I0407 22:24:57.000056 359 upgrade_proto.cpp:1082] Attempting to upgrade input file specified using deprecated 'solver_type' field (enum)': /mnt/bigdisk/DIGITS-TAM-1/digits/jobs/20210407-222455-6e13/solver.prototxt I0407 22:24:57.000228 359 upgrade_proto.cpp:1089] Successfully upgraded file specified using deprecated 'solver_type' field (enum) to 'type' field (string). W0407 22:24:57.000236 359 upgrade_proto.cpp:1091] Note that future Caffe releases will only support 'type' field (string) for a solver's type. I0407 22:24:57.002756 359 caffe.cpp:218] Using GPUs 2 I0407 22:24:57.045344 359 caffe.cpp:223] GPU 2: GeForce RTX 2080 I0407 22:24:57.335467 359 solver.cpp:44] Initializing solver from parameters: test_iter: 51 test_interval: 102 base_lr: 0.01 display: 12 max_iter: 10200 lr_policy: "sigmoid" gamma: -0.0019607844 momentum: 0.9 weight_decay: 0.0001 stepsize: 5100 snapshot: 102 snapshot_prefix: "snapshot" solver_mode: GPU device_id: 2 net: "train_val.prototxt" train_state { level: 0 stage: "" } type: "SGD" I0407 22:24:57.336266 359 solver.cpp:87] Creating training net from net file: train_val.prototxt I0407 22:24:57.336860 359 net.cpp:294] The NetState phase (0) differed from the phase (1) specified by a rule in layer val-data I0407 22:24:57.336874 359 net.cpp:294] The NetState phase (0) differed from the phase (1) specified by a rule in layer accuracy I0407 22:24:57.337000 359 net.cpp:51] Initializing net from parameters: state { phase: TRAIN level: 0 stage: "" } layer { name: "train-data" type: "Data" top: "data" top: "label" include { phase: TRAIN } transform_param { mirror: true crop_size: 227 mean_file: "/mnt/bigdisk/DIGITS-TAM-1/digits/jobs/20210407-221849-d51c/mean.binaryproto" } data_param { source: "/mnt/bigdisk/DIGITS-TAM-1/digits/jobs/20210407-221849-d51c/train_db" batch_size: 128 backend: LMDB } } layer { name: "conv1" type: "Convolution" bottom: "data" top: "conv1" param { lr_mult: 1 decay_mult: 1 } param { lr_mult: 2 decay_mult: 0 } convolution_param { num_output: 96 kernel_size: 11 stride: 4 weight_filler { type: "gaussian" std: 0.01 } bias_filler { type: "constant" value: 0 } } } layer { name: "relu1" type: "ReLU" bottom: "conv1" top: "conv1" } layer { name: "norm1" type: "LRN" bottom: "conv1" top: "norm1" lrn_param { local_size: 5 alpha: 0.0001 beta: 0.75 } } layer { name: "pool1" type: "Pooling" bottom: "norm1" top: "pool1" pooling_param { pool: MAX kernel_size: 3 stride: 2 } } layer { name: "conv2" type: "Convolution" bottom: "pool1" top: "conv2" param { lr_mult: 1 decay_mult: 1 } param { lr_mult: 2 decay_mult: 0 } convolution_param { num_output: 256 pad: 2 kernel_size: 5 group: 2 weight_filler { type: "gaussian" std: 0.01 } bias_filler { type: "constant" value: 0.1 } } } layer { name: "relu2" type: "ReLU" bottom: "conv2" top: "conv2" } layer { name: "norm2" type: "LRN" bottom: "conv2" top: "norm2" lrn_param { local_size: 5 alpha: 0.0001 beta: 0.75 } } layer { name: "pool2" type: "Pooling" bottom: "norm2" top: "pool2" pooling_param { pool: MAX kernel_size: 3 stride: 2 } } layer { name: "conv3" type: "Convolution" bottom: "pool2" top: "conv3" param { lr_mult: 1 decay_mult: 1 } param { lr_mult: 2 decay_mult: 0 } convolution_param { num_output: 384 pad: 1 kernel_size: 3 weight_filler { type: "gaussian" std: 0.01 } bias_filler { type: "constant" value: 0 } } } layer { name: "relu3" type: "ReLU" bottom: "conv3" top: "conv3" } layer { name: "conv4" type: "Convolution" bottom: "conv3" top: "conv4" param { lr_mult: 1 decay_mult: 1 } param { lr_mult: 2 decay_mult: 0 } convolution_param { num_output: 384 pad: 1 kernel_size: 3 group: 2 weight_filler { type: "gaussian" std: 0.01 } bias_filler { type: "constant" value: 0.1 } } } layer { name: "relu4" type: "ReLU" bottom: "conv4" top: "conv4" } layer { name: "conv5" type: "Convolution" bottom: "conv4" top: "conv5" param { lr_mult: 1 decay_mult: 1 } param { lr_mult: 2 decay_mult: 0 } convolution_param { num_output: 256 pad: 1 kernel_size: 3 group: 2 weight_filler { type: "gaussian" std: 0.01 } bias_filler { type: "constant" value: 0.1 } } } layer { name: "relu5" type: "ReLU" bottom: "conv5" top: "conv5" } layer { name: "pool5" type: "Pooling" bottom: "conv5" top: "pool5" pooling_param { pool: MAX kernel_size: 3 stride: 2 } } layer { name: "fc6" type: "InnerProduct" bottom: "pool5" top: "fc6" param { lr_mult: 1 decay_mult: 1 } param { lr_mult: 2 decay_mult: 0 } inner_product_param { num_output: 4096 weight_filler { type: "gaussian" std: 0.005 } bias_filler { type: "constant" value: 0.1 } } } layer { name: "relu6" type: "ReLU" bottom: "fc6" top: "fc6" } layer { name: "drop6" type: "Dropout" bottom: "fc6" top: "fc6" dropout_param { dropout_ratio: 0.5 } } layer { name: "fc7" type: "InnerProduct" bottom: "fc6" top: "fc7" param { lr_mult: 1 decay_mult: 1 } param { lr_mult: 2 decay_mult: 0 } inner_product_param { num_output: 4096 weight_filler { type: "gaussian" std: 0.005 } bias_filler { type: "constant" value: 0.1 } } } layer { name: "relu7" type: "ReLU" bottom: "fc7" top: "fc7" } layer { name: "drop7" type: "Dropout" bottom: "fc7" top: "fc7" dropout_param { dropout_ratio: 0.5 } } layer { name: "fc8" type: "InnerProduct" bottom: "fc7" top: "fc8" param { lr_mult: 1 decay_mult: 1 } param { lr_mult: 2 decay_mult: 0 } inner_product_param { num_output: 196 weight_filler { type: "gaussian" std: 0.01 } bias_filler { type: "constant" value: 0 } } } layer { name: "loss" type: "SoftmaxWithLoss" bottom: "fc8" bottom: "label" top: "loss" } I0407 22:24:57.337081 359 layer_factory.hpp:77] Creating layer train-data I0407 22:24:57.339022 359 db_lmdb.cpp:35] Opened lmdb /mnt/bigdisk/DIGITS-TAM-1/digits/jobs/20210407-221849-d51c/train_db I0407 22:24:57.339668 359 net.cpp:84] Creating Layer train-data I0407 22:24:57.339680 359 net.cpp:380] train-data -> data I0407 22:24:57.339700 359 net.cpp:380] train-data -> label I0407 22:24:57.339711 359 data_transformer.cpp:25] Loading mean file from: /mnt/bigdisk/DIGITS-TAM-1/digits/jobs/20210407-221849-d51c/mean.binaryproto I0407 22:24:57.344272 359 data_layer.cpp:45] output data size: 128,3,227,227 I0407 22:24:57.471490 359 net.cpp:122] Setting up train-data I0407 22:24:57.471513 359 net.cpp:129] Top shape: 128 3 227 227 (19787136) I0407 22:24:57.471518 359 net.cpp:129] Top shape: 128 (128) I0407 22:24:57.471519 359 net.cpp:137] Memory required for data: 79149056 I0407 22:24:57.471529 359 layer_factory.hpp:77] Creating layer conv1 I0407 22:24:57.471549 359 net.cpp:84] Creating Layer conv1 I0407 22:24:57.471554 359 net.cpp:406] conv1 <- data I0407 22:24:57.471566 359 net.cpp:380] conv1 -> conv1 I0407 22:24:58.376600 359 net.cpp:122] Setting up conv1 I0407 22:24:58.376621 359 net.cpp:129] Top shape: 128 96 55 55 (37171200) I0407 22:24:58.376624 359 net.cpp:137] Memory required for data: 227833856 I0407 22:24:58.376643 359 layer_factory.hpp:77] Creating layer relu1 I0407 22:24:58.376653 359 net.cpp:84] Creating Layer relu1 I0407 22:24:58.376657 359 net.cpp:406] relu1 <- conv1 I0407 22:24:58.376662 359 net.cpp:367] relu1 -> conv1 (in-place) I0407 22:24:58.376996 359 net.cpp:122] Setting up relu1 I0407 22:24:58.377004 359 net.cpp:129] Top shape: 128 96 55 55 (37171200) I0407 22:24:58.377007 359 net.cpp:137] Memory required for data: 376518656 I0407 22:24:58.377010 359 layer_factory.hpp:77] Creating layer norm1 I0407 22:24:58.377019 359 net.cpp:84] Creating Layer norm1 I0407 22:24:58.377038 359 net.cpp:406] norm1 <- conv1 I0407 22:24:58.377043 359 net.cpp:380] norm1 -> norm1 I0407 22:24:58.377588 359 net.cpp:122] Setting up norm1 I0407 22:24:58.377597 359 net.cpp:129] Top shape: 128 96 55 55 (37171200) I0407 22:24:58.377600 359 net.cpp:137] Memory required for data: 525203456 I0407 22:24:58.377604 359 layer_factory.hpp:77] Creating layer pool1 I0407 22:24:58.377610 359 net.cpp:84] Creating Layer pool1 I0407 22:24:58.377614 359 net.cpp:406] pool1 <- norm1 I0407 22:24:58.377619 359 net.cpp:380] pool1 -> pool1 I0407 22:24:58.377650 359 net.cpp:122] Setting up pool1 I0407 22:24:58.377655 359 net.cpp:129] Top shape: 128 96 27 27 (8957952) I0407 22:24:58.377657 359 net.cpp:137] Memory required for data: 561035264 I0407 22:24:58.377660 359 layer_factory.hpp:77] Creating layer conv2 I0407 22:24:58.377669 359 net.cpp:84] Creating Layer conv2 I0407 22:24:58.377672 359 net.cpp:406] conv2 <- pool1 I0407 22:24:58.377676 359 net.cpp:380] conv2 -> conv2 I0407 22:24:58.388175 359 net.cpp:122] Setting up conv2 I0407 22:24:58.388195 359 net.cpp:129] Top shape: 128 256 27 27 (23887872) I0407 22:24:58.388200 359 net.cpp:137] Memory required for data: 656586752 I0407 22:24:58.388213 359 layer_factory.hpp:77] Creating layer relu2 I0407 22:24:58.388221 359 net.cpp:84] Creating Layer relu2 I0407 22:24:58.388226 359 net.cpp:406] relu2 <- conv2 I0407 22:24:58.388233 359 net.cpp:367] relu2 -> conv2 (in-place) I0407 22:24:58.388893 359 net.cpp:122] Setting up relu2 I0407 22:24:58.388906 359 net.cpp:129] Top shape: 128 256 27 27 (23887872) I0407 22:24:58.388911 359 net.cpp:137] Memory required for data: 752138240 I0407 22:24:58.388916 359 layer_factory.hpp:77] Creating layer norm2 I0407 22:24:58.388924 359 net.cpp:84] Creating Layer norm2 I0407 22:24:58.388929 359 net.cpp:406] norm2 <- conv2 I0407 22:24:58.388936 359 net.cpp:380] norm2 -> norm2 I0407 22:24:58.389379 359 net.cpp:122] Setting up norm2 I0407 22:24:58.389391 359 net.cpp:129] Top shape: 128 256 27 27 (23887872) I0407 22:24:58.389395 359 net.cpp:137] Memory required for data: 847689728 I0407 22:24:58.389400 359 layer_factory.hpp:77] Creating layer pool2 I0407 22:24:58.389408 359 net.cpp:84] Creating Layer pool2 I0407 22:24:58.389413 359 net.cpp:406] pool2 <- norm2 I0407 22:24:58.389420 359 net.cpp:380] pool2 -> pool2 I0407 22:24:58.389453 359 net.cpp:122] Setting up pool2 I0407 22:24:58.389459 359 net.cpp:129] Top shape: 128 256 13 13 (5537792) I0407 22:24:58.389463 359 net.cpp:137] Memory required for data: 869840896 I0407 22:24:58.389467 359 layer_factory.hpp:77] Creating layer conv3 I0407 22:24:58.389477 359 net.cpp:84] Creating Layer conv3 I0407 22:24:58.389482 359 net.cpp:406] conv3 <- pool2 I0407 22:24:58.389487 359 net.cpp:380] conv3 -> conv3 I0407 22:24:58.400347 359 net.cpp:122] Setting up conv3 I0407 22:24:58.400363 359 net.cpp:129] Top shape: 128 384 13 13 (8306688) I0407 22:24:58.400367 359 net.cpp:137] Memory required for data: 903067648 I0407 22:24:58.400375 359 layer_factory.hpp:77] Creating layer relu3 I0407 22:24:58.400382 359 net.cpp:84] Creating Layer relu3 I0407 22:24:58.400386 359 net.cpp:406] relu3 <- conv3 I0407 22:24:58.400391 359 net.cpp:367] relu3 -> conv3 (in-place) I0407 22:24:58.400890 359 net.cpp:122] Setting up relu3 I0407 22:24:58.400900 359 net.cpp:129] Top shape: 128 384 13 13 (8306688) I0407 22:24:58.400903 359 net.cpp:137] Memory required for data: 936294400 I0407 22:24:58.400907 359 layer_factory.hpp:77] Creating layer conv4 I0407 22:24:58.400915 359 net.cpp:84] Creating Layer conv4 I0407 22:24:58.400918 359 net.cpp:406] conv4 <- conv3 I0407 22:24:58.400923 359 net.cpp:380] conv4 -> conv4 I0407 22:24:58.411839 359 net.cpp:122] Setting up conv4 I0407 22:24:58.411855 359 net.cpp:129] Top shape: 128 384 13 13 (8306688) I0407 22:24:58.411859 359 net.cpp:137] Memory required for data: 969521152 I0407 22:24:58.411867 359 layer_factory.hpp:77] Creating layer relu4 I0407 22:24:58.411876 359 net.cpp:84] Creating Layer relu4 I0407 22:24:58.411896 359 net.cpp:406] relu4 <- conv4 I0407 22:24:58.411902 359 net.cpp:367] relu4 -> conv4 (in-place) I0407 22:24:58.412392 359 net.cpp:122] Setting up relu4 I0407 22:24:58.412402 359 net.cpp:129] Top shape: 128 384 13 13 (8306688) I0407 22:24:58.412405 359 net.cpp:137] Memory required for data: 1002747904 I0407 22:24:58.412408 359 layer_factory.hpp:77] Creating layer conv5 I0407 22:24:58.412417 359 net.cpp:84] Creating Layer conv5 I0407 22:24:58.412420 359 net.cpp:406] conv5 <- conv4 I0407 22:24:58.412426 359 net.cpp:380] conv5 -> conv5 I0407 22:24:58.421608 359 net.cpp:122] Setting up conv5 I0407 22:24:58.421627 359 net.cpp:129] Top shape: 128 256 13 13 (5537792) I0407 22:24:58.421629 359 net.cpp:137] Memory required for data: 1024899072 I0407 22:24:58.421643 359 layer_factory.hpp:77] Creating layer relu5 I0407 22:24:58.421653 359 net.cpp:84] Creating Layer relu5 I0407 22:24:58.421655 359 net.cpp:406] relu5 <- conv5 I0407 22:24:58.421661 359 net.cpp:367] relu5 -> conv5 (in-place) I0407 22:24:58.422227 359 net.cpp:122] Setting up relu5 I0407 22:24:58.422238 359 net.cpp:129] Top shape: 128 256 13 13 (5537792) I0407 22:24:58.422241 359 net.cpp:137] Memory required for data: 1047050240 I0407 22:24:58.422245 359 layer_factory.hpp:77] Creating layer pool5 I0407 22:24:58.422250 359 net.cpp:84] Creating Layer pool5 I0407 22:24:58.422253 359 net.cpp:406] pool5 <- conv5 I0407 22:24:58.422259 359 net.cpp:380] pool5 -> pool5 I0407 22:24:58.422294 359 net.cpp:122] Setting up pool5 I0407 22:24:58.422300 359 net.cpp:129] Top shape: 128 256 6 6 (1179648) I0407 22:24:58.422303 359 net.cpp:137] Memory required for data: 1051768832 I0407 22:24:58.422307 359 layer_factory.hpp:77] Creating layer fc6 I0407 22:24:58.422315 359 net.cpp:84] Creating Layer fc6 I0407 22:24:58.422318 359 net.cpp:406] fc6 <- pool5 I0407 22:24:58.422323 359 net.cpp:380] fc6 -> fc6 I0407 22:24:58.781308 359 net.cpp:122] Setting up fc6 I0407 22:24:58.781327 359 net.cpp:129] Top shape: 128 4096 (524288) I0407 22:24:58.781330 359 net.cpp:137] Memory required for data: 1053865984 I0407 22:24:58.781339 359 layer_factory.hpp:77] Creating layer relu6 I0407 22:24:58.781348 359 net.cpp:84] Creating Layer relu6 I0407 22:24:58.781352 359 net.cpp:406] relu6 <- fc6 I0407 22:24:58.781359 359 net.cpp:367] relu6 -> fc6 (in-place) I0407 22:24:58.782140 359 net.cpp:122] Setting up relu6 I0407 22:24:58.782151 359 net.cpp:129] Top shape: 128 4096 (524288) I0407 22:24:58.782155 359 net.cpp:137] Memory required for data: 1055963136 I0407 22:24:58.782158 359 layer_factory.hpp:77] Creating layer drop6 I0407 22:24:58.782164 359 net.cpp:84] Creating Layer drop6 I0407 22:24:58.782167 359 net.cpp:406] drop6 <- fc6 I0407 22:24:58.782172 359 net.cpp:367] drop6 -> fc6 (in-place) I0407 22:24:58.782200 359 net.cpp:122] Setting up drop6 I0407 22:24:58.782205 359 net.cpp:129] Top shape: 128 4096 (524288) I0407 22:24:58.782207 359 net.cpp:137] Memory required for data: 1058060288 I0407 22:24:58.782210 359 layer_factory.hpp:77] Creating layer fc7 I0407 22:24:58.782218 359 net.cpp:84] Creating Layer fc7 I0407 22:24:58.782222 359 net.cpp:406] fc7 <- fc6 I0407 22:24:58.782225 359 net.cpp:380] fc7 -> fc7 I0407 22:24:58.941969 359 net.cpp:122] Setting up fc7 I0407 22:24:58.941987 359 net.cpp:129] Top shape: 128 4096 (524288) I0407 22:24:58.941990 359 net.cpp:137] Memory required for data: 1060157440 I0407 22:24:58.941999 359 layer_factory.hpp:77] Creating layer relu7 I0407 22:24:58.942008 359 net.cpp:84] Creating Layer relu7 I0407 22:24:58.942011 359 net.cpp:406] relu7 <- fc7 I0407 22:24:58.942018 359 net.cpp:367] relu7 -> fc7 (in-place) I0407 22:24:58.942502 359 net.cpp:122] Setting up relu7 I0407 22:24:58.942517 359 net.cpp:129] Top shape: 128 4096 (524288) I0407 22:24:58.942519 359 net.cpp:137] Memory required for data: 1062254592 I0407 22:24:58.942523 359 layer_factory.hpp:77] Creating layer drop7 I0407 22:24:58.942529 359 net.cpp:84] Creating Layer drop7 I0407 22:24:58.942548 359 net.cpp:406] drop7 <- fc7 I0407 22:24:58.942554 359 net.cpp:367] drop7 -> fc7 (in-place) I0407 22:24:58.942576 359 net.cpp:122] Setting up drop7 I0407 22:24:58.942581 359 net.cpp:129] Top shape: 128 4096 (524288) I0407 22:24:58.942584 359 net.cpp:137] Memory required for data: 1064351744 I0407 22:24:58.942586 359 layer_factory.hpp:77] Creating layer fc8 I0407 22:24:58.942595 359 net.cpp:84] Creating Layer fc8 I0407 22:24:58.942597 359 net.cpp:406] fc8 <- fc7 I0407 22:24:58.942602 359 net.cpp:380] fc8 -> fc8 I0407 22:24:58.950403 359 net.cpp:122] Setting up fc8 I0407 22:24:58.950412 359 net.cpp:129] Top shape: 128 196 (25088) I0407 22:24:58.950415 359 net.cpp:137] Memory required for data: 1064452096 I0407 22:24:58.950421 359 layer_factory.hpp:77] Creating layer loss I0407 22:24:58.950428 359 net.cpp:84] Creating Layer loss I0407 22:24:58.950431 359 net.cpp:406] loss <- fc8 I0407 22:24:58.950435 359 net.cpp:406] loss <- label I0407 22:24:58.950441 359 net.cpp:380] loss -> loss I0407 22:24:58.950449 359 layer_factory.hpp:77] Creating layer loss I0407 22:24:58.951120 359 net.cpp:122] Setting up loss I0407 22:24:58.951129 359 net.cpp:129] Top shape: (1) I0407 22:24:58.951133 359 net.cpp:132] with loss weight 1 I0407 22:24:58.951148 359 net.cpp:137] Memory required for data: 1064452100 I0407 22:24:58.951151 359 net.cpp:198] loss needs backward computation. I0407 22:24:58.951157 359 net.cpp:198] fc8 needs backward computation. I0407 22:24:58.951160 359 net.cpp:198] drop7 needs backward computation. I0407 22:24:58.951164 359 net.cpp:198] relu7 needs backward computation. I0407 22:24:58.951165 359 net.cpp:198] fc7 needs backward computation. I0407 22:24:58.951169 359 net.cpp:198] drop6 needs backward computation. I0407 22:24:58.951170 359 net.cpp:198] relu6 needs backward computation. I0407 22:24:58.951174 359 net.cpp:198] fc6 needs backward computation. I0407 22:24:58.951176 359 net.cpp:198] pool5 needs backward computation. I0407 22:24:58.951179 359 net.cpp:198] relu5 needs backward computation. I0407 22:24:58.951182 359 net.cpp:198] conv5 needs backward computation. I0407 22:24:58.951185 359 net.cpp:198] relu4 needs backward computation. I0407 22:24:58.951189 359 net.cpp:198] conv4 needs backward computation. I0407 22:24:58.951191 359 net.cpp:198] relu3 needs backward computation. I0407 22:24:58.951193 359 net.cpp:198] conv3 needs backward computation. I0407 22:24:58.951203 359 net.cpp:198] pool2 needs backward computation. I0407 22:24:58.951207 359 net.cpp:198] norm2 needs backward computation. I0407 22:24:58.951210 359 net.cpp:198] relu2 needs backward computation. I0407 22:24:58.951212 359 net.cpp:198] conv2 needs backward computation. I0407 22:24:58.951215 359 net.cpp:198] pool1 needs backward computation. I0407 22:24:58.951218 359 net.cpp:198] norm1 needs backward computation. I0407 22:24:58.951221 359 net.cpp:198] relu1 needs backward computation. I0407 22:24:58.951223 359 net.cpp:198] conv1 needs backward computation. I0407 22:24:58.951226 359 net.cpp:200] train-data does not need backward computation. I0407 22:24:58.951229 359 net.cpp:242] This network produces output loss I0407 22:24:58.951246 359 net.cpp:255] Network initialization done. I0407 22:24:58.951777 359 solver.cpp:172] Creating test net (#0) specified by net file: train_val.prototxt I0407 22:24:58.951807 359 net.cpp:294] The NetState phase (1) differed from the phase (0) specified by a rule in layer train-data I0407 22:24:58.951942 359 net.cpp:51] Initializing net from parameters: state { phase: TEST } layer { name: "val-data" type: "Data" top: "data" top: "label" include { phase: TEST } transform_param { crop_size: 227 mean_file: "/mnt/bigdisk/DIGITS-TAM-1/digits/jobs/20210407-221849-d51c/mean.binaryproto" } data_param { source: "/mnt/bigdisk/DIGITS-TAM-1/digits/jobs/20210407-221849-d51c/val_db" batch_size: 32 backend: LMDB } } layer { name: "conv1" type: "Convolution" bottom: "data" top: "conv1" param { lr_mult: 1 decay_mult: 1 } param { lr_mult: 2 decay_mult: 0 } convolution_param { num_output: 96 kernel_size: 11 stride: 4 weight_filler { type: "gaussian" std: 0.01 } bias_filler { type: "constant" value: 0 } } } layer { name: "relu1" type: "ReLU" bottom: "conv1" top: "conv1" } layer { name: "norm1" type: "LRN" bottom: "conv1" top: "norm1" lrn_param { local_size: 5 alpha: 0.0001 beta: 0.75 } } layer { name: "pool1" type: "Pooling" bottom: "norm1" top: "pool1" pooling_param { pool: MAX kernel_size: 3 stride: 2 } } layer { name: "conv2" type: "Convolution" bottom: "pool1" top: "conv2" param { lr_mult: 1 decay_mult: 1 } param { lr_mult: 2 decay_mult: 0 } convolution_param { num_output: 256 pad: 2 kernel_size: 5 group: 2 weight_filler { type: "gaussian" std: 0.01 } bias_filler { type: "constant" value: 0.1 } } } layer { name: "relu2" type: "ReLU" bottom: "conv2" top: "conv2" } layer { name: "norm2" type: "LRN" bottom: "conv2" top: "norm2" lrn_param { local_size: 5 alpha: 0.0001 beta: 0.75 } } layer { name: "pool2" type: "Pooling" bottom: "norm2" top: "pool2" pooling_param { pool: MAX kernel_size: 3 stride: 2 } } layer { name: "conv3" type: "Convolution" bottom: "pool2" top: "conv3" param { lr_mult: 1 decay_mult: 1 } param { lr_mult: 2 decay_mult: 0 } convolution_param { num_output: 384 pad: 1 kernel_size: 3 weight_filler { type: "gaussian" std: 0.01 } bias_filler { type: "constant" value: 0 } } } layer { name: "relu3" type: "ReLU" bottom: "conv3" top: "conv3" } layer { name: "conv4" type: "Convolution" bottom: "conv3" top: "conv4" param { lr_mult: 1 decay_mult: 1 } param { lr_mult: 2 decay_mult: 0 } convolution_param { num_output: 384 pad: 1 kernel_size: 3 group: 2 weight_filler { type: "gaussian" std: 0.01 } bias_filler { type: "constant" value: 0.1 } } } layer { name: "relu4" type: "ReLU" bottom: "conv4" top: "conv4" } layer { name: "conv5" type: "Convolution" bottom: "conv4" top: "conv5" param { lr_mult: 1 decay_mult: 1 } param { lr_mult: 2 decay_mult: 0 } convolution_param { num_output: 256 pad: 1 kernel_size: 3 group: 2 weight_filler { type: "gaussian" std: 0.01 } bias_filler { type: "constant" value: 0.1 } } } layer { name: "relu5" type: "ReLU" bottom: "conv5" top: "conv5" } layer { name: "pool5" type: "Pooling" bottom: "conv5" top: "pool5" pooling_param { pool: MAX kernel_size: 3 stride: 2 } } layer { name: "fc6" type: "InnerProduct" bottom: "pool5" top: "fc6" param { lr_mult: 1 decay_mult: 1 } param { lr_mult: 2 decay_mult: 0 } inner_product_param { num_output: 4096 weight_filler { type: "gaussian" std: 0.005 } bias_filler { type: "constant" value: 0.1 } } } layer { name: "relu6" type: "ReLU" bottom: "fc6" top: "fc6" } layer { name: "drop6" type: "Dropout" bottom: "fc6" top: "fc6" dropout_param { dropout_ratio: 0.5 } } layer { name: "fc7" type: "InnerProduct" bottom: "fc6" top: "fc7" param { lr_mult: 1 decay_mult: 1 } param { lr_mult: 2 decay_mult: 0 } inner_product_param { num_output: 4096 weight_filler { type: "gaussian" std: 0.005 } bias_filler { type: "constant" value: 0.1 } } } layer { name: "relu7" type: "ReLU" bottom: "fc7" top: "fc7" } layer { name: "drop7" type: "Dropout" bottom: "fc7" top: "fc7" dropout_param { dropout_ratio: 0.5 } } layer { name: "fc8" type: "InnerProduct" bottom: "fc7" top: "fc8" param { lr_mult: 1 decay_mult: 1 } param { lr_mult: 2 decay_mult: 0 } inner_product_param { num_output: 196 weight_filler { type: "gaussian" std: 0.01 } bias_filler { type: "constant" value: 0 } } } layer { name: "accuracy" type: "Accuracy" bottom: "fc8" bottom: "label" top: "accuracy" include { phase: TEST } } layer { name: "loss" type: "SoftmaxWithLoss" bottom: "fc8" bottom: "label" top: "loss" } I0407 22:24:58.952044 359 layer_factory.hpp:77] Creating layer val-data I0407 22:24:58.953537 359 db_lmdb.cpp:35] Opened lmdb /mnt/bigdisk/DIGITS-TAM-1/digits/jobs/20210407-221849-d51c/val_db I0407 22:24:58.954196 359 net.cpp:84] Creating Layer val-data I0407 22:24:58.954207 359 net.cpp:380] val-data -> data I0407 22:24:58.954216 359 net.cpp:380] val-data -> label I0407 22:24:58.954221 359 data_transformer.cpp:25] Loading mean file from: /mnt/bigdisk/DIGITS-TAM-1/digits/jobs/20210407-221849-d51c/mean.binaryproto I0407 22:24:58.957778 359 data_layer.cpp:45] output data size: 32,3,227,227 I0407 22:24:58.991655 359 net.cpp:122] Setting up val-data I0407 22:24:58.991675 359 net.cpp:129] Top shape: 32 3 227 227 (4946784) I0407 22:24:58.991679 359 net.cpp:129] Top shape: 32 (32) I0407 22:24:58.991683 359 net.cpp:137] Memory required for data: 19787264 I0407 22:24:58.991688 359 layer_factory.hpp:77] Creating layer label_val-data_1_split I0407 22:24:58.991699 359 net.cpp:84] Creating Layer label_val-data_1_split I0407 22:24:58.991703 359 net.cpp:406] label_val-data_1_split <- label I0407 22:24:58.991708 359 net.cpp:380] label_val-data_1_split -> label_val-data_1_split_0 I0407 22:24:58.991717 359 net.cpp:380] label_val-data_1_split -> label_val-data_1_split_1 I0407 22:24:58.991762 359 net.cpp:122] Setting up label_val-data_1_split I0407 22:24:58.991768 359 net.cpp:129] Top shape: 32 (32) I0407 22:24:58.991771 359 net.cpp:129] Top shape: 32 (32) I0407 22:24:58.991773 359 net.cpp:137] Memory required for data: 19787520 I0407 22:24:58.991776 359 layer_factory.hpp:77] Creating layer conv1 I0407 22:24:58.991787 359 net.cpp:84] Creating Layer conv1 I0407 22:24:58.991789 359 net.cpp:406] conv1 <- data I0407 22:24:58.991794 359 net.cpp:380] conv1 -> conv1 I0407 22:24:58.994642 359 net.cpp:122] Setting up conv1 I0407 22:24:58.994653 359 net.cpp:129] Top shape: 32 96 55 55 (9292800) I0407 22:24:58.994657 359 net.cpp:137] Memory required for data: 56958720 I0407 22:24:58.994665 359 layer_factory.hpp:77] Creating layer relu1 I0407 22:24:58.994671 359 net.cpp:84] Creating Layer relu1 I0407 22:24:58.994674 359 net.cpp:406] relu1 <- conv1 I0407 22:24:58.994679 359 net.cpp:367] relu1 -> conv1 (in-place) I0407 22:24:58.995002 359 net.cpp:122] Setting up relu1 I0407 22:24:58.995012 359 net.cpp:129] Top shape: 32 96 55 55 (9292800) I0407 22:24:58.995014 359 net.cpp:137] Memory required for data: 94129920 I0407 22:24:58.995018 359 layer_factory.hpp:77] Creating layer norm1 I0407 22:24:58.995024 359 net.cpp:84] Creating Layer norm1 I0407 22:24:58.995028 359 net.cpp:406] norm1 <- conv1 I0407 22:24:58.995033 359 net.cpp:380] norm1 -> norm1 I0407 22:24:58.995978 359 net.cpp:122] Setting up norm1 I0407 22:24:58.995988 359 net.cpp:129] Top shape: 32 96 55 55 (9292800) I0407 22:24:58.995990 359 net.cpp:137] Memory required for data: 131301120 I0407 22:24:58.995995 359 layer_factory.hpp:77] Creating layer pool1 I0407 22:24:58.996001 359 net.cpp:84] Creating Layer pool1 I0407 22:24:58.996003 359 net.cpp:406] pool1 <- norm1 I0407 22:24:58.996008 359 net.cpp:380] pool1 -> pool1 I0407 22:24:58.996034 359 net.cpp:122] Setting up pool1 I0407 22:24:58.996039 359 net.cpp:129] Top shape: 32 96 27 27 (2239488) I0407 22:24:58.996042 359 net.cpp:137] Memory required for data: 140259072 I0407 22:24:58.996044 359 layer_factory.hpp:77] Creating layer conv2 I0407 22:24:58.996052 359 net.cpp:84] Creating Layer conv2 I0407 22:24:58.996054 359 net.cpp:406] conv2 <- pool1 I0407 22:24:58.996076 359 net.cpp:380] conv2 -> conv2 I0407 22:24:59.004604 359 net.cpp:122] Setting up conv2 I0407 22:24:59.004626 359 net.cpp:129] Top shape: 32 256 27 27 (5971968) I0407 22:24:59.004629 359 net.cpp:137] Memory required for data: 164146944 I0407 22:24:59.004640 359 layer_factory.hpp:77] Creating layer relu2 I0407 22:24:59.004648 359 net.cpp:84] Creating Layer relu2 I0407 22:24:59.004652 359 net.cpp:406] relu2 <- conv2 I0407 22:24:59.004657 359 net.cpp:367] relu2 -> conv2 (in-place) I0407 22:24:59.005244 359 net.cpp:122] Setting up relu2 I0407 22:24:59.005254 359 net.cpp:129] Top shape: 32 256 27 27 (5971968) I0407 22:24:59.005256 359 net.cpp:137] Memory required for data: 188034816 I0407 22:24:59.005259 359 layer_factory.hpp:77] Creating layer norm2 I0407 22:24:59.005270 359 net.cpp:84] Creating Layer norm2 I0407 22:24:59.005272 359 net.cpp:406] norm2 <- conv2 I0407 22:24:59.005277 359 net.cpp:380] norm2 -> norm2 I0407 22:24:59.006044 359 net.cpp:122] Setting up norm2 I0407 22:24:59.006053 359 net.cpp:129] Top shape: 32 256 27 27 (5971968) I0407 22:24:59.006057 359 net.cpp:137] Memory required for data: 211922688 I0407 22:24:59.006060 359 layer_factory.hpp:77] Creating layer pool2 I0407 22:24:59.006067 359 net.cpp:84] Creating Layer pool2 I0407 22:24:59.006070 359 net.cpp:406] pool2 <- norm2 I0407 22:24:59.006078 359 net.cpp:380] pool2 -> pool2 I0407 22:24:59.006105 359 net.cpp:122] Setting up pool2 I0407 22:24:59.006111 359 net.cpp:129] Top shape: 32 256 13 13 (1384448) I0407 22:24:59.006114 359 net.cpp:137] Memory required for data: 217460480 I0407 22:24:59.006116 359 layer_factory.hpp:77] Creating layer conv3 I0407 22:24:59.006126 359 net.cpp:84] Creating Layer conv3 I0407 22:24:59.006129 359 net.cpp:406] conv3 <- pool2 I0407 22:24:59.006135 359 net.cpp:380] conv3 -> conv3 I0407 22:24:59.018026 359 net.cpp:122] Setting up conv3 I0407 22:24:59.018043 359 net.cpp:129] Top shape: 32 384 13 13 (2076672) I0407 22:24:59.018045 359 net.cpp:137] Memory required for data: 225767168 I0407 22:24:59.018056 359 layer_factory.hpp:77] Creating layer relu3 I0407 22:24:59.018066 359 net.cpp:84] Creating Layer relu3 I0407 22:24:59.018069 359 net.cpp:406] relu3 <- conv3 I0407 22:24:59.018075 359 net.cpp:367] relu3 -> conv3 (in-place) I0407 22:24:59.018676 359 net.cpp:122] Setting up relu3 I0407 22:24:59.018685 359 net.cpp:129] Top shape: 32 384 13 13 (2076672) I0407 22:24:59.018688 359 net.cpp:137] Memory required for data: 234073856 I0407 22:24:59.018692 359 layer_factory.hpp:77] Creating layer conv4 I0407 22:24:59.018702 359 net.cpp:84] Creating Layer conv4 I0407 22:24:59.018705 359 net.cpp:406] conv4 <- conv3 I0407 22:24:59.018712 359 net.cpp:380] conv4 -> conv4 I0407 22:24:59.029397 359 net.cpp:122] Setting up conv4 I0407 22:24:59.029410 359 net.cpp:129] Top shape: 32 384 13 13 (2076672) I0407 22:24:59.029414 359 net.cpp:137] Memory required for data: 242380544 I0407 22:24:59.029422 359 layer_factory.hpp:77] Creating layer relu4 I0407 22:24:59.029428 359 net.cpp:84] Creating Layer relu4 I0407 22:24:59.029431 359 net.cpp:406] relu4 <- conv4 I0407 22:24:59.029438 359 net.cpp:367] relu4 -> conv4 (in-place) I0407 22:24:59.029821 359 net.cpp:122] Setting up relu4 I0407 22:24:59.029830 359 net.cpp:129] Top shape: 32 384 13 13 (2076672) I0407 22:24:59.029832 359 net.cpp:137] Memory required for data: 250687232 I0407 22:24:59.029835 359 layer_factory.hpp:77] Creating layer conv5 I0407 22:24:59.029846 359 net.cpp:84] Creating Layer conv5 I0407 22:24:59.029850 359 net.cpp:406] conv5 <- conv4 I0407 22:24:59.029856 359 net.cpp:380] conv5 -> conv5 I0407 22:24:59.039891 359 net.cpp:122] Setting up conv5 I0407 22:24:59.039907 359 net.cpp:129] Top shape: 32 256 13 13 (1384448) I0407 22:24:59.039911 359 net.cpp:137] Memory required for data: 256225024 I0407 22:24:59.039922 359 layer_factory.hpp:77] Creating layer relu5 I0407 22:24:59.039929 359 net.cpp:84] Creating Layer relu5 I0407 22:24:59.039948 359 net.cpp:406] relu5 <- conv5 I0407 22:24:59.039954 359 net.cpp:367] relu5 -> conv5 (in-place) I0407 22:24:59.040582 359 net.cpp:122] Setting up relu5 I0407 22:24:59.040592 359 net.cpp:129] Top shape: 32 256 13 13 (1384448) I0407 22:24:59.040596 359 net.cpp:137] Memory required for data: 261762816 I0407 22:24:59.040598 359 layer_factory.hpp:77] Creating layer pool5 I0407 22:24:59.040608 359 net.cpp:84] Creating Layer pool5 I0407 22:24:59.040611 359 net.cpp:406] pool5 <- conv5 I0407 22:24:59.040616 359 net.cpp:380] pool5 -> pool5 I0407 22:24:59.040652 359 net.cpp:122] Setting up pool5 I0407 22:24:59.040657 359 net.cpp:129] Top shape: 32 256 6 6 (294912) I0407 22:24:59.040659 359 net.cpp:137] Memory required for data: 262942464 I0407 22:24:59.040663 359 layer_factory.hpp:77] Creating layer fc6 I0407 22:24:59.040670 359 net.cpp:84] Creating Layer fc6 I0407 22:24:59.040673 359 net.cpp:406] fc6 <- pool5 I0407 22:24:59.040678 359 net.cpp:380] fc6 -> fc6 I0407 22:24:59.399106 359 net.cpp:122] Setting up fc6 I0407 22:24:59.399128 359 net.cpp:129] Top shape: 32 4096 (131072) I0407 22:24:59.399132 359 net.cpp:137] Memory required for data: 263466752 I0407 22:24:59.399140 359 layer_factory.hpp:77] Creating layer relu6 I0407 22:24:59.399152 359 net.cpp:84] Creating Layer relu6 I0407 22:24:59.399154 359 net.cpp:406] relu6 <- fc6 I0407 22:24:59.399160 359 net.cpp:367] relu6 -> fc6 (in-place) I0407 22:24:59.399969 359 net.cpp:122] Setting up relu6 I0407 22:24:59.399978 359 net.cpp:129] Top shape: 32 4096 (131072) I0407 22:24:59.399981 359 net.cpp:137] Memory required for data: 263991040 I0407 22:24:59.399984 359 layer_factory.hpp:77] Creating layer drop6 I0407 22:24:59.399991 359 net.cpp:84] Creating Layer drop6 I0407 22:24:59.399994 359 net.cpp:406] drop6 <- fc6 I0407 22:24:59.400000 359 net.cpp:367] drop6 -> fc6 (in-place) I0407 22:24:59.400023 359 net.cpp:122] Setting up drop6 I0407 22:24:59.400028 359 net.cpp:129] Top shape: 32 4096 (131072) I0407 22:24:59.400032 359 net.cpp:137] Memory required for data: 264515328 I0407 22:24:59.400033 359 layer_factory.hpp:77] Creating layer fc7 I0407 22:24:59.400040 359 net.cpp:84] Creating Layer fc7 I0407 22:24:59.400043 359 net.cpp:406] fc7 <- fc6 I0407 22:24:59.400048 359 net.cpp:380] fc7 -> fc7 I0407 22:24:59.559554 359 net.cpp:122] Setting up fc7 I0407 22:24:59.559576 359 net.cpp:129] Top shape: 32 4096 (131072) I0407 22:24:59.559579 359 net.cpp:137] Memory required for data: 265039616 I0407 22:24:59.559588 359 layer_factory.hpp:77] Creating layer relu7 I0407 22:24:59.559597 359 net.cpp:84] Creating Layer relu7 I0407 22:24:59.559599 359 net.cpp:406] relu7 <- fc7 I0407 22:24:59.559605 359 net.cpp:367] relu7 -> fc7 (in-place) I0407 22:24:59.560138 359 net.cpp:122] Setting up relu7 I0407 22:24:59.560149 359 net.cpp:129] Top shape: 32 4096 (131072) I0407 22:24:59.560153 359 net.cpp:137] Memory required for data: 265563904 I0407 22:24:59.560155 359 layer_factory.hpp:77] Creating layer drop7 I0407 22:24:59.560161 359 net.cpp:84] Creating Layer drop7 I0407 22:24:59.560164 359 net.cpp:406] drop7 <- fc7 I0407 22:24:59.560169 359 net.cpp:367] drop7 -> fc7 (in-place) I0407 22:24:59.560194 359 net.cpp:122] Setting up drop7 I0407 22:24:59.560199 359 net.cpp:129] Top shape: 32 4096 (131072) I0407 22:24:59.560202 359 net.cpp:137] Memory required for data: 266088192 I0407 22:24:59.560204 359 layer_factory.hpp:77] Creating layer fc8 I0407 22:24:59.560211 359 net.cpp:84] Creating Layer fc8 I0407 22:24:59.560214 359 net.cpp:406] fc8 <- fc7 I0407 22:24:59.560218 359 net.cpp:380] fc8 -> fc8 I0407 22:24:59.568007 359 net.cpp:122] Setting up fc8 I0407 22:24:59.568018 359 net.cpp:129] Top shape: 32 196 (6272) I0407 22:24:59.568020 359 net.cpp:137] Memory required for data: 266113280 I0407 22:24:59.568027 359 layer_factory.hpp:77] Creating layer fc8_fc8_0_split I0407 22:24:59.568032 359 net.cpp:84] Creating Layer fc8_fc8_0_split I0407 22:24:59.568035 359 net.cpp:406] fc8_fc8_0_split <- fc8 I0407 22:24:59.568056 359 net.cpp:380] fc8_fc8_0_split -> fc8_fc8_0_split_0 I0407 22:24:59.568063 359 net.cpp:380] fc8_fc8_0_split -> fc8_fc8_0_split_1 I0407 22:24:59.568091 359 net.cpp:122] Setting up fc8_fc8_0_split I0407 22:24:59.568095 359 net.cpp:129] Top shape: 32 196 (6272) I0407 22:24:59.568099 359 net.cpp:129] Top shape: 32 196 (6272) I0407 22:24:59.568101 359 net.cpp:137] Memory required for data: 266163456 I0407 22:24:59.568104 359 layer_factory.hpp:77] Creating layer accuracy I0407 22:24:59.568111 359 net.cpp:84] Creating Layer accuracy I0407 22:24:59.568114 359 net.cpp:406] accuracy <- fc8_fc8_0_split_0 I0407 22:24:59.568117 359 net.cpp:406] accuracy <- label_val-data_1_split_0 I0407 22:24:59.568122 359 net.cpp:380] accuracy -> accuracy I0407 22:24:59.568128 359 net.cpp:122] Setting up accuracy I0407 22:24:59.568132 359 net.cpp:129] Top shape: (1) I0407 22:24:59.568135 359 net.cpp:137] Memory required for data: 266163460 I0407 22:24:59.568137 359 layer_factory.hpp:77] Creating layer loss I0407 22:24:59.568141 359 net.cpp:84] Creating Layer loss I0407 22:24:59.568145 359 net.cpp:406] loss <- fc8_fc8_0_split_1 I0407 22:24:59.568147 359 net.cpp:406] loss <- label_val-data_1_split_1 I0407 22:24:59.568151 359 net.cpp:380] loss -> loss I0407 22:24:59.568158 359 layer_factory.hpp:77] Creating layer loss I0407 22:24:59.568816 359 net.cpp:122] Setting up loss I0407 22:24:59.568826 359 net.cpp:129] Top shape: (1) I0407 22:24:59.568830 359 net.cpp:132] with loss weight 1 I0407 22:24:59.568838 359 net.cpp:137] Memory required for data: 266163464 I0407 22:24:59.568842 359 net.cpp:198] loss needs backward computation. I0407 22:24:59.568846 359 net.cpp:200] accuracy does not need backward computation. I0407 22:24:59.568850 359 net.cpp:198] fc8_fc8_0_split needs backward computation. I0407 22:24:59.568852 359 net.cpp:198] fc8 needs backward computation. I0407 22:24:59.568856 359 net.cpp:198] drop7 needs backward computation. I0407 22:24:59.568858 359 net.cpp:198] relu7 needs backward computation. I0407 22:24:59.568861 359 net.cpp:198] fc7 needs backward computation. I0407 22:24:59.568863 359 net.cpp:198] drop6 needs backward computation. I0407 22:24:59.568866 359 net.cpp:198] relu6 needs backward computation. I0407 22:24:59.568868 359 net.cpp:198] fc6 needs backward computation. I0407 22:24:59.568872 359 net.cpp:198] pool5 needs backward computation. I0407 22:24:59.568876 359 net.cpp:198] relu5 needs backward computation. I0407 22:24:59.568877 359 net.cpp:198] conv5 needs backward computation. I0407 22:24:59.568881 359 net.cpp:198] relu4 needs backward computation. I0407 22:24:59.568883 359 net.cpp:198] conv4 needs backward computation. I0407 22:24:59.568886 359 net.cpp:198] relu3 needs backward computation. I0407 22:24:59.568889 359 net.cpp:198] conv3 needs backward computation. I0407 22:24:59.568892 359 net.cpp:198] pool2 needs backward computation. I0407 22:24:59.568894 359 net.cpp:198] norm2 needs backward computation. I0407 22:24:59.568897 359 net.cpp:198] relu2 needs backward computation. I0407 22:24:59.568900 359 net.cpp:198] conv2 needs backward computation. I0407 22:24:59.568902 359 net.cpp:198] pool1 needs backward computation. I0407 22:24:59.568905 359 net.cpp:198] norm1 needs backward computation. I0407 22:24:59.568909 359 net.cpp:198] relu1 needs backward computation. I0407 22:24:59.568912 359 net.cpp:198] conv1 needs backward computation. I0407 22:24:59.568914 359 net.cpp:200] label_val-data_1_split does not need backward computation. I0407 22:24:59.568918 359 net.cpp:200] val-data does not need backward computation. I0407 22:24:59.568920 359 net.cpp:242] This network produces output accuracy I0407 22:24:59.568923 359 net.cpp:242] This network produces output loss I0407 22:24:59.568939 359 net.cpp:255] Network initialization done. I0407 22:24:59.569005 359 solver.cpp:56] Solver scaffolding done. I0407 22:24:59.569339 359 caffe.cpp:248] Starting Optimization I0407 22:24:59.569347 359 solver.cpp:272] Solving I0407 22:24:59.569358 359 solver.cpp:273] Learning Rate Policy: sigmoid I0407 22:24:59.571000 359 solver.cpp:330] Iteration 0, Testing net (#0) I0407 22:24:59.571009 359 net.cpp:676] Ignoring source layer train-data I0407 22:24:59.658704 359 blocking_queue.cpp:49] Waiting for data I0407 22:25:03.936782 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:25:03.982277 359 solver.cpp:397] Test net output #0: accuracy = 0.00245098 I0407 22:25:03.982334 359 solver.cpp:397] Test net output #1: loss = 5.28174 (* 1 = 5.28174 loss) I0407 22:25:04.081135 359 solver.cpp:218] Iteration 0 (0 iter/s, 4.51172s/12 iters), loss = 5.28806 I0407 22:25:04.082665 359 solver.cpp:237] Train net output #0: loss = 5.28806 (* 1 = 5.28806 loss) I0407 22:25:04.082700 359 sgd_solver.cpp:105] Iteration 0, lr = 0.00999955 I0407 22:25:07.896353 359 solver.cpp:218] Iteration 12 (3.14657 iter/s, 3.81367s/12 iters), loss = 5.28562 I0407 22:25:07.896394 359 solver.cpp:237] Train net output #0: loss = 5.28562 (* 1 = 5.28562 loss) I0407 22:25:07.896401 359 sgd_solver.cpp:105] Iteration 12, lr = 0.00999954 I0407 22:25:12.851897 359 solver.cpp:218] Iteration 24 (2.42156 iter/s, 4.95549s/12 iters), loss = 5.29167 I0407 22:25:12.851936 359 solver.cpp:237] Train net output #0: loss = 5.29167 (* 1 = 5.29167 loss) I0407 22:25:12.851944 359 sgd_solver.cpp:105] Iteration 24, lr = 0.00999952 I0407 22:25:17.794610 359 solver.cpp:218] Iteration 36 (2.42785 iter/s, 4.94265s/12 iters), loss = 5.28802 I0407 22:25:17.794651 359 solver.cpp:237] Train net output #0: loss = 5.28802 (* 1 = 5.28802 loss) I0407 22:25:17.794661 359 sgd_solver.cpp:105] Iteration 36, lr = 0.00999951 I0407 22:25:22.766485 359 solver.cpp:218] Iteration 48 (2.41361 iter/s, 4.97181s/12 iters), loss = 5.2704 I0407 22:25:22.766522 359 solver.cpp:237] Train net output #0: loss = 5.2704 (* 1 = 5.2704 loss) I0407 22:25:22.766530 359 sgd_solver.cpp:105] Iteration 48, lr = 0.0099995 I0407 22:25:27.680014 359 solver.cpp:218] Iteration 60 (2.44226 iter/s, 4.91347s/12 iters), loss = 5.26748 I0407 22:25:27.680188 359 solver.cpp:237] Train net output #0: loss = 5.26748 (* 1 = 5.26748 loss) I0407 22:25:27.680197 359 sgd_solver.cpp:105] Iteration 60, lr = 0.00999949 I0407 22:25:32.628957 359 solver.cpp:218] Iteration 72 (2.42485 iter/s, 4.94875s/12 iters), loss = 5.29883 I0407 22:25:32.628998 359 solver.cpp:237] Train net output #0: loss = 5.29883 (* 1 = 5.29883 loss) I0407 22:25:32.629009 359 sgd_solver.cpp:105] Iteration 72, lr = 0.00999948 I0407 22:25:37.554020 359 solver.cpp:218] Iteration 84 (2.43655 iter/s, 4.92501s/12 iters), loss = 5.30654 I0407 22:25:37.554062 359 solver.cpp:237] Train net output #0: loss = 5.30654 (* 1 = 5.30654 loss) I0407 22:25:37.554071 359 sgd_solver.cpp:105] Iteration 84, lr = 0.00999946 I0407 22:25:42.509307 359 solver.cpp:218] Iteration 96 (2.42169 iter/s, 4.95522s/12 iters), loss = 5.29047 I0407 22:25:42.509352 359 solver.cpp:237] Train net output #0: loss = 5.29047 (* 1 = 5.29047 loss) I0407 22:25:42.509361 359 sgd_solver.cpp:105] Iteration 96, lr = 0.00999945 I0407 22:25:44.187778 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:25:44.489156 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_102.caffemodel I0407 22:25:50.035602 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_102.solverstate I0407 22:25:53.812662 359 solver.cpp:330] Iteration 102, Testing net (#0) I0407 22:25:53.812680 359 net.cpp:676] Ignoring source layer train-data I0407 22:25:58.538599 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:25:58.625958 359 solver.cpp:397] Test net output #0: accuracy = 0.00551471 I0407 22:25:58.626006 359 solver.cpp:397] Test net output #1: loss = 5.28653 (* 1 = 5.28653 loss) I0407 22:26:00.399950 359 solver.cpp:218] Iteration 108 (0.670745 iter/s, 17.8906s/12 iters), loss = 5.29295 I0407 22:26:00.399988 359 solver.cpp:237] Train net output #0: loss = 5.29295 (* 1 = 5.29295 loss) I0407 22:26:00.399996 359 sgd_solver.cpp:105] Iteration 108, lr = 0.00999944 I0407 22:26:05.338814 359 solver.cpp:218] Iteration 120 (2.42974 iter/s, 4.9388s/12 iters), loss = 5.26551 I0407 22:26:05.338848 359 solver.cpp:237] Train net output #0: loss = 5.26551 (* 1 = 5.26551 loss) I0407 22:26:05.338856 359 sgd_solver.cpp:105] Iteration 120, lr = 0.00999943 I0407 22:26:10.315505 359 solver.cpp:218] Iteration 132 (2.41127 iter/s, 4.97663s/12 iters), loss = 5.30808 I0407 22:26:10.315548 359 solver.cpp:237] Train net output #0: loss = 5.30808 (* 1 = 5.30808 loss) I0407 22:26:10.315557 359 sgd_solver.cpp:105] Iteration 132, lr = 0.00999941 I0407 22:26:15.271690 359 solver.cpp:218] Iteration 144 (2.42125 iter/s, 4.95612s/12 iters), loss = 5.29402 I0407 22:26:15.271726 359 solver.cpp:237] Train net output #0: loss = 5.29402 (* 1 = 5.29402 loss) I0407 22:26:15.271734 359 sgd_solver.cpp:105] Iteration 144, lr = 0.0099994 I0407 22:26:20.201300 359 solver.cpp:218] Iteration 156 (2.4343 iter/s, 4.92955s/12 iters), loss = 5.26807 I0407 22:26:20.201336 359 solver.cpp:237] Train net output #0: loss = 5.26807 (* 1 = 5.26807 loss) I0407 22:26:20.201344 359 sgd_solver.cpp:105] Iteration 156, lr = 0.00999938 I0407 22:26:25.184865 359 solver.cpp:218] Iteration 168 (2.40794 iter/s, 4.98351s/12 iters), loss = 5.28113 I0407 22:26:25.184901 359 