56:43.222638 4221 solver.cpp:397] Test net output #0: accuracy = 0.09375 I0409 22:56:43.222681 4221 solver.cpp:397] Test net output #1: loss = 4.33436 (* 1 = 4.33436 loss) I0409 22:56:44.950589 4221 solver.cpp:218] Iteration 1128 (1.22073 iter/s, 9.83019s/12 iters), loss = 4.15879 I0409 22:56:44.950644 4221 solver.cpp:237] Train net output #0: loss = 4.15879 (* 1 = 4.15879 loss) I0409 22:56:44.950654 4221 sgd_solver.cpp:105] Iteration 1128, lr = 0.00799762 I0409 22:56:49.826668 4221 solver.cpp:218] Iteration 1140 (2.46113 iter/s, 4.87582s/12 iters), loss = 4.29467 I0409 22:56:49.826710 4221 solver.cpp:237] Train net output #0: loss = 4.29467 (* 1 = 4.29467 loss) I0409 22:56:49.826720 4221 sgd_solver.cpp:105] Iteration 1140, lr = 0.00797863 I0409 22:56:54.660005 4221 solver.cpp:218] Iteration 1152 (2.48288 iter/s, 4.83309s/12 iters), loss = 3.96556 I0409 22:56:54.660043 4221 solver.cpp:237] Train net output #0: loss = 3.96556 (* 1 = 3.96556 loss) I0409 22:56:54.660051 4221 sgd_solver.cpp:105] Iteration 1152, lr = 0.00795969 I0409 22:56:59.635874 4221 solver.cpp:218] Iteration 1164 (2.41176 iter/s, 4.97561s/12 iters), loss = 4.13805 I0409 22:56:59.636023 4221 solver.cpp:237] Train net output #0: loss = 4.13805 (* 1 = 4.13805 loss) I0409 22:56:59.636037 4221 sgd_solver.cpp:105] Iteration 1164, lr = 0.00794079 I0409 22:57:04.660001 4221 solver.cpp:218] Iteration 1176 (2.38865 iter/s, 5.02377s/12 iters), loss = 4.31163 I0409 22:57:04.660059 4221 solver.cpp:237] Train net output #0: loss = 4.31163 (* 1 = 4.31163 loss) I0409 22:57:04.660071 4221 sgd_solver.cpp:105] Iteration 1176, lr = 0.00792194 I0409 22:57:09.637526 4221 solver.cpp:218] Iteration 1188 (2.41097 iter/s, 4.97725s/12 iters), loss = 4.17388 I0409 22:57:09.637583 4221 solver.cpp:237] Train net output #0: loss = 4.17388 (* 1 = 4.17388 loss) I0409 22:57:09.637593 4221 sgd_solver.cpp:105] Iteration 1188, lr = 0.00790313 I0409 22:57:14.748162 4221 solver.cpp:218] Iteration 1200 (2.34817 iter/s, 5.11036s/12 iters), loss = 4.14007 I0409 22:57:14.748211 4221 solver.cpp:237] Train net output #0: loss = 4.14007 (* 1 = 4.14007 loss) I0409 22:57:14.748220 4221 sgd_solver.cpp:105] Iteration 1200, lr = 0.00788437 I0409 22:57:20.250819 4221 solver.cpp:218] Iteration 1212 (2.18088 iter/s, 5.50236s/12 iters), loss = 4.08061 I0409 22:57:20.250874 4221 solver.cpp:237] Train net output #0: loss = 4.08061 (* 1 = 4.08061 loss) I0409 22:57:20.250885 4221 sgd_solver.cpp:105] Iteration 1212, lr = 0.00786565 I0409 22:57:20.529253 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 22:57:24.722190 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_1224.caffemodel I0409 22:57:25.987706 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_1224.solverstate I0409 22:57:27.781911 4221 solver.cpp:330] Iteration 1224, Testing net (#0) I0409 22:57:27.781937 4221 net.cpp:676] Ignoring source layer train-data I0409 22:57:31.647060 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 22:57:32.157392 4221 solver.cpp:397] Test net output #0: accuracy = 0.112132 I0409 22:57:32.157438 4221 solver.cpp:397] Test net output #1: loss = 4.26772 (* 1 = 4.26772 loss) I0409 22:57:32.239333 4221 solver.cpp:218] Iteration 1224 (1.001 iter/s, 11.988s/12 iters), loss = 4.2109 I0409 22:57:32.239392 4221 solver.cpp:237] Train net output #0: loss = 4.2109 (* 1 = 4.2109 loss) I0409 22:57:32.239403 4221 sgd_solver.cpp:105] Iteration 1224, lr = 0.00784697 I0409 22:57:36.387217 4221 solver.cpp:218] Iteration 1236 (2.89321 iter/s, 4.14764s/12 iters), loss = 4.15357 I0409 22:57:36.387274 4221 solver.cpp:237] Train net output #0: loss = 4.15357 (* 1 = 4.15357 loss) I0409 22:57:36.387285 4221 sgd_solver.cpp:105] Iteration 1236, lr = 0.00782834 I0409 22:57:41.242990 4221 solver.cpp:218] Iteration 1248 (2.47142 iter/s, 4.85551s/12 iters), loss = 4.06025 I0409 22:57:41.243046 4221 solver.cpp:237] Train net output #0: loss = 4.06025 (* 1 = 4.06025 loss) I0409 22:57:41.243058 4221 sgd_solver.cpp:105] Iteration 1248, lr = 0.00780976 I0409 22:57:46.080520 4221 solver.cpp:218] Iteration 1260 (2.48074 iter/s, 4.83727s/12 iters), loss = 4.11742 I0409 22:57:46.080565 4221 solver.cpp:237] Train net output #0: loss = 4.11742 (* 1 = 4.11742 loss) I0409 22:57:46.080574 4221 sgd_solver.cpp:105] Iteration 1260, lr = 0.00779122 I0409 22:57:51.029592 4221 solver.cpp:218] Iteration 1272 (2.42482 iter/s, 4.94881s/12 iters), loss = 3.91593 I0409 22:57:51.029637 4221 solver.cpp:237] Train net output #0: loss = 3.91593 (* 1 = 3.91593 loss) I0409 22:57:51.029644 4221 sgd_solver.cpp:105] Iteration 1272, lr = 0.00777272 I0409 22:57:55.870749 4221 solver.cpp:218] Iteration 1284 (2.47888 iter/s, 4.8409s/12 iters), loss = 4.09871 I0409 22:57:55.870796 4221 solver.cpp:237] Train net output #0: loss = 4.09871 (* 1 = 4.09871 loss) I0409 22:57:55.870808 4221 sgd_solver.cpp:105] Iteration 1284, lr = 0.00775426 I0409 22:58:00.776712 4221 solver.cpp:218] Iteration 1296 (2.44613 iter/s, 4.9057s/12 iters), loss = 3.82859 I0409 22:58:00.776760 4221 solver.cpp:237] Train net output #0: loss = 3.82859 (* 1 = 3.82859 loss) I0409 22:58:00.776768 4221 sgd_solver.cpp:105] Iteration 1296, lr = 0.00773585 I0409 22:58:05.679172 4221 solver.cpp:218] Iteration 1308 (2.44789 iter/s, 4.90219s/12 iters), loss = 4.02337 I0409 22:58:05.679308 4221 solver.cpp:237] Train net output #0: loss = 4.02337 (* 1 = 4.02337 loss) I0409 22:58:05.679317 4221 sgd_solver.cpp:105] Iteration 1308, lr = 0.00771749 I0409 22:58:08.154258 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 22:58:10.586241 4221 solver.cpp:218] Iteration 1320 (2.44563 iter/s, 4.90672s/12 iters), loss = 3.83804 I0409 22:58:10.586288 4221 solver.cpp:237] Train net output #0: loss = 3.83804 (* 1 = 3.83804 loss) I0409 22:58:10.586297 4221 sgd_solver.cpp:105] Iteration 1320, lr = 0.00769916 I0409 22:58:12.588752 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_1326.caffemodel I0409 22:58:13.274370 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_1326.solverstate I0409 22:58:13.783205 4221 solver.cpp:330] Iteration 1326, Testing net (#0) I0409 22:58:13.783224 4221 net.cpp:676] Ignoring source layer train-data I0409 22:58:17.661478 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 22:58:18.376577 4221 solver.cpp:397] Test net output #0: accuracy = 0.125 I0409 22:58:18.376607 4221 solver.cpp:397] Test net output #1: loss = 4.06494 (* 1 = 4.06494 loss) I0409 22:58:20.211143 4221 solver.cpp:218] Iteration 1332 (1.24682 iter/s, 9.62445s/12 iters), loss = 3.71032 I0409 22:58:20.211187 4221 solver.cpp:237] Train net output #0: loss = 3.71032 (* 1 = 3.71032 loss) I0409 22:58:20.211195 4221 sgd_solver.cpp:105] Iteration 1332, lr = 0.00768088 I0409 22:58:25.189129 4221 solver.cpp:218] Iteration 1344 (2.41074 iter/s, 4.97772s/12 iters), loss = 3.72013 I0409 22:58:25.189184 4221 solver.cpp:237] Train net output #0: loss = 3.72013 (* 1 = 3.72013 loss) I0409 22:58:25.189195 4221 sgd_solver.cpp:105] Iteration 1344, lr = 0.00766265 I0409 22:58:30.094604 4221 solver.cpp:218] Iteration 1356 (2.44638 iter/s, 4.90521s/12 iters), loss = 3.87589 I0409 22:58:30.094664 4221 solver.cpp:237] Train net output #0: loss = 3.87589 (* 1 = 3.87589 loss) I0409 22:58:30.094676 4221 sgd_solver.cpp:105] Iteration 1356, lr = 0.00764446 I0409 22:58:34.886184 4221 solver.cpp:218] Iteration 1368 (2.50454 iter/s, 4.79131s/12 iters), loss = 3.86457 I0409 22:58:34.886242 4221 solver.cpp:237] Train net output #0: loss = 3.86457 (* 1 = 3.86457 loss) I0409 22:58:34.886253 4221 sgd_solver.cpp:105] Iteration 1368, lr = 0.00762631 I0409 22:58:34.886507 4221 blocking_queue.cpp:49] Waiting for data I0409 22:58:39.841614 4221 solver.cpp:218] Iteration 1380 (2.42172 iter/s, 4.95516s/12 iters), loss = 3.85109 I0409 22:58:39.841730 4221 solver.cpp:237] Train net output #0: loss = 3.85109 (* 1 = 3.85109 loss) I0409 22:58:39.841742 4221 sgd_solver.cpp:105] Iteration 1380, lr = 0.0076082 I0409 22:58:44.724138 4221 solver.cpp:218] Iteration 1392 (2.45791 iter/s, 4.88219s/12 iters), loss = 3.78409 I0409 22:58:44.724195 4221 solver.cpp:237] Train net output #0: loss = 3.78409 (* 1 = 3.78409 loss) I0409 22:58:44.724206 4221 sgd_solver.cpp:105] Iteration 1392, lr = 0.00759014 I0409 22:58:49.811645 4221 solver.cpp:218] Iteration 1404 (2.35885 iter/s, 5.08723s/12 iters), loss = 3.99824 I0409 22:58:49.811702 4221 solver.cpp:237] Train net output #0: loss = 3.99824 (* 1 = 3.99824 loss) I0409 22:58:49.811712 4221 sgd_solver.cpp:105] Iteration 1404, lr = 0.00757212 I0409 22:58:54.326213 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 22:58:54.667807 4221 solver.cpp:218] Iteration 1416 (2.47122 iter/s, 4.8559s/12 iters), loss = 3.59998 I0409 22:58:54.667856 4221 solver.cpp:237] Train net output #0: loss = 3.59998 (* 1 = 3.59998 loss) I0409 22:58:54.667865 4221 sgd_solver.cpp:105] Iteration 1416, lr = 0.00755414 I0409 22:58:59.035248 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_1428.caffemodel I0409 22:59:00.481017 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_1428.solverstate I0409 22:59:01.641072 4221 solver.cpp:330] Iteration 1428, Testing net (#0) I0409 22:59:01.641098 4221 net.cpp:676] Ignoring source layer train-data I0409 22:59:05.538228 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 22:59:06.126914 4221 solver.cpp:397] Test net output #0: accuracy = 0.139706 I0409 22:59:06.126943 4221 solver.cpp:397] Test net output #1: loss = 3.97348 (* 1 = 3.97348 loss) I0409 22:59:06.210090 4221 solver.cpp:218] Iteration 1428 (1.0397 iter/s, 11.5417s/12 iters), loss = 3.63388 I0409 22:59:06.210144 4221 solver.cpp:237] Train net output #0: loss = 3.63388 (* 1 = 3.63388 loss) I0409 22:59:06.210152 4221 sgd_solver.cpp:105] Iteration 1428, lr = 0.0075362 I0409 22:59:11.103802 4221 solver.cpp:218] Iteration 1440 (2.45226 iter/s, 4.89345s/12 iters), loss = 3.79379 I0409 22:59:11.103924 4221 solver.cpp:237] Train net output #0: loss = 3.79379 (* 1 = 3.79379 loss) I0409 22:59:11.103933 4221 sgd_solver.cpp:105] Iteration 1440, lr = 0.00751831 I0409 22:59:16.011729 4221 solver.cpp:218] Iteration 1452 (2.44519 iter/s, 4.90759s/12 iters), loss = 3.95379 I0409 22:59:16.011778 4221 solver.cpp:237] Train net output #0: loss = 3.95379 (* 1 = 3.95379 loss) I0409 22:59:16.011787 4221 sgd_solver.cpp:105] Iteration 1452, lr = 0.00750046 I0409 22:59:20.957409 4221 solver.cpp:218] Iteration 1464 (2.42649 iter/s, 4.94542s/12 iters), loss = 3.64374 I0409 22:59:20.957450 4221 solver.cpp:237] Train net output #0: loss = 3.64374 (* 1 = 3.64374 loss) I0409 22:59:20.957458 4221 sgd_solver.cpp:105] Iteration 1464, lr = 0.00748265 I0409 22:59:25.988780 4221 solver.cpp:218] Iteration 1476 (2.38516 iter/s, 5.03111s/12 iters), loss = 3.62728 I0409 22:59:25.988829 4221 solver.cpp:237] Train net output #0: loss = 3.62728 (* 1 = 3.62728 loss) I0409 22:59:25.988838 4221 sgd_solver.cpp:105] Iteration 1476, lr = 0.00746489 I0409 22:59:30.868441 4221 solver.cpp:218] Iteration 1488 (2.45932 iter/s, 4.8794s/12 iters), loss = 3.46235 I0409 22:59:30.868494 4221 solver.cpp:237] Train net output #0: loss = 3.46235 (* 1 = 3.46235 loss) I0409 22:59:30.868503 4221 sgd_solver.cpp:105] Iteration 1488, lr = 0.00744716 I0409 22:59:35.843410 4221 solver.cpp:218] Iteration 1500 (2.41221 iter/s, 4.97469s/12 iters), loss = 3.73683 I0409 22:59:35.843484 4221 solver.cpp:237] Train net output #0: loss = 3.73683 (* 1 = 3.73683 loss) I0409 22:59:35.843503 4221 sgd_solver.cpp:105] Iteration 1500, lr = 0.00742948 I0409 22:59:40.773986 4221 solver.cpp:218] Iteration 1512 (2.43394 iter/s, 4.93027s/12 iters), loss = 3.49215 I0409 22:59:40.774041 4221 solver.cpp:237] Train net output #0: loss = 3.49215 (* 1 = 3.49215 loss) I0409 22:59:40.774053 4221 sgd_solver.cpp:105] Iteration 1512, lr = 0.00741184 I0409 22:59:42.549768 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 22:59:45.682965 4221 solver.cpp:218] Iteration 1524 (2.44464 iter/s, 4.9087s/12 iters), loss = 3.73207 I0409 22:59:45.683017 4221 solver.cpp:237] Train net output #0: loss = 3.73207 (* 1 = 3.73207 loss) I0409 22:59:45.683027 4221 sgd_solver.cpp:105] Iteration 1524, lr = 0.00739425 I0409 22:59:47.648345 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_1530.caffemodel I0409 22:59:54.016691 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_1530.solverstate I0409 22:59:55.397037 4221 solver.cpp:330] Iteration 1530, Testing net (#0) I0409 22:59:55.397068 4221 net.cpp:676] Ignoring source layer train-data I0409 22:59:59.213768 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 22:59:59.849623 4221 solver.cpp:397] Test net output #0: accuracy = 0.134191 I0409 22:59:59.849658 4221 solver.cpp:397] Test net output #1: loss = 3.91781 (* 1 = 3.91781 loss) I0409 23:00:01.778877 4221 solver.cpp:218] Iteration 1536 (0.745564 iter/s, 16.0952s/12 iters), loss = 3.60845 I0409 23:00:01.778926 4221 solver.cpp:237] Train net output #0: loss = 3.60845 (* 1 = 3.60845 loss) I0409 23:00:01.778935 4221 sgd_solver.cpp:105] Iteration 1536, lr = 0.00737669 I0409 23:00:06.674006 4221 solver.cpp:218] Iteration 1548 (2.45155 iter/s, 4.89486s/12 iters), loss = 3.23984 I0409 23:00:06.674052 4221 solver.cpp:237] Train net output #0: loss = 3.23984 (* 1 = 3.23984 loss) I0409 23:00:06.674060 4221 sgd_solver.cpp:105] Iteration 1548, lr = 0.00735918 I0409 23:00:11.505342 4221 solver.cpp:218] Iteration 1560 (2.48392 iter/s, 4.83108s/12 iters), loss = 3.57996 I0409 23:00:11.505388 4221 solver.cpp:237] Train net output #0: loss = 3.57996 (* 1 = 3.57996 loss) I0409 23:00:11.505396 4221 sgd_solver.cpp:105] Iteration 1560, lr = 0.00734171 I0409 23:00:16.423166 4221 solver.cpp:218] Iteration 1572 (2.44024 iter/s, 4.91756s/12 iters), loss = 3.59601 I0409 23:00:16.425985 4221 solver.cpp:237] Train net output #0: loss = 3.59601 (* 1 = 3.59601 loss) I0409 23:00:16.426002 4221 sgd_solver.cpp:105] Iteration 1572, lr = 0.00732427 I0409 23:00:21.318006 4221 solver.cpp:218] Iteration 1584 (2.45306 iter/s, 4.89184s/12 iters), loss = 3.45653 I0409 23:00:21.318068 4221 solver.cpp:237] Train net output #0: loss = 3.45653 (* 1 = 3.45653 loss) I0409 23:00:21.318079 4221 sgd_solver.cpp:105] Iteration 1584, lr = 0.00730688 I0409 23:00:26.159416 4221 solver.cpp:218] Iteration 1596 (2.47876 iter/s, 4.84114s/12 iters), loss = 3.61984 I0409 23:00:26.159472 4221 solver.cpp:237] Train net output #0: loss = 3.61984 (* 1 = 3.61984 loss) I0409 23:00:26.159483 4221 sgd_solver.cpp:105] Iteration 1596, lr = 0.00728954 I0409 23:00:30.967566 4221 solver.cpp:218] Iteration 1608 (2.4959 iter/s, 4.80788s/12 iters), loss = 3.48152 I0409 23:00:30.967619 4221 solver.cpp:237] Train net output #0: loss = 3.48152 (* 1 = 3.48152 loss) I0409 23:00:30.967631 4221 sgd_solver.cpp:105] Iteration 1608, lr = 0.00727223 I0409 23:00:34.738517 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:00:35.781924 4221 solver.cpp:218] Iteration 1620 (2.49268 iter/s, 4.81409s/12 iters), loss = 3.34632 I0409 23:00:35.781997 4221 solver.cpp:237] Train net output #0: loss = 3.34632 (* 1 = 3.34632 loss) I0409 23:00:35.782006 4221 sgd_solver.cpp:105] Iteration 1620, lr = 0.00725496 I0409 23:00:40.192273 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_1632.caffemodel I0409 23:00:41.123222 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_1632.solverstate I0409 23:00:41.715690 4221 solver.cpp:330] Iteration 1632, Testing net (#0) I0409 23:00:41.715714 4221 net.cpp:676] Ignoring source layer train-data I0409 23:00:45.560245 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:00:46.258227 4221 solver.cpp:397] Test net output #0: accuracy = 0.16299 I0409 23:00:46.258275 4221 solver.cpp:397] Test net output #1: loss = 3.80218 (* 1 = 3.80218 loss) I0409 23:00:46.342689 4221 solver.cpp:218] Iteration 1632 (1.13634 iter/s, 10.5603s/12 iters), loss = 3.28325 I0409 23:00:46.342751 4221 solver.cpp:237] Train net output #0: loss = 3.28325 (* 1 = 3.28325 loss) I0409 23:00:46.342765 4221 sgd_solver.cpp:105] Iteration 1632, lr = 0.00723774 I0409 23:00:50.344692 4221 solver.cpp:218] Iteration 1644 (2.99867 iter/s, 4.00177s/12 iters), loss = 3.53041 I0409 23:00:50.344832 4221 solver.cpp:237] Train net output #0: loss = 3.53041 (* 1 = 3.53041 loss) I0409 23:00:50.344844 4221 sgd_solver.cpp:105] Iteration 1644, lr = 0.00722056 I0409 23:00:55.218924 4221 solver.cpp:218] Iteration 1656 (2.46211 iter/s, 4.87388s/12 iters), loss = 3.23318 I0409 23:00:55.218978 4221 solver.cpp:237] Train net output #0: loss = 3.23318 (* 1 = 3.23318 loss) I0409 23:00:55.218989 4221 sgd_solver.cpp:105] Iteration 1656, lr = 0.00720341 I0409 23:01:00.153124 4221 solver.cpp:218] Iteration 1668 (2.43214 iter/s, 4.93393s/12 iters), loss = 3.08833 I0409 23:01:00.153172 4221 solver.cpp:237] Train net output #0: loss = 3.08833 (* 1 = 3.08833 loss) I0409 23:01:00.153180 4221 sgd_solver.cpp:105] Iteration 1668, lr = 0.00718631 I0409 23:01:04.990857 4221 solver.cpp:218] Iteration 1680 (2.48063 iter/s, 4.83748s/12 iters), loss = 3.42887 I0409 23:01:04.990900 4221 solver.cpp:237] Train net output #0: loss = 3.42887 (* 1 = 3.42887 loss) I0409 23:01:04.990909 4221 sgd_solver.cpp:105] Iteration 1680, lr = 0.00716925 I0409 23:01:09.908602 4221 solver.cpp:218] Iteration 1692 (2.44027 iter/s, 4.91748s/12 iters), loss = 3.13693 I0409 23:01:09.908658 4221 solver.cpp:237] Train net output #0: loss = 3.13693 (* 1 = 3.13693 loss) I0409 23:01:09.908669 4221 sgd_solver.cpp:105] Iteration 1692, lr = 0.00715223 I0409 23:01:14.796869 4221 solver.cpp:218] Iteration 1704 (2.45499 iter/s, 4.888s/12 iters), loss = 2.88456 I0409 23:01:14.796926 4221 solver.cpp:237] Train net output #0: loss = 2.88456 (* 1 = 2.88456 loss) I0409 23:01:14.796937 4221 sgd_solver.cpp:105] Iteration 1704, lr = 0.00713525 I0409 23:01:19.684229 4221 solver.cpp:218] Iteration 1716 (2.45586 iter/s, 4.88628s/12 iters), loss = 3.37315 I0409 23:01:19.684278 4221 solver.cpp:237] Train net output #0: loss = 3.37315 (* 1 = 3.37315 loss) I0409 23:01:19.684289 4221 sgd_solver.cpp:105] Iteration 1716, lr = 0.00711831 I0409 23:01:20.843592 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:01:24.770344 4221 solver.cpp:218] Iteration 1728 (2.35949 iter/s, 5.08584s/12 iters), loss = 3.24472 I0409 23:01:24.770401 4221 solver.cpp:237] Train net output #0: loss = 3.24472 (* 1 = 3.24472 loss) I0409 23:01:24.770412 4221 sgd_solver.cpp:105] Iteration 1728, lr = 0.00710141 I0409 23:01:26.728034 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_1734.caffemodel I0409 23:01:27.713567 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_1734.solverstate I0409 23:01:28.390269 4221 solver.cpp:330] Iteration 1734, Testing net (#0) I0409 23:01:28.390298 4221 net.cpp:676] Ignoring source layer train-data I0409 23:01:32.089346 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:01:32.793486 4221 solver.cpp:397] Test net output #0: accuracy = 0.189338 I0409 23:01:32.793524 4221 solver.cpp:397] Test net output #1: loss = 3.66122 (* 1 = 3.66122 loss) I0409 23:01:34.497769 4221 solver.cpp:218] Iteration 1740 (1.23369 iter/s, 9.72695s/12 iters), loss = 3.20286 I0409 23:01:34.497836 4221 solver.cpp:237] Train net output #0: loss = 3.20286 (* 1 = 3.20286 loss) I0409 23:01:34.497848 4221 sgd_solver.cpp:105] Iteration 1740, lr = 0.00708455 I0409 23:01:39.438226 4221 solver.cpp:218] Iteration 1752 (2.42906 iter/s, 4.94017s/12 iters), loss = 3.32501 I0409 23:01:39.438279 4221 solver.cpp:237] Train net output #0: loss = 3.32501 (* 1 = 3.32501 loss) I0409 23:01:39.438289 4221 sgd_solver.cpp:105] Iteration 1752, lr = 0.00706773 I0409 23:01:44.356863 4221 solver.cpp:218] Iteration 1764 (2.43983 iter/s, 4.91837s/12 iters), loss = 3.01739 I0409 23:01:44.356920 4221 solver.cpp:237] Train net output #0: loss = 3.01739 (* 1 = 3.01739 loss) I0409 23:01:44.356930 4221 sgd_solver.cpp:105] Iteration 1764, lr = 0.00705094 I0409 23:01:49.393169 4221 solver.cpp:218] Iteration 1776 (2.38283 iter/s, 5.03603s/12 iters), loss = 3.09368 I0409 23:01:49.393226 4221 solver.cpp:237] Train net output #0: loss = 3.09368 (* 1 = 3.09368 loss) I0409 23:01:49.393237 4221 sgd_solver.cpp:105] Iteration 1776, lr = 0.0070342 I0409 23:01:54.266402 4221 solver.cpp:218] Iteration 1788 (2.46257 iter/s, 4.87296s/12 iters), loss = 3.48686 I0409 23:01:54.266527 4221 solver.cpp:237] Train net output #0: loss = 3.48686 (* 1 = 3.48686 loss) I0409 23:01:54.266536 4221 sgd_solver.cpp:105] Iteration 1788, lr = 0.0070175 I0409 23:01:59.158047 4221 solver.cpp:218] Iteration 1800 (2.45333 iter/s, 4.8913s/12 iters), loss = 3.09484 I0409 23:01:59.158097 4221 solver.cpp:237] Train net output #0: loss = 3.09484 (* 1 = 3.09484 loss) I0409 23:01:59.158109 4221 sgd_solver.cpp:105] Iteration 1800, lr = 0.00700084 I0409 23:02:04.166287 4221 solver.cpp:218] Iteration 1812 (2.39618 iter/s, 5.00797s/12 iters), loss = 3.02298 I0409 23:02:04.166337 4221 solver.cpp:237] Train net output #0: loss = 3.02298 (* 1 = 3.02298 loss) I0409 23:02:04.166350 4221 sgd_solver.cpp:105] Iteration 1812, lr = 0.00698422 I0409 23:02:07.228256 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:02:08.947100 4221 solver.cpp:218] Iteration 1824 (2.51017 iter/s, 4.78055s/12 iters), loss = 3.35961 I0409 23:02:08.947158 4221 solver.cpp:237] Train net output #0: loss = 3.35961 (* 1 = 3.35961 loss) I0409 23:02:08.947171 4221 sgd_solver.cpp:105] Iteration 1824, lr = 0.00696764 I0409 23:02:13.307348 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_1836.caffemodel I0409 23:02:14.947307 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_1836.solverstate I0409 23:02:16.446231 4221 solver.cpp:330] Iteration 1836, Testing net (#0) I0409 23:02:16.446256 4221 net.cpp:676] Ignoring source layer train-data I0409 23:02:20.192449 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:02:21.049209 4221 solver.cpp:397] Test net output #0: accuracy = 0.206495 I0409 23:02:21.049260 4221 solver.cpp:397] Test net output #1: loss = 3.57501 (* 1 = 3.57501 loss) I0409 23:02:21.132441 4221 solver.cpp:218] Iteration 1836 (0.984836 iter/s, 12.1848s/12 iters), loss = 2.96911 I0409 23:02:21.132498 4221 solver.cpp:237] Train net output #0: loss = 2.96911 (* 1 = 2.96911 loss) I0409 23:02:21.132509 4221 sgd_solver.cpp:105] Iteration 1836, lr = 0.0069511 I0409 23:02:25.225992 4221 solver.cpp:218] Iteration 1848 (2.93161 iter/s, 4.09331s/12 iters), loss = 2.85631 I0409 23:02:25.226073 4221 solver.cpp:237] Train net output #0: loss = 2.85631 (* 1 = 2.85631 loss) I0409 23:02:25.226086 4221 sgd_solver.cpp:105] Iteration 1848, lr = 0.00693459 I0409 23:02:30.021986 4221 solver.cpp:218] Iteration 1860 (2.50224 iter/s, 4.79571s/12 iters), loss = 3.00068 I0409 23:02:30.022032 4221 solver.cpp:237] Train net output #0: loss = 3.00068 (* 1 = 3.00068 loss) I0409 23:02:30.022040 4221 sgd_solver.cpp:105] Iteration 1860, lr = 0.00691813 I0409 23:02:34.927350 4221 solver.cpp:218] Iteration 1872 (2.44643 iter/s, 4.9051s/12 iters), loss = 3.04068 I0409 23:02:34.927400 4221 solver.cpp:237] Train net output #0: loss = 3.04068 (* 1 = 3.04068 loss) I0409 23:02:34.927412 4221 sgd_solver.cpp:105] Iteration 1872, lr = 0.0069017 I0409 23:02:39.820534 4221 solver.cpp:218] Iteration 1884 (2.45252 iter/s, 4.89292s/12 iters), loss = 3.04749 I0409 23:02:39.820590 4221 solver.cpp:237] Train net output #0: loss = 3.04749 (* 1 = 3.04749 loss) I0409 23:02:39.820603 4221 sgd_solver.cpp:105] Iteration 1884, lr = 0.00688532 I0409 23:02:44.673069 4221 solver.cpp:218] Iteration 1896 (2.47307 iter/s, 4.85226s/12 iters), loss = 3.04142 I0409 23:02:44.673120 4221 solver.cpp:237] Train net output #0: loss = 3.04142 (* 1 = 3.04142 loss) I0409 23:02:44.673130 4221 sgd_solver.cpp:105] Iteration 1896, lr = 0.00686897 I0409 23:02:49.554177 4221 solver.cpp:218] Iteration 1908 (2.45859 iter/s, 4.88084s/12 iters), loss = 2.89947 I0409 23:02:49.554229 4221 solver.cpp:237] Train net output #0: loss = 2.89947 (* 1 = 2.89947 loss) I0409 23:02:49.554241 4221 sgd_solver.cpp:105] Iteration 1908, lr = 0.00685266 I0409 23:02:54.460137 4221 solver.cpp:218] Iteration 1920 (2.44614 iter/s, 4.90569s/12 iters), loss = 2.98247 I0409 23:02:54.460188 4221 solver.cpp:237] Train net output #0: loss = 2.98247 (* 1 = 2.98247 loss) I0409 23:02:54.460199 4221 sgd_solver.cpp:105] Iteration 1920, lr = 0.00683639 I0409 23:02:54.774709 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:02:59.304410 4221 solver.cpp:218] Iteration 1932 (2.47728 iter/s, 4.84402s/12 iters), loss = 2.90548 I0409 23:02:59.304535 4221 solver.cpp:237] Train net output #0: loss = 2.90548 (* 1 = 2.90548 loss) I0409 23:02:59.304546 4221 sgd_solver.cpp:105] Iteration 1932, lr = 0.00682016 I0409 23:03:01.274400 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_1938.caffemodel I0409 23:03:02.003221 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_1938.solverstate I0409 23:03:02.522507 4221 solver.cpp:330] Iteration 1938, Testing net (#0) I0409 23:03:02.522534 4221 net.cpp:676] Ignoring source layer train-data I0409 23:03:06.184479 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:03:06.966831 4221 solver.cpp:397] Test net output #0: accuracy = 0.214461 I0409 23:03:06.966881 4221 solver.cpp:397] Test net output #1: loss = 3.55468 (* 1 = 3.55468 loss) I0409 23:03:08.805305 4221 solver.cpp:218] Iteration 1944 (1.26311 iter/s, 9.50036s/12 iters), loss = 2.64986 I0409 23:03:08.805379 4221 solver.cpp:237] Train net output #0: loss = 2.64986 (* 1 = 2.64986 loss) I0409 23:03:08.805393 4221 sgd_solver.cpp:105] Iteration 1944, lr = 0.00680397 I0409 23:03:13.689787 4221 solver.cpp:218] Iteration 1956 (2.4569 iter/s, 4.8842s/12 iters), loss = 2.68214 I0409 23:03:13.689836 4221 solver.cpp:237] Train net output #0: loss = 2.68214 (* 1 = 2.68214 loss) I0409 23:03:13.689846 4221 sgd_solver.cpp:105] Iteration 1956, lr = 0.00678782 I0409 23:03:18.521872 4221 solver.cpp:218] Iteration 1968 (2.48353 iter/s, 4.83182s/12 iters), loss = 2.62908 I0409 23:03:18.521922 4221 solver.cpp:237] Train net output #0: loss = 2.62908 (* 1 = 2.62908 loss) I0409 23:03:18.521932 4221 sgd_solver.cpp:105] Iteration 1968, lr = 0.0067717 I0409 23:03:23.448889 4221 solver.cpp:218] Iteration 1980 (2.43569 iter/s, 4.92674s/12 iters), loss = 2.93979 I0409 23:03:23.448956 4221 solver.cpp:237] Train net output #0: loss = 2.93979 (* 1 = 2.93979 loss) I0409 23:03:23.448968 4221 sgd_solver.cpp:105] Iteration 1980, lr = 0.00675562 I0409 23:03:28.214217 4221 solver.cpp:218] Iteration 1992 (2.51833 iter/s, 4.76505s/12 iters), loss = 2.83412 I0409 23:03:28.214267 4221 solver.cpp:237] Train net output #0: loss = 2.83412 (* 1 = 2.83412 loss) I0409 23:03:28.214278 4221 sgd_solver.cpp:105] Iteration 1992, lr = 0.00673958 I0409 23:03:32.983358 4221 solver.cpp:218] Iteration 2004 (2.51631 iter/s, 4.76888s/12 iters), loss = 2.69359 I0409 23:03:32.983458 4221 solver.cpp:237] Train net output #0: loss = 2.69359 (* 1 = 2.69359 loss) I0409 23:03:32.983469 4221 sgd_solver.cpp:105] Iteration 2004, lr = 0.00672358 I0409 23:03:37.777755 4221 solver.cpp:218] Iteration 2016 (2.50308 iter/s, 4.79409s/12 iters), loss = 2.88323 I0409 23:03:37.777815 4221 solver.cpp:237] Train net output #0: loss = 2.88323 (* 1 = 2.88323 loss) I0409 23:03:37.777827 4221 sgd_solver.cpp:105] Iteration 2016, lr = 0.00670762 I0409 23:03:40.283658 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:03:42.653092 4221 solver.cpp:218] Iteration 2028 (2.46151 iter/s, 4.87506s/12 iters), loss = 2.44453 I0409 23:03:42.653148 4221 solver.cpp:237] Train net output #0: loss = 2.44453 (* 1 = 2.44453 loss) I0409 23:03:42.653159 4221 sgd_solver.cpp:105] Iteration 2028, lr = 0.00669169 I0409 23:03:47.150702 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_2040.caffemodel I0409 23:03:47.931728 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_2040.solverstate I0409 23:03:48.432040 4221 solver.cpp:330] Iteration 2040, Testing net (#0) I0409 23:03:48.432065 4221 net.cpp:676] Ignoring source layer train-data I0409 23:03:52.293066 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:03:53.119355 4221 solver.cpp:397] Test net output #0: accuracy = 0.218137 I0409 23:03:53.119392 4221 solver.cpp:397] Test net output #1: loss = 3.50765 (* 1 = 3.50765 loss) I0409 23:03:53.202672 4221 solver.cpp:218] Iteration 2040 (1.13754 iter/s, 10.5491s/12 iters), loss = 2.74597 I0409 23:03:53.202733 4221 solver.cpp:237] Train net output #0: loss = 2.74597 (* 1 = 2.74597 loss) I0409 23:03:53.202745 4221 sgd_solver.cpp:105] Iteration 2040, lr = 0.00667581 I0409 23:03:57.359491 4221 solver.cpp:218] Iteration 2052 (2.88699 iter/s, 4.15658s/12 iters), loss = 3.0551 I0409 23:03:57.359537 4221 solver.cpp:237] Train net output #0: loss = 3.0551 (* 1 = 3.0551 loss) I0409 23:03:57.359546 4221 sgd_solver.cpp:105] Iteration 2052, lr = 0.00665996 I0409 23:03:57.731212 4221 blocking_queue.cpp:49] Waiting for data I0409 23:04:02.164252 4221 solver.cpp:218] Iteration 2064 (2.49766 iter/s, 4.8045s/12 iters), loss = 2.89701 I0409 23:04:02.164314 4221 solver.cpp:237] Train net output #0: loss = 2.89701 (* 1 = 2.89701 loss) I0409 23:04:02.164330 4221 sgd_solver.cpp:105] Iteration 2064, lr = 0.00664414 I0409 23:04:07.027915 4221 solver.cpp:218] Iteration 2076 (2.46741 iter/s, 4.86339s/12 iters), loss = 2.9996 I0409 23:04:07.028066 4221 solver.cpp:237] Train net output #0: loss = 2.9996 (* 1 = 2.9996 loss) I0409 23:04:07.028080 4221 sgd_solver.cpp:105] Iteration 2076, lr = 0.00662837 I0409 23:04:11.908614 4221 solver.cpp:218] Iteration 2088 (2.45885 iter/s, 4.88034s/12 iters), loss = 2.64106 I0409 23:04:11.908672 4221 solver.cpp:237] Train net output #0: loss = 2.64106 (* 1 = 2.64106 loss) I0409 23:04:11.908685 4221 sgd_solver.cpp:105] Iteration 2088, lr = 0.00661263 I0409 23:04:16.791747 4221 solver.cpp:218] Iteration 2100 (2.45757 iter/s, 4.88287s/12 iters), loss = 2.73515 I0409 23:04:16.791798 4221 solver.cpp:237] Train net output #0: loss = 2.73515 (* 1 = 2.73515 loss) I0409 23:04:16.791810 4221 sgd_solver.cpp:105] Iteration 2100, lr = 0.00659693 I0409 23:04:21.708158 4221 solver.cpp:218] Iteration 2112 (2.44094 iter/s, 4.91614s/12 iters), loss = 2.70102 I0409 23:04:21.708217 4221 solver.cpp:237] Train net output #0: loss = 2.70102 (* 1 = 2.70102 loss) I0409 23:04:21.708235 4221 sgd_solver.cpp:105] Iteration 2112, lr = 0.00658127 I0409 23:04:26.257009 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:04:26.566001 4221 solver.cpp:218] Iteration 2124 (2.47037 iter/s, 4.85758s/12 iters), loss = 2.5845 I0409 23:04:26.566046 4221 solver.cpp:237] Train net output #0: loss = 2.5845 (* 1 = 2.5845 loss) I0409 23:04:26.566057 4221 sgd_solver.cpp:105] Iteration 2124, lr = 0.00656564 I0409 23:04:31.408893 4221 solver.cpp:218] Iteration 2136 (2.47799 iter/s, 4.84264s/12 iters), loss = 2.35223 I0409 23:04:31.408943 4221 solver.cpp:237] Train net output #0: loss = 2.35223 (* 1 = 2.35223 loss) I0409 23:04:31.408955 4221 sgd_solver.cpp:105] Iteration 2136, lr = 0.00655006 I0409 23:04:33.392257 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_2142.caffemodel I0409 23:04:36.479161 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_2142.solverstate I0409 23:04:40.077123 4221 solver.cpp:330] Iteration 2142, Testing net (#0) I0409 23:04:40.077221 4221 net.cpp:676] Ignoring source layer train-data I0409 23:04:43.918247 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:04:44.780165 4221 solver.cpp:397] Test net output #0: accuracy = 0.223652 I0409 23:04:44.780206 4221 solver.cpp:397] Test net output #1: loss = 3.51202 (* 1 = 3.51202 loss) I0409 23:04:46.507182 4221 solver.cpp:218] Iteration 2148 (0.794827 iter/s, 15.0976s/12 iters), loss = 2.40508 I0409 23:04:46.507225 4221 solver.cpp:237] Train net output #0: loss = 2.40508 (* 1 = 2.40508 loss) I0409 23:04:46.507232 4221 sgd_solver.cpp:105] Iteration 2148, lr = 0.00653451 I0409 23:04:51.357843 4221 solver.cpp:218] Iteration 2160 (2.47402 iter/s, 4.85041s/12 iters), loss = 2.89141 I0409 23:04:51.357885 4221 solver.cpp:237] Train net output #0: loss = 2.89141 (* 1 = 2.89141 loss) I0409 23:04:51.357897 4221 sgd_solver.cpp:105] Iteration 2160, lr = 0.00651899 I0409 23:04:56.181560 4221 solver.cpp:218] Iteration 2172 (2.48784 iter/s, 4.82346s/12 iters), loss = 2.34389 I0409 23:04:56.181607 4221 solver.cpp:237] Train net output #0: loss = 2.34389 (* 1 = 2.34389 loss) I0409 23:04:56.181617 4221 sgd_solver.cpp:105] Iteration 2172, lr = 0.00650351 I0409 23:05:01.011524 4221 solver.cpp:218] Iteration 2184 (2.48462 iter/s, 4.8297s/12 iters), loss = 2.60752 I0409 23:05:01.011581 4221 solver.cpp:237] Train net output #0: loss = 2.60752 (* 1 = 2.60752 loss) I0409 23:05:01.011593 4221 sgd_solver.cpp:105] Iteration 2184, lr = 0.00648807 I0409 23:05:05.934476 4221 solver.cpp:218] Iteration 2196 (2.4377 iter/s, 4.92268s/12 iters), loss = 2.42197 I0409 23:05:05.934535 4221 solver.cpp:237] Train net output #0: loss = 2.42197 (* 1 = 2.42197 loss) I0409 23:05:05.934548 4221 sgd_solver.cpp:105] Iteration 2196, lr = 0.00647267 I0409 23:05:10.937059 4221 solver.cpp:218] Iteration 2208 (2.39889 iter/s, 5.00231s/12 iters), loss = 2.36415 I0409 23:05:10.937193 4221 solver.cpp:237] Train net output #0: loss = 2.36415 (* 1 = 2.36415 loss) I0409 23:05:10.937204 4221 sgd_solver.cpp:105] Iteration 2208, lr = 0.0064573 I0409 23:05:15.780432 4221 solver.cpp:218] Iteration 2220 (2.47779 iter/s, 4.84303s/12 iters), loss = 2.43993 I0409 23:05:15.780478 4221 solver.cpp:237] Train net output #0: loss = 2.43993 (* 1 = 2.43993 loss) I0409 23:05:15.780486 4221 sgd_solver.cpp:105] Iteration 2220, lr = 0.00644197 I0409 23:05:17.524461 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:05:20.611536 4221 solver.cpp:218] Iteration 2232 (2.48404 iter/s, 4.83085s/12 iters), loss = 2.48292 I0409 23:05:20.611590 4221 solver.cpp:237] Train net output #0: loss = 2.48292 (* 1 = 2.48292 loss) I0409 23:05:20.611603 4221 sgd_solver.cpp:105] Iteration 2232, lr = 0.00642668 I0409 23:05:25.177143 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_2244.caffemodel I0409 23:05:25.919067 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_2244.solverstate I0409 23:05:26.419538 4221 solver.cpp:330] Iteration 2244, Testing net (#0) I0409 23:05:26.419565 4221 net.cpp:676] Ignoring source layer train-data I0409 23:05:29.946548 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:05:30.851783 4221 solver.cpp:397] Test net output #0: accuracy = 0.250613 I0409 23:05:30.851830 4221 solver.cpp:397] Test net output #1: loss = 3.30737 (* 1 = 3.30737 loss) I0409 23:05:30.935035 4221 solver.cpp:218] Iteration 2244 (1.16245 iter/s, 10.323s/12 iters), loss = 2.43112 I0409 23:05:30.935086 4221 solver.cpp:237] Train net output #0: loss = 2.43112 (* 1 = 2.43112 loss) I0409 23:05:30.935096 4221 sgd_solver.cpp:105] Iteration 2244, lr = 0.00641142 I0409 23:05:35.239392 4221 solver.cpp:218] Iteration 2256 (2.78803 iter/s, 4.30411s/12 iters), loss = 2.17861 I0409 23:05:35.239444 4221 solver.cpp:237] Train net output #0: loss = 2.17861 (* 1 = 2.17861 loss) I0409 23:05:35.239454 4221 sgd_solver.cpp:105] Iteration 2256, lr = 0.0063962 