Created
October 22, 2015 09:05
-
-
Save ducha-aiki/c0d1325f0cebe0b05c36 to your computer and use it in GitHub Desktop.
Examples of how to use batch_norm in caffe
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| # The train/test net protocol buffer definition | |
| net: "examples/cifar10/cifar10_full_sigmoid_train_test.prototxt" | |
| # test_iter specifies how many forward passes the test should carry out. | |
| # In the case of CIFAR10, we have test batch size 100 and 100 test iterations, | |
| # covering the full 10,000 testing images. | |
| test_iter: 100 | |
| # Carry out testing every 1000 training iterations. | |
| test_interval: 1000 | |
| # The base learning rate, momentum and the weight decay of the network. | |
| base_lr: 0.001 | |
| momentum: 0.9 | |
| weight_decay: 0.004 | |
| # The learning rate policy | |
| lr_policy: "multistep" | |
| gamma: 0.1 | |
| # Display every 200 iterations | |
| display: 100 | |
| # The maximum number of iterations | |
| max_iter: 40000 | |
| stepvalue: 30000 | |
| stepvalue: 35000 | |
| # snapshot intermediate results | |
| snapshot_prefix: "examples/cifar10_full_sigmoid" | |
| # solver mode: CPU or GPU | |
| solver_mode: GPU |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| # The train/test net protocol buffer definition | |
| net: "examples/cifar10/cifar10_full_sigmoid_train_test_bn.prototxt" | |
| # test_iter specifies how many forward passes the test should carry out. | |
| # In the case of CIFAR10, we have test batch size 100 and 100 test iterations, | |
| # covering the full 10,000 testing images. | |
| test_iter: 100 | |
| test_initialization: false | |
| # Carry out testing every 1000 training iterations. | |
| test_interval: 1000 | |
| # The base learning rate, momentum and the weight decay of the network. | |
| base_lr: 0.001 | |
| momentum: 0.9 | |
| weight_decay: 0.004 | |
| # The learning rate policy | |
| lr_policy: "multistep" | |
| gamma: 0.1 | |
| # Display every 200 iterations | |
| display: 100 | |
| # The maximum number of iterations | |
| max_iter: 40000 | |
| stepvalue: 30000 | |
| stepvalue: 35000 | |
| # snapshot intermediate results | |
| snapshot_prefix: "examples/cifar10_full_sigmoid_bn" | |
| # solver mode: CPU or GPU | |
| solver_mode: GPU |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| name: "CIFAR10_full" | |
| layer { | |
| name: "cifar" | |
| type: "Data" | |
| top: "data" | |
| top: "label" | |
| include { | |
| phase: TRAIN | |
| } | |
| transform_param { | |
| mean_file: "examples/cifar10/mean.binaryproto" | |
| } | |
| data_param { | |
| source: "examples/cifar10/cifar10_train_lmdb" | |
| batch_size: 100 | |
| backend: LMDB | |
| } | |
| } | |
| layer { | |
| name: "cifar" | |
| type: "Data" | |
| top: "data" | |
| top: "label" | |
| include { | |
| phase: TEST | |
| } | |
| transform_param { | |
| mean_file: "examples/cifar10/mean.binaryproto" | |
| } | |
| data_param { | |
| source: "examples/cifar10/cifar10_test_lmdb" | |
| batch_size: 100 | |
| 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: 32 | |
| pad: 2 | |
| kernel_size: 5 | |
| stride: 1 | |
| weight_filler { | |
| type: "gaussian" | |
| std: 0.0001 | |
| } | |
| bias_filler { | |
| type: "constant" | |
| } | |
| } | |
| } | |
| layer { | |
| name: "pool1" | |
| type: "Pooling" | |
| bottom: "conv1" | |
| top: "pool1" | |