solver.cpp:237] Train net output #0: loss = 5.28113 (* 1 = 5.28113 loss) I0407 22:26:25.184909 359 sgd_solver.cpp:105] Iteration 168, lr = 0.00999937 I0407 22:26:30.109724 359 solver.cpp:218] Iteration 180 (2.43665 iter/s, 4.9248s/12 iters), loss = 5.29121 I0407 22:26:30.109831 359 solver.cpp:237] Train net output #0: loss = 5.29121 (* 1 = 5.29121 loss) I0407 22:26:30.109840 359 sgd_solver.cpp:105] Iteration 180, lr = 0.00999935 I0407 22:26:35.060994 359 solver.cpp:218] Iteration 192 (2.42368 iter/s, 4.95114s/12 iters), loss = 5.27884 I0407 22:26:35.061033 359 solver.cpp:237] Train net output #0: loss = 5.27884 (* 1 = 5.27884 loss) I0407 22:26:35.061039 359 sgd_solver.cpp:105] Iteration 192, lr = 0.00999934 I0407 22:26:38.851106 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:26:39.524173 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_204.caffemodel I0407 22:26:42.585925 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_204.solverstate I0407 22:26:44.957439 359 solver.cpp:330] Iteration 204, Testing net (#0) I0407 22:26:44.957466 359 net.cpp:676] Ignoring source layer train-data I0407 22:26:49.607137 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:26:49.742899 359 solver.cpp:397] Test net output #0: accuracy = 0.00796569 I0407 22:26:49.742942 359 solver.cpp:397] Test net output #1: loss = 5.19495 (* 1 = 5.19495 loss) I0407 22:26:49.839781 359 solver.cpp:218] Iteration 204 (0.811979 iter/s, 14.7787s/12 iters), loss = 5.24342 I0407 22:26:49.839821 359 solver.cpp:237] Train net output #0: loss = 5.24342 (* 1 = 5.24342 loss) I0407 22:26:49.839828 359 sgd_solver.cpp:105] Iteration 204, lr = 0.00999932 I0407 22:26:53.951974 359 solver.cpp:218] Iteration 216 (2.91819 iter/s, 4.11213s/12 iters), loss = 5.2751 I0407 22:26:53.952008 359 solver.cpp:237] Train net output #0: loss = 5.2751 (* 1 = 5.2751 loss) I0407 22:26:53.952015 359 sgd_solver.cpp:105] Iteration 216, lr = 0.00999931 I0407 22:26:58.894639 359 solver.cpp:218] Iteration 228 (2.42787 iter/s, 4.94261s/12 iters), loss = 5.2056 I0407 22:26:58.894675 359 solver.cpp:237] Train net output #0: loss = 5.2056 (* 1 = 5.2056 loss) I0407 22:26:58.894683 359 sgd_solver.cpp:105] Iteration 228, lr = 0.00999929 I0407 22:27:03.779997 359 solver.cpp:218] Iteration 240 (2.45635 iter/s, 4.8853s/12 iters), loss = 5.21841 I0407 22:27:03.780162 359 solver.cpp:237] Train net output #0: loss = 5.21841 (* 1 = 5.21841 loss) I0407 22:27:03.780172 359 sgd_solver.cpp:105] Iteration 240, lr = 0.00999927 I0407 22:27:08.759843 359 solver.cpp:218] Iteration 252 (2.4098 iter/s, 4.97966s/12 iters), loss = 5.21303 I0407 22:27:08.759881 359 solver.cpp:237] Train net output #0: loss = 5.21303 (* 1 = 5.21303 loss) I0407 22:27:08.759888 359 sgd_solver.cpp:105] Iteration 252, lr = 0.00999926 I0407 22:27:13.693011 359 solver.cpp:218] Iteration 264 (2.43255 iter/s, 4.9331s/12 iters), loss = 5.17119 I0407 22:27:13.693055 359 solver.cpp:237] Train net output #0: loss = 5.17119 (* 1 = 5.17119 loss) I0407 22:27:13.693063 359 sgd_solver.cpp:105] Iteration 264, lr = 0.00999924 I0407 22:27:18.646687 359 solver.cpp:218] Iteration 276 (2.42248 iter/s, 4.95361s/12 iters), loss = 5.10664 I0407 22:27:18.646729 359 solver.cpp:237] Train net output #0: loss = 5.10664 (* 1 = 5.10664 loss) I0407 22:27:18.646737 359 sgd_solver.cpp:105] Iteration 276, lr = 0.00999922 I0407 22:27:23.590082 359 solver.cpp:218] Iteration 288 (2.42751 iter/s, 4.94333s/12 iters), loss = 5.15481 I0407 22:27:23.590126 359 solver.cpp:237] Train net output #0: loss = 5.15481 (* 1 = 5.15481 loss) I0407 22:27:23.590133 359 sgd_solver.cpp:105] Iteration 288, lr = 0.0099992 I0407 22:27:28.508064 359 solver.cpp:218] Iteration 300 (2.44006 iter/s, 4.91792s/12 iters), loss = 5.1391 I0407 22:27:28.508106 359 solver.cpp:237] Train net output #0: loss = 5.1391 (* 1 = 5.1391 loss) I0407 22:27:28.508114 359 sgd_solver.cpp:105] Iteration 300, lr = 0.00999918 I0407 22:27:29.468757 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:27:30.625247 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_306.caffemodel I0407 22:27:33.680151 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_306.solverstate I0407 22:27:36.049736 359 solver.cpp:330] Iteration 306, Testing net (#0) I0407 22:27:36.049819 359 net.cpp:676] Ignoring source layer train-data I0407 22:27:40.323453 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:27:40.484102 359 solver.cpp:397] Test net output #0: accuracy = 0.00980392 I0407 22:27:40.484138 359 solver.cpp:397] Test net output #1: loss = 5.14301 (* 1 = 5.14301 loss) I0407 22:27:42.278084 359 solver.cpp:218] Iteration 312 (0.871464 iter/s, 13.7699s/12 iters), loss = 5.16477 I0407 22:27:42.278126 359 solver.cpp:237] Train net output #0: loss = 5.16477 (* 1 = 5.16477 loss) I0407 22:27:42.278133 359 sgd_solver.cpp:105] Iteration 312, lr = 0.00999916 I0407 22:27:47.238617 359 solver.cpp:218] Iteration 324 (2.41913 iter/s, 4.96046s/12 iters), loss = 5.10394 I0407 22:27:47.238658 359 solver.cpp:237] Train net output #0: loss = 5.10394 (* 1 = 5.10394 loss) I0407 22:27:47.238667 359 sgd_solver.cpp:105] Iteration 324, lr = 0.00999914 I0407 22:27:52.168494 359 solver.cpp:218] Iteration 336 (2.43417 iter/s, 4.92981s/12 iters), loss = 5.16258 I0407 22:27:52.168537 359 solver.cpp:237] Train net output #0: loss = 5.16258 (* 1 = 5.16258 loss) I0407 22:27:52.168546 359 sgd_solver.cpp:105] Iteration 336, lr = 0.00999912 I0407 22:27:57.110576 359 solver.cpp:218] Iteration 348 (2.42816 iter/s, 4.94201s/12 iters), loss = 5.16971 I0407 22:27:57.110622 359 solver.cpp:237] Train net output #0: loss = 5.16971 (* 1 = 5.16971 loss) I0407 22:27:57.110631 359 sgd_solver.cpp:105] Iteration 348, lr = 0.0099991 I0407 22:28:02.080013 359 solver.cpp:218] Iteration 360 (2.4148 iter/s, 4.96936s/12 iters), loss = 5.08358 I0407 22:28:02.080055 359 solver.cpp:237] Train net output #0: loss = 5.08358 (* 1 = 5.08358 loss) I0407 22:28:02.080065 359 sgd_solver.cpp:105] Iteration 360, lr = 0.00999908 I0407 22:28:07.004658 359 solver.cpp:218] Iteration 372 (2.43676 iter/s, 4.92458s/12 iters), loss = 5.17792 I0407 22:28:07.004829 359 solver.cpp:237] Train net output #0: loss = 5.17792 (* 1 = 5.17792 loss) I0407 22:28:07.004838 359 sgd_solver.cpp:105] Iteration 372, lr = 0.00999906 I0407 22:28:11.959836 359 solver.cpp:218] Iteration 384 (2.4218 iter/s, 4.95499s/12 iters), loss = 5.14355 I0407 22:28:11.959878 359 solver.cpp:237] Train net output #0: loss = 5.14355 (* 1 = 5.14355 loss) I0407 22:28:11.959887 359 sgd_solver.cpp:105] Iteration 384, lr = 0.00999904 I0407 22:28:16.881438 359 solver.cpp:218] Iteration 396 (2.43826 iter/s, 4.92154s/12 iters), loss = 5.17333 I0407 22:28:16.881474 359 solver.cpp:237] Train net output #0: loss = 5.17333 (* 1 = 5.17333 loss) I0407 22:28:16.881481 359 sgd_solver.cpp:105] Iteration 396, lr = 0.00999901 I0407 22:28:19.977671 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:28:21.349093 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_408.caffemodel I0407 22:28:25.832041 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_408.solverstate I0407 22:28:29.169131 359 solver.cpp:330] Iteration 408, Testing net (#0) I0407 22:28:29.169149 359 net.cpp:676] Ignoring source layer train-data I0407 22:28:33.587616 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:28:33.794279 359 solver.cpp:397] Test net output #0: accuracy = 0.0171569 I0407 22:28:33.794325 359 solver.cpp:397] Test net output #1: loss = 5.09501 (* 1 = 5.09501 loss) I0407 22:28:33.890836 359 solver.cpp:218] Iteration 408 (0.705496 iter/s, 17.0093s/12 iters), loss = 5.08627 I0407 22:28:33.890900 359 solver.cpp:237] Train net output #0: loss = 5.08627 (* 1 = 5.08627 loss) I0407 22:28:33.890913 359 sgd_solver.cpp:105] Iteration 408, lr = 0.00999899 I0407 22:28:38.042706 359 solver.cpp:218] Iteration 420 (2.89032 iter/s, 4.15179s/12 iters), loss = 5.02218 I0407 22:28:38.042827 359 solver.cpp:237] Train net output #0: loss = 5.02218 (* 1 = 5.02218 loss) I0407 22:28:38.042836 359 sgd_solver.cpp:105] Iteration 420, lr = 0.00999897 I0407 22:28:42.962224 359 solver.cpp:218] Iteration 432 (2.43933 iter/s, 4.91938s/12 iters), loss = 5.16687 I0407 22:28:42.962265 359 solver.cpp:237] Train net output #0: loss = 5.16687 (* 1 = 5.16687 loss) I0407 22:28:42.962272 359 sgd_solver.cpp:105] Iteration 432, lr = 0.00999894 I0407 22:28:47.891464 359 solver.cpp:218] Iteration 444 (2.43448 iter/s, 4.92917s/12 iters), loss = 5.01469 I0407 22:28:47.891505 359 solver.cpp:237] Train net output #0: loss = 5.01469 (* 1 = 5.01469 loss) I0407 22:28:47.891512 359 sgd_solver.cpp:105] Iteration 444, lr = 0.00999892 I0407 22:28:52.833674 359 solver.cpp:218] Iteration 456 (2.4281 iter/s, 4.94215s/12 iters), loss = 5.04734 I0407 22:28:52.833719 359 solver.cpp:237] Train net output #0: loss = 5.04734 (* 1 = 5.04734 loss) I0407 22:28:52.833726 359 sgd_solver.cpp:105] Iteration 456, lr = 0.00999889 I0407 22:28:57.785079 359 solver.cpp:218] Iteration 468 (2.42359 iter/s, 4.95134s/12 iters), loss = 5.1518 I0407 22:28:57.785120 359 solver.cpp:237] Train net output #0: loss = 5.1518 (* 1 = 5.1518 loss) I0407 22:28:57.785128 359 sgd_solver.cpp:105] Iteration 468, lr = 0.00999886 I0407 22:29:02.684864 359 solver.cpp:218] Iteration 480 (2.44912 iter/s, 4.89972s/12 iters), loss = 5.13715 I0407 22:29:02.684908 359 solver.cpp:237] Train net output #0: loss = 5.13715 (* 1 = 5.13715 loss) I0407 22:29:02.684917 359 sgd_solver.cpp:105] Iteration 480, lr = 0.00999884 I0407 22:29:07.649720 359 solver.cpp:218] Iteration 492 (2.41702 iter/s, 4.96479s/12 iters), loss = 5.02726 I0407 22:29:07.649762 359 solver.cpp:237] Train net output #0: loss = 5.02726 (* 1 = 5.02726 loss) I0407 22:29:07.649771 359 sgd_solver.cpp:105] Iteration 492, lr = 0.00999881 I0407 22:29:12.556727 359 solver.cpp:218] Iteration 504 (2.44552 iter/s, 4.90694s/12 iters), loss = 5.06168 I0407 22:29:12.556864 359 solver.cpp:237] Train net output #0: loss = 5.06168 (* 1 = 5.06168 loss) I0407 22:29:12.556874 359 sgd_solver.cpp:105] Iteration 504, lr = 0.00999878 I0407 22:29:12.793963 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:29:14.535104 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_510.caffemodel I0407 22:29:17.622326 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_510.solverstate I0407 22:29:19.984289 359 solver.cpp:330] Iteration 510, Testing net (#0) I0407 22:29:19.984308 359 net.cpp:676] Ignoring source layer train-data I0407 22:29:24.211604 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:29:24.449582 359 solver.cpp:397] Test net output #0: accuracy = 0.0196078 I0407 22:29:24.449625 359 solver.cpp:397] Test net output #1: loss = 5.03868 (* 1 = 5.03868 loss) I0407 22:29:26.382103 359 solver.cpp:218] Iteration 516 (0.86798 iter/s, 13.8252s/12 iters), loss = 5.04089 I0407 22:29:26.382153 359 solver.cpp:237] Train net output #0: loss = 5.04089 (* 1 = 5.04089 loss) I0407 22:29:26.382162 359 sgd_solver.cpp:105] Iteration 516, lr = 0.00999875 I0407 22:29:31.556311 359 solver.cpp:218] Iteration 528 (2.31923 iter/s, 5.17413s/12 iters), loss = 5.11859 I0407 22:29:31.556351 359 solver.cpp:237] Train net output #0: loss = 5.11859 (* 1 = 5.11859 loss) I0407 22:29:31.556360 359 sgd_solver.cpp:105] Iteration 528, lr = 0.00999872 I0407 22:29:36.519137 359 solver.cpp:218] Iteration 540 (2.41801 iter/s, 4.96276s/12 iters), loss = 4.98663 I0407 22:29:36.519182 359 solver.cpp:237] Train net output #0: loss = 4.98663 (* 1 = 4.98663 loss) I0407 22:29:36.519191 359 sgd_solver.cpp:105] Iteration 540, lr = 0.00999869 I0407 22:29:41.679147 359 solver.cpp:218] Iteration 552 (2.32561 iter/s, 5.15994s/12 iters), loss = 4.89262 I0407 22:29:41.679185 359 solver.cpp:237] Train net output #0: loss = 4.89262 (* 1 = 4.89262 loss) I0407 22:29:41.679193 359 sgd_solver.cpp:105] Iteration 552, lr = 0.00999866 I0407 22:29:46.902901 359 solver.cpp:218] Iteration 564 (2.29723 iter/s, 5.22369s/12 iters), loss = 5.01488 I0407 22:29:46.903031 359 solver.cpp:237] Train net output #0: loss = 5.01488 (* 1 = 5.01488 loss) I0407 22:29:46.903041 359 sgd_solver.cpp:105] Iteration 564, lr = 0.00999863 I0407 22:29:51.820363 359 solver.cpp:218] Iteration 576 (2.44036 iter/s, 4.91731s/12 iters), loss = 5.11047 I0407 22:29:51.820405 359 solver.cpp:237] Train net output #0: loss = 5.11047 (* 1 = 5.11047 loss) I0407 22:29:51.820415 359 sgd_solver.cpp:105] Iteration 576, lr = 0.0099986 I0407 22:29:56.786377 359 solver.cpp:218] Iteration 588 (2.41646 iter/s, 4.96595s/12 iters), loss = 5.0596 I0407 22:29:56.786419 359 solver.cpp:237] Train net output #0: loss = 5.0596 (* 1 = 5.0596 loss) I0407 22:29:56.786428 359 sgd_solver.cpp:105] Iteration 588, lr = 0.00999856 I0407 22:30:01.696429 359 solver.cpp:218] Iteration 600 (2.444 iter/s, 4.90998s/12 iters), loss = 5.00307 I0407 22:30:01.696473 359 solver.cpp:237] Train net output #0: loss = 5.00307 (* 1 = 5.00307 loss) I0407 22:30:01.696481 359 sgd_solver.cpp:105] Iteration 600, lr = 0.00999853 I0407 22:30:04.082912 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:30:06.179540 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_612.caffemodel I0407 22:30:09.268191 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_612.solverstate I0407 22:30:11.650826 359 solver.cpp:330] Iteration 612, Testing net (#0) I0407 22:30:11.650844 359 net.cpp:676] Ignoring source layer train-data I0407 22:30:16.100869 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:30:16.416594 359 solver.cpp:397] Test net output #0: accuracy = 0.0300245 I0407 22:30:16.416641 359 solver.cpp:397] Test net output #1: loss = 4.99111 (* 1 = 4.99111 loss) I0407 22:30:16.513608 359 solver.cpp:218] Iteration 612 (0.809875 iter/s, 14.8171s/12 iters), loss = 4.94482 I0407 22:30:16.513654 359 solver.cpp:237] Train net output #0: loss = 4.94482 (* 1 = 4.94482 loss) I0407 22:30:16.513662 359 sgd_solver.cpp:105] Iteration 612, lr = 0.00999849 I0407 22:30:20.595207 359 solver.cpp:218] Iteration 624 (2.94008 iter/s, 4.08153s/12 iters), loss = 5.06131 I0407 22:30:20.595358 359 solver.cpp:237] Train net output #0: loss = 5.06131 (* 1 = 5.06131 loss) I0407 22:30:20.595368 359 sgd_solver.cpp:105] Iteration 624, lr = 0.00999846 I0407 22:30:25.532660 359 solver.cpp:218] Iteration 636 (2.43049 iter/s, 4.93728s/12 iters), loss = 4.99104 I0407 22:30:25.532699 359 solver.cpp:237] Train net output #0: loss = 4.99104 (* 1 = 4.99104 loss) I0407 22:30:25.532707 359 sgd_solver.cpp:105] Iteration 636, lr = 0.00999842 I0407 22:30:30.427517 359 solver.cpp:218] Iteration 648 (2.45158 iter/s, 4.8948s/12 iters), loss = 4.97093 I0407 22:30:30.427562 359 solver.cpp:237] Train net output #0: loss = 4.97093 (* 1 = 4.97093 loss) I0407 22:30:30.427570 359 sgd_solver.cpp:105] Iteration 648, lr = 0.00999838 I0407 22:30:35.438525 359 solver.cpp:218] Iteration 660 (2.39476 iter/s, 5.01094s/12 iters), loss = 4.91551 I0407 22:30:35.438557 359 solver.cpp:237] Train net output #0: loss = 4.91551 (* 1 = 4.91551 loss) I0407 22:30:35.438565 359 sgd_solver.cpp:105] Iteration 660, lr = 0.00999834 I0407 22:30:40.433089 359 solver.cpp:218] Iteration 672 (2.40264 iter/s, 4.9945s/12 iters), loss = 4.99966 I0407 22:30:40.433125 359 solver.cpp:237] Train net output #0: loss = 4.99966 (* 1 = 4.99966 loss) I0407 22:30:40.433135 359 sgd_solver.cpp:105] Iteration 672, lr = 0.0099983 I0407 22:30:45.276988 359 solver.cpp:218] Iteration 684 (2.47737 iter/s, 4.84384s/12 iters), loss = 4.84708 I0407 22:30:45.277024 359 solver.cpp:237] Train net output #0: loss = 4.84708 (* 1 = 4.84708 loss) I0407 22:30:45.277032 359 sgd_solver.cpp:105] Iteration 684, lr = 0.00999826 I0407 22:30:46.048619 359 blocking_queue.cpp:49] Waiting for data I0407 22:30:50.219533 359 solver.cpp:218] Iteration 696 (2.42793 iter/s, 4.94248s/12 iters), loss = 4.90099 I0407 22:30:50.219574 359 solver.cpp:237] Train net output #0: loss = 4.90099 (* 1 = 4.90099 loss) I0407 22:30:50.219583 359 sgd_solver.cpp:105] Iteration 696, lr = 0.00999822 I0407 22:30:54.781718 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:30:55.152725 359 solver.cpp:218] Iteration 708 (2.43253 iter/s, 4.93313s/12 iters), loss = 5.01674 I0407 22:30:55.152765 359 solver.cpp:237] Train net output #0: loss = 5.01674 (* 1 = 5.01674 loss) I0407 22:30:55.152773 359 sgd_solver.cpp:105] Iteration 708, lr = 0.00999818 I0407 22:30:57.139627 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_714.caffemodel I0407 22:31:02.480870 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_714.solverstate I0407 22:31:04.856963 359 solver.cpp:330] Iteration 714, Testing net (#0) I0407 22:31:04.856981 359 net.cpp:676] Ignoring source layer train-data I0407 22:31:09.474417 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:31:09.809592 359 solver.cpp:397] Test net output #0: accuracy = 0.0349265 I0407 22:31:09.809638 359 solver.cpp:397] Test net output #1: loss = 4.88016 (* 1 = 4.88016 loss) I0407 22:31:11.613674 359 solver.cpp:218] Iteration 720 (0.729002 iter/s, 16.4609s/12 iters), loss = 4.89186 I0407 22:31:11.613708 359 solver.cpp:237] Train net output #0: loss = 4.89186 (* 1 = 4.89186 loss) I0407 22:31:11.613714 359 sgd_solver.cpp:105] Iteration 720, lr = 0.00999814 I0407 22:31:16.581562 359 solver.cpp:218] Iteration 732 (2.41554 iter/s, 4.96783s/12 iters), loss = 4.8736 I0407 22:31:16.581598 359 solver.cpp:237] Train net output #0: loss = 4.8736 (* 1 = 4.8736 loss) I0407 22:31:16.581605 359 sgd_solver.cpp:105] Iteration 732, lr = 0.00999809 I0407 22:31:21.489038 359 solver.cpp:218] Iteration 744 (2.44528 iter/s, 4.90742s/12 iters), loss = 4.85927 I0407 22:31:21.489078 359 solver.cpp:237] Train net output #0: loss = 4.85927 (* 1 = 4.85927 loss) I0407 22:31:21.489084 359 sgd_solver.cpp:105] Iteration 744, lr = 0.00999805 I0407 22:31:26.414517 359 solver.cpp:218] Iteration 756 (2.43634 iter/s, 4.92542s/12 iters), loss = 4.65793 I0407 22:31:26.414664 359 solver.cpp:237] Train net output #0: loss = 4.65793 (* 1 = 4.65793 loss) I0407 22:31:26.414674 359 sgd_solver.cpp:105] Iteration 756, lr = 0.009998 I0407 22:31:31.361675 359 solver.cpp:218] Iteration 768 (2.42572 iter/s, 4.94699s/12 iters), loss = 4.72027 I0407 22:31:31.361716 359 solver.cpp:237] Train net output #0: loss = 4.72027 (* 1 = 4.72027 loss) I0407 22:31:31.361726 359 sgd_solver.cpp:105] Iteration 768, lr = 0.00999795 I0407 22:31:36.293408 359 solver.cpp:218] Iteration 780 (2.43326 iter/s, 4.93166s/12 iters), loss = 4.8095 I0407 22:31:36.293455 359 solver.cpp:237] Train net output #0: loss = 4.8095 (* 1 = 4.8095 loss) I0407 22:31:36.293463 359 sgd_solver.cpp:105] Iteration 780, lr = 0.00999791 I0407 22:31:41.203377 359 solver.cpp:218] Iteration 792 (2.44404 iter/s, 4.9099s/12 iters), loss = 4.91927 I0407 22:31:41.203415 359 solver.cpp:237] Train net output #0: loss = 4.91927 (* 1 = 4.91927 loss) I0407 22:31:41.203423 359 sgd_solver.cpp:105] Iteration 792, lr = 0.00999785 I0407 22:31:46.121747 359 solver.cpp:218] Iteration 804 (2.43986 iter/s, 4.91831s/12 iters), loss = 4.78404 I0407 22:31:46.121783 359 solver.cpp:237] Train net output #0: loss = 4.78404 (* 1 = 4.78404 loss) I0407 22:31:46.121789 359 sgd_solver.cpp:105] Iteration 804, lr = 0.0099978 I0407 22:31:47.830135 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:31:50.604972 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_816.caffemodel I0407 22:31:55.502758 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_816.solverstate I0407 22:31:57.982527 359 solver.cpp:330] Iteration 816, Testing net (#0) I0407 22:31:57.982635 359 net.cpp:676] Ignoring source layer train-data I0407 22:32:02.365247 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:32:02.721709 359 solver.cpp:397] Test net output #0: accuracy = 0.0465686 I0407 22:32:02.721740 359 solver.cpp:397] Test net output #1: loss = 4.76026 (* 1 = 4.76026 loss) I0407 22:32:02.817966 359 solver.cpp:218] Iteration 816 (0.71873 iter/s, 16.6961s/12 iters), loss = 4.67944 I0407 22:32:02.818039 359 solver.cpp:237] Train net output #0: loss = 4.67944 (* 1 = 4.67944 loss) I0407 22:32:02.818055 359 sgd_solver.cpp:105] Iteration 816, lr = 0.00999775 I0407 22:32:06.937144 359 solver.cpp:218] Iteration 828 (2.91327 iter/s, 4.11909s/12 iters), loss = 4.74997 I0407 22:32:06.937181 359 solver.cpp:237] Train net output #0: loss = 4.74997 (* 1 = 4.74997 loss) I0407 22:32:06.937191 359 sgd_solver.cpp:105] Iteration 828, lr = 0.0099977 I0407 22:32:11.848239 359 solver.cpp:218] Iteration 840 (2.44348 iter/s, 4.91104s/12 iters), loss = 4.78239 I0407 22:32:11.848280 359 solver.cpp:237] Train net output #0: loss = 4.78239 (* 1 = 4.78239 loss) I0407 22:32:11.848289 359 sgd_solver.cpp:105] Iteration 840, lr = 0.00999764 I0407 22:32:16.808430 359 solver.cpp:218] Iteration 852 (2.4193 iter/s, 4.96012s/12 iters), loss = 4.65832 I0407 22:32:16.808476 359 solver.cpp:237] Train net output #0: loss = 4.65832 (* 1 = 4.65832 loss) I0407 22:32:16.808485 359 sgd_solver.cpp:105] Iteration 852, lr = 0.00999759 I0407 22:32:21.736987 359 solver.cpp:218] Iteration 864 (2.43482 iter/s, 4.92849s/12 iters), loss = 4.8104 I0407 22:32:21.737030 359 solver.cpp:237] Train net output #0: loss = 4.8104 (* 1 = 4.8104 loss) I0407 22:32:21.737040 359 sgd_solver.cpp:105] Iteration 864, lr = 0.00999753 I0407 22:32:26.718418 359 solver.cpp:218] Iteration 876 (2.40898 iter/s, 4.98136s/12 iters), loss = 4.80148 I0407 22:32:26.718461 359 solver.cpp:237] Train net output #0: loss = 4.80148 (* 1 = 4.80148 loss) I0407 22:32:26.718469 359 sgd_solver.cpp:105] Iteration 876, lr = 0.00999747 I0407 22:32:31.612859 359 solver.cpp:218] Iteration 888 (2.45179 iter/s, 4.89438s/12 iters), loss = 4.79278 I0407 22:32:31.613009 359 solver.cpp:237] Train net output #0: loss = 4.79278 (* 1 = 4.79278 loss) I0407 22:32:31.613018 359 sgd_solver.cpp:105] Iteration 888, lr = 0.00999741 I0407 22:32:36.590762 359 solver.cpp:218] Iteration 900 (2.41074 iter/s, 4.97773s/12 iters), loss = 4.67766 I0407 22:32:36.590807 359 solver.cpp:237] Train net output #0: loss = 4.67766 (* 1 = 4.67766 loss) I0407 22:32:36.590816 359 sgd_solver.cpp:105] Iteration 900, lr = 0.00999735 I0407 22:32:40.446219 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:32:41.531286 359 solver.cpp:218] Iteration 912 (2.42893 iter/s, 4.94045s/12 iters), loss = 4.83716 I0407 22:32:41.531327 359 solver.cpp:237] Train net output #0: loss = 4.83716 (* 1 = 4.83716 loss) I0407 22:32:41.531335 359 sgd_solver.cpp:105] Iteration 912, lr = 0.00999729 I0407 22:32:43.534842 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_918.caffemodel I0407 22:32:48.348703 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_918.solverstate I0407 22:32:53.865550 359 solver.cpp:330] Iteration 918, Testing net (#0) I0407 22:32:53.865566 359 net.cpp:676] Ignoring source layer train-data I0407 22:32:57.934777 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:32:58.338527 359 solver.cpp:397] Test net output #0: accuracy = 0.0526961 I0407 22:32:58.338574 359 solver.cpp:397] Test net output #1: loss = 4.64697 (* 1 = 4.64697 loss) I0407 22:33:00.147841 359 solver.cpp:218] Iteration 924 (0.644591 iter/s, 18.6165s/12 iters), loss = 4.59192 I0407 22:33:00.147883 359 solver.cpp:237] Train net output #0: loss = 4.59192 (* 1 = 4.59192 loss) I0407 22:33:00.147892 359 sgd_solver.cpp:105] Iteration 924, lr = 0.00999722 I0407 22:33:05.110913 359 solver.cpp:218] Iteration 936 (2.41789 iter/s, 4.963s/12 iters), loss = 4.47134 I0407 22:33:05.111032 359 solver.cpp:237] Train net output #0: loss = 4.47134 (* 1 = 4.47134 loss) I0407 22:33:05.111042 359 sgd_solver.cpp:105] Iteration 936, lr = 0.00999716 I0407 22:33:10.071822 359 solver.cpp:218] Iteration 948 (2.41898 iter/s, 4.96077s/12 iters), loss = 4.64827 I0407 22:33:10.071862 359 solver.cpp:237] Train net output #0: loss = 4.64827 (* 1 = 4.64827 loss) I0407 22:33:10.071871 359 sgd_solver.cpp:105] Iteration 948, lr = 0.00999709 I0407 22:33:14.996212 359 solver.cpp:218] Iteration 960 (2.43688 iter/s, 4.92432s/12 iters), loss = 4.65115 I0407 22:33:14.996250 359 solver.cpp:237] Train net output #0: loss = 4.65115 (* 1 = 4.65115 loss) I0407 22:33:14.996259 359 sgd_solver.cpp:105] Iteration 960, lr = 0.00999702 I0407 22:33:19.943707 359 solver.cpp:218] Iteration 972 (2.4255 iter/s, 4.94743s/12 iters), loss = 4.75913 I0407 22:33:19.943754 359 solver.cpp:237] Train net output #0: loss = 4.75913 (* 1 = 4.75913 loss) I0407 22:33:19.943763 359 sgd_solver.cpp:105] Iteration 972, lr = 0.00999695 I0407 22:33:24.741403 359 solver.cpp:218] Iteration 984 (2.50124 iter/s, 4.79763s/12 iters), loss = 4.52512 I0407 22:33:24.741446 359 solver.cpp:237] Train net output #0: loss = 4.52512 (* 1 = 4.52512 loss) I0407 22:33:24.741453 359 sgd_solver.cpp:105] Iteration 984, lr = 0.00999688 I0407 22:33:29.687659 359 solver.cpp:218] Iteration 996 (2.42611 iter/s, 4.94619s/12 iters), loss = 4.73446 I0407 22:33:29.687703 359 solver.cpp:237] Train net output #0: loss = 4.73446 (* 1 = 4.73446 loss) I0407 22:33:29.687711 359 sgd_solver.cpp:105] Iteration 996, lr = 0.0099968 I0407 22:33:34.659121 359 solver.cpp:218] Iteration 1008 (2.41381 iter/s, 4.97139s/12 iters), loss = 4.23608 I0407 22:33:34.659162 359 solver.cpp:237] Train net output #0: loss = 4.23608 (* 1 = 4.23608 loss) I0407 22:33:34.659169 359 sgd_solver.cpp:105] Iteration 1008, lr = 0.00999672 I0407 22:33:35.640033 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:33:39.085248 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_1020.caffemodel I0407 22:33:41.908701 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_1020.solverstate I0407 22:33:44.267123 359 solver.cpp:330] Iteration 1020, Testing net (#0) I0407 22:33:44.267141 359 net.cpp:676] Ignoring source layer train-data I0407 22:33:48.382591 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:33:48.822115 359 solver.cpp:397] Test net output #0: accuracy = 0.0612745 I0407 22:33:48.822146 359 solver.cpp:397] Test net output #1: loss = 4.49162 (* 1 = 4.49162 loss) I0407 22:33:48.918799 359 solver.cpp:218] Iteration 1020 (0.841539 iter/s, 14.2596s/12 iters), loss = 4.52069 I0407 22:33:48.918857 359 solver.cpp:237] Train net output #0: loss = 4.52069 (* 1 = 4.52069 loss) I0407 22:33:48.918869 359 sgd_solver.cpp:105] Iteration 1020, lr = 0.00999665 I0407 22:33:53.033442 359 solver.cpp:218] Iteration 1032 (2.91647 iter/s, 4.11456s/12 iters), loss = 4.36611 I0407 22:33:53.033484 359 solver.cpp:237] Train net output #0: loss = 4.36611 (* 1 = 4.36611 loss) I0407 22:33:53.033493 359 sgd_solver.cpp:105] Iteration 1032, lr = 0.00999657 I0407 22:33:57.986956 359 solver.cpp:218] Iteration 1044 (2.42255 iter/s, 4.95345s/12 iters), loss = 4.5116 I0407 22:33:57.986992 359 solver.cpp:237] Train net output #0: loss = 4.5116 (* 1 = 4.5116 loss) I0407 22:33:57.987000 359 sgd_solver.cpp:105] Iteration 1044, lr = 0.00999648 I0407 22:34:02.841928 359 solver.cpp:218] Iteration 1056 (2.47172 iter/s, 4.85491s/12 iters), loss = 4.39033 I0407 22:34:02.841967 359 solver.cpp:237] Train net output #0: loss = 4.39033 (* 1 = 4.39033 loss) I0407 22:34:02.841975 359 sgd_solver.cpp:105] Iteration 1056, lr = 0.0099964 I0407 22:34:07.756639 359 solver.cpp:218] Iteration 1068 (2.44168 iter/s, 4.91465s/12 iters), loss = 4.31935 I0407 22:34:07.756770 359 solver.cpp:237] Train net output #0: loss = 4.31935 (* 1 = 4.31935 loss) I0407 22:34:07.756779 359 sgd_solver.cpp:105] Iteration 1068, lr = 0.00999632 I0407 22:34:12.726732 359 solver.cpp:218] Iteration 1080 (2.41452 iter/s, 4.96994s/12 iters), loss = 4.47299 I0407 22:34:12.726776 359 solver.cpp:237] Train net output #0: loss = 4.47299 (* 1 = 4.47299 loss) I0407 22:34:12.726785 359 sgd_solver.cpp:105] Iteration 1080, lr = 0.00999623 I0407 22:34:17.582882 359 solver.cpp:218] Iteration 1092 (2.47113 iter/s, 4.85609s/12 iters), loss = 4.21391 I0407 22:34:17.582921 359 solver.cpp:237] Train net output #0: loss = 4.21391 (* 1 = 4.21391 loss) I0407 22:34:17.582929 359 sgd_solver.cpp:105] Iteration 1092, lr = 0.00999614 I0407 22:34:22.533805 359 solver.cpp:218] Iteration 1104 (2.42382 iter/s, 4.95086s/12 iters), loss = 4.69628 I0407 22:34:22.533843 359 solver.cpp:237] Train net output #0: loss = 4.69628 (* 1 = 4.69628 loss) I0407 22:34:22.533850 359 sgd_solver.cpp:105] Iteration 1104, lr = 0.00999605 I0407 22:34:25.619745 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:34:27.427727 359 solver.cpp:218] Iteration 1116 (2.45205 iter/s, 4.89386s/12 iters), loss = 4.34636 I0407 22:34:27.427764 359 solver.cpp:237] Train net output #0: loss = 4.34636 (* 1 = 4.34636 loss) I0407 22:34:27.427772 359 sgd_solver.cpp:105] Iteration 1116, lr = 0.00999595 I0407 22:34:29.422978 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_1122.caffemodel I0407 22:34:32.512549 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_1122.solverstate I0407 22:34:36.219897 359 solver.cpp:330] Iteration 1122, Testing net (#0) I0407 22:34:36.219918 359 net.cpp:676] Ignoring source layer train-data I0407 22:34:40.466763 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:34:40.992725 359 solver.cpp:397] Test net output #0: accuracy = 0.0778186 I0407 22:34:40.992771 359 solver.cpp:397] Test net output #1: loss = 4.36458 (* 1 = 4.36458 loss) I0407 22:34:42.793181 359 solver.cpp:218] Iteration 1128 (0.780977 iter/s, 15.3654s/12 iters), loss = 4.1891 I0407 22:34:42.793221 359 solver.cpp:237] Train net output #0: loss = 4.1891 (* 1 = 4.1891 loss) I0407 22:34:42.793229 359 sgd_solver.cpp:105] Iteration 1128, lr = 0.00999586 I0407 22:34:47.728471 359 solver.cpp:218] Iteration 1140 (2.4315 iter/s, 4.93522s/12 iters), loss = 4.35371 I0407 22:34:47.728518 359 solver.cpp:237] Train net output #0: loss = 4.35371 (* 1 = 4.35371 loss) I0407 22:34:47.728528 359 sgd_solver.cpp:105] Iteration 1140, lr = 0.00999576 I0407 22:34:52.632295 359 solver.cpp:218] Iteration 1152 (2.4471 iter/s, 4.90376s/12 iters), loss = 3.96796 I0407 22:34:52.632331 359 solver.cpp:237] Train net output #0: loss = 3.96796 (* 1 = 3.96796 loss) I0407 22:34:52.632339 359 sgd_solver.cpp:105] Iteration 1152, lr = 0.00999566 I0407 22:34:57.603988 359 solver.cpp:218] Iteration 1164 (2.41369 iter/s, 4.97163s/12 iters), loss = 4.15408 I0407 22:34:57.604030 359 solver.cpp:237] Train net output #0: loss = 4.15408 (* 1 = 4.15408 loss) I0407 22:34:57.604039 359 sgd_solver.cpp:105] Iteration 1164, lr = 0.00999555 I0407 22:35:02.549484 359 solver.cpp:218] Iteration 1176 (2.42648 iter/s, 4.94543s/12 iters), loss = 4.15201 I0407 22:35:02.549527 359 solver.cpp:237] Train net output #0: loss = 4.15201 (* 1 = 4.15201 loss) I0407 22:35:02.549535 359 sgd_solver.cpp:105] Iteration 1176, lr = 0.00999545 I0407 22:35:07.461488 359 solver.cpp:218] Iteration 1188 (2.44303 iter/s, 4.91194s/12 iters), loss = 3.98278 I0407 22:35:07.461529 359 solver.cpp:237] Train net output #0: loss = 3.98278 (* 1 = 3.98278 loss) I0407 22:35:07.461537 359 sgd_solver.cpp:105] Iteration 1188, lr = 0.00999534 I0407 22:35:12.443728 359 solver.cpp:218] Iteration 1200 (2.40859 iter/s, 4.98218s/12 iters), loss = 4.23676 I0407 22:35:12.443876 359 solver.cpp:237] Train net output #0: loss = 4.23676 (* 1 = 4.23676 loss) I0407 22:35:12.443886 359 sgd_solver.cpp:105] Iteration 1200, lr = 0.00999523 I0407 22:35:17.360339 359 solver.cpp:218] Iteration 1212 (2.44079 iter/s, 4.91644s/12 iters), loss = 4.04756 I0407 22:35:17.360383 359 solver.cpp:237] Train net output #0: loss = 4.04756 (* 1 = 4.04756 loss) I0407 22:35:17.360391 359 sgd_solver.cpp:105] Iteration 1212, lr = 0.00999511 I0407 22:35:17.625289 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:35:21.844188 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_1224.caffemodel I0407 22:35:24.919265 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_1224.solverstate I0407 22:35:27.275514 359 solver.cpp:330] Iteration 1224, Testing net (#0) I0407 22:35:27.275533 359 net.cpp:676] Ignoring source layer train-data I0407 22:35:31.362200 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:35:31.937912 359 solver.cpp:397] Test net output #0: accuracy = 0.0808824 I0407 22:35:31.937959 359 solver.cpp:397] Test net output #1: loss = 4.26455 (* 1 = 4.26455 loss) I0407 22:35:32.033993 359 solver.cpp:218] Iteration 1224 (0.817797 iter/s, 14.6736s/12 iters), loss = 4.16022 I0407 22:35:32.034037 359 solver.cpp:237] Train net output #0: loss = 4.16022 (* 1 = 4.16022 loss) I0407 22:35:32.034045 359 sgd_solver.cpp:105] Iteration 1224, lr = 0.009995 I0407 22:35:36.215714 359 solver.cpp:218] Iteration 1236 (2.86968 iter/s, 4.18166s/12 iters), loss = 4.14807 I0407 22:35:36.215751 359 solver.cpp:237] Train net output #0: loss = 4.14807 (* 1 = 4.14807 loss) I0407 22:35:36.215762 359 sgd_solver.cpp:105] Iteration 1236, lr = 0.00999488 I0407 22:35:41.153381 359 solver.cpp:218] Iteration 1248 (2.43033 iter/s, 4.93761s/12 iters), loss = 4.01501 I0407 22:35:41.153419 359 solver.cpp:237] Train net output #0: loss = 4.01501 (* 1 = 4.01501 loss) I0407 22:35:41.153427 359 sgd_solver.cpp:105] Iteration 1248, lr = 0.00999476 I0407 22:35:46.095556 359 solver.cpp:218] Iteration 1260 (2.42811 iter/s, 4.94211s/12 iters), loss = 3.90919 I0407 22:35:46.095719 359 solver.cpp:237] Train net output #0: loss = 3.90919 (* 1 = 3.90919 loss) I0407 22:35:46.095729 359 sgd_solver.cpp:105] Iteration 1260, lr = 0.00999463 I0407 22:35:50.956984 359 solver.cpp:218] Iteration 1272 (2.4685 iter/s, 4.86125s/12 iters), loss = 3.89158 I0407 22:35:50.957023 359 solver.cpp:237] Train net output #0: loss = 3.89158 (* 1 = 3.89158 loss) I0407 22:35:50.957031 359 sgd_solver.cpp:105] Iteration 1272, lr = 0.0099945 I0407 22:35:55.918812 359 solver.cpp:218] Iteration 1284 (2.41849 iter/s, 4.96177s/12 iters), loss = 3.97335 I0407 22:35:55.918848 359 solver.cpp:237] Train net output #0: loss = 3.97335 (* 1 = 3.97335 loss) I0407 22:35:55.918855 359 sgd_solver.cpp:105] Iteration 1284, lr = 0.00999437 I0407 22:36:00.840694 359 solver.cpp:218] Iteration 1296 (2.43812 iter/s, 4.92182s/12 iters), loss = 4.00292 I0407 22:36:00.840731 359 solver.cpp:237] Train net output #0: loss = 4.00292 (* 1 = 4.00292 loss) I0407 22:36:00.840739 359 sgd_solver.cpp:105] Iteration 1296, lr = 0.00999424 I0407 22:36:05.786229 359 solver.cpp:218] Iteration 1308 (2.42646 iter/s, 4.94548s/12 iters), loss = 4.08033 I0407 22:36:05.786267 359 solver.cpp:237] Train net output #0: loss = 4.08033 (* 1 = 4.08033 loss) I0407 22:36:05.786275 359 sgd_solver.cpp:105] Iteration 1308, lr = 0.0099941 I0407 22:36:08.250079 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:36:10.687924 359 solver.cpp:218] Iteration 1320 (2.44816 iter/s, 4.90163s/12 iters), loss = 4.1993 I0407 22:36:10.687971 359 solver.cpp:237] Train net output #0: loss = 4.1993 (* 1 = 4.1993 loss) I0407 22:36:10.687980 359 sgd_solver.cpp:105] Iteration 1320, lr = 0.00999396 I0407 22:36:12.724865 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_1326.caffemodel I0407 22:36:16.704424 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_1326.solverstate I0407 22:36:19.068617 359 solver.cpp:330] Iteration 1326, Testing net (#0) I0407 22:36:19.068635 359 net.cpp:676] Ignoring source layer train-data I0407 22:36:23.350137 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:36:24.033901 359 solver.cpp:397] Test net output #0: accuracy = 0.115196 I0407 22:36:24.033941 359 solver.cpp:397] Test net output #1: loss = 4.0472 (* 1 = 4.0472 loss) I0407 22:36:25.826490 359 solver.cpp:218] Iteration 1332 (0.792682 iter/s, 15.1385s/12 iters), loss = 3.8674 I0407 22:36:25.826531 359 solver.cpp:237] Train net output #0: loss = 3.8674 (* 1 = 3.8674 loss) I0407 22:36:25.826539 359 sgd_solver.cpp:105] Iteration 1332, lr = 0.00999382 I0407 22:36:30.738754 359 solver.cpp:218] Iteration 1344 (2.4429 iter/s, 4.9122s/12 iters), loss = 3.83042 I0407 22:36:30.738801 359 solver.cpp:237] Train net output #0: loss = 3.83042 (* 1 = 3.83042 loss) I0407 22:36:30.738811 359 sgd_solver.cpp:105] Iteration 1344, lr = 0.00999367 I0407 22:36:35.691824 359 solver.cpp:218] Iteration 1356 (2.42277 iter/s, 4.953s/12 iters), loss = 4.03057 I0407 22:36:35.691864 359 solver.cpp:237] Train net output #0: loss = 4.03057 (* 1 = 4.03057 loss) I0407 22:36:35.691872 359 sgd_solver.cpp:105] Iteration 1356, lr = 0.00999352 I0407 22:36:40.628046 359 solver.cpp:218] Iteration 1368 (2.43104 iter/s, 4.93616s/12 iters), loss = 3.99038 I0407 22:36:40.628091 359 solver.cpp:237] Train net output #0: loss = 3.99038 (* 1 = 3.99038 loss) I0407 22:36:40.628099 359 sgd_solver.cpp:105] Iteration 1368, lr = 0.00999337 I0407 22:36:41.805462 359 blocking_queue.cpp:49] Waiting for data I0407 22:36:45.591382 359 solver.cpp:218] Iteration 1380 (2.41776 iter/s, 4.96326s/12 iters), loss = 3.90185 I0407 22:36:45.591426 359 solver.cpp:237] Train net output #0: loss = 3.90185 (* 1 = 3.90185 loss) I0407 22:36:45.591434 359 sgd_solver.cpp:105] Iteration 1380, lr = 0.00999321 I0407 22:36:50.489408 359 solver.cpp:218] Iteration 1392 (2.45 iter/s, 4.89796s/12 iters), loss = 4.01423 I0407 22:36:50.489535 359 solver.cpp:237] Train net output #0: loss = 4.01423 (* 1 = 4.01423 loss) I0407 22:36:50.489545 359 sgd_solver.cpp:105] Iteration 1392, lr = 0.00999305 I0407 22:36:55.458361 359 solver.cpp:218] Iteration 1404 (2.41507 iter/s, 4.9688s/12 iters), loss = 3.78515 I0407 22:36:55.458403 359 solver.cpp:237] Train net output #0: loss = 3.78515 (* 1 = 3.78515 loss) I0407 22:36:55.458412 359 sgd_solver.cpp:105] Iteration 1404, lr = 0.00999288 I0407 22:37:00.012928 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:37:00.354723 359 solver.cpp:218] Iteration 1416 (2.45083 iter/s, 4.8963s/12 iters), loss = 3.86286 I0407 22:37:00.354764 359 solver.cpp:237] Train net output #0: loss = 3.86286 (* 1 = 3.86286 loss) I0407 22:37:00.354773 359 sgd_solver.cpp:105] Iteration 1416, lr = 0.00999271 I0407 22:37:04.854851 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_1428.caffemodel I0407 22:37:08.827879 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_1428.solverstate I0407 22:37:11.759481 359 solver.cpp:330] Iteration 1428, Testing net (#0) I0407 22:37:11.759498 359 net.cpp:676] Ignoring source layer train-data I0407 22:37:15.826653 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:37:16.482923 359 solver.cpp:397] Test net output #0: accuracy = 0.110294 I0407 22:37:16.482971 359 solver.cpp:397] Test net output #1: loss = 4.01474 (* 1 = 4.01474 loss) I0407 22:37:16.579562 359 solver.cpp:218] Iteration 1428 (0.739611 iter/s, 16.2248s/12 iters), loss = 3.76934 I0407 22:37:16.579612 359 solver.cpp:237] Train net output #0: loss = 3.76934 (* 1 = 3.76934 loss) I0407 22:37:16.579619 359 sgd_solver.cpp:105] Iteration 1428, lr = 0.00999254 I0407 22:37:20.879366 359 solver.cpp:218] Iteration 1440 (2.79087 iter/s, 4.29974s/12 iters), loss = 3.79479 I0407 22:37:20.879532 359 solver.cpp:237] Train net output #0: loss = 3.79479 (* 1 = 3.79479 loss) I0407 22:37:20.879541 359 sgd_solver.cpp:105] Iteration 1440, lr = 0.00999236 I0407 22:37:25.823222 359 solver.cpp:218] Iteration 1452 (2.42735 iter/s, 4.94367s/12 iters), loss = 3.83387 I0407 22:37:25.823263 359 solver.cpp:237] Train net output #0: loss = 3.83387 (* 1 = 3.83387 loss) I0407 22:37:25.823272 359 sgd_solver.cpp:105] Iteration 1452, lr = 0.00999218 I0407 22:37:30.762645 359 solver.cpp:218] Iteration 1464 (2.42946 iter/s, 4.93936s/12 iters), loss = 3.49704 I0407 22:37:30.762682 359 solver.cpp:237] Train net output #0: loss = 3.49704 (* 1 = 3.49704 loss) I0407 22:37:30.762691 359 sgd_solver.cpp:105] Iteration 1464, lr = 0.00999199 I0407 22:37:35.785454 359 solver.cpp:218] Iteration 1476 (2.38913 iter/s, 5.02275s/12 iters), loss = 3.77816 I0407 22:37:35.785496 359 solver.cpp:237] Train net output #0: loss = 3.77816 (* 1 = 3.77816 loss) I0407 22:37:35.785504 359 sgd_solver.cpp:105] Iteration 1476, lr = 0.0099918 I0407 22:37:40.759732 359 solver.cpp:218] Iteration 1488 (2.41244 iter/s, 4.97421s/12 iters), loss = 3.68157 I0407 22:37:40.759775 359 solver.cpp:237] Train net output #0: loss = 3.68157 (* 1 = 3.68157 loss) I0407 22:37:40.759783 359 sgd_solver.cpp:105] Iteration 1488, lr = 0.00999161 I0407 22:37:45.792701 359 solver.cpp:218] Iteration 1500 (2.38431 iter/s, 5.0329s/12 iters), loss = 3.85981 I0407 22:37:45.792744 359 solver.cpp:237] Train net output #0: loss = 3.85981 (* 1 = 3.85981 loss) I0407 22:37:45.792752 359 sgd_solver.cpp:105] Iteration 1500, lr = 0.00999141 I0407 22:37:50.967263 359 solver.cpp:218] Iteration 1512 (2.31907 iter/s, 5.1745s/12 iters), loss = 3.58055 I0407 22:37:50.967398 359 solver.cpp:237] Train net output #0: loss = 3.58055 (* 1 = 3.58055 loss) I0407 22:37:50.967407 359 sgd_solver.cpp:105] Iteration 1512, lr = 0.00999121 I0407 22:37:52.703716 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:37:55.821451 359 solver.cpp:218] Iteration 1524 (2.47217 iter/s, 4.85404s/12 iters), loss = 3.73534 I0407 22:37:55.821487 359 solver.cpp:237] Train net output #0: loss = 3.73534 (* 1 = 3.73534 loss) I0407 22:37:55.821494 359 sgd_solver.cpp:105] Iteration 1524, lr = 0.009991 I0407 22:37:57.910384 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_1530.caffemodel I0407 22:38:01.042002 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_1530.solverstate I0407 22:38:03.403066 359 solver.cpp:330] Iteration 1530, Testing net (#0) I0407 22:38:03.403086 359 net.cpp:676] Ignoring source layer train-data I0407 22:38:07.249096 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:38:07.957537 359 solver.cpp:397] Test net output #0: accuracy = 0.131127 I0407 22:38:07.957581 359 solver.cpp:397] Test net output #1: loss = 3.87971 (* 1 = 3.87971 loss) I0407 22:38:09.773028 359 solver.cpp:218] Iteration 1536 (0.860122 iter/s, 13.9515s/12 iters), loss = 3.37834 I0407 22:38:09.773062 359 solver.cpp:237] Train net output #0: loss = 3.37834 (* 1 = 3.37834 loss) I0407 22:38:09.773070 359 sgd_solver.cpp:105] Iteration 1536, lr = 0.00999078 I0407 22:38:14.798009 359 solver.cpp:218] Iteration 1548 (2.3881 iter/s, 5.02493s/12 iters), loss = 3.56426 I0407 22:38:14.798050 359 solver.cpp:237] Train net output #0: loss = 3.56426 (* 1 = 3.56426 loss) I0407 22:38:14.798058 359 sgd_solver.cpp:105] Iteration 1548, lr = 0.00999056 I0407 22:38:19.762490 359 solver.cpp:218] Iteration 1560 (2.4172 iter/s, 4.96441s/12 iters), loss = 3.63487 I0407 22:38:19.762537 359 solver.cpp:237] Train net output #0: loss = 3.63487 (* 1 = 3.63487 loss) I0407 22:38:19.762545 359 sgd_solver.cpp:105] Iteration 1560, lr = 0.00999034 I0407 22:38:24.742230 359 solver.cpp:218] Iteration 1572 (2.4098 iter/s, 4.97967s/12 iters), loss = 3.66285 I0407 22:38:24.742388 359 solver.cpp:237] Train net output #0: loss = 3.66285 (* 1 = 3.66285 loss) I0407 22:38:24.742404 359 sgd_solver.cpp:105] Iteration 1572, lr = 0.00999011 I0407 22:38:29.863627 359 solver.cpp:218] Iteration 1584 (2.34319 iter/s, 5.12122s/12 iters), loss = 3.51581 I0407 22:38:29.863664 359 solver.cpp:237] Train net output #0: loss = 3.51581 (* 1 = 3.51581 loss) I0407 22:38:29.863672 359 sgd_solver.cpp:105] Iteration 1584, lr = 0.00998987 I0407 22:38:34.847849 359 solver.cpp:218] Iteration 1596 (2.40763 iter/s, 4.98416s/12 iters), loss = 3.26739 I0407 22:38:34.847888 359 solver.cpp:237] Train net output #0: loss = 3.26739 (* 1 = 3.26739 loss) I0407 22:38:34.847895 359 sgd_solver.cpp:105] Iteration 1596, lr = 0.00998963 I0407 22:38:40.061777 359 solver.cpp:218] Iteration 1608 (2.30155 iter/s, 5.21387s/12 iters), loss = 3.56462 I0407 22:38:40.061815 359 solver.cpp:237] Train net output #0: loss = 3.56462 (* 1 = 3.56462 loss) I0407 22:38:40.061822 359 sgd_solver.cpp:105] Iteration 1608, lr = 0.00998939 I0407 22:38:43.895316 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:38:44.947158 359 solver.cpp:218] Iteration 1620 (2.45634 iter/s, 4.88532s/12 iters), loss = 3.76939 I0407 22:38:44.947208 359 solver.cpp:237] Train net output #0: loss = 3.76939 (* 1 = 3.76939 loss) I0407 22:38:44.947218 359 sgd_solver.cpp:105] Iteration 1620, lr = 0.00998913 I0407 22:38:49.420012 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_1632.caffemodel I0407 22:38:52.522264 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_1632.solverstate I0407 22:38:54.877012 359 solver.cpp:330] Iteration 1632, Testing net (#0) I0407 22:38:54.877141 359 net.cpp:676] Ignoring source layer train-data I0407 22:38:58.904289 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:38:59.651060 359 solver.cpp:397] Test net output #0: accuracy = 0.150735 I0407 22:38:59.651093 359 solver.cpp:397] Test net output #1: loss = 3.72563 (* 1 = 3.72563 loss) I0407 22:38:59.747587 359 solver.cpp:218] Iteration 1632 (0.810792 iter/s, 14.8003s/12 iters), loss = 3.5402 I0407 22:38:59.747635 359 solver.cpp:237] Train net output #0: loss = 3.5402 (* 1 = 3.5402 loss) I0407 22:38:59.747643 359 sgd_solver.cpp:105] Iteration 1632, lr = 0.00998887 I0407 22:39:03.867409 359 solver.cpp:218] Iteration 1644 (2.91279 iter/s, 4.11976s/12 iters), loss = 3.39617 I0407 22:39:03.867441 359 solver.cpp:237] Train net output #0: loss = 3.39617 (* 1 = 3.39617 loss) I0407 22:39:03.867449 359 sgd_solver.cpp:105] Iteration 1644, lr = 0.00998861 I0407 22:39:08.843333 359 solver.cpp:218] Iteration 1656 (2.41164 iter/s, 4.97587s/12 iters), loss = 3.47303 I0407 22:39:08.843375 359 solver.cpp:237] Train net output #0: loss = 3.47303 (* 1 = 3.47303 loss) I0407 22:39:08.843384 359 sgd_solver.cpp:105] Iteration 1656, lr = 0.00998834 I0407 22:39:13.711283 359 solver.cpp:218] Iteration 1668 (2.46514 iter/s, 4.86789s/12 iters), loss = 3.42902 I0407 22:39:13.711319 359 solver.cpp:237] Train net output #0: loss = 3.42902 (* 1 = 3.42902 loss) I0407 22:39:13.711328 359 sgd_solver.cpp:105] Iteration 1668, lr = 0.00998806 I0407 22:39:18.650694 359 solver.cpp:218] Iteration 1680 (2.42947 iter/s, 4.93935s/12 iters), loss = 3.12802 I0407 22:39:18.650739 359 solver.cpp:237] Train net output #0: loss = 3.12802 (* 1 = 3.12802 loss) I0407 22:39:18.650748 359 sgd_solver.cpp:105] Iteration 1680, lr = 0.00998778 I0407 22:39:23.563797 359 solver.cpp:218] Iteration 1692 (2.44248 iter/s, 4.91303s/12 iters), loss = 3.12681 I0407 22:39:23.563850 359 solver.cpp:237] Train net output #0: loss = 3.12681 (* 1 = 3.12681 loss) I0407 22:39:23.563860 359 sgd_solver.cpp:105] Iteration 1692, lr = 0.00998749 I0407 22:39:28.526150 359 solver.cpp:218] Iteration 1704 (2.41824 iter/s, 4.96228s/12 iters), loss = 3.151 I0407 22:39:28.526324 359 solver.cpp:237] Train net output #0: loss = 3.151 (* 1 = 3.151 loss) I0407 22:39:28.526332 359 sgd_solver.cpp:105] Iteration 1704, lr = 0.00998719 I0407 22:39:33.444808 359 solver.cpp:218] Iteration 1716 (2.43978 iter/s, 4.91847s/12 iters), loss = 3.32572 I0407 22:39:33.444845 359 solver.cpp:237] Train net output #0: loss = 3.32572 (* 1 = 3.32572 loss) I0407 22:39:33.444852 359 sgd_solver.cpp:105] Iteration 1716, lr = 0.00998688 I0407 22:39:34.456984 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:39:38.346740 359 solver.cpp:218] Iteration 1728 (2.44804 iter/s, 4.90187s/12 iters), loss = 3.33658 I0407 22:39:38.346781 359 solver.cpp:237] Train net output #0: loss = 3.33658 (* 1 = 3.33658 loss) I0407 22:39:38.346788 359 sgd_solver.cpp:105] Iteration 1728, lr = 0.00998657 I0407 22:39:40.381992 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_1734.caffemodel I0407 22:39:43.566239 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_1734.solverstate I0407 22:39:45.939190 359 solver.cpp:330] Iteration 1734, Testing net (#0) I0407 22:39:45.939219 359 net.cpp:676] Ignoring source layer train-data I0407 22:39:49.854434 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:39:50.574000 359 solver.cpp:397] Test net output #0: accuracy = 0.175245 I0407 22:39:50.574049 359 solver.cpp:397] Test net output #1: loss = 3.626 (* 1 = 3.626 loss) I0407 22:39:52.444792 359 solver.cpp:218] Iteration 1740 (0.851186 iter/s, 14.098s/12 iters), loss = 3.19325 I0407 22:39:52.444828 359 solver.cpp:237] Train net output #0: loss = 3.19325 (* 1 = 3.19325 loss) I0407 22:39:52.444835 359 sgd_solver.cpp:105] Iteration 1740, lr = 0.00998625 I0407 22:39:57.374161 359 solver.cpp:218] Iteration 1752 (2.43441 iter/s, 4.92932s/12 iters), loss = 3.13466 I0407 22:39:57.374197 359 solver.cpp:237] Train net output #0: loss = 3.13466 (* 1 = 3.13466 loss) I0407 22:39:57.374204 359 sgd_solver.cpp:105] Iteration 1752, lr = 0.00998593 I0407 22:40:02.352735 359 solver.cpp:218] Iteration 1764 (2.41036 iter/s, 4.97852s/12 iters), loss = 3.22794 I0407 22:40:02.352860 359 solver.cpp:237] Train net output #0: loss = 3.22794 (* 1 = 3.22794 loss) I0407 22:40:02.352869 359 sgd_solver.cpp:105] Iteration 1764, lr = 0.00998559 I0407 22:40:07.304322 359 solver.cpp:218] Iteration 1776 (2.42354 iter/s, 4.95144s/12 iters), loss = 3.18758 I0407 22:40:07.304366 359 solver.cpp:237] Train net output #0: loss = 3.18758 (* 1 = 3.18758 loss) I0407 22:40:07.304374 359 sgd_solver.cpp:105] Iteration 1776, lr = 0.00998525 I0407 22:40:12.302299 359 solver.cpp:218] Iteration 1788 (2.40101 iter/s, 4.9979s/12 iters), loss = 3.0682 I0407 22:40:12.302345 359 solver.cpp:237] Train net output #0: loss = 3.0682 (* 1 = 3.0682 loss) I0407 22:40:12.302353 359 sgd_solver.cpp:105] Iteration 1788, lr = 0.0099849 I0407 22:40:17.184746 359 solver.cpp:218] Iteration 1800 (2.45782 iter/s, 4.88238s/12 iters), loss = 2.91501 I0407 22:40:17.184787 359 solver.cpp:237] Train net output #0: loss = 2.91501 (* 1 = 2.91501 loss) I0407 22:40:17.184796 359 sgd_solver.cpp:105] Iteration 1800, lr = 0.00998454 I0407 22:40:22.162169 359 solver.cpp:218] Iteration 1812 (2.41092 iter/s, 4.97735s/12 iters), loss = 3.29513 I0407 22:40:22.162211 359 solver.cpp:237] Train net output #0: loss = 3.29513 (* 1 = 3.29513 loss) I0407 22:40:22.162220 359 sgd_solver.cpp:105] Iteration 1812, lr = 0.00998417 I0407 22:40:25.314308 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:40:27.094609 359 solver.cpp:218] Iteration 1824 (2.43291 iter/s, 4.93237s/12 iters), loss = 3.4254 I0407 22:40:27.094653 359 solver.cpp:237] Train net output #0: loss = 3.4254 (* 1 = 3.4254 loss) I0407 22:40:27.094661 359 sgd_solver.cpp:105] Iteration 1824, lr = 0.0099838 I0407 22:40:31.569416 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_1836.caffemodel I0407 22:40:34.641258 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_1836.solverstate I0407 22:40:37.078544 359 solver.cpp:330] Iteration 1836, Testing net (#0) I0407 22:40:37.078564 359 net.cpp:676] Ignoring source layer train-data I0407 22:40:41.028375 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:40:41.862079 359 solver.cpp:397] Test net output #0: accuracy = 0.181373 I0407 22:40:41.862124 359 solver.cpp:397] Test net output #1: loss = 3.53646 (* 1 = 3.53646 loss) I0407 22:40:41.959156 359 solver.cpp:218] Iteration 1836 (0.807295 iter/s, 14.8645s/12 iters), loss = 3.25231 I0407 22:40:41.959223 359 solver.cpp:237] Train net output #0: loss = 3.25231 (* 1 = 3.25231 loss) I0407 22:40:41.959233 359 sgd_solver.cpp:105] Iteration 1836, lr = 0.00998341 I0407 22:40:46.102376 359 solver.cpp:218] Iteration 1848 (2.89636 iter/s, 4.14314s/12 iters), loss = 3.28534 I0407 22:40:46.102421 359 solver.cpp:237] Train net output #0: loss = 3.28534 (* 1 = 3.28534 loss) I0407 22:40:46.102429 359 sgd_solver.cpp:105] Iteration 1848, lr = 0.00998302 I0407 22:40:51.040313 359 solver.cpp:218] Iteration 1860 (2.4302 iter/s, 4.93787s/12 iters), loss = 2.97721 I0407 22:40:51.040352 359 solver.cpp:237] Train net output #0: loss = 2.97721 (* 1 = 2.97721 loss) I0407 22:40:51.040360 359 sgd_solver.cpp:105] Iteration 1860, lr = 0.00998261 I0407 22:40:55.966456 359 solver.cpp:218] Iteration 1872 (2.43601 iter/s, 4.92609s/12 iters), loss = 2.8339 I0407 22:40:55.966493 359 solver.cpp:237] Train net output #0: loss = 2.8339 (* 1 = 2.8339 loss) I0407 22:40:55.966501 359 sgd_solver.cpp:105] Iteration 1872, lr = 0.0099822 I0407 22:41:00.919643 359 solver.cpp:218] Iteration 1884 (2.42271 iter/s, 4.95313s/12 iters), loss = 3.01308 I0407 22:41:00.919684 359 solver.cpp:237] Train net output #0: loss = 3.01308 (* 1 = 3.01308 loss) I0407 22:41:00.919692 359 sgd_solver.cpp:105] Iteration 1884, lr = 0.00998178 I0407 22:41:05.817286 359 solver.cpp:218] Iteration 1896 (2.45019 iter/s, 4.89758s/12 iters), loss = 3.05372 I0407 22:41:05.817401 359 solver.cpp:237] Train net output #0: loss = 3.05372 (* 1 = 3.05372 loss) I0407 22:41:05.817411 359 sgd_solver.cpp:105] Iteration 1896, lr = 0.00998134 I0407 22:41:10.788357 359 solver.cpp:218] Iteration 1908 (2.41403 iter/s, 4.97094s/12 iters), loss = 2.96224 I0407 22:41:10.788396 359 solver.cpp:237] Train net output #0: loss = 2.96224 (* 1 = 2.96224 loss) I0407 22:41:10.788403 359 sgd_solver.cpp:105] Iteration 1908, lr = 0.0099809 I0407 22:41:15.696400 359 solver.cpp:218] Iteration 1920 (2.445 iter/s, 4.90798s/12 iters), loss = 2.92034 I0407 22:41:15.696446 359 solver.cpp:237] Train net output #0: loss = 2.92034 (* 1 = 2.92034 loss) I0407 22:41:15.696455 359 sgd_solver.cpp:105] Iteration 1920, lr = 0.00998045 I0407 22:41:15.991915 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:41:20.653021 359 solver.cpp:218] Iteration 1932 (2.42104 iter/s, 4.95655s/12 iters), loss = 3.14069 I0407 22:41:20.653064 359 solver.cpp:237] Train net output #0: loss = 3.14069 (* 1 = 3.14069 loss) I0407 22:41:20.653072 359 sgd_solver.cpp:105] Iteration 1932, lr = 0.00997998 I0407 22:41:22.773319 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_1938.caffemodel I0407 22:41:25.864236 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_1938.solverstate I0407 22:41:28.227759 359 solver.cpp:330] Iteration 1938, Testing net (#0) I0407 22:41:28.227777 359 net.cpp:676] Ignoring source layer train-data I0407 22:41:32.138288 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:41:32.936493 359 solver.cpp:397] Test net output #0: accuracy = 0.193015 I0407 22:41:32.936539 359 solver.cpp:397] Test net output #1: loss = 3.47011 (* 1 = 3.47011 loss) I0407 22:41:34.749590 359 solver.cpp:218] Iteration 1944 (0.851276 iter/s, 14.0965s/12 iters), loss = 3.1482 I0407 22:41:34.749629 359 solver.cpp:237] Train net output #0: loss = 3.1482 (* 1 = 3.1482 loss) I0407 22:41:34.749636 359 sgd_solver.cpp:105] Iteration 1944, lr = 0.00997951 I0407 22:41:39.694242 359 solver.cpp:218] Iteration 1956 (2.42689 iter/s, 4.9446s/12 iters), loss = 2.96406 I0407 22:41:39.694414 359 solver.cpp:237] Train net output #0: loss = 2.96406 (* 1 = 2.96406 loss) I0407 22:41:39.694424 359 sgd_solver.cpp:105] Iteration 1956, lr = 0.00997902 I0407 22:41:44.628644 359 solver.cpp:218] Iteration 1968 (2.432 iter/s, 4.93421s/12 iters), loss = 2.81329 I0407 22:41:44.628688 359 solver.cpp:237] Train net output #0: loss = 2.81329 (* 1 = 2.81329 loss) I0407 22:41:44.628696 359 sgd_solver.cpp:105] Iteration 1968, lr = 0.00997852 I0407 22:41:49.602478 359 solver.cpp:218] Iteration 1980 (2.41266 iter/s, 4.97377s/12 iters), loss = 2.97557 I0407 22:41:49.602519 359 solver.cpp:237] Train net output #0: loss = 2.97557 (* 1 = 2.97557 loss) I0407 22:41:49.602526 359 sgd_solver.cpp:105] Iteration 1980, lr = 0.00997801 I0407 22:41:54.548319 359 solver.cpp:218] Iteration 1992 (2.42631 iter/s, 4.94578s/12 iters), loss = 2.91323 I0407 22:41:54.548359 359 solver.cpp:237] Train net output #0: loss = 2.91323 (* 1 = 2.91323 loss) I0407 22:41:54.548368 359 sgd_solver.cpp:105] Iteration 1992, lr = 0.00997749 I0407 22:41:59.498108 359 solver.cpp:218] Iteration 2004 (2.42438 iter/s, 4.94973s/12 iters), loss = 2.86267 I0407 22:41:59.498152 359 solver.cpp:237] Train net output #0: loss = 2.86267 (* 1 = 2.86267 loss) I0407 22:41:59.498162 359 sgd_solver.cpp:105] Iteration 2004, lr = 0.00997696 I0407 22:42:04.465466 359 solver.cpp:218] Iteration 2016 (2.41581 iter/s, 4.96729s/12 iters), loss = 2.82093 I0407 22:42:04.465524 359 solver.cpp:237] Train net output #0: loss = 2.82093 (* 1 = 2.82093 loss) I0407 22:42:04.465538 359 sgd_solver.cpp:105] Iteration 2016, lr = 0.00997641 I0407 22:42:06.967725 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:42:09.377447 359 solver.cpp:218] Iteration 2028 (2.44304 iter/s, 4.91191s/12 iters), loss = 3.11125 I0407 22:42:09.377490 359 solver.cpp:237] Train net output #0: loss = 3.11125 (* 1 = 3.11125 loss) I0407 22:42:09.377499 359 sgd_solver.cpp:105] Iteration 2028, lr = 0.00997585 I0407 22:42:13.875124 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_2040.caffemodel I0407 22:42:17.015957 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_2040.solverstate I0407 22:42:19.374635 359 solver.cpp:330] Iteration 2040, Testing net (#0) I0407 22:42:19.374653 359 net.cpp:676] Ignoring source layer train-data I0407 22:42:23.261555 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:42:24.114142 359 solver.cpp:397] Test net output #0: accuracy = 0.227328 I0407 22:42:24.114192 359 solver.cpp:397] Test net output #1: loss = 3.24905 (* 1 = 3.24905 loss) I0407 22:42:24.210613 359 solver.cpp:218] Iteration 2040 (0.809002 iter/s, 14.8331s/12 iters), loss = 2.63953 I0407 22:42:24.210654 359 solver.cpp:237] Train net output #0: loss = 2.63953 (* 1 = 2.63953 loss) I0407 22:42:24.210661 359 sgd_solver.cpp:105] Iteration 2040, lr = 0.00997527 I0407 22:42:28.373304 359 solver.cpp:218] Iteration 2052 (2.88279 iter/s, 4.16263s/12 iters), loss = 2.96616 I0407 22:42:28.373347 359 solver.cpp:237] Train net output #0: loss = 2.96616 (* 1 = 2.96616 loss) I0407 22:42:28.373355 359 sgd_solver.cpp:105] Iteration 2052, lr = 0.00997469 I0407 22:42:29.953039 359 blocking_queue.cpp:49] Waiting for data I0407 22:42:33.262485 359 solver.cpp:218] Iteration 2064 (2.45443 iter/s, 4.88912s/12 iters), loss = 2.97788 I0407 22:42:33.262521 359 solver.cpp:237] Train net output #0: loss = 2.97788 (* 1 = 2.97788 loss) I0407 22:42:33.262529 359 sgd_solver.cpp:105] Iteration 2064, lr = 0.00997408 I0407 22:42:38.240365 359 solver.cpp:218] Iteration 2076 (2.41069 iter/s, 4.97782s/12 iters), loss = 3.04856 I0407 22:42:38.240401 359 solver.cpp:237] Train net output #0: loss = 3.04856 (* 1 = 3.04856 loss) I0407 22:42:38.240408 359 sgd_solver.cpp:105] Iteration 2076, lr = 0.00997347 I0407 22:42:43.194154 359 solver.cpp:218] Iteration 2088 (2.42241 iter/s, 4.95374s/12 iters), loss = 2.89056 I0407 22:42:43.194188 359 solver.cpp:237] Train net output #0: loss = 2.89056 (* 1 = 2.89056 loss) I0407 22:42:43.194196 359 sgd_solver.cpp:105] Iteration 2088, lr = 0.00997284 I0407 22:42:48.136507 359 solver.cpp:218] Iteration 2100 (2.42802 iter/s, 4.9423s/12 iters), loss = 3.05797 I0407 22:42:48.136633 359 solver.cpp:237] Train net output #0: loss = 3.05797 (* 1 = 3.05797 loss) I0407 22:42:48.136642 359 sgd_solver.cpp:105] Iteration 2100, lr = 0.0099722 I0407 22:42:53.081180 359 solver.cpp:218] Iteration 2112 (2.42692 iter/s, 4.94453s/12 iters), loss = 2.60516 I0407 22:42:53.081218 359 solver.cpp:237] Train net output #0: loss = 2.60516 (* 1 = 2.60516 loss) I0407 22:42:53.081224 359 sgd_solver.cpp:105] Iteration 2112, lr = 0.00997153 I0407 22:42:57.717440 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:42:58.030218 359 solver.cpp:218] Iteration 2124 (2.42474 iter/s, 4.94898s/12 iters), loss = 2.90187 I0407 22:42:58.030259 359 solver.cpp:237] Train net output #0: loss = 2.90187 (* 1 = 2.90187 loss) I0407 22:42:58.030268 359 sgd_solver.cpp:105] Iteration 2124, lr = 0.00997086 I0407 22:43:02.951489 359 solver.cpp:218] Iteration 2136 (2.43843 iter/s, 4.92121s/12 iters), loss = 3.1684 I0407 22:43:02.951524 359 solver.cpp:237] Train net output #0: loss = 3.1684 (* 1 = 3.1684 loss) I0407 22:43:02.951532 359 sgd_solver.cpp:105] Iteration 2136, lr = 0.00997017 I0407 22:43:04.986749 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_2142.caffemodel I0407 22:43:08.056052 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_2142.solverstate I0407 22:43:10.412533 359 solver.cpp:330] Iteration 2142, Testing net (#0) I0407 22:43:10.412554 359 net.cpp:676] Ignoring source layer train-data I0407 22:43:14.105819 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:43:14.981710 359 solver.cpp:397] Test net output #0: accuracy = 0.262255 I0407 22:43:14.981756 359 solver.cpp:397] Test net output #1: loss = 3.15991 (* 1 = 3.15991 loss) I0407 22:43:16.770299 359 solver.cpp:218] Iteration 2148 (0.868386 iter/s, 13.8187s/12 iters), loss = 2.6207 I0407 22:43:16.770344 359 solver.cpp:237] Train net output #0: loss = 2.6207 (* 1 = 2.6207 loss) I0407 22:43:16.770351 359 sgd_solver.cpp:105] Iteration 2148, lr = 0.00996946 I0407 22:43:21.637567 359 solver.cpp:218] Iteration 2160 (2.46548 iter/s, 4.86721s/12 iters), loss = 3.12555 I0407 22:43:21.637723 359 solver.cpp:237] Train net output #0: loss = 3.12555 (* 1 = 3.12555 loss) I0407 22:43:21.637732 359 sgd_solver.cpp:105] Iteration 2160, lr = 0.00996873 I0407 22:43:26.586370 359 solver.cpp:218] Iteration 2172 (2.42491 iter/s, 4.94863s/12 iters), loss = 2.9212 I0407 22:43:26.586406 359 solver.cpp:237] Train net output #0: loss = 2.9212 (* 1 = 2.9212 loss) I0407 22:43:26.586413 359 sgd_solver.cpp:105] Iteration 2172, lr = 0.00996799 I0407 22:43:31.510375 359 solver.cpp:218] Iteration 2184 (2.43707 iter/s, 4.92395s/12 iters), loss = 2.61246 I0407 22:43:31.510416 359 solver.cpp:237] Train net output #0: loss = 2.61246 (* 1 = 2.61246 loss) I0407 22:43:31.510423 359 sgd_solver.cpp:105] Iteration 2184, lr = 0.00996723 I0407 22:43:36.463024 359 solver.cpp:218] Iteration 2196 (2.42298 iter/s, 4.95259s/12 iters), loss = 2.73664 I0407 22:43:36.463070 359 solver.cpp:237] Train net output #0: loss = 2.73664 (* 1 = 2.73664 loss) I0407 22:43:36.463078 359 sgd_solver.cpp:105] Iteration 2196, lr = 0.00996646 I0407 22:43:41.413550 359 solver.cpp:218] Iteration 2208 (2.42402 iter/s, 4.95046s/12 iters), loss = 2.63206 I0407 22:43:41.413590 359 solver.cpp:237] Train net output #0: loss = 2.63206 (* 1 = 2.63206 loss) I0407 22:43:41.413599 359 sgd_solver.cpp:105] Iteration 2208, lr = 0.00996566 I0407 22:43:46.349711 359 solver.cpp:218] Iteration 2220 (2.43107 iter/s, 4.9361s/12 iters), loss = 2.47727 I0407 22:43:46.349750 359 solver.cpp:237] Train net output #0: loss = 2.47727 (* 1 = 2.47727 loss) I0407 22:43:46.349757 359 sgd_solver.cpp:105] Iteration 2220, lr = 0.00996485 I0407 22:43:48.122831 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:43:51.218158 359 solver.cpp:218] Iteration 2232 (2.46488 iter/s, 4.86839s/12 iters), loss = 2.48663 I0407 22:43:51.218200 359 solver.cpp:237] Train net output #0: loss = 2.48663 (* 1 = 2.48663 loss) I0407 22:43:51.218209 359 sgd_solver.cpp:105] Iteration 2232, lr = 0.00996401 I0407 22:43:55.724633 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_2244.caffemodel I0407 22:43:58.784642 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_2244.solverstate I0407 22:44:02.889842 359 solver.cpp:330] Iteration 2244, Testing net (#0) I0407 22:44:02.889859 359 net.cpp:676] Ignoring source layer train-data I0407 22:44:06.451486 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:44:07.376595 359 solver.cpp:397] Test net output #0: accuracy = 0.273897 I0407 22:44:07.376646 359 solver.cpp:397] Test net output #1: loss = 3.17993 (* 1 = 3.17993 loss) I0407 22:44:07.473145 359 solver.cpp:218] Iteration 2244 (0.738239 iter/s, 16.2549s/12 iters), loss = 2.46823 I0407 22:44:07.473194 359 solver.cpp:237] Train net output #0: loss = 2.46823 (* 1 = 2.46823 loss) I0407 22:44:07.473203 359 sgd_solver.cpp:105] Iteration 2244, lr = 0.00996316 I0407 22:44:11.575385 359 solver.cpp:218] Iteration 2256 (2.92528 iter/s, 4.10217s/12 iters), loss = 2.4896 I0407 22:44:11.575423 359 solver.cpp:237] Train net output #0: loss = 2.4896 (* 1 = 2.4896 loss) I0407 22:44:11.575433 359 sgd_solver.cpp:105] Iteration 2256, lr = 0.00996228 I0407 22:44:16.564204 359 solver.cpp:218] Iteration 2268 (2.40541 iter/s, 4.98876s/12 iters), loss = 2.43589 I0407 22:44:16.564240 359 solver.cpp:237] Train net output #0: loss = 2.43589 (* 1 = 2.43589 loss) I0407 22:44:16.564249 359 sgd_solver.cpp:105] Iteration 2268, lr = 0.00996139 I0407 22:44:21.590222 359 solver.cpp:218] Iteration 2280 (2.38761 iter/s, 5.02595s/12 iters), loss = 2.43739 I0407 22:44:21.590266 359 solver.cpp:237] Train net output #0: loss = 2.43739 (* 1 = 2.43739 loss) I0407 22:44:21.590276 359 sgd_solver.cpp:105] Iteration 2280, lr = 0.00996047 I0407 22:44:26.538518 359 solver.cpp:218] Iteration 2292 (2.42511 iter/s, 4.94823s/12 iters), loss = 2.44551 I0407 22:44:26.538664 359 solver.cpp:237] Train net output #0: loss = 2.44551 (* 1 = 2.44551 loss) I0407 22:44:26.538674 359 sgd_solver.cpp:105] Iteration 2292, lr = 0.00995954 I0407 22:44:31.483511 359 solver.cpp:218] Iteration 2304 (2.42678 iter/s, 4.94483s/12 iters), loss = 2.43083 I0407 22:44:31.483551 359 solver.cpp:237] Train net output #0: loss = 2.43083 (* 1 = 2.43083 loss) I0407 22:44:31.483559 359 sgd_solver.cpp:105] Iteration 2304, lr = 0.00995858 I0407 22:44:36.507511 359 solver.cpp:218] Iteration 2316 (2.38857 iter/s, 5.02394s/12 iters), loss = 2.68947 I0407 22:44:36.507551 359 solver.cpp:237] Train net output #0: loss = 2.68947 (* 1 = 2.68947 loss) I0407 22:44:36.507560 359 sgd_solver.cpp:105] Iteration 2316, lr = 0.00995759 I0407 22:44:40.379441 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:44:41.403112 359 solver.cpp:218] Iteration 2328 (2.45121 iter/s, 4.89554s/12 iters), loss = 2.6636 I0407 22:44:41.403156 359 solver.cpp:237] Train net output #0: loss = 2.6636 (* 1 = 2.6636 loss) I0407 22:44:41.403163 359 sgd_solver.cpp:105] Iteration 2328, lr = 0.00995659 I0407 22:44:46.347402 359 solver.cpp:218] Iteration 2340 (2.42708 iter/s, 4.94422s/12 iters), loss = 2.40418 I0407 22:44:46.347448 359 solver.cpp:237] Train net output #0: loss = 2.40418 (* 1 = 2.40418 loss) I0407 22:44:46.347457 359 sgd_solver.cpp:105] Iteration 2340, lr = 0.00995556 I0407 22:44:48.363399 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_2346.caffemodel I0407 22:44:51.524742 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_2346.solverstate I0407 22:44:53.978636 359 solver.cpp:330] Iteration 2346, Testing net (#0) I0407 22:44:53.978653 359 net.cpp:676] Ignoring source layer train-data I0407 22:44:57.708062 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:44:58.705417 359 solver.cpp:397] Test net output #0: accuracy = 0.252451 I0407 22:44:58.705457 359 solver.cpp:397] Test net output #1: loss = 3.26199 (* 1 = 3.26199 loss) I0407 22:45:00.496237 359 solver.cpp:218] Iteration 2352 (0.848131 iter/s, 14.1488s/12 iters), loss = 2.34774 I0407 22:45:00.496279 359 solver.cpp:237] Train net output #0: loss = 2.34774 (* 1 = 2.34774 loss) I0407 22:45:00.496289 359 sgd_solver.cpp:105] Iteration 2352, lr = 0.00995451 I0407 22:45:05.401335 359 solver.cpp:218] Iteration 2364 (2.44647 iter/s, 4.90503s/12 iters), loss = 2.51786 I0407 22:45:05.401373 359 solver.cpp:237] Train net output #0: loss = 2.51786 (* 1 = 2.51786 loss) I0407 22:45:05.401381 359 sgd_solver.cpp:105] Iteration 2364, lr = 0.00995343 I0407 22:45:10.361184 359 solver.cpp:218] Iteration 2376 (2.41946 iter/s, 4.95979s/12 iters), loss = 2.59695 I0407 22:45:10.361224 359 solver.cpp:237] Train net output #0: loss = 2.59695 (* 1 = 2.59695 loss) I0407 22:45:10.361232 359 sgd_solver.cpp:105] Iteration 2376, lr = 0.00995233 I0407 22:45:15.275897 359 solver.cpp:218] Iteration 2388 (2.44168 iter/s, 4.91465s/12 iters), loss = 2.55361 I0407 22:45:15.275935 359 solver.cpp:237] Train net output #0: loss = 2.55361 (* 1 = 2.55361 loss) I0407 22:45:15.275943 359 sgd_solver.cpp:105] Iteration 2388, lr = 0.0099512 I0407 22:45:20.219977 359 solver.cpp:218] Iteration 2400 (2.42717 iter/s, 4.94402s/12 iters), loss = 2.29065 I0407 22:45:20.220019 359 solver.cpp:237] Train net output #0: loss = 2.29065 (* 1 = 2.29065 loss) I0407 22:45:20.220027 359 sgd_solver.cpp:105] Iteration 2400, lr = 0.00995004 I0407 22:45:25.143514 359 solver.cpp:218] Iteration 2412 (2.43731 iter/s, 4.92347s/12 iters), loss = 2.76306 I0407 22:45:25.143559 359 solver.cpp:237] Train net output #0: loss = 2.76306 (* 1 = 2.76306 loss) I0407 22:45:25.143568 359 sgd_solver.cpp:105] Iteration 2412, lr = 0.00994886 I0407 22:45:30.110707 359 solver.cpp:218] Iteration 2424 (2.41588 iter/s, 4.96713s/12 iters), loss = 2.06413 I0407 22:45:30.110862 359 solver.cpp:237] Train net output #0: loss = 2.06413 (* 1 = 2.06413 loss) I0407 22:45:30.110872 359 sgd_solver.cpp:105] Iteration 2424, lr = 0.00994765 I0407 22:45:31.152237 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:45:35.006736 359 solver.cpp:218] Iteration 2436 (2.45105 iter/s, 4.89586s/12 iters), loss = 2.38329 I0407 22:45:35.006778 359 solver.cpp:237] Train net output #0: loss = 2.38329 (* 1 = 2.38329 loss) I0407 22:45:35.006785 359 sgd_solver.cpp:105] Iteration 2436, lr = 0.00994641 I0407 22:45:39.507761 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_2448.caffemodel I0407 22:45:43.694573 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_2448.solverstate I0407 22:45:46.619050 359 solver.cpp:330] Iteration 2448, Testing net (#0) I0407 22:45:46.619069 359 net.cpp:676] Ignoring source layer train-data I0407 22:45:50.383879 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:45:51.464020 359 solver.cpp:397] Test net output #0: accuracy = 0.239583 I0407 22:45:51.464053 359 solver.cpp:397] Test net output #1: loss = 3.32404 (* 1 = 3.32404 loss) I0407 22:45:51.560308 359 solver.cpp:218] Iteration 2448 (0.724923 iter/s, 16.5535s/12 iters), loss = 2.68792 I0407 22:45:51.560364 359 solver.cpp:237] Train net output #0: loss = 2.68792 (* 1 = 2.68792 loss) I0407 22:45:51.560374 359 sgd_solver.cpp:105] Iteration 2448, lr = 0.00994514 I0407 22:45:55.595785 359 solver.cpp:218] Iteration 2460 (2.97368 iter/s, 4.0354s/12 iters), loss = 2.49981 I0407 22:45:55.595826 359 solver.cpp:237] Train net output #0: loss = 2.49981 (* 1 = 2.49981 loss) I0407 22:45:55.595835 359 sgd_solver.cpp:105] Iteration 2460, lr = 0.00994384 I0407 22:46:00.445024 359 solver.cpp:218] Iteration 2472 (2.47464 iter/s, 4.84918s/12 iters), loss = 2.17352 I0407 22:46:00.445158 359 solver.cpp:237] Train net output #0: loss = 2.17352 (* 1 = 2.17352 loss) I0407 22:46:00.445165 359 sgd_solver.cpp:105] Iteration 2472, lr = 0.00994251 I0407 22:46:05.404966 359 solver.cpp:218] Iteration 2484 (2.41946 iter/s, 4.95979s/12 iters), loss = 2.29499 I0407 22:46:05.405000 359 solver.cpp:237] Train net output #0: loss = 2.29499 (* 1 = 2.29499 loss) I0407 22:46:05.405007 359 sgd_solver.cpp:105] Iteration 2484, lr = 0.00994115 I0407 22:46:10.333974 359 solver.cpp:218] Iteration 2496 (2.43459 iter/s, 4.92896s/12 iters), loss = 2.5319 I0407 22:46:10.334015 359 solver.cpp:237] Train net output #0: loss = 2.5319 (* 1 = 2.5319 loss) I0407 22:46:10.334023 359 sgd_solver.cpp:105] Iteration 2496, lr = 0.00993976 I0407 22:46:15.308440 359 solver.cpp:218] Iteration 2508 (2.41235 iter/s, 4.97441s/12 iters), loss = 2.37591 I0407 22:46:15.308475 359 solver.cpp:237] Train net output #0: loss = 2.37591 (* 1 = 2.37591 loss) I0407 22:46:15.308482 359 sgd_solver.cpp:105] Iteration 2508, lr = 0.00993833 I0407 22:46:20.222555 359 solver.cpp:218] Iteration 2520 (2.44197 iter/s, 4.91406s/12 iters), loss = 2.29151 I0407 22:46:20.222589 359 solver.cpp:237] Train net output #0: loss = 2.29151 (* 1 = 2.29151 loss) I0407 22:46:20.222596 359 sgd_solver.cpp:105] Iteration 2520, lr = 0.00993687 I0407 22:46:23.421361 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:46:25.167135 359 solver.cpp:218] Iteration 2532 (2.42693 iter/s, 4.94453s/12 iters), loss = 2.35582 I0407 22:46:25.167168 359 solver.cpp:237] Train net output #0: loss = 2.35582 (* 1 = 2.35582 loss) I0407 22:46:25.167176 359 sgd_solver.cpp:105] Iteration 2532, lr = 0.00993538 I0407 22:46:30.112251 359 solver.cpp:218] Iteration 2544 (2.42666 iter/s, 4.94506s/12 iters), loss = 2.42731 I0407 22:46:30.112291 359 solver.cpp:237] Train net output #0: loss = 2.42731 (* 1 = 2.42731 loss) I0407 22:46:30.112300 359 sgd_solver.cpp:105] Iteration 2544, lr = 0.00993385 I0407 22:46:32.149363 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_2550.caffemodel I0407 22:46:35.288813 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_2550.solverstate I0407 22:46:38.053391 359 solver.cpp:330] Iteration 2550, Testing net (#0) I0407 22:46:38.053411 359 net.cpp:676] Ignoring source layer train-data I0407 22:46:41.696125 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:46:42.827939 359 solver.cpp:397] Test net output #0: accuracy = 0.284314 I0407 22:46:42.827984 359 solver.cpp:397] Test net output #1: loss = 3.15071 (* 1 = 3.15071 loss) I0407 22:46:44.611464 359 solver.cpp:218] Iteration 2556 (0.827636 iter/s, 14.4991s/12 iters), loss = 2.2578 I0407 22:46:44.611512 359 solver.cpp:237] Train net output #0: loss = 2.2578 (* 1 = 2.2578 loss) I0407 22:46:44.611521 359 sgd_solver.cpp:105] Iteration 2556, lr = 0.00993228 I0407 22:46:49.662487 359 solver.cpp:218] Iteration 2568 (2.3758 iter/s, 5.05094s/12 iters), loss = 2.19678 I0407 22:46:49.662552 359 solver.cpp:237] Train net output #0: loss = 2.19678 (* 1 = 2.19678 loss) I0407 22:46:49.662567 359 sgd_solver.cpp:105] Iteration 2568, lr = 0.00993068 I0407 22:46:54.587882 359 solver.cpp:218] Iteration 2580 (2.4364 iter/s, 4.92531s/12 iters), loss = 1.96284 I0407 22:46:54.587924 359 solver.cpp:237] Train net output #0: loss = 1.96284 (* 1 = 1.96284 loss) I0407 22:46:54.587931 359 sgd_solver.cpp:105] Iteration 2580, lr = 0.00992905 I0407 22:46:59.540197 359 solver.cpp:218] Iteration 2592 (2.42314 iter/s, 4.95226s/12 iters), loss = 2.35947 I0407 22:46:59.540233 359 solver.cpp:237] Train net output #0: loss = 2.35947 (* 1 = 2.35947 loss) I0407 22:46:59.540241 359 sgd_solver.cpp:105] Iteration 2592, lr = 0.00992737 I0407 22:47:04.440917 359 solver.cpp:218] Iteration 2604 (2.44865 iter/s, 4.90066s/12 iters), loss = 2.09648 I0407 22:47:04.441054 359 solver.cpp:237] Train net output #0: loss = 2.09648 (* 1 = 2.09648 loss) I0407 22:47:04.441063 359 sgd_solver.cpp:105] Iteration 2604, lr = 0.00992565 I0407 22:47:09.410863 359 solver.cpp:218] Iteration 2616 (2.41459 iter/s, 4.96979s/12 iters), loss = 2.16389 I0407 22:47:09.410907 359 solver.cpp:237] Train net output #0: loss = 2.16389 (* 1 = 2.16389 loss) I0407 22:47:09.410914 359 sgd_solver.cpp:105] Iteration 2616, lr = 0.00992389 I0407 22:47:14.319983 359 solver.cpp:218] Iteration 2628 (2.44446 iter/s, 4.90905s/12 iters), loss = 1.97931 I0407 22:47:14.320024 359 solver.cpp:237] Train net output #0: loss = 1.97931 (* 1 = 1.97931 loss) I0407 22:47:14.320032 359 sgd_solver.cpp:105] Iteration 2628, lr = 0.0099221 I0407 22:47:14.738675 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:47:19.256965 359 solver.cpp:218] Iteration 2640 (2.43067 iter/s, 4.93692s/12 iters), loss = 2.25741 I0407 22:47:19.257009 359 solver.cpp:237] Train net output #0: loss = 2.25741 (* 1 = 2.25741 loss) I0407 22:47:19.257016 359 sgd_solver.cpp:105] Iteration 2640, lr = 0.00992026 I0407 22:47:23.661909 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_2652.caffemodel I0407 22:47:26.747251 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_2652.solverstate I0407 22:47:29.112129 359 solver.cpp:330] Iteration 2652, Testing net (#0) I0407 22:47:29.112147 359 net.cpp:676] Ignoring source layer train-data I0407 22:47:32.740947 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:47:33.908448 359 solver.cpp:397] Test net output #0: accuracy = 0.272059 I0407 22:47:33.908484 359 solver.cpp:397] Test net output #1: loss = 3.15613 (* 1 = 3.15613 loss) I0407 22:47:34.004735 359 solver.cpp:218] Iteration 2652 (0.813687 iter/s, 14.7477s/12 iters), loss = 2.32786 I0407 22:47:34.004778 359 solver.cpp:237] Train net output #0: loss = 2.32786 (* 1 = 2.32786 loss) I0407 22:47:34.004787 359 sgd_solver.cpp:105] Iteration 2652, lr = 0.00991837 I0407 22:47:38.125959 359 solver.cpp:218] Iteration 2664 (2.91181 iter/s, 4.12115s/12 iters), loss = 2.42141 I0407 22:47:38.126102 359 solver.cpp:237] Train net output #0: loss = 2.42141 (* 1 = 2.42141 loss) I0407 22:47:38.126111 359 sgd_solver.cpp:105] Iteration 2664, lr = 0.00991645 I0407 22:47:42.995462 359 solver.cpp:218] Iteration 2676 (2.4644 iter/s, 4.86934s/12 iters), loss = 1.88912 I0407 22:47:42.995507 359 solver.cpp:237] Train net output #0: loss = 1.88912 (* 1 = 1.88912 loss) I0407 22:47:42.995515 359 sgd_solver.cpp:105] Iteration 2676, lr = 0.00991447 I0407 22:47:47.908500 359 solver.cpp:218] Iteration 2688 (2.44251 iter/s, 4.91298s/12 iters), loss = 2.26359 I0407 22:47:47.908532 359 solver.cpp:237] Train net output #0: loss = 2.26359 (* 1 = 2.26359 loss) I0407 22:47:47.908540 359 sgd_solver.cpp:105] Iteration 2688, lr = 0.00991246 I0407 22:47:52.857479 359 solver.cpp:218] Iteration 2700 (2.42477 iter/s, 4.94892s/12 iters), loss = 2.33243 I0407 22:47:52.857511 359 solver.cpp:237] Train net output #0: loss = 2.33243 (* 1 = 2.33243 loss) I0407 22:47:52.857518 359 sgd_solver.cpp:105] Iteration 2700, lr = 0.00991039 I0407 22:47:57.767093 359 solver.cpp:218] Iteration 2712 (2.44421 iter/s, 4.90956s/12 iters), loss = 2.30132 I0407 22:47:57.767129 359 solver.cpp:237] Train net output #0: loss = 2.30132 (* 1 = 2.30132 loss) I0407 22:47:57.767138 359 sgd_solver.cpp:105] Iteration 2712, lr = 0.00990828 I0407 22:48:02.710635 359 solver.cpp:218] Iteration 2724 (2.42744 iter/s, 4.94349s/12 iters), loss = 2.0974 I0407 22:48:02.710675 359 solver.cpp:237] Train net output #0: loss = 2.0974 (* 1 = 2.0974 loss) I0407 22:48:02.710681 359 sgd_solver.cpp:105] Iteration 2724, lr = 0.00990611 I0407 22:48:05.232297 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:48:07.613812 359 solver.cpp:218] Iteration 2736 (2.44742 iter/s, 4.90311s/12 iters), loss = 2.26774 I0407 22:48:07.613849 359 solver.cpp:237] Train net output #0: loss = 2.26774 (* 1 = 2.26774 loss) I0407 22:48:07.613857 359 sgd_solver.cpp:105] Iteration 2736, lr = 0.0099039 I0407 22:48:12.586241 359 solver.cpp:218] Iteration 2748 (2.41334 iter/s, 4.97237s/12 iters), loss = 2.08512 I0407 22:48:12.586366 359 solver.cpp:237] Train net output #0: loss = 2.08512 (* 1 = 2.08512 loss) I0407 22:48:12.586375 359 sgd_solver.cpp:105] Iteration 2748, lr = 0.00990163 I0407 22:48:14.588271 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_2754.caffemodel I0407 22:48:17.679343 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_2754.solverstate I0407 22:48:20.043385 359 solver.cpp:330] Iteration 2754, Testing net (#0) I0407 22:48:20.043403 359 net.cpp:676] Ignoring source layer train-data I0407 22:48:23.333578 359 blocking_queue.cpp:49] Waiting for data I0407 22:48:23.601482 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:48:24.816747 359 solver.cpp:397] Test net output #0: accuracy = 0.283088 I0407 22:48:24.816795 359 solver.cpp:397] Test net output #1: loss = 3.17317 (* 1 = 3.17317 loss) I0407 22:48:26.637207 359 solver.cpp:218] Iteration 2760 (0.854044 iter/s, 14.0508s/12 iters), loss = 2.25691 I0407 22:48:26.637259 359 solver.cpp:237] Train net output #0: loss = 2.25691 (* 1 = 2.25691 loss) I0407 22:48:26.637269 359 sgd_solver.cpp:105] Iteration 2760, lr = 0.00989932 I0407 22:48:31.531042 359 solver.cpp:218] Iteration 2772 (2.4521 iter/s, 4.89376s/12 iters), loss = 2.27579 I0407 22:48:31.531082 359 solver.cpp:237] Train net output #0: loss = 2.27579 (* 1 = 2.27579 loss) I0407 22:48:31.531090 359 sgd_solver.cpp:105] Iteration 2772, lr = 0.00989694 I0407 22:48:36.490770 359 solver.cpp:218] Iteration 2784 (2.41952 iter/s, 4.95966s/12 iters), loss = 1.85031 I0407 22:48:36.490814 359 solver.cpp:237] Train net output #0: loss = 1.85031 (* 1 = 1.85031 loss) I0407 22:48:36.490823 359 sgd_solver.cpp:105] Iteration 2784, lr = 0.00989452 I0407 22:48:41.406931 359 solver.cpp:218] Iteration 2796 (2.44096 iter/s, 4.9161s/12 iters), loss = 2.35421 I0407 22:48:41.406965 359 solver.cpp:237] Train net output #0: loss = 2.35421 (* 1 = 2.35421 loss) I0407 22:48:41.406973 359 sgd_solver.cpp:105] Iteration 2796, lr = 0.00989203 I0407 22:48:46.345438 359 solver.cpp:218] Iteration 2808 (2.42991 iter/s, 4.93846s/12 iters), loss = 2.1644 I0407 22:48:46.345599 359 solver.cpp:237] Train net output #0: loss = 2.1644 (* 1 = 2.1644 loss) I0407 22:48:46.345609 359 sgd_solver.cpp:105] Iteration 2808, lr = 0.00988949 I0407 22:48:51.282537 359 solver.cpp:218] Iteration 2820 (2.43066 iter/s, 4.93693s/12 iters), loss = 1.87829 I0407 22:48:51.282572 359 solver.cpp:237] Train net output #0: loss = 1.87829 (* 1 = 1.87829 loss) I0407 22:48:51.282579 359 sgd_solver.cpp:105] Iteration 2820, lr = 0.00988689 I0407 22:48:55.939746 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:48:56.224982 359 solver.cpp:218] Iteration 2832 (2.42797 iter/s, 4.94239s/12 iters), loss = 2.01649 I0407 22:48:56.225016 359 solver.cpp:237] Train net output #0: loss = 2.01649 (* 1 = 2.01649 loss) I0407 22:48:56.225023 359 sgd_solver.cpp:105] Iteration 2832, lr = 0.00988423 I0407 22:49:01.161550 359 solver.cpp:218] Iteration 2844 (2.43087 iter/s, 4.93651s/12 iters), loss = 1.9606 I0407 22:49:01.161585 359 solver.cpp:237] Train net output #0: loss = 1.9606 (* 1 = 1.9606 loss) I0407 22:49:01.161593 359 sgd_solver.cpp:105] Iteration 2844, lr = 0.0098815 I0407 22:49:05.659430 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_2856.caffemodel I0407 22:49:08.729086 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_2856.solverstate I0407 22:49:11.087358 359 solver.cpp:330] Iteration 2856, Testing net (#0) I0407 22:49:11.087375 359 net.cpp:676] Ignoring source layer train-data I0407 22:49:14.623360 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:49:15.875644 359 solver.cpp:397] Test net output #0: accuracy = 0.304534 I0407 22:49:15.875692 359 solver.cpp:397] Test net output #1: loss = 3.03383 (* 1 = 3.03383 loss) I0407 22:49:15.972419 359 solver.cpp:218] Iteration 2856 (0.81022 iter/s, 14.8108s/12 iters), loss = 1.85983 I0407 22:49:15.972462 359 solver.cpp:237] Train net output #0: loss = 1.85983 (* 1 = 1.85983 loss) I0407 22:49:15.972471 359 sgd_solver.cpp:105] Iteration 2856, lr = 0.00987872 I0407 22:49:20.129029 359 solver.cpp:218] Iteration 2868 (2.88701 iter/s, 4.15655s/12 iters), loss = 2.07138 I0407 22:49:20.129158 359 solver.cpp:237] Train net output #0: loss = 2.07138 (* 1 = 2.07138 loss) I0407 22:49:20.129168 359 sgd_solver.cpp:105] Iteration 2868, lr = 0.00987586 I0407 22:49:25.049413 359 solver.cpp:218] Iteration 2880 (2.43891 iter/s, 4.92024s/12 iters), loss = 2.13079 I0407 22:49:25.049450 359 solver.cpp:237] Train net output #0: loss = 2.13079 (* 1 = 2.13079 loss) I0407 22:49:25.049458 359 sgd_solver.cpp:105] Iteration 2880, lr = 0.00987295 I0407 22:49:30.000919 359 solver.cpp:218] Iteration 2892 (2.42353 iter/s, 4.95145s/12 iters), loss = 2.06035 I0407 22:49:30.000954 359 solver.cpp:237] Train net output #0: loss = 2.06035 (* 1 = 2.06035 loss) I0407 22:49:30.000962 359 sgd_solver.cpp:105] Iteration 2892, lr = 0.00986996 I0407 22:49:34.916677 359 solver.cpp:218] Iteration 2904 (2.44116 iter/s, 4.9157s/12 iters), loss = 2.31742 I0407 22:49:34.916713 359 solver.cpp:237] Train net output #0: loss = 2.31742 (* 1 = 2.31742 loss) I0407 22:49:34.916721 359 sgd_solver.cpp:105] Iteration 2904, lr = 0.00986691 I0407 22:49:39.877827 359 solver.cpp:218] Iteration 2916 (2.41882 iter/s, 4.96109s/12 iters), loss = 1.92919 I0407 22:49:39.877864 359 solver.cpp:237] Train net output #0: loss = 1.92919 (* 1 = 1.92919 loss) I0407 22:49:39.877873 359 sgd_solver.cpp:105] Iteration 2916, lr = 0.00986378 I0407 22:49:44.795217 359 solver.cpp:218] Iteration 2928 (2.44035 iter/s, 4.91733s/12 iters), loss = 1.81033 I0407 22:49:44.795258 359 solver.cpp:237] Train net output #0: loss = 1.81033 (* 1 = 1.81033 loss) I0407 22:49:44.795266 359 sgd_solver.cpp:105] Iteration 2928, lr = 0.00986058 I0407 22:49:46.627804 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:49:49.728142 359 solver.cpp:218] Iteration 2940 (2.43267 iter/s, 4.93285s/12 iters), loss = 2.03943 I0407 22:49:49.728201 359 solver.cpp:237] Train net output #0: loss = 2.03943 (* 1 = 2.03943 loss) I0407 22:49:49.728214 359 sgd_solver.cpp:105] Iteration 2940, lr = 0.00985731 I0407 22:49:54.657171 359 solver.cpp:218] Iteration 2952 (2.43459 iter/s, 4.92895s/12 iters), loss = 2.02725 I0407 22:49:54.657321 359 solver.cpp:237] Train net output #0: loss = 2.02725 (* 1 = 2.02725 loss) I0407 22:49:54.657332 359 sgd_solver.cpp:105] Iteration 2952, lr = 0.00985396 I0407 22:49:56.686234 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_2958.caffemodel I0407 22:49:59.871939 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_2958.solverstate I0407 22:50:02.232760 359 solver.cpp:330] Iteration 2958, Testing net (#0) I0407 22:50:02.232780 359 net.cpp:676] Ignoring source layer train-data I0407 22:50:05.661442 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:50:06.950773 359 solver.cpp:397] Test net output #0: accuracy = 0.299632 I0407 22:50:06.950816 359 solver.cpp:397] Test net output #1: loss = 3.07793 (* 1 = 3.07793 loss) I0407 22:50:08.736313 359 solver.cpp:218] Iteration 2964 (0.852336 iter/s, 14.079s/12 iters), loss = 2.05118 I0407 22:50:08.736361 359 solver.cpp:237] Train net output #0: loss = 2.05118 (* 1 = 2.05118 loss) I0407 22:50:08.736371 359 sgd_solver.cpp:105] Iteration 2964, lr = 0.00985054 I0407 22:50:13.663187 359 solver.cpp:218] Iteration 2976 (2.43565 iter/s, 4.92681s/12 iters), loss = 1.90589 I0407 22:50:13.663225 359 solver.cpp:237] Train net output #0: loss = 1.90589 (* 1 = 1.90589 loss) I0407 22:50:13.663233 359 sgd_solver.cpp:105] Iteration 2976, lr = 0.00984703 I0407 22:50:18.610044 359 solver.cpp:218] Iteration 2988 (2.42581 iter/s, 4.9468s/12 iters), loss = 2.09538 I0407 22:50:18.610085 359 solver.cpp:237] Train net output #0: loss = 2.09538 (* 1 = 2.09538 loss) I0407 22:50:18.610093 359 sgd_solver.cpp:105] Iteration 2988, lr = 0.00984345 I0407 22:50:23.547987 359 solver.cpp:218] Iteration 3000 (2.43019 iter/s, 4.93788s/12 iters), loss = 1.83884 I0407 22:50:23.548025 359 solver.cpp:237] Train net output #0: loss = 1.83884 (* 1 = 1.83884 loss) I0407 22:50:23.548033 359 sgd_solver.cpp:105] Iteration 3000, lr = 0.00983978 I0407 22:50:28.594018 359 solver.cpp:218] Iteration 3012 (2.37813 iter/s, 5.04598s/12 iters), loss = 1.87208 I0407 22:50:28.594077 359 solver.cpp:237] Train net output #0: loss = 1.87208 (* 1 = 1.87208 loss) I0407 22:50:28.594085 359 sgd_solver.cpp:105] Iteration 3012, lr = 0.00983603 I0407 22:50:33.816968 359 solver.cpp:218] Iteration 3024 (2.29759 iter/s, 5.22287s/12 iters), loss = 1.93008 I0407 22:50:33.817006 359 solver.cpp:237] Train net output #0: loss = 1.93008 (* 1 = 1.93008 loss) I0407 22:50:33.817014 359 sgd_solver.cpp:105] Iteration 3024, lr = 0.00983219 I0407 22:50:37.810873 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:50:38.804452 359 solver.cpp:218] Iteration 3036 (2.40605 iter/s, 4.98743s/12 iters), loss = 1.70603 I0407 22:50:38.804488 359 solver.cpp:237] Train net output #0: loss = 1.70603 (* 1 = 1.70603 loss) I0407 22:50:38.804495 359 sgd_solver.cpp:105] Iteration 3036, lr = 0.00982826 I0407 22:50:43.883749 359 solver.cpp:218] Iteration 3048 (2.36256 iter/s, 5.07924s/12 iters), loss = 1.91323 I0407 22:50:43.883790 359 solver.cpp:237] Train net output #0: loss = 1.91323 (* 1 = 1.91323 loss) I0407 22:50:43.883797 359 sgd_solver.cpp:105] Iteration 3048, lr = 0.00982425 I0407 22:50:48.389130 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_3060.caffemodel I0407 22:50:52.510715 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_3060.solverstate I0407 22:50:54.960774 359 solver.cpp:330] Iteration 3060, Testing net (#0) I0407 22:50:54.960794 359 net.cpp:676] Ignoring source layer train-data I0407 22:50:58.439750 