I0409 23:05:40.026422 4221 solver.cpp:218] Iteration 2268 (2.50692 iter/s, 4.78676s/12 iters), loss = 2.80787 I0409 23:05:40.026479 4221 solver.cpp:237] Train net output #0: loss = 2.80787 (* 1 = 2.80787 loss) I0409 23:05:40.026492 4221 sgd_solver.cpp:105] Iteration 2268, lr = 0.00638101 I0409 23:05:44.840906 4221 solver.cpp:218] Iteration 2280 (2.49262 iter/s, 4.81421s/12 iters), loss = 2.46849 I0409 23:05:44.841022 4221 solver.cpp:237] Train net output #0: loss = 2.46849 (* 1 = 2.46849 loss) I0409 23:05:44.841033 4221 sgd_solver.cpp:105] Iteration 2280, lr = 0.00636586 I0409 23:05:49.675835 4221 solver.cpp:218] Iteration 2292 (2.48211 iter/s, 4.8346s/12 iters), loss = 2.53695 I0409 23:05:49.675889 4221 solver.cpp:237] Train net output #0: loss = 2.53695 (* 1 = 2.53695 loss) I0409 23:05:49.675899 4221 sgd_solver.cpp:105] Iteration 2292, lr = 0.00635075 I0409 23:05:54.484436 4221 solver.cpp:218] Iteration 2304 (2.49567 iter/s, 4.80834s/12 iters), loss = 2.53656 I0409 23:05:54.484483 4221 solver.cpp:237] Train net output #0: loss = 2.53656 (* 1 = 2.53656 loss) I0409 23:05:54.484494 4221 sgd_solver.cpp:105] Iteration 2304, lr = 0.00633567 I0409 23:05:59.310828 4221 solver.cpp:218] Iteration 2316 (2.48646 iter/s, 4.82614s/12 iters), loss = 2.36224 I0409 23:05:59.310871 4221 solver.cpp:237] Train net output #0: loss = 2.36224 (* 1 = 2.36224 loss) I0409 23:05:59.310879 4221 sgd_solver.cpp:105] Iteration 2316, lr = 0.00632063 I0409 23:06:03.119390 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:06:04.126931 4221 solver.cpp:218] Iteration 2328 (2.49177 iter/s, 4.81585s/12 iters), loss = 1.99291 I0409 23:06:04.126986 4221 solver.cpp:237] Train net output #0: loss = 1.99291 (* 1 = 1.99291 loss) I0409 23:06:04.126996 4221 sgd_solver.cpp:105] Iteration 2328, lr = 0.00630562 I0409 23:06:08.981230 4221 solver.cpp:218] Iteration 2340 (2.47217 iter/s, 4.85403s/12 iters), loss = 2.12249 I0409 23:06:08.981276 4221 solver.cpp:237] Train net output #0: loss = 2.12249 (* 1 = 2.12249 loss) I0409 23:06:08.981284 4221 sgd_solver.cpp:105] Iteration 2340, lr = 0.00629065 I0409 23:06:10.945909 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_2346.caffemodel I0409 23:06:11.647964 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_2346.solverstate I0409 23:06:12.142987 4221 solver.cpp:330] Iteration 2346, Testing net (#0) I0409 23:06:12.143004 4221 net.cpp:676] Ignoring source layer train-data I0409 23:06:15.960016 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:06:16.929265 4221 solver.cpp:397] Test net output #0: accuracy = 0.261029 I0409 23:06:16.929293 4221 solver.cpp:397] Test net output #1: loss = 3.37592 (* 1 = 3.37592 loss) I0409 23:06:18.696380 4221 solver.cpp:218] Iteration 2352 (1.23524 iter/s, 9.71469s/12 iters), loss = 2.46093 I0409 23:06:18.696431 4221 solver.cpp:237] Train net output #0: loss = 2.46093 (* 1 = 2.46093 loss) I0409 23:06:18.696444 4221 sgd_solver.cpp:105] Iteration 2352, lr = 0.00627571 I0409 23:06:23.707955 4221 solver.cpp:218] Iteration 2364 (2.39459 iter/s, 5.0113s/12 iters), loss = 2.45224 I0409 23:06:23.708012 4221 solver.cpp:237] Train net output #0: loss = 2.45224 (* 1 = 2.45224 loss) I0409 23:06:23.708025 4221 sgd_solver.cpp:105] Iteration 2364, lr = 0.00626081 I0409 23:06:28.584050 4221 solver.cpp:218] Iteration 2376 (2.46112 iter/s, 4.87583s/12 iters), loss = 2.00844 I0409 23:06:28.584095 4221 solver.cpp:237] Train net output #0: loss = 2.00844 (* 1 = 2.00844 loss) I0409 23:06:28.584105 4221 sgd_solver.cpp:105] Iteration 2376, lr = 0.00624595 I0409 23:06:33.436560 4221 solver.cpp:218] Iteration 2388 (2.47308 iter/s, 4.85225s/12 iters), loss = 2.2414 I0409 23:06:33.436607 4221 solver.cpp:237] Train net output #0: loss = 2.2414 (* 1 = 2.2414 loss) I0409 23:06:33.436619 4221 sgd_solver.cpp:105] Iteration 2388, lr = 0.00623112 I0409 23:06:38.292928 4221 solver.cpp:218] Iteration 2400 (2.47112 iter/s, 4.8561s/12 iters), loss = 2.19074 I0409 23:06:38.292992 4221 solver.cpp:237] Train net output #0: loss = 2.19074 (* 1 = 2.19074 loss) I0409 23:06:38.293004 4221 sgd_solver.cpp:105] Iteration 2400, lr = 0.00621633 I0409 23:06:43.151742 4221 solver.cpp:218] Iteration 2412 (2.46988 iter/s, 4.85854s/12 iters), loss = 2.12521 I0409 23:06:43.151783 4221 solver.cpp:237] Train net output #0: loss = 2.12521 (* 1 = 2.12521 loss) I0409 23:06:43.151791 4221 sgd_solver.cpp:105] Iteration 2412, lr = 0.00620157 I0409 23:06:48.011646 4221 solver.cpp:218] Iteration 2424 (2.46932 iter/s, 4.85964s/12 iters), loss = 2.14101 I0409 23:06:48.011801 4221 solver.cpp:237] Train net output #0: loss = 2.14101 (* 1 = 2.14101 loss) I0409 23:06:48.011816 4221 sgd_solver.cpp:105] Iteration 2424, lr = 0.00618684 I0409 23:06:49.059224 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:06:52.832018 4221 solver.cpp:218] Iteration 2436 (2.48962 iter/s, 4.82001s/12 iters), loss = 2.26537 I0409 23:06:52.832068 4221 solver.cpp:237] Train net output #0: loss = 2.26537 (* 1 = 2.26537 loss) I0409 23:06:52.832080 4221 sgd_solver.cpp:105] Iteration 2436, lr = 0.00617215 I0409 23:06:57.200848 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_2448.caffemodel I0409 23:06:58.482965 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_2448.solverstate I0409 23:06:59.537976 4221 solver.cpp:330] Iteration 2448, Testing net (#0) I0409 23:06:59.538007 4221 net.cpp:676] Ignoring source layer train-data I0409 23:07:03.007979 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:07:03.982746 4221 solver.cpp:397] Test net output #0: accuracy = 0.269608 I0409 23:07:03.982790 4221 solver.cpp:397] Test net output #1: loss = 3.302 (* 1 = 3.302 loss) I0409 23:07:04.066547 4221 solver.cpp:218] Iteration 2448 (1.06819 iter/s, 11.234s/12 iters), loss = 2.0459 I0409 23:07:04.066604 4221 solver.cpp:237] Train net output #0: loss = 2.0459 (* 1 = 2.0459 loss) I0409 23:07:04.066617 4221 sgd_solver.cpp:105] Iteration 2448, lr = 0.0061575 I0409 23:07:08.310890 4221 solver.cpp:218] Iteration 2460 (2.82746 iter/s, 4.2441s/12 iters), loss = 2.04063 I0409 23:07:08.310945 4221 solver.cpp:237] Train net output #0: loss = 2.04063 (* 1 = 2.04063 loss) I0409 23:07:08.310956 4221 sgd_solver.cpp:105] Iteration 2460, lr = 0.00614288 I0409 23:07:13.161761 4221 solver.cpp:218] Iteration 2472 (2.47392 iter/s, 4.8506s/12 iters), loss = 2.36591 I0409 23:07:13.161818 4221 solver.cpp:237] Train net output #0: loss = 2.36591 (* 1 = 2.36591 loss) I0409 23:07:13.161829 4221 sgd_solver.cpp:105] Iteration 2472, lr = 0.0061283 I0409 23:07:17.961061 4221 solver.cpp:218] Iteration 2484 (2.5005 iter/s, 4.79903s/12 iters), loss = 2.18935 I0409 23:07:17.961109 4221 solver.cpp:237] Train net output #0: loss = 2.18935 (* 1 = 2.18935 loss) I0409 23:07:17.961120 4221 sgd_solver.cpp:105] Iteration 2484, lr = 0.00611375 I0409 23:07:22.760542 4221 solver.cpp:218] Iteration 2496 (2.50041 iter/s, 4.79922s/12 iters), loss = 2.20747 I0409 23:07:22.760648 4221 solver.cpp:237] Train net output #0: loss = 2.20747 (* 1 = 2.20747 loss) I0409 23:07:22.760659 4221 sgd_solver.cpp:105] Iteration 2496, lr = 0.00609923 I0409 23:07:27.610536 4221 solver.cpp:218] Iteration 2508 (2.47439 iter/s, 4.84968s/12 iters), loss = 2.23414 I0409 23:07:27.610579 4221 solver.cpp:237] Train net output #0: loss = 2.23414 (* 1 = 2.23414 loss) I0409 23:07:27.610586 4221 sgd_solver.cpp:105] Iteration 2508, lr = 0.00608475 I0409 23:07:32.453457 4221 solver.cpp:218] Iteration 2520 (2.47797 iter/s, 4.84266s/12 iters), loss = 2.22019 I0409 23:07:32.453505 4221 solver.cpp:237] Train net output #0: loss = 2.22019 (* 1 = 2.22019 loss) I0409 23:07:32.453513 4221 sgd_solver.cpp:105] Iteration 2520, lr = 0.0060703 I0409 23:07:35.582684 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:07:37.301455 4221 solver.cpp:218] Iteration 2532 (2.47538 iter/s, 4.84774s/12 iters), loss = 2.09782 I0409 23:07:37.301506 4221 solver.cpp:237] Train net output #0: loss = 2.09782 (* 1 = 2.09782 loss) I0409 23:07:37.301518 4221 sgd_solver.cpp:105] Iteration 2532, lr = 0.00605589 I0409 23:07:42.470921 4221 solver.cpp:218] Iteration 2544 (2.32145 iter/s, 5.16919s/12 iters), loss = 2.01636 I0409 23:07:42.470979 4221 solver.cpp:237] Train net output #0: loss = 2.01636 (* 1 = 2.01636 loss) I0409 23:07:42.470993 4221 sgd_solver.cpp:105] Iteration 2544, lr = 0.00604151 I0409 23:07:44.426290 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_2550.caffemodel I0409 23:07:45.218546 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_2550.solverstate I0409 23:07:45.713943 4221 solver.cpp:330] Iteration 2550, Testing net (#0) I0409 23:07:45.713985 4221 net.cpp:676] Ignoring source layer train-data I0409 23:07:49.228451 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:07:50.248536 4221 solver.cpp:397] Test net output #0: accuracy = 0.268995 I0409 23:07:50.248584 4221 solver.cpp:397] Test net output #1: loss = 3.37247 (* 1 = 3.37247 loss) I0409 23:07:52.050474 4221 solver.cpp:218] Iteration 2556 (1.25273 iter/s, 9.57909s/12 iters), loss = 2.16173 I0409 23:07:52.050527 4221 solver.cpp:237] Train net output #0: loss = 2.16173 (* 1 = 2.16173 loss) I0409 23:07:52.050537 4221 sgd_solver.cpp:105] Iteration 2556, lr = 0.00602717 I0409 23:07:56.882403 4221 solver.cpp:218] Iteration 2568 (2.48362 iter/s, 4.83167s/12 iters), loss = 2.24688 I0409 23:07:56.882529 4221 solver.cpp:237] Train net output #0: loss = 2.24688 (* 1 = 2.24688 loss) I0409 23:07:56.882544 4221 sgd_solver.cpp:105] Iteration 2568, lr = 0.00601286 I0409 23:08:01.752511 4221 solver.cpp:218] Iteration 2580 (2.46419 iter/s, 4.86976s/12 iters), loss = 2.13058 I0409 23:08:01.752564 4221 solver.cpp:237] Train net output #0: loss = 2.13058 (* 1 = 2.13058 loss) I0409 23:08:01.752578 4221 sgd_solver.cpp:105] Iteration 2580, lr = 0.00599858 I0409 23:08:06.534072 4221 solver.cpp:218] Iteration 2592 (2.50978 iter/s, 4.7813s/12 iters), loss = 2.2466 I0409 23:08:06.534121 4221 solver.cpp:237] Train net output #0: loss = 2.2466 (* 1 = 2.2466 loss) I0409 23:08:06.534133 4221 sgd_solver.cpp:105] Iteration 2592, lr = 0.00598434 I0409 23:08:11.530032 4221 solver.cpp:218] Iteration 2604 (2.40207 iter/s, 4.9957s/12 iters), loss = 2.35307 I0409 23:08:11.530072 4221 solver.cpp:237] Train net output #0: loss = 2.35307 (* 1 = 2.35307 loss) I0409 23:08:11.530081 4221 sgd_solver.cpp:105] Iteration 2604, lr = 0.00597013 I0409 23:08:16.582602 4221 solver.cpp:218] Iteration 2616 (2.37515 iter/s, 5.05231s/12 iters), loss = 1.94119 I0409 23:08:16.582648 4221 solver.cpp:237] Train net output #0: loss = 1.94119 (* 1 = 1.94119 loss) I0409 23:08:16.582659 4221 sgd_solver.cpp:105] Iteration 2616, lr = 0.00595596 I0409 23:08:21.402753 4221 solver.cpp:218] Iteration 2628 (2.48968 iter/s, 4.81989s/12 iters), loss = 1.9786 I0409 23:08:21.402797 4221 solver.cpp:237] Train net output #0: loss = 1.9786 (* 1 = 1.9786 loss) I0409 23:08:21.402807 4221 sgd_solver.cpp:105] Iteration 2628, lr = 0.00594182 I0409 23:08:21.822044 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:08:26.234345 4221 solver.cpp:218] Iteration 2640 (2.48379 iter/s, 4.83133s/12 iters), loss = 2.30832 I0409 23:08:26.234395 4221 solver.cpp:237] Train net output #0: loss = 2.30832 (* 1 = 2.30832 loss) I0409 23:08:26.234402 4221 sgd_solver.cpp:105] Iteration 2640, lr = 0.00592771 I0409 23:08:30.640142 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_2652.caffemodel I0409 23:08:31.343240 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_2652.solverstate I0409 23:08:31.907001 4221 solver.cpp:330] Iteration 2652, Testing net (#0) I0409 23:08:31.907027 4221 net.cpp:676] Ignoring source layer train-data I0409 23:08:35.374286 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:08:36.438161 4221 solver.cpp:397] Test net output #0: accuracy = 0.276961 I0409 23:08:36.438202 4221 solver.cpp:397] Test net output #1: loss = 3.32635 (* 1 = 3.32635 loss) I0409 23:08:36.521405 4221 solver.cpp:218] Iteration 2652 (1.16657 iter/s, 10.2866s/12 iters), loss = 1.98099 I0409 23:08:36.521457 4221 solver.cpp:237] Train net output #0: loss = 1.98099 (* 1 = 1.98099 loss) I0409 23:08:36.521468 4221 sgd_solver.cpp:105] Iteration 2652, lr = 0.00591364 I0409 23:08:40.651286 4221 solver.cpp:218] Iteration 2664 (2.90582 iter/s, 4.12964s/12 iters), loss = 2.01121 I0409 23:08:40.651335 4221 solver.cpp:237] Train net output #0: loss = 2.01121 (* 1 = 2.01121 loss) I0409 23:08:40.651346 4221 sgd_solver.cpp:105] Iteration 2664, lr = 0.0058996 I0409 23:08:45.502904 4221 solver.cpp:218] Iteration 2676 (2.47354 iter/s, 4.85135s/12 iters), loss = 2.21237 I0409 23:08:45.502956 4221 solver.cpp:237] Train net output #0: loss = 2.21237 (* 1 = 2.21237 loss) I0409 23:08:45.502966 4221 sgd_solver.cpp:105] Iteration 2676, lr = 0.00588559 I0409 23:08:50.443037 4221 solver.cpp:218] Iteration 2688 (2.42922 iter/s, 4.93986s/12 iters), loss = 2.00698 I0409 23:08:50.443092 4221 solver.cpp:237] Train net output #0: loss = 2.00698 (* 1 = 2.00698 loss) I0409 23:08:50.443104 4221 sgd_solver.cpp:105] Iteration 2688, lr = 0.00587162 I0409 23:08:55.412714 4221 solver.cpp:218] Iteration 2700 (2.41478 iter/s, 4.96941s/12 iters), loss = 2.09146 I0409 23:08:55.412757 4221 solver.cpp:237] Train net output #0: loss = 2.09146 (* 1 = 2.09146 loss) I0409 23:08:55.412766 4221 sgd_solver.cpp:105] Iteration 2700, lr = 0.00585768 I0409 23:09:00.298285 4221 solver.cpp:218] Iteration 2712 (2.45634 iter/s, 4.88531s/12 iters), loss = 1.70095 I0409 23:09:00.298337 4221 solver.cpp:237] Train net output #0: loss = 1.70095 (* 1 = 1.70095 loss) I0409 23:09:00.298349 4221 sgd_solver.cpp:105] Iteration 2712, lr = 0.00584377 I0409 23:09:05.214813 4221 solver.cpp:218] Iteration 2724 (2.44088 iter/s, 4.91625s/12 iters), loss = 1.72599 I0409 23:09:05.214965 4221 solver.cpp:237] Train net output #0: loss = 1.72599 (* 1 = 1.72599 loss) I0409 23:09:05.214979 4221 sgd_solver.cpp:105] Iteration 2724, lr = 0.0058299 I0409 23:09:07.737661 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:09:10.105505 4221 solver.cpp:218] Iteration 2736 (2.45382 iter/s, 4.89033s/12 iters), loss = 1.71215 I0409 23:09:10.105551 4221 solver.cpp:237] Train net output #0: loss = 1.71215 (* 1 = 1.71215 loss) I0409 23:09:10.105561 4221 sgd_solver.cpp:105] Iteration 2736, lr = 0.00581605 I0409 23:09:15.019745 4221 solver.cpp:218] Iteration 2748 (2.44201 iter/s, 4.91398s/12 iters), loss = 2.38395 I0409 23:09:15.019795 4221 solver.cpp:237] Train net output #0: loss = 2.38395 (* 1 = 2.38395 loss) I0409 23:09:15.019805 4221 sgd_solver.cpp:105] Iteration 2748, lr = 0.00580225 I0409 23:09:17.003062 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_2754.caffemodel I0409 23:09:17.669128 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_2754.solverstate I0409 23:09:18.157436 4221 solver.cpp:330] Iteration 2754, Testing net (#0) I0409 23:09:18.157465 4221 net.cpp:676] Ignoring source layer train-data I0409 23:09:20.858141 4221 blocking_queue.cpp:49] Waiting for data I0409 23:09:21.364727 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:09:22.465420 4221 solver.cpp:397] Test net output #0: accuracy = 0.276961 I0409 23:09:22.465466 4221 solver.cpp:397] Test net output #1: loss = 3.29186 (* 1 = 3.29186 loss) I0409 23:09:24.365828 4221 solver.cpp:218] Iteration 2760 (1.28402 iter/s, 9.34564s/12 iters), loss = 2.09203 I0409 23:09:24.365871 4221 solver.cpp:237] Train net output #0: loss = 2.09203 (* 1 = 2.09203 loss) I0409 23:09:24.365880 4221 sgd_solver.cpp:105] Iteration 2760, lr = 0.00578847 I0409 23:09:29.244546 4221 solver.cpp:218] Iteration 2772 (2.45979 iter/s, 4.87846s/12 iters), loss = 2.2907 I0409 23:09:29.244596 4221 solver.cpp:237] Train net output #0: loss = 2.2907 (* 1 = 2.2907 loss) I0409 23:09:29.244609 4221 sgd_solver.cpp:105] Iteration 2772, lr = 0.00577473 I0409 23:09:34.133077 4221 solver.cpp:218] Iteration 2784 (2.45486 iter/s, 4.88827s/12 iters), loss = 2.13466 I0409 23:09:34.133124 4221 solver.cpp:237] Train net output #0: loss = 2.13466 (* 1 = 2.13466 loss) I0409 23:09:34.133134 4221 sgd_solver.cpp:105] Iteration 2784, lr = 0.00576102 I0409 23:09:38.959347 4221 solver.cpp:218] Iteration 2796 (2.48653 iter/s, 4.82601s/12 iters), loss = 1.9318 I0409 23:09:38.959421 4221 solver.cpp:237] Train net output #0: loss = 1.9318 (* 1 = 1.9318 loss) I0409 23:09:38.959430 4221 sgd_solver.cpp:105] Iteration 2796, lr = 0.00574734 I0409 23:09:43.878307 4221 solver.cpp:218] Iteration 2808 (2.43968 iter/s, 4.91867s/12 iters), loss = 1.71923 I0409 23:09:43.878358 4221 solver.cpp:237] Train net output #0: loss = 1.71923 (* 1 = 1.71923 loss) I0409 23:09:43.878369 4221 sgd_solver.cpp:105] Iteration 2808, lr = 0.00573369 I0409 23:09:48.773047 4221 solver.cpp:218] Iteration 2820 (2.45174 iter/s, 4.89448s/12 iters), loss = 1.67349 I0409 23:09:48.773098 4221 solver.cpp:237] Train net output #0: loss = 1.67349 (* 1 = 1.67349 loss) I0409 23:09:48.773109 4221 sgd_solver.cpp:105] Iteration 2820, lr = 0.00572008 I0409 23:09:53.294226 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:09:53.574440 4221 solver.cpp:218] Iteration 2832 (2.49941 iter/s, 4.80113s/12 iters), loss = 1.4849 I0409 23:09:53.574499 4221 solver.cpp:237] Train net output #0: loss = 1.4849 (* 1 = 1.4849 loss) I0409 23:09:53.574512 4221 sgd_solver.cpp:105] Iteration 2832, lr = 0.0057065 I0409 23:09:58.378664 4221 solver.cpp:218] Iteration 2844 (2.49795 iter/s, 4.80395s/12 iters), loss = 1.93694 I0409 23:09:58.378724 4221 solver.cpp:237] Train net output #0: loss = 1.93694 (* 1 = 1.93694 loss) I0409 23:09:58.378737 4221 sgd_solver.cpp:105] Iteration 2844, lr = 0.00569295 I0409 23:10:02.724948 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_2856.caffemodel I0409 23:10:03.393484 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_2856.solverstate I0409 23:10:03.884052 4221 solver.cpp:330] Iteration 2856, Testing net (#0) I0409 23:10:03.884080 4221 net.cpp:676] Ignoring source layer train-data I0409 23:10:07.360874 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:10:08.494976 4221 solver.cpp:397] Test net output #0: accuracy = 0.275123 I0409 23:10:08.495031 4221 solver.cpp:397] Test net output #1: loss = 3.31038 (* 1 = 3.31038 loss) I0409 23:10:08.578217 4221 solver.cpp:218] Iteration 2856 (1.17658 iter/s, 10.1991s/12 iters), loss = 1.84354 I0409 23:10:08.578272 4221 solver.cpp:237] Train net output #0: loss = 1.84354 (* 1 = 1.84354 loss) I0409 23:10:08.578284 4221 sgd_solver.cpp:105] Iteration 2856, lr = 0.00567944 I0409 23:10:12.681093 4221 solver.cpp:218] Iteration 2868 (2.92494 iter/s, 4.10264s/12 iters), loss = 2.10282 I0409 23:10:12.681216 4221 solver.cpp:237] Train net output #0: loss = 2.10282 (* 1 = 2.10282 loss) I0409 23:10:12.681226 4221 sgd_solver.cpp:105] Iteration 2868, lr = 0.00566595 I0409 23:10:17.779551 4221 solver.cpp:218] Iteration 2880 (2.35381 iter/s, 5.09812s/12 iters), loss = 1.4749 I0409 23:10:17.779580 4221 solver.cpp:237] Train net output #0: loss = 1.4749 (* 1 = 1.4749 loss) I0409 23:10:17.779590 4221 sgd_solver.cpp:105] Iteration 2880, lr = 0.0056525 I0409 23:10:22.551591 4221 solver.cpp:218] Iteration 2892 (2.51477 iter/s, 4.7718s/12 iters), loss = 2.14451 I0409 23:10:22.551637 4221 solver.cpp:237] Train net output #0: loss = 2.14451 (* 1 = 2.14451 loss) I0409 23:10:22.551649 4221 sgd_solver.cpp:105] Iteration 2892, lr = 0.00563908 I0409 23:10:27.500689 4221 solver.cpp:218] Iteration 2904 (2.42482 iter/s, 4.94883s/12 iters), loss = 1.41881 I0409 23:10:27.500744 4221 solver.cpp:237] Train net output #0: loss = 1.41881 (* 1 = 1.41881 loss) I0409 23:10:27.500756 4221 sgd_solver.cpp:105] Iteration 2904, lr = 0.00562569 I0409 23:10:32.616322 4221 solver.cpp:218] Iteration 2916 (2.34588 iter/s, 5.11536s/12 iters), loss = 1.58917 I0409 23:10:32.616374 4221 solver.cpp:237] Train net output #0: loss = 1.58917 (* 1 = 1.58917 loss) I0409 23:10:32.616386 4221 sgd_solver.cpp:105] Iteration 2916, lr = 0.00561233 I0409 23:10:37.611433 4221 solver.cpp:218] Iteration 2928 (2.40248 iter/s, 4.99484s/12 iters), loss = 1.62183 I0409 23:10:37.611495 4221 solver.cpp:237] Train net output #0: loss = 1.62183 (* 1 = 1.62183 loss) I0409 23:10:37.611510 4221 sgd_solver.cpp:105] Iteration 2928, lr = 0.00559901 I0409 23:10:39.404417 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:10:42.485608 4221 solver.cpp:218] Iteration 2940 (2.46209 iter/s, 4.8739s/12 iters), loss = 1.89774 I0409 23:10:42.485657 4221 solver.cpp:237] Train net output #0: loss = 1.89774 (* 1 = 1.89774 loss) I0409 23:10:42.485667 4221 sgd_solver.cpp:105] Iteration 2940, lr = 0.00558572 I0409 23:10:47.411288 4221 solver.cpp:218] Iteration 2952 (2.43635 iter/s, 4.92541s/12 iters), loss = 1.79051 I0409 23:10:47.411414 4221 solver.cpp:237] Train net output #0: loss = 1.79051 (* 1 = 1.79051 loss) I0409 23:10:47.411428 4221 sgd_solver.cpp:105] Iteration 2952, lr = 0.00557245 I0409 23:10:49.407912 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_2958.caffemodel I0409 23:10:50.122573 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_2958.solverstate I0409 23:10:50.616812 4221 solver.cpp:330] Iteration 2958, Testing net (#0) I0409 23:10:50.616833 4221 net.cpp:676] Ignoring source layer train-data I0409 23:10:53.876490 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:10:55.057287 4221 solver.cpp:397] Test net output #0: accuracy = 0.289828 I0409 23:10:55.057324 4221 solver.cpp:397] Test net output #1: loss = 3.45094 (* 1 = 3.45094 loss) I0409 23:10:56.814637 4221 solver.cpp:218] Iteration 2964 (1.27621 iter/s, 9.40283s/12 iters), loss = 1.46535 I0409 23:10:56.814693 4221 solver.cpp:237] Train net output #0: loss = 1.46535 (* 1 = 1.46535 loss) I0409 23:10:56.814705 4221 sgd_solver.cpp:105] Iteration 2964, lr = 0.00555922 I0409 23:11:01.705849 4221 solver.cpp:218] Iteration 2976 (2.45352 iter/s, 4.89094s/12 iters), loss = 1.78577 I0409 23:11:01.705907 4221 solver.cpp:237] Train net output #0: loss = 1.78577 (* 1 = 1.78577 loss) I0409 23:11:01.705920 4221 sgd_solver.cpp:105] Iteration 2976, lr = 0.00554603 I0409 23:11:06.579291 4221 solver.cpp:218] Iteration 2988 (2.46246 iter/s, 4.87317s/12 iters), loss = 1.63153 I0409 23:11:06.579344 4221 solver.cpp:237] Train net output #0: loss = 1.63153 (* 1 = 1.63153 loss) I0409 23:11:06.579356 4221 sgd_solver.cpp:105] Iteration 2988, lr = 0.00553286 I0409 23:11:11.406796 4221 solver.cpp:218] Iteration 3000 (2.48589 iter/s, 4.82724s/12 iters), loss = 1.67912 I0409 23:11:11.406847 4221 solver.cpp:237] Train net output #0: loss = 1.67912 (* 1 = 1.67912 loss) I0409 23:11:11.406858 4221 sgd_solver.cpp:105] Iteration 3000, lr = 0.00551972 I0409 23:11:16.267976 4221 solver.cpp:218] Iteration 3012 (2.46867 iter/s, 4.86092s/12 iters), loss = 1.45149 I0409 23:11:16.268028 4221 solver.cpp:237] Train net output #0: loss = 1.45149 (* 1 = 1.45149 loss) I0409 23:11:16.268039 4221 sgd_solver.cpp:105] Iteration 3012, lr = 0.00550662 I0409 23:11:21.179301 4221 solver.cpp:218] Iteration 3024 (2.44346 iter/s, 4.91106s/12 iters), loss = 1.63855 I0409 23:11:21.179397 4221 solver.cpp:237] Train net output #0: loss = 1.63855 (* 1 = 1.63855 loss) I0409 23:11:21.179410 4221 sgd_solver.cpp:105] Iteration 3024, lr = 0.00549354 I0409 23:11:25.152159 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:11:26.149963 4221 solver.cpp:218] Iteration 3036 (2.41432 iter/s, 4.97034s/12 iters), loss = 1.25533 I0409 23:11:26.150018 4221 solver.cpp:237] Train net output #0: loss = 1.25533 (* 1 = 1.25533 loss) I0409 23:11:26.150030 4221 sgd_solver.cpp:105] Iteration 3036, lr = 0.0054805 I0409 23:11:31.070171 4221 solver.cpp:218] Iteration 3048 (2.43905 iter/s, 4.91994s/12 iters), loss = 1.92059 I0409 23:11:31.070225 4221 solver.cpp:237] Train net output #0: loss = 1.92059 (* 1 = 1.92059 loss) I0409 23:11:31.070237 4221 sgd_solver.cpp:105] Iteration 3048, lr = 0.00546749 I0409 23:11:35.482775 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_3060.caffemodel I0409 23:11:36.754539 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_3060.solverstate I0409 23:11:37.824357 4221 solver.cpp:330] Iteration 3060, Testing net (#0) I0409 23:11:37.824386 4221 net.cpp:676] Ignoring source layer train-data I0409 23:11:41.035357 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:11:42.256968 4221 solver.cpp:397] Test net output #0: accuracy = 0.279412 I0409 23:11:42.257017 4221 solver.cpp:397] Test net output #1: loss = 3.29006 (* 1 = 3.29006 loss) I0409 23:11:42.340235 4221 solver.cpp:218] Iteration 3060 (1.06482 iter/s, 11.2695s/12 iters), loss = 1.60915 I0409 23:11:42.340288 4221 solver.cpp:237] Train net output #0: loss = 1.60915 (* 1 = 1.60915 loss) I0409 23:11:42.340299 4221 sgd_solver.cpp:105] Iteration 3060, lr = 0.00545451 I0409 23:11:46.475131 4221 solver.cpp:218] Iteration 3072 (2.90229 iter/s, 4.13466s/12 iters), loss = 1.62427 I0409 23:11:46.475178 4221 solver.cpp:237] Train net output #0: loss = 1.62427 (* 1 = 1.62427 loss) I0409 23:11:46.475186 4221 sgd_solver.cpp:105] Iteration 3072, lr = 0.00544156 I0409 23:11:51.395234 4221 solver.cpp:218] Iteration 3084 (2.4391 iter/s, 4.91984s/12 iters), loss = 1.38564 I0409 23:11:51.395355 4221 solver.cpp:237] Train net output #0: loss = 1.38564 (* 1 = 1.38564 loss) I0409 23:11:51.395370 4221 sgd_solver.cpp:105] Iteration 3084, lr = 0.00542864 I0409 23:11:56.283068 4221 solver.cpp:218] Iteration 3096 (2.45524 iter/s, 4.8875s/12 iters), loss = 1.61764 I0409 23:11:56.283128 4221 solver.cpp:237] Train net output #0: loss = 1.61764 (* 1 = 1.61764 loss) I0409 23:11:56.283140 4221 sgd_solver.cpp:105] Iteration 3096, lr = 0.00541575 I0409 23:12:01.165993 4221 solver.cpp:218] Iteration 3108 (2.45768 iter/s, 4.88265s/12 iters), loss = 1.37679 I0409 23:12:01.166044 4221 solver.cpp:237] Train net output #0: loss = 1.37679 (* 1 = 1.37679 loss) I0409 23:12:01.166055 4221 sgd_solver.cpp:105] Iteration 3108, lr = 0.00540289 I0409 23:12:06.035097 4221 solver.cpp:218] Iteration 3120 (2.46465 iter/s, 4.86884s/12 iters), loss = 1.29491 I0409 23:12:06.035145 4221 solver.cpp:237] Train net output #0: loss = 1.29491 (* 1 = 1.29491 loss) I0409 23:12:06.035152 4221 sgd_solver.cpp:105] Iteration 3120, lr = 0.00539006 I0409 23:12:10.842269 4221 solver.cpp:218] Iteration 3132 (2.4964 iter/s, 4.80692s/12 iters), loss = 1.65298 I0409 23:12:10.842311 4221 solver.cpp:237] Train net output #0: loss = 1.65298 (* 1 = 1.65298 loss) I0409 23:12:10.842327 4221 sgd_solver.cpp:105] Iteration 3132, lr = 0.00537727 I0409 23:12:11.911909 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:12:15.677443 4221 solver.cpp:218] Iteration 3144 (2.48194 iter/s, 4.83492s/12 iters), loss = 1.2053 I0409 23:12:15.677487 4221 solver.cpp:237] Train net output #0: loss = 1.2053 (* 1 = 1.2053 loss) I0409 23:12:15.677496 4221 sgd_solver.cpp:105] Iteration 3144, lr = 0.0053645 I0409 23:12:20.468220 4221 solver.cpp:218] Iteration 3156 (2.50495 iter/s, 4.79052s/12 iters), loss = 1.43333 I0409 23:12:20.468276 4221 solver.cpp:237] Train net output #0: loss = 1.43333 (* 1 = 1.43333 loss) I0409 23:12:20.468287 4221 sgd_solver.cpp:105] Iteration 3156, lr = 0.00535176 I0409 23:12:22.454236 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_3162.caffemodel I0409 23:12:23.801739 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_3162.solverstate I0409 23:12:24.595197 4221 solver.cpp:330] Iteration 3162, Testing net (#0) I0409 23:12:24.595227 4221 net.cpp:676] Ignoring source layer train-data I0409 23:12:27.949049 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:12:29.282124 4221 solver.cpp:397] Test net output #0: accuracy = 0.309436 I0409 23:12:29.282168 4221 solver.cpp:397] Test net output #1: loss = 3.32163 (* 1 = 3.32163 loss) I0409 23:12:31.209327 4221 solver.cpp:218] Iteration 3168 (1.11726 iter/s, 10.7406s/12 iters), loss = 1.35752 I0409 23:12:31.209370 4221 solver.cpp:237] Train net output #0: loss = 1.35752 (* 1 = 1.35752 loss) I0409 23:12:31.209378 4221 sgd_solver.cpp:105] Iteration 3168, lr = 0.00533906 I0409 23:12:36.015151 4221 solver.cpp:218] Iteration 3180 (2.4971 iter/s, 4.80556s/12 iters), loss = 1.54348 I0409 23:12:36.015203 4221 solver.cpp:237] Train net output #0: loss = 1.54348 (* 1 = 1.54348 loss) I0409 23:12:36.015215 4221 sgd_solver.cpp:105] Iteration 3180, lr = 0.00532638 I0409 23:12:40.872479 4221 solver.cpp:218] Iteration 3192 (2.47063 iter/s, 4.85706s/12 iters), loss = 1.40338 I0409 23:12:40.872543 4221 solver.cpp:237] Train net output #0: loss = 1.40338 (* 1 = 1.40338 loss) I0409 23:12:40.872558 4221 sgd_solver.cpp:105] Iteration 3192, lr = 0.00531374 I0409 23:12:45.679613 4221 solver.cpp:218] Iteration 3204 (2.49643 iter/s, 4.80686s/12 iters), loss = 1.4964 I0409 23:12:45.679663 4221 solver.cpp:237] Train net output #0: loss = 1.4964 (* 1 = 1.4964 loss) I0409 23:12:45.679675 4221 sgd_solver.cpp:105] Iteration 3204, lr = 0.00530112 I0409 23:12:50.609246 4221 solver.cpp:218] Iteration 3216 (2.43439 iter/s, 4.92936s/12 iters), loss = 1.52899 I0409 23:12:50.609304 4221 solver.cpp:237] Train net output #0: loss = 1.52899 (* 1 = 1.52899 loss) I0409 23:12:50.609318 4221 sgd_solver.cpp:105] Iteration 3216, lr = 0.00528853 I0409 23:12:55.483669 4221 solver.cpp:218] Iteration 3228 (2.46197 iter/s, 4.87415s/12 iters), loss = 1.42196 I0409 23:12:55.483793 4221 solver.cpp:237] Train net output #0: loss = 1.42196 (* 1 = 1.42196 loss) I0409 23:12:55.483803 4221 sgd_solver.cpp:105] Iteration 3228, lr = 0.00527598 I0409 23:12:58.646945 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:13:00.374066 4221 solver.cpp:218] Iteration 3240 (2.45395 iter/s, 4.89007s/12 iters), loss = 1.48414 I0409 23:13:00.374104 4221 solver.cpp:237] Train net output #0: loss = 1.48414 (* 1 = 1.48414 loss) I0409 23:13:00.374112 4221 sgd_solver.cpp:105] Iteration 3240, lr = 0.00526345 I0409 23:13:05.341627 4221 solver.cpp:218] Iteration 3252 (2.4158 iter/s, 4.9673s/12 iters), loss = 1.21814 I0409 23:13:05.341684 4221 solver.cpp:237] Train net output #0: loss = 1.21814 (* 1 = 1.21814 loss) I0409 23:13:05.341696 4221 sgd_solver.cpp:105] Iteration 3252, lr = 0.00525095 I0409 23:13:09.778790 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_3264.caffemodel I0409 23:13:10.492182 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_3264.solverstate I0409 23:13:10.991935 4221 solver.cpp:330] Iteration 3264, Testing net (#0) I0409 23:13:10.991962 4221 net.cpp:676] Ignoring source layer train-data I0409 23:13:14.222894 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:13:15.522719 4221 solver.cpp:397] Test net output #0: accuracy = 0.307598 I0409 23:13:15.522759 4221 solver.cpp:397] Test net output #1: loss = 3.34859 (* 1 = 3.34859 loss) I0409 23:13:15.605863 4221 solver.cpp:218] Iteration 3264 (1.16916 iter/s, 10.2638s/12 iters), loss = 1.33637 I0409 23:13:15.605912 4221 solver.cpp:237] Train net output #0: loss = 1.33637 (* 1 = 1.33637 loss) I0409 23:13:15.605923 4221 sgd_solver.cpp:105] Iteration 3264, lr = 0.00523849 I0409 23:13:19.662003 4221 solver.cpp:218] Iteration 3276 (2.95865 iter/s, 4.0559s/12 iters), loss = 1.45009 I0409 23:13:19.662060 4221 solver.cpp:237] Train net output #0: loss = 1.45009 (* 1 = 1.45009 loss) I0409 23:13:19.662071 4221 sgd_solver.cpp:105] Iteration 3276, lr = 0.00522605 I0409 23:13:24.525048 4221 solver.cpp:218] Iteration 3288 (2.46772 iter/s, 4.86278s/12 iters), loss = 1.30379 I0409 23:13:24.525094 4221 solver.cpp:237] Train net output #0: loss = 1.30379 (* 1 = 1.30379 loss) I0409 23:13:24.525104 4221 sgd_solver.cpp:105] Iteration 3288, lr = 0.00521364 I0409 23:13:29.397294 4221 solver.cpp:218] Iteration 3300 (2.46306 iter/s, 4.87199s/12 iters), loss = 1.34009 I0409 23:13:29.397367 4221 solver.cpp:237] Train net output #0: loss = 1.34009 (* 1 = 1.34009 loss) I0409 23:13:29.397377 4221 sgd_solver.cpp:105] Iteration 3300, lr = 0.00520126 I0409 23:13:34.316865 4221 solver.cpp:218] Iteration 3312 (2.43938 iter/s, 4.91928s/12 iters), loss = 1.52958 I0409 23:13:34.316921 4221 solver.cpp:237] Train net output #0: loss = 1.52958 (* 1 = 1.52958 loss) I0409 23:13:34.316934 4221 sgd_solver.cpp:105] Iteration 3312, lr = 0.00518892 I0409 23:13:39.222867 4221 solver.cpp:218] Iteration 3324 (2.44612 iter/s, 4.90573s/12 iters), loss = 1.30851 I0409 23:13:39.222931 4221 solver.cpp:237] Train net output #0: loss = 1.30851 (* 1 = 1.30851 loss) I0409 23:13:39.222944 4221 sgd_solver.cpp:105] Iteration 3324, lr = 0.0051766 I0409 23:13:43.995028 4221 solver.cpp:218] Iteration 3336 (2.51473 iter/s, 4.77189s/12 iters), loss = 1.32827 I0409 23:13:43.995091 4221 solver.cpp:237] Train net output #0: loss = 1.32827 (* 1 = 1.32827 loss) I0409 23:13:43.995103 4221 sgd_solver.cpp:105] Iteration 3336, lr = 0.00516431 I0409 23:13:44.442965 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:13:48.777993 4221 solver.cpp:218] Iteration 3348 (2.50907 iter/s, 4.78265s/12 iters), loss = 1.25635 I0409 23:13:48.778038 4221 solver.cpp:237] Train net output #0: loss = 1.25635 (* 1 = 1.25635 loss) I0409 23:13:48.778046 4221 sgd_solver.cpp:105] Iteration 3348, lr = 0.00515204 I0409 23:13:53.801877 4221 solver.cpp:218] Iteration 3360 (2.38872 iter/s, 5.02362s/12 iters), loss = 1.32086 I0409 23:13:53.801929 4221 solver.cpp:237] Train net output #0: loss = 1.32086 (* 1 = 1.32086 loss) I0409 23:13:53.801940 4221 sgd_solver.cpp:105] Iteration 3360, lr = 0.00513981 I0409 23:13:55.759985 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_3366.caffemodel I0409 23:13:56.449887 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_3366.solverstate I0409 23:13:56.941103 4221 solver.cpp:330] Iteration 3366, Testing net (#0) I0409 23:13:56.941133 4221 net.cpp:676] Ignoring source layer train-data I0409 23:14:00.051872 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:14:01.386399 4221 solver.cpp:397] Test net output #0: accuracy = 0.296569 I0409 23:14:01.386440 4221 solver.cpp:397] Test net output #1: loss = 3.36251 (* 1 = 3.36251 loss) I0409 23:14:03.223316 4221 solver.cpp:218] Iteration 3372 (1.27375 iter/s, 9.42099s/12 iters), loss = 1.26528 I0409 23:14:03.223371 4221 solver.cpp:237] Train net output #0: loss = 1.26528 (* 1 = 1.26528 loss) I0409 23:14:03.223382 4221 sgd_solver.cpp:105] Iteration 3372, lr = 0.00512761 I0409 23:14:08.106837 4221 solver.cpp:218] Iteration 3384 (2.45738 iter/s, 4.88326s/12 iters), loss = 1.25061 I0409 23:14:08.106886 4221 solver.cpp:237] Train net output #0: loss = 1.25061 (* 1 = 1.25061 loss) I0409 23:14:08.106899 4221 sgd_solver.cpp:105] Iteration 3384, lr = 0.00511544 I0409 23:14:13.038358 4221 solver.cpp:218] Iteration 3396 (2.43346 iter/s, 4.93126s/12 iters), loss = 1.22118 I0409 23:14:13.038414 4221 solver.cpp:237] Train net output #0: loss = 1.22118 (* 1 = 1.22118 loss) I0409 23:14:13.038424 4221 sgd_solver.cpp:105] Iteration 3396, lr = 0.00510329 I0409 23:14:17.925722 4221 solver.cpp:218] Iteration 3408 (2.45545 iter/s, 4.88709s/12 iters), loss = 1.33619 I0409 23:14:17.925779 4221 solver.cpp:237] Train net output #0: loss = 1.33619 (* 1 = 1.33619 loss) I0409 23:14:17.925791 4221 sgd_solver.cpp:105] Iteration 3408, lr = 