| pooling_param { | |
| pool: MAX | |
| kernel_size: 3 | |
| stride: 2 | |
| } | |
| } | |
| layer { | |
| name: "Sigmoid1" | |
| type: "Sigmoid" | |
| bottom: "pool1" | |
| top: "pool1" | |
| } | |
| 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: 32 | |
| pad: 2 | |
| kernel_size: 5 | |
| stride: 1 | |
| weight_filler { | |
| type: "gaussian" | |
| std: 0.01 | |
| } | |
| bias_filler { | |
| type: "constant" | |
| } | |
| } | |
| } | |
| layer { | |
| name: "Sigmoid2" | |
| type: "Sigmoid" | |
| bottom: "conv2" | |
| top: "conv2" | |
| } | |
| layer { | |
| name: "pool2" | |
| type: "Pooling" | |
| bottom: "conv2" | |
| top: "pool2" | |
| pooling_param { | |
| pool: AVE | |
| kernel_size: 3 | |
| stride: 2 | |
| } | |
| } | |
| layer { | |
| name: "conv3" | |
| type: "Convolution" | |
| bottom: "pool2" | |
| top: "conv3" | |
| convolution_param { | |
| num_output: 64 | |
| pad: 2 | |
| kernel_size: 5 | |
| stride: 1 | |
| weight_filler { | |
| type: "gaussian" | |
| std: 0.01 | |
| } | |
| bias_filler { | |
| type: "constant" | |
| } | |
| } | |
| param { | |
| lr_mult: 1 | |
| decay_mult: 1 | |
| } | |
| param { | |
| lr_mult: 2 | |
| decay_mult: 0 | |
| } | |
| } | |
| layer { | |
| name: "Sigmoid3" | |
| type: "Sigmoid" | |
| bottom: "conv3" | |
| top: "conv3" | |
| } | |
| layer { | |
| name: "pool3" | |
| type: "Pooling" | |
| bottom: "conv3" | |
| top: "pool3" | |
| pooling_param { | |
| pool: AVE | |
| kernel_size: 3 | |
| stride: 2 | |
| } | |
| } | |
| layer { | |
| name: "ip1" | |
| type: "InnerProduct" | |
| bottom: "pool3" | |
| top: "ip1" | |
| param { | |
| lr_mult: 1 | |
| decay_mult: 1 | |
| } | |
| param { | |
| lr_mult: 2 | |
| decay_mult: 0 | |
| } | |
| inner_product_param { | |
| num_output: 10 | |
| weight_filler { | |
| type: "gaussian" | |
| std: 0.01 | |
| } | |
| bias_filler { | |
| type: "constant" | |
| } | |
| } | |
| } | |
| layer { | |
| name: "accuracy" | |
| type: "Accuracy" | |
| bottom: "ip1" | |
| bottom: "label" | |
| top: "accuracy" | |
| include { | |
| phase: TEST | |
| } | |
| } | |
| layer { | |
| name: "loss" | |
| type: "SoftmaxWithLoss" | |
| bottom: "ip1" | |
| bottom: "label" | |
| top: "loss" | |
| } |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| name: "CIFAR10_full" | |
| layer { | |
| name: "cifar" | |
| type: "Data" | |
| top: "data" | |
| top: "label" | |
| include { | |
| phase: TRAIN | |
| } | |
| transform_param { | |
| mean_file: "examples/cifar10/mean.binaryproto" | |
| } | |
| data_param { | |
| source: "examples/cifar10/cifar10_train_lmdb" | |
| batch_size: 100 | |
| backend: LMDB | |
| } | |
| } | |
| layer { | |
| name: "cifar" | |
| type: "Data" | |
| top: "data" | |
| top: "label" | |
| include { | |
| phase: TEST | |
| } | |
| transform_param { | |
| mean_file: "examples/cifar10/mean.binaryproto" | |
| } | |
| data_param { | |
| source: "examples/cifar10/cifar10_test_lmdb" | |
| batch_size: 100 | |
| backend: LMDB | |
| } | |
| } | |
| layer { | |
| name: "conv1" | |
| type: "Convolution" | |
| bottom: "data" | |
| top: "conv1" | |
| param { | |
| lr_mult: 1 | |
| decay_mult: 1 | |
| } | |
| param { | |
| lr_mult: 0 | |
| decay_mult: 0 | |
| } | |
| convolution_param { | |
| num_output: 32 | |
| pad: 2 | |
| kernel_size: 5 | |
| stride: 1 | |
| weight_filler { | |
| type: "gaussian" | |
| std: 0.0001 | |
| } | |