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:50:59.740667 359 solver.cpp:397] Test net output #0: accuracy = 0.313113 I0407 22:50:59.740809 359 solver.cpp:397] Test net output #1: loss = 2.98991 (* 1 = 2.98991 loss) I0407 22:50:59.837545 359 solver.cpp:218] Iteration 3060 (0.752176 iter/s, 15.9537s/12 iters), loss = 1.75386 I0407 22:50:59.837592 359 solver.cpp:237] Train net output #0: loss = 1.75386 (* 1 = 1.75386 loss) I0407 22:50:59.837601 359 sgd_solver.cpp:105] Iteration 3060, lr = 0.00982014 I0407 22:51:04.122287 359 solver.cpp:218] Iteration 3072 (2.80068 iter/s, 4.28468s/12 iters), loss = 1.54107 I0407 22:51:04.122321 359 solver.cpp:237] Train net output #0: loss = 1.54107 (* 1 = 1.54107 loss) I0407 22:51:04.122328 359 sgd_solver.cpp:105] Iteration 3072, lr = 0.00981593 I0407 22:51:09.134703 359 solver.cpp:218] Iteration 3084 (2.39408 iter/s, 5.01236s/12 iters), loss = 1.87213 I0407 22:51:09.134748 359 solver.cpp:237] Train net output #0: loss = 1.87213 (* 1 = 1.87213 loss) I0407 22:51:09.134757 359 sgd_solver.cpp:105] Iteration 3084, lr = 0.00981163 I0407 22:51:14.128625 359 solver.cpp:218] Iteration 3096 (2.40295 iter/s, 4.99385s/12 iters), loss = 1.77313 I0407 22:51:14.128671 359 solver.cpp:237] Train net output #0: loss = 1.77313 (* 1 = 1.77313 loss) I0407 22:51:14.128680 359 sgd_solver.cpp:105] Iteration 3096, lr = 0.00980724 I0407 22:51:18.986593 359 solver.cpp:218] Iteration 3108 (2.47021 iter/s, 4.8579s/12 iters), loss = 1.94607 I0407 22:51:18.986634 359 solver.cpp:237] Train net output #0: loss = 1.94607 (* 1 = 1.94607 loss) I0407 22:51:18.986642 359 sgd_solver.cpp:105] Iteration 3108, lr = 0.00980274 I0407 22:51:23.914330 359 solver.cpp:218] Iteration 3120 (2.43522 iter/s, 4.92768s/12 iters), loss = 2.19232 I0407 22:51:23.914369 359 solver.cpp:237] Train net output #0: loss = 2.19232 (* 1 = 2.19232 loss) I0407 22:51:23.914376 359 sgd_solver.cpp:105] Iteration 3120, lr = 0.00979814 I0407 22:51:28.876279 359 solver.cpp:218] Iteration 3132 (2.41843 iter/s, 4.96189s/12 iters), loss = 1.86277 I0407 22:51:28.876312 359 solver.cpp:237] Train net output #0: loss = 1.86277 (* 1 = 1.86277 loss) I0407 22:51:28.876319 359 sgd_solver.cpp:105] Iteration 3132, lr = 0.00979343 I0407 22:51:29.947422 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:51:33.763401 359 solver.cpp:218] Iteration 3144 (2.45546 iter/s, 4.88706s/12 iters), loss = 1.78341 I0407 22:51:33.763442 359 solver.cpp:237] Train net output #0: loss = 1.78341 (* 1 = 1.78341 loss) I0407 22:51:33.763450 359 sgd_solver.cpp:105] Iteration 3144, lr = 0.00978861 I0407 22:51:38.727494 359 solver.cpp:218] Iteration 3156 (2.41739 iter/s, 4.96403s/12 iters), loss = 1.60958 I0407 22:51:38.727533 359 solver.cpp:237] Train net output #0: loss = 1.60958 (* 1 = 1.60958 loss) I0407 22:51:38.727540 359 sgd_solver.cpp:105] Iteration 3156, lr = 0.00978369 I0407 22:51:40.710310 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_3162.caffemodel I0407 22:51:45.026276 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_3162.solverstate I0407 22:51:48.847363 359 solver.cpp:330] Iteration 3162, Testing net (#0) I0407 22:51:48.847381 359 net.cpp:676] Ignoring source layer train-data I0407 22:51:52.225072 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:51:53.618080 359 solver.cpp:397] Test net output #0: accuracy = 0.319853 I0407 22:51:53.618124 359 solver.cpp:397] Test net output #1: loss = 2.96404 (* 1 = 2.96404 loss) I0407 22:51:55.414750 359 solver.cpp:218] Iteration 3168 (0.719115 iter/s, 16.6872s/12 iters), loss = 1.72166 I0407 22:51:55.414785 359 solver.cpp:237] Train net output #0: loss = 1.72166 (* 1 = 1.72166 loss) I0407 22:51:55.414794 359 sgd_solver.cpp:105] Iteration 3168, lr = 0.00977866 I0407 22:52:00.319183 359 solver.cpp:218] Iteration 3180 (2.4468 iter/s, 4.90437s/12 iters), loss = 1.62192 I0407 22:52:00.319308 359 solver.cpp:237] Train net output #0: loss = 1.62192 (* 1 = 1.62192 loss) I0407 22:52:00.319317 359 sgd_solver.cpp:105] Iteration 3180, lr = 0.0097735 I0407 22:52:05.216598 359 solver.cpp:218] Iteration 3192 (2.45034 iter/s, 4.89727s/12 iters), loss = 1.81731 I0407 22:52:05.216640 359 solver.cpp:237] Train net output #0: loss = 1.81731 (* 1 = 1.81731 loss) I0407 22:52:05.216648 359 sgd_solver.cpp:105] Iteration 3192, lr = 0.00976824 I0407 22:52:10.143244 359 solver.cpp:218] Iteration 3204 (2.43577 iter/s, 4.92657s/12 iters), loss = 1.89343 I0407 22:52:10.143287 359 solver.cpp:237] Train net output #0: loss = 1.89343 (* 1 = 1.89343 loss) I0407 22:52:10.143296 359 sgd_solver.cpp:105] Iteration 3204, lr = 0.00976285 I0407 22:52:15.104357 359 solver.cpp:218] Iteration 3216 (2.41884 iter/s, 4.96105s/12 iters), loss = 1.50783 I0407 22:52:15.104393 359 solver.cpp:237] Train net output #0: loss = 1.50783 (* 1 = 1.50783 loss) I0407 22:52:15.104399 359 sgd_solver.cpp:105] Iteration 3216, lr = 0.00975734 I0407 22:52:20.017047 359 solver.cpp:218] Iteration 3228 (2.44268 iter/s, 4.91264s/12 iters), loss = 1.82193 I0407 22:52:20.017084 359 solver.cpp:237] Train net output #0: loss = 1.82193 (* 1 = 1.82193 loss) I0407 22:52:20.017092 359 sgd_solver.cpp:105] Iteration 3228, lr = 0.00975171 I0407 22:52:23.233575 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:52:24.955689 359 solver.cpp:218] Iteration 3240 (2.42985 iter/s, 4.93858s/12 iters), loss = 2.01245 I0407 22:52:24.955731 359 solver.cpp:237] Train net output #0: loss = 2.01245 (* 1 = 2.01245 loss) I0407 22:52:24.955740 359 sgd_solver.cpp:105] Iteration 3240, lr = 0.00974595 I0407 22:52:29.877923 359 solver.cpp:218] Iteration 3252 (2.43795 iter/s, 4.92217s/12 iters), loss = 1.8181 I0407 22:52:29.877959 359 solver.cpp:237] Train net output #0: loss = 1.8181 (* 1 = 1.8181 loss) I0407 22:52:29.877965 359 sgd_solver.cpp:105] Iteration 3252, lr = 0.00974005 I0407 22:52:34.374315 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_3264.caffemodel I0407 22:52:37.939534 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_3264.solverstate I0407 22:52:42.060667 359 solver.cpp:330] Iteration 3264, Testing net (#0) I0407 22:52:42.060688 359 net.cpp:676] Ignoring source layer train-data I0407 22:52:45.365583 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:52:46.692616 359 solver.cpp:397] Test net output #0: accuracy = 0.311274 I0407 22:52:46.692664 359 solver.cpp:397] Test net output #1: loss = 3.02124 (* 1 = 3.02124 loss) I0407 22:52:46.789216 359 solver.cpp:218] Iteration 3264 (0.709588 iter/s, 16.9112s/12 iters), loss = 1.81548 I0407 22:52:46.789254 359 solver.cpp:237] Train net output #0: loss = 1.81548 (* 1 = 1.81548 loss) I0407 22:52:46.789263 359 sgd_solver.cpp:105] Iteration 3264, lr = 0.00973403 I0407 22:52:50.907032 359 solver.cpp:218] Iteration 3276 (2.9142 iter/s, 4.11776s/12 iters), loss = 1.84454 I0407 22:52:50.907068 359 solver.cpp:237] Train net output #0: loss = 1.84454 (* 1 = 1.84454 loss) I0407 22:52:50.907074 359 sgd_solver.cpp:105] Iteration 3276, lr = 0.00972787 I0407 22:52:55.844164 359 solver.cpp:218] Iteration 3288 (2.43059 iter/s, 4.93707s/12 iters), loss = 1.45229 I0407 22:52:55.844206 359 solver.cpp:237] Train net output #0: loss = 1.45229 (* 1 = 1.45229 loss) I0407 22:52:55.844214 359 sgd_solver.cpp:105] Iteration 3288, lr = 0.00972157 I0407 22:53:00.785689 359 solver.cpp:218] Iteration 3300 (2.42843 iter/s, 4.94146s/12 iters), loss = 1.59583 I0407 22:53:00.785729 359 solver.cpp:237] Train net output #0: loss = 1.59583 (* 1 = 1.59583 loss) I0407 22:53:00.785737 359 sgd_solver.cpp:105] Iteration 3300, lr = 0.00971513 I0407 22:53:05.713490 359 solver.cpp:218] Iteration 3312 (2.43519 iter/s, 4.92774s/12 iters), loss = 1.94547 I0407 22:53:05.713624 359 solver.cpp:237] Train net output #0: loss = 1.94547 (* 1 = 1.94547 loss) I0407 22:53:05.713632 359 sgd_solver.cpp:105] Iteration 3312, lr = 0.00970855 I0407 22:53:10.669137 359 solver.cpp:218] Iteration 3324 (2.42156 iter/s, 4.95549s/12 iters), loss = 1.7866 I0407 22:53:10.669180 359 solver.cpp:237] Train net output #0: loss = 1.7866 (* 1 = 1.7866 loss) I0407 22:53:10.669188 359 sgd_solver.cpp:105] Iteration 3324, lr = 0.00970181 I0407 22:53:15.620919 359 solver.cpp:218] Iteration 3336 (2.4234 iter/s, 4.95171s/12 iters), loss = 1.43149 I0407 22:53:15.620962 359 solver.cpp:237] Train net output #0: loss = 1.43149 (* 1 = 1.43149 loss) I0407 22:53:15.620971 359 sgd_solver.cpp:105] Iteration 3336, lr = 0.00969493 I0407 22:53:16.067467 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:53:20.512449 359 solver.cpp:218] Iteration 3348 (2.45325 iter/s, 4.89146s/12 iters), loss = 1.57974 I0407 22:53:20.512490 359 solver.cpp:237] Train net output #0: loss = 1.57974 (* 1 = 1.57974 loss) I0407 22:53:20.512499 359 sgd_solver.cpp:105] Iteration 3348, lr = 0.00968789 I0407 22:53:25.477869 359 solver.cpp:218] Iteration 3360 (2.41675 iter/s, 4.96535s/12 iters), loss = 1.89577 I0407 22:53:25.477910 359 solver.cpp:237] Train net output #0: loss = 1.89577 (* 1 = 1.89577 loss) I0407 22:53:25.477918 359 sgd_solver.cpp:105] Iteration 3360, lr = 0.0096807 I0407 22:53:27.475548 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_3366.caffemodel I0407 22:53:30.577740 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_3366.solverstate I0407 22:53:34.989799 359 solver.cpp:330] Iteration 3366, Testing net (#0) I0407 22:53:34.989815 359 net.cpp:676] Ignoring source layer train-data I0407 22:53:38.015815 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:53:39.354274 359 solver.cpp:397] Test net output #0: accuracy = 0.303922 I0407 22:53:39.354315 359 solver.cpp:397] Test net output #1: loss = 2.98053 (* 1 = 2.98053 loss) I0407 22:53:41.153334 359 solver.cpp:218] Iteration 3372 (0.765531 iter/s, 15.6754s/12 iters), loss = 1.74271 I0407 22:53:41.153375 359 solver.cpp:237] Train net output #0: loss = 1.74271 (* 1 = 1.74271 loss) I0407 22:53:41.153383 359 sgd_solver.cpp:105] Iteration 3372, lr = 0.00967335 I0407 22:53:46.088308 359 solver.cpp:218] Iteration 3384 (2.43165 iter/s, 4.93491s/12 iters), loss = 1.61365 I0407 22:53:46.088351 359 solver.cpp:237] Train net output #0: loss = 1.61365 (* 1 = 1.61365 loss) I0407 22:53:46.088357 359 sgd_solver.cpp:105] Iteration 3384, lr = 0.00966583 I0407 22:53:51.047209 359 solver.cpp:218] Iteration 3396 (2.41992 iter/s, 4.95884s/12 iters), loss = 1.52177 I0407 22:53:51.047253 359 solver.cpp:237] Train net output #0: loss = 1.52177 (* 1 = 1.52177 loss) I0407 22:53:51.047261 359 sgd_solver.cpp:105] Iteration 3396, lr = 0.00965815 I0407 22:53:55.980113 359 solver.cpp:218] Iteration 3408 (2.43268 iter/s, 4.93283s/12 iters), loss = 1.84539 I0407 22:53:55.980157 359 solver.cpp:237] Train net output #0: loss = 1.84539 (* 1 = 1.84539 loss) I0407 22:53:55.980165 359 sgd_solver.cpp:105] Iteration 3408, lr = 0.00965029 I0407 22:54:00.915119 359 solver.cpp:218] Iteration 3420 (2.43164 iter/s, 4.93494s/12 iters), loss = 1.69592 I0407 22:54:00.915166 359 solver.cpp:237] Train net output #0: loss = 1.69592 (* 1 = 1.69592 loss) I0407 22:54:00.915174 359 sgd_solver.cpp:105] Iteration 3420, lr = 0.00964226 I0407 22:54:05.856339 359 solver.cpp:218] Iteration 3432 (2.42858 iter/s, 4.94115s/12 iters), loss = 1.71626 I0407 22:54:05.856387 359 solver.cpp:237] Train net output #0: loss = 1.71626 (* 1 = 1.71626 loss) I0407 22:54:05.856395 359 sgd_solver.cpp:105] Iteration 3432, lr = 0.00963406 I0407 22:54:08.425891 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:54:10.768893 359 solver.cpp:218] Iteration 3444 (2.44276 iter/s, 4.91248s/12 iters), loss = 1.78531 I0407 22:54:10.768937 359 solver.cpp:237] Train net output #0: loss = 1.78531 (* 1 = 1.78531 loss) I0407 22:54:10.768946 359 sgd_solver.cpp:105] Iteration 3444, lr = 0.00962567 I0407 22:54:15.631819 359 solver.cpp:218] Iteration 3456 (2.46768 iter/s, 4.86286s/12 iters), loss = 1.8038 I0407 22:54:15.631856 359 solver.cpp:237] Train net output #0: loss = 1.8038 (* 1 = 1.8038 loss) I0407 22:54:15.631863 359 sgd_solver.cpp:105] Iteration 3456, lr = 0.0096171 I0407 22:54:20.116326 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_3468.caffemodel I0407 22:54:23.230362 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_3468.solverstate I0407 22:54:25.599941 359 solver.cpp:330] Iteration 3468, Testing net (#0) I0407 22:54:25.599961 359 net.cpp:676] Ignoring source layer train-data I0407 22:54:25.979009 359 blocking_queue.cpp:49] Waiting for data I0407 22:54:28.577630 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:54:29.964434 359 solver.cpp:397] Test net output #0: accuracy = 0.297181 I0407 22:54:29.964470 359 solver.cpp:397] Test net output #1: loss = 3.10223 (* 1 = 3.10223 loss) I0407 22:54:30.060896 359 solver.cpp:218] Iteration 3468 (0.831658 iter/s, 14.429s/12 iters), loss = 1.71592 I0407 22:54:30.060936 359 solver.cpp:237] Train net output #0: loss = 1.71592 (* 1 = 1.71592 loss) I0407 22:54:30.060943 359 sgd_solver.cpp:105] Iteration 3468, lr = 0.00960834 I0407 22:54:34.042717 359 solver.cpp:218] Iteration 3480 (3.01374 iter/s, 3.98176s/12 iters), loss = 1.78015 I0407 22:54:34.042757 359 solver.cpp:237] Train net output #0: loss = 1.78015 (* 1 = 1.78015 loss) I0407 22:54:34.042765 359 sgd_solver.cpp:105] Iteration 3480, lr = 0.00959939 I0407 22:54:38.969643 359 solver.cpp:218] Iteration 3492 (2.43563 iter/s, 4.92686s/12 iters), loss = 1.64672 I0407 22:54:38.969753 359 solver.cpp:237] Train net output #0: loss = 1.64672 (* 1 = 1.64672 loss) I0407 22:54:38.969760 359 sgd_solver.cpp:105] Iteration 3492, lr = 0.00959024 I0407 22:54:43.925062 359 solver.cpp:218] Iteration 3504 (2.42166 iter/s, 4.95528s/12 iters), loss = 1.89541 I0407 22:54:43.925109 359 solver.cpp:237] Train net output #0: loss = 1.89541 (* 1 = 1.89541 loss) I0407 22:54:43.925118 359 sgd_solver.cpp:105] Iteration 3504, lr = 0.0095809 I0407 22:54:48.824512 359 solver.cpp:218] Iteration 3516 (2.44929 iter/s, 4.89939s/12 iters), loss = 1.91145 I0407 22:54:48.824544 359 solver.cpp:237] Train net output #0: loss = 1.91145 (* 1 = 1.91145 loss) I0407 22:54:48.824553 359 sgd_solver.cpp:105] Iteration 3516, lr = 0.00957135 I0407 22:54:53.741083 359 solver.cpp:218] Iteration 3528 (2.44075 iter/s, 4.91651s/12 iters), loss = 1.44124 I0407 22:54:53.741124 359 solver.cpp:237] Train net output #0: loss = 1.44124 (* 1 = 1.44124 loss) I0407 22:54:53.741133 359 sgd_solver.cpp:105] Iteration 3528, lr = 0.00956159 I0407 22:54:58.331110 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:54:58.587404 359 solver.cpp:218] Iteration 3540 (2.47614 iter/s, 4.84626s/12 iters), loss = 1.56631 I0407 22:54:58.587445 359 solver.cpp:237] Train net output #0: loss = 1.56631 (* 1 = 1.56631 loss) I0407 22:54:58.587453 359 sgd_solver.cpp:105] Iteration 3540, lr = 0.00955162 I0407 22:55:03.489917 359 solver.cpp:218] Iteration 3552 (2.44775 iter/s, 4.90246s/12 iters), loss = 1.99541 I0407 22:55:03.489951 359 solver.cpp:237] Train net output #0: loss = 1.99541 (* 1 = 1.99541 loss) I0407 22:55:03.489959 359 sgd_solver.cpp:105] Iteration 3552, lr = 0.00954143 I0407 22:55:08.438408 359 solver.cpp:218] Iteration 3564 (2.42501 iter/s, 4.94844s/12 iters), loss = 1.75829 I0407 22:55:08.438442 359 solver.cpp:237] Train net output #0: loss = 1.75829 (* 1 = 1.75829 loss) I0407 22:55:08.438449 359 sgd_solver.cpp:105] Iteration 3564, lr = 0.00953103 I0407 22:55:10.426291 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_3570.caffemodel I0407 22:55:13.532604 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_3570.solverstate I0407 22:55:15.900574 359 solver.cpp:330] Iteration 3570, Testing net (#0) I0407 22:55:15.900593 359 net.cpp:676] Ignoring source layer train-data I0407 22:55:18.917930 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:55:20.338215 359 solver.cpp:397] Test net output #0: accuracy = 0.317402 I0407 22:55:20.338258 359 solver.cpp:397] Test net output #1: loss = 3.03856 (* 1 = 3.03856 loss) I0407 22:55:22.125113 359 solver.cpp:218] Iteration 3576 (0.876768 iter/s, 13.6866s/12 iters), loss = 1.82348 I0407 22:55:22.125156 359 solver.cpp:237] Train net output #0: loss = 1.82348 (* 1 = 1.82348 loss) I0407 22:55:22.125164 359 sgd_solver.cpp:105] Iteration 3576, lr = 0.0095204 I0407 22:55:27.052601 359 solver.cpp:218] Iteration 3588 (2.43535 iter/s, 4.92742s/12 iters), loss = 1.5205 I0407 22:55:27.052639 359 solver.cpp:237] Train net output #0: loss = 1.5205 (* 1 = 1.5205 loss) I0407 22:55:27.052646 359 sgd_solver.cpp:105] Iteration 3588, lr = 0.00950954 I0407 22:55:32.014633 359 solver.cpp:218] Iteration 3600 (2.41839 iter/s, 4.96197s/12 iters), loss = 1.56863 I0407 22:55:32.014675 359 solver.cpp:237] Train net output #0: loss = 1.56863 (* 1 = 1.56863 loss) I0407 22:55:32.014683 359 sgd_solver.cpp:105] Iteration 3600, lr = 0.00949845 I0407 22:55:36.968760 359 solver.cpp:218] Iteration 3612 (2.42225 iter/s, 4.95406s/12 iters), loss = 1.74721 I0407 22:55:36.968801 359 solver.cpp:237] Train net output #0: loss = 1.74721 (* 1 = 1.74721 loss) I0407 22:55:36.968809 359 sgd_solver.cpp:105] Iteration 3612, lr = 0.00948712 I0407 22:55:41.915572 359 solver.cpp:218] Iteration 3624 (2.42583 iter/s, 4.94675s/12 iters), loss = 1.32363 I0407 22:55:41.915701 359 solver.cpp:237] Train net output #0: loss = 1.32363 (* 1 = 1.32363 loss) I0407 22:55:41.915710 359 sgd_solver.cpp:105] Iteration 3624, lr = 0.00947555 I0407 22:55:46.881790 359 solver.cpp:218] Iteration 3636 (2.4164 iter/s, 4.96607s/12 iters), loss = 1.42952 I0407 22:55:46.881827 359 solver.cpp:237] Train net output #0: loss = 1.42952 (* 1 = 1.42952 loss) I0407 22:55:46.881835 359 sgd_solver.cpp:105] Iteration 3636, lr = 0.00946373 I0407 22:55:48.705510 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:55:51.770571 359 solver.cpp:218] Iteration 3648 (2.45463 iter/s, 4.88872s/12 iters), loss = 1.4643 I0407 22:55:51.770614 359 solver.cpp:237] Train net output #0: loss = 1.4643 (* 1 = 1.4643 loss) I0407 22:55:51.770622 359 sgd_solver.cpp:105] Iteration 3648, lr = 0.00945166 I0407 22:55:56.723450 359 solver.cpp:218] Iteration 3660 (2.42286 iter/s, 4.95282s/12 iters), loss = 1.42625 I0407 22:55:56.723487 359 solver.cpp:237] Train net output #0: loss = 1.42625 (* 1 = 1.42625 loss) I0407 22:55:56.723495 359 sgd_solver.cpp:105] Iteration 3660, lr = 0.00943934 I0407 22:56:01.163729 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_3672.caffemodel I0407 22:56:04.252982 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_3672.solverstate I0407 22:56:06.616099 359 solver.cpp:330] Iteration 3672, Testing net (#0) I0407 22:56:06.616117 359 net.cpp:676] Ignoring source layer train-data I0407 22:56:09.556977 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:56:11.049046 359 solver.cpp:397] Test net output #0: accuracy = 0.322304 I0407 22:56:11.049093 359 solver.cpp:397] Test net output #1: loss = 3.03693 (* 1 = 3.03693 loss) I0407 22:56:11.145462 359 solver.cpp:218] Iteration 3672 (0.832065 iter/s, 14.4219s/12 iters), loss = 1.59599 I0407 22:56:11.145499 359 solver.cpp:237] Train net output #0: loss = 1.59599 (* 1 = 1.59599 loss) I0407 22:56:11.145506 359 sgd_solver.cpp:105] Iteration 3672, lr = 0.00942676 I0407 22:56:15.304809 359 solver.cpp:218] Iteration 3684 (2.88511 iter/s, 4.15929s/12 iters), loss = 1.48452 I0407 22:56:15.304971 359 solver.cpp:237] Train net output #0: loss = 1.48452 (* 1 = 1.48452 loss) I0407 22:56:15.304980 359 sgd_solver.cpp:105] Iteration 3684, lr = 0.00941391 I0407 22:56:20.271034 359 solver.cpp:218] Iteration 3696 (2.41641 iter/s, 4.96604s/12 iters), loss = 1.43322 I0407 22:56:20.271077 359 solver.cpp:237] Train net output #0: loss = 1.43322 (* 1 = 1.43322 loss) I0407 22:56:20.271086 359 sgd_solver.cpp:105] Iteration 3696, lr = 0.00940079 I0407 22:56:25.196681 359 solver.cpp:218] Iteration 3708 (2.43626 iter/s, 4.92558s/12 iters), loss = 1.16543 I0407 22:56:25.196725 359 solver.cpp:237] Train net output #0: loss = 1.16543 (* 1 = 1.16543 loss) I0407 22:56:25.196735 359 sgd_solver.cpp:105] Iteration 3708, lr = 0.0093874 I0407 22:56:30.150964 359 solver.cpp:218] Iteration 3720 (2.42218 iter/s, 4.95422s/12 iters), loss = 1.78309 I0407 22:56:30.151001 359 solver.cpp:237] Train net output #0: loss = 1.78309 (* 1 = 1.78309 loss) I0407 22:56:30.151010 359 sgd_solver.cpp:105] Iteration 3720, lr = 0.00937373 I0407 22:56:35.094007 359 solver.cpp:218] Iteration 3732 (2.42768 iter/s, 4.94298s/12 iters), loss = 1.28626 I0407 22:56:35.094048 359 solver.cpp:237] Train net output #0: loss = 1.28626 (* 1 = 1.28626 loss) I0407 22:56:35.094056 359 sgd_solver.cpp:105] Iteration 3732, lr = 0.00935977 I0407 22:56:39.051785 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:56:40.023315 359 solver.cpp:218] Iteration 3744 (2.43445 iter/s, 4.92925s/12 iters), loss = 1.57416 I0407 22:56:40.023353 359 solver.cpp:237] Train net output #0: loss = 1.57416 (* 1 = 1.57416 loss) I0407 22:56:40.023362 359 sgd_solver.cpp:105] Iteration 3744, lr = 0.00934553 I0407 22:56:45.009591 359 solver.cpp:218] Iteration 3756 (2.40663 iter/s, 4.98622s/12 iters), loss = 1.47913 I0407 22:56:45.009625 359 solver.cpp:237] Train net output #0: loss = 1.47913 (* 1 = 1.47913 loss) I0407 22:56:45.009634 359 sgd_solver.cpp:105] Iteration 3756, lr = 0.00933099 I0407 22:56:49.940285 359 solver.cpp:218] Iteration 3768 (2.43376 iter/s, 4.93064s/12 iters), loss = 1.3796 I0407 22:56:49.940412 359 solver.cpp:237] Train net output #0: loss = 1.3796 (* 1 = 1.3796 loss) I0407 22:56:49.940421 359 sgd_solver.cpp:105] Iteration 3768, lr = 0.00931615 I0407 22:56:51.956657 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_3774.caffemodel I0407 22:56:55.111392 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_3774.solverstate I0407 22:56:57.535313 359 solver.cpp:330] Iteration 3774, Testing net (#0) I0407 22:56:57.535332 359 net.cpp:676] Ignoring source layer train-data I0407 22:57:00.533215 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:57:02.193050 359 solver.cpp:397] Test net output #0: accuracy = 0.308211 I0407 22:57:02.193097 359 solver.cpp:397] Test net output #1: loss = 3.07509 (* 1 = 3.07509 loss) I0407 22:57:03.967428 359 solver.cpp:218] Iteration 3780 (0.855493 iter/s, 14.027s/12 iters), loss = 1.33665 I0407 22:57:03.967471 359 solver.cpp:237] Train net output #0: loss = 1.33665 (* 1 = 1.33665 loss) I0407 22:57:03.967479 359 sgd_solver.cpp:105] Iteration 3780, lr = 0.00930101 I0407 22:57:08.915087 359 solver.cpp:218] Iteration 3792 (2.42542 iter/s, 4.94759s/12 iters), loss = 1.36408 I0407 22:57:08.915132 359 solver.cpp:237] Train net output #0: loss = 1.36408 (* 1 = 1.36408 loss) I0407 22:57:08.915140 359 sgd_solver.cpp:105] Iteration 3792, lr = 0.00928555 I0407 22:57:13.822329 359 solver.cpp:218] Iteration 3804 (2.4454 iter/s, 4.90717s/12 iters), loss = 1.28694 I0407 22:57:13.822369 359 solver.cpp:237] Train net output #0: loss = 1.28694 (* 1 = 1.28694 loss) I0407 22:57:13.822377 359 sgd_solver.cpp:105] Iteration 3804, lr = 0.00926979 I0407 22:57:18.775259 359 solver.cpp:218] Iteration 3816 (2.42284 iter/s, 4.95287s/12 iters), loss = 1.4879 I0407 22:57:18.775296 359 solver.cpp:237] Train net output #0: loss = 1.4879 (* 1 = 1.4879 loss) I0407 22:57:18.775305 359 sgd_solver.cpp:105] Iteration 3816, lr = 0.0092537 I0407 22:57:23.694917 359 solver.cpp:218] Iteration 3828 (2.43922 iter/s, 4.9196s/12 iters), loss = 1.40069 I0407 22:57:23.695066 359 solver.cpp:237] Train net output #0: loss = 1.40069 (* 1 = 1.40069 loss) I0407 22:57:23.695076 359 sgd_solver.cpp:105] Iteration 3828, lr = 0.00923728 I0407 22:57:28.654698 359 solver.cpp:218] Iteration 3840 (2.41955 iter/s, 4.95961s/12 iters), loss = 1.30807 I0407 22:57:28.654744 359 solver.cpp:237] Train net output #0: loss = 1.30807 (* 1 = 1.30807 loss) I0407 22:57:28.654753 359 sgd_solver.cpp:105] Iteration 3840, lr = 0.00922054 I0407 22:57:29.756131 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:57:33.486899 359 solver.cpp:218] Iteration 3852 (2.48337 iter/s, 4.83214s/12 iters), loss = 1.5054 I0407 22:57:33.486941 359 solver.cpp:237] Train net output #0: loss = 1.5054 (* 1 = 1.5054 loss) I0407 22:57:33.486949 359 sgd_solver.cpp:105] Iteration 3852, lr = 0.00920346 I0407 22:57:38.419373 359 solver.cpp:218] Iteration 3864 (2.43289 iter/s, 4.93241s/12 iters), loss = 1.32235 I0407 22:57:38.419415 359 solver.cpp:237] Train net output #0: loss = 1.32235 (* 1 = 1.32235 loss) I0407 22:57:38.419423 359 sgd_solver.cpp:105] Iteration 3864, lr = 0.00918604 I0407 22:57:42.897219 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_3876.caffemodel I0407 22:57:46.039988 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_3876.solverstate I0407 22:57:48.402369 359 solver.cpp:330] Iteration 3876, Testing net (#0) I0407 22:57:48.402385 359 net.cpp:676] Ignoring source layer train-data I0407 22:57:51.336009 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:57:53.038373 359 solver.cpp:397] Test net output #0: accuracy = 0.322304 I0407 22:57:53.038420 359 solver.cpp:397] Test net output #1: loss = 3.15085 (* 1 = 3.15085 loss) I0407 22:57:53.135298 359 solver.cpp:218] Iteration 3876 (0.815447 iter/s, 14.7159s/12 iters), loss = 1.29692 I0407 22:57:53.135342 359 solver.cpp:237] Train net output #0: loss = 1.29692 (* 1 = 1.29692 loss) I0407 22:57:53.135350 359 sgd_solver.cpp:105] Iteration 3876, lr = 0.00916827 I0407 22:57:57.240638 359 solver.cpp:218] Iteration 3888 (2.92307 iter/s, 4.10527s/12 iters), loss = 1.1866 I0407 22:57:57.240765 359 solver.cpp:237] Train net output #0: loss = 1.1866 (* 1 = 1.1866 loss) I0407 22:57:57.240774 359 sgd_solver.cpp:105] Iteration 3888, lr = 0.00915015 I0407 22:58:02.198922 359 solver.cpp:218] Iteration 3900 (2.42027 iter/s, 4.95813s/12 iters), loss = 1.35107 I0407 22:58:02.198976 359 solver.cpp:237] Train net output #0: loss = 1.35107 (* 1 = 1.35107 loss) I0407 22:58:02.198987 359 sgd_solver.cpp:105] Iteration 3900, lr = 0.00913168 I0407 22:58:07.118937 359 solver.cpp:218] Iteration 3912 (2.43905 iter/s, 4.91994s/12 iters), loss = 1.32408 I0407 22:58:07.118981 359 solver.cpp:237] Train net output #0: loss = 1.32408 (* 1 = 1.32408 loss) I0407 22:58:07.118989 359 sgd_solver.cpp:105] Iteration 3912, lr = 0.00911284 I0407 22:58:12.036504 359 solver.cpp:218] Iteration 3924 (2.44026 iter/s, 4.9175s/12 iters), loss = 1.04679 I0407 22:58:12.036551 359 solver.cpp:237] Train net output #0: loss = 1.04679 (* 1 = 1.04679 loss) I0407 22:58:12.036559 359 sgd_solver.cpp:105] Iteration 3924, lr = 0.00909363 I0407 22:58:16.966168 359 solver.cpp:218] Iteration 3936 (2.43428 iter/s, 4.9296s/12 iters), loss = 1.03738 I0407 22:58:16.966207 359 solver.cpp:237] Train net output #0: loss = 1.03738 (* 1 = 1.03738 loss) I0407 22:58:16.966214 359 sgd_solver.cpp:105] Iteration 3936, lr = 0.00907405 I0407 22:58:20.266736 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:58:21.915828 359 solver.cpp:218] Iteration 3948 (2.42444 iter/s, 4.9496s/12 iters), loss = 1.55883 I0407 22:58:21.915870 359 solver.cpp:237] Train net output #0: loss = 1.55883 (* 1 = 1.55883 loss) I0407 22:58:21.915879 359 sgd_solver.cpp:105] Iteration 3948, lr = 0.00905409 I0407 22:58:26.839010 359 solver.cpp:218] Iteration 3960 (2.43748 iter/s, 4.92312s/12 iters), loss = 1.16935 I0407 22:58:26.839046 359 solver.cpp:237] Train net output #0: loss = 1.16935 (* 1 = 1.16935 loss) I0407 22:58:26.839053 359 sgd_solver.cpp:105] Iteration 3960, lr = 0.00903374 I0407 22:58:31.700322 359 solver.cpp:218] Iteration 3972 (2.4685 iter/s, 4.86125s/12 iters), loss = 1.00229 I0407 22:58:31.700532 359 solver.cpp:237] Train net output #0: loss = 1.00229 (* 1 = 1.00229 loss) I0407 22:58:31.700547 359 sgd_solver.cpp:105] Iteration 3972, lr = 0.00901301 I0407 22:58:33.714874 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_3978.caffemodel I0407 22:58:36.835970 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_3978.solverstate I0407 22:58:39.232069 359 solver.cpp:330] Iteration 3978, Testing net (#0) I0407 22:58:39.232086 359 net.cpp:676] Ignoring source layer train-data I0407 22:58:42.364213 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:58:44.073710 359 solver.cpp:397] Test net output #0: accuracy = 0.311274 I0407 22:58:44.073755 359 solver.cpp:397] Test net output #1: loss = 3.19912 (* 1 = 3.19912 loss) I0407 22:58:45.872015 359 solver.cpp:218] Iteration 3984 (0.846772 iter/s, 14.1715s/12 iters), loss = 1.46864 I0407 22:58:45.872053 359 solver.cpp:237] Train net output #0: loss = 1.46864 (* 1 = 1.46864 loss) I0407 22:58:45.872061 359 sgd_solver.cpp:105] Iteration 3984, lr = 0.00899188 I0407 22:58:50.795529 359 solver.cpp:218] Iteration 3996 (2.43731 iter/s, 4.92346s/12 iters), loss = 1.06736 I0407 22:58:50.795572 359 solver.cpp:237] Train net output #0: loss = 1.06736 (* 1 = 1.06736 loss) I0407 22:58:50.795580 359 sgd_solver.cpp:105] Iteration 3996, lr = 0.00897035 I0407 22:58:55.753439 359 solver.cpp:218] Iteration 4008 (2.42041 iter/s, 4.95784s/12 iters), loss = 1.24715 I0407 22:58:55.753484 359 solver.cpp:237] Train net output #0: loss = 1.24715 (* 1 = 1.24715 loss) I0407 22:58:55.753494 359 sgd_solver.cpp:105] Iteration 4008, lr = 0.00894841 I0407 22:59:00.688710 359 solver.cpp:218] Iteration 4020 (2.43151 iter/s, 4.93521s/12 iters), loss = 1.62328 I0407 22:59:00.688757 359 solver.cpp:237] Train net output #0: loss = 1.62328 (* 1 = 1.62328 loss) I0407 22:59:00.688766 359 sgd_solver.cpp:105] Iteration 4020, lr = 0.00892607 I0407 22:59:05.590694 359 solver.cpp:218] Iteration 4032 (2.44802 iter/s, 4.90192s/12 iters), loss = 1.28528 I0407 22:59:05.590814 359 solver.cpp:237] Train net output #0: loss = 1.28528 (* 1 = 1.28528 loss) I0407 22:59:05.590822 359 sgd_solver.cpp:105] Iteration 4032, lr = 0.0089033 I0407 22:59:10.494958 359 solver.cpp:218] Iteration 4044 (2.44692 iter/s, 4.90412s/12 iters), loss = 1.08686 I0407 22:59:10.495004 359 solver.cpp:237] Train net output #0: loss = 1.08686 (* 1 = 1.08686 loss) I0407 22:59:10.495012 359 sgd_solver.cpp:105] Iteration 4044, lr = 0.00888011 I0407 22:59:10.971974 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:59:15.421900 359 solver.cpp:218] Iteration 4056 (2.43562 iter/s, 4.92688s/12 iters), loss = 1.43536 I0407 22:59:15.421945 359 solver.cpp:237] Train net output #0: loss = 1.43536 (* 1 = 1.43536 loss) I0407 22:59:15.421953 359 sgd_solver.cpp:105] Iteration 4056, lr = 0.0088565 I0407 22:59:20.351946 359 solver.cpp:218] Iteration 4068 (2.43409 iter/s, 4.92998s/12 iters), loss = 1.2736 I0407 22:59:20.351989 359 solver.cpp:237] Train net output #0: loss = 1.2736 (* 1 = 1.2736 loss) I0407 22:59:20.351996 359 sgd_solver.cpp:105] Iteration 4068, lr = 0.00883245 I0407 22:59:24.777037 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_4080.caffemodel I0407 22:59:29.177081 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_4080.solverstate I0407 22:59:31.699450 359 solver.cpp:330] Iteration 4080, Testing net (#0) I0407 22:59:31.699465 359 net.cpp:676] Ignoring source layer train-data I0407 22:59:34.641240 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 22:59:36.284500 359 solver.cpp:397] Test net output #0: accuracy = 0.323529 I0407 22:59:36.284646 359 solver.cpp:397] Test net output #1: loss = 3.14619 (* 1 = 3.14619 loss) I0407 22:59:36.381498 359 solver.cpp:218] Iteration 4080 (0.748621 iter/s, 16.0295s/12 iters), loss = 1.19587 I0407 22:59:36.381542 359 solver.cpp:237] Train net output #0: loss = 1.19587 (* 1 = 1.19587 loss) I0407 22:59:36.381549 359 sgd_solver.cpp:105] Iteration 4080, lr = 0.00880797 I0407 22:59:40.510213 359 solver.cpp:218] Iteration 4092 (2.90652 iter/s, 4.12865s/12 iters), loss = 1.26657 I0407 22:59:40.510258 359 solver.cpp:237] Train net output #0: loss = 1.26657 (* 1 = 1.26657 loss) I0407 22:59:40.510267 359 sgd_solver.cpp:105] Iteration 4092, lr = 0.00878304 I0407 22:59:45.452402 359 solver.cpp:218] Iteration 4104 (2.42811 iter/s, 4.94212s/12 iters), loss = 1.40396 I0407 22:59:45.452447 359 solver.cpp:237] Train net output #0: loss = 1.40396 (* 1 = 1.40396 loss) I0407 22:59:45.452455 359 sgd_solver.cpp:105] Iteration 4104, lr = 0.00875767 I0407 22:59:50.409837 359 solver.cpp:218] Iteration 4116 (2.42064 iter/s, 4.95736s/12 iters), loss = 1.27666 I0407 22:59:50.409880 359 solver.cpp:237] Train net output #0: loss = 1.27666 (* 1 = 1.27666 loss) I0407 22:59:50.409889 359 sgd_solver.cpp:105] Iteration 4116, lr = 0.00873184 I0407 22:59:55.328012 359 solver.cpp:218] Iteration 4128 (2.43996 iter/s, 4.91811s/12 iters), loss = 1.10314 I0407 22:59:55.328045 359 solver.cpp:237] Train net output #0: loss = 1.10314 (* 1 = 1.10314 loss) I0407 22:59:55.328052 359 sgd_solver.cpp:105] Iteration 4128, lr = 0.00870556 I0407 23:00:00.289180 359 solver.cpp:218] Iteration 4140 (2.41881 iter/s, 4.96111s/12 iters), loss = 1.34791 I0407 23:00:00.289219 359 solver.cpp:237] Train net output #0: loss = 1.34791 (* 1 = 1.34791 loss) I0407 23:00:00.289227 359 sgd_solver.cpp:105] Iteration 4140, lr = 0.00867881 I0407 23:00:02.866626 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:00:05.180112 359 solver.cpp:218] Iteration 4152 (2.45355 iter/s, 4.89087s/12 iters), loss = 1.45668 I0407 23:00:05.180158 359 solver.cpp:237] Train net output #0: loss = 1.45668 (* 1 = 1.45668 loss) I0407 23:00:05.180166 359 sgd_solver.cpp:105] Iteration 4152, lr = 0.0086516 I0407 23:00:06.758963 359 blocking_queue.cpp:49] Waiting for data I0407 23:00:10.133764 359 solver.cpp:218] Iteration 4164 (2.42249 iter/s, 4.95358s/12 iters), loss = 0.80139 I0407 23:00:10.133810 359 solver.cpp:237] Train net output #0: loss = 0.80139 (* 1 = 0.80139 loss) I0407 23:00:10.133818 359 sgd_solver.cpp:105] Iteration 4164, lr = 0.00862391 I0407 23:00:15.034997 359 solver.cpp:218] Iteration 4176 (2.4484 iter/s, 4.90117s/12 iters), loss = 1.35264 I0407 23:00:15.035044 359 solver.cpp:237] Train net output #0: loss = 1.35264 (* 1 = 1.35264 loss) I0407 23:00:15.035053 359 sgd_solver.cpp:105] Iteration 4176, lr = 0.00859575 I0407 23:00:17.019186 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_4182.caffemodel I0407 23:00:21.290120 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_4182.solverstate I0407 23:00:24.105581 359 solver.cpp:330] Iteration 4182, Testing net (#0) I0407 23:00:24.105600 359 net.cpp:676] Ignoring source layer train-data I0407 23:00:27.069430 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:00:28.889843 359 solver.cpp:397] Test net output #0: accuracy = 0.316176 I0407 23:00:28.889892 359 solver.cpp:397] Test net output #1: loss = 3.2268 (* 1 = 3.2268 loss) I0407 23:00:30.651959 359 solver.cpp:218] Iteration 4188 (0.768399 iter/s, 15.6169s/12 iters), loss = 1.26156 I0407 23:00:30.651999 359 solver.cpp:237] Train net output #0: loss = 1.26156 (* 1 = 1.26156 loss) I0407 23:00:30.652007 359 sgd_solver.cpp:105] Iteration 4188, lr = 0.00856711 I0407 23:00:35.575906 359 solver.cpp:218] Iteration 4200 (2.4371 iter/s, 4.92388s/12 iters), loss = 1.15826 I0407 23:00:35.575951 359 solver.cpp:237] Train net output #0: loss = 1.15826 (* 1 = 1.15826 loss) I0407 23:00:35.575959 359 sgd_solver.cpp:105] Iteration 4200, lr = 0.00853798 I0407 23:00:40.534448 359 solver.cpp:218] Iteration 4212 (2.4201 iter/s, 4.95848s/12 iters), loss = 1.24572 I0407 23:00:40.534613 359 solver.cpp:237] Train net output #0: loss = 1.24572 (* 1 = 1.24572 loss) I0407 23:00:40.534622 359 sgd_solver.cpp:105] Iteration 4212, lr = 0.00850836 I0407 23:00:45.426296 359 solver.cpp:218] Iteration 4224 (2.45315 iter/s, 4.89167s/12 iters), loss = 0.929474 I0407 23:00:45.426337 359 solver.cpp:237] Train net output #0: loss = 0.929474 (* 1 = 0.929474 loss) I0407 23:00:45.426344 359 sgd_solver.cpp:105] Iteration 4224, lr = 0.00847826 I0407 23:00:50.392489 359 solver.cpp:218] Iteration 4236 (2.41637 iter/s, 4.96614s/12 iters), loss = 1.17604 I0407 23:00:50.392526 359 solver.cpp:237] Train net output #0: loss = 1.17604 (* 1 = 1.17604 loss) I0407 23:00:50.392534 359 sgd_solver.cpp:105] Iteration 4236, lr = 0.00844765 I0407 23:00:55.056447 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:00:55.281594 359 solver.cpp:218] Iteration 4248 (2.45447 iter/s, 4.88904s/12 iters), loss = 1.20396 I0407 23:00:55.281641 359 solver.cpp:237] Train net output #0: loss = 1.20396 (* 1 = 1.20396 loss) I0407 23:00:55.281648 359 sgd_solver.cpp:105] Iteration 4248, lr = 0.00841654 I0407 23:01:00.224630 359 solver.cpp:218] Iteration 4260 (2.42769 iter/s, 4.94297s/12 iters), loss = 1.30071 I0407 23:01:00.224678 359 solver.cpp:237] Train net output #0: loss = 1.30071 (* 1 = 1.30071 loss) I0407 23:01:00.224687 359 sgd_solver.cpp:105] Iteration 4260, lr = 0.00838493 I0407 23:01:05.163117 359 solver.cpp:218] Iteration 4272 (2.42993 iter/s, 4.93842s/12 iters), loss = 0.907478 I0407 23:01:05.163162 359 solver.cpp:237] Train net output #0: loss = 0.907478 (* 1 = 0.907478 loss) I0407 23:01:05.163172 359 sgd_solver.cpp:105] Iteration 4272, lr = 0.00835281 I0407 23:01:09.643810 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_4284.caffemodel I0407 23:01:12.732884 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_4284.solverstate I0407 23:01:15.492810 359 solver.cpp:330] Iteration 4284, Testing net (#0) I0407 23:01:15.492828 359 net.cpp:676] Ignoring source layer train-data I0407 23:01:18.268864 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:01:19.993580 359 solver.cpp:397] Test net output #0: accuracy = 0.311274 I0407 23:01:19.993623 359 solver.cpp:397] Test net output #1: loss = 3.17454 (* 1 = 3.17454 loss) I0407 23:01:20.090034 359 solver.cpp:218] Iteration 4284 (0.803921 iter/s, 14.9268s/12 iters), loss = 1.1332 I0407 23:01:20.090072 359 solver.cpp:237] Train net output #0: loss = 1.1332 (* 1 = 1.1332 loss) I0407 23:01:20.090080 359 sgd_solver.cpp:105] Iteration 4284, lr = 0.00832018 I0407 23:01:24.208346 359 solver.cpp:218] Iteration 4296 (2.91386 iter/s, 4.11825s/12 iters), loss = 1.09216 I0407 23:01:24.208396 359 solver.cpp:237] Train net output #0: loss = 1.09216 (* 1 = 1.09216 loss) I0407 23:01:24.208410 359 sgd_solver.cpp:105] Iteration 4296, lr = 0.00828704 I0407 23:01:29.155835 359 solver.cpp:218] Iteration 4308 (2.42551 iter/s, 4.94742s/12 iters), loss = 1.20225 I0407 23:01:29.155875 359 solver.cpp:237] Train net output #0: loss = 1.20225 (* 1 = 1.20225 loss) I0407 23:01:29.155884 359 sgd_solver.cpp:105] Iteration 4308, lr = 0.00825338 I0407 23:01:34.067395 359 solver.cpp:218] Iteration 4320 (2.44324 iter/s, 4.9115s/12 iters), loss = 1.19316 I0407 23:01:34.067433 359 solver.cpp:237] Train net output #0: loss = 1.19316 (* 1 = 1.19316 loss) I0407 23:01:34.067441 359 sgd_solver.cpp:105] Iteration 4320, lr = 0.0082192 I0407 23:01:39.042028 359 solver.cpp:218] Iteration 4332 (2.41227 iter/s, 4.97458s/12 iters), loss = 1.08762 I0407 23:01:39.042062 359 solver.cpp:237] Train net output #0: loss = 1.08762 (* 1 = 1.08762 loss) I0407 23:01:39.042069 359 sgd_solver.cpp:105] Iteration 4332, lr = 0.0081845 I0407 23:01:43.957597 359 solver.cpp:218] Iteration 4344 (2.44125 iter/s, 4.91551s/12 iters), loss = 0.923437 I0407 23:01:43.957777 359 solver.cpp:237] Train net output #0: loss = 0.923437 (* 1 = 0.923437 loss) I0407 23:01:43.957798 359 sgd_solver.cpp:105] Iteration 4344, lr = 0.00814928 I0407 23:01:45.825765 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:01:48.904299 359 solver.cpp:218] Iteration 4356 (2.42595 iter/s, 4.94651s/12 iters), loss = 1.15211 I0407 23:01:48.904340 359 solver.cpp:237] Train net output #0: loss = 1.15211 (* 1 = 1.15211 loss) I0407 23:01:48.904348 359 sgd_solver.cpp:105] Iteration 4356, lr = 0.00811353 I0407 23:01:53.825569 359 solver.cpp:218] Iteration 4368 (2.43843 iter/s, 4.9212s/12 iters), loss = 1.17825 I0407 23:01:53.825613 359 solver.cpp:237] Train net output #0: loss = 1.17825 (* 1 = 1.17825 loss) I0407 23:01:53.825620 359 sgd_solver.cpp:105] Iteration 4368, lr = 0.00807725 I0407 23:01:58.777667 359 solver.cpp:218] Iteration 4380 (2.42325 iter/s, 4.95203s/12 iters), loss = 1.07236 I0407 23:01:58.777707 359 solver.cpp:237] Train net output #0: loss = 1.07236 (* 1 = 1.07236 loss) I0407 23:01:58.777716 359 sgd_solver.cpp:105] Iteration 4380, lr = 0.00804044 I0407 23:02:00.772776 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_4386.caffemodel I0407 23:02:03.842010 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_4386.solverstate I0407 23:02:06.212311 359 solver.cpp:330] Iteration 4386, Testing net (#0) I0407 23:02:06.212329 359 net.cpp:676] Ignoring source layer train-data I0407 23:02:08.871178 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:02:10.609735 359 solver.cpp:397] Test net output #0: accuracy = 0.352328 I0407 23:02:10.609783 359 solver.cpp:397] Test net output #1: loss = 3.0698 (* 1 = 3.0698 loss) I0407 23:02:12.402021 359 solver.cpp:218] Iteration 4392 (0.88078 iter/s, 13.6243s/12 iters), loss = 1.30571 I0407 23:02:12.402076 359 solver.cpp:237] Train net output #0: loss = 1.30571 (* 1 = 1.30571 loss) I0407 23:02:12.402084 359 sgd_solver.cpp:105] Iteration 4392, lr = 0.0080031 I0407 23:02:17.323110 359 solver.cpp:218] Iteration 4404 (2.43852 iter/s, 4.92101s/12 iters), loss = 1.05286 I0407 23:02:17.323215 359 solver.cpp:237] Train net output #0: loss = 1.05286 (* 1 = 1.05286 loss) I0407 23:02:17.323225 359 sgd_solver.cpp:105] Iteration 4404, lr = 0.00796523 I0407 23:02:22.288081 359 solver.cpp:218] Iteration 4416 (2.41699 iter/s, 4.96485s/12 iters), loss = 1.1881 I0407 23:02:22.288121 359 solver.cpp:237] Train net output #0: loss = 1.1881 (* 1 = 1.1881 loss) I0407 23:02:22.288130 359 sgd_solver.cpp:105] Iteration 4416, lr = 0.00792683 I0407 23:02:27.180815 359 solver.cpp:218] Iteration 4428 (2.45265 iter/s, 4.89267s/12 iters), loss = 1.18597 I0407 23:02:27.180855 359 solver.cpp:237] Train net output #0: loss = 1.18597 (* 1 = 1.18597 loss) I0407 23:02:27.180863 359 sgd_solver.cpp:105] Iteration 4428, lr = 0.0078879 I0407 23:02:32.143658 359 solver.cpp:218] Iteration 4440 (2.418 iter/s, 4.96278s/12 iters), loss = 0.850651 I0407 23:02:32.143699 359 solver.cpp:237] Train net output #0: loss = 0.850651 (* 1 = 0.850651 loss) I0407 23:02:32.143707 359 sgd_solver.cpp:105] Iteration 4440, lr = 0.00784843 I0407 23:02:36.092711 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:02:37.025038 359 solver.cpp:218] Iteration 4452 (2.45835 iter/s, 4.88131s/12 iters), loss = 1.04663 I0407 23:02:37.025085 359 solver.cpp:237] Train net output #0: loss = 1.04663 (* 1 = 1.04663 loss) I0407 23:02:37.025094 359 sgd_solver.cpp:105] Iteration 4452, lr = 0.00780843 I0407 23:02:41.990219 359 solver.cpp:218] Iteration 4464 (2.41686 iter/s, 4.96511s/12 iters), loss = 1.07428 I0407 23:02:41.990262 359 solver.cpp:237] Train net output #0: loss = 1.07428 (* 1 = 1.07428 loss) I0407 23:02:41.990269 359 sgd_solver.cpp:105] Iteration 4464, lr = 0.0077679 I0407 23:02:46.901567 359 solver.cpp:218] Iteration 4476 (2.44335 iter/s, 4.91128s/12 iters), loss = 0.912665 I0407 23:02:46.901614 359 solver.cpp:237] Train net output #0: loss = 0.912665 (* 1 = 0.912665 loss) I0407 23:02:46.901624 359 sgd_solver.cpp:105] Iteration 4476, lr = 0.00772684 I0407 23:02:51.394461 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_4488.caffemodel I0407 23:02:55.574654 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_4488.solverstate I0407 23:02:59.708824 359 solver.cpp:330] Iteration 4488, Testing net (#0) I0407 23:02:59.708842 359 net.cpp:676] Ignoring source layer train-data I0407 23:03:02.337529 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:03:04.114943 359 solver.cpp:397] Test net output #0: accuracy = 0.344363 I0407 23:03:04.114993 359 solver.cpp:397] Test net output #1: loss = 3.10434 (* 1 = 3.10434 loss) I0407 23:03:04.211478 359 solver.cpp:218] Iteration 4488 (0.693248 iter/s, 17.3098s/12 iters), loss = 0.779825 I0407 23:03:04.211525 359 solver.cpp:237] Train net output #0: loss = 0.779825 (* 1 = 0.779825 loss) I0407 23:03:04.211534 359 sgd_solver.cpp:105] Iteration 4488, lr = 