0.00509117 I0409 23:14:22.835863 4221 solver.cpp:218] Iteration 3420 (2.44406 iter/s, 4.90987s/12 iters), loss = 1.40319 I0409 23:14:22.835916 4221 solver.cpp:237] Train net output #0: loss = 1.40319 (* 1 = 1.40319 loss) I0409 23:14:22.835928 4221 sgd_solver.cpp:105] Iteration 3420, lr = 0.00507909 I0409 23:14:27.694635 4221 solver.cpp:218] Iteration 3432 (2.46989 iter/s, 4.85851s/12 iters), loss = 1.40962 I0409 23:14:27.694690 4221 solver.cpp:237] Train net output #0: loss = 1.40962 (* 1 = 1.40962 loss) I0409 23:14:27.694701 4221 sgd_solver.cpp:105] Iteration 3432, lr = 0.00506703 I0409 23:14:30.242180 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:14:32.564316 4221 solver.cpp:218] Iteration 3444 (2.46436 iter/s, 4.86941s/12 iters), loss = 1.12618 I0409 23:14:32.564361 4221 solver.cpp:237] Train net output #0: loss = 1.12618 (* 1 = 1.12618 loss) I0409 23:14:32.564373 4221 sgd_solver.cpp:105] Iteration 3444, lr = 0.005055 I0409 23:14:37.439669 4221 solver.cpp:218] Iteration 3456 (2.46149 iter/s, 4.8751s/12 iters), loss = 1.30675 I0409 23:14:37.439713 4221 solver.cpp:237] Train net output #0: loss = 1.30675 (* 1 = 1.30675 loss) I0409 23:14:37.439724 4221 sgd_solver.cpp:105] Iteration 3456, lr = 0.005043 I0409 23:14:41.906412 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_3468.caffemodel I0409 23:14:42.608737 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_3468.solverstate I0409 23:14:43.093484 4221 solver.cpp:330] Iteration 3468, Testing net (#0) I0409 23:14:43.093506 4221 net.cpp:676] Ignoring source layer train-data I0409 23:14:43.229176 4221 blocking_queue.cpp:49] Waiting for data I0409 23:14:46.339756 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:14:47.719772 4221 solver.cpp:397] Test net output #0: accuracy = 0.315564 I0409 23:14:47.719821 4221 solver.cpp:397] Test net output #1: loss = 3.45428 (* 1 = 3.45428 loss) I0409 23:14:47.803205 4221 solver.cpp:218] Iteration 3468 (1.15796 iter/s, 10.3631s/12 iters), loss = 0.8945 I0409 23:14:47.803256 4221 solver.cpp:237] Train net output #0: loss = 0.8945 (* 1 = 0.8945 loss) I0409 23:14:47.803267 4221 sgd_solver.cpp:105] Iteration 3468, lr = 0.00503102 I0409 23:14:52.000216 4221 solver.cpp:218] Iteration 3480 (2.85934 iter/s, 4.19677s/12 iters), loss = 1.52594 I0409 23:14:52.000265 4221 solver.cpp:237] Train net output #0: loss = 1.52594 (* 1 = 1.52594 loss) I0409 23:14:52.000277 4221 sgd_solver.cpp:105] Iteration 3480, lr = 0.00501908 I0409 23:14:56.794953 4221 solver.cpp:218] Iteration 3492 (2.50288 iter/s, 4.79448s/12 iters), loss = 1.60377 I0409 23:14:56.795003 4221 solver.cpp:237] Train net output #0: loss = 1.60377 (* 1 = 1.60377 loss) I0409 23:14:56.795015 4221 sgd_solver.cpp:105] Iteration 3492, lr = 0.00500716 I0409 23:15:01.765120 4221 solver.cpp:218] Iteration 3504 (2.41454 iter/s, 4.96989s/12 iters), loss = 1.51569 I0409 23:15:01.765255 4221 solver.cpp:237] Train net output #0: loss = 1.51569 (* 1 = 1.51569 loss) I0409 23:15:01.765270 4221 sgd_solver.cpp:105] Iteration 3504, lr = 0.00499527 I0409 23:15:06.576894 4221 solver.cpp:218] Iteration 3516 (2.49406 iter/s, 4.81143s/12 iters), loss = 1.14858 I0409 23:15:06.576946 4221 solver.cpp:237] Train net output #0: loss = 1.14858 (* 1 = 1.14858 loss) I0409 23:15:06.576961 4221 sgd_solver.cpp:105] Iteration 3516, lr = 0.00498341 I0409 23:15:11.534580 4221 solver.cpp:218] Iteration 3528 (2.42061 iter/s, 4.95742s/12 iters), loss = 1.26479 I0409 23:15:11.534637 4221 solver.cpp:237] Train net output #0: loss = 1.26479 (* 1 = 1.26479 loss) I0409 23:15:11.534651 4221 sgd_solver.cpp:105] Iteration 3528, lr = 0.00497158 I0409 23:15:16.177668 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:15:16.442155 4221 solver.cpp:218] Iteration 3540 (2.44534 iter/s, 4.9073s/12 iters), loss = 1.11939 I0409 23:15:16.442207 4221 solver.cpp:237] Train net output #0: loss = 1.11939 (* 1 = 1.11939 loss) I0409 23:15:16.442219 4221 sgd_solver.cpp:105] Iteration 3540, lr = 0.00495978 I0409 23:15:21.393115 4221 solver.cpp:218] Iteration 3552 (2.4239 iter/s, 4.95069s/12 iters), loss = 1.29188 I0409 23:15:21.393162 4221 solver.cpp:237] Train net output #0: loss = 1.29188 (* 1 = 1.29188 loss) I0409 23:15:21.393170 4221 sgd_solver.cpp:105] Iteration 3552, lr = 0.004948 I0409 23:15:26.181843 4221 solver.cpp:218] Iteration 3564 (2.50602 iter/s, 4.78847s/12 iters), loss = 1.15304 I0409 23:15:26.181900 4221 solver.cpp:237] Train net output #0: loss = 1.15304 (* 1 = 1.15304 loss) I0409 23:15:26.181913 4221 sgd_solver.cpp:105] Iteration 3564, lr = 0.00493626 I0409 23:15:28.148808 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_3570.caffemodel I0409 23:15:28.850962 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_3570.solverstate I0409 23:15:29.362701 4221 solver.cpp:330] Iteration 3570, Testing net (#0) I0409 23:15:29.362731 4221 net.cpp:676] Ignoring source layer train-data I0409 23:15:32.301457 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:15:33.713325 4221 solver.cpp:397] Test net output #0: accuracy = 0.313726 I0409 23:15:33.713376 4221 solver.cpp:397] Test net output #1: loss = 3.35799 (* 1 = 3.35799 loss) I0409 23:15:35.452772 4221 solver.cpp:218] Iteration 3576 (1.29443 iter/s, 9.27048s/12 iters), loss = 1.4353 I0409 23:15:35.452811 4221 solver.cpp:237] Train net output #0: loss = 1.4353 (* 1 = 1.4353 loss) I0409 23:15:35.452818 4221 sgd_solver.cpp:105] Iteration 3576, lr = 0.00492454 I0409 23:15:40.377439 4221 solver.cpp:218] Iteration 3588 (2.43684 iter/s, 4.92441s/12 iters), loss = 1.03514 I0409 23:15:40.377490 4221 solver.cpp:237] Train net output #0: loss = 1.03514 (* 1 = 1.03514 loss) I0409 23:15:40.377501 4221 sgd_solver.cpp:105] Iteration 3588, lr = 0.00491284 I0409 23:15:45.199770 4221 solver.cpp:218] Iteration 3600 (2.48856 iter/s, 4.82207s/12 iters), loss = 1.1912 I0409 23:15:45.199815 4221 solver.cpp:237] Train net output #0: loss = 1.1912 (* 1 = 1.1912 loss) I0409 23:15:45.199824 4221 sgd_solver.cpp:105] Iteration 3600, lr = 0.00490118 I0409 23:15:50.025194 4221 solver.cpp:218] Iteration 3612 (2.48696 iter/s, 4.82517s/12 iters), loss = 1.21471 I0409 23:15:50.025243 4221 solver.cpp:237] Train net output #0: loss = 1.21471 (* 1 = 1.21471 loss) I0409 23:15:50.025252 4221 sgd_solver.cpp:105] Iteration 3612, lr = 0.00488954 I0409 23:15:54.822758 4221 solver.cpp:218] Iteration 3624 (2.50141 iter/s, 4.7973s/12 iters), loss = 1.40644 I0409 23:15:54.822803 4221 solver.cpp:237] Train net output #0: loss = 1.40644 (* 1 = 1.40644 loss) I0409 23:15:54.822811 4221 sgd_solver.cpp:105] Iteration 3624, lr = 0.00487793 I0409 23:15:59.676302 4221 solver.cpp:218] Iteration 3636 (2.47256 iter/s, 4.85327s/12 iters), loss = 1.14139 I0409 23:15:59.676378 4221 solver.cpp:237] Train net output #0: loss = 1.14139 (* 1 = 1.14139 loss) I0409 23:15:59.676395 4221 sgd_solver.cpp:105] Iteration 3636, lr = 0.00486635 I0409 23:16:01.467571 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:16:04.453486 4221 solver.cpp:218] Iteration 3648 (2.51209 iter/s, 4.77691s/12 iters), loss = 1.01513 I0409 23:16:04.453589 4221 solver.cpp:237] Train net output #0: loss = 1.01513 (* 1 = 1.01513 loss) I0409 23:16:04.453605 4221 sgd_solver.cpp:105] Iteration 3648, lr = 0.0048548 I0409 23:16:09.404412 4221 solver.cpp:218] Iteration 3660 (2.42395 iter/s, 4.95061s/12 iters), loss = 1.15631 I0409 23:16:09.404466 4221 solver.cpp:237] Train net output #0: loss = 1.15631 (* 1 = 1.15631 loss) I0409 23:16:09.404479 4221 sgd_solver.cpp:105] Iteration 3660, lr = 0.00484327 I0409 23:16:13.769826 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_3672.caffemodel I0409 23:16:14.871091 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_3672.solverstate I0409 23:16:16.142602 4221 solver.cpp:330] Iteration 3672, Testing net (#0) I0409 23:16:16.142638 4221 net.cpp:676] Ignoring source layer train-data I0409 23:16:19.674230 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:16:21.150480 4221 solver.cpp:397] Test net output #0: accuracy = 0.306373 I0409 23:16:21.150527 4221 solver.cpp:397] Test net output #1: loss = 3.48413 (* 1 = 3.48413 loss) I0409 23:16:21.233773 4221 solver.cpp:218] Iteration 3672 (1.01447 iter/s, 11.8288s/12 iters), loss = 1.01722 I0409 23:16:21.233821 4221 solver.cpp:237] Train net output #0: loss = 1.01722 (* 1 = 1.01722 loss) I0409 23:16:21.233831 4221 sgd_solver.cpp:105] Iteration 3672, lr = 0.00483177 I0409 23:16:25.357784 4221 solver.cpp:218] Iteration 3684 (2.90995 iter/s, 4.12378s/12 iters), loss = 1.35962 I0409 23:16:25.357827 4221 solver.cpp:237] Train net output #0: loss = 1.35962 (* 1 = 1.35962 loss) I0409 23:16:25.357837 4221 sgd_solver.cpp:105] Iteration 3684, lr = 0.0048203 I0409 23:16:30.288116 4221 solver.cpp:218] Iteration 3696 (2.43404 iter/s, 4.93007s/12 iters), loss = 1.05659 I0409 23:16:30.288169 4221 solver.cpp:237] Train net output #0: loss = 1.05659 (* 1 = 1.05659 loss) I0409 23:16:30.288179 4221 sgd_solver.cpp:105] Iteration 3696, lr = 0.00480886 I0409 23:16:35.243929 4221 solver.cpp:218] Iteration 3708 (2.42153 iter/s, 4.95554s/12 iters), loss = 1.115 I0409 23:16:35.244285 4221 solver.cpp:237] Train net output #0: loss = 1.115 (* 1 = 1.115 loss) I0409 23:16:35.244298 4221 sgd_solver.cpp:105] Iteration 3708, lr = 0.00479744 I0409 23:16:40.270032 4221 solver.cpp:218] Iteration 3720 (2.38781 iter/s, 5.02553s/12 iters), loss = 1.25646 I0409 23:16:40.270076 4221 solver.cpp:237] Train net output #0: loss = 1.25646 (* 1 = 1.25646 loss) I0409 23:16:40.270085 4221 sgd_solver.cpp:105] Iteration 3720, lr = 0.00478605 I0409 23:16:45.605739 4221 solver.cpp:218] Iteration 3732 (2.24912 iter/s, 5.33543s/12 iters), loss = 1.28186 I0409 23:16:45.605794 4221 solver.cpp:237] Train net output #0: loss = 1.28186 (* 1 = 1.28186 loss) I0409 23:16:45.605805 4221 sgd_solver.cpp:105] Iteration 3732, lr = 0.00477469 I0409 23:16:49.728176 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:16:50.704139 4221 solver.cpp:218] Iteration 3744 (2.35381 iter/s, 5.09812s/12 iters), loss = 0.816851 I0409 23:16:50.704197 4221 solver.cpp:237] Train net output #0: loss = 0.816851 (* 1 = 0.816851 loss) I0409 23:16:50.704208 4221 sgd_solver.cpp:105] Iteration 3744, lr = 0.00476335 I0409 23:16:55.655580 4221 solver.cpp:218] Iteration 3756 (2.42367 iter/s, 4.95116s/12 iters), loss = 0.978124 I0409 23:16:55.655637 4221 solver.cpp:237] Train net output #0: loss = 0.978124 (* 1 = 0.978124 loss) I0409 23:16:55.655650 4221 sgd_solver.cpp:105] Iteration 3756, lr = 0.00475204 I0409 23:17:00.911104 4221 solver.cpp:218] Iteration 3768 (2.28343 iter/s, 5.25524s/12 iters), loss = 0.899812 I0409 23:17:00.911149 4221 solver.cpp:237] Train net output #0: loss = 0.899812 (* 1 = 0.899812 loss) I0409 23:17:00.911157 4221 sgd_solver.cpp:105] Iteration 3768, lr = 0.00474076 I0409 23:17:03.047052 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_3774.caffemodel I0409 23:17:03.902853 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_3774.solverstate I0409 23:17:04.453230 4221 solver.cpp:330] Iteration 3774, Testing net (#0) I0409 23:17:04.453260 4221 net.cpp:676] Ignoring source layer train-data I0409 23:17:07.463512 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:17:09.407240 4221 solver.cpp:397] Test net output #0: accuracy = 0.29902 I0409 23:17:09.407280 4221 solver.cpp:397] Test net output #1: loss = 3.56888 (* 1 = 3.56888 loss) I0409 23:17:11.457082 4221 solver.cpp:218] Iteration 3780 (1.13793 iter/s, 10.5455s/12 iters), loss = 1.00701 I0409 23:17:11.457125 4221 solver.cpp:237] Train net output #0: loss = 1.00701 (* 1 = 1.00701 loss) I0409 23:17:11.457134 4221 sgd_solver.cpp:105] Iteration 3780, lr = 0.00472951 I0409 23:17:16.688984 4221 solver.cpp:218] Iteration 3792 (2.29374 iter/s, 5.23163s/12 iters), loss = 1.00221 I0409 23:17:16.689038 4221 solver.cpp:237] Train net output #0: loss = 1.00221 (* 1 = 1.00221 loss) I0409 23:17:16.689049 4221 sgd_solver.cpp:105] Iteration 3792, lr = 0.00471828 I0409 23:17:21.605410 4221 solver.cpp:218] Iteration 3804 (2.44093 iter/s, 4.91616s/12 iters), loss = 1.13428 I0409 23:17:21.605469 4221 solver.cpp:237] Train net output #0: loss = 1.13428 (* 1 = 1.13428 loss) I0409 23:17:21.605481 4221 sgd_solver.cpp:105] Iteration 3804, lr = 0.00470707 I0409 23:17:26.516705 4221 solver.cpp:218] Iteration 3816 (2.44348 iter/s, 4.91103s/12 iters), loss = 0.783854 I0409 23:17:26.516746 4221 solver.cpp:237] Train net output #0: loss = 0.783854 (* 1 = 0.783854 loss) I0409 23:17:26.516754 4221 sgd_solver.cpp:105] Iteration 3816, lr = 0.0046959 I0409 23:17:31.534390 4221 solver.cpp:218] Iteration 3828 (2.39167 iter/s, 5.01742s/12 iters), loss = 1.11264 I0409 23:17:31.534446 4221 solver.cpp:237] Train net output #0: loss = 1.11264 (* 1 = 1.11264 loss) I0409 23:17:31.534456 4221 sgd_solver.cpp:105] Iteration 3828, lr = 0.00468475 I0409 23:17:36.424088 4221 solver.cpp:218] Iteration 3840 (2.45427 iter/s, 4.88943s/12 iters), loss = 1.0604 I0409 23:17:36.424135 4221 solver.cpp:237] Train net output #0: loss = 1.0604 (* 1 = 1.0604 loss) I0409 23:17:36.424145 4221 sgd_solver.cpp:105] Iteration 3840, lr = 0.00467363 I0409 23:17:37.557211 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:17:41.351204 4221 solver.cpp:218] Iteration 3852 (2.43563 iter/s, 4.92685s/12 iters), loss = 1.03206 I0409 23:17:41.351248 4221 solver.cpp:237] Train net output #0: loss = 1.03206 (* 1 = 1.03206 loss) I0409 23:17:41.351256 4221 sgd_solver.cpp:105] Iteration 3852, lr = 0.00466253 I0409 23:17:46.220636 4221 solver.cpp:218] Iteration 3864 (2.46448 iter/s, 4.86918s/12 iters), loss = 1.08033 I0409 23:17:46.220682 4221 solver.cpp:237] Train net output #0: loss = 1.08033 (* 1 = 1.08033 loss) I0409 23:17:46.220691 4221 sgd_solver.cpp:105] Iteration 3864, lr = 0.00465146 I0409 23:17:50.645150 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_3876.caffemodel I0409 23:17:51.335295 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_3876.solverstate I0409 23:17:51.828464 4221 solver.cpp:330] Iteration 3876, Testing net (#0) I0409 23:17:51.828490 4221 net.cpp:676] Ignoring source layer train-data I0409 23:17:54.789288 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:17:56.365468 4221 solver.cpp:397] Test net output #0: accuracy = 0.326593 I0409 23:17:56.365500 4221 solver.cpp:397] Test net output #1: loss = 3.38648 (* 1 = 3.38648 loss) I0409 23:17:56.448915 4221 solver.cpp:218] Iteration 3876 (1.17327 iter/s, 10.2278s/12 iters), loss = 0.796197 I0409 23:17:56.448963 4221 solver.cpp:237] Train net output #0: loss = 0.796197 (* 1 = 0.796197 loss) I0409 23:17:56.448972 4221 sgd_solver.cpp:105] Iteration 3876, lr = 0.00464042 I0409 23:18:00.627496 4221 solver.cpp:218] Iteration 3888 (2.87195 iter/s, 4.17834s/12 iters), loss = 0.857789 I0409 23:18:00.627557 4221 solver.cpp:237] Train net output #0: loss = 0.857789 (* 1 = 0.857789 loss) I0409 23:18:00.627568 4221 sgd_solver.cpp:105] Iteration 3888, lr = 0.0046294 I0409 23:18:05.541501 4221 solver.cpp:218] Iteration 3900 (2.44214 iter/s, 4.91373s/12 iters), loss = 0.928314 I0409 23:18:05.541560 4221 solver.cpp:237] Train net output #0: loss = 0.928314 (* 1 = 0.928314 loss) I0409 23:18:05.541571 4221 sgd_solver.cpp:105] Iteration 3900, lr = 0.00461841 I0409 23:18:10.323570 4221 solver.cpp:218] Iteration 3912 (2.50952 iter/s, 4.7818s/12 iters), loss = 0.783376 I0409 23:18:10.323678 4221 solver.cpp:237] Train net output #0: loss = 0.783376 (* 1 = 0.783376 loss) I0409 23:18:10.323690 4221 sgd_solver.cpp:105] Iteration 3912, lr = 0.00460744 I0409 23:18:15.258968 4221 solver.cpp:218] Iteration 3924 (2.43157 iter/s, 4.93508s/12 iters), loss = 0.815188 I0409 23:18:15.259013 4221 solver.cpp:237] Train net output #0: loss = 0.815188 (* 1 = 0.815188 loss) I0409 23:18:15.259022 4221 sgd_solver.cpp:105] Iteration 3924, lr = 0.0045965 I0409 23:18:20.299573 4221 solver.cpp:218] Iteration 3936 (2.38079 iter/s, 5.04034s/12 iters), loss = 0.629147 I0409 23:18:20.299612 4221 solver.cpp:237] Train net output #0: loss = 0.629147 (* 1 = 0.629147 loss) I0409 23:18:20.299620 4221 sgd_solver.cpp:105] Iteration 3936, lr = 0.00458559 I0409 23:18:23.523042 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:18:25.086717 4221 solver.cpp:218] Iteration 3948 (2.50684 iter/s, 4.78689s/12 iters), loss = 0.716686 I0409 23:18:25.086766 4221 solver.cpp:237] Train net output #0: loss = 0.716686 (* 1 = 0.716686 loss) I0409 23:18:25.086778 4221 sgd_solver.cpp:105] Iteration 3948, lr = 0.0045747 I0409 23:18:29.929390 4221 solver.cpp:218] Iteration 3960 (2.47811 iter/s, 4.84241s/12 iters), loss = 0.854722 I0409 23:18:29.929450 4221 solver.cpp:237] Train net output #0: loss = 0.854722 (* 1 = 0.854722 loss) I0409 23:18:29.929461 4221 sgd_solver.cpp:105] Iteration 3960, lr = 0.00456384 I0409 23:18:34.758891 4221 solver.cpp:218] Iteration 3972 (2.48487 iter/s, 4.82923s/12 iters), loss = 0.909898 I0409 23:18:34.758942 4221 solver.cpp:237] Train net output #0: loss = 0.909898 (* 1 = 0.909898 loss) I0409 23:18:34.758951 4221 sgd_solver.cpp:105] Iteration 3972, lr = 0.00455301 I0409 23:18:36.710407 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_3978.caffemodel I0409 23:18:37.411151 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_3978.solverstate I0409 23:18:37.913784 4221 solver.cpp:330] Iteration 3978, Testing net (#0) I0409 23:18:37.913808 4221 net.cpp:676] Ignoring source layer train-data I0409 23:18:40.790786 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:18:42.427486 4221 solver.cpp:397] Test net output #0: accuracy = 0.293505 I0409 23:18:42.427522 4221 solver.cpp:397] Test net output #1: loss = 3.46502 (* 1 = 3.46502 loss) I0409 23:18:44.343214 4221 solver.cpp:218] Iteration 3984 (1.25211 iter/s, 9.58386s/12 iters), loss = 0.881151 I0409 23:18:44.343289 4221 solver.cpp:237] Train net output #0: loss = 0.881151 (* 1 = 0.881151 loss) I0409 23:18:44.343304 4221 sgd_solver.cpp:105] Iteration 3984, lr = 0.0045422 I0409 23:18:49.212872 4221 solver.cpp:218] Iteration 3996 (2.46438 iter/s, 4.86937s/12 iters), loss = 1.07481 I0409 23:18:49.212922 4221 solver.cpp:237] Train net output #0: loss = 1.07481 (* 1 = 1.07481 loss) I0409 23:18:49.212934 4221 sgd_solver.cpp:105] Iteration 3996, lr = 0.00453141 I0409 23:18:54.063922 4221 solver.cpp:218] Iteration 4008 (2.47383 iter/s, 4.85078s/12 iters), loss = 1.04675 I0409 23:18:54.063980 4221 solver.cpp:237] Train net output #0: loss = 1.04675 (* 1 = 1.04675 loss) I0409 23:18:54.063990 4221 sgd_solver.cpp:105] Iteration 4008, lr = 0.00452066 I0409 23:18:58.865020 4221 solver.cpp:218] Iteration 4020 (2.49957 iter/s, 4.80083s/12 iters), loss = 1.28259 I0409 23:18:58.865072 4221 solver.cpp:237] Train net output #0: loss = 1.28259 (* 1 = 1.28259 loss) I0409 23:18:58.865082 4221 sgd_solver.cpp:105] Iteration 4020, lr = 0.00450992 I0409 23:19:03.757196 4221 solver.cpp:218] Iteration 4032 (2.45303 iter/s, 4.8919s/12 iters), loss = 1.08895 I0409 23:19:03.757256 4221 solver.cpp:237] Train net output #0: loss = 1.08895 (* 1 = 1.08895 loss) I0409 23:19:03.757268 4221 sgd_solver.cpp:105] Iteration 4032, lr = 0.00449921 I0409 23:19:08.660498 4221 solver.cpp:218] Iteration 4044 (2.44747 iter/s, 4.90302s/12 iters), loss = 1.09347 I0409 23:19:08.660573 4221 solver.cpp:237] Train net output #0: loss = 1.09347 (* 1 = 1.09347 loss) I0409 23:19:08.660589 4221 sgd_solver.cpp:105] Iteration 4044, lr = 0.00448853 I0409 23:19:09.210808 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:19:13.646075 4221 solver.cpp:218] Iteration 4056 (2.40708 iter/s, 4.98529s/12 iters), loss = 0.712282 I0409 23:19:13.646176 4221 solver.cpp:237] Train net output #0: loss = 0.712282 (* 1 = 0.712282 loss) I0409 23:19:13.646190 4221 sgd_solver.cpp:105] Iteration 4056, lr = 0.00447788 I0409 23:19:18.507871 4221 solver.cpp:218] Iteration 4068 (2.46838 iter/s, 4.86149s/12 iters), loss = 0.750674 I0409 23:19:18.507917 4221 solver.cpp:237] Train net output #0: loss = 0.750674 (* 1 = 0.750674 loss) I0409 23:19:18.507925 4221 sgd_solver.cpp:105] Iteration 4068, lr = 0.00446724 I0409 23:19:22.978427 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_4080.caffemodel I0409 23:19:25.518889 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_4080.solverstate I0409 23:19:27.548887 4221 solver.cpp:330] Iteration 4080, Testing net (#0) I0409 23:19:27.548921 4221 net.cpp:676] Ignoring source layer train-data I0409 23:19:30.414477 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:19:32.078531 4221 solver.cpp:397] Test net output #0: accuracy = 0.33701 I0409 23:19:32.078581 4221 solver.cpp:397] Test net output #1: loss = 3.41812 (* 1 = 3.41812 loss) I0409 23:19:32.159873 4221 solver.cpp:218] Iteration 4080 (0.879032 iter/s, 13.6514s/12 iters), loss = 0.989332 I0409 23:19:32.159922 4221 solver.cpp:237] Train net output #0: loss = 0.989332 (* 1 = 0.989332 loss) I0409 23:19:32.159935 4221 sgd_solver.cpp:105] Iteration 4080, lr = 0.00445664 I0409 23:19:36.275444 4221 solver.cpp:218] Iteration 4092 (2.91592 iter/s, 4.11534s/12 iters), loss = 0.898653 I0409 23:19:36.275502 4221 solver.cpp:237] Train net output #0: loss = 0.898653 (* 1 = 0.898653 loss) I0409 23:19:36.275514 4221 sgd_solver.cpp:105] Iteration 4092, lr = 0.00444606 I0409 23:19:41.128767 4221 solver.cpp:218] Iteration 4104 (2.47267 iter/s, 4.85306s/12 iters), loss = 0.927848 I0409 23:19:41.128821 4221 solver.cpp:237] Train net output #0: loss = 0.927848 (* 1 = 0.927848 loss) I0409 23:19:41.128834 4221 sgd_solver.cpp:105] Iteration 4104, lr = 0.0044355 I0409 23:19:45.961712 4221 solver.cpp:218] Iteration 4116 (2.4831 iter/s, 4.83267s/12 iters), loss = 0.906812 I0409 23:19:45.961866 4221 solver.cpp:237] Train net output #0: loss = 0.906812 (* 1 = 0.906812 loss) I0409 23:19:45.961876 4221 sgd_solver.cpp:105] Iteration 4116, lr = 0.00442497 I0409 23:19:50.881386 4221 solver.cpp:218] Iteration 4128 (2.43937 iter/s, 4.91931s/12 iters), loss = 0.778899 I0409 23:19:50.881426 4221 solver.cpp:237] Train net output #0: loss = 0.778899 (* 1 = 0.778899 loss) I0409 23:19:50.881435 4221 sgd_solver.cpp:105] Iteration 4128, lr = 0.00441447 I0409 23:19:56.013878 4221 solver.cpp:218] Iteration 4140 (2.33817 iter/s, 5.13222s/12 iters), loss = 0.714774 I0409 23:19:56.013944 4221 solver.cpp:237] Train net output #0: loss = 0.714774 (* 1 = 0.714774 loss) I0409 23:19:56.013974 4221 sgd_solver.cpp:105] Iteration 4140, lr = 0.00440398 I0409 23:19:58.540141 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:20:00.804951 4221 solver.cpp:218] Iteration 4152 (2.5048 iter/s, 4.7908s/12 iters), loss = 0.691466 I0409 23:20:00.805017 4221 solver.cpp:237] Train net output #0: loss = 0.691466 (* 1 = 0.691466 loss) I0409 23:20:00.805028 4221 sgd_solver.cpp:105] Iteration 4152, lr = 0.00439353 I0409 23:20:01.158149 4221 blocking_queue.cpp:49] Waiting for data I0409 23:20:05.786095 4221 solver.cpp:218] Iteration 4164 (2.40922 iter/s, 4.98086s/12 iters), loss = 0.558491 I0409 23:20:05.786154 4221 solver.cpp:237] Train net output #0: loss = 0.558491 (* 1 = 0.558491 loss) I0409 23:20:05.786167 4221 sgd_solver.cpp:105] Iteration 4164, lr = 0.0043831 I0409 23:20:10.622707 4221 solver.cpp:218] Iteration 4176 (2.48122 iter/s, 4.83633s/12 iters), loss = 0.981476 I0409 23:20:10.622761 4221 solver.cpp:237] Train net output #0: loss = 0.981476 (* 1 = 0.981476 loss) I0409 23:20:10.622771 4221 sgd_solver.cpp:105] Iteration 4176, lr = 0.00437269 I0409 23:20:12.581066 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_4182.caffemodel I0409 23:20:14.469529 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_4182.solverstate I0409 23:20:14.951153 4221 solver.cpp:330] Iteration 4182, Testing net (#0) I0409 23:20:14.951180 4221 net.cpp:676] Ignoring source layer train-data I0409 23:20:17.819680 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:20:19.503291 4221 solver.cpp:397] Test net output #0: accuracy = 0.336397 I0409 23:20:19.503342 4221 solver.cpp:397] Test net output #1: loss = 3.40506 (* 1 = 3.40506 loss) I0409 23:20:21.377859 4221 solver.cpp:218] Iteration 4188 (1.1158 iter/s, 10.7546s/12 iters), loss = 0.768805 I0409 23:20:21.377915 4221 solver.cpp:237] Train net output #0: loss = 0.768805 (* 1 = 0.768805 loss) I0409 23:20:21.377926 4221 sgd_solver.cpp:105] Iteration 4188, lr = 0.00436231 I0409 23:20:26.337993 4221 solver.cpp:218] Iteration 4200 (2.41943 iter/s, 4.95984s/12 iters), loss = 0.629804 I0409 23:20:26.338044 4221 solver.cpp:237] Train net output #0: loss = 0.629804 (* 1 = 0.629804 loss) I0409 23:20:26.338054 4221 sgd_solver.cpp:105] Iteration 4200, lr = 0.00435195 I0409 23:20:31.221566 4221 solver.cpp:218] Iteration 4212 (2.45735 iter/s, 4.88331s/12 iters), loss = 0.61607 I0409 23:20:31.221614 4221 solver.cpp:237] Train net output #0: loss = 0.61607 (* 1 = 0.61607 loss) I0409 23:20:31.221623 4221 sgd_solver.cpp:105] Iteration 4212, lr = 0.00434162 I0409 23:20:36.215588 4221 solver.cpp:218] Iteration 4224 (2.40301 iter/s, 4.99374s/12 iters), loss = 0.828761 I0409 23:20:36.215653 4221 solver.cpp:237] Train net output #0: loss = 0.828761 (* 1 = 0.828761 loss) I0409 23:20:36.215667 4221 sgd_solver.cpp:105] Iteration 4224, lr = 0.00433131 I0409 23:20:41.080302 4221 solver.cpp:218] Iteration 4236 (2.46689 iter/s, 4.86443s/12 iters), loss = 0.995863 I0409 23:20:41.080363 4221 solver.cpp:237] Train net output #0: loss = 0.995863 (* 1 = 0.995863 loss) I0409 23:20:41.080376 4221 sgd_solver.cpp:105] Iteration 4236, lr = 0.00432103 I0409 23:20:45.698457 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:20:45.915709 4221 solver.cpp:218] Iteration 4248 (2.48183 iter/s, 4.83514s/12 iters), loss = 0.722569 I0409 23:20:45.915758 4221 solver.cpp:237] Train net output #0: loss = 0.722569 (* 1 = 0.722569 loss) I0409 23:20:45.915767 4221 sgd_solver.cpp:105] Iteration 4248, lr = 0.00431077 I0409 23:20:50.974889 4221 solver.cpp:218] Iteration 4260 (2.37206 iter/s, 5.0589s/12 iters), loss = 0.757109 I0409 23:20:50.975031 4221 solver.cpp:237] Train net output #0: loss = 0.757109 (* 1 = 0.757109 loss) I0409 23:20:50.975042 4221 sgd_solver.cpp:105] Iteration 4260, lr = 0.00430053 I0409 23:20:55.856675 4221 solver.cpp:218] Iteration 4272 (2.4583 iter/s, 4.88143s/12 iters), loss = 0.673616 I0409 23:20:55.856722 4221 solver.cpp:237] Train net output #0: loss = 0.673616 (* 1 = 0.673616 loss) I0409 23:20:55.856730 4221 sgd_solver.cpp:105] Iteration 4272, lr = 0.00429032 I0409 23:21:00.209753 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_4284.caffemodel I0409 23:21:01.187553 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_4284.solverstate I0409 23:21:01.677384 4221 solver.cpp:330] Iteration 4284, Testing net (#0) I0409 23:21:01.677418 4221 net.cpp:676] Ignoring source layer train-data I0409 23:21:04.442154 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:21:06.248253 4221 solver.cpp:397] Test net output #0: accuracy = 0.343137 I0409 23:21:06.248291 4221 solver.cpp:397] Test net output #1: loss = 3.45565 (* 1 = 3.45565 loss) I0409 23:21:06.331308 4221 solver.cpp:218] Iteration 4284 (1.14568 iter/s, 10.4741s/12 iters), loss = 0.853698 I0409 23:21:06.331368 4221 solver.cpp:237] Train net output #0: loss = 0.853698 (* 1 = 0.853698 loss) I0409 23:21:06.331379 4221 sgd_solver.cpp:105] Iteration 4284, lr = 0.00428014 I0409 23:21:10.492673 4221 solver.cpp:218] Iteration 4296 (2.88384 iter/s, 4.16112s/12 iters), loss = 0.615812 I0409 23:21:10.492733 4221 solver.cpp:237] Train net output #0: loss = 0.615812 (* 1 = 0.615812 loss) I0409 23:21:10.492745 4221 sgd_solver.cpp:105] Iteration 4296, lr = 0.00426998 I0409 23:21:15.473731 4221 solver.cpp:218] Iteration 4308 (2.40926 iter/s, 4.98078s/12 iters), loss = 0.914488 I0409 23:21:15.473780 4221 solver.cpp:237] Train net output #0: loss = 0.914488 (* 1 = 0.914488 loss) I0409 23:21:15.473789 4221 sgd_solver.cpp:105] Iteration 4308, lr = 0.00425984 I0409 23:21:20.563357 4221 solver.cpp:218] Iteration 4320 (2.35786 iter/s, 5.08935s/12 iters), loss = 0.84824 I0409 23:21:20.563417 4221 solver.cpp:237] Train net output #0: loss = 0.84824 (* 1 = 0.84824 loss) I0409 23:21:20.563429 4221 sgd_solver.cpp:105] Iteration 4320, lr = 0.00424972 I0409 23:21:25.525517 4221 solver.cpp:218] Iteration 4332 (2.41844 iter/s, 4.96188s/12 iters), loss = 0.59668 I0409 23:21:25.525672 4221 solver.cpp:237] Train net output #0: loss = 0.59668 (* 1 = 0.59668 loss) I0409 23:21:25.525684 4221 sgd_solver.cpp:105] Iteration 4332, lr = 0.00423964 I0409 23:21:30.351503 4221 solver.cpp:218] Iteration 4344 (2.48673 iter/s, 4.82562s/12 iters), loss = 0.821203 I0409 23:21:30.351569 4221 solver.cpp:237] Train net output #0: loss = 0.821203 (* 1 = 0.821203 loss) I0409 23:21:30.351583 4221 sgd_solver.cpp:105] Iteration 4344, lr = 0.00422957 I0409 23:21:32.182065 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:21:35.144568 4221 solver.cpp:218] Iteration 4356 (2.50376 iter/s, 4.79279s/12 iters), loss = 0.835945 I0409 23:21:35.144632 4221 solver.cpp:237] Train net output #0: loss = 0.835945 (* 1 = 0.835945 loss) I0409 23:21:35.144644 4221 sgd_solver.cpp:105] Iteration 4356, lr = 0.00421953 I0409 23:21:39.941709 4221 solver.cpp:218] Iteration 4368 (2.50164 iter/s, 4.79686s/12 iters), loss = 0.81258 I0409 23:21:39.941771 4221 solver.cpp:237] Train net output #0: loss = 0.81258 (* 1 = 0.81258 loss) I0409 23:21:39.941783 4221 sgd_solver.cpp:105] Iteration 4368, lr = 0.00420951 I0409 23:21:44.879176 4221 solver.cpp:218] Iteration 4380 (2.43053 iter/s, 4.93719s/12 iters), loss = 0.589477 I0409 23:21:44.879223 4221 solver.cpp:237] Train net output #0: loss = 0.589477 (* 1 = 0.589477 loss) I0409 23:21:44.879232 4221 sgd_solver.cpp:105] Iteration 4380, lr = 0.00419952 I0409 23:21:46.839591 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_4386.caffemodel I0409 23:21:47.740705 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_4386.solverstate I0409 23:21:48.227643 4221 solver.cpp:330] Iteration 4386, Testing net (#0) I0409 23:21:48.227669 4221 net.cpp:676] Ignoring source layer train-data I0409 23:21:50.921452 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:21:52.679304 4221 solver.cpp:397] Test net output #0: accuracy = 0.35049 I0409 23:21:52.679342 4221 solver.cpp:397] Test net output #1: loss = 3.43285 (* 1 = 3.43285 loss) I0409 23:21:55.024296 4221 solver.cpp:218] Iteration 4392 (1.18289 iter/s, 10.1446s/12 iters), loss = 0.620185 I0409 23:21:55.024355 4221 solver.cpp:237] Train net output #0: loss = 0.620185 (* 1 = 0.620185 loss) I0409 23:21:55.024367 4221 sgd_solver.cpp:105] Iteration 4392, lr = 0.00418954 I0409 23:22:00.213850 4221 solver.cpp:218] Iteration 4404 (2.31246 iter/s, 5.18927s/12 iters), loss = 0.759502 I0409 23:22:00.213948 4221 solver.cpp:237] Train net output #0: loss = 0.759502 (* 1 = 0.759502 loss) I0409 23:22:00.213984 4221 sgd_solver.cpp:105] Iteration 4404, lr = 0.0041796 I0409 23:22:05.082741 4221 solver.cpp:218] Iteration 4416 (2.46479 iter/s, 4.86858s/12 iters), loss = 0.698907 I0409 23:22:05.082803 4221 solver.cpp:237] Train net output #0: loss = 0.698907 (* 1 = 0.698907 loss) I0409 23:22:05.082815 4221 sgd_solver.cpp:105] Iteration 4416, lr = 0.00416967 I0409 23:22:10.012223 4221 solver.cpp:218] Iteration 4428 (2.43447 iter/s, 4.9292s/12 iters), loss = 0.687902 I0409 23:22:10.012282 4221 solver.cpp:237] Train net output #0: loss = 0.687902 (* 1 = 0.687902 loss) I0409 23:22:10.012293 4221 sgd_solver.cpp:105] Iteration 4428, lr = 0.00415977 I0409 23:22:15.666867 4221 solver.cpp:218] Iteration 4440 (2.12226 iter/s, 5.65434s/12 iters), loss = 0.506901 I0409 23:22:15.666914 4221 solver.cpp:237] Train net output #0: loss = 0.506901 (* 1 = 0.506901 loss) I0409 23:22:15.666924 4221 sgd_solver.cpp:105] Iteration 4440, lr = 0.0041499 I0409 23:22:19.654296 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:22:20.602468 4221 solver.cpp:218] Iteration 4452 (2.43145 iter/s, 4.93533s/12 iters), loss = 0.670309 I0409 23:22:20.602517 4221 solver.cpp:237] Train net output #0: loss = 0.670309 (* 1 = 0.670309 loss) I0409 23:22:20.602526 4221 sgd_solver.cpp:105] Iteration 4452, lr = 0.00414005 I0409 23:22:25.574250 4221 solver.cpp:218] Iteration 4464 (2.41375 iter/s, 4.97151s/12 iters), loss = 0.624443 I0409 23:22:25.574297 4221 solver.cpp:237] Train net output #0: loss = 0.624443 (* 1 = 0.624443 loss) I0409 23:22:25.574306 4221 sgd_solver.cpp:105] Iteration 4464, lr = 0.00413022 I0409 23:22:30.549131 4221 solver.cpp:218] Iteration 4476 (2.41225 iter/s, 4.97462s/12 iters), loss = 0.573783 I0409 23:22:30.549232 4221 solver.cpp:237] Train net output #0: loss = 0.573783 (* 1 = 0.573783 loss) I0409 23:22:30.549242 4221 sgd_solver.cpp:105] Iteration 4476, lr = 0.00412041 I0409 23:22:35.090117 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_4488.caffemodel I0409 23:22:35.741505 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_4488.solverstate I0409 23:22:36.226748 4221 solver.cpp:330] Iteration 4488, Testing net (#0) I0409 23:22:36.226781 4221 net.cpp:676] Ignoring source layer train-data I0409 23:22:38.912214 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:22:40.747992 4221 solver.cpp:397] Test net output #0: accuracy = 0.360294 I0409 23:22:40.748032 4221 solver.cpp:397] Test net output #1: loss = 3.30097 (* 1 = 3.30097 loss) I0409 23:22:40.831276 4221 solver.cpp:218] Iteration 4488 (1.16713 iter/s, 10.2816s/12 iters), loss = 0.86071 I0409 23:22:40.831319 4221 solver.cpp:237] Train net output #0: loss = 0.86071 (* 1 = 0.86071 loss) I0409 23:22:40.831328 4221 sgd_solver.cpp:105] Iteration 4488, lr = 0.00411063 I0409 23:22:44.975066 4221 solver.cpp:218] Iteration 4500 (2.89606 iter/s, 4.14356s/12 iters), loss = 0.780343 I0409 23:22:44.975121 4221 solver.cpp:237] Train net output #0: loss = 0.780343 (* 1 = 0.780343 loss) I0409 23:22:44.975132 4221 sgd_solver.cpp:105] Iteration 4500, lr = 0.00410087 I0409 23:22:49.877141 4221 solver.cpp:218] Iteration 4512 (2.44808 iter/s, 4.9018s/12 iters), loss = 0.570364 I0409 23:22:49.877198 4221 solver.cpp:237] Train net output #0: loss = 0.570364 (* 1 = 0.570364 loss) I0409 23:22:49.877211 4221 sgd_solver.cpp:105] Iteration 4512, lr = 0.00409113 I0409 23:22:54.731041 4221 solver.cpp:218] Iteration 4524 (2.47238 iter/s, 4.85363s/12 iters), loss = 0.530567 I0409 23:22:54.731093 4221 solver.cpp:237] Train net output #0: loss = 0.530567 (* 1 = 0.530567 loss) I0409 23:22:54.731103 4221 sgd_solver.cpp:105] Iteration 4524, lr = 0.00408142 I0409 23:22:59.763264 4221 solver.cpp:218] Iteration 4536 (2.38476 iter/s, 5.03195s/12 iters), loss = 0.448092 I0409 23:22:59.763319 4221 solver.cpp:237] Train net output #0: loss = 0.448092 (* 1 = 0.448092 loss) I0409 23:22:59.763330 4221 sgd_solver.cpp:105] Iteration 4536, lr = 0.00407173 I0409 23:23:04.837185 4221 solver.cpp:218] Iteration 4548 (2.36516 iter/s, 5.07365s/12 iters), loss = 0.427565 I0409 23:23:04.838558 4221 solver.cpp:237] Train net output #0: loss = 0.427565 (* 1 = 0.427565 loss) I0409 23:23:04.838572 4221 sgd_solver.cpp:105] Iteration 4548, lr = 0.00406206 I0409 23:23:06.052544 