| bias_filler { | |
| type: "constant" | |
| } | |
| } | |
| } | |
| layer { | |
| name: "pool1" | |
| type: "Pooling" | |
| bottom: "conv1" | |
| top: "pool1" | |
| pooling_param { | |
| pool: MAX | |
| kernel_size: 3 | |
| stride: 2 | |
| } | |
| } | |
| layer { | |
| name: "bn1" | |
| type: "BatchNorm" | |
| bottom: "pool1" | |
| top: "bn1" | |
| batch_norm_param { | |
| use_global_stats: false | |
| } | |
| param { | |
| lr_mult: 0 | |
| } | |
| param { | |
| lr_mult: 0 | |
| } | |
| param { | |
| lr_mult: 0 | |
| } | |
| include { | |
| phase: TRAIN | |
| } | |
| } | |
| layer { | |
| name: "bn1" | |
| type: "BatchNorm" | |
| bottom: "pool1" | |
| top: "bn1" | |
| batch_norm_param { | |
| use_global_stats: true | |
| } | |
| param { | |
| lr_mult: 0 | |
| } | |
| param { | |
| lr_mult: 0 | |
| } | |
| param { | |
| lr_mult: 0 | |
| } | |
| include { | |
| phase: TEST | |
| } | |
| } | |
| layer { | |
| name: "Sigmoid1" | |
| type: "Sigmoid" | |
| bottom: "bn1" | |
| top: "bn1" | |
| } | |
| layer { | |
| name: "conv2" | |
| type: "Convolution" | |
| bottom: "bn1" | |
| top: "conv2" | |
| param { | |
| lr_mult: 1 | |
| decay_mult: 1 | |
| } | |
| param { | |
| lr_mult: 0 | |
| decay_mult: 0 | |
| } | |
| convolution_param { | |
| num_output: 32 | |
| pad: 2 | |
| kernel_size: 5 | |
| stride: 1 | |
| weight_filler { | |
| type: "gaussian" | |
| std: 0.01 | |
| } | |
| bias_filler { | |
| type: "constant" | |
| } | |
| } | |
| } | |
| layer { | |
| name: "bn2" | |
| type: "BatchNorm" | |
| bottom: "conv2" | |
| top: "conv2" | |
| batch_norm_param { | |
| use_global_stats: false | |
| } | |
| param { | |
| lr_mult: 0 | |
| } | |
| param { | |
| lr_mult: 0 | |
| } | |
| param { | |
| lr_mult: 0 | |
| } | |
| include { | |
| phase: TRAIN | |
| } | |
| } | |
| layer { | |
| name: "bn2" | |
| type: "BatchNorm" | |
| bottom: "conv2" | |
| top: "conv2" | |
| batch_norm_param { | |
| use_global_stats: true | |
| } | |
| param { | |
| lr_mult: 0 | |
| } | |
| param { | |
| lr_mult: 0 | |
| } | |
| param { | |
| lr_mult: 0 | |
| } | |
| include { | |
| phase: TEST | |
| } | |
| } | |
| layer { | |
| name: "Sigmoid2" | |
| type: "Sigmoid" | |
| bottom: "conv2" | |
| top: "conv2" | |
| } | |
| layer { | |
| name: "pool2" | |
| type: "Pooling" | |
| bottom: "conv2" | |
| top: "pool2" | |
| pooling_param { | |
| pool: AVE | |
| kernel_size: 3 | |
| stride: 2 | |
| } | |
| } | |
| layer { | |
| name: "conv3" | |
| type: "Convolution" | |
| bottom: "pool2" | |
| top: "conv3" | |
| convolution_param { | |
| num_output: 64 | |
| pad: 2 | |
| kernel_size: 5 | |
| stride: 1 | |
| weight_filler { | |
| type: "gaussian" | |
| std: 0.01 | |
| } | |
| bias_filler { | |
| type: "constant" | |
| } | |
| } | |
| param { | |
| lr_mult: 1 | |
| decay_mult: 1 | |
| } | |
| param { | |
| lr_mult: 0 | |
| decay_mult: 0 | |
| } | |
| } | |
| layer { | |
| name: "bn3" | |
| type: "BatchNorm" | |
| bottom: "conv3" | |
| top: "conv3" | |
| batch_norm_param { | |
| use_global_stats: false | |
| } | |
| param { | |
| lr_mult: 0 | |
| } | |
| param { | |
| lr_mult: 0 | |
| } | |
| param { | |
| lr_mult: 0 | |
| } | |
| include { | |
| phase: TRAIN | |
| } | |
| } | |
| layer { | |
| name: "bn3" | |
| type: "BatchNorm" | |
| bottom: "conv3" | |
| top: "conv3" | |
| batch_norm_param { | |
| use_global_stats: true | |
| } | |