0.00768525 I0407 23:03:08.291023 359 solver.cpp:218] Iteration 4500 (2.94155 iter/s, 4.07948s/12 iters), loss = 0.777058 I0407 23:03:08.291064 359 solver.cpp:237] Train net output #0: loss = 0.777058 (* 1 = 0.777058 loss) I0407 23:03:08.291071 359 sgd_solver.cpp:105] Iteration 4500, lr = 0.00764313 I0407 23:03:13.197472 359 solver.cpp:218] Iteration 4512 (2.44579 iter/s, 4.90639s/12 iters), loss = 0.890965 I0407 23:03:13.197516 359 solver.cpp:237] Train net output #0: loss = 0.890965 (* 1 = 0.890965 loss) I0407 23:03:13.197525 359 sgd_solver.cpp:105] Iteration 4512, lr = 0.00760048 I0407 23:03:18.064895 359 solver.cpp:218] Iteration 4524 (2.4654 iter/s, 4.86736s/12 iters), loss = 1.10968 I0407 23:03:18.064934 359 solver.cpp:237] Train net output #0: loss = 1.10968 (* 1 = 1.10968 loss) I0407 23:03:18.064944 359 sgd_solver.cpp:105] Iteration 4524, lr = 0.0075573 I0407 23:03:23.037331 359 solver.cpp:218] Iteration 4536 (2.41333 iter/s, 4.97237s/12 iters), loss = 0.852607 I0407 23:03:23.037443 359 solver.cpp:237] Train net output #0: loss = 0.852607 (* 1 = 0.852607 loss) I0407 23:03:23.037451 359 sgd_solver.cpp:105] Iteration 4536, lr = 0.00751361 I0407 23:03:28.023815 359 solver.cpp:218] Iteration 4548 (2.40657 iter/s, 4.98635s/12 iters), loss = 1.0859 I0407 23:03:28.023852 359 solver.cpp:237] Train net output #0: loss = 1.0859 (* 1 = 1.0859 loss) I0407 23:03:28.023860 359 sgd_solver.cpp:105] Iteration 4548, lr = 0.00746939 I0407 23:03:29.331871 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:03:33.036319 359 solver.cpp:218] Iteration 4560 (2.39404 iter/s, 5.01245s/12 iters), loss = 1.12001 I0407 23:03:33.036357 359 solver.cpp:237] Train net output #0: loss = 1.12001 (* 1 = 1.12001 loss) I0407 23:03:33.036365 359 sgd_solver.cpp:105] Iteration 4560, lr = 0.00742466 I0407 23:03:37.971357 359 solver.cpp:218] Iteration 4572 (2.43162 iter/s, 4.93498s/12 iters), loss = 1.08174 I0407 23:03:37.971395 359 solver.cpp:237] Train net output #0: loss = 1.08174 (* 1 = 1.08174 loss) I0407 23:03:37.971402 359 sgd_solver.cpp:105] Iteration 4572, lr = 0.00737941 I0407 23:03:42.927739 359 solver.cpp:218] Iteration 4584 (2.42115 iter/s, 4.95633s/12 iters), loss = 0.919457 I0407 23:03:42.927776 359 solver.cpp:237] Train net output #0: loss = 0.919457 (* 1 = 0.919457 loss) I0407 23:03:42.927783 359 sgd_solver.cpp:105] Iteration 4584, lr = 0.00733365 I0407 23:03:45.004302 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_4590.caffemodel I0407 23:03:48.106365 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_4590.solverstate I0407 23:03:50.478353 359 solver.cpp:330] Iteration 4590, Testing net (#0) I0407 23:03:50.478371 359 net.cpp:676] Ignoring source layer train-data I0407 23:03:53.257681 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:03:55.208550 359 solver.cpp:397] Test net output #0: accuracy = 0.355392 I0407 23:03:55.208596 359 solver.cpp:397] Test net output #1: loss = 3.0353 (* 1 = 3.0353 loss) I0407 23:03:57.061884 359 solver.cpp:218] Iteration 4596 (0.849012 iter/s, 14.1341s/12 iters), loss = 1.01907 I0407 23:03:57.061933 359 solver.cpp:237] Train net output #0: loss = 1.01907 (* 1 = 1.01907 loss) I0407 23:03:57.061941 359 sgd_solver.cpp:105] Iteration 4596, lr = 0.00728739 I0407 23:04:02.015368 359 solver.cpp:218] Iteration 4608 (2.42257 iter/s, 4.95342s/12 iters), loss = 1.03239 I0407 23:04:02.015406 359 solver.cpp:237] Train net output #0: loss = 1.03239 (* 1 = 1.03239 loss) I0407 23:04:02.015414 359 sgd_solver.cpp:105] Iteration 4608, lr = 0.00724063 I0407 23:04:07.018111 359 solver.cpp:218] Iteration 4620 (2.39871 iter/s, 5.00269s/12 iters), loss = 0.820322 I0407 23:04:07.018154 359 solver.cpp:237] Train net output #0: loss = 0.820322 (* 1 = 0.820322 loss) I0407 23:04:07.018162 359 sgd_solver.cpp:105] Iteration 4620, lr = 0.00719337 I0407 23:04:11.869942 359 solver.cpp:218] Iteration 4632 (2.47332 iter/s, 4.85177s/12 iters), loss = 1.01248 I0407 23:04:11.869976 359 solver.cpp:237] Train net output #0: loss = 1.01248 (* 1 = 1.01248 loss) I0407 23:04:11.869984 359 sgd_solver.cpp:105] Iteration 4632, lr = 0.00714562 I0407 23:04:16.785553 359 solver.cpp:218] Iteration 4644 (2.44123 iter/s, 4.91555s/12 iters), loss = 0.868681 I0407 23:04:16.785595 359 solver.cpp:237] Train net output #0: loss = 0.868681 (* 1 = 0.868681 loss) I0407 23:04:16.785604 359 sgd_solver.cpp:105] Iteration 4644, lr = 0.00709739 I0407 23:04:20.158550 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:04:21.725051 359 solver.cpp:218] Iteration 4656 (2.42943 iter/s, 4.93943s/12 iters), loss = 0.895408 I0407 23:04:21.725095 359 solver.cpp:237] Train net output #0: loss = 0.895408 (* 1 = 0.895408 loss) I0407 23:04:21.725102 359 sgd_solver.cpp:105] Iteration 4656, lr = 0.00704868 I0407 23:04:26.661913 359 solver.cpp:218] Iteration 4668 (2.43072 iter/s, 4.9368s/12 iters), loss = 1.19164 I0407 23:04:26.662040 359 solver.cpp:237] Train net output #0: loss = 1.19164 (* 1 = 1.19164 loss) I0407 23:04:26.662048 359 sgd_solver.cpp:105] Iteration 4668, lr = 0.0069995 I0407 23:04:31.571571 359 solver.cpp:218] Iteration 4680 (2.44423 iter/s, 4.90951s/12 iters), loss = 0.866611 I0407 23:04:31.571610 359 solver.cpp:237] Train net output #0: loss = 0.866611 (* 1 = 0.866611 loss) I0407 23:04:31.571619 359 sgd_solver.cpp:105] Iteration 4680, lr = 0.00694985 I0407 23:04:36.015635 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_4692.caffemodel I0407 23:04:40.339160 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_4692.solverstate I0407 23:04:43.731345 359 solver.cpp:330] Iteration 4692, Testing net (#0) I0407 23:04:43.731362 359 net.cpp:676] Ignoring source layer train-data I0407 23:04:46.538678 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:04:48.667326 359 solver.cpp:397] Test net output #0: accuracy = 0.341299 I0407 23:04:48.667382 359 solver.cpp:397] Test net output #1: loss = 3.18518 (* 1 = 3.18518 loss) I0407 23:04:48.764021 359 solver.cpp:218] Iteration 4692 (0.697984 iter/s, 17.1924s/12 iters), loss = 1.11231 I0407 23:04:48.764063 359 solver.cpp:237] Train net output #0: loss = 1.11231 (* 1 = 1.11231 loss) I0407 23:04:48.764072 359 sgd_solver.cpp:105] Iteration 4692, lr = 0.00689974 I0407 23:04:52.885627 359 solver.cpp:218] Iteration 4704 (2.91153 iter/s, 4.12154s/12 iters), loss = 0.821108 I0407 23:04:52.885666 359 solver.cpp:237] Train net output #0: loss = 0.821108 (* 1 = 0.821108 loss) I0407 23:04:52.885673 359 sgd_solver.cpp:105] Iteration 4704, lr = 0.00684919 I0407 23:04:57.789563 359 solver.cpp:218] Iteration 4716 (2.44704 iter/s, 4.90388s/12 iters), loss = 0.731509 I0407 23:04:57.789701 359 solver.cpp:237] Train net output #0: loss = 0.731509 (* 1 = 0.731509 loss) I0407 23:04:57.789710 359 sgd_solver.cpp:105] Iteration 4716, lr = 0.00679819 I0407 23:05:02.699545 359 solver.cpp:218] Iteration 4728 (2.44408 iter/s, 4.90982s/12 iters), loss = 1.09188 I0407 23:05:02.699590 359 solver.cpp:237] Train net output #0: loss = 1.09188 (* 1 = 1.09188 loss) I0407 23:05:02.699599 359 sgd_solver.cpp:105] Iteration 4728, lr = 0.00674676 I0407 23:05:07.689182 359 solver.cpp:218] Iteration 4740 (2.40501 iter/s, 4.98957s/12 iters), loss = 0.68834 I0407 23:05:07.689220 359 solver.cpp:237] Train net output #0: loss = 0.68834 (* 1 = 0.68834 loss) I0407 23:05:07.689229 359 sgd_solver.cpp:105] Iteration 4740, lr = 0.00669491 I0407 23:05:12.675359 359 solver.cpp:218] Iteration 4752 (2.40668 iter/s, 4.98612s/12 iters), loss = 0.783096 I0407 23:05:12.675407 359 solver.cpp:237] Train net output #0: loss = 0.783096 (* 1 = 0.783096 loss) I0407 23:05:12.675416 359 sgd_solver.cpp:105] Iteration 4752, lr = 0.00664264 I0407 23:05:13.182657 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:05:17.643110 359 solver.cpp:218] Iteration 4764 (2.41561 iter/s, 4.96769s/12 iters), loss = 0.976755 I0407 23:05:17.643149 359 solver.cpp:237] Train net output #0: loss = 0.976755 (* 1 = 0.976755 loss) I0407 23:05:17.643157 359 sgd_solver.cpp:105] Iteration 4764, lr = 0.00658996 I0407 23:05:22.654903 359 solver.cpp:218] Iteration 4776 (2.39438 iter/s, 5.01173s/12 iters), loss = 0.813546 I0407 23:05:22.654948 359 solver.cpp:237] Train net output #0: loss = 0.813546 (* 1 = 0.813546 loss) I0407 23:05:22.654956 359 sgd_solver.cpp:105] Iteration 4776, lr = 0.00653689 I0407 23:05:27.560628 359 solver.cpp:218] Iteration 4788 (2.44615 iter/s, 4.90566s/12 iters), loss = 0.932335 I0407 23:05:27.560667 359 solver.cpp:237] Train net output #0: loss = 0.932335 (* 1 = 0.932335 loss) I0407 23:05:27.560675 359 sgd_solver.cpp:105] Iteration 4788, lr = 0.00648343 I0407 23:05:29.552392 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_4794.caffemodel I0407 23:05:33.405416 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_4794.solverstate I0407 23:05:36.520601 359 solver.cpp:330] Iteration 4794, Testing net (#0) I0407 23:05:36.520617 359 net.cpp:676] Ignoring source layer train-data I0407 23:05:39.249437 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:05:41.456301 359 solver.cpp:397] Test net output #0: accuracy = 0.364583 I0407 23:05:41.456336 359 solver.cpp:397] Test net output #1: loss = 3.15568 (* 1 = 3.15568 loss) I0407 23:05:43.324231 359 solver.cpp:218] Iteration 4800 (0.761251 iter/s, 15.7635s/12 iters), loss = 0.874731 I0407 23:05:43.324280 359 solver.cpp:237] Train net output #0: loss = 0.874731 (* 1 = 0.874731 loss) I0407 23:05:43.324287 359 sgd_solver.cpp:105] Iteration 4800, lr = 0.0064296 I0407 23:05:48.240773 359 solver.cpp:218] Iteration 4812 (2.44078 iter/s, 4.91647s/12 iters), loss = 0.70028 I0407 23:05:48.240818 359 solver.cpp:237] Train net output #0: loss = 0.70028 (* 1 = 0.70028 loss) I0407 23:05:48.240828 359 sgd_solver.cpp:105] Iteration 4812, lr = 0.00637541 I0407 23:05:53.208781 359 solver.cpp:218] Iteration 4824 (2.41549 iter/s, 4.96793s/12 iters), loss = 0.785002 I0407 23:05:53.208829 359 solver.cpp:237] Train net output #0: loss = 0.785002 (* 1 = 0.785002 loss) I0407 23:05:53.208838 359 sgd_solver.cpp:105] Iteration 4824, lr = 0.00632086 I0407 23:05:58.198913 359 solver.cpp:218] Iteration 4836 (2.40478 iter/s, 4.99007s/12 iters), loss = 0.711259 I0407 23:05:58.198953 359 solver.cpp:237] Train net output #0: loss = 0.711259 (* 1 = 0.711259 loss) I0407 23:05:58.198962 359 sgd_solver.cpp:105] Iteration 4836, lr = 0.00626597 I0407 23:06:00.234081 359 blocking_queue.cpp:49] Waiting for data I0407 23:06:03.153045 359 solver.cpp:218] Iteration 4848 (2.42225 iter/s, 4.95407s/12 iters), loss = 0.77453 I0407 23:06:03.153090 359 solver.cpp:237] Train net output #0: loss = 0.77453 (* 1 = 0.77453 loss) I0407 23:06:03.153097 359 sgd_solver.cpp:105] Iteration 4848, lr = 0.00621076 I0407 23:06:05.812907 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:06:08.101795 359 solver.cpp:218] Iteration 4860 (2.42489 iter/s, 4.94868s/12 iters), loss = 0.931621 I0407 23:06:08.101838 359 solver.cpp:237] Train net output #0: loss = 0.931621 (* 1 = 0.931621 loss) I0407 23:06:08.101847 359 sgd_solver.cpp:105] Iteration 4860, lr = 0.00615523 I0407 23:06:13.024924 359 solver.cpp:218] Iteration 4872 (2.43751 iter/s, 4.92306s/12 iters), loss = 0.982474 I0407 23:06:13.024971 359 solver.cpp:237] Train net output #0: loss = 0.982474 (* 1 = 0.982474 loss) I0407 23:06:13.024979 359 sgd_solver.cpp:105] Iteration 4872, lr = 0.0060994 I0407 23:06:17.970460 359 solver.cpp:218] Iteration 4884 (2.42646 iter/s, 4.94547s/12 iters), loss = 0.737415 I0407 23:06:17.970504 359 solver.cpp:237] Train net output #0: loss = 0.737415 (* 1 = 0.737415 loss) I0407 23:06:17.970512 359 sgd_solver.cpp:105] Iteration 4884, lr = 0.00604327 I0407 23:06:22.354007 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_4896.caffemodel I0407 23:06:25.483212 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_4896.solverstate I0407 23:06:28.072077 359 solver.cpp:330] Iteration 4896, Testing net (#0) I0407 23:06:28.072094 359 net.cpp:676] Ignoring source layer train-data I0407 23:06:30.701858 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:06:32.736217 359 solver.cpp:397] Test net output #0: accuracy = 0.367647 I0407 23:06:32.736244 359 solver.cpp:397] Test net output #1: loss = 3.02149 (* 1 = 3.02149 loss) I0407 23:06:32.832558 359 solver.cpp:218] Iteration 4896 (0.807427 iter/s, 14.862s/12 iters), loss = 0.760391 I0407 23:06:32.832603 359 solver.cpp:237] Train net output #0: loss = 0.760391 (* 1 = 0.760391 loss) I0407 23:06:32.832612 359 sgd_solver.cpp:105] Iteration 4896, lr = 0.00598688 I0407 23:06:36.929620 359 solver.cpp:218] Iteration 4908 (2.92897 iter/s, 4.097s/12 iters), loss = 0.764885 I0407 23:06:36.929656 359 solver.cpp:237] Train net output #0: loss = 0.764885 (* 1 = 0.764885 loss) I0407 23:06:36.929664 359 sgd_solver.cpp:105] Iteration 4908, lr = 0.00593022 I0407 23:06:41.883965 359 solver.cpp:218] Iteration 4920 (2.42215 iter/s, 4.95428s/12 iters), loss = 0.940427 I0407 23:06:41.884011 359 solver.cpp:237] Train net output #0: loss = 0.940427 (* 1 = 0.940427 loss) I0407 23:06:41.884019 359 sgd_solver.cpp:105] Iteration 4920, lr = 0.00587331 I0407 23:06:46.776434 359 solver.cpp:218] Iteration 4932 (2.45279 iter/s, 4.8924s/12 iters), loss = 0.897402 I0407 23:06:46.776484 359 solver.cpp:237] Train net output #0: loss = 0.897402 (* 1 = 0.897402 loss) I0407 23:06:46.776494 359 sgd_solver.cpp:105] Iteration 4932, lr = 0.00581616 I0407 23:06:51.752323 359 solver.cpp:218] Iteration 4944 (2.41166 iter/s, 4.97582s/12 iters), loss = 0.832853 I0407 23:06:51.752368 359 solver.cpp:237] Train net output #0: loss = 0.832853 (* 1 = 0.832853 loss) I0407 23:06:51.752377 359 sgd_solver.cpp:105] Iteration 4944, lr = 0.0057588 I0407 23:06:56.452111 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:06:56.648890 359 solver.cpp:218] Iteration 4956 (2.45073 iter/s, 4.8965s/12 iters), loss = 0.689222 I0407 23:06:56.648924 359 solver.cpp:237] Train net output #0: loss = 0.689222 (* 1 = 0.689222 loss) I0407 23:06:56.648932 359 sgd_solver.cpp:105] Iteration 4956, lr = 0.00570123 I0407 23:07:01.592567 359 solver.cpp:218] Iteration 4968 (2.42737 iter/s, 4.94362s/12 iters), loss = 0.366809 I0407 23:07:01.592710 359 solver.cpp:237] Train net output #0: loss = 0.366809 (* 1 = 0.366809 loss) I0407 23:07:01.592718 359 sgd_solver.cpp:105] Iteration 4968, lr = 0.00564347 I0407 23:07:06.515650 359 solver.cpp:218] Iteration 4980 (2.43757 iter/s, 4.92293s/12 iters), loss = 0.479785 I0407 23:07:06.515688 359 solver.cpp:237] Train net output #0: loss = 0.479785 (* 1 = 0.479785 loss) I0407 23:07:06.515695 359 sgd_solver.cpp:105] Iteration 4980, lr = 0.00558554 I0407 23:07:11.458390 359 solver.cpp:218] Iteration 4992 (2.42783 iter/s, 4.94268s/12 iters), loss = 0.729526 I0407 23:07:11.458427 359 solver.cpp:237] Train net output #0: loss = 0.729526 (* 1 = 0.729526 loss) I0407 23:07:11.458434 359 sgd_solver.cpp:105] Iteration 4992, lr = 0.00552744 I0407 23:07:13.449579 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_4998.caffemodel I0407 23:07:16.561566 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_4998.solverstate I0407 23:07:19.019897 359 solver.cpp:330] Iteration 4998, Testing net (#0) I0407 23:07:19.019917 359 net.cpp:676] Ignoring source layer train-data I0407 23:07:21.678143 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:07:23.734504 359 solver.cpp:397] Test net output #0: accuracy = 0.379902 I0407 23:07:23.734550 359 solver.cpp:397] Test net output #1: loss = 3.08703 (* 1 = 3.08703 loss) I0407 23:07:25.583346 359 solver.cpp:218] Iteration 5004 (0.849564 iter/s, 14.1249s/12 iters), loss = 0.401737 I0407 23:07:25.583397 359 solver.cpp:237] Train net output #0: loss = 0.401737 (* 1 = 0.401737 loss) I0407 23:07:25.583405 359 sgd_solver.cpp:105] Iteration 5004, lr = 0.0054692 I0407 23:07:30.571161 359 solver.cpp:218] Iteration 5016 (2.4059 iter/s, 4.98775s/12 iters), loss = 0.620913 I0407 23:07:30.571208 359 solver.cpp:237] Train net output #0: loss = 0.620913 (* 1 = 0.620913 loss) I0407 23:07:30.571216 359 sgd_solver.cpp:105] Iteration 5016, lr = 0.00541084 I0407 23:07:35.521492 359 solver.cpp:218] Iteration 5028 (2.42411 iter/s, 4.95026s/12 iters), loss = 0.647608 I0407 23:07:35.521632 359 solver.cpp:237] Train net output #0: loss = 0.647608 (* 1 = 0.647608 loss) I0407 23:07:35.521642 359 sgd_solver.cpp:105] Iteration 5028, lr = 0.00535236 I0407 23:07:40.458485 359 solver.cpp:218] Iteration 5040 (2.43071 iter/s, 4.93683s/12 iters), loss = 0.65534 I0407 23:07:40.458532 359 solver.cpp:237] Train net output #0: loss = 0.65534 (* 1 = 0.65534 loss) I0407 23:07:40.458540 359 sgd_solver.cpp:105] Iteration 5040, lr = 0.00529378 I0407 23:07:45.444732 359 solver.cpp:218] Iteration 5052 (2.40665 iter/s, 4.98618s/12 iters), loss = 0.575499 I0407 23:07:45.444774 359 solver.cpp:237] Train net output #0: loss = 0.575499 (* 1 = 0.575499 loss) I0407 23:07:45.444782 359 sgd_solver.cpp:105] Iteration 5052, lr = 0.00523512 I0407 23:07:47.281433 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:07:50.295415 359 solver.cpp:218] Iteration 5064 (2.47391 iter/s, 4.85062s/12 iters), loss = 0.906331 I0407 23:07:50.295460 359 solver.cpp:237] Train net output #0: loss = 0.906331 (* 1 = 0.906331 loss) I0407 23:07:50.295469 359 sgd_solver.cpp:105] Iteration 5064, lr = 0.0051764 I0407 23:07:55.257282 359 solver.cpp:218] Iteration 5076 (2.41848 iter/s, 4.9618s/12 iters), loss = 0.596021 I0407 23:07:55.257321 359 solver.cpp:237] Train net output #0: loss = 0.596021 (* 1 = 0.596021 loss) I0407 23:07:55.257329 359 sgd_solver.cpp:105] Iteration 5076, lr = 0.00511763 I0407 23:08:00.217500 359 solver.cpp:218] Iteration 5088 (2.41928 iter/s, 4.96016s/12 iters), loss = 0.823575 I0407 23:08:00.217536 359 solver.cpp:237] Train net output #0: loss = 0.823575 (* 1 = 0.823575 loss) I0407 23:08:00.217545 359 sgd_solver.cpp:105] Iteration 5088, lr = 0.00505882 I0407 23:08:04.707762 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_5100.caffemodel I0407 23:08:07.831751 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_5100.solverstate I0407 23:08:10.195988 359 solver.cpp:330] Iteration 5100, Testing net (#0) I0407 23:08:10.196007 359 net.cpp:676] Ignoring source layer train-data I0407 23:08:12.549998 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:08:14.568877 359 solver.cpp:397] Test net output #0: accuracy = 0.370098 I0407 23:08:14.568924 359 solver.cpp:397] Test net output #1: loss = 3.10449 (* 1 = 3.10449 loss) I0407 23:08:14.665768 359 solver.cpp:218] Iteration 5100 (0.830553 iter/s, 14.4482s/12 iters), loss = 0.744585 I0407 23:08:14.665807 359 solver.cpp:237] Train net output #0: loss = 0.744585 (* 1 = 0.744585 loss) I0407 23:08:14.665817 359 sgd_solver.cpp:105] Iteration 5100, lr = 0.005 I0407 23:08:18.806478 359 solver.cpp:218] Iteration 5112 (2.89809 iter/s, 4.14065s/12 iters), loss = 0.670824 I0407 23:08:18.806510 359 solver.cpp:237] Train net output #0: loss = 0.670824 (* 1 = 0.670824 loss) I0407 23:08:18.806517 359 sgd_solver.cpp:105] Iteration 5112, lr = 0.00494118 I0407 23:08:23.739210 359 solver.cpp:218] Iteration 5124 (2.43275 iter/s, 4.93268s/12 iters), loss = 0.651989 I0407 23:08:23.739246 359 solver.cpp:237] Train net output #0: loss = 0.651989 (* 1 = 0.651989 loss) I0407 23:08:23.739253 359 sgd_solver.cpp:105] Iteration 5124, lr = 0.00488237 I0407 23:08:28.681856 359 solver.cpp:218] Iteration 5136 (2.42787 iter/s, 4.94259s/12 iters), loss = 0.458284 I0407 23:08:28.681891 359 solver.cpp:237] Train net output #0: loss = 0.458284 (* 1 = 0.458284 loss) I0407 23:08:28.681900 359 sgd_solver.cpp:105] Iteration 5136, lr = 0.0048236 I0407 23:08:33.633774 359 solver.cpp:218] Iteration 5148 (2.42333 iter/s, 4.95186s/12 iters), loss = 0.590291 I0407 23:08:33.633818 359 solver.cpp:237] Train net output #0: loss = 0.590291 (* 1 = 0.590291 loss) I0407 23:08:33.633827 359 sgd_solver.cpp:105] Iteration 5148, lr = 0.00476488 I0407 23:08:37.657568 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:08:38.561758 359 solver.cpp:218] Iteration 5160 (2.43511 iter/s, 4.92792s/12 iters), loss = 0.577291 I0407 23:08:38.561887 359 solver.cpp:237] Train net output #0: loss = 0.577291 (* 1 = 0.577291 loss) I0407 23:08:38.561895 359 sgd_solver.cpp:105] Iteration 5160, lr = 0.00470622 I0407 23:08:43.500888 359 solver.cpp:218] Iteration 5172 (2.42965 iter/s, 4.93898s/12 iters), loss = 0.661538 I0407 23:08:43.500932 359 solver.cpp:237] Train net output #0: loss = 0.661538 (* 1 = 0.661538 loss) I0407 23:08:43.500941 359 sgd_solver.cpp:105] Iteration 5172, lr = 0.00464764 I0407 23:08:48.400684 359 solver.cpp:218] Iteration 5184 (2.44911 iter/s, 4.89973s/12 iters), loss = 0.358573 I0407 23:08:48.400730 359 solver.cpp:237] Train net output #0: loss = 0.358573 (* 1 = 0.358573 loss) I0407 23:08:48.400739 359 sgd_solver.cpp:105] Iteration 5184, lr = 0.00458916 I0407 23:08:53.309351 359 solver.cpp:218] Iteration 5196 (2.44469 iter/s, 4.9086s/12 iters), loss = 0.27501 I0407 23:08:53.309396 359 solver.cpp:237] Train net output #0: loss = 0.27501 (* 1 = 0.27501 loss) I0407 23:08:53.309404 359 sgd_solver.cpp:105] Iteration 5196, lr = 0.0045308 I0407 23:08:55.294013 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_5202.caffemodel I0407 23:08:58.374043 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_5202.solverstate I0407 23:09:00.744858 359 solver.cpp:330] Iteration 5202, Testing net (#0) I0407 23:09:00.744876 359 net.cpp:676] Ignoring source layer train-data I0407 23:09:03.268515 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:09:05.517493 359 solver.cpp:397] Test net output #0: accuracy = 0.381127 I0407 23:09:05.517540 359 solver.cpp:397] Test net output #1: loss = 3.10979 (* 1 = 3.10979 loss) I0407 23:09:07.335219 359 solver.cpp:218] Iteration 5208 (0.855567 iter/s, 14.0258s/12 iters), loss = 0.510135 I0407 23:09:07.335266 359 solver.cpp:237] Train net output #0: loss = 0.510135 (* 1 = 0.510135 loss) I0407 23:09:07.335274 359 sgd_solver.cpp:105] Iteration 5208, lr = 0.00447256 I0407 23:09:12.320524 359 solver.cpp:218] Iteration 5220 (2.40711 iter/s, 4.98524s/12 iters), loss = 0.437447 I0407 23:09:12.320667 359 solver.cpp:237] Train net output #0: loss = 0.437447 (* 1 = 0.437447 loss) I0407 23:09:12.320677 359 sgd_solver.cpp:105] Iteration 5220, lr = 0.00441446 I0407 23:09:17.229663 359 solver.cpp:218] Iteration 5232 (2.4445 iter/s, 4.90898s/12 iters), loss = 0.348768 I0407 23:09:17.229701 359 solver.cpp:237] Train net output #0: loss = 0.348768 (* 1 = 0.348768 loss) I0407 23:09:17.229709 359 sgd_solver.cpp:105] Iteration 5232, lr = 0.00435653 I0407 23:09:22.266436 359 solver.cpp:218] Iteration 5244 (2.3825 iter/s, 5.03672s/12 iters), loss = 0.47217 I0407 23:09:22.266476 359 solver.cpp:237] Train net output #0: loss = 0.47217 (* 1 = 0.47217 loss) I0407 23:09:22.266484 359 sgd_solver.cpp:105] Iteration 5244, lr = 0.00429877 I0407 23:09:27.221897 359 solver.cpp:218] Iteration 5256 (2.4216 iter/s, 4.9554s/12 iters), loss = 0.531433 I0407 23:09:27.221944 359 solver.cpp:237] Train net output #0: loss = 0.531433 (* 1 = 0.531433 loss) I0407 23:09:27.221952 359 sgd_solver.cpp:105] Iteration 5256, lr = 0.0042412 I0407 23:09:28.476162 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:09:32.108680 359 solver.cpp:218] Iteration 5268 (2.45564 iter/s, 4.88672s/12 iters), loss = 0.521499 I0407 23:09:32.108714 359 solver.cpp:237] Train net output #0: loss = 0.521499 (* 1 = 0.521499 loss) I0407 23:09:32.108722 359 sgd_solver.cpp:105] Iteration 5268, lr = 0.00418384 I0407 23:09:37.075927 359 solver.cpp:218] Iteration 5280 (2.41585 iter/s, 4.9672s/12 iters), loss = 0.657227 I0407 23:09:37.075966 359 solver.cpp:237] Train net output #0: loss = 0.657227 (* 1 = 0.657227 loss) I0407 23:09:37.075973 359 sgd_solver.cpp:105] Iteration 5280, lr = 0.00412669 I0407 23:09:41.969507 359 solver.cpp:218] Iteration 5292 (2.45222 iter/s, 4.89352s/12 iters), loss = 0.364044 I0407 23:09:41.969550 359 solver.cpp:237] Train net output #0: loss = 0.364044 (* 1 = 0.364044 loss) I0407 23:09:41.969558 359 sgd_solver.cpp:105] Iteration 5292, lr = 0.00406978 I0407 23:09:46.410845 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_5304.caffemodel I0407 23:09:49.721287 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_5304.solverstate I0407 23:09:52.087182 359 solver.cpp:330] Iteration 5304, Testing net (#0) I0407 23:09:52.087206 359 net.cpp:676] Ignoring source layer train-data I0407 23:09:54.421185 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:09:56.533413 359 solver.cpp:397] Test net output #0: accuracy = 0.390319 I0407 23:09:56.533447 359 solver.cpp:397] Test net output #1: loss = 3.06499 (* 1 = 3.06499 loss) I0407 23:09:56.629779 359 solver.cpp:218] Iteration 5304 (0.818544 iter/s, 14.6602s/12 iters), loss = 0.290723 I0407 23:09:56.629854 359 solver.cpp:237] Train net output #0: loss = 0.290723 (* 1 = 0.290723 loss) I0407 23:09:56.629870 359 sgd_solver.cpp:105] Iteration 5304, lr = 0.00401312 I0407 23:10:00.773187 359 solver.cpp:218] Iteration 5316 (2.89623 iter/s, 4.14332s/12 iters), loss = 0.678481 I0407 23:10:00.773232 359 solver.cpp:237] Train net output #0: loss = 0.678481 (* 1 = 0.678481 loss) I0407 23:10:00.773241 359 sgd_solver.cpp:105] Iteration 5316, lr = 0.00395672 I0407 23:10:05.697723 359 solver.cpp:218] Iteration 5328 (2.43681 iter/s, 4.92448s/12 iters), loss = 0.40485 I0407 23:10:05.697759 359 solver.cpp:237] Train net output #0: loss = 0.40485 (* 1 = 0.40485 loss) I0407 23:10:05.697767 359 sgd_solver.cpp:105] Iteration 5328, lr = 0.0039006 I0407 23:10:10.650027 359 solver.cpp:218] Iteration 5340 (2.42314 iter/s, 4.95225s/12 iters), loss = 0.476173 I0407 23:10:10.650063 359 solver.cpp:237] Train net output #0: loss = 0.476173 (* 1 = 0.476173 loss) I0407 23:10:10.650071 359 sgd_solver.cpp:105] Iteration 5340, lr = 0.00384477 I0407 23:10:15.558791 359 solver.cpp:218] Iteration 5352 (2.44463 iter/s, 4.90871s/12 iters), loss = 0.45152 I0407 23:10:15.558828 359 solver.cpp:237] Train net output #0: loss = 0.45152 (* 1 = 0.45152 loss) I0407 23:10:15.558835 359 sgd_solver.cpp:105] Iteration 5352, lr = 0.00378924 I0407 23:10:18.931526 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:10:20.503518 359 solver.cpp:218] Iteration 5364 (2.42685 iter/s, 4.94467s/12 iters), loss = 0.472535 I0407 23:10:20.503556 359 solver.cpp:237] Train net output #0: loss = 0.472535 (* 1 = 0.472535 loss) I0407 23:10:20.503563 359 sgd_solver.cpp:105] Iteration 5364, lr = 0.00373403 I0407 23:10:25.455345 359 solver.cpp:218] Iteration 5376 (2.42338 iter/s, 4.95177s/12 iters), loss = 0.339578 I0407 23:10:25.455384 359 solver.cpp:237] Train net output #0: loss = 0.339578 (* 1 = 0.339578 loss) I0407 23:10:25.455391 359 sgd_solver.cpp:105] Iteration 5376, lr = 0.00367914 I0407 23:10:30.403879 359 solver.cpp:218] Iteration 5388 (2.42499 iter/s, 4.94848s/12 iters), loss = 0.329 I0407 23:10:30.403918 359 solver.cpp:237] Train net output #0: loss = 0.329 (* 1 = 0.329 loss) I0407 23:10:30.403925 359 sgd_solver.cpp:105] Iteration 5388, lr = 0.00362459 I0407 23:10:35.242074 359 solver.cpp:218] Iteration 5400 (2.48029 iter/s, 4.83814s/12 iters), loss = 0.300492 I0407 23:10:35.242113 359 solver.cpp:237] Train net output #0: loss = 0.300492 (* 1 = 0.300492 loss) I0407 23:10:35.242121 359 sgd_solver.cpp:105] Iteration 5400, lr = 0.0035704 I0407 23:10:37.239481 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_5406.caffemodel I0407 23:10:41.514129 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_5406.solverstate I0407 23:10:43.912750 359 solver.cpp:330] Iteration 5406, Testing net (#0) I0407 23:10:43.912768 359 net.cpp:676] Ignoring source layer train-data I0407 23:10:46.360626 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:10:48.694677 359 solver.cpp:397] Test net output #0: accuracy = 0.415441 I0407 23:10:48.694725 359 solver.cpp:397] Test net output #1: loss = 2.93582 (* 1 = 2.93582 loss) I0407 23:10:50.504892 359 solver.cpp:218] Iteration 5412 (0.786228 iter/s, 15.2627s/12 iters), loss = 0.316812 I0407 23:10:50.505004 359 solver.cpp:237] Train net output #0: loss = 0.316812 (* 1 = 0.316812 loss) I0407 23:10:50.505012 359 sgd_solver.cpp:105] Iteration 5412, lr = 0.00351657 I0407 23:10:55.414255 359 solver.cpp:218] Iteration 5424 (2.44438 iter/s, 4.90923s/12 iters), loss = 0.392943 I0407 23:10:55.414299 359 solver.cpp:237] Train net output #0: loss = 0.392943 (* 1 = 0.392943 loss) I0407 23:10:55.414307 359 sgd_solver.cpp:105] Iteration 5424, lr = 0.00346311 I0407 23:11:00.315161 359 solver.cpp:218] Iteration 5436 (2.44856 iter/s, 4.90084s/12 iters), loss = 0.486893 I0407 23:11:00.315220 359 solver.cpp:237] Train net output #0: loss = 0.486893 (* 1 = 0.486893 loss) I0407 23:11:00.315233 359 sgd_solver.cpp:105] Iteration 5436, lr = 0.00341004 I0407 23:11:05.259418 359 solver.cpp:218] Iteration 5448 (2.4271 iter/s, 4.94418s/12 iters), loss = 0.304622 I0407 23:11:05.259452 359 solver.cpp:237] Train net output #0: loss = 0.304622 (* 1 = 0.304622 loss) I0407 23:11:05.259459 359 sgd_solver.cpp:105] Iteration 5448, lr = 0.00335736 I0407 23:11:10.148643 359 solver.cpp:218] Iteration 5460 (2.4544 iter/s, 4.88917s/12 iters), loss = 0.433443 I0407 23:11:10.148679 359 solver.cpp:237] Train net output #0: loss = 0.433443 (* 1 = 0.433443 loss) I0407 23:11:10.148687 359 sgd_solver.cpp:105] Iteration 5460, lr = 0.00330509 I0407 23:11:10.683794 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:11:14.939079 359 solver.cpp:218] Iteration 5472 (2.50502 iter/s, 4.79038s/12 iters), loss = 0.227015 I0407 23:11:14.939128 359 solver.cpp:237] Train net output #0: loss = 0.227015 (* 1 = 0.227015 loss) I0407 23:11:14.939137 359 sgd_solver.cpp:105] Iteration 5472, lr = 0.00325324 I0407 23:11:19.948105 359 solver.cpp:218] Iteration 5484 (2.39571 iter/s, 5.00895s/12 iters), loss = 0.63865 I0407 23:11:19.948150 359 solver.cpp:237] Train net output #0: loss = 0.63865 (* 1 = 0.63865 loss) I0407 23:11:19.948159 359 sgd_solver.cpp:105] Iteration 5484, lr = 0.00320181 I0407 23:11:24.890550 359 solver.cpp:218] Iteration 5496 (2.42798 iter/s, 4.94237s/12 iters), loss = 0.474697 I0407 23:11:24.890722 359 solver.cpp:237] Train net output #0: loss = 0.474697 (* 1 = 0.474697 loss) I0407 23:11:24.890731 359 sgd_solver.cpp:105] Iteration 5496, lr = 0.00315081 I0407 23:11:29.351128 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_5508.caffemodel I0407 23:11:33.410147 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_5508.solverstate I0407 23:11:36.455431 359 solver.cpp:330] Iteration 5508, Testing net (#0) I0407 23:11:36.455449 359 net.cpp:676] Ignoring source layer train-data I0407 23:11:38.838516 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:11:41.222564 359 solver.cpp:397] Test net output #0: accuracy = 0.427696 I0407 23:11:41.222604 359 solver.cpp:397] Test net output #1: loss = 2.84702 (* 1 = 2.84702 loss) I0407 23:11:41.318938 359 solver.cpp:218] Iteration 5508 (0.730452 iter/s, 16.4282s/12 iters), loss = 0.303231 I0407 23:11:41.318982 359 solver.cpp:237] Train net output #0: loss = 0.303231 (* 1 = 0.303231 loss) I0407 23:11:41.318991 359 sgd_solver.cpp:105] Iteration 5508, lr = 0.00310026 I0407 23:11:45.424505 359 solver.cpp:218] Iteration 5520 (2.9229 iter/s, 4.10551s/12 iters), loss = 0.404447 I0407 23:11:45.424542 359 solver.cpp:237] Train net output #0: loss = 0.404447 (* 1 = 0.404447 loss) I0407 23:11:45.424551 359 sgd_solver.cpp:105] Iteration 5520, lr = 0.00305015 I0407 23:11:47.838690 359 blocking_queue.cpp:49] Waiting for data I0407 23:11:50.383118 359 solver.cpp:218] Iteration 5532 (2.42006 iter/s, 4.95855s/12 iters), loss = 0.238364 I0407 23:11:50.383157 359 solver.cpp:237] Train net output #0: loss = 0.238364 (* 1 = 0.238364 loss) I0407 23:11:50.383165 359 sgd_solver.cpp:105] Iteration 5532, lr = 0.0030005 I0407 23:11:55.311910 359 solver.cpp:218] Iteration 5544 (2.4347 iter/s, 4.92873s/12 iters), loss = 0.375532 I0407 23:11:55.312031 359 solver.cpp:237] Train net output #0: loss = 0.375532 (* 1 = 0.375532 loss) I0407 23:11:55.312041 359 sgd_solver.cpp:105] Iteration 5544, lr = 0.00295132 I0407 23:12:00.365229 359 solver.cpp:218] Iteration 5556 (2.37474 iter/s, 5.05318s/12 iters), loss = 0.372204 I0407 23:12:00.365263 359 solver.cpp:237] Train net output #0: loss = 0.372204 (* 1 = 0.372204 loss) I0407 23:12:00.365270 359 sgd_solver.cpp:105] Iteration 5556, lr = 0.00290261 I0407 23:12:03.050983 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:12:05.316218 359 solver.cpp:218] Iteration 5568 (2.42378 iter/s, 4.95094s/12 iters), loss = 0.242734 I0407 23:12:05.316255 359 solver.cpp:237] Train net output #0: loss = 0.242734 (* 1 = 0.242734 loss) I0407 23:12:05.316263 359 sgd_solver.cpp:105] Iteration 5568, lr = 0.00285438 I0407 23:12:10.244213 359 solver.cpp:218] Iteration 5580 (2.4351 iter/s, 4.92794s/12 iters), loss = 0.20582 I0407 23:12:10.244254 359 solver.cpp:237] Train net output #0: loss = 0.20582 (* 1 = 0.20582 loss) I0407 23:12:10.244263 359 sgd_solver.cpp:105] Iteration 5580, lr = 0.00280663 I0407 23:12:15.203958 359 solver.cpp:218] Iteration 5592 (2.41951 iter/s, 4.95969s/12 iters), loss = 0.297734 I0407 23:12:15.203994 359 solver.cpp:237] Train net output #0: loss = 0.297734 (* 1 = 0.297734 loss) I0407 23:12:15.204002 359 sgd_solver.cpp:105] Iteration 5592, lr = 0.00275937 I0407 23:12:20.126761 359 solver.cpp:218] Iteration 5604 (2.43766 iter/s, 4.92275s/12 iters), loss = 0.439124 I0407 23:12:20.126802 359 solver.cpp:237] Train net output #0: loss = 0.439124 (* 1 = 0.439124 loss) I0407 23:12:20.126811 359 sgd_solver.cpp:105] Iteration 5604, lr = 0.00271261 I0407 23:12:22.170949 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_5610.caffemodel I0407 23:12:26.229061 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_5610.solverstate I0407 23:12:30.063935 359 solver.cpp:330] Iteration 5610, Testing net (#0) I0407 23:12:30.063956 359 net.cpp:676] Ignoring source layer train-data I0407 23:12:32.436645 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:12:34.856134 359 solver.cpp:397] Test net output #0: accuracy = 0.435662 I0407 23:12:34.856169 359 solver.cpp:397] Test net output #1: loss = 2.89615 (* 1 = 2.89615 loss) I0407 23:12:36.632707 359 solver.cpp:218] Iteration 5616 (0.727014 iter/s, 16.5059s/12 iters), loss = 0.331044 I0407 23:12:36.632753 359 solver.cpp:237] Train net output #0: loss = 0.331044 (* 1 = 0.331044 loss) I0407 23:12:36.632761 359 sgd_solver.cpp:105] Iteration 5616, lr = 0.00266635 I0407 23:12:41.692314 359 solver.cpp:218] Iteration 5628 (2.37175 iter/s, 5.05955s/12 iters), loss = 0.389382 I0407 23:12:41.692351 359 solver.cpp:237] Train net output #0: loss = 0.389382 (* 1 = 0.389382 loss) I0407 23:12:41.692358 359 sgd_solver.cpp:105] Iteration 5628, lr = 0.00262059 I0407 23:12:46.717353 359 solver.cpp:218] Iteration 5640 (2.38807 iter/s, 5.02498s/12 iters), loss = 0.374075 I0407 23:12:46.717396 359 solver.cpp:237] Train net output #0: loss = 0.374075 (* 1 = 0.374075 loss) I0407 23:12:46.717402 359 sgd_solver.cpp:105] Iteration 5640, lr = 0.00257534 I0407 23:12:51.776077 359 solver.cpp:218] Iteration 5652 (2.37217 iter/s, 5.05867s/12 iters), loss = 0.245748 I0407 23:12:51.776108 359 solver.cpp:237] Train net output #0: loss = 0.245748 (* 1 = 0.245748 loss) I0407 23:12:51.776116 359 sgd_solver.cpp:105] Iteration 5652, lr = 0.00253061 I0407 23:12:56.510118 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:12:56.678118 359 solver.cpp:218] Iteration 5664 (2.44798 iter/s, 4.902s/12 iters), loss = 0.179748 I0407 23:12:56.678148 359 solver.cpp:237] Train net output #0: loss = 0.179748 (* 1 = 0.179748 loss) I0407 23:12:56.678155 359 sgd_solver.cpp:105] Iteration 5664, lr = 0.00248639 I0407 23:13:01.638792 359 solver.cpp:218] Iteration 5676 (2.41905 iter/s, 4.96063s/12 iters), loss = 0.356952 I0407 23:13:01.638820 359 solver.cpp:237] Train net output #0: loss = 0.356952 (* 1 = 0.356952 loss) I0407 23:13:01.638828 359 sgd_solver.cpp:105] Iteration 5676, lr = 0.0024427 I0407 23:13:06.605708 359 solver.cpp:218] Iteration 5688 (2.41601 iter/s, 4.96687s/12 iters), loss = 0.308913 I0407 23:13:06.605746 359 solver.cpp:237] Train net output #0: loss = 0.308913 (* 1 = 0.308913 loss) I0407 23:13:06.605753 359 sgd_solver.cpp:105] Iteration 5688, lr = 0.00239952 I0407 23:13:11.461937 359 solver.cpp:218] Iteration 5700 (2.47108 iter/s, 4.85617s/12 iters), loss = 0.389904 I0407 23:13:11.461978 359 solver.cpp:237] Train net output #0: loss = 0.389904 (* 1 = 0.389904 loss) I0407 23:13:11.461987 359 sgd_solver.cpp:105] Iteration 5700, lr = 0.00235687 I0407 23:13:15.926808 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_5712.caffemodel I0407 23:13:20.141396 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_5712.solverstate I0407 23:13:23.874502 359 solver.cpp:330] Iteration 5712, Testing net (#0) I0407 23:13:23.874519 359 net.cpp:676] Ignoring source layer train-data I0407 23:13:26.197837 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:13:28.663111 359 solver.cpp:397] Test net output #0: accuracy = 0.436887 I0407 23:13:28.663331 359 solver.cpp:397] Test net output #1: loss = 2.88431 (* 1 = 2.88431 loss) I0407 23:13:28.759913 359 solver.cpp:218] Iteration 5712 (0.693726 iter/s, 17.2979s/12 iters), loss = 0.310276 I0407 23:13:28.759960 359 solver.cpp:237] Train net output #0: loss = 0.310276 (* 1 = 0.310276 loss) I0407 23:13:28.759968 359 sgd_solver.cpp:105] Iteration 5712, lr = 0.00231475 I0407 23:13:32.957828 359 solver.cpp:218] Iteration 5724 (2.85861 iter/s, 4.19785s/12 iters), loss = 0.385243 I0407 23:13:32.957868 359 solver.cpp:237] Train net output #0: loss = 0.385243 (* 1 = 0.385243 loss) I0407 23:13:32.957876 359 sgd_solver.cpp:105] Iteration 5724, lr = 0.00227316 I0407 23:13:37.826519 359 solver.cpp:218] Iteration 5736 (2.46476 iter/s, 4.86863s/12 iters), loss = 0.323 I0407 23:13:37.826558 359 solver.cpp:237] Train net output #0: loss = 0.323 (* 1 = 0.323 loss) I0407 23:13:37.826567 359 sgd_solver.cpp:105] Iteration 5736, lr = 0.0022321 I0407 23:13:42.735463 359 solver.cpp:218] Iteration 5748 (2.44455 iter/s, 4.90888s/12 iters), loss = 0.269936 I0407 23:13:42.735502 359 solver.cpp:237] Train net output #0: loss = 0.269936 (* 1 = 0.269936 loss) I0407 23:13:42.735509 359 sgd_solver.cpp:105] Iteration 5748, lr = 0.00219157 I0407 23:13:47.683583 359 solver.cpp:218] Iteration 5760 (2.42519 iter/s, 4.94806s/12 iters), loss = 0.227433 I0407 23:13:47.683625 359 solver.cpp:237] Train net output #0: loss = 0.227433 (* 1 = 0.227433 loss) I0407 23:13:47.683634 359 sgd_solver.cpp:105] Iteration 5760, lr = 0.00215157 I0407 23:13:49.597967 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:13:52.574467 359 solver.cpp:218] Iteration 5772 (2.45357 iter/s, 4.89083s/12 iters), loss = 0.158779 I0407 23:13:52.574502 359 solver.cpp:237] Train net output #0: loss = 0.158779 (* 1 = 0.158779 loss) I0407 23:13:52.574509 359 sgd_solver.cpp:105] Iteration 5772, lr = 0.0021121 I0407 23:13:57.531615 359 solver.cpp:218] Iteration 5784 (2.42077 iter/s, 4.95709s/12 iters), loss = 0.225406 I0407 23:13:57.531652 359 solver.cpp:237] Train net output #0: loss = 0.225406 (* 1 = 0.225406 loss) I0407 23:13:57.531661 359 sgd_solver.cpp:105] Iteration 5784, lr = 0.00207317 I0407 23:14:02.442128 359 solver.cpp:218] Iteration 5796 (2.44376 iter/s, 4.91046s/12 iters), loss = 0.260227 I0407 23:14:02.442282 359 solver.cpp:237] Train net output #0: loss = 0.260227 (* 1 = 0.260227 loss) I0407 23:14:02.442293 359 sgd_solver.cpp:105] Iteration 5796, lr = 0.00203477 I0407 23:14:07.388200 359 solver.cpp:218] Iteration 5808 (2.42625 iter/s, 4.9459s/12 iters), loss = 0.212148 I0407 23:14:07.388238 359 solver.cpp:237] Train net output #0: loss = 0.212148 (* 1 = 0.212148 loss) I0407 23:14:07.388247 359 sgd_solver.cpp:105] Iteration 5808, lr = 0.0019969 I0407 23:14:09.397231 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_5814.caffemodel I0407 23:14:12.773272 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_5814.solverstate I0407 23:14:16.778278 359 solver.cpp:330] Iteration 5814, Testing net (#0) I0407 23:14:16.778295 359 net.cpp:676] Ignoring source layer train-data I0407 23:14:19.008105 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:14:21.349109 359 solver.cpp:397] Test net output #0: accuracy = 0.443015 I0407 23:14:21.349156 359 solver.cpp:397] Test net output #1: loss = 2.83557 (* 1 = 2.83557 loss) I0407 23:14:23.160087 359 solver.cpp:218] Iteration 5820 (0.760851 iter/s, 15.7718s/12 iters), loss = 0.406367 I0407 23:14:23.160125 359 solver.cpp:237] Train net output #0: loss = 0.406367 (* 1 = 0.406367 loss) I0407 23:14:23.160133 359 sgd_solver.cpp:105] Iteration 5820, lr = 0.00195956 I0407 23:14:28.109808 359 solver.cpp:218] Iteration 5832 (2.42441 iter/s, 4.94966s/12 iters), loss = 0.310057 I0407 23:14:28.109850 359 solver.cpp:237] Train net output #0: loss = 0.310057 (* 1 = 0.310057 loss) I0407 23:14:28.109858 359 sgd_solver.cpp:105] Iteration 5832, lr = 0.00192275 I0407 23:14:33.014621 359 solver.cpp:218] Iteration 5844 (2.44661 iter/s, 4.90475s/12 iters), loss = 0.263185 I0407 23:14:33.014744 359 solver.cpp:237] Train net output #0: loss = 0.263185 (* 1 = 0.263185 loss) I0407 23:14:33.014755 359 sgd_solver.cpp:105] Iteration 5844, lr = 0.00188647 I0407 23:14:37.971206 359 solver.cpp:218] Iteration 5856 (2.42109 iter/s, 4.95645s/12 iters), loss = 0.19987 I0407 23:14:37.971242 359 solver.cpp:237] Train net output #0: loss = 0.19987 (* 1 = 0.19987 loss) I0407 23:14:37.971251 359 sgd_solver.cpp:105] Iteration 5856, lr = 0.00185072 I0407 23:14:42.130307 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:14:42.918393 359 solver.cpp:218] Iteration 5868 (2.42565 iter/s, 4.94713s/12 iters), loss = 0.199371 I0407 23:14:42.918435 359 solver.cpp:237] Train net output #0: loss = 0.199371 (* 1 = 0.199371 loss) I0407 23:14:42.918444 359 sgd_solver.cpp:105] Iteration 5868, lr = 0.0018155 I0407 23:14:47.815058 359 solver.cpp:218] Iteration 5880 (2.45068 iter/s, 4.8966s/12 iters), loss = 0.184291 I0407 23:14:47.815104 359 solver.cpp:237] Train net output #0: loss = 0.184291 (* 1 = 0.184291 loss) I0407 23:14:47.815114 359 sgd_solver.cpp:105] Iteration 5880, lr = 0.0017808 I0407 23:14:52.780167 359 solver.cpp:218] Iteration 5892 (2.4169 iter/s, 4.96504s/12 iters), loss = 0.264258 I0407 23:14:52.780211 359 solver.cpp:237] Train net output #0: loss = 0.264258 (* 1 = 0.264258 loss) I0407 23:14:52.780220 359 sgd_solver.cpp:105] Iteration 5892, lr = 0.00174662 I0407 23:14:57.756559 359 solver.cpp:218] Iteration 5904 (2.41142 iter/s, 4.97632s/12 iters), loss = 0.284302 I0407 23:14:57.756620 359 solver.cpp:237] Train net output #0: loss = 0.284302 (* 1 = 0.284302 loss) I0407 23:14:57.756631 359 sgd_solver.cpp:105] Iteration 5904, lr = 0.00171296 I0407 23:15:02.260032 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_5916.caffemodel I0407 23:15:05.350765 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_5916.solverstate I0407 23:15:08.007269 359 solver.cpp:330] Iteration 5916, Testing net (#0) I0407 23:15:08.007289 359 net.cpp:676] Ignoring source layer train-data I0407 23:15:10.244650 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:15:12.810101 359 solver.cpp:397] Test net output #0: accuracy = 0.455882 I0407 23:15:12.810149 359 solver.cpp:397] Test net output #1: loss = 2.83536 (* 1 = 2.83536 loss) I0407 23:15:12.906216 359 solver.cpp:218] Iteration 5916 (0.792102 iter/s, 15.1496s/12 iters), loss = 0.143237 I0407 23:15:12.906260 359 solver.cpp:237] Train net output #0: loss = 0.143237 (* 1 = 0.143237 loss) I0407 23:15:12.906268 359 sgd_solver.cpp:105] Iteration 5916, lr = 0.00167982 I0407 23:15:16.993103 359 solver.cpp:218] Iteration 5928 (2.93626 iter/s, 4.08682s/12 iters), loss = 0.145722 I0407 23:15:16.993140 359 solver.cpp:237] Train net output #0: loss = 0.145722 (* 1 = 0.145722 loss) I0407 23:15:16.993149 359 sgd_solver.cpp:105] Iteration 5928, lr = 0.00164719 I0407 23:15:21.842190 359 solver.cpp:218] Iteration 5940 (2.47472 iter/s, 4.84903s/12 iters), loss = 0.133358 I0407 23:15:21.842231 359 solver.cpp:237] Train net output #0: loss = 0.133358 (* 1 = 0.133358 loss) I0407 23:15:21.842239 359 sgd_solver.cpp:105] Iteration 5940, lr = 0.00161507 I0407 23:15:26.799206 359 solver.cpp:218] Iteration 5952 (2.42085 iter/s, 4.95693s/12 iters), loss = 0.313479 I0407 23:15:26.799253 359 solver.cpp:237] Train net output #0: loss = 0.313479 (* 1 = 0.313479 loss) I0407 23:15:26.799263 359 sgd_solver.cpp:105] Iteration 5952, lr = 0.00158346 I0407 23:15:31.779289 359 solver.cpp:218] Iteration 5964 (2.40964 iter/s, 4.98001s/12 iters), loss = 0.124743 I0407 23:15:31.779335 359 solver.cpp:237] Train net output #0: loss = 0.124743 (* 1 = 0.124743 loss) I0407 23:15:31.779345 359 sgd_solver.cpp:105] Iteration 5964, lr = 0.00155235 I0407 23:15:33.063752 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:15:36.666793 359 solver.cpp:218] Iteration 5976 (2.45528 iter/s, 4.88743s/12 iters), loss = 0.146947 I0407 23:15:36.666956 359 solver.cpp:237] Train net output #0: loss = 0.146947 (* 1 = 0.146947 loss) I0407 23:15:36.666966 359 sgd_solver.cpp:105] Iteration 5976, lr = 0.00152174 I0407 23:15:41.544632 359 solver.cpp:218] Iteration 5988 (2.4602 iter/s, 4.87766s/12 iters), loss = 0.148596 I0407 23:15:41.544668 359 solver.cpp:237] Train net output #0: loss = 0.148596 (* 1 = 0.148596 loss) I0407 23:15:41.544677 359 sgd_solver.cpp:105] Iteration 5988, lr = 0.00149164 I0407 23:15:46.482853 359 solver.cpp:218] Iteration 6000 (2.43006 iter/s, 4.93815s/12 iters), loss = 0.233873 I0407 23:15:46.482905 359 solver.cpp:237] Train net output #0: loss = 0.233873 (* 1 = 0.233873 loss) I0407 23:15:46.482916 359 sgd_solver.cpp:105] Iteration 6000, lr = 0.00146202 I0407 23:15:51.396286 359 solver.cpp:218] Iteration 6012 (2.44232 iter/s, 4.91336s/12 iters), loss = 0.278592 I0407 23:15:51.396332 359 solver.cpp:237] Train net output #0: loss = 0.278592 (* 1 = 0.278592 loss) I0407 23:15:51.396342 359 sgd_solver.cpp:105] Iteration 6012, lr = 0.00143289 I0407 23:15:53.396051 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_6018.caffemodel I0407 23:15:57.885305 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_6018.solverstate I0407 23:16:00.275216 359 solver.cpp:330] Iteration 6018, Testing net (#0) I0407 23:16:00.275236 359 net.cpp:676] Ignoring source layer train-data I0407 23:16:02.468118 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:16:05.065683 359 solver.cpp:397] Test net output #0: accuracy = 0.447917 I0407 23:16:05.065723 359 solver.cpp:397] Test net output #1: loss = 2.83763 (* 1 = 2.83763 loss) I0407 23:16:06.846371 359 solver.cpp:218] Iteration 6024 (0.776699 iter/s, 15.45s/12 iters), loss = 0.245023 I0407 23:16:06.846490 359 solver.cpp:237] Train net output #0: loss = 0.245023 (* 1 = 0.245023 loss) I0407 23:16:06.846499 359 sgd_solver.cpp:105] Iteration 6024, lr = 0.00140425 I0407 23:16:11.799628 359 solver.cpp:218] Iteration 6036 (2.42272 iter/s, 4.95312s/12 iters), loss = 0.170038 I0407 23:16:11.799665 359 solver.cpp:237] Train net output #0: loss = 0.170038 (* 1 = 0.170038 loss) I0407 23:16:11.799674 359 sgd_solver.cpp:105] Iteration 6036, lr = 0.00137609 I0407 23:16:16.704784 359 solver.cpp:218] Iteration 6048 (2.44643 iter/s, 4.9051s/12 iters), loss = 0.217919 I0407 23:16:16.704824 359 solver.cpp:237] Train net output #0: loss = 0.217919 (* 1 = 0.217919 loss) I0407 23:16:16.704833 359 sgd_solver.cpp:105] Iteration 6048, lr = 0.0013484 I0407 23:16:21.646584 359 solver.cpp:218] Iteration 6060 (2.42829 iter/s, 4.94174s/12 iters), loss = 0.134132 I0407 23:16:21.646625 359 solver.cpp:237] Train net output #0: loss = 0.134132 (* 1 = 0.134132 loss) I0407 23:16:21.646634 359 sgd_solver.cpp:105] Iteration 6060, lr = 0.00132119 I0407 23:16:25.035243 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:16:26.537616 359 solver.cpp:218] Iteration 6072 (2.4535 iter/s, 4.89097s/12 iters), loss = 0.237463 I0407 23:16:26.537659 359 solver.cpp:237] Train net output #0: loss = 0.237463 (* 1 = 0.237463 loss) I0407 23:16:26.537668 359 sgd_solver.cpp:105] Iteration 6072, lr = 0.00129444 I0407 23:16:31.477509 359 solver.cpp:218] Iteration 6084 (2.42924 iter/s, 4.93982s/12 iters), loss = 0.241037 I0407 23:16:31.477552 359 solver.cpp:237] Train net output #0: loss = 0.241037 (* 1 = 0.241037 loss) I0407 23:16:31.477560 359 sgd_solver.cpp:105] Iteration 6084, lr = 0.00126816 I0407 23:16:36.413036 359 solver.cpp:218] Iteration 6096 (2.43138 iter/s, 4.93546s/12 iters), loss = 0.269924 I0407 23:16:36.413079 359 solver.cpp:237] Train net output #0: loss = 0.269924 (* 1 = 0.269924 loss) I0407 23:16:36.413089 359 sgd_solver.cpp:105] Iteration 6096, lr = 0.00124233 I0407 23:16:41.308813 359 solver.cpp:218] Iteration 6108 (2.45113 iter/s, 4.89571s/12 iters), loss = 0.185467 I0407 23:16:41.308938 359 solver.cpp:237] Train net output #0: loss = 0.185467 (* 1 = 0.185467 loss) I0407 23:16:41.308948 359 sgd_solver.cpp:105] Iteration 6108, lr = 0.00121696 I0407 23:16:45.815301 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_6120.caffemodel I0407 23:16:48.882534 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_6120.solverstate I0407 23:16:51.347414 359 solver.cpp:330] Iteration 6120, Testing net (#0) I0407 23:16:51.347432 359 net.cpp:676] Ignoring source layer train-data I0407 23:16:53.483459 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:16:56.131894 359 solver.cpp:397] Test net output #0: accuracy = 0.445466 I0407 23:16:56.131943 359 solver.cpp:397] Test net output #1: loss = 2.83952 (* 1 = 2.83952 loss) I0407 23:16:56.228446 359 solver.cpp:218] Iteration 6120 (0.804318 iter/s, 14.9195s/12 iters), loss = 0.145287 I0407 23:16:56.228493 359 solver.cpp:237] Train net output #0: loss = 0.145287 (* 1 = 0.145287 loss) I0407 23:16:56.228502 359 sgd_solver.cpp:105] Iteration 6120, lr = 0.00119203 I0407 23:17:00.337146 359 solver.cpp:218] Iteration 6132 (2.92068 iter/s, 4.10864s/12 iters), loss = 0.132761 I0407 23:17:00.337185 359 solver.cpp:237] Train net output #0: loss = 0.132761 (* 1 = 0.132761 loss) I0407 23:17:00.337193 359 sgd_solver.cpp:105] Iteration 6132, lr = 0.00116755 I0407 23:17:05.298271 359 solver.cpp:218] Iteration 6144 (2.41883 iter/s, 4.96107s/12 iters), loss = 0.162121 I0407 23:17:05.298308 359 solver.cpp:237] Train net output #0: loss = 0.162121 (* 1 = 0.162121 loss) I0407 23:17:05.298317 359 sgd_solver.cpp:105] Iteration 6144, lr = 0.0011435 I0407 23:17:10.255321 359 solver.cpp:218] Iteration 6156 (2.42082 iter/s, 4.95699s/12 iters), loss = 0.186738 I0407 23:17:10.255363 359 solver.cpp:237] Train net output #0: loss = 0.186738 (* 1 = 0.186738 loss) I0407 23:17:10.255371 359 sgd_solver.cpp:105] Iteration 6156, lr = 0.00111989 I0407 23:17:15.203159 359 solver.cpp:218] Iteration 6168 (2.42533 iter/s, 4.94778s/12 iters), loss = 0.137813 I0407 23:17:15.203291 359 solver.cpp:237] Train net output #0: loss = 0.137812 (* 1 = 0.137812 loss) I0407 23:17:15.203300 359 sgd_solver.cpp:105] Iteration 6168, lr = 0.0010967 I0407 23:17:15.768159 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:17:20.098423 359 solver.cpp:218] Iteration 6180 (2.45142 iter/s, 4.89512s/12 iters), loss = 0.253244 I0407 23:17:20.098469 359 solver.cpp:237] Train net output #0: loss = 0.253244 (* 1 = 0.253244 loss) I0407 23:17:20.098477 359 sgd_solver.cpp:105] Iteration 6180, lr = 0.00107393 I0407 23:17:25.054626 359 solver.cpp:218] Iteration 6192 (2.42124 iter/s, 4.95613s/12 iters), loss = 0.189062 I0407 23:17:25.054667 359 solver.cpp:237] Train net output #0: loss = 0.189062 (* 1 = 0.189062 loss) I0407 23:17:25.054675 359 sgd_solver.cpp:105] Iteration 6192, lr = 0.00105159 I0407 23:17:29.953361 359 solver.cpp:218] Iteration 6204 (2.44964 iter/s, 4.89868s/12 iters), loss = 0.222306 I0407 23:17:29.953399 359 solver.cpp:237] Train net output #0: loss = 0.222306 (* 1 = 0.222306 loss) I0407 23:17:29.953406 359 sgd_solver.cpp:105] Iteration 6204, lr = 0.00102965 I0407 23:17:34.915452 359 solver.cpp:218] Iteration 6216 (2.41836 iter/s, 4.96204s/12 iters), loss = 0.200733 I0407 23:17:34.915488 359 solver.cpp:237] Train net output #0: loss = 0.200733 (* 1 = 0.200733 loss) I0407 23:17:34.915498 359 sgd_solver.cpp:105] Iteration 6216, lr = 0.00100812 I0407 23:17:36.912931 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_6222.caffemodel I0407 23:17:40.000715 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_6222.solverstate I0407 23:17:42.434772 359 solver.cpp:330] Iteration 6222, Testing net (#0) I0407 23:17:42.434790 359 net.cpp:676] Ignoring source layer train-data I0407 23:17:44.584822 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:17:46.061904 359 blocking_queue.cpp:49] Waiting for data I0407 23:17:47.348780 359 solver.cpp:397] Test net output #0: accuracy = 0.441789 I0407 23:17:47.348807 359 solver.cpp:397] Test net output #1: loss = 2.85965 (* 1 = 2.85965 loss) I0407 23:17:49.122150 359 solver.cpp:218] Iteration 6228 (0.844676 iter/s, 14.2066s/12 iters), loss = 0.129902 I0407 23:17:49.122196 359 solver.cpp:237] Train net output #0: loss = 0.129902 (* 1 = 0.129902 loss) I0407 23:17:49.122205 359 sgd_solver.cpp:105] Iteration 6228, lr = 0.00098699 I0407 23:17:54.039830 359 solver.cpp:218] Iteration 6240 (2.44021 iter/s, 4.91762s/12 iters), loss = 0.153368 I0407 23:17:54.039870 359 solver.cpp:237] Train net output #0: loss = 0.153368 (* 1 = 0.153368 loss) I0407 23:17:54.039878 359 sgd_solver.cpp:105] Iteration 6240, lr = 0.000966255 I0407 23:17:58.976161 359 solver.cpp:218] Iteration 6252 (2.43099 iter/s, 4.93627s/12 iters), loss = 0.0896474 I0407 23:17:58.976199 359 solver.cpp:237] Train net output #0: loss = 0.0896474 (* 1 = 0.0896474 loss) I0407 23:17:58.976208 359 sgd_solver.cpp:105] Iteration 6252, lr = 0.000945911 I0407 23:18:03.978098 359 solver.cpp:218] Iteration 6264 (2.3991 iter/s, 5.00188s/12 iters), loss = 0.259995 I0407 23:18:03.978142 359 solver.cpp:237] Train net output #0: loss = 0.259995 (* 1 = 0.259995 loss) I0407 23:18:03.978149 359 sgd_solver.cpp:105] Iteration 6264, lr = 0.00092595 I0407 23:18:06.676151 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:18:08.906256 359 solver.cpp:218] Iteration 6276 (2.43502 iter/s, 4.92809s/12 iters), loss = 0.273486 I0407 23:18:08.906299 359 solver.cpp:237] Train net output #0: loss = 0.273486 (* 1 = 0.273486 loss) I0407 23:18:08.906307 359 sgd_solver.cpp:105] Iteration 6276, lr = 0.000906369 I0407 23:18:13.853348 359 solver.cpp:218] Iteration 6288 (2.4257 iter/s, 4.94703s/12 iters), loss = 0.139224 I0407 23:18:13.853389 359 solver.cpp:237] Train net output #0: loss = 0.139224 (* 1 = 0.139224 loss) I0407 23:18:13.853399 359 sgd_solver.cpp:105] Iteration 6288, lr = 0.000887162 I0407 23:18:18.791798 359 solver.cpp:218] Iteration 6300 (2.42994 iter/s, 4.93839s/12 iters), loss = 0.134716 I0407 23:18:18.791942 359 solver.cpp:237] Train net output #0: loss = 0.134716 (* 1 = 0.134716 loss) I0407 23:18:18.791951 359 sgd_solver.cpp:105] Iteration 6300, lr = 0.000868323 I0407 23:18:23.745375 359 solver.cpp:218] Iteration 6312 (2.42257 iter/s, 4.95341s/12 iters), loss = 0.139055 I0407 23:18:23.745416 359 solver.cpp:237] Train net output #0: loss = 0.139055 (* 1 = 0.139055 loss) I0407 23:18:23.745424 359 sgd_solver.cpp:105] Iteration 6312, lr = 0.000849846 I0407 23:18:28.225522 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_6324.caffemodel I0407 23:18:31.307723 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_6324.solverstate I0407 23:18:33.675303 359 solver.cpp:330] Iteration 6324, Testing net (#0) I0407 23:18:33.675323 359 net.cpp:676] Ignoring source layer train-data I0407 23:18:35.731346 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:18:38.459969 359 solver.cpp:397] Test net output #0: accuracy = 0.457721 I0407 23:18:38.460003 359 solver.cpp:397] Test net output #1: loss = 2.81101 (* 1 = 2.81101 loss) I0407 23:18:38.556648 359 solver.cpp:218] Iteration 6324 (0.810198 iter/s, 14.8112s/12 iters), loss = 0.153668 I0407 23:18:38.556690 359 solver.cpp:237] Train net output #0: loss = 0.153668 (* 1 = 0.153668 loss) I0407 23:18:38.556699 359 sgd_solver.cpp:105] Iteration 6324, lr = 0.000831727 I0407 23:18:42.664073 359 solver.cpp:218] Iteration 6336 (2.92159 iter/s, 4.10736s/12 iters), loss = 0.163364 I0407 23:18:42.664115 359 solver.cpp:237] Train net output #0: loss = 0.163364 (* 1 = 0.163364 loss) I0407 23:18:42.664122 359 sgd_solver.cpp:105] Iteration 6336, lr = 0.00081396 I0407 23:18:47.559902 359 solver.cpp:218] Iteration 6348 (2.4511 iter/s, 4.89576s/12 iters), loss = 0.231072 I0407 23:18:47.559952 359 solver.cpp:237] Train net output #0: loss = 0.231071 (* 1 = 0.231071 loss) I0407 23:18:47.559959 359 sgd_solver.cpp:105] Iteration 6348, lr = 0.000796539 I0407 23:18:52.520370 359 solver.cpp:218] Iteration 6360 (2.41916 iter/s, 4.9604s/12 iters), loss = 0.120645 I0407 23:18:52.520499 359 solver.cpp:237] Train net output #0: loss = 0.120645 (* 1 = 0.120645 loss) I0407 23:18:52.520507 359 sgd_solver.cpp:105] Iteration 6360, lr = 0.000779459 I0407 23:18:57.274106 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:18:57.413798 359 solver.cpp:218] Iteration 6372 (2.45234 iter/s, 4.89328s/12 iters), loss = 0.153222 I0407 23:18:57.413831 359 solver.cpp:237] Train net output #0: loss = 0.153222 (* 1 = 0.153222 loss) I0407 23:18:57.413839 359 sgd_solver.cpp:105] Iteration 6372, lr = 0.000762716 I0407 23:19:02.350708 359 solver.cpp:218] Iteration 6384 (2.4307 iter/s, 4.93686s/12 iters), loss = 0.0881359 I0407 23:19:02.350745 359 solver.cpp:237] Train net output #0: loss = 0.0881359 (* 1 = 0.0881359 loss) I0407 23:19:02.350752 359 sgd_solver.cpp:105] Iteration 6384, lr = 0.000746303 I0407 23:19:07.270803 359 solver.cpp:218] Iteration 6396 (2.439 iter/s, 4.92004s/12 iters), loss = 0.135171 I0407 23:19:07.270850 359 solver.cpp:237] Train net output #0: loss = 0.135171 (* 1 = 0.135171 loss) I0407 23:19:07.270859 359 sgd_solver.cpp:105] Iteration 6396, lr = 0.000730215 I0407 23:19:12.212148 359 solver.cpp:218] Iteration 6408 (2.42852 iter/s, 4.94128s/12 iters), loss = 0.274553 I0407 23:19:12.212185 359 solver.cpp:237] Train net output #0: loss = 0.274553 (* 1 = 0.274553 loss) I0407 23:19:12.212193 359 sgd_solver.cpp:105] Iteration 6408, lr = 0.000714447 I0407 23:19:17.134399 359 solver.cpp:218] Iteration 6420 (2.43794 iter/s, 4.9222s/12 iters), loss = 0.138166 I0407 23:19:17.134439 359 solver.cpp:237] Train net output #0: loss = 0.138166 (* 1 = 0.138166 loss) I0407 23:19:17.134447 359 sgd_solver.cpp:105] Iteration 6420, lr = 0.000698994 I0407 23:19:19.124998 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_6426.caffemodel I0407 23:19:22.204881 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_6426.solverstate I0407 23:19:24.673915 359 solver.cpp:330] Iteration 6426, Testing net (#0) I0407 23:19:24.674012 359 net.cpp:676] Ignoring source layer train-data I0407 23:19:26.581594 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:19:29.139668 359 solver.cpp:397] Test net output #0: accuracy = 0.450368 I0407 23:19:29.139700 359 solver.cpp:397] Test net output #1: loss = 2.84339 (* 1 = 2.84339 loss) I0407 23:19:30.940819 359 solver.cpp:218] Iteration 6432 (0.869165 iter/s, 13.8063s/12 iters), loss = 0.150292 I0407 23:19:30.940857 359 solver.cpp:237] Train net output #0: loss = 0.150292 (* 1 = 0.150292 loss) I0407 23:19:30.940865 359 sgd_solver.cpp:105] Iteration 6432, lr = 0.000683851 I0407 23:19:35.942068 359 solver.cpp:218] Iteration 6444 (2.39943 iter/s, 5.00119s/12 iters), loss = 0.246275 I0407 23:19:35.942111 359 solver.cpp:237] Train net output #0: loss = 0.246274 (* 1 = 0.246274 loss) I0407 23:19:35.942119 359 sgd_solver.cpp:105] Iteration 6444, lr = 0.000669012 I0407 23:19:40.886096 359 solver.cpp:218] Iteration 6456 (2.4272 iter/s, 4.94396s/12 iters), loss = 0.215359 I0407 23:19:40.886143 359 solver.cpp:237] Train net output #0: loss = 0.215359 (* 1 = 0.215359 loss) I0407 23:19:40.886152 359 sgd_solver.cpp:105] Iteration 6456, lr = 0.000654472 I0407 23:19:45.861039 359 solver.cpp:218] Iteration 6468 (2.41212 iter/s, 4.97487s/12 iters), loss = 0.126235 I0407 23:19:45.861086 359 solver.cpp:237] Train net output #0: loss = 0.126235 (* 1 = 0.126235 loss) I0407 23:19:45.861094 359 sgd_solver.cpp:105] Iteration 6468, lr = 0.000640227 I0407 23:19:47.810267 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:19:50.822371 359 solver.cpp:218] Iteration 6480 (2.41874 iter/s, 4.96127s/12 iters), loss = 0.105952 I0407 23:19:50.822408 359 solver.cpp:237] Train net output #0: loss = 0.105952 (* 1 = 0.105952 loss) I0407 23:19:50.822417 359 sgd_solver.cpp:105] Iteration 6480, lr = 0.000626271 I0407 23:19:55.739878 359 solver.cpp:218] Iteration 6492 (2.44029 iter/s, 4.91745s/12 iters), loss = 0.238412 I0407 23:19:55.740053 359 solver.cpp:237] Train net output #0: loss = 0.238412 (* 1 = 0.238412 loss) I0407 23:19:55.740069 359 sgd_solver.cpp:105] Iteration 6492, lr = 0.0006126 I0407 23:20:00.695417 359 solver.cpp:218] Iteration 6504 (2.42162 iter/s, 4.95535s/12 iters), loss = 0.251793 I0407 23:20:00.695458 359 solver.cpp:237] Train net output #0: loss = 0.251793 (* 1 = 0.251793 loss) I0407 23:20:00.695467 359 sgd_solver.cpp:105] Iteration 6504, lr = 0.000599207 I0407 23:20:05.594112 359 solver.cpp:218] Iteration 6516 (2.44966 iter/s, 4.89864s/12 iters), loss = 0.196607 I0407 23:20:05.594148 359 solver.cpp:237] Train net output #0: loss = 0.196607 (* 1 = 0.196607 loss) I0407 23:20:05.594156 359 sgd_solver.cpp:105] Iteration 6516, lr = 0.00058609 I0407 23:20:10.109151 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_6528.caffemodel I0407 23:20:13.216063 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_6528.solverstate I0407 23:20:15.603492 359 solver.cpp:330] Iteration 6528, Testing net (#0) I0407 23:20:15.603508 359 net.cpp:676] Ignoring source layer train-data I0407 23:20:17.508338 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:20:20.119869 359 solver.cpp:397] Test net output #0: accuracy = 0.457721 I0407 23:20:20.119915 359 solver.cpp:397] Test net output #1: loss = 2.87348 (* 1 = 2.87348 loss) I0407 23:20:20.216583 359 solver.cpp:218] Iteration 6528 (0.820659 iter/s, 14.6224s/12 iters), loss = 0.0990301 I0407 23:20:20.216629 359 solver.cpp:237] Train net output #0: loss = 0.0990301 (* 1 = 0.0990301 loss) I0407 23:20:20.216637 359 sgd_solver.cpp:105] Iteration 6528, lr = 0.000573242 I0407 23:20:24.354866 359 solver.cpp:218] Iteration 6540 (2.8998 iter/s, 4.13822s/12 iters), loss = 0.175362 I0407 23:20:24.354910 359 solver.cpp:237] Train net output #0: loss = 0.175362 (* 1 = 0.175362 loss) I0407 23:20:24.354919 359 sgd_solver.cpp:105] Iteration 6540, lr = 0.000560659 I0407 23:20:29.286823 359 solver.cpp:218] Iteration 6552 (2.43315 iter/s, 4.93189s/12 iters), loss = 0.13445 I0407 23:20:29.286944 359 solver.cpp:237] Train net output #0: loss = 0.13445 (* 1 = 0.13445 loss) I0407 23:20:29.286954 359 sgd_solver.cpp:105] Iteration 6552, lr = 0.000548335 I0407 23:20:34.175601 359 solver.cpp:218] Iteration 6564 (2.45467 iter/s, 4.88864s/12 iters), loss = 0.184569 I0407 23:20:34.175639 359 solver.cpp:237] Train net output #0: loss = 0.184569 (* 1 = 0.184569 loss) I0407 23:20:34.175648 359 sgd_solver.cpp:105] Iteration 6564, lr = 0.000536268 I0407 23:20:38.373523 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:20:39.128399 359 solver.cpp:218] Iteration 6576 (2.4229 iter/s, 4.95274s/12 iters), loss = 0.22993 I0407 23:20:39.128435 359 solver.cpp:237] Train net output #0: loss = 0.22993 (* 1 = 0.22993 loss) I0407 23:20:39.128443 359 sgd_solver.cpp:105] Iteration 6576, lr = 0.000524451 I0407 23:20:44.056602 359 solver.cpp:218] Iteration 6588 (2.43499 iter/s, 4.92815s/12 iters), loss = 0.13458 I0407 23:20:44.056636 359 solver.cpp:237] Train net output #0: loss = 0.13458 (* 1 = 0.13458 loss) I0407 23:20:44.056643 359 sgd_solver.cpp:105] Iteration 6588, lr = 0.000512881 I0407 23:20:49.049531 359 solver.cpp:218] Iteration 6600 (2.40343 iter/s, 4.99287s/12 iters), loss = 0.151401 I0407 23:20:49.049571 359 solver.cpp:237] Train net output #0: loss = 0.151401 (* 1 = 0.151401 loss) I0407 23:20:49.049580 359 sgd_solver.cpp:105] Iteration 6600, lr = 0.000501552 I0407 23:20:53.991566 359 solver.cpp:218] Iteration 6612 (2.42818 iter/s, 4.94197s/12 iters), loss = 0.177412 I0407 23:20:53.991613 359 solver.cpp:237] Train net output #0: loss = 0.177412 (* 1 = 0.177412 loss) I0407 23:20:53.991621 359 sgd_solver.cpp:105] Iteration 6612, lr = 0.00049046 I0407 23:20:58.907235 359 solver.cpp:218] Iteration 6624 (2.44121 iter/s, 4.9156s/12 iters), loss = 0.160151 I0407 23:20:58.907280 359 solver.cpp:237] Train net output #0: loss = 0.160151 (* 1 = 0.160151 loss) I0407 23:20:58.907289 359 sgd_solver.cpp:105] Iteration 6624, lr = 0.000479602 I0407 23:21:00.898057 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_6630.caffemodel I0407 23:21:03.982336 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_6630.solverstate I0407 23:21:06.351004 359 solver.cpp:330] Iteration 6630, Testing net (#0) I0407 23:21:06.351023 359 net.cpp:676] Ignoring source layer train-data I0407 23:21:08.286592 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:21:11.145256 359 solver.cpp:397] Test net output #0: accuracy = 0.454044 I0407 23:21:11.145303 359 solver.cpp:397] Test net output #1: loss = 2.86725 (* 1 = 2.86725 loss) I0407 23:21:12.957787 359 solver.cpp:218] Iteration 6636 (0.854064 iter/s, 14.0505s/12 iters), loss = 0.115546 I0407 23:21:12.957829 359 solver.cpp:237] Train net output #0: loss = 0.115546 (* 1 = 0.115546 loss) I0407 23:21:12.957837 359 sgd_solver.cpp:105] Iteration 6636, lr = 0.000468972 I0407 23:21:17.907053 359 solver.cpp:218] Iteration 6648 (2.42463 iter/s, 4.94921s/12 iters), loss = 0.131054 I0407 23:21:17.907088 359 solver.cpp:237] Train net output #0: loss = 0.131054 (* 1 = 0.131054 loss) I0407 23:21:17.907095 359 sgd_solver.cpp:105] Iteration 6648, lr = 0.000458566 I0407 23:21:22.831107 359 solver.cpp:218] Iteration 6660 (2.43704 iter/s, 4.924s/12 iters), loss = 0.0919114 I0407 23:21:22.831142 359 solver.cpp:237] Train net output #0: loss = 0.0919114 (* 1 = 0.0919114 loss) I0407 23:21:22.831151 359 sgd_solver.cpp:105] Iteration 6660, lr = 0.00044838 I0407 23:21:27.782884 359 solver.cpp:218] Iteration 6672 (2.4234 iter/s, 4.95172s/12 iters), loss = 0.109363 I0407 23:21:27.782924 359 solver.cpp:237] Train net output #0: loss = 0.109363 (* 1 = 0.109363 loss) I0407 23:21:27.782932 359 sgd_solver.cpp:105] Iteration 6672, lr = 0.000438411 I0407 23:21:29.098017 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:21:32.668825 359 solver.cpp:218] Iteration 6684 (2.45606 iter/s, 4.88588s/12 iters), loss = 0.0794411 I0407 23:21:32.668939 359 solver.cpp:237] Train net output #0: loss = 0.079441 (* 1 = 0.079441 loss) I0407 23:21:32.668947 359 sgd_solver.cpp:105] Iteration 6684, lr = 0.000428653 I0407 23:21:37.626327 359 solver.cpp:218] Iteration 6696 (2.42064 iter/s, 4.95737s/12 iters), loss = 0.192344 I0407 23:21:37.626370 359 solver.cpp:237] Train net output #0: loss = 0.192344 (* 1 = 0.192344 loss) I0407 23:21:37.626377 359 sgd_solver.cpp:105] Iteration 6696, lr = 0.000419102 I0407 23:21:42.536317 359 solver.cpp:218] Iteration 6708 (2.44403 iter/s, 4.90992s/12 iters), loss = 0.189281 I0407 23:21:42.536360 359 solver.cpp:237] Train net output #0: loss = 0.189281 (* 1 = 0.189281 loss) I0407 23:21:42.536370 359 sgd_solver.cpp:105] Iteration 6708, lr = 0.000409755 I0407 23:21:47.443796 359 solver.cpp:218] Iteration 6720 (2.44528 iter/s, 4.90741s/12 iters), loss = 0.0837439 I0407 23:21:47.443840 359 solver.cpp:237] Train net output #0: loss = 0.0837439 (* 1 = 0.0837439 loss) I0407 23:21:47.443847 359 sgd_solver.cpp:105] Iteration 6720, lr = 0.000400608 I0407 23:21:51.929875 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_6732.caffemodel I0407 23:21:55.912889 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_6732.solverstate I0407 23:21:58.291445 359 solver.cpp:330] Iteration 6732, Testing net (#0) I0407 23:21:58.291465 359 net.cpp:676] Ignoring source layer train-data I0407 23:22:00.027452 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:22:02.689129 359 solver.cpp:397] Test net output #0: accuracy = 0.456495 I0407 23:22:02.689332 359 solver.cpp:397] Test net output #1: loss = 2.84666 (* 1 = 2.84666 loss) I0407 23:22:02.785645 359 solver.cpp:218] Iteration 6732 (0.782178 iter/s, 15.3418s/12 iters), loss = 0.0341394 I0407 23:22:02.785689 359 solver.cpp:237] Train net output #0: loss = 0.0341394 (* 1 = 0.0341394 loss) I0407 23:22:02.785698 359 sgd_solver.cpp:105] Iteration 6732, lr = 0.000391657 I0407 23:22:06.918531 359 solver.cpp:218] Iteration 6744 (2.90358 iter/s, 4.13283s/12 iters), loss = 0.172641 I0407 23:22:06.918565 359 solver.cpp:237] Train net output #0: loss = 0.172641 (* 1 = 0.172641 loss) I0407 23:22:06.918573 359 sgd_solver.cpp:105] Iteration 6744, lr = 0.000382898 I0407 23:22:11.868175 359 solver.cpp:218] Iteration 6756 (2.42444 iter/s, 4.94959s/12 iters), loss = 0.241404 I0407 23:22:11.868216 359 solver.cpp:237] Train net output #0: loss = 0.241404 (* 1 = 0.241404 loss) I0407 23:22:11.868223 359 sgd_solver.cpp:105] Iteration 6756, lr = 0.000374327 I0407 23:22:16.728605 359 solver.cpp:218] Iteration 6768 (2.46895 iter/s, 4.86037s/12 iters), loss = 0.279273 I0407 23:22:16.728642 359 solver.cpp:237] Train net output #0: loss = 0.279273 (* 1 = 0.279273 loss) I0407 23:22:16.728650 359 sgd_solver.cpp:105] Iteration 6768, lr = 0.000365941 I0407 23:22:20.148819 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:22:21.670934 359 solver.cpp:218] Iteration 6780 (2.42803 iter/s, 4.94227s/12 iters), loss = 0.128633 I0407 23:22:21.670977 359 solver.cpp:237] Train net output #0: loss = 0.128633 (* 1 = 0.128633 loss) I0407 23:22:21.670986 359 sgd_solver.cpp:105] Iteration 6780, lr = 0.000357735 I0407 23:22:26.599380 359 solver.cpp:218] Iteration 6792 (2.43487 iter/s, 4.92839s/12 iters), loss = 0.183782 I0407 23:22:26.599417 359 solver.cpp:237] Train net output #0: loss = 0.183782 (* 1 = 0.183782 loss) I0407 23:22:26.599426 359 sgd_solver.cpp:105] Iteration 6792, lr = 0.000349707 I0407 23:22:31.549271 359 solver.cpp:218] Iteration 6804 (2.42432 iter/s, 4.94984s/12 iters), loss = 0.152019 I0407 23:22:31.549310 359 solver.cpp:237] Train net output #0: loss = 0.152019 (* 1 = 0.152019 loss) I0407 23:22:31.549319 359 sgd_solver.cpp:105] Iteration 6804, lr = 0.000341853 I0407 23:22:36.446985 359 solver.cpp:218] Iteration 6816 (2.45015 iter/s, 4.89766s/12 iters), loss = 0.167886 I0407 23:22:36.447109 359 solver.cpp:237] Train net output #0: loss = 0.167886 (* 1 = 0.167886 loss) I0407 23:22:36.447118 359 sgd_solver.cpp:105] Iteration 6816, lr = 0.000334169 I0407 23:22:41.410583 359 solver.cpp:218] Iteration 6828 (2.41767 iter/s, 4.96346s/12 iters), loss = 0.105224 I0407 23:22:41.410617 359 solver.cpp:237] Train net output #0: loss = 0.105224 (* 1 = 0.105224 loss) I0407 23:22:41.410625 359 sgd_solver.cpp:105] Iteration 6828, lr = 0.000326652 I0407 23:22:43.398420 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_6834.caffemodel I0407 23:22:47.631776 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_6834.solverstate I0407 23:22:51.158975 359 solver.cpp:330] Iteration 6834, Testing net (#0) I0407 23:22:51.158994 359 net.cpp:676] Ignoring source layer train-data I0407 23:22:52.835144 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:22:55.525240 359 solver.cpp:397] Test net output #0: accuracy = 0.456495 I0407 23:22:55.525287 359 solver.cpp:397] Test net output #1: loss = 2.84198 (* 1 = 2.84198 loss) I0407 23:22:57.320017 359 solver.cpp:218] Iteration 6840 (0.754273 iter/s, 15.9094s/12 iters), loss = 0.213791 I0407 23:22:57.320061 359 solver.cpp:237] Train net output #0: loss = 0.213791 (* 1 = 0.213791 loss) I0407 23:22:57.320070 359 sgd_solver.cpp:105] Iteration 6840, lr = 0.000319298 I0407 23:23:02.217681 359 solver.cpp:218] Iteration 6852 (2.45018 iter/s, 4.8976s/12 iters), loss = 0.144369 I0407 23:23:02.217728 359 solver.cpp:237] Train net output #0: loss = 0.144369 (* 1 = 0.144369 loss) I0407 23:23:02.217736 359 sgd_solver.cpp:105] Iteration 6852, lr = 0.000312105 I0407 23:23:07.196313 359 solver.cpp:218] Iteration 6864 (2.41033 iter/s, 4.97857s/12 iters), loss = 0.196857 I0407 23:23:07.197065 359 solver.cpp:237] Train net output #0: loss = 0.196857 (* 1 = 0.196857 loss) I0407 23:23:07.197077 359 sgd_solver.cpp:105] Iteration 6864, lr = 0.000305068 I0407 23:23:12.125314 359 solver.cpp:218] Iteration 6876 (2.43495 iter/s, 4.92823s/12 iters), loss = 0.215531 I0407 23:23:12.125355 359 solver.cpp:237] Train net output #0: loss = 0.215531 (* 1 = 0.215531 loss) I0407 23:23:12.125361 359 sgd_solver.cpp:105] Iteration 6876, lr = 0.000298185 I0407 23:23:12.732437 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:23:17.067513 359 solver.cpp:218] Iteration 6888 (2.4281 iter/s, 4.94213s/12 iters), loss = 0.170668 I0407 23:23:17.067559 359 solver.cpp:237] Train net output #0: loss = 0.170668 (* 1 = 0.170668 loss) I0407 23:23:17.067569 359 sgd_solver.cpp:105] Iteration 6888, lr = 0.000291453 I0407 23:23:22.020543 359 solver.cpp:218] Iteration 6900 (2.42279 iter/s, 4.95296s/12 iters), loss = 0.184094 I0407 23:23:22.020581 359 solver.cpp:237] Train net output #0: loss = 0.184094 (* 1 = 0.184094 loss) I0407 23:23:22.020588 359 sgd_solver.cpp:105] Iteration 6900, lr = 0.000284869 I0407 23:23:26.911700 359 solver.cpp:218] Iteration 6912 (2.45344 iter/s, 4.8911s/12 iters), loss = 0.305297 I0407 23:23:26.911737 359 solver.cpp:237] Train net output #0: loss = 0.305297 (* 1 = 0.305297 loss) I0407 23:23:26.911747 359 sgd_solver.cpp:105] Iteration 6912, lr = 0.000278428 I0407 23:23:31.828889 359 solver.cpp:218] Iteration 6924 (2.44045 iter/s, 4.91713s/12 iters), loss = 0.262044 I0407 23:23:31.828924 359 solver.cpp:237] Train net output #0: loss = 0.262044 (* 1 = 0.262044 loss) I0407 23:23:31.828933 359 sgd_solver.cpp:105] Iteration 6924, lr = 0.00027213 I0407 23:23:36.315490 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_6936.caffemodel I0407 23:23:40.034473 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_6936.solverstate I0407 23:23:43.089866 359 solver.cpp:330] Iteration 6936, Testing net (#0) I0407 23:23:43.089884 359 net.cpp:676] Ignoring source layer train-data I0407 23:23:43.654500 359 blocking_queue.cpp:49] Waiting for data I0407 23:23:44.722661 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:23:47.463395 359 solver.cpp:397] Test net output #0: accuracy = 0.458333 I0407 23:23:47.463443 359 solver.cpp:397] Test net output #1: loss = 2.82497 (* 1 = 2.82497 loss) I0407 23:23:47.560518 359 solver.cpp:218] Iteration 6936 (0.762799 iter/s, 15.7315s/12 iters), loss = 0.118046 I0407 23:23:47.560585 359 solver.cpp:237] Train net output #0: loss = 0.118046 (* 1 = 0.118046 loss) I0407 23:23:47.560600 359 sgd_solver.cpp:105] Iteration 6936, lr = 0.00026597 I0407 23:23:51.671131 359 solver.cpp:218] Iteration 6948 (2.91933 iter/s, 4.11054s/12 iters), loss = 0.097086 I0407 23:23:51.671175 359 solver.cpp:237] Train net output #0: loss = 0.097086 (* 1 = 0.097086 loss) I0407 23:23:51.671185 359 sgd_solver.cpp:105] Iteration 6948, lr = 0.000259946 I0407 23:23:56.623329 359 solver.cpp:218] Iteration 6960 (2.4232 iter/s, 4.95213s/12 iters), loss = 0.112395 I0407 23:23:56.623373 359 solver.cpp:237] Train net output #0: loss = 0.112395 (* 1 = 0.112395 loss) I0407 23:23:56.623380 359 sgd_solver.cpp:105] Iteration 6960, lr = 0.000254054 I0407 23:24:01.587177 359 solver.cpp:218] Iteration 6972 (2.41751 iter/s, 4.96379s/12 iters), loss = 0.14027 I0407 23:24:01.587222 359 solver.cpp:237] Train net output #0: loss = 0.14027 (* 1 = 0.14027 loss) I0407 23:24:01.587231 359 sgd_solver.cpp:105] Iteration 6972, lr = 0.000248293 I0407 23:24:04.314172 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:24:06.516044 359 solver.cpp:218] Iteration 6984 (2.43467 iter/s, 4.92881s/12 iters), loss = 0.167093 I0407 23:24:06.516083 359 solver.cpp:237] Train net output #0: loss = 0.167093 (* 1 = 0.167093 loss) I0407 23:24:06.516090 359 sgd_solver.cpp:105] Iteration 6984, lr = 0.000242659 I0407 23:24:11.514892 359 solver.cpp:218] Iteration 6996 (2.40058 iter/s, 4.99878s/12 iters), loss = 0.201238 I0407 23:24:11.515044 359 solver.cpp:237] Train net output #0: loss = 0.201238 (* 1 = 0.201238 loss) I0407 23:24:11.515054 359 sgd_solver.cpp:105] Iteration 6996, lr = 0.00023715 I0407 23:24:16.443599 359 solver.cpp:218] Iteration 7008 (2.4348 iter/s, 4.92853s/12 iters), loss = 0.0924055 I0407 23:24:16.443646 359 solver.cpp:237] Train net output #0: loss = 0.0924055 (* 1 = 0.0924055 loss) I0407 23:24:16.443655 359 sgd_solver.cpp:105] Iteration 7008, lr = 0.000231763 I0407 23:24:21.383993 359 solver.cpp:218] Iteration 7020 (2.42899 iter/s, 4.94032s/12 iters), loss = 0.153556 I0407 23:24:21.384040 359 solver.cpp:237] Train net output #0: loss = 0.153556 (* 1 = 0.153556 loss) I0407 23:24:21.384048 359 sgd_solver.cpp:105] Iteration 7020, lr = 0.000226495 I0407 23:24:26.370350 359 solver.cpp:218] Iteration 7032 (2.4066 iter/s, 4.98628s/12 iters), loss = 0.136955 I0407 23:24:26.370398 359 solver.cpp:237] Train net output #0: loss = 0.136955 (* 1 = 0.136955 loss) I0407 23:24:26.370407 359 sgd_solver.cpp:105] Iteration 7032, lr = 0.000221345 I0407 23:24:28.369777 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_7038.caffemodel I0407 23:24:31.476897 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_7038.solverstate I0407 23:24:33.903177 359 solver.cpp:330] Iteration 7038, Testing net (#0) I0407 23:24:33.903194 359 net.cpp:676] Ignoring source layer train-data I0407 23:24:35.673597 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:24:38.711179 359 solver.cpp:397] Test net output #0: accuracy = 0.455882 I0407 23:24:38.711243 359 solver.cpp:397] Test net output #1: loss = 2.83419 (* 1 = 2.83419 loss) I0407 23:24:40.521961 359 solver.cpp:218] Iteration 7044 (0.847965 iter/s, 14.1515s/12 iters), loss = 0.114693 I0407 23:24:40.522001 359 solver.cpp:237] Train net output #0: loss = 0.114693 (* 1 = 0.114693 loss) I0407 23:24:40.522007 359 sgd_solver.cpp:105] Iteration 7044, lr = 0.000216309 I0407 23:24:45.452225 359 solver.cpp:218] Iteration 7056 (2.43398 iter/s, 4.9302s/12 iters), loss = 0.170558 I0407 23:24:45.452363 359 solver.cpp:237] Train net output #0: loss = 0.170558 (* 1 = 0.170558 loss) I0407 23:24:45.452373 359 sgd_solver.cpp:105] Iteration 7056, lr = 0.000211385 I0407 23:24:50.392465 359 solver.cpp:218] Iteration 7068 (2.42911 iter/s, 4.94009s/12 iters), loss = 0.188994 I0407 23:24:50.392504 359 solver.cpp:237] Train net output #0: loss = 0.188994 (* 1 = 0.188994 loss) I0407 23:24:50.392513 359 sgd_solver.cpp:105] Iteration 7068, lr = 0.000206571 I0407 23:24:55.213446 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:24:55.323170 359 solver.cpp:218] Iteration 7080 (2.43376 iter/s, 4.93065s/12 iters), loss = 0.216889 I0407 23:24:55.323213 359 solver.cpp:237] Train net output #0: loss = 0.216889 (* 1 = 0.216889 loss) I0407 23:24:55.323221 359 sgd_solver.cpp:105] Iteration 7080, lr = 0.000201864 I0407 23:25:00.184571 359 solver.cpp:218] Iteration 7092 (2.46846 iter/s, 4.86134s/12 iters), loss = 0.159934 I0407 23:25:00.184615 359 solver.cpp:237] Train net output #0: loss = 0.159934 (* 1 = 0.159934 loss) I0407 23:25:00.184623 359 sgd_solver.cpp:105] Iteration 7092, lr = 0.000197262 I0407 23:25:05.111347 359 solver.cpp:218] Iteration 7104 (2.4357 iter/s, 4.92671s/12 iters), loss = 0.0967507 I0407 23:25:05.111384 359 solver.cpp:237] Train net output #0: loss = 0.0967507 (* 1 = 0.0967507 loss) I0407 23:25:05.111393 359 sgd_solver.cpp:105] Iteration 7104, lr = 0.000192763 I0407 23:25:10.049810 359 solver.cpp:218] Iteration 7116 (2.42993 iter/s, 4.93841s/12 iters), loss = 0.212849 I0407 23:25:10.049844 359 solver.cpp:237] Train net output #0: loss = 0.212849 (* 1 = 0.212849 loss) I0407 23:25:10.049851 359 sgd_solver.cpp:105] Iteration 7116, lr = 0.000188365 I0407 23:25:14.970294 359 solver.cpp:218] Iteration 7128 (2.43881 iter/s, 4.92043s/12 iters), loss = 0.216186 I0407 23:25:14.970332 359 solver.cpp:237] Train net output #0: loss = 0.216186 (* 1 = 0.216186 loss) I0407 23:25:14.970340 359 sgd_solver.cpp:105] Iteration 7128, lr = 0.000184065 I0407 23:25:19.462628 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_7140.caffemodel I0407 23:25:22.590390 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_7140.solverstate I0407 23:25:24.958521 359 solver.cpp:330] Iteration 7140, Testing net (#0) I0407 23:25:24.958539 359 net.cpp:676] Ignoring source layer train-data I0407 23:25:26.613893 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:25:29.475989 359 solver.cpp:397] Test net output #0: accuracy = 0.45527 I0407 23:25:29.476022 359 solver.cpp:397] Test net output #1: loss = 2.85037 (* 1 = 2.85037 loss) I0407 23:25:29.572391 359 solver.cpp:218] Iteration 7140 (0.821804 iter/s, 14.602s/12 iters), loss = 0.174071 I0407 23:25:29.572436 359 solver.cpp:237] Train net output #0: loss = 0.174071 (* 1 = 0.174071 loss) I0407 23:25:29.572444 359 sgd_solver.cpp:105] Iteration 7140, lr = 0.000179862 I0407 23:25:33.689715 359 solver.cpp:218] Iteration 7152 (2.91456 iter/s, 4.11726s/12 iters), loss = 0.145695 I0407 23:25:33.689760 359 solver.cpp:237] Train net output #0: loss = 0.145695 (* 1 = 0.145695 loss) I0407 23:25:33.689769 359 sgd_solver.cpp:105] Iteration 7152, lr = 0.000175753 I0407 23:25:38.629235 359 solver.cpp:218] Iteration 7164 (2.42942 iter/s, 4.93945s/12 iters), loss = 0.0993061 I0407 23:25:38.629281 359 solver.cpp:237] Train net output #0: loss = 0.0993061 (* 1 = 0.0993061 loss) I0407 23:25:38.629288 359 sgd_solver.cpp:105] Iteration 7164, lr = 0.000171736 I0407 23:25:43.583881 359 solver.cpp:218] Iteration 7176 (2.422 iter/s, 4.95458s/12 iters), loss = 0.0892541 I0407 23:25:43.583933 359 solver.cpp:237] Train net output #0: loss = 0.0892541 (* 1 = 0.0892541 loss) I0407 23:25:43.583942 359 sgd_solver.cpp:105] Iteration 7176, lr = 0.000167809 I0407 23:25:45.655647 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:25:48.480268 359 solver.cpp:218] Iteration 7188 (2.45083 iter/s, 4.89631s/12 iters), loss = 0.188101 I0407 23:25:48.480309 359 solver.cpp:237] Train net output #0: loss = 0.188101 (* 1 = 0.188101 loss) I0407 23:25:48.480319 359 sgd_solver.cpp:105] Iteration 7188, lr = 0.000163971 I0407 23:25:53.413872 359 solver.cpp:218] Iteration 7200 (2.43233 iter/s, 4.93354s/12 iters), loss = 0.188401 I0407 23:25:53.413934 359 solver.cpp:237] Train net output #0: loss = 0.188401 (* 1 = 0.188401 loss) I0407 23:25:53.413942 359 sgd_solver.cpp:105] Iteration 7200, lr = 0.000160219 I0407 23:25:58.344467 359 solver.cpp:218] Iteration 7212 (2.43382 iter/s, 4.93052s/12 iters), loss = 0.127247 I0407 23:25:58.344501 359 solver.cpp:237] Train net output #0: loss = 0.127247 (* 1 = 0.127247 loss) I0407 23:25:58.344509 359 sgd_solver.cpp:105] Iteration 7212, lr = 0.000156551 I0407 23:26:03.242328 359 solver.cpp:218] Iteration 7224 (2.45008 iter/s, 4.89781s/12 iters), loss = 0.152345 I0407 23:26:03.242363 359 solver.cpp:237] Train net output #0: loss = 0.152345 (* 1 = 0.152345 loss) I0407 23:26:03.242372 359 sgd_solver.cpp:105] Iteration 7224, lr = 0.000152967 I0407 23:26:08.191819 359 solver.cpp:218] Iteration 7236 (2.42452 iter/s, 4.94944s/12 iters), loss = 0.101342 I0407 23:26:08.191854 359 solver.cpp:237] Train net output #0: loss = 0.101342 (* 1 = 0.101342 loss) I0407 23:26:08.191862 359 sgd_solver.cpp:105] Iteration 7236, lr = 0.000149463 I0407 23:26:10.191319 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_7242.caffemodel I0407 23:26:13.329421 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_7242.solverstate I0407 23:26:15.743396 359 solver.cpp:330] Iteration 7242, Testing net (#0) I0407 23:26:15.743415 359 net.cpp:676] Ignoring source layer train-data I0407 23:26:17.394312 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:26:20.311830 359 solver.cpp:397] Test net output #0: accuracy = 0.457721 I0407 23:26:20.311866 359 solver.cpp:397] Test net output #1: loss = 2.8281 (* 1 = 2.8281 loss) I0407 23:26:22.118371 359 solver.cpp:218] Iteration 7248 (0.861668 iter/s, 13.9265s/12 iters), loss = 0.161142 I0407 23:26:22.118412 359 solver.cpp:237] Train net output #0: loss = 0.161142 (* 1 = 0.161142 loss) I0407 23:26:22.118422 359 sgd_solver.cpp:105] Iteration 7248, lr = 0.000146038 I0407 23:26:27.061333 359 solver.cpp:218] Iteration 7260 (2.42772 iter/s, 4.9429s/12 iters), loss = 0.0641212 I0407 23:26:27.061465 359 solver.cpp:237] Train net output #0: loss = 0.0641212 (* 1 = 0.0641212 loss) I0407 23:26:27.061473 359 sgd_solver.cpp:105] Iteration 7260, lr = 0.00014269 I0407 23:26:31.940481 359 solver.cpp:218] Iteration 7272 (2.45952 iter/s, 4.879s/12 iters), loss = 0.097481 I0407 23:26:31.940527 359 solver.cpp:237] Train net output #0: loss = 0.0974811 (* 1 = 0.0974811 loss) I0407 23:26:31.940536 359 sgd_solver.cpp:105] Iteration 7272, lr = 0.000139418 I0407 23:26:36.102092 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:26:36.841253 359 solver.cpp:218] Iteration 7284 (2.44863 iter/s, 4.90069s/12 iters), loss = 0.119114 I0407 23:26:36.841300 359 solver.cpp:237] Train net output #0: loss = 0.119114 (* 1 = 0.119114 loss) I0407 23:26:36.841307 359 sgd_solver.cpp:105] Iteration 7284, lr = 0.00013622 I0407 23:26:41.810151 359 solver.cpp:218] Iteration 7296 (2.41506 iter/s, 4.96883s/12 iters), loss = 0.0925093 I0407 23:26:41.810197 359 solver.cpp:237] Train net output #0: loss = 0.0925093 (* 1 = 0.0925093 loss) I0407 23:26:41.810206 359 sgd_solver.cpp:105] Iteration 7296, lr = 0.000133094 I0407 23:26:46.759048 359 solver.cpp:218] Iteration 7308 (2.42482 iter/s, 4.94883s/12 iters), loss = 0.153806 I0407 23:26:46.759092 359 solver.cpp:237] Train net output #0: loss = 0.153806 (* 1 = 0.153806 loss) I0407 23:26:46.759101 359 sgd_solver.cpp:105] Iteration 7308, lr = 0.00013004 I0407 23:26:51.659354 359 solver.cpp:218] Iteration 7320 (2.44886 iter/s, 4.90024s/12 iters), loss = 0.263687 I0407 23:26:51.659399 359 solver.cpp:237] Train net output #0: loss = 0.263687 (* 1 = 0.263687 loss) I0407 23:26:51.659407 359 sgd_solver.cpp:105] Iteration 7320, lr = 0.000127054 I0407 23:26:56.606472 359 solver.cpp:218] Iteration 7332 (2.42569 iter/s, 4.94705s/12 iters), loss = 0.123806 I0407 23:26:56.606519 359 solver.cpp:237] Train net output #0: loss = 0.123806 (* 1 = 0.123806 loss) I0407 23:26:56.606531 359 sgd_solver.cpp:105] Iteration 7332, lr = 0.000124136 I0407 23:27:01.061975 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_7344.caffemodel I0407 23:27:04.517289 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_7344.solverstate I0407 23:27:07.873240 359 solver.cpp:330] Iteration 7344, Testing net (#0) I0407 23:27:07.873258 359 net.cpp:676] Ignoring source layer train-data I0407 23:27:09.503477 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:27:12.685493 359 solver.cpp:397] Test net output #0: accuracy = 0.458333 I0407 23:27:12.685520 359 solver.cpp:397] Test net output #1: loss = 2.81908 (* 1 = 2.81908 loss) I0407 23:27:12.782277 359 solver.cpp:218] Iteration 7344 (0.741852 iter/s, 16.1757s/12 iters), loss = 0.103389 I0407 23:27:12.782321 359 solver.cpp:237] Train net output #0: loss = 0.10339 (* 1 = 0.10339 loss) I0407 23:27:12.782330 359 sgd_solver.cpp:105] Iteration 7344, lr = 0.000121284 I0407 23:27:16.921480 359 solver.cpp:218] Iteration 7356 (2.89915 iter/s, 4.13914s/12 iters), loss = 0.128865 I0407 23:27:16.921522 359 solver.cpp:237] Train net output #0: loss = 0.128865 (* 1 = 0.128865 loss) I0407 23:27:16.921531 359 sgd_solver.cpp:105] Iteration 7356, lr = 0.000118497 I0407 23:27:21.841703 359 solver.cpp:218] Iteration 7368 (2.43895 iter/s, 4.92016s/12 iters), loss = 0.146078 I0407 23:27:21.841749 359 solver.cpp:237] Train net output #0: loss = 0.146078 (* 1 = 0.146078 loss) I0407 23:27:21.841758 359 sgd_solver.cpp:105] Iteration 7368, lr = 0.000115774 I0407 23:27:26.791333 359 solver.cpp:218] Iteration 7380 (2.42446 iter/s, 4.94957s/12 iters), loss = 0.0587999 I0407 23:27:26.791374 359 solver.cpp:237] Train net output #0: loss = 0.0587999 (* 1 = 0.0587999 loss) I0407 23:27:26.791383 359 sgd_solver.cpp:105] Iteration 7380, lr = 0.000113112 I0407 23:27:28.137576 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:27:31.679008 359 solver.cpp:218] Iteration 7392 (2.45518 iter/s, 4.88762s/12 iters), loss = 0.196448 I0407 23:27:31.679145 359 solver.cpp:237] Train net output #0: loss = 0.196448 (* 1 = 0.196448 loss) I0407 23:27:31.679155 359 sgd_solver.cpp:105] Iteration 7392, lr = 0.00011051 I0407 23:27:36.649873 359 solver.cpp:218] Iteration 7404 (2.41414 iter/s, 4.97072s/12 iters), loss = 0.104589 I0407 23:27:36.649904 359 solver.cpp:237] Train net output #0: loss = 0.104589 (* 1 = 0.104589 loss) I0407 23:27:36.649912 359 sgd_solver.cpp:105] Iteration 7404, lr = 0.000107968 I0407 23:27:41.546430 359 solver.cpp:218] Iteration 7416 (2.45073 iter/s, 4.89651s/12 iters), loss = 0.0983313 I0407 23:27:41.546465 359 solver.cpp:237] Train net output #0: loss = 0.0983313 (* 1 = 0.0983313 loss) I0407 23:27:41.546473 359 sgd_solver.cpp:105] Iteration 7416, lr = 0.000105484 I0407 23:27:46.514627 359 solver.cpp:218] Iteration 7428 (2.41539 iter/s, 4.96814s/12 iters), loss = 0.15201 I0407 23:27:46.514665 359 solver.cpp:237] Train net output #0: loss = 0.15201 (* 1 = 0.15201 loss) I0407 23:27:46.514673 359 sgd_solver.cpp:105] Iteration 7428, lr = 0.000103056 I0407 23:27:51.431454 359 solver.cpp:218] Iteration 7440 (2.44063 iter/s, 4.91677s/12 iters), loss = 0.167093 I0407 23:27:51.431491 359 solver.cpp:237] Train net output #0: loss = 0.167093 (* 1 = 0.167093 loss) I0407 23:27:51.431499 359 sgd_solver.cpp:105] Iteration 7440, lr = 0.000100684 I0407 23:27:53.427947 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_7446.caffemodel I0407 23:27:56.502161 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_7446.solverstate I0407 23:27:58.871912 359 solver.cpp:330] Iteration 7446, Testing net (#0) I0407 23:27:58.871929 359 net.cpp:676] Ignoring source layer train-data I0407 23:28:00.456387 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:28:03.699553 359 solver.cpp:397] Test net output #0: accuracy = 0.452206 I0407 23:28:03.699710 359 solver.cpp:397] Test net output #1: loss = 2.84002 (* 1 = 2.84002 loss) I0407 23:28:05.497198 359 solver.cpp:218] Iteration 7452 (0.853141 iter/s, 14.0657s/12 iters), loss = 0.0823287 I0407 23:28:05.497236 359 solver.cpp:237] Train net output #0: loss = 0.0823287 (* 1 = 0.0823287 loss) I0407 23:28:05.497244 359 sgd_solver.cpp:105] Iteration 7452, lr = 9.83655e-05 I0407 23:28:10.405025 359 solver.cpp:218] Iteration 7464 (2.4451 iter/s, 4.90777s/12 iters), loss = 0.0612384 I0407 23:28:10.405064 359 solver.cpp:237] Train net output #0: loss = 0.0612384 (* 1 = 0.0612384 loss) I0407 23:28:10.405071 359 sgd_solver.cpp:105] Iteration 7464, lr = 9.61e-05 I0407 23:28:15.322244 359 solver.cpp:218] Iteration 7476 (2.44043 iter/s, 4.91716s/12 iters), loss = 0.10629 I0407 23:28:15.322284 359 solver.cpp:237] Train net output #0: loss = 0.10629 (* 1 = 0.10629 loss) I0407 23:28:15.322293 359 sgd_solver.cpp:105] Iteration 7476, lr = 9.38862e-05 I0407 23:28:18.794477 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:28:20.268381 359 solver.cpp:218] Iteration 7488 (2.42617 iter/s, 4.94608s/12 iters), loss = 0.157365 I0407 23:28:20.268424 359 solver.cpp:237] Train net output #0: loss = 0.157365 (* 1 = 0.157365 loss) I0407 23:28:20.268432 359 sgd_solver.cpp:105] Iteration 7488, lr = 9.1723e-05 I0407 23:28:25.195538 359 solver.cpp:218] Iteration 7500 (2.43552 iter/s, 4.92709s/12 iters), loss = 0.0981837 I0407 23:28:25.195590 359 solver.cpp:237] Train net output #0: loss = 0.0981837 (* 1 = 0.0981837 loss) I0407 23:28:25.195603 359 sgd_solver.cpp:105] Iteration 7500, lr = 8.96091e-05 I0407 23:28:30.147296 359 solver.cpp:218] Iteration 7512 (2.42342 iter/s, 4.95169s/12 iters), loss = 0.100816 I0407 23:28:30.147337 359 solver.cpp:237] Train net output #0: loss = 0.100816 (* 1 = 0.100816 loss) I0407 23:28:30.147346 359 sgd_solver.cpp:105] Iteration 7512, lr = 8.75435e-05 I0407 23:28:35.043851 359 solver.cpp:218] Iteration 7524 (2.45073 iter/s, 4.89649s/12 iters), loss = 0.128969 I0407 23:28:35.044023 359 solver.cpp:237] Train net output #0: loss = 0.128969 (* 1 = 0.128969 loss) I0407 23:28:35.044035 359 sgd_solver.cpp:105] Iteration 7524, lr = 8.55251e-05 I0407 23:28:39.916781 359 solver.cpp:218] Iteration 7536 (2.46268 iter/s, 4.87274s/12 iters), loss = 0.0964068 I0407 23:28:39.916826 359 solver.cpp:237] Train net output #0: loss = 0.0964068 (* 1 = 0.0964068 loss) I0407 23:28:39.916834 359 sgd_solver.cpp:105] Iteration 7536, lr = 8.35528e-05 I0407 23:28:44.411808 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_7548.caffemodel I0407 23:28:47.514835 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_7548.solverstate I0407 23:28:49.876381 359 solver.cpp:330] Iteration 7548, Testing net (#0) I0407 23:28:49.876400 359 net.cpp:676] Ignoring source layer train-data I0407 23:28:51.413739 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:28:54.714722 359 solver.cpp:397] Test net output #0: accuracy = 0.450368 I0407 23:28:54.714771 359 solver.cpp:397] Test net output #1: loss = 2.82414 (* 1 = 2.82414 loss) I0407 23:28:54.811138 359 solver.cpp:218] Iteration 7548 (0.805679 iter/s, 14.8943s/12 iters), loss = 0.215424 I0407 23:28:54.811184 359 solver.cpp:237] Train net output #0: loss = 0.215424 (* 1 = 0.215424 loss) I0407 23:28:54.811192 359 sgd_solver.cpp:105] Iteration 7548, lr = 8.16257e-05 I0407 23:28:58.913132 359 solver.cpp:218] Iteration 7560 (2.92545 iter/s, 4.10193s/12 iters), loss = 0.160461 I0407 23:28:58.913178 359 solver.cpp:237] Train net output #0: loss = 0.160461 (* 1 = 0.160461 loss) I0407 23:28:58.913187 359 sgd_solver.cpp:105] Iteration 7560, lr = 7.97426e-05 I0407 23:29:03.854527 359 solver.cpp:218] Iteration 7572 (2.4285 iter/s, 4.94133s/12 iters), loss = 0.0862786 I0407 23:29:03.854568 359 solver.cpp:237] Train net output #0: loss = 0.0862786 (* 1 = 0.0862786 loss) I0407 23:29:03.854578 359 sgd_solver.cpp:105] Iteration 7572, lr = 7.79027e-05 I0407 23:29:08.836550 359 solver.cpp:218] Iteration 7584 (2.40869 iter/s, 4.98196s/12 iters), loss = 0.0929231 I0407 23:29:08.836692 359 solver.cpp:237] Train net output #0: loss = 0.0929231 (* 1 = 0.0929231 loss) I0407 23:29:08.836701 359 sgd_solver.cpp:105] Iteration 7584, lr = 7.61049e-05 I0407 23:29:09.457921 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:29:13.764643 359 solver.cpp:218] Iteration 7596 (2.4351 iter/s, 4.92793s/12 iters), loss = 0.130177 I0407 23:29:13.764689 359 solver.cpp:237] Train net output #0: loss = 0.130177 (* 1 = 0.130177 loss) I0407 23:29:13.764698 359 sgd_solver.cpp:105] Iteration 7596, lr = 7.43482e-05 I0407 23:29:18.683015 359 solver.cpp:218] Iteration 7608 (2.43986 iter/s, 4.91831s/12 iters), loss = 0.124939 I0407 23:29:18.683053 359 solver.cpp:237] Train net output #0: loss = 0.124939 (* 1 = 0.124939 loss) I0407 23:29:18.683063 359 sgd_solver.cpp:105] Iteration 7608, lr = 7.26318e-05 I0407 23:29:23.533573 359 solver.cpp:218] Iteration 7620 (2.47397 iter/s, 4.8505s/12 iters), loss = 0.0977765 I0407 23:29:23.533617 359 solver.cpp:237] Train net output #0: loss = 0.0977765 (* 1 = 0.0977765 loss) I0407 23:29:23.533625 359 sgd_solver.cpp:105] Iteration 7620, lr = 7.09548e-05 I0407 23:29:25.874089 359 blocking_queue.cpp:49] Waiting for data I0407 23:29:28.392765 359 solver.cpp:218] Iteration 7632 (2.46958 iter/s, 4.85913s/12 iters), loss = 0.0783972 I0407 23:29:28.392810 359 solver.cpp:237] Train net output #0: loss = 0.0783972 (* 1 = 0.0783972 loss) I0407 23:29:28.392818 359 sgd_solver.cpp:105] Iteration 7632, lr = 6.93162e-05 I0407 23:29:33.337855 359 solver.cpp:218] Iteration 7644 (2.42668 iter/s, 4.94502s/12 iters), loss = 0.186941 I0407 23:29:33.337898 359 solver.cpp:237] Train net output #0: loss = 0.186941 (* 1 = 0.186941 loss) I0407 23:29:33.337905 359 sgd_solver.cpp:105] Iteration 7644, lr = 6.77152e-05 I0407 23:29:35.343619 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_7650.caffemodel I0407 23:29:38.440865 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_7650.solverstate I0407 23:29:40.827283 359 solver.cpp:330] Iteration 7650, Testing net (#0) I0407 23:29:40.827399 359 net.cpp:676] Ignoring source layer train-data I0407 23:29:42.337857 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:29:45.688679 359 solver.cpp:397] Test net output #0: accuracy = 0.452819 I0407 23:29:45.688707 359 solver.cpp:397] Test net output #1: loss = 2.83726 (* 1 = 2.83726 loss) I0407 23:29:47.478560 359 solver.cpp:218] Iteration 7656 (0.848618 iter/s, 14.1406s/12 iters), loss = 0.100135 I0407 23:29:47.478597 359 solver.cpp:237] Train net output #0: loss = 0.100135 (* 1 = 0.100135 loss) I0407 23:29:47.478605 359 sgd_solver.cpp:105] Iteration 7656, lr = 6.61509e-05 I0407 23:29:52.431851 359 solver.cpp:218] Iteration 7668 (2.42266 iter/s, 4.95324s/12 iters), loss = 0.133091 I0407 23:29:52.431883 359 solver.cpp:237] Train net output #0: loss = 0.133091 (* 1 = 0.133091 loss) I0407 23:29:52.431890 359 sgd_solver.cpp:105] Iteration 7668, lr = 6.46225e-05 I0407 23:29:57.359477 359 solver.cpp:218] Iteration 7680 (2.43528 iter/s, 4.92757s/12 iters), loss = 0.138438 I0407 23:29:57.359513 359 solver.cpp:237] Train net output #0: loss = 0.138438 (* 1 = 0.138438 loss) I0407 23:29:57.359520 359 sgd_solver.cpp:105] Iteration 7680, lr = 6.31292e-05 I0407 23:30:00.120311 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:30:02.288190 359 solver.cpp:218] Iteration 7692 (2.43474 iter/s, 4.92866s/12 iters), loss = 0.171719 I0407 23:30:02.288228 359 solver.cpp:237] Train net output #0: loss = 0.171719 (* 1 = 0.171719 loss) I0407 23:30:02.288235 359 sgd_solver.cpp:105] Iteration 7692, lr = 6.16702e-05 I0407 23:30:07.204720 359 solver.cpp:218] Iteration 7704 (2.44078 iter/s, 4.91647s/12 iters), loss = 0.0526268 I0407 23:30:07.204764 359 solver.cpp:237] Train net output #0: loss = 0.0526269 (* 1 = 0.0526269 loss) I0407 23:30:07.204773 359 sgd_solver.cpp:105] Iteration 7704, lr = 6.02447e-05 I0407 23:30:12.156980 359 solver.cpp:218] Iteration 7716 (2.42317 iter/s, 4.95219s/12 iters), loss = 0.0985625 I0407 23:30:12.157104 359 solver.cpp:237] Train net output #0: loss = 0.0985625 (* 1 = 0.0985625 loss) I0407 23:30:12.157112 359 sgd_solver.cpp:105] Iteration 7716, lr = 5.8852e-05 I0407 23:30:17.069025 359 solver.cpp:218] Iteration 7728 (2.44305 iter/s, 4.9119s/12 iters), loss = 0.136825 I0407 23:30:17.069072 359 solver.cpp:237] Train net output #0: loss = 0.136825 (* 1 = 0.136825 loss) I0407 23:30:17.069079 359 sgd_solver.cpp:105] Iteration 7728, lr = 5.74913e-05 I0407 23:30:22.034132 359 solver.cpp:218] Iteration 7740 (2.4169 iter/s, 4.96505s/12 iters), loss = 0.132538 I0407 23:30:22.034163 359 solver.cpp:237] Train net output #0: loss = 0.132538 (* 1 = 0.132538 loss) I0407 23:30:22.034171 359 sgd_solver.cpp:105] Iteration 7740, lr = 5.61618e-05 I0407 23:30:26.498157 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_7752.caffemodel I0407 23:30:29.626694 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_7752.solverstate I0407 23:30:31.996178 359 solver.cpp:330] Iteration 7752, Testing net (#0) I0407 23:30:31.996196 359 net.cpp:676] Ignoring source layer train-data I0407 23:30:33.442696 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:30:36.751013 359 solver.cpp:397] Test net output #0: accuracy = 0.451593 I0407 23:30:36.751061 359 solver.cpp:397] Test net output #1: loss = 2.82652 (* 1 = 2.82652 loss) I0407 23:30:36.847776 359 solver.cpp:218] Iteration 7752 (0.810067 iter/s, 14.8136s/12 iters), loss = 0.161823 I0407 23:30:36.847820 359 solver.cpp:237] Train net output #0: loss = 0.161824 (* 1 = 0.161824 loss) I0407 23:30:36.847829 359 sgd_solver.cpp:105] Iteration 7752, lr = 5.4863e-05 I0407 23:30:40.917726 359 solver.cpp:218] Iteration 7764 (2.94848 iter/s, 4.06989s/12 iters), loss = 0.180322 I0407 23:30:40.917771 359 solver.cpp:237] Train net output #0: loss = 0.180322 (* 1 = 0.180322 loss) I0407 23:30:40.917780 359 sgd_solver.cpp:105] Iteration 7764, lr = 5.3594e-05 I0407 23:30:45.918315 359 solver.cpp:218] Iteration 7776 (2.39975 iter/s, 5.00052s/12 iters), loss = 0.166381 I0407 23:30:45.918469 359 solver.cpp:237] Train net output #0: loss = 0.166381 (* 1 = 0.166381 loss) I0407 23:30:45.918479 359 sgd_solver.cpp:105] Iteration 7776, lr = 5.23542e-05 I0407 23:30:50.836014 359 solver.cpp:218] Iteration 7788 (2.44025 iter/s, 4.91752s/12 iters), loss = 0.13637 I0407 23:30:50.836061 359 solver.cpp:237] Train net output #0: loss = 0.13637 (* 1 = 0.13637 loss) I0407 23:30:50.836068 359 sgd_solver.cpp:105] Iteration 7788, lr = 5.11429e-05 I0407 23:30:50.842371 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:30:55.787715 359 solver.cpp:218] Iteration 7800 (2.42344 iter/s, 4.95163s/12 iters), loss = 0.115531 I0407 23:30:55.787762 359 solver.cpp:237] Train net output #0: loss = 0.115531 (* 1 = 0.115531 loss) I0407 23:30:55.787771 359 sgd_solver.cpp:105] Iteration 7800, lr = 4.99596e-05 I0407 23:31:00.731448 359 solver.cpp:218] Iteration 7812 (2.42735 iter/s, 4.94367s/12 iters), loss = 0.159319 I0407 23:31:00.731490 359 solver.cpp:237] Train net output #0: loss = 0.159319 (* 1 = 0.159319 loss) I0407 23:31:00.731498 359 sgd_solver.cpp:105] Iteration 7812, lr = 4.88034e-05 I0407 23:31:05.690143 359 solver.cpp:218] Iteration 7824 (2.42003 iter/s, 4.95862s/12 iters), loss = 0.190898 I0407 23:31:05.690202 359 solver.cpp:237] Train net output #0: loss = 0.190898 (* 1 = 0.190898 loss) I0407 23:31:05.690213 359 sgd_solver.cpp:105] Iteration 7824, lr = 4.76739e-05 I0407 23:31:10.649706 359 solver.cpp:218] Iteration 7836 (2.4196 iter/s, 4.95949s/12 iters), loss = 0.131701 I0407 23:31:10.649742 359 solver.cpp:237] Train net output #0: loss = 0.131701 (* 1 = 0.131701 loss) I0407 23:31:10.649749 359 sgd_solver.cpp:105] Iteration 7836, lr = 4.65705e-05 I0407 23:31:15.593343 359 solver.cpp:218] Iteration 7848 (2.42739 iter/s, 4.94358s/12 iters), loss = 0.100684 I0407 23:31:15.593381 359 solver.cpp:237] Train net output #0: loss = 0.100684 (* 1 = 0.100684 loss) I0407 23:31:15.593389 359 sgd_solver.cpp:105] Iteration 7848, lr = 4.54924e-05 I0407 23:31:17.607589 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_7854.caffemodel I0407 23:31:20.820639 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_7854.solverstate I0407 23:31:23.295486 359 solver.cpp:330] Iteration 7854, Testing net (#0) I0407 23:31:23.295502 359 net.cpp:676] Ignoring source layer train-data I0407 23:31:24.710500 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:31:28.186986 359 solver.cpp:397] Test net output #0: accuracy = 0.450368 I0407 23:31:28.187026 359 solver.cpp:397] Test net output #1: loss = 2.83505 (* 1 = 2.83505 loss) I0407 23:31:29.999689 359 solver.cpp:218] Iteration 7860 (0.83297 iter/s, 14.4063s/12 iters), loss = 0.110066 I0407 23:31:29.999734 359 solver.cpp:237] Train net output #0: loss = 0.110066 (* 1 = 0.110066 loss) I0407 23:31:29.999742 359 sgd_solver.cpp:105] Iteration 7860, lr = 4.44392e-05 I0407 23:31:34.907042 359 solver.cpp:218] Iteration 7872 (2.44534 iter/s, 4.90729s/12 iters), loss = 0.135548 I0407 23:31:34.907086 359 solver.cpp:237] Train net output #0: loss = 0.135548 (* 1 = 0.135548 loss) I0407 23:31:34.907094 359 sgd_solver.cpp:105] Iteration 7872, lr = 4.34102e-05 I0407 23:31:39.873056 359 solver.cpp:218] Iteration 7884 (2.41646 iter/s, 4.96595s/12 iters), loss = 0.054767 I0407 23:31:39.873098 359 solver.cpp:237] Train net output #0: loss = 0.054767 (* 1 = 0.054767 loss) I0407 23:31:39.873106 359 sgd_solver.cpp:105] Iteration 7884, lr = 4.2405e-05 I0407 23:31:41.984211 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:31:44.824985 359 solver.cpp:218] Iteration 7896 (2.42333 iter/s, 4.95186s/12 iters), loss = 0.199566 I0407 23:31:44.825026 359 solver.cpp:237] Train net output #0: loss = 0.199566 (* 1 = 0.199566 loss) I0407 23:31:44.825033 359 sgd_solver.cpp:105] Iteration 7896, lr = 4.1423e-05 I0407 23:31:49.743180 359 solver.cpp:218] Iteration 7908 (2.43995 iter/s, 4.91813s/12 iters), loss = 0.127736 I0407 23:31:49.743315 359 solver.cpp:237] Train net output #0: loss = 0.127736 (* 1 = 0.127736 loss) I0407 23:31:49.743324 359 sgd_solver.cpp:105] Iteration 7908, lr = 4.04636e-05 I0407 23:31:54.692584 359 solver.cpp:218] Iteration 7920 (2.42461 iter/s, 4.94925s/12 iters), loss = 0.0679647 I0407 23:31:54.692621 359 solver.cpp:237] Train net output #0: loss = 0.0679648 (* 1 = 0.0679648 loss) I0407 23:31:54.692629 359 sgd_solver.cpp:105] Iteration 7920, lr = 3.95264e-05 I0407 23:31:59.591023 359 solver.cpp:218] Iteration 7932 (2.44979 iter/s, 4.89839s/12 iters), loss = 0.146791 I0407 23:31:59.591060 359 solver.cpp:237] Train net output #0: loss = 0.146791 (* 1 = 0.146791 loss) I0407 23:31:59.591068 359 sgd_solver.cpp:105] Iteration 7932, lr = 3.86107e-05 I0407 23:32:04.563835 359 solver.cpp:218] Iteration 7944 (2.41315 iter/s, 4.97275s/12 iters), loss = 0.112108 I0407 23:32:04.563874 359 solver.cpp:237] Train net output #0: loss = 0.112108 (* 1 = 0.112108 loss) I0407 23:32:04.563882 359 sgd_solver.cpp:105] Iteration 7944, lr = 3.77162e-05 I0407 23:32:09.028057 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_7956.caffemodel I0407 23:32:12.097879 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_7956.solverstate I0407 23:32:14.462730 359 solver.cpp:330] Iteration 7956, Testing net (#0) I0407 23:32:14.462759 359 net.cpp:676] Ignoring source layer train-data I0407 23:32:15.847432 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:32:19.101797 359 solver.cpp:397] Test net output #0: accuracy = 0.45527 I0407 23:32:19.101836 359 solver.cpp:397] Test net output #1: loss = 2.82293 (* 1 = 2.82293 loss) I0407 23:32:19.198400 359 solver.cpp:218] Iteration 7956 (0.819981 iter/s, 14.6345s/12 iters), loss = 0.0440542 I0407 23:32:19.198443 359 solver.cpp:237] Train net output #0: loss = 0.0440543 (* 1 = 0.0440543 loss) I0407 23:32:19.198451 359 sgd_solver.cpp:105] Iteration 7956, lr = 3.68424e-05 I0407 23:32:23.230445 359 solver.cpp:218] Iteration 7968 (2.9762 iter/s, 4.03199s/12 iters), loss = 0.182227 I0407 23:32:23.230568 359 solver.cpp:237] Train net output #0: loss = 0.182227 (* 1 = 0.182227 loss) I0407 23:32:23.230577 359 sgd_solver.cpp:105] Iteration 7968, lr = 3.59887e-05 I0407 23:32:28.163815 359 solver.cpp:218] Iteration 7980 (2.43248 iter/s, 4.93323s/12 iters), loss = 0.199141 I0407 23:32:28.163852 359 solver.cpp:237] Train net output #0: loss = 0.199141 (* 1 = 0.199141 loss) I0407 23:32:28.163861 359 sgd_solver.cpp:105] Iteration 7980, lr = 3.51547e-05 I0407 23:32:32.363152 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:32:33.021293 359 solver.cpp:218] Iteration 7992 (2.47045 iter/s, 4.85742s/12 iters), loss = 0.152924 I0407 23:32:33.021327 359 solver.cpp:237] Train net output #0: loss = 0.152924 (* 1 = 0.152924 loss) I0407 23:32:33.021335 359 sgd_solver.cpp:105] Iteration 7992, lr = 3.434e-05 I0407 23:32:37.985229 359 solver.cpp:218] Iteration 8004 (2.41746 iter/s, 4.96389s/12 iters), loss = 0.0793058 I0407 23:32:37.985263 359 solver.cpp:237] Train net output #0: loss = 0.0793058 (* 1 = 0.0793058 loss) I0407 23:32:37.985271 359 sgd_solver.cpp:105] Iteration 8004, lr = 3.35441e-05 I0407 23:32:42.897962 359 solver.cpp:218] Iteration 8016 (2.44266 iter/s, 4.91268s/12 iters), loss = 0.14935 I0407 23:32:42.897997 359 solver.cpp:237] Train net output #0: loss = 0.14935 (* 1 = 0.14935 loss) I0407 23:32:42.898005 359 sgd_solver.cpp:105] Iteration 8016, lr = 3.27666e-05 I0407 23:32:47.788993 359 solver.cpp:218] Iteration 8028 (2.4535 iter/s, 4.89098s/12 iters), loss = 0.115184 I0407 23:32:47.789032 359 solver.cpp:237] Train net output #0: loss = 0.115184 (* 1 = 0.115184 loss) I0407 23:32:47.789041 359 sgd_solver.cpp:105] Iteration 8028, lr = 3.20071e-05 I0407 23:32:52.683718 359 solver.cpp:218] Iteration 8040 (2.45165 iter/s, 4.89466s/12 iters), loss = 0.169018 I0407 23:32:52.683758 359 solver.cpp:237] Train net output #0: loss = 0.169018 (* 1 = 0.169018 loss) I0407 23:32:52.683766 359 sgd_solver.cpp:105] Iteration 8040, lr = 3.12651e-05 I0407 23:32:57.533496 359 solver.cpp:218] Iteration 8052 (2.47437 iter/s, 4.84972s/12 iters), loss = 0.0941235 I0407 23:32:57.533654 359 solver.cpp:237] Train net output #0: loss = 0.0941235 (* 1 = 0.0941235 loss) I0407 23:32:57.533664 359 sgd_solver.cpp:105] Iteration 8052, lr = 3.05403e-05 I0407 23:32:59.543113 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_8058.caffemodel I0407 23:33:02.643108 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_8058.solverstate I0407 23:33:05.006484 359 solver.cpp:330] Iteration 8058, Testing net (#0) I0407 23:33:05.006502 359 net.cpp:676] Ignoring source layer train-data I0407 23:33:06.332792 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:33:09.763608 359 solver.cpp:397] Test net output #0: accuracy = 0.451593 I0407 23:33:09.763656 359 solver.cpp:397] Test net output #1: loss = 2.83435 (* 1 = 2.83435 loss) I0407 23:33:11.501518 359 solver.cpp:218] Iteration 8064 (0.859117 iter/s, 13.9678s/12 iters), loss = 0.109753 I0407 23:33:11.501559 359 solver.cpp:237] Train net output #0: loss = 0.109753 (* 1 = 0.109753 loss) I0407 23:33:11.501567 359 sgd_solver.cpp:105] Iteration 8064, lr = 2.98322e-05 I0407 23:33:16.439133 359 solver.cpp:218] Iteration 8076 (2.43035 iter/s, 4.93756s/12 iters), loss = 0.0864122 I0407 23:33:16.439169 359 solver.cpp:237] Train net output #0: loss = 0.0864122 (* 1 = 0.0864122 loss) I0407 23:33:16.439177 359 sgd_solver.cpp:105] Iteration 8076, lr = 2.91405e-05 I0407 23:33:21.394717 359 solver.cpp:218] Iteration 8088 (2.42154 iter/s, 4.95553s/12 iters), loss = 0.139236 I0407 23:33:21.394753 359 solver.cpp:237] Train net output #0: loss = 0.139236 (* 1 = 0.139236 loss) I0407 23:33:21.394760 359 sgd_solver.cpp:105] Iteration 8088, lr = 2.84647e-05 I0407 23:33:22.773025 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:33:26.297230 359 solver.cpp:218] Iteration 8100 (2.44775 iter/s, 4.90245s/12 iters), loss = 0.171817 I0407 23:33:26.297267 359 solver.cpp:237] Train net output #0: loss = 0.171817 (* 1 = 0.171817 loss) I0407 23:33:26.297276 359 sgd_solver.cpp:105] Iteration 8100, lr = 2.78046e-05 I0407 23:33:31.270700 359 solver.cpp:218] Iteration 8112 (2.41283 iter/s, 4.97342s/12 iters), loss = 0.177519 I0407 23:33:31.270818 359 solver.cpp:237] Train net output #0: loss = 0.177519 (* 1 = 0.177519 loss) I0407 23:33:31.270828 359 sgd_solver.cpp:105] Iteration 8112, lr = 2.71598e-05 I0407 23:33:36.160349 359 solver.cpp:218] Iteration 8124 (2.45423 iter/s, 4.88952s/12 iters), loss = 0.138745 I0407 23:33:36.160383 359 solver.cpp:237] Train net output #0: loss = 0.138745 (* 1 = 0.138745 loss) I0407 23:33:36.160392 359 sgd_solver.cpp:105] Iteration 8124, lr = 2.65299e-05 I0407 23:33:41.049350 359 solver.cpp:218] Iteration 8136 (2.45452 iter/s, 4.88895s/12 iters), loss = 0.144014 I0407 23:33:41.049391 359 solver.cpp:237] Train net output #0: loss = 0.144014 (* 1 = 0.144014 loss) I0407 23:33:41.049398 359 sgd_solver.cpp:105] Iteration 8136, lr = 2.59145e-05 I0407 23:33:46.007997 359 solver.cpp:218] Iteration 8148 (2.42004 iter/s, 4.95859s/12 iters), loss = 0.115102 I0407 23:33:46.008036 359 solver.cpp:237] Train net output #0: loss = 0.115102 (* 1 = 0.115102 loss) I0407 23:33:46.008044 359 sgd_solver.cpp:105] Iteration 8148, lr = 2.53134e-05 I0407 23:33:50.515622 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_8160.caffemodel I0407 23:33:53.589617 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_8160.solverstate I0407 23:33:55.999799 359 solver.cpp:330] Iteration 8160, Testing net (#0) I0407 23:33:55.999816 359 net.cpp:676] Ignoring source layer train-data I0407 23:33:57.269635 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:34:00.778331 359 solver.cpp:397] Test net output #0: accuracy = 0.450368 I0407 23:34:00.778373 359 solver.cpp:397] Test net output #1: loss = 2.84238 (* 1 = 2.84238 loss) I0407 23:34:00.874678 359 solver.cpp:218] Iteration 8160 (0.807178 iter/s, 14.8666s/12 iters), loss = 0.199744 I0407 23:34:00.874722 359 solver.cpp:237] Train net output #0: loss = 0.199744 (* 1 = 0.199744 loss) I0407 23:34:00.874732 359 sgd_solver.cpp:105] Iteration 8160, lr = 2.47262e-05 I0407 23:34:04.965214 359 solver.cpp:218] Iteration 8172 (2.93365 iter/s, 4.09047s/12 iters), loss = 0.231639 I0407 23:34:04.965356 359 solver.cpp:237] Train net output #0: loss = 0.231639 (* 1 = 0.231639 loss) I0407 23:34:04.965365 359 sgd_solver.cpp:105] Iteration 8172, lr = 2.41526e-05 I0407 23:34:09.918386 359 solver.cpp:218] Iteration 8184 (2.42277 iter/s, 4.95301s/12 iters), loss = 0.143868 I0407 23:34:09.918423 359 solver.cpp:237] Train net output #0: loss = 0.143868 (* 1 = 0.143868 loss) I0407 23:34:09.918431 359 sgd_solver.cpp:105] Iteration 8184, lr = 2.35923e-05 I0407 23:34:13.396580 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:34:14.833173 359 solver.cpp:218] Iteration 8196 (2.44164 iter/s, 4.91473s/12 iters), loss = 0.0685172 I0407 23:34:14.833211 359 solver.cpp:237] Train net output #0: loss = 0.0685173 (* 1 = 0.0685173 loss) I0407 23:34:14.833218 359 sgd_solver.cpp:105] Iteration 8196, lr = 2.30449e-05 I0407 23:34:19.807173 359 solver.cpp:218] Iteration 8208 (2.41258 iter/s, 4.97394s/12 iters), loss = 0.14359 I0407 23:34:19.807224 359 solver.cpp:237] Train net output #0: loss = 0.14359 (* 1 = 0.14359 loss) I0407 23:34:19.807233 359 sgd_solver.cpp:105] Iteration 8208, lr = 2.25102e-05 I0407 23:34:24.754192 359 solver.cpp:218] Iteration 8220 (2.42574 iter/s, 4.94695s/12 iters), loss = 0.0174887 I0407 23:34:24.754232 359 solver.cpp:237] Train net output #0: loss = 0.0174888 (* 1 = 0.0174888 loss) I0407 23:34:24.754240 359 sgd_solver.cpp:105] Iteration 8220, lr = 2.19879e-05 I0407 23:34:29.654958 359 solver.cpp:218] Iteration 8232 (2.44862 iter/s, 4.90071s/12 iters), loss = 0.133079 I0407 23:34:29.654996 359 solver.cpp:237] Train net output #0: loss = 0.133079 (* 1 = 0.133079 loss) I0407 23:34:29.655007 359 sgd_solver.cpp:105] Iteration 8232, lr = 2.14777e-05 I0407 23:34:34.613402 359 solver.cpp:218] Iteration 8244 (2.42014 iter/s, 4.95839s/12 iters), loss = 0.0642758 I0407 23:34:34.613440 359 solver.cpp:237] Train net output #0: loss = 0.0642759 (* 1 = 0.0642759 loss) I0407 23:34:34.613447 359 sgd_solver.cpp:105] Iteration 8244, lr = 2.09793e-05 I0407 23:34:39.521891 359 solver.cpp:218] Iteration 8256 (2.44477 iter/s, 4.90843s/12 iters), loss = 0.0774939 I0407 23:34:39.522049 359 solver.cpp:237] Train net output #0: loss = 0.077494 (* 1 = 0.077494 loss) I0407 23:34:39.522058 359 sgd_solver.cpp:105] Iteration 8256, lr = 2.04924e-05 I0407 23:34:41.525800 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_8262.caffemodel I0407 23:34:44.588554 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_8262.solverstate I0407 23:34:46.951015 359 solver.cpp:330] Iteration 8262, Testing net (#0) I0407 23:34:46.951032 359 net.cpp:676] Ignoring source layer train-data I0407 23:34:48.098776 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:34:51.387286 359 solver.cpp:397] Test net output #0: accuracy = 0.447917 I0407 23:34:51.387331 359 solver.cpp:397] Test net output #1: loss = 2.83274 (* 1 = 2.83274 loss) I0407 23:34:53.118649 359 solver.cpp:218] Iteration 8268 (0.882576 iter/s, 13.5966s/12 iters), loss = 0.091885 I0407 23:34:53.118693 359 solver.cpp:237] Train net output #0: loss = 0.0918851 (* 1 = 0.0918851 loss) I0407 23:34:53.118702 359 sgd_solver.cpp:105] Iteration 8268, lr = 2.00168e-05 I0407 23:34:58.066449 359 solver.cpp:218] Iteration 8280 (2.42535 iter/s, 4.94773s/12 iters), loss = 0.147981 I0407 23:34:58.066495 359 solver.cpp:237] Train net output #0: loss = 0.147981 (* 1 = 0.147981 loss) I0407 23:34:58.066504 359 sgd_solver.cpp:105] Iteration 8280, lr = 1.95522e-05 I0407 23:35:02.989535 359 solver.cpp:218] Iteration 8292 (2.43753 iter/s, 4.92302s/12 iters), loss = 0.144756 I0407 23:35:02.989579 359 solver.cpp:237] Train net output #0: loss = 0.144756 (* 1 = 0.144756 loss) I0407 23:35:02.989588 359 sgd_solver.cpp:105] Iteration 8292, lr = 1.90984e-05 I0407 23:35:03.587188 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:35:07.839907 359 solver.cpp:218] Iteration 8304 (2.47407 iter/s, 4.85031s/12 iters), loss = 0.122598 I0407 23:35:07.839943 359 solver.cpp:237] Train net output #0: loss = 0.122599 (* 1 = 0.122599 loss) I0407 23:35:07.839951 359 sgd_solver.cpp:105] Iteration 8304, lr = 1.86551e-05 I0407 23:35:10.672946 359 blocking_queue.cpp:49] Waiting for data I0407 23:35:12.806915 359 solver.cpp:218] Iteration 8316 (2.41597 iter/s, 4.96695s/12 iters), loss = 0.089899 I0407 23:35:12.806954 359 solver.cpp:237] Train net output #0: loss = 0.0898991 (* 1 = 0.0898991 loss) I0407 23:35:12.806962 359 sgd_solver.cpp:105] Iteration 8316, lr = 1.82221e-05 I0407 23:35:17.669307 359 solver.cpp:218] Iteration 8328 (2.46795 iter/s, 4.86234s/12 iters), loss = 0.124047 I0407 23:35:17.669342 359 solver.cpp:237] Train net output #0: loss = 0.124047 (* 1 = 0.124047 loss) I0407 23:35:17.669350 359 sgd_solver.cpp:105] Iteration 8328, lr = 1.77991e-05 I0407 23:35:22.583503 359 solver.cpp:218] Iteration 8340 (2.44193 iter/s, 4.91414s/12 iters), loss = 0.0786122 I0407 23:35:22.583547 359 solver.cpp:237] Train net output #0: loss = 0.0786123 (* 1 = 0.0786123 loss) I0407 23:35:22.583555 359 sgd_solver.cpp:105] Iteration 8340, lr = 1.73859e-05 I0407 23:35:27.570639 359 solver.cpp:218] Iteration 8352 (2.40622 iter/s, 4.98707s/12 iters), loss = 0.123852 I0407 23:35:27.570683 359 solver.cpp:237] Train net output #0: loss = 0.123852 (* 1 = 0.123852 loss) I0407 23:35:27.570693 359 sgd_solver.cpp:105] Iteration 8352, lr = 1.69823e-05 I0407 23:35:32.068830 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_8364.caffemodel I0407 23:35:35.198611 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_8364.solverstate I0407 23:35:37.552906 359 solver.cpp:330] Iteration 8364, Testing net (#0) I0407 23:35:37.552925 359 net.cpp:676] Ignoring source layer train-data I0407 23:35:38.627948 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:35:41.923678 359 solver.cpp:397] Test net output #0: accuracy = 0.447917 I0407 23:35:41.923846 359 solver.cpp:397] Test net output #1: loss = 2.83897 (* 1 = 2.83897 loss) I0407 23:35:42.020275 359 solver.cpp:218] Iteration 8364 (0.830475 iter/s, 14.4496s/12 iters), loss = 0.0775367 I0407 23:35:42.020320 359 solver.cpp:237] Train net output #0: loss = 0.0775368 (* 1 = 0.0775368 loss) I0407 23:35:42.020330 359 sgd_solver.cpp:105] Iteration 8364, lr = 1.6588e-05 I0407 23:35:46.119009 359 solver.cpp:218] Iteration 8376 (2.92778 iter/s, 4.09866s/12 iters), loss = 0.118774 I0407 23:35:46.119052 359 solver.cpp:237] Train net output #0: loss = 0.118774 (* 1 = 0.118774 loss) I0407 23:35:46.119060 359 sgd_solver.cpp:105] Iteration 8376, lr = 1.62029e-05 I0407 23:35:51.069802 359 solver.cpp:218] Iteration 8388 (2.42389 iter/s, 4.95073s/12 iters), loss = 0.0871738 I0407 23:35:51.069844 359 solver.cpp:237] Train net output #0: loss = 0.0871739 (* 1 = 0.0871739 loss) I0407 23:35:51.069852 359 sgd_solver.cpp:105] Iteration 8388, lr = 1.58267e-05 I0407 23:35:53.821846 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:35:55.958904 359 solver.cpp:218] Iteration 8400 (2.45447 iter/s, 4.88905s/12 iters), loss = 0.149992 I0407 23:35:55.958937 359 solver.cpp:237] Train net output #0: loss = 0.149992 (* 1 = 0.149992 loss) I0407 23:35:55.958945 359 sgd_solver.cpp:105] Iteration 8400, lr = 1.54592e-05 I0407 23:36:00.911273 359 solver.cpp:218] Iteration 8412 (2.42311 iter/s, 4.95232s/12 iters), loss = 0.0321403 I0407 23:36:00.911309 359 solver.cpp:237] Train net output #0: loss = 0.0321404 (* 1 = 0.0321404 loss) I0407 23:36:00.911315 359 sgd_solver.cpp:105] Iteration 8412, lr = 1.51002e-05 I0407 23:36:05.826252 359 solver.cpp:218] Iteration 8424 (2.44155 iter/s, 4.91492s/12 iters), loss = 0.0848399 I0407 23:36:05.826297 359 solver.cpp:237] Train net output #0: loss = 0.08484 (* 1 = 0.08484 loss) I0407 23:36:05.826306 359 sgd_solver.cpp:105] Iteration 8424, lr = 1.47496e-05 I0407 23:36:10.771190 359 solver.cpp:218] Iteration 8436 (2.42675 iter/s, 4.94488s/12 iters), loss = 0.118401 I0407 23:36:10.771235 359 solver.cpp:237] Train net output #0: loss = 0.118401 (* 1 = 0.118401 loss) I0407 23:36:10.771245 359 sgd_solver.cpp:105] Iteration 8436, lr = 1.44071e-05 I0407 23:36:15.706318 359 solver.cpp:218] Iteration 8448 (2.43158 iter/s, 4.93506s/12 iters), loss = 0.199423 I0407 23:36:15.706467 359 solver.cpp:237] Train net output #0: loss = 0.199423 (* 1 = 0.199423 loss) I0407 23:36:15.706477 359 sgd_solver.cpp:105] Iteration 8448, lr = 1.40725e-05 I0407 23:36:20.670230 359 solver.cpp:218] Iteration 8460 (2.41753 iter/s, 4.96375s/12 iters), loss = 0.115787 I0407 23:36:20.670269 359 solver.cpp:237] Train net output #0: loss = 0.115787 (* 1 = 0.115787 loss) I0407 23:36:20.670276 359 sgd_solver.cpp:105] Iteration 8460, lr = 1.37457e-05 I0407 23:36:22.669723 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_8466.caffemodel I0407 23:36:25.751446 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_8466.solverstate I0407 23:36:28.113206 359 solver.cpp:330] Iteration 8466, Testing net (#0) I0407 23:36:28.113232 359 net.cpp:676] Ignoring source layer train-data I0407 23:36:29.256511 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:36:32.916054 359 solver.cpp:397] Test net output #0: accuracy = 0.45098 I0407 23:36:32.916097 359 solver.cpp:397] Test net output #1: loss = 2.82356 (* 1 = 2.82356 loss) I0407 23:36:34.718864 359 solver.cpp:218] Iteration 8472 (0.85418 iter/s, 14.0486s/12 iters), loss = 0.125319 I0407 23:36:34.718904 359 solver.cpp:237] Train net output #0: loss = 0.125319 (* 1 = 0.125319 loss) I0407 23:36:34.718911 359 sgd_solver.cpp:105] Iteration 8472, lr = 1.34265e-05 I0407 23:36:39.697643 359 solver.cpp:218] Iteration 8484 (2.41026 iter/s, 4.97872s/12 iters), loss = 0.144618 I0407 23:36:39.697682 359 solver.cpp:237] Train net output #0: loss = 0.144618 (* 1 = 0.144618 loss) I0407 23:36:39.697690 359 sgd_solver.cpp:105] Iteration 8484, lr = 1.31147e-05 I0407 23:36:44.599232 359 solver.cpp:218] Iteration 8496 (2.44822 iter/s, 4.90153s/12 iters), loss = 0.069995 I0407 23:36:44.599278 359 solver.cpp:237] Train net output #0: loss = 0.0699951 (* 1 = 0.0699951 loss) I0407 23:36:44.599287 359 sgd_solver.cpp:105] Iteration 8496, lr = 1.28101e-05 I0407 23:36:44.635639 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:36:49.501340 359 solver.cpp:218] Iteration 8508 (2.44796 iter/s, 4.90204s/12 iters), loss = 0.0924 I0407 23:36:49.501523 359 solver.cpp:237] Train net output #0: loss = 0.0924001 (* 1 = 0.0924001 loss) I0407 23:36:49.501533 359 sgd_solver.cpp:105] Iteration 8508, lr = 1.25126e-05 I0407 23:36:54.462152 359 solver.cpp:218] Iteration 8520 (2.41906 iter/s, 4.96061s/12 iters), loss = 0.105946 I0407 23:36:54.462196 359 solver.cpp:237] Train net output #0: loss = 0.105946 (* 1 = 0.105946 loss) I0407 23:36:54.462204 359 sgd_solver.cpp:105] Iteration 8520, lr = 1.2222e-05 I0407 23:36:59.358388 359 solver.cpp:218] Iteration 8532 (2.45089 iter/s, 4.89617s/12 iters), loss = 0.191108 I0407 23:36:59.358426 359 solver.cpp:237] Train net output #0: loss = 0.191108 (* 1 = 0.191108 loss) I0407 23:36:59.358434 359 sgd_solver.cpp:105] Iteration 8532, lr = 1.19381e-05 I0407 23:37:04.319348 359 solver.cpp:218] Iteration 8544 (2.41892 iter/s, 4.9609s/12 iters), loss = 0.0866535 I0407 23:37:04.319396 359 solver.cpp:237] Train net output #0: loss = 0.0866536 (* 1 = 0.0866536 loss) I0407 23:37:04.319403 359 sgd_solver.cpp:105] Iteration 8544, lr = 1.16608e-05 I0407 23:37:09.266324 359 solver.cpp:218] Iteration 8556 (2.42576 iter/s, 4.9469s/12 iters), loss = 0.0468395 I0407 23:37:09.266379 359 solver.cpp:237] Train net output #0: loss = 0.0468396 (* 1 = 0.0468396 loss) I0407 23:37:09.266391 359 sgd_solver.cpp:105] Iteration 8556, lr = 1.13899e-05 I0407 23:37:13.731248 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_8568.caffemodel I0407 23:37:16.826028 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_8568.solverstate I0407 23:37:19.188450 359 solver.cpp:330] Iteration 8568, Testing net (#0) I0407 23:37:19.188469 359 net.cpp:676] Ignoring source layer train-data I0407 23:37:20.282374 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:37:23.907529 359 solver.cpp:397] Test net output #0: accuracy = 0.451593 I0407 23:37:23.907574 359 solver.cpp:397] Test net output #1: loss = 2.83594 (* 1 = 2.83594 loss) I0407 23:37:24.004022 359 solver.cpp:218] Iteration 8568 (0.814243 iter/s, 14.7376s/12 iters), loss = 0.0781373 I0407 23:37:24.004073 359 solver.cpp:237] Train net output #0: loss = 0.0781374 (* 1 = 0.0781374 loss) I0407 23:37:24.004081 359 sgd_solver.cpp:105] Iteration 8568, lr = 1.11254e-05 I0407 23:37:28.215729 359 solver.cpp:218] Iteration 8580 (2.84925 iter/s, 4.21164s/12 iters), loss = 0.0775844 I0407 23:37:28.215766 359 solver.cpp:237] Train net output #0: loss = 0.0775845 (* 1 = 0.0775845 loss) I0407 23:37:28.215773 359 sgd_solver.cpp:105] Iteration 8580, lr = 1.08669e-05 I0407 23:37:33.258900 359 solver.cpp:218] Iteration 8592 (2.37948 iter/s, 5.04311s/12 iters), loss = 0.110503 I0407 23:37:33.258944 359 solver.cpp:237] Train net output #0: loss = 0.110503 (* 1 = 0.110503 loss) I0407 23:37:33.258953 359 sgd_solver.cpp:105] Iteration 8592, lr = 1.06145e-05 I0407 23:37:35.387615 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:37:38.152248 359 solver.cpp:218] Iteration 8604 (2.45234 iter/s, 4.89328s/12 iters), loss = 0.165899 I0407 23:37:38.152295 359 solver.cpp:237] Train net output #0: loss = 0.165899 (* 1 = 0.165899 loss) I0407 23:37:38.152303 359 sgd_solver.cpp:105] Iteration 8604, lr = 1.03679e-05 I0407 23:37:43.119252 359 solver.cpp:218] Iteration 8616 (2.41598 iter/s, 4.96693s/12 iters), loss = 0.0955551 I0407 23:37:43.119297 359 solver.cpp:237] Train net output #0: loss = 0.0955551 (* 1 = 0.0955551 loss) I0407 23:37:43.119304 359 sgd_solver.cpp:105] Iteration 8616, lr = 1.0127e-05 I0407 23:37:48.027545 359 solver.cpp:218] Iteration 8628 (2.44487 iter/s, 4.90823s/12 iters), loss = 0.0759271 I0407 23:37:48.027586 359 solver.cpp:237] Train net output #0: loss = 0.0759272 (* 1 = 0.0759272 loss) I0407 23:37:48.027595 359 sgd_solver.cpp:105] Iteration 8628, lr = 9.89177e-06 I0407 23:37:52.936188 359 solver.cpp:218] Iteration 8640 (2.4447 iter/s, 4.90858s/12 iters), loss = 0.0806167 I0407 23:37:52.936350 359 solver.cpp:237] Train net output #0: loss = 0.0806167 (* 1 = 0.0806167 loss) I0407 23:37:52.936360 359 sgd_solver.cpp:105] Iteration 8640, lr = 9.66196e-06 I0407 23:37:57.858758 359 solver.cpp:218] Iteration 8652 (2.43784 iter/s, 4.92238s/12 iters), loss = 0.154482 I0407 23:37:57.858803 359 solver.cpp:237] Train net output #0: loss = 0.154482 (* 1 = 0.154482 loss) I0407 23:37:57.858810 359 sgd_solver.cpp:105] Iteration 8652, lr = 9.43749e-06 I0407 23:38:02.821100 359 solver.cpp:218] Iteration 8664 (2.41824 iter/s, 4.96228s/12 iters), loss = 0.0870906 I0407 23:38:02.821141 359 solver.cpp:237] Train net output #0: loss = 0.0870906 (* 1 = 0.0870906 loss) I0407 23:38:02.821148 359 sgd_solver.cpp:105] Iteration 8664, lr = 9.21823e-06 I0407 23:38:04.811540 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_8670.caffemodel I0407 23:38:07.887387 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_8670.solverstate I0407 23:38:10.309214 359 solver.cpp:330] Iteration 8670, Testing net (#0) I0407 23:38:10.309235 359 net.cpp:676] Ignoring source layer train-data I0407 23:38:11.371522 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:38:15.141376 359 solver.cpp:397] Test net output #0: accuracy = 0.449755 I0407 23:38:15.141422 359 solver.cpp:397] Test net output #1: loss = 2.82661 (* 1 = 2.82661 loss) I0407 23:38:16.930375 359 solver.cpp:218] Iteration 8676 (0.850509 iter/s, 14.1092s/12 iters), loss = 0.0449423 I0407 23:38:16.930423 359 solver.cpp:237] Train net output #0: loss = 0.0449423 (* 1 = 0.0449423 loss) I0407 23:38:16.930431 359 sgd_solver.cpp:105] Iteration 8676, lr = 9.00405e-06 I0407 23:38:21.901751 359 solver.cpp:218] Iteration 8688 (2.41385 iter/s, 4.97131s/12 iters), loss = 0.0630664 I0407 23:38:21.901794 359 solver.cpp:237] Train net output #0: loss = 0.0630664 (* 1 = 0.0630664 loss) I0407 23:38:21.901803 359 sgd_solver.cpp:105] Iteration 8688, lr = 8.79485e-06 I0407 23:38:26.116315 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:38:26.788120 359 solver.cpp:218] Iteration 8700 (2.45584 iter/s, 4.88631s/12 iters), loss = 0.0818737 I0407 23:38:26.788162 359 solver.cpp:237] Train net output #0: loss = 0.0818738 (* 1 = 0.0818738 loss) I0407 23:38:26.788172 359 sgd_solver.cpp:105] Iteration 8700, lr = 8.5905e-06 I0407 23:38:31.746891 359 solver.cpp:218] Iteration 8712 (2.41999 iter/s, 4.9587s/12 iters), loss = 0.0961585 I0407 23:38:31.746935 359 solver.cpp:237] Train net output #0: loss = 0.0961586 (* 1 = 0.0961586 loss) I0407 23:38:31.746944 359 sgd_solver.cpp:105] Iteration 8712, lr = 8.3909e-06 I0407 23:38:36.654266 359 solver.cpp:218] Iteration 8724 (2.44533 iter/s, 4.90731s/12 iters), loss = 0.188736 I0407 23:38:36.654305 359 solver.cpp:237] Train net output #0: loss = 0.188736 (* 1 = 0.188736 loss) I0407 23:38:36.654314 359 sgd_solver.cpp:105] Iteration 8724, lr = 8.19593e-06 I0407 23:38:41.599942 359 solver.cpp:218] Iteration 8736 (2.42639 iter/s, 4.94562s/12 iters), loss = 0.175551 I0407 23:38:41.599982 359 solver.cpp:237] Train net output #0: loss = 0.175551 (* 1 = 0.175551 loss) I0407 23:38:41.599989 359 sgd_solver.cpp:105] Iteration 8736, lr = 8.00549e-06 I0407 23:38:46.522909 359 solver.cpp:218] Iteration 8748 (2.43758 iter/s, 4.92291s/12 iters), loss = 0.283982 I0407 23:38:46.522948 359 solver.cpp:237] Train net output #0: loss = 0.283982 (* 1 = 0.283982 loss) I0407 23:38:46.522954 359 sgd_solver.cpp:105] Iteration 8748, lr = 7.81947e-06 I0407 23:38:51.479146 359 solver.cpp:218] Iteration 8760 (2.42122 iter/s, 4.95618s/12 iters), loss = 0.11002 I0407 23:38:51.479183 359 solver.cpp:237] Train net output #0: loss = 0.11002 (* 1 = 0.11002 loss) I0407 23:38:51.479192 359 sgd_solver.cpp:105] Iteration 8760, lr = 7.63777e-06 I0407 23:38:55.937175 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_8772.caffemodel I0407 23:38:59.037871 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_8772.solverstate I0407 23:39:01.429013 359 solver.cpp:330] Iteration 8772, Testing net (#0) I0407 23:39:01.429029 359 net.cpp:676] Ignoring source layer train-data I0407 23:39:02.386000 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:39:05.864818 359 solver.cpp:397] Test net output #0: accuracy = 0.448529 I0407 23:39:05.864850 359 solver.cpp:397] Test net output #1: loss = 2.83201 (* 1 = 2.83201 loss) I0407 23:39:05.961374 359 solver.cpp:218] Iteration 8772 (0.828607 iter/s, 14.4821s/12 iters), loss = 0.0881024 I0407 23:39:05.961450 359 solver.cpp:237] Train net output #0: loss = 0.0881025 (* 1 = 0.0881025 loss) I0407 23:39:05.961467 359 sgd_solver.cpp:105] Iteration 8772, lr = 7.46029e-06 I0407 23:39:10.080976 359 solver.cpp:218] Iteration 8784 (2.91297 