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:23:09.675827 4221 solver.cpp:218] Iteration 4560 (2.48085 iter/s, 4.83706s/12 iters), loss = 0.521 I0409 23:23:09.675886 4221 solver.cpp:237] Train net output #0: loss = 0.521 (* 1 = 0.521 loss) I0409 23:23:09.675897 4221 sgd_solver.cpp:105] Iteration 4560, lr = 0.00405242 I0409 23:23:14.782881 4221 solver.cpp:218] Iteration 4572 (2.34982 iter/s, 5.10678s/12 iters), loss = 0.645564 I0409 23:23:14.782922 4221 solver.cpp:237] Train net output #0: loss = 0.645564 (* 1 = 0.645564 loss) I0409 23:23:14.782929 4221 sgd_solver.cpp:105] Iteration 4572, lr = 0.0040428 I0409 23:23:19.727870 4221 solver.cpp:218] Iteration 4584 (2.42683 iter/s, 4.94472s/12 iters), loss = 0.732224 I0409 23:23:19.727926 4221 solver.cpp:237] Train net output #0: loss = 0.732224 (* 1 = 0.732224 loss) I0409 23:23:19.727936 4221 sgd_solver.cpp:105] Iteration 4584, lr = 0.0040332 I0409 23:23:21.667538 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_4590.caffemodel I0409 23:23:22.368150 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_4590.solverstate I0409 23:23:22.849401 4221 solver.cpp:330] Iteration 4590, Testing net (#0) I0409 23:23:22.849429 4221 net.cpp:676] Ignoring source layer train-data I0409 23:23:25.360913 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:23:27.179646 4221 solver.cpp:397] Test net output #0: accuracy = 0.352941 I0409 23:23:27.179685 4221 solver.cpp:397] Test net output #1: loss = 3.59511 (* 1 = 3.59511 loss) I0409 23:23:29.125429 4221 solver.cpp:218] Iteration 4596 (1.27699 iter/s, 9.39711s/12 iters), loss = 0.731362 I0409 23:23:29.125473 4221 solver.cpp:237] Train net output #0: loss = 0.731362 (* 1 = 0.731362 loss) I0409 23:23:29.125481 4221 sgd_solver.cpp:105] Iteration 4596, lr = 0.00402362 I0409 23:23:34.057653 4221 solver.cpp:218] Iteration 4608 (2.43311 iter/s, 4.93196s/12 iters), loss = 0.544891 I0409 23:23:34.057703 4221 solver.cpp:237] Train net output #0: loss = 0.544891 (* 1 = 0.544891 loss) I0409 23:23:34.057710 4221 sgd_solver.cpp:105] Iteration 4608, lr = 0.00401407 I0409 23:23:38.952849 4221 solver.cpp:218] Iteration 4620 (2.45152 iter/s, 4.89493s/12 iters), loss = 0.706475 I0409 23:23:38.952986 4221 solver.cpp:237] Train net output #0: loss = 0.706475 (* 1 = 0.706475 loss) I0409 23:23:38.952998 4221 sgd_solver.cpp:105] Iteration 4620, lr = 0.00400454 I0409 23:23:43.870784 4221 solver.cpp:218] Iteration 4632 (2.44022 iter/s, 4.91759s/12 iters), loss = 0.593291 I0409 23:23:43.870832 4221 solver.cpp:237] Train net output #0: loss = 0.593291 (* 1 = 0.593291 loss) I0409 23:23:43.870841 4221 sgd_solver.cpp:105] Iteration 4632, lr = 0.00399503 I0409 23:23:48.792714 4221 solver.cpp:218] Iteration 4644 (2.4382 iter/s, 4.92166s/12 iters), loss = 0.65448 I0409 23:23:48.792763 4221 solver.cpp:237] Train net output #0: loss = 0.65448 (* 1 = 0.65448 loss) I0409 23:23:48.792773 4221 sgd_solver.cpp:105] Iteration 4644, lr = 0.00398555 I0409 23:23:52.041788 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:23:53.580265 4221 solver.cpp:218] Iteration 4656 (2.50664 iter/s, 4.78728s/12 iters), loss = 0.813777 I0409 23:23:53.580328 4221 solver.cpp:237] Train net output #0: loss = 0.813777 (* 1 = 0.813777 loss) I0409 23:23:53.580340 4221 sgd_solver.cpp:105] Iteration 4656, lr = 0.00397608 I0409 23:23:58.393311 4221 solver.cpp:218] Iteration 4668 (2.49337 iter/s, 4.81277s/12 iters), loss = 0.452823 I0409 23:23:58.393375 4221 solver.cpp:237] Train net output #0: loss = 0.452823 (* 1 = 0.452823 loss) I0409 23:23:58.393388 4221 sgd_solver.cpp:105] Iteration 4668, lr = 0.00396664 I0409 23:24:03.158584 4221 solver.cpp:218] Iteration 4680 (2.51836 iter/s, 4.765s/12 iters), loss = 0.808291 I0409 23:24:03.158641 4221 solver.cpp:237] Train net output #0: loss = 0.808291 (* 1 = 0.808291 loss) I0409 23:24:03.158653 4221 sgd_solver.cpp:105] Iteration 4680, lr = 0.00395723 I0409 23:24:07.672996 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_4692.caffemodel I0409 23:24:09.983656 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_4692.solverstate I0409 23:24:10.772471 4221 solver.cpp:330] Iteration 4692, Testing net (#0) I0409 23:24:10.772493 4221 net.cpp:676] Ignoring source layer train-data I0409 23:24:13.352787 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:24:15.231876 4221 solver.cpp:397] Test net output #0: accuracy = 0.365196 I0409 23:24:15.231917 4221 solver.cpp:397] Test net output #1: loss = 3.4974 (* 1 = 3.4974 loss) I0409 23:24:15.315045 4221 solver.cpp:218] Iteration 4692 (0.987175 iter/s, 12.1559s/12 iters), loss = 0.664155 I0409 23:24:15.315094 4221 solver.cpp:237] Train net output #0: loss = 0.664155 (* 1 = 0.664155 loss) I0409 23:24:15.315104 4221 sgd_solver.cpp:105] Iteration 4692, lr = 0.00394783 I0409 23:24:19.533450 4221 solver.cpp:218] Iteration 4704 (2.84484 iter/s, 4.21817s/12 iters), loss = 0.490401 I0409 23:24:19.533500 4221 solver.cpp:237] Train net output #0: loss = 0.490401 (* 1 = 0.490401 loss) I0409 23:24:19.533509 4221 sgd_solver.cpp:105] Iteration 4704, lr = 0.00393846 I0409 23:24:24.402936 4221 solver.cpp:218] Iteration 4716 (2.46446 iter/s, 4.86922s/12 iters), loss = 0.770133 I0409 23:24:24.402993 4221 solver.cpp:237] Train net output #0: loss = 0.770133 (* 1 = 0.770133 loss) I0409 23:24:24.403004 4221 sgd_solver.cpp:105] Iteration 4716, lr = 0.00392911 I0409 23:24:29.237784 4221 solver.cpp:218] Iteration 4728 (2.48212 iter/s, 4.83458s/12 iters), loss = 0.6628 I0409 23:24:29.237828 4221 solver.cpp:237] Train net output #0: loss = 0.6628 (* 1 = 0.6628 loss) I0409 23:24:29.237835 4221 sgd_solver.cpp:105] Iteration 4728, lr = 0.00391978 I0409 23:24:34.019176 4221 solver.cpp:218] Iteration 4740 (2.50986 iter/s, 4.78114s/12 iters), loss = 0.593828 I0409 23:24:34.019227 4221 solver.cpp:237] Train net output #0: loss = 0.593828 (* 1 = 0.593828 loss) I0409 23:24:34.019237 4221 sgd_solver.cpp:105] Iteration 4740, lr = 0.00391047 I0409 23:24:38.832620 4221 solver.cpp:218] Iteration 4752 (2.49315 iter/s, 4.81318s/12 iters), loss = 0.61338 I0409 23:24:38.832661 4221 solver.cpp:237] Train net output #0: loss = 0.61338 (* 1 = 0.61338 loss) I0409 23:24:38.832669 4221 sgd_solver.cpp:105] Iteration 4752, lr = 0.00390119 I0409 23:24:39.337716 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:24:43.850343 4221 solver.cpp:218] Iteration 4764 (2.39165 iter/s, 5.01746s/12 iters), loss = 0.655319 I0409 23:24:43.850504 4221 solver.cpp:237] Train net output #0: loss = 0.655319 (* 1 = 0.655319 loss) I0409 23:24:43.850517 4221 sgd_solver.cpp:105] Iteration 4764, lr = 0.00389193 I0409 23:24:48.653097 4221 solver.cpp:218] Iteration 4776 (2.49876 iter/s, 4.80238s/12 iters), loss = 0.604197 I0409 23:24:48.653156 4221 solver.cpp:237] Train net output #0: loss = 0.604197 (* 1 = 0.604197 loss) I0409 23:24:48.653168 4221 sgd_solver.cpp:105] Iteration 4776, lr = 0.00388269 I0409 23:24:53.478011 4221 solver.cpp:218] Iteration 4788 (2.48723 iter/s, 4.82464s/12 iters), loss = 0.557006 I0409 23:24:53.478062 4221 solver.cpp:237] Train net output #0: loss = 0.557006 (* 1 = 0.557006 loss) I0409 23:24:53.478072 4221 sgd_solver.cpp:105] Iteration 4788, lr = 0.00387347 I0409 23:24:55.431450 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_4794.caffemodel I0409 23:24:56.120111 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_4794.solverstate I0409 23:24:56.849151 4221 solver.cpp:330] Iteration 4794, Testing net (#0) I0409 23:24:56.849171 4221 net.cpp:676] Ignoring source layer train-data I0409 23:24:59.443711 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:25:01.354233 4221 solver.cpp:397] Test net output #0: accuracy = 0.387255 I0409 23:25:01.354276 4221 solver.cpp:397] Test net output #1: loss = 3.41566 (* 1 = 3.41566 loss) I0409 23:25:03.088922 4221 solver.cpp:218] Iteration 4800 (1.24864 iter/s, 9.61045s/12 iters), loss = 0.539119 I0409 23:25:03.088974 4221 solver.cpp:237] Train net output #0: loss = 0.539119 (* 1 = 0.539119 loss) I0409 23:25:03.088982 4221 sgd_solver.cpp:105] Iteration 4800, lr = 0.00386427 I0409 23:25:07.993470 4221 solver.cpp:218] Iteration 4812 (2.44684 iter/s, 4.90428s/12 iters), loss = 0.653143 I0409 23:25:07.993525 4221 solver.cpp:237] Train net output #0: loss = 0.653143 (* 1 = 0.653143 loss) I0409 23:25:07.993536 4221 sgd_solver.cpp:105] Iteration 4812, lr = 0.0038551 I0409 23:25:12.842229 4221 solver.cpp:218] Iteration 4824 (2.47499 iter/s, 4.8485s/12 iters), loss = 0.663112 I0409 23:25:12.842267 4221 solver.cpp:237] Train net output #0: loss = 0.663112 (* 1 = 0.663112 loss) I0409 23:25:12.842276 4221 sgd_solver.cpp:105] Iteration 4824, lr = 0.00384594 I0409 23:25:17.852418 4221 solver.cpp:218] Iteration 4836 (2.39524 iter/s, 5.00993s/12 iters), loss = 0.593891 I0409 23:25:17.852550 4221 solver.cpp:237] Train net output #0: loss = 0.593891 (* 1 = 0.593891 loss) I0409 23:25:17.852558 4221 sgd_solver.cpp:105] Iteration 4836, lr = 0.00383681 I0409 23:25:18.686944 4221 blocking_queue.cpp:49] Waiting for data I0409 23:25:22.862013 4221 solver.cpp:218] Iteration 4848 (2.39557 iter/s, 5.00925s/12 iters), loss = 0.683318 I0409 23:25:22.862061 4221 solver.cpp:237] Train net output #0: loss = 0.683318 (* 1 = 0.683318 loss) I0409 23:25:22.862071 4221 sgd_solver.cpp:105] Iteration 4848, lr = 0.0038277 I0409 23:25:25.472913 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:25:27.716477 4221 solver.cpp:218] Iteration 4860 (2.47208 iter/s, 4.8542s/12 iters), loss = 0.497385 I0409 23:25:27.716536 4221 solver.cpp:237] Train net output #0: loss = 0.497385 (* 1 = 0.497385 loss) I0409 23:25:27.716547 4221 sgd_solver.cpp:105] Iteration 4860, lr = 0.00381862 I0409 23:25:32.502257 4221 solver.cpp:218] Iteration 4872 (2.50757 iter/s, 4.78551s/12 iters), loss = 0.643173 I0409 23:25:32.502324 4221 solver.cpp:237] Train net output #0: loss = 0.643173 (* 1 = 0.643173 loss) I0409 23:25:32.502336 4221 sgd_solver.cpp:105] Iteration 4872, lr = 0.00380955 I0409 23:25:37.272581 4221 solver.cpp:218] Iteration 4884 (2.5157 iter/s, 4.77005s/12 iters), loss = 0.458412 I0409 23:25:37.272642 4221 solver.cpp:237] Train net output #0: loss = 0.458412 (* 1 = 0.458412 loss) I0409 23:25:37.272655 4221 sgd_solver.cpp:105] Iteration 4884, lr = 0.0038005 I0409 23:25:41.718730 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_4896.caffemodel I0409 23:25:42.540235 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_4896.solverstate I0409 23:25:43.173173 4221 solver.cpp:330] Iteration 4896, Testing net (#0) I0409 23:25:43.173202 4221 net.cpp:676] Ignoring source layer train-data I0409 23:25:45.605562 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:25:47.563360 4221 solver.cpp:397] Test net output #0: accuracy = 0.378064 I0409 23:25:47.563410 4221 solver.cpp:397] Test net output #1: loss = 3.42426 (* 1 = 3.42426 loss) I0409 23:25:47.645035 4221 solver.cpp:218] Iteration 4896 (1.15697 iter/s, 10.372s/12 iters), loss = 0.542082 I0409 23:25:47.645095 4221 solver.cpp:237] Train net output #0: loss = 0.542082 (* 1 = 0.542082 loss) I0409 23:25:47.645107 4221 sgd_solver.cpp:105] Iteration 4896, lr = 0.00379148 I0409 23:25:51.649413 4221 solver.cpp:218] Iteration 4908 (2.9969 iter/s, 4.00414s/12 iters), loss = 0.59926 I0409 23:25:51.649536 4221 solver.cpp:237] Train net output #0: loss = 0.59926 (* 1 = 0.59926 loss) I0409 23:25:51.649549 4221 sgd_solver.cpp:105] Iteration 4908, lr = 0.00378248 I0409 23:25:56.499491 4221 solver.cpp:218] Iteration 4920 (2.47436 iter/s, 4.84974s/12 iters), loss = 0.542728 I0409 23:25:56.499550 4221 solver.cpp:237] Train net output #0: loss = 0.542728 (* 1 = 0.542728 loss) I0409 23:25:56.499562 4221 sgd_solver.cpp:105] Iteration 4920, lr = 0.0037735 I0409 23:26:01.341331 4221 solver.cpp:218] Iteration 4932 (2.47854 iter/s, 4.84157s/12 iters), loss = 0.561028 I0409 23:26:01.341394 4221 solver.cpp:237] Train net output #0: loss = 0.561028 (* 1 = 0.561028 loss) I0409 23:26:01.341408 4221 sgd_solver.cpp:105] Iteration 4932, lr = 0.00376454 I0409 23:26:06.210237 4221 solver.cpp:218] Iteration 4944 (2.46476 iter/s, 4.86863s/12 iters), loss = 0.545163 I0409 23:26:06.210278 4221 solver.cpp:237] Train net output #0: loss = 0.545163 (* 1 = 0.545163 loss) I0409 23:26:06.210286 4221 sgd_solver.cpp:105] Iteration 4944, lr = 0.0037556 I0409 23:26:10.898403 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:26:11.084967 4221 solver.cpp:218] Iteration 4956 (2.46181 iter/s, 4.87447s/12 iters), loss = 0.497671 I0409 23:26:11.085024 4221 solver.cpp:237] Train net output #0: loss = 0.497671 (* 1 = 0.497671 loss) I0409 23:26:11.085036 4221 sgd_solver.cpp:105] Iteration 4956, lr = 0.00374669 I0409 23:26:15.995543 4221 solver.cpp:218] Iteration 4968 (2.44385 iter/s, 4.91029s/12 iters), loss = 0.577721 I0409 23:26:15.995615 4221 solver.cpp:237] Train net output #0: loss = 0.577721 (* 1 = 0.577721 loss) I0409 23:26:15.995630 4221 sgd_solver.cpp:105] Iteration 4968, lr = 0.00373779 I0409 23:26:20.879884 4221 solver.cpp:218] Iteration 4980 (2.45697 iter/s, 4.88406s/12 iters), loss = 0.593796 I0409 23:26:20.879928 4221 solver.cpp:237] Train net output #0: loss = 0.593796 (* 1 = 0.593796 loss) I0409 23:26:20.879936 4221 sgd_solver.cpp:105] Iteration 4980, lr = 0.00372892 I0409 23:26:25.791893 4221 solver.cpp:218] Iteration 4992 (2.44312 iter/s, 4.91175s/12 iters), loss = 0.766923 I0409 23:26:25.792060 4221 solver.cpp:237] Train net output #0: loss = 0.766923 (* 1 = 0.766923 loss) I0409 23:26:25.792073 4221 sgd_solver.cpp:105] Iteration 4992, lr = 0.00372006 I0409 23:26:27.761269 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_4998.caffemodel I0409 23:26:28.459877 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_4998.solverstate I0409 23:26:28.952126 4221 solver.cpp:330] Iteration 4998, Testing net (#0) I0409 23:26:28.952148 4221 net.cpp:676] Ignoring source layer train-data I0409 23:26:31.577860 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:26:33.564231 4221 solver.cpp:397] Test net output #0: accuracy = 0.372549 I0409 23:26:33.564282 4221 solver.cpp:397] Test net output #1: loss = 3.55572 (* 1 = 3.55572 loss) I0409 23:26:35.350697 4221 solver.cpp:218] Iteration 5004 (1.25546 iter/s, 9.55824s/12 iters), loss = 0.359861 I0409 23:26:35.350744 4221 solver.cpp:237] Train net output #0: loss = 0.359861 (* 1 = 0.359861 loss) I0409 23:26:35.350752 4221 sgd_solver.cpp:105] Iteration 5004, lr = 0.00371123 I0409 23:26:40.263375 4221 solver.cpp:218] Iteration 5016 (2.44279 iter/s, 4.91241s/12 iters), loss = 0.621054 I0409 23:26:40.263425 4221 solver.cpp:237] Train net output #0: loss = 0.621054 (* 1 = 0.621054 loss) I0409 23:26:40.263434 4221 sgd_solver.cpp:105] Iteration 5016, lr = 0.00370242 I0409 23:26:45.191565 4221 solver.cpp:218] Iteration 5028 (2.4351 iter/s, 4.92792s/12 iters), loss = 0.431745 I0409 23:26:45.191610 4221 solver.cpp:237] Train net output #0: loss = 0.431745 (* 1 = 0.431745 loss) I0409 23:26:45.191619 4221 sgd_solver.cpp:105] Iteration 5028, lr = 0.00369363 I0409 23:26:50.109467 4221 solver.cpp:218] Iteration 5040 (2.4402 iter/s, 4.91764s/12 iters), loss = 0.408379 I0409 23:26:50.109517 4221 solver.cpp:237] Train net output #0: loss = 0.408379 (* 1 = 0.408379 loss) I0409 23:26:50.109527 4221 sgd_solver.cpp:105] Iteration 5040, lr = 0.00368486 I0409 23:26:55.115567 4221 solver.cpp:218] Iteration 5052 (2.3972 iter/s, 5.00583s/12 iters), loss = 0.366023 I0409 23:26:55.115620 4221 solver.cpp:237] Train net output #0: loss = 0.366023 (* 1 = 0.366023 loss) I0409 23:26:55.115630 4221 sgd_solver.cpp:105] Iteration 5052, lr = 0.00367611 I0409 23:26:57.399236 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:27:00.568356 4221 solver.cpp:218] Iteration 5064 (2.20083 iter/s, 5.4525s/12 iters), loss = 0.560331 I0409 23:27:00.568413 4221 solver.cpp:237] Train net output #0: loss = 0.560331 (* 1 = 0.560331 loss) I0409 23:27:00.568425 4221 sgd_solver.cpp:105] Iteration 5064, lr = 0.00366738 I0409 23:27:05.770856 4221 solver.cpp:218] Iteration 5076 (2.30671 iter/s, 5.20221s/12 iters), loss = 0.520092 I0409 23:27:05.770920 4221 solver.cpp:237] Train net output #0: loss = 0.520092 (* 1 = 0.520092 loss) I0409 23:27:05.770932 4221 sgd_solver.cpp:105] Iteration 5076, lr = 0.00365868 I0409 23:27:10.618556 4221 solver.cpp:218] Iteration 5088 (2.47554 iter/s, 4.84743s/12 iters), loss = 0.470144 I0409 23:27:10.618611 4221 solver.cpp:237] Train net output #0: loss = 0.470144 (* 1 = 0.470144 loss) I0409 23:27:10.618621 4221 sgd_solver.cpp:105] Iteration 5088, lr = 0.00364999 I0409 23:27:15.118647 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_5100.caffemodel I0409 23:27:15.815026 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_5100.solverstate I0409 23:27:16.324880 4221 solver.cpp:330] Iteration 5100, Testing net (#0) I0409 23:27:16.324908 4221 net.cpp:676] Ignoring source layer train-data I0409 23:27:18.800308 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:27:20.814832 4221 solver.cpp:397] Test net output #0: accuracy = 0.370711 I0409 23:27:20.814882 4221 solver.cpp:397] Test net output #1: loss = 3.61005 (* 1 = 3.61005 loss) I0409 23:27:20.897815 4221 solver.cpp:218] Iteration 5100 (1.16745 iter/s, 10.2788s/12 iters), loss = 0.712708 I0409 23:27:20.897859 4221 solver.cpp:237] Train net output #0: loss = 0.712708 (* 1 = 0.712708 loss) I0409 23:27:20.897869 4221 sgd_solver.cpp:105] Iteration 5100, lr = 0.00364132 I0409 23:27:25.038020 4221 solver.cpp:218] Iteration 5112 (2.89857 iter/s, 4.13998s/12 iters), loss = 0.57111 I0409 23:27:25.038069 4221 solver.cpp:237] Train net output #0: loss = 0.57111 (* 1 = 0.57111 loss) I0409 23:27:25.038076 4221 sgd_solver.cpp:105] Iteration 5112, lr = 0.00363268 I0409 23:27:29.866515 4221 solver.cpp:218] Iteration 5124 (2.48538 iter/s, 4.82823s/12 iters), loss = 0.659307 I0409 23:27:29.866663 4221 solver.cpp:237] Train net output #0: loss = 0.659307 (* 1 = 0.659307 loss) I0409 23:27:29.866675 4221 sgd_solver.cpp:105] Iteration 5124, lr = 0.00362405 I0409 23:27:35.142374 4221 solver.cpp:218] Iteration 5136 (2.27467 iter/s, 5.27548s/12 iters), loss = 0.469113 I0409 23:27:35.142434 4221 solver.cpp:237] Train net output #0: loss = 0.469113 (* 1 = 0.469113 loss) I0409 23:27:35.142452 4221 sgd_solver.cpp:105] Iteration 5136, lr = 0.00361545 I0409 23:27:39.978962 4221 solver.cpp:218] Iteration 5148 (2.48123 iter/s, 4.83631s/12 iters), loss = 0.346556 I0409 23:27:39.979023 4221 solver.cpp:237] Train net output #0: loss = 0.346556 (* 1 = 0.346556 loss) I0409 23:27:39.979038 4221 sgd_solver.cpp:105] Iteration 5148, lr = 0.00360687 I0409 23:27:43.907882 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:27:44.798153 4221 solver.cpp:218] Iteration 5160 (2.49018 iter/s, 4.81892s/12 iters), loss = 0.625366 I0409 23:27:44.798211 4221 solver.cpp:237] Train net output #0: loss = 0.625366 (* 1 = 0.625366 loss) I0409 23:27:44.798223 4221 sgd_solver.cpp:105] Iteration 5160, lr = 0.0035983 I0409 23:27:49.639787 4221 solver.cpp:218] Iteration 5172 (2.47864 iter/s, 4.84136s/12 iters), loss = 0.395186 I0409 23:27:49.639845 4221 solver.cpp:237] Train net output #0: loss = 0.395186 (* 1 = 0.395186 loss) I0409 23:27:49.639858 4221 sgd_solver.cpp:105] Iteration 5172, lr = 0.00358976 I0409 23:27:54.429239 4221 solver.cpp:218] Iteration 5184 (2.50565 iter/s, 4.78918s/12 iters), loss = 0.368724 I0409 23:27:54.429291 4221 solver.cpp:237] Train net output #0: loss = 0.368724 (* 1 = 0.368724 loss) I0409 23:27:54.429301 4221 sgd_solver.cpp:105] Iteration 5184, lr = 0.00358124 I0409 23:27:59.254760 4221 solver.cpp:218] Iteration 5196 (2.48691 iter/s, 4.82526s/12 iters), loss = 0.645713 I0409 23:27:59.254808 4221 solver.cpp:237] Train net output #0: loss = 0.645713 (* 1 = 0.645713 loss) I0409 23:27:59.254817 4221 sgd_solver.cpp:105] Iteration 5196, lr = 0.00357273 I0409 23:28:01.257632 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_5202.caffemodel I0409 23:28:01.954147 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_5202.solverstate I0409 23:28:02.447686 4221 solver.cpp:330] Iteration 5202, Testing net (#0) I0409 23:28:02.447705 4221 net.cpp:676] Ignoring source layer train-data I0409 23:28:05.043699 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:28:07.122689 4221 solver.cpp:397] Test net output #0: accuracy = 0.381127 I0409 23:28:07.122727 4221 solver.cpp:397] Test net output #1: loss = 3.53147 (* 1 = 3.53147 loss) I0409 23:28:09.081487 4221 solver.cpp:218] Iteration 5208 (1.22122 iter/s, 9.82626s/12 iters), loss = 0.748553 I0409 23:28:09.081544 4221 solver.cpp:237] Train net output #0: loss = 0.748553 (* 1 = 0.748553 loss) I0409 23:28:09.081558 4221 sgd_solver.cpp:105] Iteration 5208, lr = 0.00356425 I0409 23:28:13.979810 4221 solver.cpp:218] Iteration 5220 (2.44995 iter/s, 4.89805s/12 iters), loss = 0.472654 I0409 23:28:13.979856 4221 solver.cpp:237] Train net output #0: loss = 0.472654 (* 1 = 0.472654 loss) I0409 23:28:13.979864 4221 sgd_solver.cpp:105] Iteration 5220, lr = 0.00355579 I0409 23:28:18.851915 4221 solver.cpp:218] Iteration 5232 (2.46313 iter/s, 4.87184s/12 iters), loss = 0.478913 I0409 23:28:18.851974 4221 solver.cpp:237] Train net output #0: loss = 0.478913 (* 1 = 0.478913 loss) I0409 23:28:18.851995 4221 sgd_solver.cpp:105] Iteration 5232, lr = 0.00354735 I0409 23:28:23.777911 4221 solver.cpp:218] Iteration 5244 (2.43619 iter/s, 4.92573s/12 iters), loss = 0.451916 I0409 23:28:23.777971 4221 solver.cpp:237] Train net output #0: loss = 0.451916 (* 1 = 0.451916 loss) I0409 23:28:23.777979 4221 sgd_solver.cpp:105] Iteration 5244, lr = 0.00353892 I0409 23:28:28.821647 4221 solver.cpp:218] Iteration 5256 (2.37932 iter/s, 5.04347s/12 iters), loss = 0.474637 I0409 23:28:28.821702 4221 solver.cpp:237] Train net output #0: loss = 0.474637 (* 1 = 0.474637 loss) I0409 23:28:28.821713 4221 sgd_solver.cpp:105] Iteration 5256, lr = 0.00353052 I0409 23:28:30.086870 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:28:33.728010 4221 solver.cpp:218] Iteration 5268 (2.44594 iter/s, 4.90609s/12 iters), loss = 0.585804 I0409 23:28:33.728164 4221 solver.cpp:237] Train net output #0: loss = 0.585804 (* 1 = 0.585804 loss) I0409 23:28:33.728178 4221 sgd_solver.cpp:105] Iteration 5268, lr = 0.00352214 I0409 23:28:38.613847 4221 solver.cpp:218] Iteration 5280 (2.45626 iter/s, 4.88547s/12 iters), loss = 0.459323 I0409 23:28:38.613903 4221 solver.cpp:237] Train net output #0: loss = 0.459323 (* 1 = 0.459323 loss) I0409 23:28:38.613914 4221 sgd_solver.cpp:105] Iteration 5280, lr = 0.00351378 I0409 23:28:43.488843 4221 solver.cpp:218] Iteration 5292 (2.46168 iter/s, 4.87472s/12 iters), loss = 0.516604 I0409 23:28:43.488915 4221 solver.cpp:237] Train net output #0: loss = 0.516604 (* 1 = 0.516604 loss) I0409 23:28:43.488927 4221 sgd_solver.cpp:105] Iteration 5292, lr = 0.00350544 I0409 23:28:47.928848 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_5304.caffemodel I0409 23:28:48.646134 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_5304.solverstate I0409 23:28:49.395503 4221 solver.cpp:330] Iteration 5304, Testing net (#0) I0409 23:28:49.395540 4221 net.cpp:676] Ignoring source layer train-data I0409 23:28:51.808326 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:28:54.069180 4221 solver.cpp:397] Test net output #0: accuracy = 0.393382 I0409 23:28:54.069223 4221 solver.cpp:397] Test net output #1: loss = 3.48645 (* 1 = 3.48645 loss) I0409 23:28:54.152742 4221 solver.cpp:218] Iteration 5304 (1.12535 iter/s, 10.6634s/12 iters), loss = 0.413056 I0409 23:28:54.152791 4221 solver.cpp:237] Train net output #0: loss = 0.413056 (* 1 = 0.413056 loss) I0409 23:28:54.152801 4221 sgd_solver.cpp:105] Iteration 5304, lr = 0.00349711 I0409 23:28:58.328595 4221 solver.cpp:218] Iteration 5316 (2.87383 iter/s, 4.17562s/12 iters), loss = 0.510089 I0409 23:28:58.328657 4221 solver.cpp:237] Train net output #0: loss = 0.510089 (* 1 = 0.510089 loss) I0409 23:28:58.328671 4221 sgd_solver.cpp:105] Iteration 5316, lr = 0.00348881 I0409 23:29:03.139823 4221 solver.cpp:218] Iteration 5328 (2.4943 iter/s, 4.81096s/12 iters), loss = 0.284596 I0409 23:29:03.139865 4221 solver.cpp:237] Train net output #0: loss = 0.284596 (* 1 = 0.284596 loss) I0409 23:29:03.139874 4221 sgd_solver.cpp:105] Iteration 5328, lr = 0.00348053 I0409 23:29:07.971210 4221 solver.cpp:218] Iteration 5340 (2.48389 iter/s, 4.83112s/12 iters), loss = 0.511792 I0409 23:29:07.971395 4221 solver.cpp:237] Train net output #0: loss = 0.511792 (* 1 = 0.511792 loss) I0409 23:29:07.971410 4221 sgd_solver.cpp:105] Iteration 5340, lr = 0.00347226 I0409 23:29:12.784657 4221 solver.cpp:218] Iteration 5352 (2.49322 iter/s, 4.81306s/12 iters), loss = 0.65867 I0409 23:29:12.784696 4221 solver.cpp:237] Train net output #0: loss = 0.65867 (* 1 = 0.65867 loss) I0409 23:29:12.784704 4221 sgd_solver.cpp:105] Iteration 5352, lr = 0.00346402 I0409 23:29:16.445575 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:29:17.950794 4221 solver.cpp:218] Iteration 5364 (2.32294 iter/s, 5.16587s/12 iters), loss = 0.677747 I0409 23:29:17.950851 4221 solver.cpp:237] Train net output #0: loss = 0.677747 (* 1 = 0.677747 loss) I0409 23:29:17.950866 4221 sgd_solver.cpp:105] Iteration 5364, lr = 0.0034558 I0409 23:29:22.806195 4221 solver.cpp:218] Iteration 5376 (2.47161 iter/s, 4.85513s/12 iters), loss = 0.497443 I0409 23:29:22.806246 4221 solver.cpp:237] Train net output #0: loss = 0.497443 (* 1 = 0.497443 loss) I0409 23:29:22.806257 4221 sgd_solver.cpp:105] Iteration 5376, lr = 0.00344759 I0409 23:29:27.614562 4221 solver.cpp:218] Iteration 5388 (2.49579 iter/s, 4.80811s/12 iters), loss = 0.310757 I0409 23:29:27.614616 4221 solver.cpp:237] Train net output #0: loss = 0.310757 (* 1 = 0.310757 loss) I0409 23:29:27.614629 4221 sgd_solver.cpp:105] Iteration 5388, lr = 0.00343941 I0409 23:29:32.458146 4221 solver.cpp:218] Iteration 5400 (2.47764 iter/s, 4.84332s/12 iters), loss = 0.500814 I0409 23:29:32.458200 4221 solver.cpp:237] Train net output #0: loss = 0.500814 (* 1 = 0.500814 loss) I0409 23:29:32.458210 4221 sgd_solver.cpp:105] Iteration 5400, lr = 0.00343124 I0409 23:29:34.459339 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_5406.caffemodel I0409 23:29:35.114312 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_5406.solverstate I0409 23:29:35.911967 4221 solver.cpp:330] Iteration 5406, Testing net (#0) I0409 23:29:35.911993 4221 net.cpp:676] Ignoring source layer train-data I0409 23:29:38.162187 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:29:40.303755 4221 solver.cpp:397] Test net output #0: accuracy = 0.380515 I0409 23:29:40.303795 4221 solver.cpp:397] Test net output #1: loss = 3.61086 (* 1 = 3.61086 loss) I0409 23:29:42.145325 4221 solver.cpp:218] Iteration 5412 (1.23881 iter/s, 9.68672s/12 iters), loss = 0.447685 I0409 23:29:42.145372 4221 solver.cpp:237] Train net output #0: loss = 0.447685 (* 1 = 0.447685 loss) I0409 23:29:42.145383 4221 sgd_solver.cpp:105] Iteration 5412, lr = 0.00342309 I0409 23:29:47.112828 4221 solver.cpp:218] Iteration 5424 (2.41583 iter/s, 4.96724s/12 iters), loss = 0.671363 I0409 23:29:47.112875 4221 solver.cpp:237] Train net output #0: loss = 0.671363 (* 1 = 0.671363 loss) I0409 23:29:47.112884 4221 sgd_solver.cpp:105] Iteration 5424, lr = 0.00341497 I0409 23:29:52.153141 4221 solver.cpp:218] Iteration 5436 (2.38093 iter/s, 5.04005s/12 iters), loss = 0.420694 I0409 23:29:52.153189 4221 solver.cpp:237] Train net output #0: loss = 0.420694 (* 1 = 0.420694 loss) I0409 23:29:52.153201 4221 sgd_solver.cpp:105] Iteration 5436, lr = 0.00340686 I0409 23:29:57.104034 4221 solver.cpp:218] Iteration 5448 (2.42393 iter/s, 4.95063s/12 iters), loss = 0.414837 I0409 23:29:57.104094 4221 solver.cpp:237] Train net output #0: loss = 0.414837 (* 1 = 0.414837 loss) I0409 23:29:57.104106 4221 sgd_solver.cpp:105] Iteration 5448, lr = 0.00339877 I0409 23:30:01.970197 4221 solver.cpp:218] Iteration 5460 (2.46615 iter/s, 4.86589s/12 iters), loss = 0.303816 I0409 23:30:01.970250 4221 solver.cpp:237] Train net output #0: loss = 0.303816 (* 1 = 0.303816 loss) I0409 23:30:01.970263 4221 sgd_solver.cpp:105] Iteration 5460, lr = 0.0033907 I0409 23:30:02.524825 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:30:06.854104 4221 solver.cpp:218] Iteration 5472 (2.45719 iter/s, 4.88364s/12 iters), loss = 0.285576 I0409 23:30:06.854158 4221 solver.cpp:237] Train net output #0: loss = 0.285576 (* 1 = 0.285576 loss) I0409 23:30:06.854168 4221 sgd_solver.cpp:105] Iteration 5472, lr = 0.00338265 I0409 23:30:11.739562 4221 solver.cpp:218] Iteration 5484 (2.4564 iter/s, 4.8852s/12 iters), loss = 0.431286 I0409 23:30:11.739691 4221 solver.cpp:237] Train net output #0: loss = 0.431286 (* 1 = 0.431286 loss) I0409 23:30:11.739702 4221 sgd_solver.cpp:105] Iteration 5484, lr = 0.00337462 I0409 23:30:16.622331 4221 solver.cpp:218] Iteration 5496 (2.45779 iter/s, 4.88243s/12 iters), loss = 0.406614 I0409 23:30:16.622375 4221 solver.cpp:237] Train net output #0: loss = 0.406614 (* 1 = 0.406614 loss) I0409 23:30:16.622385 4221 sgd_solver.cpp:105] Iteration 5496, lr = 0.00336661 I0409 23:30:21.061981 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_5508.caffemodel I0409 23:30:21.735723 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_5508.solverstate I0409 23:30:22.498241 4221 solver.cpp:330] Iteration 5508, Testing net (#0) I0409 23:30:22.498276 4221 net.cpp:676] Ignoring source layer train-data I0409 23:30:24.689167 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:30:26.861002 4221 solver.cpp:397] Test net output #0: accuracy = 0.375613 I0409 23:30:26.861052 4221 solver.cpp:397] Test net output #1: loss = 3.64398 (* 1 = 3.64398 loss) I0409 23:30:26.944479 4221 solver.cpp:218] Iteration 5508 (1.1626 iter/s, 10.3217s/12 iters), loss = 0.285909 I0409 23:30:26.944531 4221 solver.cpp:237] Train net output #0: loss = 0.285909 (* 1 = 0.285909 loss) I0409 23:30:26.944543 4221 sgd_solver.cpp:105] Iteration 5508, lr = 0.00335861 I0409 23:30:31.020926 4221 solver.cpp:218] Iteration 5520 (2.94391 iter/s, 4.07621s/12 iters), loss = 0.291211 I0409 23:30:31.020985 4221 solver.cpp:237] Train net output #0: loss = 0.291211 (* 1 = 0.291211 loss) I0409 23:30:31.020996 4221 sgd_solver.cpp:105] Iteration 5520, lr = 0.00335064 I0409 23:30:32.161286 4221 blocking_queue.cpp:49] Waiting for data I0409 23:30:35.832799 4221 solver.cpp:218] Iteration 5532 (2.49397 iter/s, 4.8116s/12 iters), loss = 0.36908 I0409 23:30:35.832854 4221 solver.cpp:237] Train net output #0: loss = 0.36908 (* 1 = 0.36908 loss) I0409 23:30:35.832864 4221 sgd_solver.cpp:105] Iteration 5532, lr = 0.00334268 I0409 23:30:40.713485 4221 solver.cpp:218] Iteration 5544 (2.45881 iter/s, 4.88042s/12 iters), loss = 0.602221 I0409 23:30:40.713536 4221 solver.cpp:237] Train net output #0: loss = 0.602221 (* 1 = 0.602221 loss) I0409 23:30:40.713549 4221 sgd_solver.cpp:105] Iteration 5544, lr = 0.00333475 I0409 23:30:45.682528 4221 solver.cpp:218] Iteration 5556 (2.41508 iter/s, 4.96877s/12 iters), loss = 0.4968 I0409 23:30:45.682655 4221 solver.cpp:237] Train net output #0: loss = 0.4968 (* 1 = 0.4968 loss) I0409 23:30:45.682670 4221 sgd_solver.cpp:105] Iteration 5556, lr = 0.00332683 I0409 23:30:48.239125 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:30:50.473804 4221 solver.cpp:218] Iteration 5568 (2.50473 iter/s, 4.79094s/12 iters), loss = 0.372444 I0409 23:30:50.473860 4221 solver.cpp:237] Train net output #0: loss = 0.372444 (* 1 = 0.372444 loss) I0409 23:30:50.473873 4221 sgd_solver.cpp:105] Iteration 5568, lr = 0.00331893 I0409 23:30:55.333864 4221 solver.cpp:218] Iteration 5580 (2.46924 iter/s, 4.8598s/12 iters), loss = 0.643093 I0409 23:30:55.333906 4221 solver.cpp:237] Train net output #0: loss = 0.643093 (* 1 = 0.643093 loss) I0409 23:30:55.333914 4221 sgd_solver.cpp:105] Iteration 5580, lr = 0.00331105 I0409 23:31:00.109287 4221 solver.cpp:218] Iteration 5592 (2.513 iter/s, 4.77517s/12 iters), loss = 0.352138 I0409 23:31:00.109344 4221 solver.cpp:237] Train net output #0: loss = 0.352138 (* 1 = 0.352138 loss) I0409 23:31:00.109356 4221 sgd_solver.cpp:105] Iteration 5592, lr = 0.00330319 I0409 23:31:04.945371 4221 solver.cpp:218] Iteration 5604 (2.48148 iter/s, 4.83582s/12 iters), loss = 0.252884 I0409 23:31:04.945411 4221 solver.cpp:237] Train net output #0: loss = 0.252884 (* 1 = 0.252884 loss) I0409 23:31:04.945420 4221 sgd_solver.cpp:105] Iteration 5604, lr = 0.00329535 I0409 23:31:06.922726 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_5610.caffemodel I0409 23:31:09.021607 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_5610.solverstate I0409 23:31:12.522927 4221 solver.cpp:330] Iteration 5610, Testing net (#0) I0409 23:31:12.522958 4221 net.cpp:676] Ignoring source layer train-data I0409 23:31:14.664773 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:31:16.872129 4221 solver.cpp:397] Test net output #0: accuracy = 0.387255 I0409 23:31:16.872239 4221 solver.cpp:397] Test net output #1: loss = 3.7244 (* 1 = 3.7244 loss) I0409 23:31:18.718422 4221 solver.cpp:218] Iteration 5616 (0.871306 iter/s, 13.7724s/12 iters), loss = 0.66027 I0409 23:31:18.718480 4221 solver.cpp:237] Train net output #0: loss = 0.66027 (* 1 = 0.66027 loss) I0409 23:31:18.718492 4221 sgd_solver.cpp:105] Iteration 5616, lr = 0.00328752 I0409 23:31:23.567684 4221 solver.cpp:218] Iteration 5628 (2.47474 iter/s, 4.84899s/12 iters), loss = 0.332484 I0409 23:31:23.567741 4221 solver.cpp:237] Train net output #0: loss = 0.332484 (* 1 = 0.332484 loss) I0409 23:31:23.567754 4221 sgd_solver.cpp:105] Iteration 5628, lr = 0.00327972 I0409 23:31:28.463816 4221 solver.cpp:218] Iteration 5640 (2.45105 iter/s, 4.89587s/12 iters), loss = 0.505121 I0409 23:31:28.463866 4221 solver.cpp:237] Train net output #0: loss = 0.505121 (* 1 = 0.505121 loss) I0409 23:31:28.463877 4221 sgd_solver.cpp:105] Iteration 5640, lr = 0.00327193 I0409 23:31:33.365720 4221 solver.cpp:218] Iteration 5652 (2.44816 iter/s, 4.90165s/12 iters), loss = 0.292218 I0409 23:31:33.365749 4221 solver.cpp:237] Train net output #0: loss = 0.292218 (* 1 = 0.292218 loss) I0409 23:31:33.365757 4221 sgd_solver.cpp:105] Iteration 5652, lr = 0.00326416 I0409 23:31:38.008112 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:31:38.169167 4221 solver.cpp:218] Iteration 5664 (2.49833 iter/s, 4.80321s/12 iters), loss = 0.398947 I0409 23:31:38.169212 4221 solver.cpp:237] Train net output #0: loss = 0.398947 (* 1 = 0.398947 loss) I0409 23:31:38.169222 4221 sgd_solver.cpp:105] Iteration 5664, lr = 0.00325641 I0409 23:31:43.005893 4221 solver.cpp:218] Iteration 5676 (2.48115 iter/s, 4.83647s/12 iters), loss = 0.318091 I0409 23:31:43.005939 4221 solver.cpp:237] Train net output #0: loss = 0.318091 (* 1 = 0.318091 loss) I0409 23:31:43.005947 4221 sgd_solver.cpp:105] Iteration 5676, lr = 0.00324868 I0409 23:31:47.816422 4221 solver.cpp:218] Iteration 5688 (2.49466 iter/s, 4.81027s/12 iters), loss = 0.245903 I0409 23:31:47.816519 4221 solver.cpp:237] Train net output #0: loss = 0.245903 (* 1 = 0.245903 loss) I0409 