| param { | |
| lr_mult: 0 | |
| } | |
| param { | |
| lr_mult: 0 | |
| } | |
| param { | |
| lr_mult: 0 | |
| } | |
| include { | |
| phase: TEST | |
| } | |
| } | |
| layer { | |
| name: "Sigmoid3" | |
| type: "Sigmoid" | |
| bottom: "conv3" | |
| top: "conv3" | |
| } | |
| layer { | |
| name: "pool3" | |
| type: "Pooling" | |
| bottom: "conv3" | |
| top: "pool3" | |
| pooling_param { | |
| pool: AVE | |
| kernel_size: 3 | |
| stride: 2 | |
| } | |
| } | |
| layer { | |
| name: "ip1" | |
| type: "InnerProduct" | |
| bottom: "pool3" | |
| top: "ip1" | |
| param { | |
| lr_mult: 1 | |
| decay_mult: 1 | |
| } | |
| param { | |
| lr_mult: 2 | |
| decay_mult: 0 | |
| } | |
| inner_product_param { | |
| num_output: 10 | |
| weight_filler { | |
| type: "gaussian" | |
| std: 0.01 | |
| } | |
| bias_filler { | |
| type: "constant" | |
| } | |
| } | |
| } | |
| layer { | |
| name: "accuracy" | |
| type: "Accuracy" | |
| bottom: "ip1" | |
| bottom: "label" | |
| top: "accuracy" | |
| include { | |
| phase: TEST | |
| } | |
| } | |
| layer { | |
| name: "loss" | |
| type: "SoftmaxWithLoss" | |
| bottom: "ip1" | |
| bottom: "label" | |
| top: "loss" | |
| } |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| I1022 10:46:51.019372 8536 caffe.cpp:184] Using GPUs 0 | |
| I1022 10:46:51.137827 8536 solver.cpp:47] Initializing solver from parameters: | |
| test_iter: 100 | |
| test_interval: 1000 | |
| base_lr: 0.001 | |
| display: 100 | |
| max_iter: 40000 | |
| lr_policy: "multistep" | |
| gamma: 0.1 | |
| momentum: 0.9 | |
| weight_decay: 0.004 | |
| snapshot_prefix: "examples/cifar10_full_sigmoid" | |
| solver_mode: GPU | |
| device_id: 0 | |
| net: "examples/cifar10/cifar10_full_sigmoid_train_test.prototxt" | |
| stepvalue: 30000 | |
| stepvalue: 35000 | |
| I1022 10:46:51.137861 8536 solver.cpp:90] Creating training net from net file: examples/cifar10/cifar10_full_sigmoid_train_test.prototxt | |
| I1022 10:46:51.138185 8536 net.cpp:322] The NetState phase (0) differed from the phase (1) specified by a rule in layer cifar | |
| I1022 10:46:51.138197 8536 net.cpp:322] The NetState phase (0) differed from the phase (1) specified by a rule in layer accuracy | |
| I1022 10:46:51.138308 8536 net.cpp:49] Initializing net from parameters: | |
| name: "CIFAR10_full" | |
| state { | |
| phase: TRAIN | |
| } | |
| layer { | |
| name: "cifar" | |
| type: "Data" | |
| top: "data" | |
| top: "label" | |
| include { | |
| phase: TRAIN | |
| } | |
| transform_param { | |
| mean_file: "examples/cifar10/mean.binaryproto" | |
| } | |
| data_param { | |
| source: "examples/cifar10/cifar10_train_lmdb" | |
| batch_size: 100 | |
| 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: 32 | |
| pad: 2 | |
| kernel_size: 5 | |
| stride: 1 | |
| weight_filler { | |
| type: "gaussian" | |
| std: 0.0001 | |
| } | |
| bias_filler { | |
| type: "constant" | |
| } | |
| } | |
| } | |
| layer { | |
| name: "pool1" | |
| type: "Pooling" | |
| bottom: "conv1" | |
| top: "pool1" | |
| pooling_param { | |
| pool: MAX | |
| kernel_size: 3 | |
| stride: 2 | |
| } | |
| } | |
| layer { | |
| name: "Sigmoid1" | |
| type: "Sigmoid" | |
| bottom: "pool1" | |
| top: "pool1" | |
| } | |
| 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: 32 | |
| pad: 2 | |
| kernel_size: 5 | |
| stride: 1 | |
| weight_filler { | |
| type: "gaussian" | |