iter/s, 4.11951s/12 iters), loss = 0.162071 I0407 23:39:10.081019 359 solver.cpp:237] Train net output #0: loss = 0.162071 (* 1 = 0.162071 loss) I0407 23:39:10.081027 359 sgd_solver.cpp:105] Iteration 8784, lr = 7.28692e-06 I0407 23:39:15.019028 359 solver.cpp:218] Iteration 8796 (2.43014 iter/s, 4.93799s/12 iters), loss = 0.136573 I0407 23:39:15.019073 359 solver.cpp:237] Train net output #0: loss = 0.136573 (* 1 = 0.136573 loss) I0407 23:39:15.019083 359 sgd_solver.cpp:105] Iteration 8796, lr = 7.11759e-06 I0407 23:39:16.425887 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:39:19.914045 359 solver.cpp:218] Iteration 8808 (2.45151 iter/s, 4.89495s/12 iters), loss = 0.0656361 I0407 23:39:19.914089 359 solver.cpp:237] Train net output #0: loss = 0.0656361 (* 1 = 0.0656361 loss) I0407 23:39:19.914098 359 sgd_solver.cpp:105] Iteration 8808, lr = 6.95219e-06 I0407 23:39:24.690186 359 solver.cpp:218] Iteration 8820 (2.51252 iter/s, 4.77608s/12 iters), loss = 0.0844655 I0407 23:39:24.690224 359 solver.cpp:237] Train net output #0: loss = 0.0844655 (* 1 = 0.0844655 loss) I0407 23:39:24.690232 359 sgd_solver.cpp:105] Iteration 8820, lr = 6.79063e-06 I0407 23:39:29.608309 359 solver.cpp:218] Iteration 8832 (2.43998 iter/s, 4.91807s/12 iters), loss = 0.190377 I0407 23:39:29.608433 359 solver.cpp:237] Train net output #0: loss = 0.190377 (* 1 = 0.190377 loss) I0407 23:39:29.608441 359 sgd_solver.cpp:105] Iteration 8832, lr = 6.63282e-06 I0407 23:39:34.551301 359 solver.cpp:218] Iteration 8844 (2.42775 iter/s, 4.94284s/12 iters), loss = 0.121329 I0407 23:39:34.551345 359 solver.cpp:237] Train net output #0: loss = 0.121329 (* 1 = 0.121329 loss) I0407 23:39:34.551354 359 sgd_solver.cpp:105] Iteration 8844, lr = 6.47867e-06 I0407 23:39:39.516156 359 solver.cpp:218] Iteration 8856 (2.41702 iter/s, 4.96479s/12 iters), loss = 0.157651 I0407 23:39:39.516202 359 solver.cpp:237] Train net output #0: loss = 0.157651 (* 1 = 0.157651 loss) I0407 23:39:39.516211 359 sgd_solver.cpp:105] Iteration 8856, lr = 6.3281e-06 I0407 23:39:44.451517 359 solver.cpp:218] Iteration 8868 (2.43147 iter/s, 4.93529s/12 iters), loss = 0.149847 I0407 23:39:44.451558 359 solver.cpp:237] Train net output #0: loss = 0.149847 (* 1 = 0.149847 loss) I0407 23:39:44.451567 359 sgd_solver.cpp:105] Iteration 8868, lr = 6.18104e-06 I0407 23:39:46.433789 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_8874.caffemodel I0407 23:39:50.187085 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_8874.solverstate I0407 23:39:52.551925 359 solver.cpp:330] Iteration 8874, Testing net (#0) I0407 23:39:52.551944 359 net.cpp:676] Ignoring source layer train-data I0407 23:39:53.509279 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:39:57.332002 359 solver.cpp:397] Test net output #0: accuracy = 0.45098 I0407 23:39:57.332049 359 solver.cpp:397] Test net output #1: loss = 2.8289 (* 1 = 2.8289 loss) I0407 23:39:59.136139 359 solver.cpp:218] Iteration 8880 (0.817186 iter/s, 14.6845s/12 iters), loss = 0.102348 I0407 23:39:59.136188 359 solver.cpp:237] Train net output #0: loss = 0.102348 (* 1 = 0.102348 loss) I0407 23:39:59.136198 359 sgd_solver.cpp:105] Iteration 8880, lr = 6.03739e-06 I0407 23:40:04.090639 359 solver.cpp:218] Iteration 8892 (2.42207 iter/s, 4.95443s/12 iters), loss = 0.0871479 I0407 23:40:04.090806 359 solver.cpp:237] Train net output #0: loss = 0.087148 (* 1 = 0.087148 loss) I0407 23:40:04.090816 359 sgd_solver.cpp:105] Iteration 8892, lr = 5.89707e-06 I0407 23:40:07.591856 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:40:08.971700 359 solver.cpp:218] Iteration 8904 (2.45857 iter/s, 4.88088s/12 iters), loss = 0.13032 I0407 23:40:08.971735 359 solver.cpp:237] Train net output #0: loss = 0.13032 (* 1 = 0.13032 loss) I0407 23:40:08.971743 359 sgd_solver.cpp:105] Iteration 8904, lr = 5.76001e-06 I0407 23:40:13.937816 359 solver.cpp:218] Iteration 8916 (2.4164 iter/s, 4.96606s/12 iters), loss = 0.0716483 I0407 23:40:13.937858 359 solver.cpp:237] Train net output #0: loss = 0.0716483 (* 1 = 0.0716483 loss) I0407 23:40:13.937867 359 sgd_solver.cpp:105] Iteration 8916, lr = 5.62614e-06 I0407 23:40:18.847456 359 solver.cpp:218] Iteration 8928 (2.4442 iter/s, 4.90958s/12 iters), loss = 0.119038 I0407 23:40:18.847499 359 solver.cpp:237] Train net output #0: loss = 0.119038 (* 1 = 0.119038 loss) I0407 23:40:18.847508 359 sgd_solver.cpp:105] Iteration 8928, lr = 5.49538e-06 I0407 23:40:23.762256 359 solver.cpp:218] Iteration 8940 (2.44164 iter/s, 4.91474s/12 iters), loss = 0.0533999 I0407 23:40:23.762291 359 solver.cpp:237] Train net output #0: loss = 0.0533999 (* 1 = 0.0533999 loss) I0407 23:40:23.762298 359 sgd_solver.cpp:105] Iteration 8940, lr = 5.36766e-06 I0407 23:40:28.687304 359 solver.cpp:218] Iteration 8952 (2.43655 iter/s, 4.92499s/12 iters), loss = 0.0766114 I0407 23:40:28.687340 359 solver.cpp:237] Train net output #0: loss = 0.0766115 (* 1 = 0.0766115 loss) I0407 23:40:28.687348 359 sgd_solver.cpp:105] Iteration 8952, lr = 5.2429e-06 I0407 23:40:33.637977 359 solver.cpp:218] Iteration 8964 (2.42394 iter/s, 4.95061s/12 iters), loss = 0.153665 I0407 23:40:33.638021 359 solver.cpp:237] Train net output #0: loss = 0.153665 (* 1 = 0.153665 loss) I0407 23:40:33.638029 359 sgd_solver.cpp:105] Iteration 8964, lr = 5.12104e-06 I0407 23:40:38.039327 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_8976.caffemodel I0407 23:40:41.170154 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_8976.solverstate I0407 23:40:43.534703 359 solver.cpp:330] Iteration 8976, Testing net (#0) I0407 23:40:43.534723 359 net.cpp:676] Ignoring source layer train-data I0407 23:40:44.420064 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:40:47.966501 359 solver.cpp:397] Test net output #0: accuracy = 0.447304 I0407 23:40:47.966533 359 solver.cpp:397] Test net output #1: loss = 2.84442 (* 1 = 2.84442 loss) I0407 23:40:48.063611 359 solver.cpp:218] Iteration 8976 (0.831857 iter/s, 14.4255s/12 iters), loss = 0.0681887 I0407 23:40:48.063649 359 solver.cpp:237] Train net output #0: loss = 0.0681888 (* 1 = 0.0681888 loss) I0407 23:40:48.063657 359 sgd_solver.cpp:105] Iteration 8976, lr = 5.00201e-06 I0407 23:40:52.189750 359 solver.cpp:218] Iteration 8988 (2.90833 iter/s, 4.12608s/12 iters), loss = 0.0454001 I0407 23:40:52.189787 359 solver.cpp:237] Train net output #0: loss = 0.0454001 (* 1 = 0.0454001 loss) I0407 23:40:52.189795 359 sgd_solver.cpp:105] Iteration 8988, lr = 4.88574e-06 I0407 23:40:55.404886 359 blocking_queue.cpp:49] Waiting for data I0407 23:40:57.128886 359 solver.cpp:218] Iteration 9000 (2.4296 iter/s, 4.93908s/12 iters), loss = 0.0909112 I0407 23:40:57.128926 359 solver.cpp:237] Train net output #0: loss = 0.0909113 (* 1 = 0.0909113 loss) I0407 23:40:57.128933 359 sgd_solver.cpp:105] Iteration 9000, lr = 4.77218e-06 I0407 23:40:57.822428 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:41:02.033378 359 solver.cpp:218] Iteration 9012 (2.44676 iter/s, 4.90444s/12 iters), loss = 0.107862 I0407 23:41:02.033414 359 solver.cpp:237] Train net output #0: loss = 0.107862 (* 1 = 0.107862 loss) I0407 23:41:02.033421 359 sgd_solver.cpp:105] Iteration 9012, lr = 4.66126e-06 I0407 23:41:06.955420 359 solver.cpp:218] Iteration 9024 (2.43804 iter/s, 4.92198s/12 iters), loss = 0.208127 I0407 23:41:06.955462 359 solver.cpp:237] Train net output #0: loss = 0.208127 (* 1 = 0.208127 loss) I0407 23:41:06.955471 359 sgd_solver.cpp:105] Iteration 9024, lr = 4.55291e-06 I0407 23:41:11.937781 359 solver.cpp:218] Iteration 9036 (2.40853 iter/s, 4.9823s/12 iters), loss = 0.114744 I0407 23:41:11.937922 359 solver.cpp:237] Train net output #0: loss = 0.114745 (* 1 = 0.114745 loss) I0407 23:41:11.937930 359 sgd_solver.cpp:105] Iteration 9036, lr = 4.44708e-06 I0407 23:41:16.900611 359 solver.cpp:218] Iteration 9048 (2.41805 iter/s, 4.96268s/12 iters), loss = 0.113315 I0407 23:41:16.900647 359 solver.cpp:237] Train net output #0: loss = 0.113315 (* 1 = 0.113315 loss) I0407 23:41:16.900655 359 sgd_solver.cpp:105] Iteration 9048, lr = 4.34371e-06 I0407 23:41:21.853181 359 solver.cpp:218] Iteration 9060 (2.42301 iter/s, 4.95251s/12 iters), loss = 0.139511 I0407 23:41:21.853236 359 solver.cpp:237] Train net output #0: loss = 0.139511 (* 1 = 0.139511 loss) I0407 23:41:21.853250 359 sgd_solver.cpp:105] Iteration 9060, lr = 4.24274e-06 I0407 23:41:26.844935 359 solver.cpp:218] Iteration 9072 (2.404 iter/s, 4.99168s/12 iters), loss = 0.0976321 I0407 23:41:26.844978 359 solver.cpp:237] Train net output #0: loss = 0.0976322 (* 1 = 0.0976322 loss) I0407 23:41:26.844986 359 sgd_solver.cpp:105] Iteration 9072, lr = 4.14412e-06 I0407 23:41:28.838894 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_9078.caffemodel I0407 23:41:31.946161 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_9078.solverstate I0407 23:41:34.308960 359 solver.cpp:330] Iteration 9078, Testing net (#0) I0407 23:41:34.308980 359 net.cpp:676] Ignoring source layer train-data I0407 23:41:35.104416 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:41:38.670513 359 solver.cpp:397] Test net output #0: accuracy = 0.448529 I0407 23:41:38.670555 359 solver.cpp:397] Test net output #1: loss = 2.83304 (* 1 = 2.83304 loss) I0407 23:41:40.500010 359 solver.cpp:218] Iteration 9084 (0.878799 iter/s, 13.655s/12 iters), loss = 0.118667 I0407 23:41:40.500056 359 solver.cpp:237] Train net output #0: loss = 0.118667 (* 1 = 0.118667 loss) I0407 23:41:40.500064 359 sgd_solver.cpp:105] Iteration 9084, lr = 4.04779e-06 I0407 23:41:45.462385 359 solver.cpp:218] Iteration 9096 (2.41823 iter/s, 4.96231s/12 iters), loss = 0.12539 I0407 23:41:45.462509 359 solver.cpp:237] Train net output #0: loss = 0.12539 (* 1 = 0.12539 loss) I0407 23:41:45.462517 359 sgd_solver.cpp:105] Iteration 9096, lr = 3.95369e-06 I0407 23:41:48.343343 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:41:50.357234 359 solver.cpp:218] Iteration 9108 (2.45163 iter/s, 4.89471s/12 iters), loss = 0.0646588 I0407 23:41:50.357275 359 solver.cpp:237] Train net output #0: loss = 0.0646588 (* 1 = 0.0646588 loss) I0407 23:41:50.357282 359 sgd_solver.cpp:105] Iteration 9108, lr = 3.86179e-06 I0407 23:41:55.320041 359 solver.cpp:218] Iteration 9120 (2.41802 iter/s, 4.96275s/12 iters), loss = 0.17825 I0407 23:41:55.320080 359 solver.cpp:237] Train net output #0: loss = 0.17825 (* 1 = 0.17825 loss) I0407 23:41:55.320088 359 sgd_solver.cpp:105] Iteration 9120, lr = 3.77202e-06 I0407 23:42:00.233377 359 solver.cpp:218] Iteration 9132 (2.44236 iter/s, 4.91328s/12 iters), loss = 0.0906506 I0407 23:42:00.233415 359 solver.cpp:237] Train net output #0: loss = 0.0906506 (* 1 = 0.0906506 loss) I0407 23:42:00.233423 359 sgd_solver.cpp:105] Iteration 9132, lr = 3.68433e-06 I0407 23:42:05.177191 359 solver.cpp:218] Iteration 9144 (2.4273 iter/s, 4.94376s/12 iters), loss = 0.0635827 I0407 23:42:05.177232 359 solver.cpp:237] Train net output #0: loss = 0.0635828 (* 1 = 0.0635828 loss) I0407 23:42:05.177240 359 sgd_solver.cpp:105] Iteration 9144, lr = 3.59868e-06 I0407 23:42:10.095171 359 solver.cpp:218] Iteration 9156 (2.44006 iter/s, 4.91791s/12 iters), loss = 0.122193 I0407 23:42:10.095223 359 solver.cpp:237] Train net output #0: loss = 0.122193 (* 1 = 0.122193 loss) I0407 23:42:10.095233 359 sgd_solver.cpp:105] Iteration 9156, lr = 3.51503e-06 I0407 23:42:15.059562 359 solver.cpp:218] Iteration 9168 (2.41725 iter/s, 4.96432s/12 iters), loss = 0.117753 I0407 23:42:15.059609 359 solver.cpp:237] Train net output #0: loss = 0.117753 (* 1 = 0.117753 loss) I0407 23:42:15.059623 359 sgd_solver.cpp:105] Iteration 9168, lr = 3.43331e-06 I0407 23:42:19.603961 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_9180.caffemodel I0407 23:42:22.757802 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_9180.solverstate I0407 23:42:25.131599 359 solver.cpp:330] Iteration 9180, Testing net (#0) I0407 23:42:25.131618 359 net.cpp:676] Ignoring source layer train-data I0407 23:42:25.961233 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:42:29.925479 359 solver.cpp:397] Test net output #0: accuracy = 0.45098 I0407 23:42:29.925540 359 solver.cpp:397] Test net output #1: loss = 2.83207 (* 1 = 2.83207 loss) I0407 23:42:30.021684 359 solver.cpp:218] Iteration 9180 (0.802029 iter/s, 14.962s/12 iters), loss = 0.0381248 I0407 23:42:30.021733 359 solver.cpp:237] Train net output #0: loss = 0.0381248 (* 1 = 0.0381248 loss) I0407 23:42:30.021741 359 sgd_solver.cpp:105] Iteration 9180, lr = 3.3535e-06 I0407 23:42:34.232491 359 solver.cpp:218] Iteration 9192 (2.84986 iter/s, 4.21074s/12 iters), loss = 0.0942718 I0407 23:42:34.232535 359 solver.cpp:237] Train net output #0: loss = 0.0942719 (* 1 = 0.0942719 loss) I0407 23:42:34.232543 359 sgd_solver.cpp:105] Iteration 9192, lr = 3.27554e-06 I0407 23:42:39.242657 359 solver.cpp:218] Iteration 9204 (2.39516 iter/s, 5.0101s/12 iters), loss = 0.169785 I0407 23:42:39.242695 359 solver.cpp:237] Train net output #0: loss = 0.169785 (* 1 = 0.169785 loss) I0407 23:42:39.242703 359 sgd_solver.cpp:105] Iteration 9204, lr = 3.19939e-06 I0407 23:42:39.316300 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:42:44.217150 359 solver.cpp:218] Iteration 9216 (2.41233 iter/s, 4.97444s/12 iters), loss = 0.100025 I0407 23:42:44.217187 359 solver.cpp:237] Train net output #0: loss = 0.100025 (* 1 = 0.100025 loss) I0407 23:42:44.217195 359 sgd_solver.cpp:105] Iteration 9216, lr = 3.12501e-06 I0407 23:42:49.171494 359 solver.cpp:218] Iteration 9228 (2.42215 iter/s, 4.95429s/12 iters), loss = 0.11473 I0407 23:42:49.171535 359 solver.cpp:237] Train net output #0: loss = 0.11473 (* 1 = 0.11473 loss) I0407 23:42:49.171541 359 sgd_solver.cpp:105] Iteration 9228, lr = 3.05237e-06 I0407 23:42:54.106451 359 solver.cpp:218] Iteration 9240 (2.43166 iter/s, 4.9349s/12 iters), loss = 0.0410729 I0407 23:42:54.106573 359 solver.cpp:237] Train net output #0: loss = 0.0410729 (* 1 = 0.0410729 loss) I0407 23:42:54.106582 359 sgd_solver.cpp:105] Iteration 9240, lr = 2.98141e-06 I0407 23:42:59.078850 359 solver.cpp:218] Iteration 9252 (2.41339 iter/s, 4.97226s/12 iters), loss = 0.0620218 I0407 23:42:59.078891 359 solver.cpp:237] Train net output #0: loss = 0.0620219 (* 1 = 0.0620219 loss) I0407 23:42:59.078899 359 sgd_solver.cpp:105] Iteration 9252, lr = 2.91209e-06 I0407 23:43:04.040300 359 solver.cpp:218] Iteration 9264 (2.41868 iter/s, 4.96139s/12 iters), loss = 0.0994477 I0407 23:43:04.040338 359 solver.cpp:237] Train net output #0: loss = 0.0994478 (* 1 = 0.0994478 loss) I0407 23:43:04.040345 359 sgd_solver.cpp:105] Iteration 9264, lr = 2.84439e-06 I0407 23:43:09.029474 359 solver.cpp:218] Iteration 9276 (2.40524 iter/s, 4.98911s/12 iters), loss = 0.273761 I0407 23:43:09.029520 359 solver.cpp:237] Train net output #0: loss = 0.273761 (* 1 = 0.273761 loss) I0407 23:43:09.029527 359 sgd_solver.cpp:105] Iteration 9276, lr = 2.77826e-06 I0407 23:43:11.017052 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_9282.caffemodel I0407 23:43:14.122756 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_9282.solverstate I0407 23:43:16.488492 359 solver.cpp:330] Iteration 9282, Testing net (#0) I0407 23:43:16.488510 359 net.cpp:676] Ignoring source layer train-data I0407 23:43:17.266289 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:43:21.331053 359 solver.cpp:397] Test net output #0: accuracy = 0.449755 I0407 23:43:21.331094 359 solver.cpp:397] Test net output #1: loss = 2.83806 (* 1 = 2.83806 loss) I0407 23:43:23.131387 359 solver.cpp:218] Iteration 9288 (0.850953 iter/s, 14.1018s/12 iters), loss = 0.116045 I0407 23:43:23.131428 359 solver.cpp:237] Train net output #0: loss = 0.116045 (* 1 = 0.116045 loss) I0407 23:43:23.131435 359 sgd_solver.cpp:105] Iteration 9288, lr = 2.71367e-06 I0407 23:43:28.078056 359 solver.cpp:218] Iteration 9300 (2.42591 iter/s, 4.94661s/12 iters), loss = 0.1753 I0407 23:43:28.078198 359 solver.cpp:237] Train net output #0: loss = 0.1753 (* 1 = 0.1753 loss) I0407 23:43:28.078207 359 sgd_solver.cpp:105] Iteration 9300, lr = 2.65059e-06 I0407 23:43:30.237421 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:43:32.972517 359 solver.cpp:218] Iteration 9312 (2.45183 iter/s, 4.8943s/12 iters), loss = 0.166904 I0407 23:43:32.972554 359 solver.cpp:237] Train net output #0: loss = 0.166904 (* 1 = 0.166904 loss) I0407 23:43:32.972563 359 sgd_solver.cpp:105] Iteration 9312, lr = 2.58896e-06 I0407 23:43:37.932895 359 solver.cpp:218] Iteration 9324 (2.4192 iter/s, 4.96032s/12 iters), loss = 0.124534 I0407 23:43:37.932934 359 solver.cpp:237] Train net output #0: loss = 0.124534 (* 1 = 0.124534 loss) I0407 23:43:37.932942 359 sgd_solver.cpp:105] Iteration 9324, lr = 2.52877e-06 I0407 23:43:42.845811 359 solver.cpp:218] Iteration 9336 (2.44257 iter/s, 4.91286s/12 iters), loss = 0.141922 I0407 23:43:42.845849 359 solver.cpp:237] Train net output #0: loss = 0.141922 (* 1 = 0.141922 loss) I0407 23:43:42.845856 359 sgd_solver.cpp:105] Iteration 9336, lr = 2.46998e-06 I0407 23:43:47.794771 359 solver.cpp:218] Iteration 9348 (2.42478 iter/s, 4.9489s/12 iters), loss = 0.112526 I0407 23:43:47.794811 359 solver.cpp:237] Train net output #0: loss = 0.112526 (* 1 = 0.112526 loss) I0407 23:43:47.794818 359 sgd_solver.cpp:105] Iteration 9348, lr = 2.41256e-06 I0407 23:43:52.770589 359 solver.cpp:218] Iteration 9360 (2.41169 iter/s, 4.97576s/12 iters), loss = 0.206143 I0407 23:43:52.770627 359 solver.cpp:237] Train net output #0: loss = 0.206143 (* 1 = 0.206143 loss) I0407 23:43:52.770634 359 sgd_solver.cpp:105] Iteration 9360, lr = 2.35647e-06 I0407 23:43:57.705350 359 solver.cpp:218] Iteration 9372 (2.43176 iter/s, 4.93471s/12 iters), loss = 0.0886094 I0407 23:43:57.705386 359 solver.cpp:237] Train net output #0: loss = 0.0886095 (* 1 = 0.0886095 loss) I0407 23:43:57.705394 359 sgd_solver.cpp:105] Iteration 9372, lr = 2.30168e-06 I0407 23:44:02.185900 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_9384.caffemodel I0407 23:44:05.282016 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_9384.solverstate I0407 23:44:07.648779 359 solver.cpp:330] Iteration 9384, Testing net (#0) I0407 23:44:07.648798 359 net.cpp:676] Ignoring source layer train-data I0407 23:44:08.388518 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:44:12.408736 359 solver.cpp:397] Test net output #0: accuracy = 0.451593 I0407 23:44:12.408782 359 solver.cpp:397] Test net output #1: loss = 2.82807 (* 1 = 2.82807 loss) I0407 23:44:12.505198 359 solver.cpp:218] Iteration 9384 (0.810823 iter/s, 14.7998s/12 iters), loss = 0.14763 I0407 23:44:12.505244 359 solver.cpp:237] Train net output #0: loss = 0.14763 (* 1 = 0.14763 loss) I0407 23:44:12.505251 359 sgd_solver.cpp:105] Iteration 9384, lr = 2.24817e-06 I0407 23:44:16.570714 359 solver.cpp:218] Iteration 9396 (2.9517 iter/s, 4.06545s/12 iters), loss = 0.164532 I0407 23:44:16.570756 359 solver.cpp:237] Train net output #0: loss = 0.164532 (* 1 = 0.164532 loss) I0407 23:44:16.570765 359 sgd_solver.cpp:105] Iteration 9396, lr = 2.1959e-06 I0407 23:44:20.831952 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:44:21.499477 359 solver.cpp:218] Iteration 9408 (2.43472 iter/s, 4.9287s/12 iters), loss = 0.0997483 I0407 23:44:21.499524 359 solver.cpp:237] Train net output #0: loss = 0.0997483 (* 1 = 0.0997483 loss) I0407 23:44:21.499533 359 sgd_solver.cpp:105] Iteration 9408, lr = 2.14484e-06 I0407 23:44:26.424047 359 solver.cpp:218] Iteration 9420 (2.43679 iter/s, 4.9245s/12 iters), loss = 0.164456 I0407 23:44:26.424093 359 solver.cpp:237] Train net output #0: loss = 0.164456 (* 1 = 0.164456 loss) I0407 23:44:26.424103 359 sgd_solver.cpp:105] Iteration 9420, lr = 2.09498e-06 I0407 23:44:31.365551 359 solver.cpp:218] Iteration 9432 (2.42844 iter/s, 4.94144s/12 iters), loss = 0.0614223 I0407 23:44:31.365597 359 solver.cpp:237] Train net output #0: loss = 0.0614224 (* 1 = 0.0614224 loss) I0407 23:44:31.365605 359 sgd_solver.cpp:105] Iteration 9432, lr = 2.04627e-06 I0407 23:44:36.270550 359 solver.cpp:218] Iteration 9444 (2.44652 iter/s, 4.90493s/12 iters), loss = 0.179541 I0407 23:44:36.270721 359 solver.cpp:237] Train net output #0: loss = 0.179541 (* 1 = 0.179541 loss) I0407 23:44:36.270731 359 sgd_solver.cpp:105] Iteration 9444, lr = 1.99869e-06 I0407 23:44:41.219415 359 solver.cpp:218] Iteration 9456 (2.4249 iter/s, 4.94867s/12 iters), loss = 0.168537 I0407 23:44:41.219463 359 solver.cpp:237] Train net output #0: loss = 0.168537 (* 1 = 0.168537 loss) I0407 23:44:41.219471 359 sgd_solver.cpp:105] Iteration 9456, lr = 1.95222e-06 I0407 23:44:46.137250 359 solver.cpp:218] Iteration 9468 (2.44013 iter/s, 4.91777s/12 iters), loss = 0.0856009 I0407 23:44:46.137285 359 solver.cpp:237] Train net output #0: loss = 0.085601 (* 1 = 0.085601 loss) I0407 23:44:46.137293 359 sgd_solver.cpp:105] Iteration 9468, lr = 1.90683e-06 I0407 23:44:51.092250 359 solver.cpp:218] Iteration 9480 (2.42182 iter/s, 4.95495s/12 iters), loss = 0.163074 I0407 23:44:51.092288 359 solver.cpp:237] Train net output #0: loss = 0.163074 (* 1 = 0.163074 loss) I0407 23:44:51.092295 359 sgd_solver.cpp:105] Iteration 9480, lr = 1.8625e-06 I0407 23:44:53.079025 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_9486.caffemodel I0407 23:44:56.168238 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_9486.solverstate I0407 23:44:58.536355 359 solver.cpp:330] Iteration 9486, Testing net (#0) I0407 23:44:58.536373 359 net.cpp:676] Ignoring source layer train-data I0407 23:44:59.243846 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:45:03.317701 359 solver.cpp:397] Test net output #0: accuracy = 0.449142 I0407 23:45:03.317749 359 solver.cpp:397] Test net output #1: loss = 2.83584 (* 1 = 2.83584 loss) I0407 23:45:05.141343 359 solver.cpp:218] Iteration 9492 (0.854152 iter/s, 14.049s/12 iters), loss = 0.10268 I0407 23:45:05.141388 359 solver.cpp:237] Train net output #0: loss = 0.10268 (* 1 = 0.10268 loss) I0407 23:45:05.141396 359 sgd_solver.cpp:105] Iteration 9492, lr = 1.81919e-06 I0407 23:45:10.049109 359 solver.cpp:218] Iteration 9504 (2.44514 iter/s, 4.9077s/12 iters), loss = 0.069278 I0407 23:45:10.049254 359 solver.cpp:237] Train net output #0: loss = 0.0692781 (* 1 = 0.0692781 loss) I0407 23:45:10.049263 359 sgd_solver.cpp:105] Iteration 9504, lr = 1.7769e-06 I0407 23:45:11.483283 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:45:14.952170 359 solver.cpp:218] Iteration 9516 (2.44753 iter/s, 4.9029s/12 iters), loss = 0.0581919 I0407 23:45:14.952208 359 solver.cpp:237] Train net output #0: loss = 0.0581919 (* 1 = 0.0581919 loss) I0407 23:45:14.952216 359 sgd_solver.cpp:105] Iteration 9516, lr = 1.73558e-06 I0407 23:45:19.905041 359 solver.cpp:218] Iteration 9528 (2.42286 iter/s, 4.95282s/12 iters), loss = 0.0740821 I0407 23:45:19.905079 359 solver.cpp:237] Train net output #0: loss = 0.0740821 (* 1 = 0.0740821 loss) I0407 23:45:19.905087 359 sgd_solver.cpp:105] Iteration 9528, lr = 1.69523e-06 I0407 23:45:24.838248 359 solver.cpp:218] Iteration 9540 (2.43253 iter/s, 4.93314s/12 iters), loss = 0.13533 I0407 23:45:24.838294 359 solver.cpp:237] Train net output #0: loss = 0.13533 (* 1 = 0.13533 loss) I0407 23:45:24.838302 359 sgd_solver.cpp:105] Iteration 9540, lr = 1.65581e-06 I0407 23:45:29.755347 359 solver.cpp:218] Iteration 9552 (2.4405 iter/s, 4.91703s/12 iters), loss = 0.111031 I0407 23:45:29.755390 359 solver.cpp:237] Train net output #0: loss = 0.111031 (* 1 = 0.111031 loss) I0407 23:45:29.755398 359 sgd_solver.cpp:105] Iteration 9552, lr = 1.61731e-06 I0407 23:45:34.718720 359 solver.cpp:218] Iteration 9564 (2.41774 iter/s, 4.96331s/12 iters), loss = 0.140934 I0407 23:45:34.718760 359 solver.cpp:237] Train net output #0: loss = 0.140934 (* 1 = 0.140934 loss) I0407 23:45:34.718768 359 sgd_solver.cpp:105] Iteration 9564, lr = 1.57971e-06 I0407 23:45:39.594220 359 solver.cpp:218] Iteration 9576 (2.46132 iter/s, 4.87544s/12 iters), loss = 0.130819 I0407 23:45:39.594259 359 solver.cpp:237] Train net output #0: loss = 0.130819 (* 1 = 0.130819 loss) I0407 23:45:39.594266 359 sgd_solver.cpp:105] Iteration 9576, lr = 1.54298e-06 I0407 23:45:43.980114 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_9588.caffemodel I0407 23:45:47.073791 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_9588.solverstate I0407 23:45:49.440476 359 solver.cpp:330] Iteration 9588, Testing net (#0) I0407 23:45:49.440495 359 net.cpp:676] Ignoring source layer train-data I0407 23:45:50.103266 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:45:53.966145 359 solver.cpp:397] Test net output #0: accuracy = 0.448529 I0407 23:45:53.966192 359 solver.cpp:397] Test net output #1: loss = 2.83786 (* 1 = 2.83786 loss) I0407 23:45:54.062616 359 solver.cpp:218] Iteration 9588 (0.829398 iter/s, 14.4683s/12 iters), loss = 0.175383 I0407 23:45:54.062656 359 solver.cpp:237] Train net output #0: loss = 0.175384 (* 1 = 0.175384 loss) I0407 23:45:54.062665 359 sgd_solver.cpp:105] Iteration 9588, lr = 1.5071e-06 I0407 23:45:58.184581 359 solver.cpp:218] Iteration 9600 (2.91127 iter/s, 4.12191s/12 iters), loss = 0.142131 I0407 23:45:58.184618 359 solver.cpp:237] Train net output #0: loss = 0.142132 (* 1 = 0.142132 loss) I0407 23:45:58.184628 359 sgd_solver.cpp:105] Iteration 9600, lr = 1.47206e-06 I0407 23:46:01.747241 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:46:03.097952 359 solver.cpp:218] Iteration 9612 (2.44234 iter/s, 4.91332s/12 iters), loss = 0.1103 I0407 23:46:03.097991 359 solver.cpp:237] Train net output #0: loss = 0.1103 (* 1 = 0.1103 loss) I0407 23:46:03.097999 359 sgd_solver.cpp:105] Iteration 9612, lr = 1.43783e-06 I0407 23:46:08.020946 359 solver.cpp:218] Iteration 9624 (2.43757 iter/s, 4.92293s/12 iters), loss = 0.0665706 I0407 23:46:08.020983 359 solver.cpp:237] Train net output #0: loss = 0.0665706 (* 1 = 0.0665706 loss) I0407 23:46:08.020992 359 sgd_solver.cpp:105] Iteration 9624, lr = 1.4044e-06 I0407 23:46:12.976524 359 solver.cpp:218] Iteration 9636 (2.42154 iter/s, 4.95552s/12 iters), loss = 0.188979 I0407 23:46:12.976560 359 solver.cpp:237] Train net output #0: loss = 0.188979 (* 1 = 0.188979 loss) I0407 23:46:12.976567 359 sgd_solver.cpp:105] Iteration 9636, lr = 1.37175e-06 I0407 23:46:17.875429 359 solver.cpp:218] Iteration 9648 (2.44956 iter/s, 4.89884s/12 iters), loss = 0.0693474 I0407 23:46:17.875550 359 solver.cpp:237] Train net output #0: loss = 0.0693475 (* 1 = 0.0693475 loss) I0407 23:46:17.875560 359 sgd_solver.cpp:105] Iteration 9648, lr = 1.33985e-06 I0407 23:46:22.846493 359 solver.cpp:218] Iteration 9660 (2.41404 iter/s, 4.97092s/12 iters), loss = 0.142862 I0407 23:46:22.846540 359 solver.cpp:237] Train net output #0: loss = 0.142862 (* 1 = 0.142862 loss) I0407 23:46:22.846549 359 sgd_solver.cpp:105] Iteration 9660, lr = 1.3087e-06 I0407 23:46:27.756002 359 solver.cpp:218] Iteration 9672 (2.44427 iter/s, 4.90944s/12 iters), loss = 0.159316 I0407 23:46:27.756042 359 solver.cpp:237] Train net output #0: loss = 0.159316 (* 1 = 0.159316 loss) I0407 23:46:27.756050 359 sgd_solver.cpp:105] Iteration 9672, lr = 1.27827e-06 I0407 23:46:32.700523 359 solver.cpp:218] Iteration 9684 (2.42696 iter/s, 4.94446s/12 iters), loss = 0.182466 I0407 23:46:32.700558 359 solver.cpp:237] Train net output #0: loss = 0.182466 (* 1 = 0.182466 loss) I0407 23:46:32.700567 359 sgd_solver.cpp:105] Iteration 9684, lr = 1.24855e-06 I0407 23:46:34.698504 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_9690.caffemodel I0407 23:46:37.793181 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_9690.solverstate I0407 23:46:40.155972 359 solver.cpp:330] Iteration 9690, Testing net (#0) I0407 23:46:40.155987 359 net.cpp:676] Ignoring source layer train-data I0407 23:46:40.768965 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:46:43.878679 359 blocking_queue.cpp:49] Waiting for data I0407 23:46:44.983029 359 solver.cpp:397] Test net output #0: accuracy = 0.450368 I0407 23:46:44.983075 359 solver.cpp:397] Test net output #1: loss = 2.83867 (* 1 = 2.83867 loss) I0407 23:46:46.763945 359 solver.cpp:218] Iteration 9696 (0.853281 iter/s, 14.0634s/12 iters), loss = 0.142083 I0407 23:46:46.763983 359 solver.cpp:237] Train net output #0: loss = 0.142083 (* 1 = 0.142083 loss) I0407 23:46:46.763990 359 sgd_solver.cpp:105] Iteration 9696, lr = 1.21951e-06 I0407 23:46:51.716920 359 solver.cpp:218] Iteration 9708 (2.42281 iter/s, 4.95292s/12 iters), loss = 0.114182 I0407 23:46:51.717041 359 solver.cpp:237] Train net output #0: loss = 0.114182 (* 1 = 0.114182 loss) I0407 23:46:51.717048 359 sgd_solver.cpp:105] Iteration 9708, lr = 1.19116e-06 I0407 23:46:52.436004 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:46:56.626152 359 solver.cpp:218] Iteration 9720 (2.44444 iter/s, 4.9091s/12 iters), loss = 0.179163 I0407 23:46:56.626191 359 solver.cpp:237] Train net output #0: loss = 0.179163 (* 1 = 0.179163 loss) I0407 23:46:56.626199 359 sgd_solver.cpp:105] Iteration 9720, lr = 1.16346e-06 I0407 23:47:01.593101 359 solver.cpp:218] Iteration 9732 (2.416 iter/s, 4.96689s/12 iters), loss = 0.178441 I0407 23:47:01.593139 359 solver.cpp:237] Train net output #0: loss = 0.178441 (* 1 = 0.178441 loss) I0407 23:47:01.593147 359 sgd_solver.cpp:105] Iteration 9732, lr = 1.13641e-06 I0407 23:47:06.506975 359 solver.cpp:218] Iteration 9744 (2.44209 iter/s, 4.91382s/12 iters), loss = 0.122313 I0407 23:47:06.507016 359 solver.cpp:237] Train net output #0: loss = 0.122313 (* 1 = 0.122313 loss) I0407 23:47:06.507023 359 sgd_solver.cpp:105] Iteration 9744, lr = 1.10999e-06 I0407 23:47:11.483729 359 solver.cpp:218] Iteration 9756 (2.41124 iter/s, 4.97669s/12 iters), loss = 0.135807 I0407 23:47:11.483770 359 solver.cpp:237] Train net output #0: loss = 0.135807 (* 1 = 0.135807 loss) I0407 23:47:11.483779 359 sgd_solver.cpp:105] Iteration 9756, lr = 1.08417e-06 I0407 23:47:16.410970 359 solver.cpp:218] Iteration 9768 (2.43547 iter/s, 4.92718s/12 iters), loss = 0.09179 I0407 23:47:16.411010 359 solver.cpp:237] Train net output #0: loss = 0.0917901 (* 1 = 0.0917901 loss) I0407 23:47:16.411018 359 sgd_solver.cpp:105] Iteration 9768, lr = 1.05897e-06 I0407 23:47:21.338914 359 solver.cpp:218] Iteration 9780 (2.43512 iter/s, 4.92789s/12 iters), loss = 0.155029 I0407 23:47:21.338954 359 solver.cpp:237] Train net output #0: loss = 0.155029 (* 1 = 0.155029 loss) I0407 23:47:21.338963 359 sgd_solver.cpp:105] Iteration 9780, lr = 1.03434e-06 I0407 23:47:25.818817 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_9792.caffemodel I0407 23:47:28.913245 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_9792.solverstate I0407 23:47:31.423696 359 solver.cpp:330] Iteration 9792, Testing net (#0) I0407 23:47:31.423713 359 net.cpp:676] Ignoring source layer train-data I0407 23:47:31.988075 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:47:36.285781 359 solver.cpp:397] Test net output #0: accuracy = 0.449142 I0407 23:47:36.285807 359 solver.cpp:397] Test net output #1: loss = 2.8326 (* 1 = 2.8326 loss) I0407 23:47:36.381963 359 solver.cpp:218] Iteration 9792 (0.797715 iter/s, 15.043s/12 iters), loss = 0.180558 I0407 23:47:36.382001 359 solver.cpp:237] Train net output #0: loss = 0.180558 (* 1 = 0.180558 loss) I0407 23:47:36.382009 359 sgd_solver.cpp:105] Iteration 9792, lr = 1.01029e-06 I0407 23:47:40.488464 359 solver.cpp:218] Iteration 9804 (2.92224 iter/s, 4.10644s/12 iters), loss = 0.0752499 I0407 23:47:40.488519 359 solver.cpp:237] Train net output #0: loss = 0.07525 (* 1 = 0.07525 loss) I0407 23:47:40.488530 359 sgd_solver.cpp:105] Iteration 9804, lr = 9.868e-07 I0407 23:47:43.394654 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:47:45.380812 359 solver.cpp:218] Iteration 9816 (2.45284 iter/s, 4.89228s/12 iters), loss = 0.155441 I0407 23:47:45.380853 359 solver.cpp:237] Train net output #0: loss = 0.155441 (* 1 = 0.155441 loss) I0407 23:47:45.380861 359 sgd_solver.cpp:105] Iteration 9816, lr = 9.63854e-07 I0407 23:47:50.344197 359 solver.cpp:218] Iteration 9828 (2.41773 iter/s, 4.96333s/12 iters), loss = 0.171004 I0407 23:47:50.344231 359 solver.cpp:237] Train net output #0: loss = 0.171004 (* 1 = 0.171004 loss) I0407 23:47:50.344238 359 sgd_solver.cpp:105] Iteration 9828, lr = 9.41442e-07 I0407 23:47:55.281509 359 solver.cpp:218] Iteration 9840 (2.4305 iter/s, 4.93726s/12 iters), loss = 0.134581 I0407 23:47:55.281543 359 solver.cpp:237] Train net output #0: loss = 0.134581 (* 1 = 0.134581 loss) I0407 23:47:55.281551 359 sgd_solver.cpp:105] Iteration 9840, lr = 9.19551e-07 I0407 23:48:00.208547 359 solver.cpp:218] Iteration 9852 (2.43557 iter/s, 4.92698s/12 iters), loss = 0.217192 I0407 23:48:00.208673 359 solver.cpp:237] Train net output #0: loss = 0.217192 (* 1 = 0.217192 loss) I0407 23:48:00.208683 359 sgd_solver.cpp:105] Iteration 9852, lr = 8.98169e-07 I0407 23:48:05.156338 359 solver.cpp:218] Iteration 9864 (2.4254 iter/s, 4.94764s/12 iters), loss = 0.0965633 I0407 23:48:05.156380 359 solver.cpp:237] Train net output #0: loss = 0.0965633 (* 1 = 0.0965633 loss) I0407 23:48:05.156388 359 sgd_solver.cpp:105] Iteration 9864, lr = 8.77284e-07 I0407 23:48:10.085152 359 solver.cpp:218] Iteration 9876 (2.43469 iter/s, 4.92875s/12 iters), loss = 0.215828 I0407 23:48:10.085189 359 solver.cpp:237] Train net output #0: loss = 0.215828 (* 1 = 0.215828 loss) I0407 23:48:10.085197 359 sgd_solver.cpp:105] Iteration 9876, lr = 8.56885e-07 I0407 23:48:15.033823 359 solver.cpp:218] Iteration 9888 (2.42492 iter/s, 4.94862s/12 iters), loss = 0.126415 I0407 23:48:15.033859 359 solver.cpp:237] Train net output #0: loss = 0.126415 (* 1 = 0.126415 loss) I0407 23:48:15.033869 359 sgd_solver.cpp:105] Iteration 9888, lr = 8.3696e-07 I0407 23:48:17.039280 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_9894.caffemodel I0407 23:48:20.152248 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_9894.solverstate I0407 23:48:22.518414 359 solver.cpp:330] Iteration 9894, Testing net (#0) I0407 23:48:22.518432 359 net.cpp:676] Ignoring source layer train-data I0407 23:48:23.000597 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:48:26.889662 359 solver.cpp:397] Test net output #0: accuracy = 0.451593 I0407 23:48:26.889706 359 solver.cpp:397] Test net output #1: loss = 2.82148 (* 1 = 2.82148 loss) I0407 23:48:28.719249 359 solver.cpp:218] Iteration 9900 (0.87685 iter/s, 13.6854s/12 iters), loss = 0.0737331 I0407 23:48:28.719292 359 solver.cpp:237] Train net output #0: loss = 0.0737331 (* 1 = 0.0737331 loss) I0407 23:48:28.719300 359 sgd_solver.cpp:105] Iteration 9900, lr = 8.17498e-07 I0407 23:48:33.615123 359 solver.cpp:218] Iteration 9912 (2.45108 iter/s, 4.89581s/12 iters), loss = 0.100437 I0407 23:48:33.615269 359 solver.cpp:237] Train net output #0: loss = 0.100437 (* 1 = 0.100437 loss) I0407 23:48:33.615278 359 sgd_solver.cpp:105] Iteration 9912, lr = 7.98489e-07 I0407 23:48:33.708283 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:48:38.555037 359 solver.cpp:218] Iteration 9924 (2.42927 iter/s, 4.93975s/12 iters), loss = 0.120905 I0407 23:48:38.555073 359 solver.cpp:237] Train net output #0: loss = 0.120905 (* 1 = 0.120905 loss) I0407 23:48:38.555080 359 sgd_solver.cpp:105] Iteration 9924, lr = 7.79922e-07 I0407 23:48:43.471093 359 solver.cpp:218] Iteration 9936 (2.44101 iter/s, 4.916s/12 iters), loss = 0.104876 I0407 23:48:43.471132 359 solver.cpp:237] Train net output #0: loss = 0.104876 (* 1 = 0.104876 loss) I0407 23:48:43.471140 359 sgd_solver.cpp:105] Iteration 9936, lr = 7.61786e-07 I0407 23:48:48.405019 359 solver.cpp:218] Iteration 9948 (2.43217 iter/s, 4.93387s/12 iters), loss = 0.121452 I0407 23:48:48.405061 359 solver.cpp:237] Train net output #0: loss = 0.121452 (* 1 = 0.121452 loss) I0407 23:48:48.405068 359 sgd_solver.cpp:105] Iteration 9948, lr = 7.44072e-07 I0407 23:48:53.327111 359 solver.cpp:218] Iteration 9960 (2.43802 iter/s, 4.92203s/12 iters), loss = 0.0796441 I0407 23:48:53.327152 359 solver.cpp:237] Train net output #0: loss = 0.0796441 (* 1 = 0.0796441 loss) I0407 23:48:53.327160 359 sgd_solver.cpp:105] Iteration 9960, lr = 7.2677e-07 I0407 23:48:58.290238 359 solver.cpp:218] Iteration 9972 (2.41786 iter/s, 4.96307s/12 iters), loss = 0.141926 I0407 23:48:58.290271 359 solver.cpp:237] Train net output #0: loss = 0.141926 (* 1 = 0.141926 loss) I0407 23:48:58.290278 359 sgd_solver.cpp:105] Iteration 9972, lr = 7.09871e-07 I0407 23:49:03.209168 359 solver.cpp:218] Iteration 9984 (2.43958 iter/s, 4.91888s/12 iters), loss = 0.152807 I0407 23:49:03.209204 359 solver.cpp:237] Train net output #0: loss = 0.152807 (* 1 = 0.152807 loss) I0407 23:49:03.209211 359 sgd_solver.cpp:105] Iteration 9984, lr = 6.93364e-07 I0407 23:49:07.649179 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_9996.caffemodel I0407 23:49:10.758925 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_9996.solverstate I0407 23:49:13.121933 359 solver.cpp:330] Iteration 9996, Testing net (#0) I0407 23:49:13.121951 359 net.cpp:676] Ignoring source layer train-data I0407 23:49:13.596901 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:49:17.666776 359 solver.cpp:397] Test net output #0: accuracy = 0.448529 I0407 23:49:17.666821 359 solver.cpp:397] Test net output #1: loss = 2.84798 (* 1 = 2.84798 loss) I0407 23:49:17.763316 359 solver.cpp:218] Iteration 9996 (0.824512 iter/s, 14.5541s/12 iters), loss = 0.174449 I0407 23:49:17.763365 359 solver.cpp:237] Train net output #0: loss = 0.174449 (* 1 = 0.174449 loss) I0407 23:49:17.763375 359 sgd_solver.cpp:105] Iteration 9996, lr = 6.77241e-07 I0407 23:49:21.880429 359 solver.cpp:218] Iteration 10008 (2.91471 iter/s, 4.11704s/12 iters), loss = 0.158894 I0407 23:49:21.880471 359 solver.cpp:237] Train net output #0: loss = 0.158895 (* 1 = 0.158895 loss) I0407 23:49:21.880479 359 sgd_solver.cpp:105] Iteration 10008, lr = 6.61493e-07 I0407 23:49:24.086469 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:49:26.825052 359 solver.cpp:218] Iteration 10020 (2.42691 iter/s, 4.94456s/12 iters), loss = 0.101659 I0407 23:49:26.825096 359 solver.cpp:237] Train net output #0: loss = 0.101659 (* 1 = 0.101659 loss) I0407 23:49:26.825104 359 sgd_solver.cpp:105] Iteration 10020, lr = 6.46111e-07 I0407 23:49:31.748116 359 solver.cpp:218] Iteration 10032 (2.43754 iter/s, 4.923s/12 iters), loss = 0.165941 I0407 23:49:31.748163 359 solver.cpp:237] Train net output #0: loss = 0.165941 (* 1 = 0.165941 loss) I0407 23:49:31.748172 359 sgd_solver.cpp:105] Iteration 10032, lr = 6.31087e-07 I0407 23:49:36.691609 359 solver.cpp:218] Iteration 10044 (2.42747 iter/s, 4.94343s/12 iters), loss = 0.108188 I0407 23:49:36.691653 359 solver.cpp:237] Train net output #0: loss = 0.108188 (* 1 = 0.108188 loss) I0407 23:49:36.691661 359 sgd_solver.cpp:105] Iteration 10044, lr = 6.16412e-07 I0407 23:49:41.592237 359 solver.cpp:218] Iteration 10056 (2.4487 iter/s, 4.90057s/12 iters), loss = 0.18827 I0407 23:49:41.592394 359 solver.cpp:237] Train net output #0: loss = 0.18827 (* 1 = 0.18827 loss) I0407 23:49:41.592404 359 sgd_solver.cpp:105] Iteration 10056, lr = 6.02079e-07 I0407 23:49:46.561509 359 solver.cpp:218] Iteration 10068 (2.41493 iter/s, 4.9691s/12 iters), loss = 0.131724 I0407 23:49:46.561554 359 solver.cpp:237] Train net output #0: loss = 0.131724 (* 1 = 0.131724 loss) I0407 23:49:46.561563 359 sgd_solver.cpp:105] Iteration 10068, lr = 5.88078e-07 I0407 23:49:51.573681 359 solver.cpp:218] Iteration 10080 (2.3942 iter/s, 5.01211s/12 iters), loss = 0.157332 I0407 23:49:51.573726 359 solver.cpp:237] Train net output #0: loss = 0.157332 (* 1 = 0.157332 loss) I0407 23:49:51.573735 359 sgd_solver.cpp:105] Iteration 10080, lr = 5.74403e-07 I0407 23:49:56.477515 359 solver.cpp:218] Iteration 10092 (2.4471 iter/s, 4.90377s/12 iters), loss = 0.0513291 I0407 23:49:56.477560 359 solver.cpp:237] Train net output #0: loss = 0.0513291 (* 1 = 0.0513291 loss) I0407 23:49:56.477567 359 sgd_solver.cpp:105] Iteration 10092, lr = 5.61047e-07 I0407 23:49:58.425443 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_10098.caffemodel I0407 23:50:01.488281 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_10098.solverstate I0407 23:50:03.855386 359 solver.cpp:330] Iteration 10098, Testing net (#0) I0407 23:50:03.855412 359 net.cpp:676] Ignoring source layer train-data I0407 23:50:04.295897 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:50:08.660059 359 solver.cpp:397] Test net output #0: accuracy = 0.449142 I0407 23:50:08.660099 359 solver.cpp:397] Test net output #1: loss = 2.8317 (* 1 = 2.8317 loss) I0407 23:50:10.435355 359 solver.cpp:218] Iteration 10104 (0.859736 iter/s, 13.9578s/12 iters), loss = 0.0973125 I0407 23:50:10.435393 359 solver.cpp:237] Train net output #0: loss = 0.0973126 (* 1 = 0.0973126 loss) I0407 23:50:10.435401 359 sgd_solver.cpp:105] Iteration 10104, lr = 5.48e-07 I0407 23:50:14.570070 364 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:50:15.170166 359 solver.cpp:218] Iteration 10116 (2.53445 iter/s, 4.73476s/12 iters), loss = 0.0722294 I0407 23:50:15.170202 359 solver.cpp:237] Train net output #0: loss = 0.0722295 (* 1 = 0.0722295 loss) I0407 23:50:15.170210 359 sgd_solver.cpp:105] Iteration 10116, lr = 5.35257e-07 I0407 23:50:19.991789 359 solver.cpp:218] Iteration 10128 (2.48882 iter/s, 4.82157s/12 iters), loss = 0.0625775 I0407 23:50:19.991825 359 solver.cpp:237] Train net output #0: loss = 0.0625775 (* 1 = 0.0625775 loss) I0407 23:50:19.991832 359 sgd_solver.cpp:105] Iteration 10128, lr = 5.22811e-07 I0407 23:50:24.814267 359 solver.cpp:218] Iteration 10140 (2.48838 iter/s, 4.82242s/12 iters), loss = 0.035246 I0407 23:50:24.814301 359 solver.cpp:237] Train net output #0: loss = 0.035246 (* 1 = 0.035246 loss) I0407 23:50:24.814309 359 sgd_solver.cpp:105] Iteration 10140, lr = 5.10653e-07 I0407 23:50:29.618716 359 solver.cpp:218] Iteration 10152 (2.49771 iter/s, 4.80439s/12 iters), loss = 0.0725636 I0407 23:50:29.618753 359 solver.cpp:237] Train net output #0: loss = 0.0725636 (* 1 = 0.0725636 loss) I0407 23:50:29.618760 359 sgd_solver.cpp:105] Iteration 10152, lr = 4.98779e-07 I0407 23:50:34.473207 359 solver.cpp:218] Iteration 10164 (2.47196 iter/s, 4.85444s/12 iters), loss = 0.131913 I0407 23:50:34.473242 359 solver.cpp:237] Train net output #0: loss = 0.131913 (* 1 = 0.131913 loss) I0407 23:50:34.473250 359 sgd_solver.cpp:105] Iteration 10164, lr = 4.87181e-07 I0407 23:50:39.397006 359 solver.cpp:218] Iteration 10176 (2.43717 iter/s, 4.92375s/12 iters), loss = 0.14953 I0407 23:50:39.397039 359 solver.cpp:237] Train net output #0: loss = 0.14953 (* 1 = 0.14953 loss) I0407 23:50:39.397047 359 sgd_solver.cpp:105] Iteration 10176, lr = 4.75852e-07 I0407 23:50:44.281131 359 solver.cpp:218] Iteration 10188 (2.45697 iter/s, 4.88407s/12 iters), loss = 0.211933 I0407 23:50:44.281167 359 solver.cpp:237] Train net output #0: loss = 0.211933 (* 1 = 0.211933 loss) I0407 23:50:44.281175 359 sgd_solver.cpp:105] Iteration 10188, lr = 4.64787e-07 I0407 23:50:48.643961 359 solver.cpp:447] Snapshotting to binary proto file snapshot_iter_10200.caffemodel I0407 23:50:51.757793 359 sgd_solver.cpp:273] Snapshotting solver state to binary proto file snapshot_iter_10200.solverstate I0407 23:50:55.223758 359 solver.cpp:310] Iteration 10200, loss = 0.0844708 I0407 23:50:55.223788 359 solver.cpp:330] Iteration 10200, Testing net (#0) I0407 23:50:55.223794 359 net.cpp:676] Ignoring source layer train-data I0407 23:50:55.608233 369 data_layer.cpp:73] Restarting data prefetching from start. I0407 23:50:59.650789 359 solver.cpp:397] Test net output #0: accuracy = 0.451593 I0407 23:50:59.650828 359 solver.cpp:397] Test net output #1: loss = 2.82001 (* 1 = 2.82001 loss) I0407 23:50:59.650838 359 solver.cpp:315] Optimization Done. I0407 23:50:59.650846 359 caffe.cpp:259] Optimization Done.