23:31:47.816529 4221 sgd_solver.cpp:105] Iteration 5688, lr = 0.00324097 I0409 23:31:52.624245 4221 solver.cpp:218] Iteration 5700 (2.49609 iter/s, 4.80752s/12 iters), loss = 0.388912 I0409 23:31:52.624300 4221 solver.cpp:237] Train net output #0: loss = 0.388912 (* 1 = 0.388912 loss) I0409 23:31:52.624310 4221 sgd_solver.cpp:105] Iteration 5700, lr = 0.00323328 I0409 23:31:57.029404 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_5712.caffemodel I0409 23:31:57.731523 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_5712.solverstate I0409 23:31:58.548178 4221 solver.cpp:330] Iteration 5712, Testing net (#0) I0409 23:31:58.548202 4221 net.cpp:676] Ignoring source layer train-data I0409 23:32:00.760257 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:32:03.001807 4221 solver.cpp:397] Test net output #0: accuracy = 0.39277 I0409 23:32:03.001852 4221 solver.cpp:397] Test net output #1: loss = 3.62481 (* 1 = 3.62481 loss) I0409 23:32:03.084893 4221 solver.cpp:218] Iteration 5712 (1.14721 iter/s, 10.4602s/12 iters), loss = 0.274273 I0409 23:32:03.084933 4221 solver.cpp:237] Train net output #0: loss = 0.274273 (* 1 = 0.274273 loss) I0409 23:32:03.084941 4221 sgd_solver.cpp:105] Iteration 5712, lr = 0.0032256 I0409 23:32:07.150684 4221 solver.cpp:218] Iteration 5724 (2.95161 iter/s, 4.06557s/12 iters), loss = 0.362037 I0409 23:32:07.150733 4221 solver.cpp:237] Train net output #0: loss = 0.362037 (* 1 = 0.362037 loss) I0409 23:32:07.150744 4221 sgd_solver.cpp:105] Iteration 5724, lr = 0.00321794 I0409 23:32:12.182215 4221 solver.cpp:218] Iteration 5736 (2.38509 iter/s, 5.03126s/12 iters), loss = 0.464331 I0409 23:32:12.182260 4221 solver.cpp:237] Train net output #0: loss = 0.464331 (* 1 = 0.464331 loss) I0409 23:32:12.182269 4221 sgd_solver.cpp:105] Iteration 5736, lr = 0.0032103 I0409 23:32:16.966310 4221 solver.cpp:218] Iteration 5748 (2.50845 iter/s, 4.78383s/12 iters), loss = 0.2977 I0409 23:32:16.966357 4221 solver.cpp:237] Train net output #0: loss = 0.2977 (* 1 = 0.2977 loss) I0409 23:32:16.966367 4221 sgd_solver.cpp:105] Iteration 5748, lr = 0.00320268 I0409 23:32:21.793808 4221 solver.cpp:218] Iteration 5760 (2.48589 iter/s, 4.82724s/12 iters), loss = 0.228483 I0409 23:32:21.793939 4221 solver.cpp:237] Train net output #0: loss = 0.228483 (* 1 = 0.228483 loss) I0409 23:32:21.793967 4221 sgd_solver.cpp:105] Iteration 5760, lr = 0.00319508 I0409 23:32:23.701416 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:32:26.610076 4221 solver.cpp:218] Iteration 5772 (2.49173 iter/s, 4.81593s/12 iters), loss = 0.449365 I0409 23:32:26.610131 4221 solver.cpp:237] Train net output #0: loss = 0.449365 (* 1 = 0.449365 loss) I0409 23:32:26.610143 4221 sgd_solver.cpp:105] Iteration 5772, lr = 0.00318749 I0409 23:32:31.476697 4221 solver.cpp:218] Iteration 5784 (2.46591 iter/s, 4.86636s/12 iters), loss = 0.335582 I0409 23:32:31.476747 4221 solver.cpp:237] Train net output #0: loss = 0.335582 (* 1 = 0.335582 loss) I0409 23:32:31.476759 4221 sgd_solver.cpp:105] Iteration 5784, lr = 0.00317992 I0409 23:32:36.351364 4221 solver.cpp:218] Iteration 5796 (2.46184 iter/s, 4.87441s/12 iters), loss = 0.314929 I0409 23:32:36.351413 4221 solver.cpp:237] Train net output #0: loss = 0.314929 (* 1 = 0.314929 loss) I0409 23:32:36.351423 4221 sgd_solver.cpp:105] Iteration 5796, lr = 0.00317237 I0409 23:32:41.190212 4221 solver.cpp:218] Iteration 5808 (2.48007 iter/s, 4.83858s/12 iters), loss = 0.418228 I0409 23:32:41.190272 4221 solver.cpp:237] Train net output #0: loss = 0.418228 (* 1 = 0.418228 loss) I0409 23:32:41.190284 4221 sgd_solver.cpp:105] Iteration 5808, lr = 0.00316484 I0409 23:32:43.162320 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_5814.caffemodel I0409 23:32:43.885344 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_5814.solverstate I0409 23:32:44.630865 4221 solver.cpp:330] Iteration 5814, Testing net (#0) I0409 23:32:44.630897 4221 net.cpp:676] Ignoring source layer train-data I0409 23:32:46.827435 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:32:49.193409 4221 solver.cpp:397] Test net output #0: accuracy = 0.396446 I0409 23:32:49.193439 4221 solver.cpp:397] Test net output #1: loss = 3.69405 (* 1 = 3.69405 loss) I0409 23:32:51.029417 4221 solver.cpp:218] Iteration 5820 (1.21967 iter/s, 9.83873s/12 iters), loss = 0.421708 I0409 23:32:51.029471 4221 solver.cpp:237] Train net output #0: loss = 0.421708 (* 1 = 0.421708 loss) I0409 23:32:51.029484 4221 sgd_solver.cpp:105] Iteration 5820, lr = 0.00315733 I0409 23:32:55.948493 4221 solver.cpp:218] Iteration 5832 (2.43961 iter/s, 4.91881s/12 iters), loss = 0.532557 I0409 23:32:55.948563 4221 solver.cpp:237] Train net output #0: loss = 0.532557 (* 1 = 0.532557 loss) I0409 23:32:55.948572 4221 sgd_solver.cpp:105] Iteration 5832, lr = 0.00314983 I0409 23:33:00.815075 4221 solver.cpp:218] Iteration 5844 (2.46594 iter/s, 4.8663s/12 iters), loss = 0.530252 I0409 23:33:00.815133 4221 solver.cpp:237] Train net output #0: loss = 0.530252 (* 1 = 0.530252 loss) I0409 23:33:00.815145 4221 sgd_solver.cpp:105] Iteration 5844, lr = 0.00314235 I0409 23:33:05.642432 4221 solver.cpp:218] Iteration 5856 (2.48597 iter/s, 4.82709s/12 iters), loss = 0.340324 I0409 23:33:05.642482 4221 solver.cpp:237] Train net output #0: loss = 0.340324 (* 1 = 0.340324 loss) I0409 23:33:05.642493 4221 sgd_solver.cpp:105] Iteration 5856, lr = 0.00313489 I0409 23:33:09.652380 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:33:10.430893 4221 solver.cpp:218] Iteration 5868 (2.50616 iter/s, 4.7882s/12 iters), loss = 0.351903 I0409 23:33:10.430945 4221 solver.cpp:237] Train net output #0: loss = 0.351903 (* 1 = 0.351903 loss) I0409 23:33:10.430958 4221 sgd_solver.cpp:105] Iteration 5868, lr = 0.00312745 I0409 23:33:15.325392 4221 solver.cpp:218] Iteration 5880 (2.45187 iter/s, 4.89423s/12 iters), loss = 0.373365 I0409 23:33:15.325440 4221 solver.cpp:237] Train net output #0: loss = 0.373365 (* 1 = 0.373365 loss) I0409 23:33:15.325450 4221 sgd_solver.cpp:105] Iteration 5880, lr = 0.00312002 I0409 23:33:20.101570 4221 solver.cpp:218] Iteration 5892 (2.5126 iter/s, 4.77592s/12 iters), loss = 0.366663 I0409 23:33:20.101629 4221 solver.cpp:237] Train net output #0: loss = 0.366663 (* 1 = 0.366663 loss) I0409 23:33:20.101640 4221 sgd_solver.cpp:105] Iteration 5892, lr = 0.00311262 I0409 23:33:24.899307 4221 solver.cpp:218] Iteration 5904 (2.50132 iter/s, 4.79748s/12 iters), loss = 0.635323 I0409 23:33:24.899348 4221 solver.cpp:237] Train net output #0: loss = 0.635323 (* 1 = 0.635323 loss) I0409 23:33:24.899354 4221 sgd_solver.cpp:105] Iteration 5904, lr = 0.00310523 I0409 23:33:29.282629 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_5916.caffemodel I0409 23:33:29.974767 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_5916.solverstate I0409 23:33:30.472808 4221 solver.cpp:330] Iteration 5916, Testing net (#0) I0409 23:33:30.472831 4221 net.cpp:676] Ignoring source layer train-data I0409 23:33:32.601187 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:33:35.045428 4221 solver.cpp:397] Test net output #0: accuracy = 0.406863 I0409 23:33:35.045478 4221 solver.cpp:397] Test net output #1: loss = 3.58028 (* 1 = 3.58028 loss) I0409 23:33:35.128791 4221 solver.cpp:218] Iteration 5916 (1.17313 iter/s, 10.229s/12 iters), loss = 0.337232 I0409 23:33:35.128841 4221 solver.cpp:237] Train net output #0: loss = 0.337232 (* 1 = 0.337232 loss) I0409 23:33:35.128852 4221 sgd_solver.cpp:105] Iteration 5916, lr = 0.00309785 I0409 23:33:39.308878 4221 solver.cpp:218] Iteration 5928 (2.87091 iter/s, 4.17986s/12 iters), loss = 0.271049 I0409 23:33:39.308919 4221 solver.cpp:237] Train net output #0: loss = 0.271049 (* 1 = 0.271049 loss) I0409 23:33:39.308929 4221 sgd_solver.cpp:105] Iteration 5928, lr = 0.0030905 I0409 23:33:44.112438 4221 solver.cpp:218] Iteration 5940 (2.49828 iter/s, 4.80331s/12 iters), loss = 0.317867 I0409 23:33:44.112491 4221 solver.cpp:237] Train net output #0: loss = 0.317867 (* 1 = 0.317867 loss) I0409 23:33:44.112504 4221 sgd_solver.cpp:105] Iteration 5940, lr = 0.00308316 I0409 23:33:48.910712 4221 solver.cpp:218] Iteration 5952 (2.50104 iter/s, 4.79801s/12 iters), loss = 0.325423 I0409 23:33:48.910773 4221 solver.cpp:237] Train net output #0: loss = 0.325423 (* 1 = 0.325423 loss) I0409 23:33:48.910786 4221 sgd_solver.cpp:105] Iteration 5952, lr = 0.00307584 I0409 23:33:53.744352 4221 solver.cpp:218] Iteration 5964 (2.48274 iter/s, 4.83337s/12 iters), loss = 0.454795 I0409 23:33:53.744403 4221 solver.cpp:237] Train net output #0: loss = 0.454795 (* 1 = 0.454795 loss) I0409 23:33:53.744415 4221 sgd_solver.cpp:105] Iteration 5964, lr = 0.00306854 I0409 23:33:55.017706 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:33:58.542318 4221 solver.cpp:218] Iteration 5976 (2.5012 iter/s, 4.79771s/12 iters), loss = 0.441786 I0409 23:33:58.542371 4221 solver.cpp:237] Train net output #0: loss = 0.441786 (* 1 = 0.441786 loss) I0409 23:33:58.542383 4221 sgd_solver.cpp:105] Iteration 5976, lr = 0.00306125 I0409 23:34:03.343873 4221 solver.cpp:218] Iteration 5988 (2.49933 iter/s, 4.80129s/12 iters), loss = 0.341505 I0409 23:34:03.344012 4221 solver.cpp:237] Train net output #0: loss = 0.341505 (* 1 = 0.341505 loss) I0409 23:34:03.344022 4221 sgd_solver.cpp:105] Iteration 5988, lr = 0.00305398 I0409 23:34:08.145918 4221 solver.cpp:218] Iteration 6000 (2.49912 iter/s, 4.8017s/12 iters), loss = 0.524947 I0409 23:34:08.145988 4221 solver.cpp:237] Train net output #0: loss = 0.524947 (* 1 = 0.524947 loss) I0409 23:34:08.145999 4221 sgd_solver.cpp:105] Iteration 6000, lr = 0.00304673 I0409 23:34:12.959229 4221 solver.cpp:218] Iteration 6012 (2.49323 iter/s, 4.81304s/12 iters), loss = 0.478789 I0409 23:34:12.959281 4221 solver.cpp:237] Train net output #0: loss = 0.478789 (* 1 = 0.478789 loss) I0409 23:34:12.959292 4221 sgd_solver.cpp:105] Iteration 6012, lr = 0.0030395 I0409 23:34:14.915972 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_6018.caffemodel I0409 23:34:15.620821 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_6018.solverstate I0409 23:34:16.121428 4221 solver.cpp:330] Iteration 6018, Testing net (#0) I0409 23:34:16.121455 4221 net.cpp:676] Ignoring source layer train-data I0409 23:34:18.124145 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:34:20.521178 4221 solver.cpp:397] Test net output #0: accuracy = 0.404412 I0409 23:34:20.521219 4221 solver.cpp:397] Test net output #1: loss = 3.5468 (* 1 = 3.5468 loss) I0409 23:34:22.263511 4221 solver.cpp:218] Iteration 6024 (1.28979 iter/s, 9.30384s/12 iters), loss = 0.326363 I0409 23:34:22.263569 4221 solver.cpp:237] Train net output #0: loss = 0.326363 (* 1 = 0.326363 loss) I0409 23:34:22.263581 4221 sgd_solver.cpp:105] Iteration 6024, lr = 0.00303228 I0409 23:34:27.029472 4221 solver.cpp:218] Iteration 6036 (2.518 iter/s, 4.7657s/12 iters), loss = 0.284308 I0409 23:34:27.029536 4221 solver.cpp:237] Train net output #0: loss = 0.284308 (* 1 = 0.284308 loss) I0409 23:34:27.029548 4221 sgd_solver.cpp:105] Iteration 6036, lr = 0.00302508 I0409 23:34:31.813122 4221 solver.cpp:218] Iteration 6048 (2.50869 iter/s, 4.78338s/12 iters), loss = 0.326483 I0409 23:34:31.813186 4221 solver.cpp:237] Train net output #0: loss = 0.326483 (* 1 = 0.326483 loss) I0409 23:34:31.813199 4221 sgd_solver.cpp:105] Iteration 6048, lr = 0.0030179 I0409 23:34:36.603488 4221 solver.cpp:218] Iteration 6060 (2.50517 iter/s, 4.79009s/12 iters), loss = 0.257412 I0409 23:34:36.603621 4221 solver.cpp:237] Train net output #0: loss = 0.257412 (* 1 = 0.257412 loss) I0409 23:34:36.603634 4221 sgd_solver.cpp:105] Iteration 6060, lr = 0.00301074 I0409 23:34:39.900995 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:34:41.377293 4221 solver.cpp:218] Iteration 6072 (2.5139 iter/s, 4.77346s/12 iters), loss = 0.43396 I0409 23:34:41.377355 4221 solver.cpp:237] Train net output #0: loss = 0.43396 (* 1 = 0.43396 loss) I0409 23:34:41.377369 4221 sgd_solver.cpp:105] Iteration 6072, lr = 0.00300359 I0409 23:34:46.161672 4221 solver.cpp:218] Iteration 6084 (2.50831 iter/s, 4.78411s/12 iters), loss = 0.333793 I0409 23:34:46.161733 4221 solver.cpp:237] Train net output #0: loss = 0.333793 (* 1 = 0.333793 loss) I0409 23:34:46.161746 4221 sgd_solver.cpp:105] Iteration 6084, lr = 0.00299646 I0409 23:34:51.036945 4221 solver.cpp:218] Iteration 6096 (2.46154 iter/s, 4.875s/12 iters), loss = 0.326601 I0409 23:34:51.036993 4221 solver.cpp:237] Train net output #0: loss = 0.326601 (* 1 = 0.326601 loss) I0409 23:34:51.037003 4221 sgd_solver.cpp:105] Iteration 6096, lr = 0.00298934 I0409 23:34:55.868232 4221 solver.cpp:218] Iteration 6108 (2.48394 iter/s, 4.83103s/12 iters), loss = 0.313291 I0409 23:34:55.868276 4221 solver.cpp:237] Train net output #0: loss = 0.313291 (* 1 = 0.313291 loss) I0409 23:34:55.868285 4221 sgd_solver.cpp:105] Iteration 6108, lr = 0.00298225 I0409 23:35:00.242182 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_6120.caffemodel I0409 23:35:00.892774 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_6120.solverstate I0409 23:35:01.383276 4221 solver.cpp:330] Iteration 6120, Testing net (#0) I0409 23:35:01.383306 4221 net.cpp:676] Ignoring source layer train-data I0409 23:35:03.413693 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:35:05.812397 4221 solver.cpp:397] Test net output #0: accuracy = 0.404412 I0409 23:35:05.812448 4221 solver.cpp:397] Test net output #1: loss = 3.51903 (* 1 = 3.51903 loss) I0409 23:35:05.895676 4221 solver.cpp:218] Iteration 6120 (1.19677 iter/s, 10.027s/12 iters), loss = 0.461135 I0409 23:35:05.895731 4221 solver.cpp:237] Train net output #0: loss = 0.461135 (* 1 = 0.461135 loss) I0409 23:35:05.895742 4221 sgd_solver.cpp:105] Iteration 6120, lr = 0.00297517 I0409 23:35:10.140980 4221 solver.cpp:218] Iteration 6132 (2.82681 iter/s, 4.24506s/12 iters), loss = 0.430738 I0409 23:35:10.141124 4221 solver.cpp:237] Train net output #0: loss = 0.430738 (* 1 = 0.430738 loss) I0409 23:35:10.141137 4221 sgd_solver.cpp:105] Iteration 6132, lr = 0.0029681 I0409 23:35:14.938141 4221 solver.cpp:218] Iteration 6144 (2.50166 iter/s, 4.79682s/12 iters), loss = 0.302292 I0409 23:35:14.938189 4221 solver.cpp:237] Train net output #0: loss = 0.302292 (* 1 = 0.302292 loss) I0409 23:35:14.938199 4221 sgd_solver.cpp:105] Iteration 6144, lr = 0.00296105 I0409 23:35:19.868307 4221 solver.cpp:218] Iteration 6156 (2.43413 iter/s, 4.9299s/12 iters), loss = 0.313818 I0409 23:35:19.868352 4221 solver.cpp:237] Train net output #0: loss = 0.313818 (* 1 = 0.313818 loss) I0409 23:35:19.868361 4221 sgd_solver.cpp:105] Iteration 6156, lr = 0.00295402 I0409 23:35:24.806658 4221 solver.cpp:218] Iteration 6168 (2.43009 iter/s, 4.93809s/12 iters), loss = 0.475929 I0409 23:35:24.806706 4221 solver.cpp:237] Train net output #0: loss = 0.475929 (* 1 = 0.475929 loss) I0409 23:35:24.806715 4221 sgd_solver.cpp:105] Iteration 6168, lr = 0.00294701 I0409 23:35:25.400918 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:35:29.643183 4221 solver.cpp:218] Iteration 6180 (2.48126 iter/s, 4.83626s/12 iters), loss = 0.542522 I0409 23:35:29.643241 4221 solver.cpp:237] Train net output #0: loss = 0.542522 (* 1 = 0.542522 loss) I0409 23:35:29.643254 4221 sgd_solver.cpp:105] Iteration 6180, lr = 0.00294001 I0409 23:35:34.484983 4221 solver.cpp:218] Iteration 6192 (2.47855 iter/s, 4.84154s/12 iters), loss = 0.294037 I0409 23:35:34.485033 4221 solver.cpp:237] Train net output #0: loss = 0.294038 (* 1 = 0.294038 loss) I0409 23:35:34.485044 4221 sgd_solver.cpp:105] Iteration 6192, lr = 0.00293303 I0409 23:35:39.301398 4221 solver.cpp:218] Iteration 6204 (2.49162 iter/s, 4.81615s/12 iters), loss = 0.273314 I0409 23:35:39.301455 4221 solver.cpp:237] Train net output #0: loss = 0.273314 (* 1 = 0.273314 loss) I0409 23:35:39.301466 4221 sgd_solver.cpp:105] Iteration 6204, lr = 0.00292607 I0409 23:35:44.132448 4221 solver.cpp:218] Iteration 6216 (2.48407 iter/s, 4.83078s/12 iters), loss = 0.311221 I0409 23:35:44.132568 4221 solver.cpp:237] Train net output #0: loss = 0.311221 (* 1 = 0.311221 loss) I0409 23:35:44.132583 4221 sgd_solver.cpp:105] Iteration 6216, lr = 0.00291912 I0409 23:35:46.115972 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_6222.caffemodel I0409 23:35:46.780268 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_6222.solverstate I0409 23:35:47.262732 4221 solver.cpp:330] Iteration 6222, Testing net (#0) I0409 23:35:47.262756 4221 net.cpp:676] Ignoring source layer train-data I0409 23:35:49.196529 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:35:50.197048 4221 blocking_queue.cpp:49] Waiting for data I0409 23:35:51.640022 4221 solver.cpp:397] Test net output #0: accuracy = 0.417892 I0409 23:35:51.640069 4221 solver.cpp:397] Test net output #1: loss = 3.60369 (* 1 = 3.60369 loss) I0409 23:35:53.580173 4221 solver.cpp:218] Iteration 6228 (1.27022 iter/s, 9.44721s/12 iters), loss = 0.185326 I0409 23:35:53.580227 4221 solver.cpp:237] Train net output #0: loss = 0.185326 (* 1 = 0.185326 loss) I0409 23:35:53.580240 4221 sgd_solver.cpp:105] Iteration 6228, lr = 0.00291219 I0409 23:35:58.391021 4221 solver.cpp:218] Iteration 6240 (2.4945 iter/s, 4.81058s/12 iters), loss = 0.49031 I0409 23:35:58.391083 4221 solver.cpp:237] Train net output #0: loss = 0.49031 (* 1 = 0.49031 loss) I0409 23:35:58.391094 4221 sgd_solver.cpp:105] Iteration 6240, lr = 0.00290528 I0409 23:36:03.169062 4221 solver.cpp:218] Iteration 6252 (2.51163 iter/s, 4.77778s/12 iters), loss = 0.281358 I0409 23:36:03.169102 4221 solver.cpp:237] Train net output #0: loss = 0.281358 (* 1 = 0.281358 loss) I0409 23:36:03.169111 4221 sgd_solver.cpp:105] Iteration 6252, lr = 0.00289838 I0409 23:36:07.968698 4221 solver.cpp:218] Iteration 6264 (2.50032 iter/s, 4.79938s/12 iters), loss = 0.229527 I0409 23:36:07.968750 4221 solver.cpp:237] Train net output #0: loss = 0.229527 (* 1 = 0.229527 loss) I0409 23:36:07.968760 4221 sgd_solver.cpp:105] Iteration 6264, lr = 0.0028915 I0409 23:36:10.602983 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:36:12.956790 4221 solver.cpp:218] Iteration 6276 (2.40586 iter/s, 4.98782s/12 iters), loss = 0.265711 I0409 23:36:12.956837 4221 solver.cpp:237] Train net output #0: loss = 0.265711 (* 1 = 0.265711 loss) I0409 23:36:12.956848 4221 sgd_solver.cpp:105] Iteration 6276, lr = 0.00288463 I0409 23:36:18.011083 4221 solver.cpp:218] Iteration 6288 (2.37434 iter/s, 5.05403s/12 iters), loss = 0.32247 I0409 23:36:18.011211 4221 solver.cpp:237] Train net output #0: loss = 0.32247 (* 1 = 0.32247 loss) I0409 23:36:18.011221 4221 sgd_solver.cpp:105] Iteration 6288, lr = 0.00287779 I0409 23:36:22.834101 4221 solver.cpp:218] Iteration 6300 (2.48824 iter/s, 4.82269s/12 iters), loss = 0.411463 I0409 23:36:22.834146 4221 solver.cpp:237] Train net output #0: loss = 0.411463 (* 1 = 0.411463 loss) I0409 23:36:22.834156 4221 sgd_solver.cpp:105] Iteration 6300, lr = 0.00287095 I0409 23:36:27.708127 4221 solver.cpp:218] Iteration 6312 (2.46216 iter/s, 4.87377s/12 iters), loss = 0.248066 I0409 23:36:27.708187 4221 solver.cpp:237] Train net output #0: loss = 0.248066 (* 1 = 0.248066 loss) I0409 23:36:27.708199 4221 sgd_solver.cpp:105] Iteration 6312, lr = 0.00286414 I0409 23:36:32.097193 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_6324.caffemodel I0409 23:36:32.812721 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_6324.solverstate I0409 23:36:33.314612 4221 solver.cpp:330] Iteration 6324, Testing net (#0) I0409 23:36:33.314641 4221 net.cpp:676] Ignoring source layer train-data I0409 23:36:35.226754 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:36:37.709126 4221 solver.cpp:397] Test net output #0: accuracy = 0.409314 I0409 23:36:37.709168 4221 solver.cpp:397] Test net output #1: loss = 3.48419 (* 1 = 3.48419 loss) I0409 23:36:37.792285 4221 solver.cpp:218] Iteration 6324 (1.19004 iter/s, 10.0837s/12 iters), loss = 0.224279 I0409 23:36:37.792335 4221 solver.cpp:237] Train net output #0: loss = 0.224279 (* 1 = 0.224279 loss) I0409 23:36:37.792346 4221 sgd_solver.cpp:105] Iteration 6324, lr = 0.00285734 I0409 23:36:41.861750 4221 solver.cpp:218] Iteration 6336 (2.94896 iter/s, 4.06923s/12 iters), loss = 0.242696 I0409 23:36:41.861804 4221 solver.cpp:237] Train net output #0: loss = 0.242696 (* 1 = 0.242696 loss) I0409 23:36:41.861815 4221 sgd_solver.cpp:105] Iteration 6336, lr = 0.00285055 I0409 23:36:46.673632 4221 solver.cpp:218] Iteration 6348 (2.49396 iter/s, 4.81162s/12 iters), loss = 0.304898 I0409 23:36:46.673684 4221 solver.cpp:237] Train net output #0: loss = 0.304898 (* 1 = 0.304898 loss) I0409 23:36:46.673696 4221 sgd_solver.cpp:105] Iteration 6348, lr = 0.00284379 I0409 23:36:51.555002 4221 solver.cpp:218] Iteration 6360 (2.45846 iter/s, 4.8811s/12 iters), loss = 0.282758 I0409 23:36:51.555143 4221 solver.cpp:237] Train net output #0: loss = 0.282758 (* 1 = 0.282758 loss) I0409 23:36:51.555158 4221 sgd_solver.cpp:105] Iteration 6360, lr = 0.00283703 I0409 23:36:56.239869 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:36:56.369742 4221 solver.cpp:218] Iteration 6372 (2.49253 iter/s, 4.81439s/12 iters), loss = 0.192877 I0409 23:36:56.369799 4221 solver.cpp:237] Train net output #0: loss = 0.192877 (* 1 = 0.192877 loss) I0409 23:36:56.369812 4221 sgd_solver.cpp:105] Iteration 6372, lr = 0.0028303 I0409 23:37:01.223701 4221 solver.cpp:218] Iteration 6384 (2.47234 iter/s, 4.8537s/12 iters), loss = 0.49388 I0409 23:37:01.223742 4221 solver.cpp:237] Train net output #0: loss = 0.49388 (* 1 = 0.49388 loss) I0409 23:37:01.223750 4221 sgd_solver.cpp:105] Iteration 6384, lr = 0.00282358 I0409 23:37:06.426589 4221 solver.cpp:218] Iteration 6396 (2.30653 iter/s, 5.20262s/12 iters), loss = 0.417026 I0409 23:37:06.426635 4221 solver.cpp:237] Train net output #0: loss = 0.417026 (* 1 = 0.417026 loss) I0409 23:37:06.426646 4221 sgd_solver.cpp:105] Iteration 6396, lr = 0.00281687 I0409 23:37:11.299844 4221 solver.cpp:218] Iteration 6408 (2.46255 iter/s, 4.87299s/12 iters), loss = 0.266814 I0409 23:37:11.299901 4221 solver.cpp:237] Train net output #0: loss = 0.266814 (* 1 = 0.266814 loss) I0409 23:37:11.299913 4221 sgd_solver.cpp:105] Iteration 6408, lr = 0.00281019 I0409 23:37:16.165364 4221 solver.cpp:218] Iteration 6420 (2.46647 iter/s, 4.86526s/12 iters), loss = 0.236475 I0409 23:37:16.165411 4221 solver.cpp:237] Train net output #0: loss = 0.236475 (* 1 = 0.236475 loss) I0409 23:37:16.165422 4221 sgd_solver.cpp:105] Iteration 6420, lr = 0.00280351 I0409 23:37:18.162205 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_6426.caffemodel I0409 23:37:18.861544 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_6426.solverstate I0409 23:37:19.363667 4221 solver.cpp:330] Iteration 6426, Testing net (#0) I0409 23:37:19.363698 4221 net.cpp:676] Ignoring source layer train-data I0409 23:37:21.207577 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:37:23.728394 4221 solver.cpp:397] Test net output #0: accuracy = 0.419118 I0409 23:37:23.728494 4221 solver.cpp:397] Test net output #1: loss = 3.62556 (* 1 = 3.62556 loss) I0409 23:37:25.652875 4221 solver.cpp:218] Iteration 6432 (1.26488 iter/s, 9.48707s/12 iters), loss = 0.270895 I0409 23:37:25.652935 4221 solver.cpp:237] Train net output #0: loss = 0.270895 (* 1 = 0.270895 loss) I0409 23:37:25.652948 4221 sgd_solver.cpp:105] Iteration 6432, lr = 0.00279686 I0409 23:37:30.515007 4221 solver.cpp:218] Iteration 6444 (2.46819 iter/s, 4.86187s/12 iters), loss = 0.245724 I0409 23:37:30.515053 4221 solver.cpp:237] Train net output #0: loss = 0.245724 (* 1 = 0.245724 loss) I0409 23:37:30.515061 4221 sgd_solver.cpp:105] Iteration 6444, lr = 0.00279022 I0409 23:37:35.357844 4221 solver.cpp:218] Iteration 6456 (2.47802 iter/s, 4.84258s/12 iters), loss = 0.138586 I0409 23:37:35.357892 4221 solver.cpp:237] Train net output #0: loss = 0.138586 (* 1 = 0.138586 loss) I0409 23:37:35.357901 4221 sgd_solver.cpp:105] Iteration 6456, lr = 0.00278359 I0409 23:37:40.205785 4221 solver.cpp:218] Iteration 6468 (2.47541 iter/s, 4.84768s/12 iters), loss = 0.15915 I0409 23:37:40.205832 4221 solver.cpp:237] Train net output #0: loss = 0.15915 (* 1 = 0.15915 loss) I0409 23:37:40.205842 4221 sgd_solver.cpp:105] Iteration 6468, lr = 0.00277698 I0409 23:37:42.119235 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:37:45.086304 4221 solver.cpp:218] Iteration 6480 (2.45888 iter/s, 4.88027s/12 iters), loss = 0.348911 I0409 23:37:45.086342 4221 solver.cpp:237] Train net output #0: loss = 0.348911 (* 1 = 0.348911 loss) I0409 23:37:45.086351 4221 sgd_solver.cpp:105] Iteration 6480, lr = 0.00277039 I0409 23:37:50.060740 4221 solver.cpp:218] Iteration 6492 (2.41246 iter/s, 4.97418s/12 iters), loss = 0.261966 I0409 23:37:50.060787 4221 solver.cpp:237] Train net output #0: loss = 0.261966 (* 1 = 0.261966 loss) I0409 23:37:50.060796 4221 sgd_solver.cpp:105] Iteration 6492, lr = 0.00276381 I0409 23:37:54.869204 4221 solver.cpp:218] Iteration 6504 (2.49573 iter/s, 4.80821s/12 iters), loss = 0.283436 I0409 23:37:54.869356 4221 solver.cpp:237] Train net output #0: loss = 0.283436 (* 1 = 0.283436 loss) I0409 23:37:54.869369 4221 sgd_solver.cpp:105] Iteration 6504, lr = 0.00275725 I0409 23:37:59.667753 4221 solver.cpp:218] Iteration 6516 (2.50094 iter/s, 4.79819s/12 iters), loss = 0.238145 I0409 23:37:59.667809 4221 solver.cpp:237] Train net output #0: loss = 0.238145 (* 1 = 0.238145 loss) I0409 23:37:59.667820 4221 sgd_solver.cpp:105] Iteration 6516, lr = 0.00275071 I0409 23:38:04.062559 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_6528.caffemodel I0409 23:38:05.943094 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_6528.solverstate I0409 23:38:06.646884 4221 solver.cpp:330] Iteration 6528, Testing net (#0) I0409 23:38:06.646906 4221 net.cpp:676] Ignoring source layer train-data I0409 23:38:08.681555 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:38:11.243002 4221 solver.cpp:397] Test net output #0: accuracy = 0.420956 I0409 23:38:11.243050 4221 solver.cpp:397] Test net output #1: loss = 3.69056 (* 1 = 3.69056 loss) I0409 23:38:11.326478 4221 solver.cpp:218] Iteration 6528 (1.02932 iter/s, 11.6582s/12 iters), loss = 0.146658 I0409 23:38:11.326550 4221 solver.cpp:237] Train net output #0: loss = 0.146658 (* 1 = 0.146658 loss) I0409 23:38:11.326566 4221 sgd_solver.cpp:105] Iteration 6528, lr = 0.00274418 I0409 23:38:15.510118 4221 solver.cpp:218] Iteration 6540 (2.86849 iter/s, 4.18339s/12 iters), loss = 0.290943 I0409 23:38:15.510180 4221 solver.cpp:237] Train net output #0: loss = 0.290943 (* 1 = 0.290943 loss) I0409 23:38:15.510193 4221 sgd_solver.cpp:105] Iteration 6540, lr = 0.00273766 I0409 23:38:20.325500 4221 solver.cpp:218] Iteration 6552 (2.49215 iter/s, 4.81511s/12 iters), loss = 0.223807 I0409 23:38:20.325546 4221 solver.cpp:237] Train net output #0: loss = 0.223807 (* 1 = 0.223807 loss) I0409 23:38:20.325554 4221 sgd_solver.cpp:105] Iteration 6552, lr = 0.00273116 I0409 23:38:25.207254 4221 solver.cpp:218] Iteration 6564 (2.45826 iter/s, 4.88149s/12 iters), loss = 0.193312 I0409 23:38:25.207350 4221 solver.cpp:237] Train net output #0: loss = 0.193312 (* 1 = 0.193312 loss) I0409 23:38:25.207360 4221 sgd_solver.cpp:105] Iteration 6564, lr = 0.00272468 I0409 23:38:29.287674 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:38:30.039460 4221 solver.cpp:218] Iteration 6576 (2.48349 iter/s, 4.83191s/12 iters), loss = 0.228068 I0409 23:38:30.039502 4221 solver.cpp:237] Train net output #0: loss = 0.228068 (* 1 = 0.228068 loss) I0409 23:38:30.039511 4221 sgd_solver.cpp:105] Iteration 6576, lr = 0.00271821 I0409 23:38:34.888108 4221 solver.cpp:218] Iteration 6588 (2.47505 iter/s, 4.84839s/12 iters), loss = 0.309051 I0409 23:38:34.888166 4221 solver.cpp:237] Train net output #0: loss = 0.309051 (* 1 = 0.309051 loss) I0409 23:38:34.888212 4221 sgd_solver.cpp:105] Iteration 6588, lr = 0.00271175 I0409 23:38:39.670137 4221 solver.cpp:218] Iteration 6600 (2.50953 iter/s, 4.78176s/12 iters), loss = 0.272748 I0409 23:38:39.670193 4221 solver.cpp:237] Train net output #0: loss = 0.272748 (* 1 = 0.272748 loss) I0409 23:38:39.670204 4221 sgd_solver.cpp:105] Iteration 6600, lr = 0.00270532 I0409 23:38:44.481144 4221 solver.cpp:218] Iteration 6612 (2.49442 iter/s, 4.81074s/12 iters), loss = 0.280012 I0409 23:38:44.481206 4221 solver.cpp:237] Train net output #0: loss = 0.280012 (* 1 = 0.280012 loss) I0409 23:38:44.481220 4221 sgd_solver.cpp:105] Iteration 6612, lr = 0.00269889 I0409 23:38:49.290405 4221 solver.cpp:218] Iteration 6624 (2.49533 iter/s, 4.80899s/12 iters), loss = 0.226441 I0409 23:38:49.290458 4221 solver.cpp:237] Train net output #0: loss = 0.226441 (* 1 = 0.226441 loss) I0409 23:38:49.290468 4221 sgd_solver.cpp:105] Iteration 6624, lr = 0.00269248 I0409 23:38:51.259810 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_6630.caffemodel I0409 23:38:51.912070 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_6630.solverstate I0409 23:38:52.390146 4221 solver.cpp:330] Iteration 6630, Testing net (#0) I0409 23:38:52.390170 4221 net.cpp:676] Ignoring source layer train-data I0409 23:38:54.217698 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:38:56.809201 4221 solver.cpp:397] Test net output #0: accuracy = 0.405024 I0409 23:38:56.809361 4221 solver.cpp:397] Test net output #1: loss = 3.61159 (* 1 = 3.61159 loss) I0409 23:38:58.726001 4221 solver.cpp:218] Iteration 6636 (1.27185 iter/s, 9.43511s/12 iters), loss = 0.258938 I0409 23:38:58.726056 4221 solver.cpp:237] Train net output #0: loss = 0.258938 (* 1 = 0.258938 loss) I0409 23:38:58.726068 4221 sgd_solver.cpp:105] Iteration 6636, lr = 0.00268609 I0409 23:39:03.589568 4221 solver.cpp:218] Iteration 6648 (2.46746 iter/s, 4.8633s/12 iters), loss = 0.16392 I0409 23:39:03.589617 4221 solver.cpp:237] Train net output #0: loss = 0.16392 (* 1 = 0.16392 loss) I0409 23:39:03.589625 4221 sgd_solver.cpp:105] Iteration 6648, lr = 0.00267971 I0409 23:39:08.423852 4221 solver.cpp:218] Iteration 6660 (2.4824 iter/s, 4.83403s/12 iters), loss = 0.263547 I0409 23:39:08.423904 4221 solver.cpp:237] Train net output #0: loss = 0.263547 (* 1 = 0.263547 loss) I0409 23:39:08.423918 4221 sgd_solver.cpp:105] Iteration 6660, lr = 0.00267335 I0409 23:39:13.378902 4221 solver.cpp:218] Iteration 6672 (2.4219 iter/s, 4.95479s/12 iters), loss = 0.208284 I0409 23:39:13.378957 4221 solver.cpp:237] Train net output #0: loss = 0.208285 (* 1 = 0.208285 loss) I0409 23:39:13.378969 4221 sgd_solver.cpp:105] Iteration 6672, lr = 0.00266701 I0409 23:39:14.839071 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:39:18.385510 4221 solver.cpp:218] Iteration 6684 (2.39696 iter/s, 5.00634s/12 iters), loss = 0.25295 I0409 23:39:18.385567 4221 solver.cpp:237] Train net output #0: loss = 0.25295 (* 1 = 0.25295 loss) I0409 23:39:18.385579 4221 sgd_solver.cpp:105] Iteration 6684, lr = 0.00266067 I0409 23:39:23.149088 4221 solver.cpp:218] Iteration 6696 (2.51925 iter/s, 4.76332s/12 iters), loss = 0.370788 I0409 23:39:23.149142 4221 solver.cpp:237] Train net output #0: loss = 0.370788 (* 1 = 0.370788 loss) I0409 23:39:23.149155 4221 sgd_solver.cpp:105] Iteration 6696, lr = 0.00265436 I0409 23:39:27.962358 4221 solver.cpp:218] Iteration 6708 (2.49324 iter/s, 4.81301s/12 iters), loss = 0.236746 I0409 23:39:27.962458 4221 solver.cpp:237] Train net output #0: loss = 0.236746 (* 1 = 0.236746 loss) I0409 23:39:27.962468 4221 sgd_solver.cpp:105] Iteration 6708, lr = 0.00264805 I0409 23:39:32.781020 4221 solver.cpp:218] Iteration 6720 (2.49048 iter/s, 4.81836s/12 iters), loss = 0.238688 I0409 23:39:32.781073 4221 solver.cpp:237] Train net output #0: loss = 0.238688 (* 1 = 0.238688 loss) I0409 23:39:32.781085 4221 sgd_solver.cpp:105] Iteration 6720, lr = 0.00264177 I0409 23:39:37.157251 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_6732.caffemodel I0409 23:39:37.869452 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_6732.solverstate I0409 23:39:38.374557 4221 solver.cpp:330] Iteration 6732, Testing net (#0) I0409 23:39:38.374579 4221 net.cpp:676] Ignoring source layer train-data I0409 23:39:40.194969 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:39:42.945350 4221 solver.cpp:397] Test net output #0: accuracy = 0.41973 I0409 23:39:42.945391 4221 solver.cpp:397] Test net output #1: loss = 3.70581 (* 1 = 3.70581 loss) I0409 23:39:43.028525 4221 solver.cpp:218] Iteration 6732 (1.17107 iter/s, 10.247s/12 iters), loss = 0.195334 I0409 23:39:43.028578 4221 solver.cpp:237] Train net output #0: loss = 0.195334 (* 1 = 0.195334 loss) I0409 23:39:43.028587 4221 sgd_solver.cpp:105] Iteration 6732, lr = 0.0026355 I0409 23:39:47.099953 4221 solver.cpp:218] Iteration 6744 (2.94754 iter/s, 4.07119s/12 iters), loss = 0.254974 I0409 23:39:47.100010 4221 solver.cpp:237] Train net output #0: loss = 0.254974 (* 1 = 0.254974 loss) I0409 23:39:47.100023 4221 sgd_solver.cpp:105] Iteration 6744, lr = 0.00262924 I0409 23:39:51.896620 4221 solver.cpp:218] Iteration 6756 (2.50188 iter/s, 4.7964s/12 iters), loss = 0.256197 I0409 23:39:51.896668 4221 solver.cpp:237] Train net output #0: loss = 0.256197 (* 1 = 0.256197 loss) I0409 23:39:51.896677 4221 sgd_solver.cpp:105] Iteration 6756, lr = 0.002623 I0409 23:39:56.732770 4221 solver.cpp:218] Iteration 6768 (2.48145 iter/s, 4.83589s/12 iters), loss = 0.156292 I0409 23:39:56.732828 4221 solver.cpp:237] Train net output #0: loss = 0.156292 (* 1 = 0.156292 loss) I0409 23:39:56.732841 4221 sgd_solver.cpp:105] Iteration 6768, lr = 0.00261677 I0409 23:40:00.076810 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:40:01.541324 4221 solver.cpp:218] Iteration 6780 (2.49569 iter/s, 4.80828s/12 iters), loss = 0.302449 I0409 23:40:01.541384 4221 solver.cpp:237] Train net output #0: loss = 0.302449 (* 1 = 0.302449 loss) I0409 23:40:01.541396 4221 sgd_solver.cpp:105] Iteration 6780, lr = 0.00261056 I0409 23:40:06.422123 4221 solver.cpp:218] Iteration 6792 (2.45875 iter/s, 4.88053s/12 iters), loss = 0.158738 I0409 23:40:06.422183 4221 solver.cpp:237] Train net output #0: loss = 0.158738 (* 1 = 0.158738 loss) I0409 23:40:06.422195 4221 sgd_solver.cpp:105] Iteration 6792, lr = 0.00260436 I0409 23:40:11.212625 4221 solver.cpp:218] Iteration 6804 (2.5051 iter/s, 4.79023s/12 iters), loss = 0.305778 I0409 23:40:11.212698 4221 solver.cpp:237] Train net output #0: loss = 0.305778 (* 1 = 0.305778 loss) I0409 23:40:11.212716 4221 sgd_solver.cpp:105] Iteration 6804, lr = 0.00259817 I0409 23:40:16.026161 4221 solver.cpp:218] Iteration 6816 (2.49311 iter/s, 4.81326s/12 iters), loss = 0.186738 I0409 23:40:16.026219 4221 solver.cpp:237] Train net output #0: loss = 0.186739 (* 1 = 0.186739 loss) I0409 23:40:16.026232 4221 sgd_solver.cpp:105] Iteration 6816, lr = 0.00259201 I0409 23:40:20.875917 4221 solver.cpp:218] Iteration 6828 (2.47449 iter/s, 4.84949s/12 iters), loss = 0.22517 I0409 23:40:20.875962 4221 solver.cpp:237] Train net output #0: loss = 0.22517 (* 1 = 0.22517 loss) I0409 23:40:20.875970 4221 sgd_solver.cpp:105] Iteration 6828, lr = 0.00258585 I0409 23:40:22.843765 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_6834.caffemodel I0409 23:40:23.549625 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_6834.solverstate I0409 23:40:24.051388 4221 solver.cpp:330] Iteration 6834, Testing net (#0) I0409 23:40:24.051419 4221 net.cpp:676] Ignoring source layer train-data I0409 23:40:25.826551 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:40:28.678952 4221 solver.cpp:397] Test net output #0: accuracy = 0.413603 I0409 23:40:28.679003 4221 solver.cpp:397] Test net output #1: loss = 3.50285 (* 1 = 3.50285 loss) I0409 23:40:30.417933 4221 solver.cpp:218] Iteration 6840 (1.25765 iter/s, 9.54157s/12 iters), loss = 0.170048 I0409 23:40:30.418061 4221 solver.cpp:237] Train net output #0: loss = 0.170048 (* 1 = 0.170048 loss) I0409 23:40:30.418074 4221 sgd_solver.cpp:105] Iteration 6840, lr = 0.00257971 I0409 23:40:35.524590 4221 solver.cpp:218] Iteration 6852 (2.35003 iter/s, 5.10631s/12 iters), loss = 0.223947 I0409 23:40:35.524650 4221 solver.cpp:237] Train net output #0: loss = 0.223947 (* 1 = 0.223947 loss) I0409 23:40:35.524663 4221 sgd_solver.cpp:105] Iteration 6852, lr = 0.00257359 I0409 23:40:40.356159 4221 solver.cpp:218] Iteration 6864 (2.4838 iter/s, 4.83131s/12 iters), loss = 0.219684 I0409 23:40:40.356200 4221 solver.cpp:237] Train net output #0: loss = 0.219684 (* 1 = 0.219684 loss) I0409 23:40:40.356209 4221 sgd_solver.cpp:105] Iteration 6864, lr = 0.00256748 I0409 23:40:45.145036 4221 solver.cpp:218] Iteration 6876 (2.50594 iter/s, 4.78863s/12 iters), loss = 0.24368 I0409 23:40:45.145123 4221 solver.cpp:237] Train net output #0: loss = 0.24368 (* 1 = 0.24368 loss) I0409 23:40:45.145133 4221 sgd_solver.cpp:105] Iteration 6876, lr = 0.00256138 I0409 23:40:45.742604 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:40:49.944075 4221 solver.cpp:218] Iteration 6888 (2.50065 iter/s, 4.79874s/12 iters), loss = 0.128262 I0409 23:40:49.944131 4221 solver.cpp:237] Train net output #0: loss = 0.128262 (* 1 = 0.128262 loss) I0409 23:40:49.944144 4221 sgd_solver.cpp:105] Iteration 6888, lr = 0.0025553 I0409 23:40:54.977398 4221 solver.cpp:218] Iteration 6900 (2.38424 iter/s, 5.03305s/12 iters), loss = 0.19551 I0409 23:40:54.977450 4221 solver.cpp:237] Train net output #0: loss = 0.19551 (* 1 = 0.19551 loss) I0409 23:40:54.977461 4221 sgd_solver.cpp:105] Iteration 6900, lr = 0.00254923 I0409 23:40:59.876482 4221 solver.cpp:218] Iteration 6912 (2.44957 iter/s, 4.89882s/12 iters), loss = 0.342614 I0409 23:40:59.876541 4221 solver.cpp:237] Train net output #0: loss = 0.342614 (* 1 = 0.342614 loss) I0409 23:40:59.876554 4221 sgd_solver.cpp:105] Iteration 6912, lr = 0.00254318 I0409 23:41:04.750459 4221 solver.cpp:218] Iteration 6924 (2.46219 iter/s, 4.87371s/12 iters), loss = 0.257678 I0409 23:41:04.750591 4221 solver.cpp:237] Train net output #0: loss = 0.257678 (* 1 = 0.257678 loss) I0409 23:41:04.750603 4221 sgd_solver.cpp:105] Iteration 6924, lr = 0.00253714 I0409 23:41:09.194020 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_6936.caffemodel I0409 23:41:10.028283 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_6936.solverstate I0409 23:41:11.819867 4221 solver.cpp:330] Iteration 6936, Testing net (#0) I0409 23:41:11.819897 4221 net.cpp:676] Ignoring source layer train-data I0409 23:41:12.187919 4221 blocking_queue.cpp:49] Waiting for data I0409 23:41:13.563094 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:41:16.396059 4221 solver.cpp:397] Test net output #0: accuracy = 0.428309 I0409 23:41:16.396108 4221 solver.cpp:397] Test net output #1: loss = 3.5908 (* 1 = 3.5908 loss) I0409 23:41:16.477499 4221 solver.cpp:218] Iteration 6936 (1.02333 iter/s, 11.7264s/12 iters), loss = 0.216227 I0409 23:41:16.477558 4221 solver.cpp:237] Train net output #0: loss = 0.216227 (* 1 = 0.216227 loss) I0409 23:41:16.477569 4221 sgd_solver.cpp:105] Iteration 6936, lr = 0.00253112 I0409 23:41:20.562753 4221 solver.cpp:218] Iteration 6948 (2.93756 iter/s, 4.08502s/12 iters), loss = 0.296967 I0409 23:41:20.562803 4221 solver.cpp:237] Train net output #0: loss = 0.296967 (* 1 = 0.296967 loss) I0409 23:41:20.562813 4221 sgd_solver.cpp:105] Iteration 6948, lr = 0.00252511 I0409 23:41:25.364877 4221 solver.cpp:218] Iteration 6960 (2.49903 iter/s, 4.80187s/12 iters), loss = 0.132421 I0409 23:41:25.364933 4221 solver.cpp:237] Train net output #0: loss = 0.132421 (* 1 = 0.132421 loss) I0409 23:41:25.364946 4221 sgd_solver.cpp:105] Iteration 6960, lr = 0.00251911 I0409 23:41:30.207470 4221 solver.cpp:218] Iteration 6972 (2.47815 iter/s, 4.84232s/12 iters), loss = 0.20033 I0409 23:41:30.207530 4221 solver.cpp:237] Train net output #0: loss = 0.20033 (* 1 = 0.20033 loss) I0409 23:41:30.207542 4221 sgd_solver.cpp:105] Iteration 6972, lr = 0.00251313 I0409 23:41:33.069409 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:41:35.279556 4221 solver.cpp:218] Iteration 6984 (2.36602 iter/s, 5.0718s/12 iters), loss = 0.149064 I0409 23:41:35.279976 4221 solver.cpp:237] Train net output #0: loss = 0.149064 (* 1 = 0.149064 loss) I0409 23:41:35.279992 4221 sgd_solver.cpp:105] Iteration 6984, lr = 0.00250717 I0409 23:41:40.169189 4221 solver.cpp:218] Iteration 6996 (2.45449 iter/s, 4.88901s/12 iters), loss = 0.264527 I0409 23:41:40.169246 4221 solver.cpp:237] Train net output #0: loss = 0.264527 (* 1 = 0.264527 loss) I0409 23:41:40.169258 4221 sgd_solver.cpp:105] Iteration 6996, lr = 0.00250121 I0409 23:41:45.105660 4221 solver.cpp:218] Iteration 7008 (2.43102 iter/s, 4.9362s/12 iters), loss = 0.27411 I0409 23:41:45.105707 4221 solver.cpp:237] Train net output #0: loss = 0.27411 (* 1 = 0.27411 loss) I0409 23:41:45.105717 4221 sgd_solver.cpp:105] Iteration 7008, lr = 0.00249528 I0409 23:41:50.124558 4221 solver.cpp:218] Iteration 7020 (2.39109 iter/s, 5.01863s/12 iters), loss = 0.228096 I0409 23:41:50.124609 4221 solver.cpp:237] Train net output #0: loss = 0.228096 (* 1 = 0.228096 loss) I0409 23:41:50.124619 4221 sgd_solver.cpp:105] Iteration 7020, lr = 0.00248935 I0409 23:41:54.987407 4221 solver.cpp:218] Iteration 7032 (2.46782 iter/s, 4.86259s/12 iters), loss = 0.197289 I0409 23:41:54.987460 4221 solver.cpp:237] Train net output #0: loss = 0.197289 (* 1 = 0.197289 loss) I0409 23:41:54.987473 4221 sgd_solver.cpp:105] Iteration 7032, lr = 0.00248344 I0409 23:41:56.947278 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_7038.caffemodel I0409 23:41:57.657882 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_7038.solverstate I0409 23:41:58.145700 4221 solver.cpp:330] Iteration 7038, Testing net (#0) I0409 23:41:58.145722 4221 net.cpp:676] Ignoring source layer train-data I0409 23:41:59.702132 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:42:02.517629 4221 solver.cpp:397] Test net output #0: accuracy = 0.423407 I0409 23:42:02.517674 4221 solver.cpp:397] Test net output #1: loss = 3.58667 (* 1 = 3.58667 loss) I0409 23:42:04.226073 4221 solver.cpp:218] Iteration 7044 (1.29895 iter/s, 9.23823s/12 iters), loss = 0.0856219 I0409 23:42:04.226123 4221 solver.cpp:237] Train net output #0: loss = 0.0856219 (* 1 = 0.0856219 loss) I0409 23:42:04.226133 4221 sgd_solver.cpp:105] Iteration 7044, lr = 0.00247755 I0409 23:42:09.125373 4221 solver.cpp:218] Iteration 7056 (2.44946 iter/s, 4.89904s/12 iters), loss = 0.222594 I0409 23:42:09.125546 4221 solver.cpp:237] Train net output #0: loss = 0.222594 (* 1 = 0.222594 loss) I0409 23:42:09.125560 4221 sgd_solver.cpp:105] Iteration 7056, lr = 0.00247166 I0409 23:42:13.988792 4221 solver.cpp:218] Iteration 7068 (2.46759 iter/s, 4.86304s/12 iters), loss = 0.203515 I0409 23:42:13.988847 4221 solver.cpp:237] Train net output #0: loss = 0.203515 (* 1 = 0.203515 loss) I0409 23:42:13.988858 4221 sgd_solver.cpp:105] Iteration 7068, lr = 0.0024658 I0409 23:42:18.698956 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:42:18.800647 4221 solver.cpp:218] Iteration 7080 (2.49398 iter/s, 4.8116s/12 iters), loss = 0.195622 I0409 23:42:18.800689 4221 solver.cpp:237] Train net output #0: loss = 0.195622 (* 1 = 0.195622 loss) I0409 23:42:18.800698 4221 sgd_solver.cpp:105] Iteration 7080, lr = 0.00245994 I0409 23:42:23.703722 4221 solver.cpp:218] Iteration 7092 (2.44757 iter/s, 4.90282s/12 iters), loss = 0.187305 I0409 23:42:23.703776 4221 solver.cpp:237] Train net output #0: loss = 0.187305 (* 1 = 0.187305 loss) I0409 23:42:23.703788 4221 sgd_solver.cpp:105] Iteration 7092, lr = 0.0024541 I0409 23:42:28.551427 4221 solver.cpp:218] Iteration 7104 (2.47553 iter/s, 4.84744s/12 iters), loss = 0.207072 I0409 23:42:28.551474 4221 solver.cpp:237] Train net output #0: loss = 0.207072 (* 1 = 0.207072 loss) I0409 23:42:28.551483 4221 sgd_solver.cpp:105] Iteration 7104, lr = 0.00244827 I0409 23:42:33.513514 4221 solver.cpp:218] Iteration 7116 (2.41847 iter/s, 4.96182s/12 iters), loss = 0.25504 I0409 23:42:33.513564 4221 solver.cpp:237] Train net output #0: loss = 0.25504 (* 1 = 0.25504 loss) I0409 23:42:33.513576 4221 sgd_solver.cpp:105] Iteration 7116, lr = 0.00244246 I0409 23:42:38.515635 4221 solver.cpp:218] Iteration 7128 (2.39911 iter/s, 5.00186s/12 iters), loss = 0.150411 I0409 23:42:38.515687 4221 solver.cpp:237] Train net output #0: loss = 0.150411 (* 1 = 0.150411 loss) I0409 23:42:38.515700 4221 sgd_solver.cpp:105] Iteration 7128, lr = 0.00243666 I0409 23:42:43.041154 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_7140.caffemodel I0409 23:42:43.789930 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_7140.solverstate I0409 23:42:44.903354 4221 solver.cpp:330] Iteration 7140, Testing net (#0) I0409 23:42:44.903380 4221 net.cpp:676] Ignoring source layer train-data I0409 23:42:46.575174 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:42:49.469573 4221 solver.cpp:397] Test net output #0: accuracy = 0.425245 I0409 23:42:49.469621 4221 solver.cpp:397] Test net output #1: loss = 3.58864 (* 1 = 3.58864 loss) I0409 23:42:49.551988 4221 solver.cpp:218] Iteration 7140 (1.08737 iter/s, 11.0358s/12 iters), loss = 0.295782 I0409 23:42:49.552045 4221 solver.cpp:237] Train net output #0: loss = 0.295782 (* 1 = 0.295782 loss) I0409 23:42:49.552057 4221 sgd_solver.cpp:105] Iteration 7140, lr = 0.00243088 I0409 23:42:53.760305 4221 solver.cpp:218] Iteration 7152 (2.85166 iter/s, 4.20807s/12 iters), loss = 0.241706 I0409 23:42:53.760363 4221 solver.cpp:237] Train net output #0: loss = 0.241706 (* 1 = 0.241706 loss) I0409 23:42:53.760375 4221 sgd_solver.cpp:105] Iteration 7152, lr = 0.00242511 I0409 23:42:58.580034 4221 solver.cpp:218] Iteration 7164 (2.4899 iter/s, 4.81947s/12 iters), loss = 0.26219 I0409 23:42:58.580080 4221 solver.cpp:237] Train net output #0: loss = 0.26219 (* 1 = 0.26219 loss) I0409 23:42:58.580090 4221 sgd_solver.cpp:105] Iteration 7164, lr = 0.00241935 I0409 23:43:03.523739 4221 solver.cpp:218] Iteration 7176 (2.42746 iter/s, 4.94345s/12 iters), loss = 0.119598 I0409 23:43:03.523780 4221 solver.cpp:237] Train net output #0: loss = 0.119598 (* 1 = 0.119598 loss) I0409 23:43:03.523789 4221 sgd_solver.cpp:105] Iteration 7176, lr = 0.0024136 I0409 23:43:05.538329 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:43:08.306746 4221 solver.cpp:218] Iteration 7188 (2.50902 iter/s, 4.78275s/12 iters), loss = 0.209205 I0409 23:43:08.306803 4221 solver.cpp:237] Train net output #0: loss = 0.209205 (* 1 = 0.209205 loss) I0409 23:43:08.306816 4221 sgd_solver.cpp:105] Iteration 7188, lr = 0.00240787 I0409 23:43:13.122157 4221 solver.cpp:218] Iteration 7200 (2.49214 iter/s, 4.81515s/12 iters), loss = 0.11195 I0409 23:43:13.122318 4221 solver.cpp:237] Train net output #0: loss = 0.11195 (* 1 = 0.11195 loss) I0409 23:43:13.122331 4221 sgd_solver.cpp:105] Iteration 7200, lr = 0.00240216 I0409 23:43:17.964982 4221 solver.cpp:218] Iteration 7212 (2.47808 iter/s, 4.84245s/12 iters), loss = 0.154869 I0409 23:43:17.965040 4221 solver.cpp:237] Train net output #0: loss = 0.154869 (* 1 = 0.154869 loss) I0409 23:43:17.965052 4221 sgd_solver.cpp:105] Iteration 7212, lr = 0.00239645 I0409 23:43:22.784576 4221 solver.cpp:218] Iteration 7224 (2.48998 iter/s, 4.81932s/12 iters), loss = 0.190166 I0409 23:43:22.784638 4221 solver.cpp:237] Train net output #0: loss = 0.190166 (* 1 = 0.190166 loss) I0409 23:43:22.784651 4221 sgd_solver.cpp:105] Iteration 7224, lr = 0.00239076 I0409 23:43:27.636211 4221 solver.cpp:218] Iteration 7236 (2.47353 iter/s, 4.85136s/12 iters), loss = 0.18287 I0409 23:43:27.636274 4221 solver.cpp:237] Train net output #0: loss = 0.18287 (* 1 = 0.18287 loss) I0409 23:43:27.636287 4221 sgd_solver.cpp:105] Iteration 7236, lr = 0.00238509 I0409 23:43:29.611148 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_7242.caffemodel I0409 23:43:32.021302 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_7242.solverstate I0409 23:43:32.515560 4221 solver.cpp:330] Iteration 7242, Testing net (#0) I0409 23:43:32.515583 4221 net.cpp:676] Ignoring source layer train-data I0409 23:43:34.129616 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:43:37.019914 4221 solver.cpp:397] Test net output #0: accuracy = 0.431373 I0409 23:43:37.019944 4221 solver.cpp:397] Test net output #1: loss = 3.62476 (* 1 = 3.62476 loss) I0409 23:43:38.855073 4221 solver.cpp:218] Iteration 7248 (1.06968 iter/s, 11.2183s/12 iters), loss = 0.230446 I0409 23:43:38.855139 4221 solver.cpp:237] Train net output #0: loss = 0.230446 (* 1 = 0.230446 loss) I0409 23:43:38.855152 4221 sgd_solver.cpp:105] Iteration 7248, lr = 0.00237942 I0409 23:43:43.734396 4221 solver.cpp:218] Iteration 7260 (2.4595 iter/s, 4.87904s/12 iters), loss = 0.174228 I0409 23:43:43.734570 4221 solver.cpp:237] Train net output #0: loss = 0.174228 (* 1 = 0.174228 loss) I0409 23:43:43.734587 4221 sgd_solver.cpp:105] Iteration 7260, lr = 0.00237378 I0409 23:43:48.607103 4221 solver.cpp:218] Iteration 7272 (2.46288 iter/s, 4.87234s/12 iters), loss = 0.178889 I0409 23:43:48.607141 4221 solver.cpp:237] Train net output #0: loss = 0.178889 (* 1 = 0.178889 loss) I0409 23:43:48.607148 4221 sgd_solver.cpp:105] Iteration 7272, lr = 0.00236814 I0409 23:43:52.746004 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:43:53.477151 4221 solver.cpp:218] Iteration 7284 (2.46417 iter/s, 4.8698s/12 iters), loss = 0.210989 I0409 23:43:53.477196 4221 solver.cpp:237] Train net output #0: loss = 0.210989 (* 1 = 0.210989 loss) I0409 23:43:53.477206 4221 sgd_solver.cpp:105] Iteration 7284, lr = 0.00236252 I0409 23:43:58.300009 4221 solver.cpp:218] Iteration 7296 (2.48828 iter/s, 4.8226s/12 iters), loss = 0.158258 I0409 23:43:58.300055 4221 solver.cpp:237] Train net output #0: loss = 0.158258 (* 1 = 0.158258 loss) I0409 23:43:58.300065 4221 sgd_solver.cpp:105] Iteration 7296, lr = 0.00235691 I0409 23:44:03.087886 4221 solver.cpp:218] Iteration 7308 (2.50646 iter/s, 4.78762s/12 iters), loss = 0.22389 I0409 23:44:03.087944 4221 solver.cpp:237] Train net output #0: loss = 0.22389 (* 1 = 0.22389 loss) I0409 23:44:03.087956 4221 sgd_solver.cpp:105] Iteration 7308, lr = 0.00235131 I0409 23:44:07.886925 4221 solver.cpp:218] Iteration 7320 (2.50064 iter/s, 4.79877s/12 iters), loss = 0.138757 I0409 23:44:07.886981 4221 solver.cpp:237] Train net output #0: loss = 0.138757 (* 1 = 0.138757 loss) I0409 23:44:07.886994 4221 sgd_solver.cpp:105] Iteration 7320, lr = 0.00234573 I0409 23:44:12.702056 4221 solver.cpp:218] Iteration 7332 (2.49228 iter/s, 4.81487s/12 iters), loss = 0.141361 I0409 23:44:12.702111 4221 solver.cpp:237] Train net output #0: loss = 0.141361 (* 1 = 0.141361 loss) I0409 23:44:12.702123 4221 sgd_solver.cpp:105] Iteration 7332, lr = 0.00234016 I0409 23:44:17.059878 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_7344.caffemodel I0409 23:44:19.444988 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_7344.solverstate I0409 23:44:21.853157 4221 solver.cpp:330] Iteration 7344, Testing net (#0) I0409 23:44:21.853181 4221 net.cpp:676] Ignoring source layer train-data I0409 23:44:23.397946 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:44:26.266232 4221 solver.cpp:397] Test net output #0: accuracy = 0.431985 I0409 23:44:26.266278 4221 solver.cpp:397] Test net output #1: loss = 3.55367 (* 1 = 3.55367 loss) I0409 23:44:26.349520 4221 solver.cpp:218] Iteration 7344 (0.879324 iter/s, 13.6468s/12 iters), loss = 0.216986 I0409 23:44:26.349575 4221 solver.cpp:237] Train net output #0: loss = 0.216986 (* 1 = 0.216986 loss) I0409 23:44:26.349587 4221 sgd_solver.cpp:105] Iteration 7344, lr = 0.0023346 I0409 23:44:30.449719 4221 solver.cpp:218] Iteration 7356 (2.92685 iter/s, 4.09997s/12 iters), loss = 0.146443 I0409 23:44:30.449767 4221 solver.cpp:237] Train net output #0: loss = 0.146443 (* 1 = 0.146443 loss) I0409 23:44:30.449776 4221 sgd_solver.cpp:105] Iteration 7356, lr = 0.00232906 I0409 23:44:35.326414 4221 solver.cpp:218] Iteration 7368 (2.46081 iter/s, 4.87643s/12 iters), loss = 0.145381 I0409 23:44:35.326470 4221 solver.cpp:237] Train net output #0: loss = 0.145381 (* 1 = 0.145381 loss) I0409 23:44:35.326483 4221 sgd_solver.cpp:105] Iteration 7368, lr = 0.00232353 I0409 23:44:40.161522 4221 solver.cpp:218] Iteration 7380 (2.48198 iter/s, 4.83484s/12 iters), loss = 0.110986 I0409 23:44:40.161581 4221 solver.cpp:237] Train net output #0: loss = 0.110986 (* 1 = 0.110986 loss) I0409 23:44:40.161593 4221 sgd_solver.cpp:105] Iteration 7380, lr = 0.00231802 I0409 23:44:41.521893 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:44:45.038826 4221 solver.cpp:218] Iteration 7392 (2.46051 iter/s, 4.87703s/12 iters), loss = 0.198342 I0409 23:44:45.038887 4221 solver.cpp:237] Train net output #0: loss = 0.198342 (* 1 = 0.198342 loss) I0409 23:44:45.038902 4221 sgd_solver.cpp:105] Iteration 7392, lr = 0.00231251 I0409 23:44:49.919385 4221 solver.cpp:218] Iteration 7404 (2.45887 iter/s, 4.88029s/12 iters), loss = 0.131322 I0409 23:44:49.919495 4221 solver.cpp:237] Train net output #0: loss = 0.131322 (* 1 = 0.131322 loss) I0409 23:44:49.919507 4221 sgd_solver.cpp:105] Iteration 7404, lr = 0.00230702 I0409 23:44:54.992954 4221 solver.cpp:218] Iteration 7416 (2.36535 iter/s, 5.07324s/12 iters), loss = 0.18395 I0409 23:44:54.993003 4221 solver.cpp:237] Train net output #0: loss = 0.18395 (* 1 = 0.18395 loss) I0409 23:44:54.993013 4221 sgd_solver.cpp:105] Iteration 7416, lr = 0.00230154 I0409 23:44:59.935809 4221 solver.cpp:218] Iteration 7428 (2.42788 iter/s, 4.94259s/12 iters), loss = 0.228775 I0409 23:44:59.935863 4221 solver.cpp:237] Train net output #0: loss = 0.228775 (* 1 = 0.228775 loss) I0409 23:44:59.935874 4221 sgd_solver.cpp:105] Iteration 7428, lr = 0.00229608 I0409 23:45:04.857369 4221 solver.cpp:218] Iteration 7440 (2.43838 iter/s, 4.9213s/12 iters), loss = 0.100797 I0409 23:45:04.857411 4221 solver.cpp:237] Train net output #0: loss = 0.100797 (* 1 = 0.100797 loss) I0409 23:45:04.857420 4221 sgd_solver.cpp:105] Iteration 7440, lr = 0.00229063 I0409 23:45:06.859800 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_7446.caffemodel I0409 23:45:07.543231 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_7446.solverstate I0409 23:45:08.042204 4221 solver.cpp:330] Iteration 7446, Testing net (#0) I0409 23:45:08.042235 4221 net.cpp:676] Ignoring source layer train-data I0409 23:45:09.585579 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:45:12.501632 4221 solver.cpp:397] Test net output #0: accuracy = 0.436274 I0409 23:45:12.501689 4221 solver.cpp:397] Test net output #1: loss = 3.57592 (* 1 = 3.57592 loss) I0409 23:45:14.450847 4221 solver.cpp:218] Iteration 7452 (1.25091 iter/s, 9.59304s/12 iters), loss = 0.154159 I0409 23:45:14.450893 4221 solver.cpp:237] Train net output #0: loss = 0.154159 (* 1 = 0.154159 loss) I0409 23:45:14.450904 4221 sgd_solver.cpp:105] Iteration 7452, lr = 0.00228519 I0409 23:45:19.335875 4221 solver.cpp:218] Iteration 7464 (2.45661 iter/s, 4.88477s/12 iters), loss = 0.266225 I0409 23:45:19.335924 4221 solver.cpp:237] Train net output #0: loss = 0.266225 (* 1 = 0.266225 loss) I0409 23:45:19.335934 4221 sgd_solver.cpp:105] Iteration 7464, lr = 0.00227976 I0409 23:45:24.197726 4221 solver.cpp:218] Iteration 7476 (2.46833 iter/s, 4.86159s/12 iters), loss = 0.172313 I0409 23:45:24.197855 4221 solver.cpp:237] Train net output #0: loss = 0.172313 (* 1 = 0.172313 loss) I0409 23:45:24.197870 4221 sgd_solver.cpp:105] Iteration 7476, lr = 0.00227435 I0409 23:45:27.690956 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:45:29.134011 4221 solver.cpp:218] Iteration 7488 (2.43115 iter/s, 4.93594s/12 iters), loss = 0.197218 I0409 23:45:29.134070 4221 solver.cpp:237] Train net output #0: loss = 0.197218 (* 1 = 0.197218 loss) I0409 23:45:29.134090 4221 sgd_solver.cpp:105] Iteration 7488, lr = 0.00226895 I0409 23:45:34.000396 4221 solver.cpp:218] Iteration 7500 (2.46603 iter/s, 4.86611s/12 iters), loss = 0.154208 I0409 23:45:34.000452 4221 solver.cpp:237] Train net output #0: loss = 0.154208 (* 1 = 0.154208 loss) I0409 23:45:34.000465 4221 sgd_solver.cpp:105] Iteration 7500, lr = 0.00226357 I0409 23:45:38.888743 4221 solver.cpp:218] Iteration 7512 (2.45495 iter/s, 4.88808s/12 iters), loss = 0.152351 I0409 23:45:38.888794 4221 solver.cpp:237] Train net output #0: loss = 0.152351 (* 1 = 0.152351 loss) I0409 23:45:38.888804 4221 sgd_solver.cpp:105] Iteration 7512, lr = 0.00225819 I0409 23:45:43.756574 4221 solver.cpp:218] Iteration 7524 (2.4653 iter/s, 4.86757s/12 iters), loss = 0.135477 I0409 23:45:43.756633 4221 solver.cpp:237] Train net output #0: loss = 0.135477 (* 1 = 0.135477 loss) I0409 23:45:43.756646 4221 sgd_solver.cpp:105] Iteration 7524, lr = 0.00225283 I0409 23:45:48.830137 4221 solver.cpp:218] Iteration 7536 (2.36533 iter/s, 5.07329s/12 iters), loss = 0.315345 I0409 23:45:48.830181 4221 solver.cpp:237] Train net output #0: loss = 0.315345 (* 1 = 0.315345 loss) I0409 23:45:48.830190 4221 sgd_solver.cpp:105] Iteration 7536, lr = 0.00224748 I0409 23:45:53.279167 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_7548.caffemodel I0409 23:45:53.973368 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_7548.solverstate I0409 23:45:54.457132 4221 solver.cpp:330] Iteration 7548, Testing net (#0) I0409 23:45:54.457242 4221 net.cpp:676] Ignoring source layer train-data I0409 23:45:55.819618 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:45:59.031566 4221 solver.cpp:397] Test net output #0: accuracy = 0.43076 I0409 23:45:59.031602 4221 solver.cpp:397] Test net output #1: loss = 3.62098 (* 1 = 3.62098 loss) I0409 23:45:59.115063 4221 solver.cpp:218] Iteration 7548 (1.16681 iter/s, 10.2844s/12 iters), loss = 0.209072 I0409 23:45:59.115146 4221 solver.cpp:237] Train net output #0: loss = 0.209072 (* 1 = 0.209072 loss) I0409 23:45:59.115159 4221 sgd_solver.cpp:105] Iteration 7548, lr = 0.00224215 I0409 23:46:03.265357 4221 solver.cpp:218] Iteration 7560 (2.89155 iter/s, 4.15003s/12 iters), loss = 0.134265 I0409 23:46:03.265421 4221 solver.cpp:237] Train net output #0: loss = 0.134265 (* 1 = 0.134265 loss) I0409 23:46:03.265435 4221 sgd_solver.cpp:105] Iteration 7560, lr = 0.00223682 I0409 23:46:08.065549 4221 solver.cpp:218] Iteration 7572 (2.50004 iter/s, 4.79993s/12 iters), loss = 0.132997 I0409 23:46:08.065603 4221 solver.cpp:237] Train net output #0: loss = 0.132997 (* 1 = 0.132997 loss) I0409 23:46:08.065613 4221 sgd_solver.cpp:105] Iteration 7572, lr = 0.00223151 I0409 23:46:12.882725 4221 solver.cpp:218] Iteration 7584 (2.49122 iter/s, 4.81691s/12 iters), loss = 0.11544 I0409 23:46:12.882787 4221 solver.cpp:237] Train net output #0: loss = 0.11544 (* 1 = 0.11544 loss) I0409 23:46:12.882800 4221 sgd_solver.cpp:105] Iteration 7584, lr = 0.00222621 I0409 23:46:13.508564 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:46:17.697394 4221 solver.cpp:218] Iteration 7596 (2.49252 iter/s, 4.8144s/12 iters), loss = 0.140876 I0409 23:46:17.697435 4221 solver.cpp:237] Train net output #0: loss = 0.140876 (* 1 = 0.140876 loss) I0409 23:46:17.697443 4221 sgd_solver.cpp:105] Iteration 7596, lr = 0.00222093 I0409 23:46:22.602507 4221 solver.cpp:218] Iteration 7608 (2.44655 iter/s, 4.90486s/12 iters), loss = 0.200055 I0409 23:46:22.602562 4221 solver.cpp:237] Train net output #0: loss = 0.200055 (* 1 = 0.200055 loss) I0409 23:46:22.602573 4221 sgd_solver.cpp:105] Iteration 7608, lr = 0.00221565 I0409 23:46:27.420639 4221 solver.cpp:218] Iteration 7620 (2.49073 iter/s, 4.81787s/12 iters), loss = 0.200589 I0409 23:46:27.420743 4221 solver.cpp:237] Train net output #0: loss = 0.200589 (* 1 = 0.200589 loss) I0409 23:46:27.420754 4221 sgd_solver.cpp:105] Iteration 7620, lr = 0.00221039 I0409 23:46:28.552729 4221 blocking_queue.cpp:49] Waiting for data I0409 23:46:32.223140 4221 solver.cpp:218] Iteration 7632 (2.49886 iter/s, 4.80219s/12 iters), loss = 0.18504 I0409 23:46:32.223189 4221 solver.cpp:237] Train net output #0: loss = 0.18504 (* 1 = 0.18504 loss) I0409 23:46:32.223201 4221 sgd_solver.cpp:105] Iteration 7632, lr = 0.00220515 I0409 23:46:37.046639 4221 solver.cpp:218] Iteration 7644 (2.48796 iter/s, 4.82324s/12 iters), loss = 0.154344 I0409 23:46:37.046697 4221 solver.cpp:237] Train net output #0: loss = 0.154344 (* 1 = 0.154344 loss) I0409 23:46:37.046710 4221 sgd_solver.cpp:105] Iteration 7644, lr = 0.00219991 I0409 23:46:39.001904 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_7650.caffemodel I0409 23:46:39.660601 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_7650.solverstate I0409 23:46:40.140445 4221 solver.cpp:330] Iteration 7650, Testing net (#0) I0409 23:46:40.140470 4221 net.cpp:676] Ignoring source layer train-data I0409 23:46:41.474118 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:46:44.565770 4221 solver.cpp:397] Test net output #0: accuracy = 0.4375 I0409 23:46:44.565816 4221 solver.cpp:397] Test net output #1: loss = 3.51767 (* 1 = 3.51767 loss) I0409 23:46:46.396759 4221 solver.cpp:218] Iteration 7656 (1.28347 iter/s, 9.34968s/12 iters), loss = 0.136833 I0409 23:46:46.396806 4221 solver.cpp:237] Train net output #0: loss = 0.136833 (* 1 = 0.136833 loss) I0409 23:46:46.396817 4221 sgd_solver.cpp:105] Iteration 7656, lr = 0.00219469 I0409 23:46:51.210040 4221 solver.cpp:218] Iteration 7668 (2.49324 iter/s, 4.81302s/12 iters), loss = 0.144081 I0409 23:46:51.210098 4221 solver.cpp:237] Train net output #0: loss = 0.144081 (* 1 = 0.144081 loss) I0409 23:46:51.210110 4221 sgd_solver.cpp:105] Iteration 7668, lr = 0.00218948 I0409 23:46:56.000118 4221 solver.cpp:218] Iteration 7680 (2.50532 iter/s, 4.78982s/12 iters), loss = 0.118507 I0409 23:46:56.000164 4221 solver.cpp:237] Train net output #0: loss = 0.118507 (* 1 = 0.118507 loss) I0409 23:46:56.000171 4221 sgd_solver.cpp:105] Iteration 7680, lr = 0.00218428 I0409 23:46:58.689349 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:47:00.838258 4221 solver.cpp:218] Iteration 7692 (2.48043 iter/s, 4.83788s/12 iters), loss = 0.132491 I0409 23:47:00.838315 4221 solver.cpp:237] Train net output #0: loss = 0.132491 (* 1 = 0.132491 loss) I0409 23:47:00.838327 4221 sgd_solver.cpp:105] Iteration 7692, lr = 0.00217909 I0409 23:47:05.701223 4221 solver.cpp:218] Iteration 7704 (2.46777 iter/s, 4.8627s/12 iters), loss = 0.188354 I0409 23:47:05.701274 4221 solver.cpp:237] Train net output #0: loss = 0.188354 (* 1 = 0.188354 loss) I0409 23:47:05.701287 4221 sgd_solver.cpp:105] Iteration 7704, lr = 0.00217392 I0409 23:47:10.522553 4221 solver.cpp:218] Iteration 7716 (2.48908 iter/s, 4.82107s/12 iters), loss = 0.21645 I0409 23:47:10.522614 4221 solver.cpp:237] Train net output #0: loss = 0.21645 (* 1 = 0.21645 loss) I0409 23:47:10.522624 4221 sgd_solver.cpp:105] Iteration 7716, lr = 0.00216876 I0409 23:47:15.341753 4221 solver.cpp:218] Iteration 7728 (2.49018 iter/s, 4.81893s/12 iters), loss = 0.145048 I0409 23:47:15.341800 4221 solver.cpp:237] Train net output #0: loss = 0.145048 (* 1 = 0.145048 loss) I0409 23:47:15.341810 4221 sgd_solver.cpp:105] Iteration 7728, lr = 0.00216361 I0409 23:47:20.147509 4221 solver.cpp:218] Iteration 7740 (2.49714 iter/s, 4.80549s/12 iters), loss = 0.189418 I0409 23:47:20.147570 4221 solver.cpp:237] Train net output #0: loss = 0.189418 (* 1 = 0.189418 loss) I0409 23:47:20.147581 4221 sgd_solver.cpp:105] Iteration 7740, lr = 0.00215847 I0409 23:47:24.943166 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_7752.caffemodel I0409 23:47:25.684469 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_7752.solverstate I0409 23:47:26.178691 4221 solver.cpp:330] Iteration 7752, Testing net (#0) I0409 23:47:26.178717 4221 net.cpp:676] Ignoring source layer train-data I0409 23:47:27.512080 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:47:30.566226 4221 solver.cpp:397] Test net output #0: accuracy = 0.439338 I0409 23:47:30.568657 4221 solver.cpp:397] Test net output #1: loss = 3.58564 (* 1 = 3.58564 loss) I0409 23:47:30.651991 4221 solver.cpp:218] Iteration 7752 (1.14242 iter/s, 10.504s/12 iters), loss = 0.225997 I0409 23:47:30.652050 4221 solver.cpp:237] Train net output #0: loss = 0.225997 (* 1 = 0.225997 loss) I0409 23:47:30.652062 4221 sgd_solver.cpp:105] Iteration 7752, lr = 0.00215335 I0409 23:47:34.852553 4221 solver.cpp:218] Iteration 7764 (2.85692 iter/s, 4.20032s/12 iters), loss = 0.124082 I0409 23:47:34.852607 4221 solver.cpp:237] Train net output #0: loss = 0.124082 (* 1 = 0.124082 loss) I0409 23:47:34.852618 4221 sgd_solver.cpp:105] Iteration 7764, lr = 0.00214823 I0409 23:47:39.665993 4221 solver.cpp:218] Iteration 7776 (2.49316 iter/s, 4.81318s/12 iters), loss = 0.0587718 I0409 23:47:39.666043 4221 solver.cpp:237] Train net output #0: loss = 0.0587718 (* 1 = 0.0587718 loss) I0409 23:47:39.666054 4221 sgd_solver.cpp:105] Iteration 7776, lr = 0.00214313 I0409 23:47:44.458467 4221 solver.cpp:218] Iteration 7788 (2.50406 iter/s, 4.79221s/12 iters), loss = 0.159454 I0409 23:47:44.458526 4221 solver.cpp:237] Train net output #0: loss = 0.159454 (* 1 = 0.159454 loss) I0409 23:47:44.458537 4221 sgd_solver.cpp:105] Iteration 7788, lr = 0.00213805 I0409 23:47:44.466572 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:47:49.265856 4221 solver.cpp:218] Iteration 7800 (2.4963 iter/s, 4.80712s/12 iters), loss = 0.158543 I0409 23:47:49.265900 4221 solver.cpp:237] Train net output #0: loss = 0.158543 (* 1 = 0.158543 loss) I0409 23:47:49.265908 4221 sgd_solver.cpp:105] Iteration 7800, lr = 0.00213297 I0409 23:47:54.184811 4221 solver.cpp:218] Iteration 7812 (2.43967 iter/s, 4.9187s/12 iters), loss = 0.171012 I0409 23:47:54.184860 4221 solver.cpp:237] Train net output #0: loss = 0.171012 (* 1 = 0.171012 loss) I0409 23:47:54.184872 4221 sgd_solver.cpp:105] Iteration 7812, lr = 0.00212791 I0409 23:47:59.054162 4221 solver.cpp:218] Iteration 7824 (2.46452 iter/s, 4.8691s/12 iters), loss = 0.144704 I0409 23:47:59.054199 4221 solver.cpp:237] Train net output #0: loss = 0.144704 (* 1 = 0.144704 loss) I0409 23:47:59.054208 4221 sgd_solver.cpp:105] Iteration 7824, lr = 0.00212285 I0409 23:48:03.869987 4221 solver.cpp:218] Iteration 7836 (2.49192 iter/s, 4.81556s/12 iters), loss = 0.0394592 I0409 23:48:03.870121 4221 solver.cpp:237] Train net output #0: loss = 0.0394592 (* 1 = 0.0394592 loss) I0409 23:48:03.870136 4221 sgd_solver.cpp:105] Iteration 7836, lr = 0.00211781 I0409 23:48:08.717403 4221 solver.cpp:218] Iteration 7848 (2.47572 iter/s, 4.84708s/12 iters), loss = 0.137545 I0409 23:48:08.717450 4221 solver.cpp:237] Train net output #0: loss = 0.137545 (* 1 = 0.137545 loss) I0409 23:48:08.717458 4221 sgd_solver.cpp:105] Iteration 7848, lr = 0.00211279 I0409 23:48:10.837282 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_7854.caffemodel I0409 23:48:12.013841 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_7854.solverstate I0409 23:48:14.346698 4221 solver.cpp:330] Iteration 7854, Testing net (#0) I0409 23:48:14.346724 4221 net.cpp:676] Ignoring source layer train-data I0409 23:48:15.722798 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:48:18.836905 4221 solver.cpp:397] Test net output #0: accuracy = 0.425245 I0409 23:48:18.836941 4221 solver.cpp:397] Test net output #1: loss = 3.59865 (* 1 = 3.59865 loss) I0409 23:48:20.673919 4221 solver.cpp:218] Iteration 7860 (1.00368 iter/s, 11.956s/12 iters), loss = 0.132149 I0409 23:48:20.673990 4221 solver.cpp:237] Train net output #0: loss = 0.132149 (* 1 = 0.132149 loss) I0409 23:48:20.674003 4221 sgd_solver.cpp:105] Iteration 7860, lr = 0.00210777 I0409 23:48:25.531540 4221 solver.cpp:218] Iteration 7872 (2.47049 iter/s, 4.85734s/12 iters), loss = 0.0925284 I0409 23:48:25.531584 4221 solver.cpp:237] Train net output #0: loss = 0.0925284 (* 1 = 0.0925284 loss) I0409 23:48:25.531594 4221 sgd_solver.cpp:105] Iteration 7872, lr = 0.00210277 I0409 23:48:30.433341 4221 solver.cpp:218] Iteration 7884 (2.44821 iter/s, 4.90155s/12 iters), loss = 0.162672 I0409 23:48:30.433384 4221 solver.cpp:237] Train net output #0: loss = 0.162672 (* 1 = 0.162672 loss) I0409 23:48:30.433391 4221 sgd_solver.cpp:105] Iteration 7884, lr = 0.00209777 I0409 23:48:32.530242 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:48:35.309146 4221 solver.cpp:218] Iteration 7896 (2.46126 iter/s, 4.87555s/12 iters), loss = 0.13625 I0409 23:48:35.309507 4221 solver.cpp:237] Train net output #0: loss = 0.13625 (* 1 = 0.13625 loss) I0409 23:48:35.309517 4221 sgd_solver.cpp:105] Iteration 7896, lr = 0.00209279 I0409 23:48:40.240113 4221 solver.cpp:218] Iteration 7908 (2.43388 iter/s, 4.93039s/12 iters), loss = 0.07567 I0409 23:48:40.240164 4221 solver.cpp:237] Train net output #0: loss = 0.07567 (* 1 = 0.07567 loss) I0409 23:48:40.240175 4221 sgd_solver.cpp:105] Iteration 7908, lr = 0.00208782 I0409 23:48:45.053694 4221 solver.cpp:218] Iteration 7920 (2.49308 iter/s, 4.81333s/12 iters), loss = 0.254592 I0409 23:48:45.053738 4221 solver.cpp:237] Train net output #0: loss = 0.254592 (* 1 = 0.254592 loss) I0409 23:48:45.053747 4221 sgd_solver.cpp:105] Iteration 7920, lr = 0.00208287 I0409 23:48:49.858495 4221 solver.cpp:218] Iteration 7932 (2.49763 iter/s, 4.80455s/12 iters), loss = 0.133601 I0409 23:48:49.858541 4221 solver.cpp:237] Train net output #0: loss = 0.133601 (* 1 = 0.133601 loss) I0409 23:48:49.858552 4221 sgd_solver.cpp:105] Iteration 7932, lr = 0.00207792 I0409 23:48:54.752969 4221 solver.cpp:218] Iteration 7944 (2.45187 iter/s, 4.89422s/12 iters), loss = 0.250585 I0409 23:48:54.753011 4221 solver.cpp:237] Train net output #0: loss = 0.250585 (* 1 = 0.250585 loss) I0409 23:48:54.753021 4221 sgd_solver.cpp:105] Iteration 7944, lr = 0.00207299 I0409 23:48:59.122354 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_7956.caffemodel I0409 23:48:59.802969 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_7956.solverstate I0409 23:49:00.301012 4221 solver.cpp:330] Iteration 7956, Testing net (#0) I0409 23:49:00.301043 4221 net.cpp:676] Ignoring source layer train-data I0409 23:49:01.652319 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:49:04.947365 4221 solver.cpp:397] Test net output #0: accuracy = 0.434436 I0409 23:49:04.947415 4221 solver.cpp:397] Test net output #1: loss = 3.60373 (* 1 = 3.60373 loss) I0409 23:49:05.030719 4221 solver.cpp:218] Iteration 7956 (1.16763 iter/s, 10.2773s/12 iters), loss = 0.305111 I0409 23:49:05.030778 4221 solver.cpp:237] Train net output #0: loss = 0.305111 (* 1 = 0.305111 loss) I0409 23:49:05.030791 4221 sgd_solver.cpp:105] Iteration 7956, lr = 0.00206807 I0409 23:49:09.138706 4221 solver.cpp:218] Iteration 7968 (2.92131 iter/s, 4.10775s/12 iters), loss = 0.242799 I0409 23:49:09.138788 4221 solver.cpp:237] Train net output #0: loss = 0.242799 (* 1 = 0.242799 loss) I0409 23:49:09.138799 4221 sgd_solver.cpp:105] Iteration 7968, lr = 0.00206316 I0409 23:49:13.969844 4221 solver.cpp:218] Iteration 7980 (2.48404 iter/s, 4.83085s/12 iters), loss = 0.124109 I0409 23:49:13.969892 4221 solver.cpp:237] Train net output #0: loss = 0.124109 (* 1 = 0.124109 loss) I0409 23:49:13.969903 4221 sgd_solver.cpp:105] Iteration 7980, lr = 0.00205826 I0409 23:49:18.098800 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:49:18.790841 4221 solver.cpp:218] Iteration 7992 (2.48925 iter/s, 4.82074s/12 iters), loss = 0.160937 I0409 23:49:18.790889 4221 solver.cpp:237] Train net output #0: loss = 0.160937 (* 1 = 0.160937 loss) I0409 23:49:18.790897 4221 sgd_solver.cpp:105] Iteration 7992, lr = 0.00205337 I0409 23:49:23.619796 4221 solver.cpp:218] Iteration 8004 (2.48514 iter/s, 4.8287s/12 iters), loss = 0.16588 I0409 23:49:23.619843 4221 solver.cpp:237] Train net output #0: loss = 0.16588 (* 1 = 0.16588 loss) I0409 23:49:23.619853 4221 sgd_solver.cpp:105] Iteration 8004, lr = 0.0020485 I0409 23:49:28.450035 4221 solver.cpp:218] Iteration 8016 (2.48448 iter/s, 4.82998s/12 iters), loss = 0.0936862 I0409 23:49:28.450098 4221 solver.cpp:237] Train net output #0: loss = 0.0936862 (* 1 = 0.0936862 loss) I0409 23:49:28.450109 4221 sgd_solver.cpp:105] Iteration 8016, lr = 0.00204363 I0409 23:49:33.259573 4221 solver.cpp:218] Iteration 8028 (2.49519 iter/s, 4.80926s/12 iters), loss = 0.110877 I0409 23:49:33.259637 4221 solver.cpp:237] Train net output #0: loss = 0.110877 (* 1 = 0.110877 loss) I0409 23:49:33.259651 4221 sgd_solver.cpp:105] Iteration 8028, lr = 0.00203878 I0409 23:49:38.126963 4221 solver.cpp:218] Iteration 8040 (2.46553 iter/s, 4.86712s/12 iters), loss = 0.204049 I0409 23:49:38.127004 4221 solver.cpp:237] Train net output #0: loss = 0.204049 (* 1 = 0.204049 loss) I0409 23:49:38.127013 4221 sgd_solver.cpp:105] Iteration 8040, lr = 0.00203394 I0409 23:49:42.929675 4221 solver.cpp:218] Iteration 8052 (2.49872 iter/s, 4.80246s/12 iters), loss = 0.119665 I0409 23:49:42.929831 4221 solver.cpp:237] Train net output #0: loss = 0.119665 (* 1 = 0.119665 loss) I0409 23:49:42.929844 4221 sgd_solver.cpp:105] Iteration 8052, lr = 0.00202911 I0409 23:49:44.893632 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_8058.caffemodel I0409 23:49:45.599575 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_8058.solverstate I0409 23:49:46.089715 4221 solver.cpp:330] Iteration 8058, Testing net (#0) I0409 23:49:46.089740 4221 net.cpp:676] Ignoring source layer train-data I0409 23:49:47.348166 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:49:50.540285 4221 solver.cpp:397] Test net output #0: accuracy = 0.44424 I0409 23:49:50.540318 4221 solver.cpp:397] Test net output #1: loss = 3.59645 (* 1 = 3.59645 loss) I0409 23:49:52.389019 4221 solver.cpp:218] Iteration 8064 (1.26866 iter/s, 9.4588s/12 iters), loss = 0.101307 I0409 23:49:52.389063 4221 solver.cpp:237] Train net output #0: loss = 0.101307 (* 1 = 0.101307 loss) I0409 23:49:52.389075 4221 sgd_solver.cpp:105] Iteration 8064, lr = 0.00202429 I0409 23:49:57.203042 4221 solver.cpp:218] Iteration 8076 (2.49285 iter/s, 4.81376s/12 iters), loss = 0.110349 I0409 23:49:57.203102 4221 solver.cpp:237] Train net output #0: loss = 0.110349 (* 1 = 0.110349 loss) I0409 23:49:57.203114 4221 sgd_solver.cpp:105] Iteration 8076, lr = 0.00201949 I0409 23:50:02.041775 4221 solver.cpp:218] Iteration 8088 (2.48013 iter/s, 4.83846s/12 iters), loss = 0.120256 I0409 23:50:02.041826 4221 solver.cpp:237] Train net output #0: loss = 0.120256 (* 1 = 0.120256 loss) I0409 23:50:02.041834 4221 sgd_solver.cpp:105] Iteration 8088, lr = 0.00201469 I0409 23:50:03.402938 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:50:06.883899 4221 solver.cpp:218] Iteration 8100 (2.47838 iter/s, 4.84186s/12 iters), loss = 0.0576071 I0409 23:50:06.883947 4221 solver.cpp:237] Train net output #0: loss = 0.0576071 (* 1 = 0.0576071 loss) I0409 23:50:06.883957 4221 sgd_solver.cpp:105] Iteration 8100, lr = 0.00200991 I0409 23:50:11.726315 4221 solver.cpp:218] Iteration 8112 (2.47823 iter/s, 4.84216s/12 iters), loss = 0.141782 I0409 23:50:11.726368 4221 solver.cpp:237] Train net output #0: loss = 0.141782 (* 1 = 0.141782 loss) I0409 23:50:11.726378 4221 sgd_solver.cpp:105] Iteration 8112, lr = 0.00200514 I0409 23:50:16.497476 4221 solver.cpp:218] Iteration 8124 (2.51525 iter/s, 4.7709s/12 iters), loss = 0.0744115 I0409 23:50:16.497587 4221 solver.cpp:237] Train net output #0: loss = 0.0744115 (* 1 = 0.0744115 loss) I0409 23:50:16.497599 4221 sgd_solver.cpp:105] Iteration 8124, lr = 0.00200038 I0409 23:50:21.331701 4221 solver.cpp:218] Iteration 8136 (2.48246 iter/s, 4.83391s/12 iters), loss = 0.21022 I0409 23:50:21.331743 4221 solver.cpp:237] Train net output #0: loss = 0.21022 (* 1 = 0.21022 loss) I0409 23:50:21.331750 4221 sgd_solver.cpp:105] Iteration 8136, lr = 0.00199563 I0409 23:50:26.196110 4221 solver.cpp:218] Iteration 8148 (2.46703 iter/s, 4.86415s/12 iters), loss = 0.132646 I0409 23:50:26.196168 4221 solver.cpp:237] Train net output #0: loss = 0.132646 (* 1 = 0.132646 loss) I0409 23:50:26.196180 4221 sgd_solver.cpp:105] Iteration 8148, lr = 0.00199089 I0409 23:50:30.611250 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_8160.caffemodel I0409 23:50:32.391036 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_8160.solverstate I0409 23:50:33.638859 4221 solver.cpp:330] Iteration 8160, Testing net (#0) I0409 23:50:33.638881 4221 net.cpp:676] Ignoring source layer train-data I0409 23:50:34.926765 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:50:38.180724 4221 solver.cpp:397] Test net output #0: accuracy = 0.435049 I0409 23:50:38.180763 4221 solver.cpp:397] Test net output #1: loss = 3.60836 (* 1 = 3.60836 loss) I0409 23:50:38.264464 4221 solver.cpp:218] Iteration 8160 (0.994383 iter/s, 12.0678s/12 iters), loss = 0.0836188 I0409 23:50:38.264541 4221 solver.cpp:237] Train net output #0: loss = 0.0836188 (* 1 = 0.0836188 loss) I0409 23:50:38.264556 4221 sgd_solver.cpp:105] Iteration 8160, lr = 0.00198616 I0409 23:50:42.350512 4221 solver.cpp:218] Iteration 8172 (2.93701 iter/s, 4.08579s/12 iters), loss = 0.0411174 I0409 23:50:42.350565 4221 solver.cpp:237] Train net output #0: loss = 0.0411174 (* 1 = 0.0411174 loss) I0409 23:50:42.350577 4221 sgd_solver.cpp:105] Iteration 8172, lr = 0.00198145 I0409 23:50:47.299314 4221 solver.cpp:218] Iteration 8184 (2.42496 iter/s, 4.94854s/12 iters), loss = 0.157954 I0409 23:50:47.299412 4221 solver.cpp:237] Train net output #0: loss = 0.157954 (* 1 = 0.157954 loss) I0409 23:50:47.299422 4221 sgd_solver.cpp:105] Iteration 8184, lr = 0.00197674 I0409 23:50:50.781782 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:50:52.215811 4221 solver.cpp:218] Iteration 8196 (2.44092 iter/s, 4.91618s/12 iters), loss = 0.124511 I0409 23:50:52.215869 4221 solver.cpp:237] Train net output #0: loss = 0.124511 (* 1 = 0.124511 loss) I0409 23:50:52.215881 4221 sgd_solver.cpp:105] Iteration 8196, lr = 0.00197205 I0409 23:50:57.016291 4221 solver.cpp:218] Iteration 8208 (2.49989 iter/s, 4.80021s/12 iters), loss = 0.0890383 I0409 23:50:57.016350 4221 solver.cpp:237] Train net output #0: loss = 0.0890383 (* 1 = 0.0890383 loss) I0409 23:50:57.016366 4221 sgd_solver.cpp:105] Iteration 8208, lr = 0.00196737 I0409 23:51:01.800505 4221 solver.cpp:218] Iteration 8220 (2.50839 iter/s, 4.78395s/12 iters), loss = 0.170584 I0409 23:51:01.800549 4221 solver.cpp:237] Train net output #0: loss = 0.170584 (* 1 = 0.170584 loss) I0409 23:51:01.800557 4221 sgd_solver.cpp:105] Iteration 8220, lr = 0.0019627 I0409 23:51:06.618831 4221 solver.cpp:218] Iteration 8232 (2.49063 iter/s, 4.81807s/12 iters), loss = 0.138016 I0409 23:51:06.618887 4221 solver.cpp:237] Train net output #0: loss = 0.138016 (* 1 = 0.138016 loss) I0409 23:51:06.618899 4221 sgd_solver.cpp:105] Iteration 8232, lr = 0.00195804 I0409 23:51:11.531154 4221 solver.cpp:218] Iteration 8244 (2.44297 iter/s, 4.91205s/12 iters), loss = 0.210531 I0409 23:51:11.531211 4221 solver.cpp:237] Train net output #0: loss = 0.210531 (* 1 = 0.210531 loss) I0409 23:51:11.531224 4221 sgd_solver.cpp:105] Iteration 8244, lr = 0.00195339 I0409 23:51:16.506376 4221 solver.cpp:218] Iteration 8256 (2.41209 iter/s, 4.97495s/12 iters), loss = 0.0989067 I0409 23:51:16.506435 4221 solver.cpp:237] Train net output #0: loss = 0.0989067 (* 1 = 0.0989067 loss) I0409 23:51:16.506446 4221 sgd_solver.cpp:105] Iteration 8256, lr = 0.00194875 I0409 23:51:18.476528 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_8262.caffemodel I0409 23:51:20.724195 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_8262.solverstate I0409 23:51:21.263162 4221 solver.cpp:330] Iteration 8262, Testing net (#0) I0409 23:51:21.263185 4221 net.cpp:676] Ignoring source layer train-data I0409 23:51:22.370327 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:51:25.648463 4221 solver.cpp:397] Test net output #0: accuracy = 0.439951 I0409 23:51:25.648507 4221 solver.cpp:397] Test net output #1: loss = 3.56412 (* 1 = 3.56412 loss) I0409 23:51:27.534420 4221 solver.cpp:218] Iteration 8268 (1.08819 iter/s, 11.0275s/12 iters), loss = 0.0784357 I0409 23:51:27.534477 4221 solver.cpp:237] Train net output #0: loss = 0.0784357 (* 1 = 0.0784357 loss) I0409 23:51:27.534488 4221 sgd_solver.cpp:105] Iteration 8268, lr = 0.00194412 I0409 23:51:32.374545 4221 solver.cpp:218] Iteration 8280 (2.47941 iter/s, 4.83986s/12 iters), loss = 0.186161 I0409 23:51:32.374600 4221 solver.cpp:237] Train net output #0: loss = 0.186161 (* 1 = 0.186161 loss) I0409 23:51:32.374611 4221 sgd_solver.cpp:105] Iteration 8280, lr = 0.00193951 I0409 23:51:37.334986 4221 solver.cpp:218] Iteration 8292 (2.41927 iter/s, 4.96017s/12 iters), loss = 0.207908 I0409 23:51:37.335031 4221 solver.cpp:237] Train net output #0: loss = 0.207908 (* 1 = 0.207908 loss) I0409 23:51:37.335039 4221 sgd_solver.cpp:105] Iteration 8292, lr = 0.0019349 I0409 23:51:38.023083 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:51:42.266873 4221 solver.cpp:218] Iteration 8304 (2.43328 iter/s, 4.93162s/12 iters), loss = 0.234659 I0409 23:51:42.266933 4221 solver.cpp:237] Train net output #0: loss = 0.234659 (* 1 = 0.234659 loss) I0409 23:51:42.266944 4221 sgd_solver.cpp:105] Iteration 8304, lr = 0.00193031 I0409 23:51:44.062458 4221 blocking_queue.cpp:49] Waiting for data I0409 23:51:47.437003 4221 solver.cpp:218] Iteration 8316 (2.32115 iter/s, 5.16985s/12 iters), loss = 0.08843 I0409 23:51:47.437053 4221 solver.cpp:237] Train net output #0: loss = 0.08843 (* 1 = 0.08843 loss) I0409 23:51:47.437065 4221 sgd_solver.cpp:105] Iteration 8316, lr = 0.00192573 I0409 23:51:52.304080 4221 solver.cpp:218] Iteration 8328 (2.46568 iter/s, 4.86682s/12 iters), loss = 0.144358 I0409 23:51:52.304172 4221 solver.cpp:237] Train net output #0: loss = 0.144358 (* 1 = 0.144358 loss) I0409 23:51:52.304179 4221 sgd_solver.cpp:105] Iteration 8328, lr = 0.00192115 I0409 23:51:57.234413 4221 solver.cpp:218] Iteration 8340 (2.43406 iter/s, 4.93003s/12 iters), loss = 0.146581 I0409 23:51:57.234467 4221 solver.cpp:237] Train net output #0: loss = 0.146581 (* 1 = 0.146581 loss) I0409 23:51:57.234477 4221 sgd_solver.cpp:105] Iteration 8340, lr = 0.00191659 I0409 23:52:02.202669 4221 solver.cpp:218] Iteration 8352 (2.41547 iter/s, 4.96799s/12 iters), loss = 0.0995088 I0409 23:52:02.202720 4221 solver.cpp:237] Train net output #0: loss = 0.0995088 (* 1 = 0.0995088 loss) I0409 23:52:02.202730 4221 sgd_solver.cpp:105] Iteration 8352, lr = 0.00191204 I0409 23:52:06.691920 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_8364.caffemodel I0409 23:52:07.399294 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_8364.solverstate I0409 23:52:07.896837 4221 solver.cpp:330] Iteration 8364, Testing net (#0) I0409 23:52:07.896862 4221 net.cpp:676] Ignoring source layer train-data I0409 23:52:08.988750 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:52:12.297942 4221 solver.cpp:397] Test net output #0: accuracy = 0.441176 I0409 23:52:12.298012 4221 solver.cpp:397] Test net output #1: loss = 3.63398 (* 1 = 3.63398 loss) I0409 23:52:12.381922 4221 solver.cpp:218] Iteration 8364 (1.17892 iter/s, 10.1788s/12 iters), loss = 0.114685 I0409 23:52:12.382000 4221 solver.cpp:237] Train net output #0: loss = 0.114685 (* 1 = 0.114685 loss) I0409 23:52:12.382014 4221 sgd_solver.cpp:105] Iteration 8364, lr = 0.0019075 I0409 23:52:16.432498 4221 solver.cpp:218] Iteration 8376 (2.96273 iter/s, 4.05032s/12 iters), loss = 0.0722024 I0409 23:52:16.432557 4221 solver.cpp:237] Train net output #0: loss = 0.0722024 (* 1 = 0.0722024 loss) I0409 23:52:16.432570 4221 sgd_solver.cpp:105] Iteration 8376, lr = 0.00190297 I0409 23:52:21.271358 4221 solver.cpp:218] Iteration 8388 (2.48006 iter/s, 4.83859s/12 iters), loss = 0.244462 I0409 23:52:21.271411 4221 solver.cpp:237] Train net output #0: loss = 0.244462 (* 1 = 0.244462 loss) I0409 23:52:21.271422 4221 sgd_solver.cpp:105] Iteration 8388, lr = 0.00189846 I0409 23:52:24.004650 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:52:26.124743 4221 solver.cpp:218] Iteration 8400 (2.47264 iter/s, 4.85312s/12 iters), loss = 0.108122 I0409 23:52:26.124804 4221 solver.cpp:237] Train net output #0: loss = 0.108122 (* 1 = 0.108122 loss) I0409 23:52:26.124815 4221 sgd_solver.cpp:105] Iteration 8400, lr = 0.00189395 I0409 23:52:30.990126 4221 solver.cpp:218] Iteration 8412 (2.46654 iter/s, 4.86511s/12 iters), loss = 0.179023 I0409 23:52:30.990177 4221 solver.cpp:237] Train net output #0: loss = 0.179023 (* 1 = 0.179023 loss) I0409 23:52:30.990190 4221 sgd_solver.cpp:105] Iteration 8412, lr = 0.00188945 I0409 23:52:35.838210 4221 solver.cpp:218] Iteration 8424 (2.47534 iter/s, 4.84782s/12 iters), loss = 0.0707992 I0409 23:52:35.838258 4221 solver.cpp:237] Train net output #0: loss = 0.0707992 (* 1 = 0.0707992 loss) I0409 23:52:35.838268 4221 sgd_solver.cpp:105] Iteration 8424, lr = 0.00188497 I0409 23:52:40.650547 4221 solver.cpp:218] Iteration 8436 (2.49373 iter/s, 4.81208s/12 iters), loss = 0.115104 I0409 23:52:40.650591 4221 solver.cpp:237] Train net output #0: loss = 0.115104 (* 1 = 0.115104 loss) I0409 23:52:40.650600 4221 sgd_solver.cpp:105] Iteration 8436, lr = 0.00188049 I0409 23:52:45.492763 4221 solver.cpp:218] Iteration 8448 (2.47834 iter/s, 4.84196s/12 iters), loss = 0.198136 I0409 23:52:45.492823 4221 solver.cpp:237] Train net output #0: loss = 0.198136 (* 1 = 0.198136 loss) I0409 23:52:45.492835 4221 sgd_solver.cpp:105] Iteration 8448, lr = 0.00187603 I0409 23:52:50.342134 4221 solver.cpp:218] Iteration 8460 (2.47469 iter/s, 4.8491s/12 iters), loss = 0.213667 I0409 23:52:50.342192 4221 solver.cpp:237] Train net output #0: loss = 0.213667 (* 1 = 0.213667 loss) I0409 23:52:50.342206 4221 sgd_solver.cpp:105] Iteration 8460, lr = 0.00187157 I0409 23:52:52.285104 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_8466.caffemodel I0409 23:52:52.948336 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_8466.solverstate I0409 23:52:53.442242 4221 solver.cpp:330] Iteration 8466, Testing net (#0) I0409 23:52:53.442281 4221 net.cpp:676] Ignoring source layer train-data I0409 23:52:54.489866 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:52:57.795214 4221 solver.cpp:397] Test net output #0: accuracy = 0.441176 I0409 23:52:57.795253 4221 solver.cpp:397] Test net output #1: loss = 3.64294 (* 1 = 3.64294 loss) I0409 23:52:59.544593 4221 solver.cpp:218] Iteration 8472 (1.30406 iter/s, 9.20201s/12 iters), loss = 0.229339 I0409 23:52:59.544648 4221 solver.cpp:237] Train net output #0: loss = 0.229339 (* 1 = 0.229339 loss) I0409 23:52:59.544661 4221 sgd_solver.cpp:105] Iteration 8472, lr = 0.00186713 I0409 23:53:04.349026 4221 solver.cpp:218] Iteration 8484 (2.49783 iter/s, 4.80417s/12 iters), loss = 0.198943 I0409 23:53:04.349076 4221 solver.cpp:237] Train net output #0: loss = 0.198943 (* 1 = 0.198943 loss) I0409 23:53:04.349088 4221 sgd_solver.cpp:105] Iteration 8484, lr = 0.0018627 I0409 23:53:09.313925 4221 solver.cpp:218] Iteration 8496 (2.41709 iter/s, 4.96464s/12 iters), loss = 0.198914 I0409 23:53:09.313980 4221 solver.cpp:237] Train net output #0: loss = 0.198914 (* 1 = 0.198914 loss) I0409 23:53:09.313990 4221 sgd_solver.cpp:105] Iteration 8496, lr = 0.00185827 I0409 23:53:09.360869 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:53:14.182979 4221 solver.cpp:218] Iteration 8508 (2.46468 iter/s, 4.86879s/12 iters), loss = 0.145573 I0409 23:53:14.183037 4221 solver.cpp:237] Train net output #0: loss = 0.145573 (* 1 = 0.145573 loss) I0409 23:53:14.183048 4221 sgd_solver.cpp:105] Iteration 8508, lr = 0.00185386 I0409 23:53:19.026932 4221 solver.cpp:218] Iteration 8520 (2.47745 iter/s, 4.84368s/12 iters), loss = 0.145563 I0409 23:53:19.026990 4221 solver.cpp:237] Train net output #0: loss = 0.145563 (* 1 = 0.145563 loss) I0409 23:53:19.027002 4221 sgd_solver.cpp:105] Iteration 8520, lr = 0.00184946 I0409 23:53:23.849222 4221 solver.cpp:218] Iteration 8532 (2.48858 iter/s, 4.82202s/12 iters), loss = 0.0557325 I0409 23:53:23.849272 4221 solver.cpp:237] Train net output #0: loss = 0.0557325 (* 1 = 0.0557325 loss) I0409 23:53:23.849282 4221 sgd_solver.cpp:105] Iteration 8532, lr = 0.00184507 I0409 23:53:28.716099 4221 solver.cpp:218] Iteration 8544 (2.46578 iter/s, 4.86662s/12 iters), loss = 0.0705072 I0409 23:53:28.716230 4221 solver.cpp:237] Train net output #0: loss = 0.0705072 (* 1 = 0.0705072 loss) I0409 23:53:28.716243 4221 sgd_solver.cpp:105] Iteration 8544, lr = 0.00184069 I0409 23:53:33.577350 4221 solver.cpp:218] Iteration 8556 (2.46867 iter/s, 4.86091s/12 iters), loss = 0.0704781 I0409 23:53:33.577409 4221 solver.cpp:237] Train net output #0: loss = 0.0704781 (* 1 = 0.0704781 loss) I0409 23:53:33.577421 4221 sgd_solver.cpp:105] Iteration 8556, lr = 0.00183632 I0409 23:53:38.399729 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_8568.caffemodel I0409 23:53:40.654259 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_8568.solverstate I0409 23:53:41.463238 4221 solver.cpp:330] Iteration 8568, Testing net (#0) I0409 23:53:41.463263 4221 net.cpp:676] Ignoring source layer train-data I0409 23:53:42.545634 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:53:45.941737 4221 solver.cpp:397] Test net output #0: accuracy = 0.433211 I0409 23:53:45.941774 4221 solver.cpp:397] Test net output #1: loss = 3.64708 (* 1 = 3.64708 loss) I0409 23:53:46.024997 4221 solver.cpp:218] Iteration 8568 (0.964082 iter/s, 12.4471s/12 iters), loss = 0.138566 I0409 23:53:46.025048 4221 solver.cpp:237] Train net output #0: loss = 0.138566 (* 1 = 0.138566 loss) I0409 23:53:46.025058 4221 sgd_solver.cpp:105] Iteration 8568, lr = 0.00183196 I0409 23:53:50.129184 4221 solver.cpp:218] Iteration 8580 (2.92401 iter/s, 4.10395s/12 iters), loss = 0.167211 I0409 23:53:50.129230 4221 solver.cpp:237] Train net output #0: loss = 0.167211 (* 1 = 0.167211 loss) I0409 23:53:50.129240 4221 sgd_solver.cpp:105] Iteration 8580, lr = 0.00182761 I0409 23:53:55.024302 4221 solver.cpp:218] Iteration 8592 (2.45155 iter/s, 4.89486s/12 iters), loss = 0.0514045 I0409 23:53:55.024354 4221 solver.cpp:237] Train net output #0: loss = 0.0514045 (* 1 = 0.0514045 loss) I0409 23:53:55.024366 4221 sgd_solver.cpp:105] Iteration 8592, lr = 0.00182327 I0409 23:53:57.166775 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:54:00.116426 4221 solver.cpp:218] Iteration 8604 (2.35671 iter/s, 5.09185s/12 iters), loss = 0.0470827 I0409 23:54:00.116529 4221 solver.cpp:237] Train net output #0: loss = 0.0470827 (* 1 = 0.0470827 loss) I0409 23:54:00.116542 4221 sgd_solver.cpp:105] Iteration 8604, lr = 0.00181894 I0409 23:54:05.023903 4221 solver.cpp:218] Iteration 8616 (2.4454 iter/s, 4.90717s/12 iters), loss = 0.02355 I0409 23:54:05.023950 4221 solver.cpp:237] Train net output #0: loss = 0.02355 (* 1 = 0.02355 loss) I0409 23:54:05.023960 4221 sgd_solver.cpp:105] Iteration 8616, lr = 0.00181462 I0409 23:54:09.886281 4221 solver.cpp:218] Iteration 8628 (2.46806 iter/s, 4.86212s/12 iters), loss = 0.0470131 I0409 23:54:09.886328 4221 solver.cpp:237] Train net output #0: loss = 0.0470131 (* 1 = 0.0470131 loss) I0409 23:54:09.886338 4221 sgd_solver.cpp:105] Iteration 8628, lr = 0.00181031 I0409 23:54:14.736182 4221 solver.cpp:218] Iteration 8640 (2.47441 iter/s, 4.84964s/12 iters), loss = 0.0427775 I0409 23:54:14.736224 4221 solver.cpp:237] Train net output #0: loss = 0.0427775 (* 1 = 0.0427775 loss) I0409 23:54:14.736233 4221 sgd_solver.cpp:105] Iteration 8640, lr = 0.00180602 I0409 23:54:19.636571 4221 solver.cpp:218] Iteration 8652 (2.44891 iter/s, 4.90013s/12 iters), loss = 0.144457 I0409 23:54:19.636616 4221 solver.cpp:237] Train net output #0: loss = 0.144457 (* 1 = 0.144457 loss) I0409 23:54:19.636626 4221 sgd_solver.cpp:105] Iteration 8652, lr = 0.00180173 I0409 23:54:24.554662 4221 solver.cpp:218] Iteration 8664 (2.4401 iter/s, 4.91783s/12 iters), loss = 0.194542 I0409 23:54:24.554716 4221 solver.cpp:237] Train net output #0: loss = 0.194543 (* 1 = 0.194543 loss) I0409 23:54:24.554728 4221 sgd_solver.cpp:105] Iteration 8664, lr = 0.00179745 I0409 23:54:26.568629 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_8670.caffemodel I0409 23:54:27.602437 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_8670.solverstate I0409 23:54:29.214511 4221 solver.cpp:330] Iteration 8670, Testing net (#0) I0409 23:54:29.214534 4221 net.cpp:676] Ignoring source layer train-data I0409 23:54:30.194173 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:54:33.894284 4221 solver.cpp:397] Test net output #0: accuracy = 0.45098 I0409 23:54:33.894320 4221 solver.cpp:397] Test net output #1: loss = 3.57218 (* 1 = 3.57218 loss) I0409 23:54:35.769881 4221 solver.cpp:218] Iteration 8676 (1.07002 iter/s, 11.2147s/12 iters), loss = 0.244128 I0409 23:54:35.769924 4221 solver.cpp:237] Train net output #0: loss = 0.244128 (* 1 = 0.244128 loss) I0409 23:54:35.769933 4221 sgd_solver.cpp:105] Iteration 8676, lr = 0.00179318 I0409 23:54:40.701423 4221 solver.cpp:218] Iteration 8688 (2.43344 iter/s, 4.93128s/12 iters), loss = 0.140979 I0409 23:54:40.701478 4221 solver.cpp:237] Train net output #0: loss = 0.140979 (* 1 = 0.140979 loss) I0409 23:54:40.701491 4221 sgd_solver.cpp:105] Iteration 8688, lr = 0.00178893 I0409 23:54:44.998208 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:54:45.700906 4221 solver.cpp:218] Iteration 8700 (2.40038 iter/s, 4.99921s/12 iters), loss = 0.188856 I0409 23:54:45.700965 4221 solver.cpp:237] Train net output #0: loss = 0.188856 (* 1 = 0.188856 loss) I0409 23:54:45.700978 4221 sgd_solver.cpp:105] Iteration 8700, lr = 0.00178468 I0409 23:54:50.548914 4221 solver.cpp:218] Iteration 8712 (2.47538 iter/s, 4.84774s/12 iters), loss = 0.129435 I0409 23:54:50.548957 4221 solver.cpp:237] Train net output #0: loss = 0.129435 (* 1 = 0.129435 loss) I0409 23:54:50.548966 4221 sgd_solver.cpp:105] Iteration 8712, lr = 0.00178044 I0409 23:54:55.412855 4221 solver.cpp:218] Iteration 8724 (2.46727 iter/s, 4.86368s/12 iters), loss = 0.0724701 I0409 23:54:55.412911 4221 solver.cpp:237] Train net output #0: loss = 0.0724701 (* 1 = 0.0724701 loss) I0409 23:54:55.412925 4221 sgd_solver.cpp:105] Iteration 8724, lr = 0.00177621 I0409 23:55:00.242245 4221 solver.cpp:218] Iteration 8736 (2.48492 iter/s, 4.82913s/12 iters), loss = 0.0535261 I0409 23:55:00.242344 4221 solver.cpp:237] Train net output #0: loss = 0.0535261 (* 1 = 0.0535261 loss) I0409 23:55:00.242354 4221 sgd_solver.cpp:105] Iteration 8736, lr = 0.001772 I0409 23:55:05.128023 4221 solver.cpp:218] Iteration 8748 (2.45627 iter/s, 4.88546s/12 iters), loss = 0.0283711 I0409 23:55:05.128084 4221 solver.cpp:237] Train net output #0: loss = 0.0283712 (* 1 = 0.0283712 loss) I0409 23:55:05.128098 4221 sgd_solver.cpp:105] Iteration 8748, lr = 0.00176779 I0409 23:55:09.996587 4221 solver.cpp:218] Iteration 8760 (2.46493 iter/s, 4.8683s/12 iters), loss = 0.10044 I0409 23:55:09.996640 4221 solver.cpp:237] Train net output #0: loss = 0.10044 (* 1 = 0.10044 loss) I0409 23:55:09.996651 4221 sgd_solver.cpp:105] Iteration 8760, lr = 0.00176359 I0409 23:55:14.486550 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_8772.caffemodel I0409 23:55:15.170516 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_8772.solverstate I0409 23:55:16.186693 4221 solver.cpp:330] Iteration 8772, Testing net (#0) I0409 23:55:16.186719 4221 net.cpp:676] Ignoring source layer train-data I0409 23:55:17.248940 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:55:20.698943 4221 solver.cpp:397] Test net output #0: accuracy = 0.45098 I0409 23:55:20.698992 4221 solver.cpp:397] Test net output #1: loss = 3.50167 (* 1 = 3.50167 loss) I0409 23:55:20.780916 4221 solver.cpp:218] Iteration 8772 (1.11278 iter/s, 10.7838s/12 iters), loss = 0.0967658 I0409 23:55:20.780967 4221 solver.cpp:237] Train net output #0: loss = 0.0967658 (* 1 = 0.0967658 loss) I0409 23:55:20.780978 4221 sgd_solver.cpp:105] Iteration 8772, lr = 0.00175941 I0409 23:55:24.976024 4221 solver.cpp:218] Iteration 8784 (2.86064 iter/s, 4.19487s/12 iters), loss = 0.125381 I0409 23:55:24.976078 4221 solver.cpp:237] Train net output #0: loss = 0.125381 (* 1 = 0.125381 loss) I0409 23:55:24.976089 4221 sgd_solver.cpp:105] Iteration 8784, lr = 0.00175523 I0409 23:55:29.855330 4221 solver.cpp:218] Iteration 8796 (2.4595 iter/s, 4.87904s/12 iters), loss = 0.158459 I0409 23:55:29.855386 4221 solver.cpp:237] Train net output #0: loss = 0.158459 (* 1 = 0.158459 loss) I0409 23:55:29.855396 4221 sgd_solver.cpp:105] Iteration 8796, lr = 0.00175106 I0409 23:55:31.264261 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:55:34.734516 4221 solver.cpp:218] Iteration 8808 (2.45956 iter/s, 4.87892s/12 iters), loss = 0.0718486 I0409 23:55:34.734575 4221 solver.cpp:237] Train net output #0: loss = 0.0718487 (* 1 = 0.0718487 loss) I0409 23:55:34.734588 4221 sgd_solver.cpp:105] Iteration 8808, lr = 0.0017469 I0409 23:55:39.853022 4221 solver.cpp:218] Iteration 8820 (2.34456 iter/s, 5.11823s/12 iters), loss = 0.153048 I0409 23:55:39.853072 4221 solver.cpp:237] Train net output #0: loss = 0.153048 (* 1 = 0.153048 loss) I0409 23:55:39.853085 4221 sgd_solver.cpp:105] Iteration 8820, lr = 0.00174276 I0409 23:55:44.634363 4221 solver.cpp:218] Iteration 8832 (2.5099 iter/s, 4.78107s/12 iters), loss = 0.11091 I0409 23:55:44.634423 4221 solver.cpp:237] Train net output #0: loss = 0.11091 (* 1 = 0.11091 loss) I0409 23:55:44.634436 4221 sgd_solver.cpp:105] Iteration 8832, lr = 0.00173862 I0409 23:55:49.396252 4221 solver.cpp:218] Iteration 8844 (2.52015 iter/s, 4.76162s/12 iters), loss = 0.176308 I0409 23:55:49.396311 4221 solver.cpp:237] Train net output #0: loss = 0.176308 (* 1 = 0.176308 loss) I0409 23:55:49.396322 4221 sgd_solver.cpp:105] Iteration 8844, lr = 0.00173449 I0409 23:55:54.276474 4221 solver.cpp:218] Iteration 8856 (2.45904 iter/s, 4.87995s/12 iters), loss = 0.163313 I0409 23:55:54.276517 4221 solver.cpp:237] Train net output #0: loss = 0.163313 (* 1 = 0.163313 loss) I0409 23:55:54.276525 4221 sgd_solver.cpp:105] Iteration 8856, lr = 0.00173037 I0409 23:55:59.098858 4221 solver.cpp:218] Iteration 8868 (2.48853 iter/s, 4.82213s/12 iters), loss = 0.0594289 I0409 23:55:59.098920 4221 solver.cpp:237] Train net output #0: loss = 0.0594289 (* 1 = 0.0594289 loss) I0409 23:55:59.098933 4221 sgd_solver.cpp:105] Iteration 8868, lr = 0.00172626 I0409 23:56:01.060945 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_8874.caffemodel I0409 23:56:01.774029 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_8874.solverstate I0409 23:56:02.308835 4221 solver.cpp:330] Iteration 8874, Testing net (#0) I0409 23:56:02.308863 4221 net.cpp:676] Ignoring source layer train-data I0409 23:56:03.313338 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:56:06.824824 4221 solver.cpp:397] Test net output #0: accuracy = 0.45527 I0409 23:56:06.824875 4221 solver.cpp:397] Test net output #1: loss = 3.59823 (* 1 = 3.59823 loss) I0409 23:56:08.716404 4221 solver.cpp:218] Iteration 8880 (1.24778 iter/s, 9.61708s/12 iters), loss = 0.138058 I0409 23:56:08.716456 4221 solver.cpp:237] Train net output #0: loss = 0.138058 (* 1 = 0.138058 loss) I0409 23:56:08.716467 4221 sgd_solver.cpp:105] Iteration 8880, lr = 0.00172217 I0409 23:56:13.576937 4221 solver.cpp:218] Iteration 8892 (2.469 iter/s, 4.86027s/12 iters), loss = 0.100353 I0409 23:56:13.576990 4221 solver.cpp:237] Train net output #0: loss = 0.100353 (* 1 = 0.100353 loss) I0409 23:56:13.577003 4221 sgd_solver.cpp:105] Iteration 8892, lr = 0.00171808 I0409 23:56:17.037803 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:56:18.419997 4221 solver.cpp:218] Iteration 8904 (2.47791 iter/s, 4.84279s/12 iters), loss = 0.133083 I0409 23:56:18.420055 4221 solver.cpp:237] Train net output #0: loss = 0.133083 (* 1 = 0.133083 loss) I0409 23:56:18.420068 4221 sgd_solver.cpp:105] Iteration 8904, lr = 0.001714 I0409 23:56:23.236667 4221 solver.cpp:218] Iteration 8916 (2.49149 iter/s, 4.8164s/12 iters), loss = 0.157719 I0409 23:56:23.236737 4221 solver.cpp:237] Train net output #0: loss = 0.157719 (* 1 = 0.157719 loss) I0409 23:56:23.236752 4221 sgd_solver.cpp:105] Iteration 8916, lr = 0.00170993 I0409 23:56:28.058147 4221 solver.cpp:218] Iteration 8928 (2.489 iter/s, 4.82121s/12 iters), loss = 0.144075 I0409 23:56:28.058198 4221 solver.cpp:237] Train net output #0: loss = 0.144076 (* 1 = 0.144076 loss) I0409 23:56:28.058212 4221 sgd_solver.cpp:105] Iteration 8928, lr = 0.00170587 I0409 23:56:32.930090 4221 solver.cpp:218] Iteration 8940 (2.46322 iter/s, 4.87168s/12 iters), loss = 0.104919 I0409 23:56:32.930244 4221 solver.cpp:237] Train net output #0: loss = 0.104919 (* 1 = 0.104919 loss) I0409 23:56:32.930255 4221 sgd_solver.cpp:105] Iteration 8940, lr = 0.00170182 I0409 23:56:37.777693 4221 solver.cpp:218] Iteration 8952 (2.47563 iter/s, 4.84725s/12 iters), loss = 0.177171 I0409 23:56:37.777736 4221 solver.cpp:237] Train net output #0: loss = 0.177171 (* 1 = 0.177171 loss) I0409 23:56:37.777745 4221 sgd_solver.cpp:105] Iteration 8952, lr = 0.00169778 I0409 23:56:42.583922 4221 solver.cpp:218] Iteration 8964 (2.49689 iter/s, 4.80598s/12 iters), loss = 0.0883841 I0409 23:56:42.583974 4221 solver.cpp:237] Train net output #0: loss = 0.0883842 (* 1 = 0.0883842 loss) I0409 23:56:42.583986 4221 sgd_solver.cpp:105] Iteration 8964, lr = 0.00169375 I0409 23:56:46.956403 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_8976.caffemodel I0409 23:56:47.661285 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_8976.solverstate I0409 23:56:48.147424 4221 solver.cpp:330] Iteration 8976, Testing net (#0) I0409 23:56:48.147454 4221 net.cpp:676] Ignoring source layer train-data I0409 23:56:49.065652 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:56:52.786033 4221 solver.cpp:397] Test net output #0: accuracy = 0.456495 I0409 23:56:52.786084 4221 solver.cpp:397] Test net output #1: loss = 3.64634 (* 1 = 3.64634 loss) I0409 23:56:52.869359 4221 solver.cpp:218] Iteration 8976 (1.16675 iter/s, 10.285s/12 iters), loss = 0.0436887 I0409 23:56:52.869415 4221 solver.cpp:237] Train net output #0: loss = 0.0436888 (* 1 = 0.0436888 loss) I0409 23:56:52.869426 4221 sgd_solver.cpp:105] Iteration 8976, lr = 0.00168973 I0409 23:56:56.897071 4221 solver.cpp:218] Iteration 8988 (2.97953 iter/s, 4.02749s/12 iters), loss = 0.0533534 I0409 23:56:56.897111 4221 solver.cpp:237] Train net output #0: loss = 0.0533535 (* 1 = 0.0533535 loss) I0409 23:56:56.897122 4221 sgd_solver.cpp:105] Iteration 8988, lr = 0.00168571 I0409 23:56:58.905822 4221 blocking_queue.cpp:49] Waiting for data I0409 23:57:01.830070 4221 solver.cpp:218] Iteration 9000 (2.43272 iter/s, 4.93275s/12 iters), loss = 0.129487 I0409 23:57:01.830123 4221 solver.cpp:237] Train net output #0: loss = 0.129487 (* 1 = 0.129487 loss) I0409 23:57:01.830134 4221 sgd_solver.cpp:105] Iteration 9000, lr = 0.00168171 I0409 23:57:02.543305 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:57:06.788272 4221 solver.cpp:218] Iteration 9012 (2.42036 iter/s, 4.95793s/12 iters), loss = 0.125857 I0409 23:57:06.788403 4221 solver.cpp:237] Train net output #0: loss = 0.125857 (* 1 = 0.125857 loss) I0409 23:57:06.788417 4221 sgd_solver.cpp:105] Iteration 9012, lr = 0.00167772 I0409 23:57:11.633334 4221 solver.cpp:218] Iteration 9024 (2.47692 iter/s, 4.84472s/12 iters), loss = 0.105557 I0409 23:57:11.633391 4221 solver.cpp:237] Train net output #0: loss = 0.105557 (* 1 = 0.105557 loss) I0409 23:57:11.633404 4221 sgd_solver.cpp:105] Iteration 9024, lr = 0.00167374 I0409 23:57:16.534279 4221 solver.cpp:218] Iteration 9036 (2.44864 iter/s, 4.90068s/12 iters), loss = 0.0374462 I0409 23:57:16.534333 4221 solver.cpp:237] Train net output #0: loss = 0.0374462 (* 1 = 0.0374462 loss) I0409 23:57:16.534346 4221 sgd_solver.cpp:105] Iteration 9036, lr = 0.00166976 I0409 23:57:21.417691 4221 solver.cpp:218] Iteration 9048 (2.45743 iter/s, 4.88314s/12 iters), loss = 0.133817 I0409 23:57:21.417752 4221 solver.cpp:237] Train net output #0: loss = 0.133817 (* 1 = 0.133817 loss) I0409 23:57:21.417763 4221 sgd_solver.cpp:105] Iteration 9048, lr = 0.0016658 I0409 23:57:26.307538 4221 solver.cpp:218] Iteration 9060 (2.4542 iter/s, 4.88957s/12 iters), loss = 0.0766303 I0409 23:57:26.307595 4221 solver.cpp:237] Train net output #0: loss = 0.0766303 (* 1 = 0.0766303 loss) I0409 23:57:26.307606 4221 sgd_solver.cpp:105] Iteration 9060, lr = 0.00166184 I0409 23:57:31.174294 4221 solver.cpp:218] Iteration 9072 (2.46585 iter/s, 4.86648s/12 iters), loss = 0.129839 I0409 23:57:31.174365 4221 solver.cpp:237] Train net output #0: loss = 0.129839 (* 1 = 0.129839 loss) I0409 23:57:31.174374 4221 sgd_solver.cpp:105] Iteration 9072, lr = 0.0016579 I0409 23:57:33.154333 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_9078.caffemodel I0409 23:57:33.804682 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_9078.solverstate I0409 23:57:34.300505 4221 solver.cpp:330] Iteration 9078, Testing net (#0) I0409 23:57:34.300534 4221 net.cpp:676] Ignoring source layer train-data I0409 23:57:35.227706 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:57:39.081130 4221 solver.cpp:397] Test net output #0: accuracy = 0.438726 I0409 23:57:39.081280 4221 solver.cpp:397] Test net output #1: loss = 3.62645 (* 1 = 3.62645 loss) I0409 23:57:40.867200 4221 solver.cpp:218] Iteration 9084 (1.23808 iter/s, 9.69243s/12 iters), loss = 0.0668187 I0409 23:57:40.867251 4221 solver.cpp:237] Train net output #0: loss = 0.0668187 (* 1 = 0.0668187 loss) I0409 23:57:40.867264 4221 sgd_solver.cpp:105] Iteration 9084, lr = 0.00165396 I0409 23:57:45.805615 4221 solver.cpp:218] Iteration 9096 (2.43006 iter/s, 4.93815s/12 iters), loss = 0.102859 I0409 23:57:45.805668 4221 solver.cpp:237] Train net output #0: loss = 0.102859 (* 1 = 0.102859 loss) I0409 23:57:45.805680 4221 sgd_solver.cpp:105] Iteration 9096, lr = 0.00165003 I0409 23:57:48.674607 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:57:50.700402 4221 solver.cpp:218] Iteration 9108 (2.45172 iter/s, 4.89452s/12 iters), loss = 0.122066 I0409 23:57:50.700453 4221 solver.cpp:237] Train net output #0: loss = 0.122066 (* 1 = 0.122066 loss) I0409 23:57:50.700464 4221 sgd_solver.cpp:105] Iteration 9108, lr = 0.00164612 I0409 23:57:55.608347 4221 solver.cpp:218] Iteration 9120 (2.44515 iter/s, 4.90768s/12 iters), loss = 0.149859 I0409 23:57:55.608402 4221 solver.cpp:237] Train net output #0: loss = 0.149859 (* 1 = 0.149859 loss) I0409 23:57:55.608413 4221 sgd_solver.cpp:105] Iteration 9120, lr = 0.00164221 I0409 23:58:00.514961 4221 solver.cpp:218] Iteration 9132 (2.44581 iter/s, 4.90635s/12 iters), loss = 0.134589 I0409 23:58:00.515010 4221 solver.cpp:237] Train net output #0: loss = 0.134589 (* 1 = 0.134589 loss) I0409 23:58:00.515022 4221 sgd_solver.cpp:105] Iteration 9132, lr = 0.00163831 I0409 23:58:05.539302 4221 solver.cpp:218] Iteration 9144 (2.3885 iter/s, 5.02407s/12 iters), loss = 0.107897 I0409 23:58:05.539355 4221 solver.cpp:237] Train net output #0: loss = 0.107897 (* 1 = 0.107897 loss) I0409 23:58:05.539366 4221 sgd_solver.cpp:105] Iteration 9144, lr = 0.00163442 I0409 23:58:10.428069 4221 solver.cpp:218] Iteration 9156 (2.45474 iter/s, 4.8885s/12 iters), loss = 0.137261 I0409 23:58:10.428220 4221 solver.cpp:237] Train net output #0: loss = 0.137261 (* 1 = 0.137261 loss) I0409 23:58:10.428234 4221 sgd_solver.cpp:105] Iteration 9156, lr = 0.00163054 I0409 23:58:15.419857 4221 solver.cpp:218] Iteration 9168 (2.40412 iter/s, 4.99142s/12 iters), loss = 0.10382 I0409 23:58:15.419919 4221 solver.cpp:237] Train net output #0: loss = 0.10382 (* 1 = 0.10382 loss) I0409 23:58:15.419930 4221 sgd_solver.cpp:105] Iteration 9168, lr = 0.00162667 I0409 23:58:20.034112 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_9180.caffemodel I0409 23:58:20.849350 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_9180.solverstate I0409 23:58:22.425833 4221 solver.cpp:330] Iteration 9180, Testing net (#0) I0409 23:58:22.425858 4221 net.cpp:676] Ignoring source layer train-data I0409 23:58:23.296926 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:58:26.872903 4221 solver.cpp:397] Test net output #0: accuracy = 0.44424 I0409 23:58:26.872953 4221 solver.cpp:397] Test net output #1: loss = 3.59802 (* 1 = 3.59802 loss) I0409 23:58:26.956080 4221 solver.cpp:218] Iteration 9180 (1.04025 iter/s, 11.5357s/12 iters), loss = 0.109185 I0409 23:58:26.956136 4221 solver.cpp:237] Train net output #0: loss = 0.109185 (* 1 = 0.109185 loss) I0409 23:58:26.956147 4221 sgd_solver.cpp:105] Iteration 9180, lr = 0.00162281 I0409 23:58:30.982887 4221 solver.cpp:218] Iteration 9192 (2.9802 iter/s, 4.02657s/12 iters), loss = 0.150417 I0409 23:58:30.982949 4221 solver.cpp:237] Train net output #0: loss = 0.150417 (* 1 = 0.150417 loss) I0409 23:58:30.982962 4221 sgd_solver.cpp:105] Iteration 9192, lr = 0.00161895 I0409 23:58:35.979480 4221 solver.cpp:218] Iteration 9204 (2.40177 iter/s, 4.99631s/12 iters), loss = 0.0683024 I0409 23:58:35.979539 4221 solver.cpp:237] Train net output #0: loss = 0.0683024 (* 1 = 0.0683024 loss) I0409 23:58:35.979552 4221 sgd_solver.cpp:105] Iteration 9204, lr = 0.00161511 I0409 23:58:36.047503 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:58:40.779458 4221 solver.cpp:218] Iteration 9216 (2.50015 iter/s, 4.79972s/12 iters), loss = 0.108963 I0409 23:58:40.779565 4221 solver.cpp:237] Train net output #0: loss = 0.108963 (* 1 = 0.108963 loss) I0409 23:58:40.779577 4221 sgd_solver.cpp:105] Iteration 9216, lr = 0.00161128 I0409 23:58:45.546072 4221 solver.cpp:218] Iteration 9228 (2.51768 iter/s, 4.7663s/12 iters), loss = 0.0625652 I0409 23:58:45.546135 4221 solver.cpp:237] Train net output #0: loss = 0.0625653 (* 1 = 0.0625653 loss) I0409 23:58:45.546147 4221 sgd_solver.cpp:105] Iteration 9228, lr = 0.00160745 I0409 23:58:50.452160 4221 solver.cpp:218] Iteration 9240 (2.44608 iter/s, 4.90581s/12 iters), loss = 0.182343 I0409 23:58:50.452208 4221 solver.cpp:237] Train net output #0: loss = 0.182343 (* 1 = 0.182343 loss) I0409 23:58:50.452216 4221 sgd_solver.cpp:105] Iteration 9240, lr = 0.00160363 I0409 23:58:55.500360 4221 solver.cpp:218] Iteration 9252 (2.37721 iter/s, 5.04794s/12 iters), loss = 0.126206 I0409 23:58:55.500414 4221 solver.cpp:237] Train net output #0: loss = 0.126206 (* 1 = 0.126206 loss) I0409 23:58:55.500425 4221 sgd_solver.cpp:105] Iteration 9252, lr = 0.00159983 I0409 23:59:00.410024 4221 solver.cpp:218] Iteration 9264 (2.44429 iter/s, 4.90939s/12 iters), loss = 0.124499 I0409 23:59:00.410089 4221 solver.cpp:237] Train net output #0: loss = 0.124499 (* 1 = 0.124499 loss) I0409 23:59:00.410105 4221 sgd_solver.cpp:105] Iteration 9264, lr = 0.00159603 I0409 23:59:05.335500 4221 solver.cpp:218] Iteration 9276 (2.43645 iter/s, 4.9252s/12 iters), loss = 0.0959589 I0409 23:59:05.335561 4221 solver.cpp:237] Train net output #0: loss = 0.0959589 (* 1 = 0.0959589 loss) I0409 23:59:05.335572 4221 sgd_solver.cpp:105] Iteration 9276, lr = 0.00159224 I0409 23:59:07.490468 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_9282.caffemodel I0409 23:59:08.866367 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_9282.solverstate I0409 23:59:10.517010 4221 solver.cpp:330] Iteration 9282, Testing net (#0) I0409 23:59:10.517035 4221 net.cpp:676] Ignoring source layer train-data I0409 23:59:11.369675 4234 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:59:15.007179 4221 solver.cpp:397] Test net output #0: accuracy = 0.450368 I0409 23:59:15.007225 4221 solver.cpp:397] Test net output #1: loss = 3.64085 (* 1 = 3.64085 loss) I0409 23:59:16.908634 4221 solver.cpp:218] Iteration 9288 (1.03693 iter/s, 11.5726s/12 iters), loss = 0.0421764 I0409 23:59:16.908689 4221 solver.cpp:237] Train net output #0: loss = 0.0421765 (* 1 = 0.0421765 loss) I0409 23:59:16.908699 4221 sgd_solver.cpp:105] Iteration 9288, lr = 0.00158846 I0409 23:59:21.904031 4221 solver.cpp:218] Iteration 9300 (2.40234 iter/s, 4.99513s/12 iters), loss = 0.0564887 I0409 23:59:21.904076 4221 solver.cpp:237] Train net output #0: loss = 0.0564887 (* 1 = 0.0564887 loss) I0409 23:59:21.904086 4221 sgd_solver.cpp:105] Iteration 9300, lr = 0.00158469 I0409 23:59:24.034085 4233 data_layer.cpp:73] Restarting data prefetching from start. I0409 23:59:26.724756 4221 solver.cpp:218] Iteration 9312 (2.48939 iter/s, 4.82046s/12 iters), loss = 0.0818096 I0409 23:59:26.724817 4221 solver.cpp:237] Train net output #0: loss = 0.0818096 (* 1 = 0.0818096 loss) I0409 23:59:26.724829 4221 sgd_solver.cpp:105] Iteration 9312, lr = 0.00158092 I0409 23:59:31.551589 4221 solver.cpp:218] Iteration 9324 (2.48624 iter/s, 4.82656s/12 iters), loss = 0.0558225 I0409 23:59:31.551643 4221 solver.cpp:237] Train net output #0: loss = 0.0558225 (* 1 = 0.0558225 loss) I0409 23:59:31.551656 4221 sgd_solver.cpp:105] Iteration 9324, lr = 0.00157717 I0409 23:59:36.423911 4221 solver.cpp:218] Iteration 9336 (2.46302 iter/s, 4.87206s/12 iters), loss = 0.0815454 I0409 23:59:36.423961 4221 solver.cpp:237] Train net output #0: loss = 0.0815454 (* 1 = 0.0815454 loss) I0409 23:59:36.423974 4221 sgd_solver.cpp:105] Iteration 9336, lr = 0.00157343 I0409 23:59:41.305264 4221 solver.cpp:218] Iteration 9348 (2.45847 iter/s, 4.88109s/12 iters), loss = 0.221949 I0409 23:59:41.305318 4221 solver.cpp:237] Train net output #0: loss = 0.221949 (* 1 = 0.221949 loss) I0409 23:59:41.305330 4221 sgd_solver.cpp:105] Iteration 9348, lr = 0.00156969 I0409 23:59:46.184140 4221 solver.cpp:218] Iteration 9360 (2.45972 iter/s, 4.87861s/12 iters), loss = 0.0917783 I0409 23:59:46.184252 4221 solver.cpp:237] Train net output #0: loss = 0.0917784 (* 1 = 0.0917784 loss) I0409 23:59:46.184263 4221 sgd_solver.cpp:105] Iteration 9360, lr = 0.00156596 I0409 23:59:51.033110 4221 solver.cpp:218] Iteration 9372 (2.47491 iter/s, 4.84865s/12 iters), loss = 0.0737103 I0409 23:59:51.033159 4221 solver.cpp:237] Train net output #0: loss = 0.0737104 (* 1 = 0.0737104 loss) I0409 23:59:51.033171 4221 sgd_solver.cpp:105] Iteration 9372, lr = 0.00156225 I0409 23:59:55.398255 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_9384.caffemodel I0409 23:59:56.075104 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_9384.solverstate I0409 23:59:56.595118 4221 solver.cpp:330] Iteration 9384, Testing net (#0) I0409 23:59:56.595137 4221 net.cpp:676] Ignoring source layer train-data I0409 23:59:57.287099 4234 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:00:00.966595 4221 solver.cpp:397] Test net output #0: accuracy = 0.453431 I0410 00:00:00.966641 4221 solver.cpp:397] Test net output #1: loss = 3.59052 (* 1 = 3.59052 loss) I0410 00:00:01.049908 4221 solver.cpp:218] Iteration 9384 (1.19804 iter/s, 10.0163s/12 iters), loss = 0.139603 I0410 00:00:01.049981 4221 solver.cpp:237] Train net output #0: loss = 0.139603 (* 1 = 0.139603 loss) I0410 00:00:01.049993 4221 sgd_solver.cpp:105] Iteration 9384, lr = 0.00155854 I0410 00:00:05.173460 4221 solver.cpp:218] Iteration 9396 (2.91028 iter/s, 4.12332s/12 iters), loss = 0.0647017 I0410 00:00:05.173509 4221 solver.cpp:237] Train net output #0: loss = 0.0647018 (* 1 = 0.0647018 loss) I0410 00:00:05.173518 4221 sgd_solver.cpp:105] Iteration 9396, lr = 0.00155484 I0410 00:00:09.425043 4233 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:00:10.068032 4221 solver.cpp:218] Iteration 9408 (2.45183 iter/s, 4.8943s/12 iters), loss = 0.0650173 I0410 00:00:10.068089 4221 solver.cpp:237] Train net output #0: loss = 0.0650174 (* 1 = 0.0650174 loss) I0410 00:00:10.068099 4221 sgd_solver.cpp:105] Iteration 9408, lr = 0.00155114 I0410 00:00:15.133400 4221 solver.cpp:218] Iteration 9420 (2.36916 iter/s, 5.06509s/12 iters), loss = 0.132631 I0410 00:00:15.133456 4221 solver.cpp:237] Train net output #0: loss = 0.132631 (* 1 = 0.132631 loss) I0410 00:00:15.133469 4221 sgd_solver.cpp:105] Iteration 9420, lr = 0.00154746 I0410 00:00:19.921399 4221 solver.cpp:218] Iteration 9432 (2.5064 iter/s, 4.78773s/12 iters), loss = 0.0900565 I0410 00:00:19.921523 4221 solver.cpp:237] Train net output #0: loss = 0.0900566 (* 1 = 0.0900566 loss) I0410 00:00:19.921536 4221 sgd_solver.cpp:105] Iteration 9432, lr = 0.00154379 I0410 00:00:24.763831 4221 solver.cpp:218] Iteration 9444 (2.47826 iter/s, 4.8421s/12 iters), loss = 0.0325093 I0410 00:00:24.763882 4221 solver.cpp:237] Train net output #0: loss = 0.0325094 (* 1 = 0.0325094 loss) I0410 00:00:24.763893 4221 sgd_solver.cpp:105] Iteration 9444, lr = 0.00154012 I0410 00:00:29.780143 4221 solver.cpp:218] Iteration 9456 (2.39233 iter/s, 5.01604s/12 iters), loss = 0.0414006 I0410 00:00:29.780200 4221 solver.cpp:237] Train net output #0: loss = 0.0414007 (* 1 = 0.0414007 loss) I0410 00:00:29.780212 4221 sgd_solver.cpp:105] Iteration 9456, lr = 0.00153647 I0410 00:00:34.571266 4221 solver.cpp:218] Iteration 9468 (2.50477 iter/s, 4.79086s/12 iters), loss = 0.129873 I0410 00:00:34.571307 4221 solver.cpp:237] Train net output #0: loss = 0.129873 (* 1 = 0.129873 loss) I0410 00:00:34.571316 4221 sgd_solver.cpp:105] Iteration 9468, lr = 0.00153282 I0410 00:00:39.437999 4221 solver.cpp:218] Iteration 9480 (2.46587 iter/s, 4.86644s/12 iters), loss = 0.0360533 I0410 00:00:39.438066 4221 solver.cpp:237] Train net output #0: loss = 0.0360534 (* 1 = 0.0360534 loss) I0410 00:00:39.438082 4221 sgd_solver.cpp:105] Iteration 9480, lr = 0.00152918 I0410 00:00:41.446321 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_9486.caffemodel I0410 00:00:42.137645 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_9486.solverstate I0410 00:00:42.619952 4221 solver.cpp:330] Iteration 9486, Testing net (#0) I0410 00:00:42.619976 4221 net.cpp:676] Ignoring source layer train-data I0410 00:00:43.361001 4234 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:00:47.134603 4221 solver.cpp:397] Test net output #0: accuracy = 0.456495 I0410 00:00:47.134652 4221 solver.cpp:397] Test net output #1: loss = 3.61173 (* 1 = 3.61173 loss) I0410 00:00:49.048877 4221 solver.cpp:218] Iteration 9492 (1.24865 iter/s, 9.61041s/12 iters), loss = 0.182035 I0410 00:00:49.048929 4221 solver.cpp:237] Train net output #0: loss = 0.182035 (* 1 = 0.182035 loss) I0410 00:00:49.048941 4221 sgd_solver.cpp:105] Iteration 9492, lr = 0.00152555 I0410 00:00:53.893874 4221 solver.cpp:218] Iteration 9504 (2.47692 iter/s, 4.84473s/12 iters), loss = 0.0339427 I0410 00:00:53.894008 4221 solver.cpp:237] Train net output #0: loss = 0.0339428 (* 1 = 0.0339428 loss) I0410 00:00:53.894021 4221 sgd_solver.cpp:105] Iteration 9504, lr = 0.00152193 I0410 00:00:55.299417 4233 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:00:58.675858 4221 solver.cpp:218] Iteration 9516 (2.50959 iter/s, 4.78165s/12 iters), loss = 0.115617 I0410 00:00:58.675910 4221 solver.cpp:237] Train net output #0: loss = 0.115618 (* 1 = 0.115618 loss) I0410 00:00:58.675921 4221 sgd_solver.cpp:105] Iteration 9516, lr = 0.00151831 I0410 00:01:03.464558 4221 solver.cpp:218] Iteration 9528 (2.50604 iter/s, 4.78844s/12 iters), loss = 0.0746661 I0410 00:01:03.464618 4221 solver.cpp:237] Train net output #0: loss = 0.0746662 (* 1 = 0.0746662 loss) I0410 00:01:03.464632 4221 sgd_solver.cpp:105] Iteration 9528, lr = 0.00151471 I0410 00:01:08.230171 4221 solver.cpp:218] Iteration 9540 (2.51818 iter/s, 4.76534s/12 iters), loss = 0.103624 I0410 00:01:08.230237 4221 solver.cpp:237] Train net output #0: loss = 0.103624 (* 1 = 0.103624 loss) I0410 00:01:08.230249 4221 sgd_solver.cpp:105] Iteration 9540, lr = 0.00151111 I0410 00:01:13.080766 4221 solver.cpp:218] Iteration 9552 (2.47406 iter/s, 4.85032s/12 iters), loss = 0.0885757 I0410 00:01:13.080808 4221 solver.cpp:237] Train net output #0: loss = 0.0885758 (* 1 = 0.0885758 loss) I0410 00:01:13.080816 4221 sgd_solver.cpp:105] Iteration 9552, lr = 0.00150752 I0410 00:01:17.934530 4221 solver.cpp:218] Iteration 9564 (2.47244 iter/s, 4.85351s/12 iters), loss = 0.130377 I0410 00:01:17.934587 4221 solver.cpp:237] Train net output #0: loss = 0.130377 (* 1 = 0.130377 loss) I0410 00:01:17.934599 4221 sgd_solver.cpp:105] Iteration 9564, lr = 0.00150395 I0410 00:01:22.743710 4221 solver.cpp:218] Iteration 9576 (2.49537 iter/s, 4.80891s/12 iters), loss = 0.0622642 I0410 00:01:22.743767 4221 solver.cpp:237] Train net output #0: loss = 0.0622643 (* 1 = 0.0622643 loss) I0410 00:01:22.743777 4221 sgd_solver.cpp:105] Iteration 9576, lr = 0.00150037 I0410 00:01:27.134626 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_9588.caffemodel I0410 00:01:29.885275 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_9588.solverstate I0410 00:01:31.877378 4221 solver.cpp:330] Iteration 9588, Testing net (#0) I0410 00:01:31.877401 4221 net.cpp:676] Ignoring source layer train-data I0410 00:01:32.572108 4234 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:01:36.447460 4221 solver.cpp:397] Test net output #0: accuracy = 0.446078 I0410 00:01:36.447494 4221 solver.cpp:397] Test net output #1: loss = 3.66892 (* 1 = 3.66892 loss) I0410 00:01:36.530702 4221 solver.cpp:218] Iteration 9588 (0.870425 iter/s, 13.7864s/12 iters), loss = 0.0465582 I0410 00:01:36.530745 4221 solver.cpp:237] Train net output #0: loss = 0.0465583 (* 1 = 0.0465583 loss) I0410 00:01:36.530755 4221 sgd_solver.cpp:105] Iteration 9588, lr = 0.00149681 I0410 00:01:40.739360 4221 solver.cpp:218] Iteration 9600 (2.85142 iter/s, 4.20843s/12 iters), loss = 0.0879691 I0410 00:01:40.739414 4221 solver.cpp:237] Train net output #0: loss = 0.0879692 (* 1 = 0.0879692 loss) I0410 00:01:40.739425 4221 sgd_solver.cpp:105] Iteration 9600, lr = 0.00149326 I0410 00:01:44.246325 4233 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:01:45.593322 4221 solver.cpp:218] Iteration 9612 (2.47234 iter/s, 4.8537s/12 iters), loss = 0.0927441 I0410 00:01:45.593376 4221 solver.cpp:237] Train net output #0: loss = 0.0927442 (* 1 = 0.0927442 loss) I0410 00:01:45.593387 4221 sgd_solver.cpp:105] Iteration 9612, lr = 0.00148971 I0410 00:01:50.482916 4221 solver.cpp:218] Iteration 9624 (2.45432 iter/s, 4.88933s/12 iters), loss = 0.105807 I0410 00:01:50.482964 4221 solver.cpp:237] Train net output #0: loss = 0.105807 (* 1 = 0.105807 loss) I0410 00:01:50.482973 4221 sgd_solver.cpp:105] Iteration 9624, lr = 0.00148618 I0410 00:01:55.365008 4221 solver.cpp:218] Iteration 9636 (2.45809 iter/s, 4.88183s/12 iters), loss = 0.165955 I0410 00:01:55.365061 4221 solver.cpp:237] Train net output #0: loss = 0.165955 (* 1 = 0.165955 loss) I0410 00:01:55.365072 4221 sgd_solver.cpp:105] Iteration 9636, lr = 0.00148265 I0410 00:02:00.394461 4221 solver.cpp:218] Iteration 9648 (2.38608 iter/s, 5.02918s/12 iters), loss = 0.112713 I0410 00:02:00.394621 4221 solver.cpp:237] Train net output #0: loss = 0.112713 (* 1 = 0.112713 loss) I0410 00:02:00.394634 4221 sgd_solver.cpp:105] Iteration 9648, lr = 0.00147913 I0410 00:02:05.314888 4221 solver.cpp:218] Iteration 9660 (2.43899 iter/s, 4.92006s/12 iters), loss = 0.084957 I0410 00:02:05.314931 4221 solver.cpp:237] Train net output #0: loss = 0.0849572 (* 1 = 0.0849572 loss) I0410 00:02:05.314941 4221 sgd_solver.cpp:105] Iteration 9660, lr = 0.00147562 I0410 00:02:10.257932 4221 solver.cpp:218] Iteration 9672 (2.42778 iter/s, 4.94279s/12 iters), loss = 0.067867 I0410 00:02:10.258009 4221 solver.cpp:237] Train net output #0: loss = 0.0678672 (* 1 = 0.0678672 loss) I0410 00:02:10.258021 4221 sgd_solver.cpp:105] Iteration 9672, lr = 0.00147211 I0410 00:02:15.063364 4221 solver.cpp:218] Iteration 9684 (2.49732 iter/s, 4.80515s/12 iters), loss = 0.0964353 I0410 00:02:15.063419 4221 solver.cpp:237] Train net output #0: loss = 0.0964354 (* 1 = 0.0964354 loss) I0410 00:02:15.063432 4221 sgd_solver.cpp:105] Iteration 9684, lr = 0.00146862 I0410 00:02:17.043468 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_9690.caffemodel I0410 00:02:19.695065 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_9690.solverstate I0410 00:02:20.302333 4221 solver.cpp:330] Iteration 9690, Testing net (#0) I0410 00:02:20.302356 4221 net.cpp:676] Ignoring source layer train-data I0410 00:02:20.948364 4234 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:02:23.634032 4221 blocking_queue.cpp:49] Waiting for data I0410 00:02:24.888607 4221 solver.cpp:397] Test net output #0: accuracy = 0.44424 I0410 00:02:24.888636 4221 solver.cpp:397] Test net output #1: loss = 3.62057 (* 1 = 3.62057 loss) I0410 00:02:26.750525 4221 solver.cpp:218] Iteration 9696 (1.02682 iter/s, 11.6866s/12 iters), loss = 0.05923 I0410 00:02:26.750572 4221 solver.cpp:237] Train net output #0: loss = 0.0592301 (* 1 = 0.0592301 loss) I0410 00:02:26.750582 4221 sgd_solver.cpp:105] Iteration 9696, lr = 0.00146513 I0410 00:02:31.600821 4221 solver.cpp:218] Iteration 9708 (2.47421 iter/s, 4.85004s/12 iters), loss = 0.0408765 I0410 00:02:31.600914 4221 solver.cpp:237] Train net output #0: loss = 0.0408766 (* 1 = 0.0408766 loss) I0410 00:02:31.600924 4221 sgd_solver.cpp:105] Iteration 9708, lr = 0.00146165 I0410 00:02:32.376245 4233 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:02:36.531013 4221 solver.cpp:218] Iteration 9720 (2.43414 iter/s, 4.92988s/12 iters), loss = 0.0747396 I0410 00:02:36.531067 4221 solver.cpp:237] Train net output #0: loss = 0.0747397 (* 1 = 0.0747397 loss) I0410 00:02:36.531078 4221 sgd_solver.cpp:105] Iteration 9720, lr = 0.00145818 I0410 00:02:41.445327 4221 solver.cpp:218] Iteration 9732 (2.44198 iter/s, 4.91404s/12 iters), loss = 0.0592839 I0410 00:02:41.445374 4221 solver.cpp:237] Train net output #0: loss = 0.0592841 (* 1 = 0.0592841 loss) I0410 00:02:41.445384 4221 sgd_solver.cpp:105] Iteration 9732, lr = 0.00145472 I0410 00:02:46.267382 4221 solver.cpp:218] Iteration 9744 (2.4887 iter/s, 4.8218s/12 iters), loss = 0.0359808 I0410 00:02:46.267431 4221 solver.cpp:237] Train net output #0: loss = 0.035981 (* 1 = 0.035981 loss) I0410 00:02:46.267441 4221 sgd_solver.cpp:105] Iteration 9744, lr = 0.00145127 I0410 00:02:51.121122 4221 solver.cpp:218] Iteration 9756 (2.47245 iter/s, 4.85348s/12 iters), loss = 0.169189 I0410 00:02:51.121167 4221 solver.cpp:237] Train net output #0: loss = 0.16919 (* 1 = 0.16919 loss) I0410 00:02:51.121177 4221 sgd_solver.cpp:105] Iteration 9756, lr = 0.00144782 I0410 00:02:56.104146 4221 solver.cpp:218] Iteration 9768 (2.4083 iter/s, 4.98276s/12 iters), loss = 0.0548992 I0410 00:02:56.104202 4221 solver.cpp:237] Train net output #0: loss = 0.0548994 (* 1 = 0.0548994 loss) I0410 00:02:56.104212 4221 sgd_solver.cpp:105] Iteration 9768, lr = 0.00144438 I0410 00:03:00.895505 4221 solver.cpp:218] Iteration 9780 (2.50465 iter/s, 4.7911s/12 iters), loss = 0.063427 I0410 00:03:00.895552 4221 solver.cpp:237] Train net output #0: loss = 0.0634272 (* 1 = 0.0634272 loss) I0410 00:03:00.895561 4221 sgd_solver.cpp:105] Iteration 9780, lr = 0.00144095 I0410 00:03:05.302227 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_9792.caffemodel I0410 00:03:06.319083 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_9792.solverstate I0410 00:03:07.179646 4221 solver.cpp:330] Iteration 9792, Testing net (#0) I0410 00:03:07.179668 4221 net.cpp:676] Ignoring source layer train-data I0410 00:03:07.680737 4234 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:03:11.485545 4221 solver.cpp:397] Test net output #0: accuracy = 0.455882 I0410 00:03:11.485594 4221 solver.cpp:397] Test net output #1: loss = 3.61416 (* 1 = 3.61416 loss) I0410 00:03:11.569041 4221 solver.cpp:218] Iteration 9792 (1.12433 iter/s, 10.673s/12 iters), loss = 0.0234258 I0410 00:03:11.569087 4221 solver.cpp:237] Train net output #0: loss = 0.023426 (* 1 = 0.023426 loss) I0410 00:03:11.569098 4221 sgd_solver.cpp:105] Iteration 9792, lr = 0.00143753 I0410 00:03:15.760047 4221 solver.cpp:218] Iteration 9804 (2.86343 iter/s, 4.19077s/12 iters), loss = 0.151635 I0410 00:03:15.760108 4221 solver.cpp:237] Train net output #0: loss = 0.151635 (* 1 = 0.151635 loss) I0410 00:03:15.760120 4221 sgd_solver.cpp:105] Iteration 9804, lr = 0.00143412 I0410 00:03:18.627348 4233 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:03:20.610623 4221 solver.cpp:218] Iteration 9816 (2.47407 iter/s, 4.8503s/12 iters), loss = 0.0177192 I0410 00:03:20.610683 4221 solver.cpp:237] Train net output #0: loss = 0.0177194 (* 1 = 0.0177194 loss) I0410 00:03:20.610697 4221 sgd_solver.cpp:105] Iteration 9816, lr = 0.00143072 I0410 00:03:25.432122 4221 solver.cpp:218] Iteration 9828 (2.48899 iter/s, 4.82123s/12 iters), loss = 0.03854 I0410 00:03:25.432173 4221 solver.cpp:237] Train net output #0: loss = 0.0385401 (* 1 = 0.0385401 loss) I0410 00:03:25.432183 4221 sgd_solver.cpp:105] Iteration 9828, lr = 0.00142732 I0410 00:03:30.286154 4221 solver.cpp:218] Iteration 9840 (2.4723 iter/s, 4.85377s/12 iters), loss = 0.0366895 I0410 00:03:30.286195 4221 solver.cpp:237] Train net output #0: loss = 0.0366897 (* 1 = 0.0366897 loss) I0410 00:03:30.286204 4221 sgd_solver.cpp:105] Iteration 9840, lr = 0.00142393 I0410 00:03:35.154664 4221 solver.cpp:218] Iteration 9852 (2.46495 iter/s, 4.86825s/12 iters), loss = 0.068602 I0410 00:03:35.154726 4221 solver.cpp:237] Train net output #0: loss = 0.0686022 (* 1 = 0.0686022 loss) I0410 00:03:35.154737 4221 sgd_solver.cpp:105] Iteration 9852, lr = 0.00142055 I0410 00:03:40.032099 4221 solver.cpp:218] Iteration 9864 (2.46045 iter/s, 4.87717s/12 iters), loss = 0.0598735 I0410 00:03:40.032218 4221 solver.cpp:237] Train net output #0: loss = 0.0598737 (* 1 = 0.0598737 loss) I0410 00:03:40.032230 4221 sgd_solver.cpp:105] Iteration 9864, lr = 0.00141718 I0410 00:03:44.896708 4221 solver.cpp:218] Iteration 9876 (2.46696 iter/s, 4.86428s/12 iters), loss = 0.0679954 I0410 00:03:44.896762 4221 solver.cpp:237] Train net output #0: loss = 0.0679955 (* 1 = 0.0679955 loss) I0410 00:03:44.896775 4221 sgd_solver.cpp:105] Iteration 9876, lr = 0.00141381 I0410 00:03:49.765491 4221 solver.cpp:218] Iteration 9888 (2.46481 iter/s, 4.86852s/12 iters), loss = 0.087371 I0410 00:03:49.765533 4221 solver.cpp:237] Train net output #0: loss = 0.0873711 (* 1 = 0.0873711 loss) I0410 00:03:49.765542 4221 sgd_solver.cpp:105] Iteration 9888, lr = 0.00141045 I0410 00:03:51.773068 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_9894.caffemodel I0410 00:03:52.695927 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_9894.solverstate I0410 00:03:53.658159 4221 solver.cpp:330] Iteration 9894, Testing net (#0) I0410 00:03:53.658186 4221 net.cpp:676] Ignoring source layer train-data I0410 00:03:54.166380 4234 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:03:58.008605 4221 solver.cpp:397] Test net output #0: accuracy = 0.454044 I0410 00:03:58.008652 4221 solver.cpp:397] Test net output #1: loss = 3.66783 (* 1 = 3.66783 loss) I0410 00:03:59.863652 4221 solver.cpp:218] Iteration 9900 (1.18839 iter/s, 10.0977s/12 iters), loss = 0.132628 I0410 00:03:59.863700 4221 solver.cpp:237] Train net output #0: loss = 0.132628 (* 1 = 0.132628 loss) I0410 00:03:59.863709 4221 sgd_solver.cpp:105] Iteration 9900, lr = 0.00140711 I0410 00:04:04.740722 4221 solver.cpp:218] Iteration 9912 (2.46063 iter/s, 4.87681s/12 iters), loss = 0.0280204 I0410 00:04:04.740772 4221 solver.cpp:237] Train net output #0: loss = 0.0280205 (* 1 = 0.0280205 loss) I0410 00:04:04.740782 4221 sgd_solver.cpp:105] Iteration 9912, lr = 0.00140377 I0410 00:04:04.838805 4233 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:04:09.745388 4221 solver.cpp:218] Iteration 9924 (2.39789 iter/s, 5.0044s/12 iters), loss = 0.08367 I0410 00:04:09.745434 4221 solver.cpp:237] Train net output #0: loss = 0.0836702 (* 1 = 0.0836702 loss) I0410 00:04:09.745445 4221 sgd_solver.cpp:105] Iteration 9924, lr = 0.00140043 I0410 00:04:14.571642 4221 solver.cpp:218] Iteration 9936 (2.48653 iter/s, 4.826s/12 iters), loss = 0.0964301 I0410 00:04:14.571772 4221 solver.cpp:237] Train net output #0: loss = 0.0964303 (* 1 = 0.0964303 loss) I0410 00:04:14.571784 4221 sgd_solver.cpp:105] Iteration 9936, lr = 0.00139711 I0410 00:04:19.425371 4221 solver.cpp:218] Iteration 9948 (2.4725 iter/s, 4.85339s/12 iters), loss = 0.105331 I0410 00:04:19.425415 4221 solver.cpp:237] Train net output #0: loss = 0.105331 (* 1 = 0.105331 loss) I0410 00:04:19.425423 4221 sgd_solver.cpp:105] Iteration 9948, lr = 0.00139379 I0410 00:04:24.255009 4221 solver.cpp:218] Iteration 9960 (2.48479 iter/s, 4.82938s/12 iters), loss = 0.103335 I0410 00:04:24.255057 4221 solver.cpp:237] Train net output #0: loss = 0.103336 (* 1 = 0.103336 loss) I0410 00:04:24.255066 4221 sgd_solver.cpp:105] Iteration 9960, lr = 0.00139048 I0410 00:04:29.162122 4221 solver.cpp:218] Iteration 9972 (2.44556 iter/s, 4.90685s/12 iters), loss = 0.0728573 I0410 00:04:29.162170 4221 solver.cpp:237] Train net output #0: loss = 0.0728574 (* 1 = 0.0728574 loss) I0410 00:04:29.162179 4221 sgd_solver.cpp:105] Iteration 9972, lr = 0.00138718 I0410 00:04:34.031471 4221 solver.cpp:218] Iteration 9984 (2.46453 iter/s, 4.86909s/12 iters), loss = 0.108933 I0410 00:04:34.031518 4221 solver.cpp:237] Train net output #0: loss = 0.108933 (* 1 = 0.108933 loss) I0410 00:04:34.031527 4221 sgd_solver.cpp:105] Iteration 9984, lr = 0.00138389 I0410 00:04:38.580258 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_9996.caffemodel I0410 00:04:39.660601 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_9996.solverstate I0410 00:04:40.511588 4221 solver.cpp:330] Iteration 9996, Testing net (#0) I0410 00:04:40.511618 4221 net.cpp:676] Ignoring source layer train-data I0410 00:04:41.008662 4234 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:04:44.932246 4221 solver.cpp:397] Test net output #0: accuracy = 0.458333 I0410 00:04:44.932366 4221 solver.cpp:397] Test net output #1: loss = 3.64024 (* 1 = 3.64024 loss) I0410 00:04:45.015547 4221 solver.cpp:218] Iteration 9996 (1.09254 iter/s, 10.9836s/12 iters), loss = 0.0388397 I0410 00:04:45.015588 4221 solver.cpp:237] Train net output #0: loss = 0.0388398 (* 1 = 0.0388398 loss) I0410 00:04:45.015599 4221 sgd_solver.cpp:105] Iteration 9996, lr = 0.0013806 I0410 00:04:49.301013 4221 solver.cpp:218] Iteration 10008 (2.80031 iter/s, 4.28524s/12 iters), loss = 0.14537 I0410 00:04:49.301055 4221 solver.cpp:237] Train net output #0: loss = 0.14537 (* 1 = 0.14537 loss) I0410 00:04:49.301066 4221 sgd_solver.cpp:105] Iteration 10008, lr = 0.00137732 I0410 00:04:51.486532 4233 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:04:54.126907 4221 solver.cpp:218] Iteration 10020 (2.48672 iter/s, 4.82564s/12 iters), loss = 0.0337345 I0410 00:04:54.126961 4221 solver.cpp:237] Train net output #0: loss = 0.0337346 (* 1 = 0.0337346 loss) I0410 00:04:54.126973 4221 sgd_solver.cpp:105] Iteration 10020, lr = 0.00137405 I0410 00:04:58.967103 4221 solver.cpp:218] Iteration 10032 (2.47937 iter/s, 4.83993s/12 iters), loss = 0.0361255 I0410 00:04:58.967151 4221 solver.cpp:237] Train net output #0: loss = 0.0361256 (* 1 = 0.0361256 loss) I0410 00:04:58.967161 4221 sgd_solver.cpp:105] Iteration 10032, lr = 0.00137079 I0410 00:05:03.787003 4221 solver.cpp:218] Iteration 10044 (2.48981 iter/s, 4.81964s/12 iters), loss = 0.127606 I0410 00:05:03.787055 4221 solver.cpp:237] Train net output #0: loss = 0.127606 (* 1 = 0.127606 loss) I0410 00:05:03.787065 4221 sgd_solver.cpp:105] Iteration 10044, lr = 0.00136754 I0410 00:05:08.828894 4221 solver.cpp:218] Iteration 10056 (2.38019 iter/s, 5.04162s/12 iters), loss = 0.040962 I0410 00:05:08.828950 4221 solver.cpp:237] Train net output #0: loss = 0.0409622 (* 1 = 0.0409622 loss) I0410 00:05:08.828963 4221 sgd_solver.cpp:105] Iteration 10056, lr = 0.00136429 I0410 00:05:13.620505 4221 solver.cpp:218] Iteration 10068 (2.50451 iter/s, 4.79135s/12 iters), loss = 0.0978475 I0410 00:05:13.620551 4221 solver.cpp:237] Train net output #0: loss = 0.0978476 (* 1 = 0.0978476 loss) I0410 00:05:13.620561 4221 sgd_solver.cpp:105] Iteration 10068, lr = 0.00136105 I0410 00:05:18.418433 4221 solver.cpp:218] Iteration 10080 (2.50122 iter/s, 4.79767s/12 iters), loss = 0.0410353 I0410 00:05:18.418620 4221 solver.cpp:237] Train net output #0: loss = 0.0410355 (* 1 = 0.0410355 loss) I0410 00:05:18.418635 4221 sgd_solver.cpp:105] Iteration 10080, lr = 0.00135782 I0410 00:05:23.256103 4221 solver.cpp:218] Iteration 10092 (2.48074 iter/s, 4.83727s/12 iters), loss = 0.0291979 I0410 00:05:23.256157 4221 solver.cpp:237] Train net output #0: loss = 0.029198 (* 1 = 0.029198 loss) I0410 00:05:23.256168 4221 sgd_solver.cpp:105] Iteration 10092, lr = 0.0013546 I0410 00:05:25.211663 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_10098.caffemodel I0410 00:05:26.201364 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_10098.solverstate I0410 00:05:27.498437 4221 solver.cpp:330] Iteration 10098, Testing net (#0) I0410 00:05:27.498461 4221 net.cpp:676] Ignoring source layer train-data I0410 00:05:27.958642 4234 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:05:32.184610 4221 solver.cpp:397] Test net output #0: accuracy = 0.449142 I0410 00:05:32.184659 4221 solver.cpp:397] Test net output #1: loss = 3.66135 (* 1 = 3.66135 loss) I0410 00:05:34.033543 4221 solver.cpp:218] Iteration 10104 (1.11349 iter/s, 10.7769s/12 iters), loss = 0.0453892 I0410 00:05:34.033589 4221 solver.cpp:237] Train net output #0: loss = 0.0453893 (* 1 = 0.0453893 loss) I0410 00:05:34.033598 4221 sgd_solver.cpp:105] Iteration 10104, lr = 0.00135138 I0410 00:05:38.240303 4233 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:05:38.846437 4221 solver.cpp:218] Iteration 10116 (2.49343 iter/s, 4.81264s/12 iters), loss = 0.195028 I0410 00:05:38.846484 4221 solver.cpp:237] Train net output #0: loss = 0.195028 (* 1 = 0.195028 loss) I0410 00:05:38.846496 4221 sgd_solver.cpp:105] Iteration 10116, lr = 0.00134817 I0410 00:05:43.693703 4221 solver.cpp:218] Iteration 10128 (2.47576 iter/s, 4.847s/12 iters), loss = 0.098644 I0410 00:05:43.693760 4221 solver.cpp:237] Train net output #0: loss = 0.0986442 (* 1 = 0.0986442 loss) I0410 00:05:43.693773 4221 sgd_solver.cpp:105] Iteration 10128, lr = 0.00134497 I0410 00:05:48.508123 4221 solver.cpp:218] Iteration 10140 (2.49265 iter/s, 4.81416s/12 iters), loss = 0.0416593 I0410 00:05:48.508224 4221 solver.cpp:237] Train net output #0: loss = 0.0416594 (* 1 = 0.0416594 loss) I0410 00:05:48.508234 4221 sgd_solver.cpp:105] Iteration 10140, lr = 0.00134178 I0410 00:05:53.347687 4221 solver.cpp:218] Iteration 10152 (2.47972 iter/s, 4.83925s/12 iters), loss = 0.0147403 I0410 00:05:53.347738 4221 solver.cpp:237] Train net output #0: loss = 0.0147405 (* 1 = 0.0147405 loss) I0410 00:05:53.347750 4221 sgd_solver.cpp:105] Iteration 10152, lr = 0.00133859 I0410 00:05:58.173467 4221 solver.cpp:218] Iteration 10164 (2.48678 iter/s, 4.82552s/12 iters), loss = 0.0319597 I0410 00:05:58.173513 4221 solver.cpp:237] Train net output #0: loss = 0.0319598 (* 1 = 0.0319598 loss) I0410 00:05:58.173522 4221 sgd_solver.cpp:105] Iteration 10164, lr = 0.00133541 I0410 00:06:03.062872 4221 solver.cpp:218] Iteration 10176 (2.45442 iter/s, 4.88914s/12 iters), loss = 0.0205677 I0410 00:06:03.062930 4221 solver.cpp:237] Train net output #0: loss = 0.0205679 (* 1 = 0.0205679 loss) I0410 00:06:03.062943 4221 sgd_solver.cpp:105] Iteration 10176, lr = 0.00133224 I0410 00:06:07.892206 4221 solver.cpp:218] Iteration 10188 (2.48495 iter/s, 4.82907s/12 iters), loss = 0.0382411 I0410 00:06:07.892249 4221 solver.cpp:237] Train net output #0: loss = 0.0382412 (* 1 = 0.0382412 loss) I0410 00:06:07.892258 4221 sgd_solver.cpp:105] Iteration 10188, lr = 0.00132908 I0410 00:06:12.353565 4221 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_10200.caffemodel I0410 00:06:13.642715 4221 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_10200.solverstate I0410 00:06:15.381127 4221 solver.cpp:310] Iteration 10200, loss = 0.106558 I0410 00:06:15.381162 4221 solver.cpp:330] Iteration 10200, Testing net (#0) I0410 00:06:15.381170 4221 net.cpp:676] Ignoring source layer train-data I0410 00:06:15.806259 4234 data_layer.cpp:73] Restarting data prefetching from start. I0410 00:06:19.898706 4221 solver.cpp:397] Test net output #0: accuracy = 0.452206 I0410 00:06:19.898815 4221 solver.cpp:397] Test net output #1: loss = 3.66147 (* 1 = 3.66147 loss) I0410 00:06:19.898821 4221 solver.cpp:315] Optimization Done. I0410 00:06:19.898824 4221 caffe.cpp:259] Optimization Done.