| std: 0.01 | |
| } | |
| bias_filler { | |
| type: "constant" | |
| } | |
| } | |
| } | |
| layer { | |
| name: "Sigmoid2" | |
| type: "Sigmoid" | |
| bottom: "conv2" | |
| top: "conv2" | |
| } | |
| layer { | |
| name: "pool2" | |
| type: "Pooling" | |
| bottom: "conv2" | |
| top: "pool2" | |
| pooling_param { | |
| pool: AVE | |
| 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: 64 | |
| pad: 2 | |
| kernel_size: 5 | |
| stride: 1 | |
| weight_filler { | |
| type: "gaussian" | |
| std: 0.01 | |
| } | |
| bias_filler { | |
| type: "constant" | |
| } | |
| } | |
| } | |
| layer { | |
| name: "Sigmoid3" | |
| type: "Sigmoid" | |
| bottom: "conv3" | |
| top: "conv3" | |
| } | |
| layer { | |
| name: "pool3" | |
| type: "Pooling" | |
| bottom: "conv3" | |
| top: "pool3" | |
| pooling_param { | |
| pool: AVE | |
| kernel_size: 3 | |
| stride: 2 | |
| } | |
| } | |
| layer { | |
| name: "ip1" | |
| type: "InnerProduct" | |
| bottom: "pool3" | |
| top: "ip1" | |
| param { | |
| lr_mult: 1 | |
| decay_mult: 1 | |
| } | |
| param { | |
| lr_mult: 2 | |
| decay_mult: 0 | |
| } | |
| inner_product_param { | |
| num_output: 10 | |
| weight_filler { | |
| type: "gaussian" | |
| std: 0.01 | |
| } | |
| bias_filler { | |
| type: "constant" | |
| } | |
| } | |
| } | |
| layer { | |
| name: "loss" | |
| type: "SoftmaxWithLoss" | |
| bottom: "ip1" | |
| bottom: "label" | |
| top: "loss" | |
| } | |
| I1022 10:46:51.138360 8536 layer_factory.hpp:76] Creating layer cifar | |
| I1022 10:46:51.138814 8536 net.cpp:106] Creating Layer cifar | |
| I1022 10:46:51.138836 8536 net.cpp:411] cifar -> data | |
| I1022 10:46:51.138877 8536 net.cpp:411] cifar -> label | |
| I1022 10:46:51.138906 8536 data_transformer.cpp:25] Loading mean file from: examples/cifar10/mean.binaryproto | |
| I1022 10:46:51.139557 8538 db_lmdb.cpp:38] Opened lmdb examples/cifar10/cifar10_train_lmdb | |
| I1022 10:46:51.151831 8536 data_layer.cpp:45] output data size: 100,3,32,32 | |
| I1022 10:46:51.154021 8536 net.cpp:150] Setting up cifar | |
| I1022 10:46:51.154062 8536 net.cpp:157] Top shape: 100 3 32 32 (307200) | |
| I1022 10:46:51.154109 8536 net.cpp:157] Top shape: 100 (100) | |
| I1022 10:46:51.154119 8536 net.cpp:165] Memory required for data: 1229200 | |
| I1022 10:46:51.154144 8536 layer_factory.hpp:76] Creating layer conv1 | |
| I1022 10:46:51.154168 8536 net.cpp:106] Creating Layer conv1 | |
| I1022 10:46:51.154193 8536 net.cpp:454] conv1 <- data | |
| I1022 10:46:51.154211 8536 net.cpp:411] conv1 -> conv1 | |
| I1022 10:46:51.154752 8536 net.cpp:150] Setting up conv1 | |
| I1022 10:46:51.154758 8536 net.cpp:157] Top shape: 100 32 32 32 (3276800) | |
| I1022 10:46:51.154773 8536 net.cpp:165] Memory required for data: 14336400 | |
| I1022 10:46:51.154786 8536 layer_factory.hpp:76] Creating layer pool1 | |
| I1022 10:46:51.154817 8536 net.cpp:106] Creating Layer pool1 | |
| I1022 10:46:51.154821 8536 net.cpp:454] pool1 <- conv1 | |
| I1022 10:46:51.154825 8536 net.cpp:411] pool1 -> pool1 | |
| I1022 10:46:51.155061 8536 net.cpp:150] Setting up pool1 | |
| I1022 10:46:51.155066 8536 net.cpp:157] Top shape: 100 32 16 16 (819200) | |
| I1022 10:46:51.155068 8536 net.cpp:165] Memory required for data: 17613200 | |
| I1022 10:46:51.155083 8536 layer_factory.hpp:76] Creating layer Sigmoid1 | |
| I1022 10:46:51.155086 8536 net.cpp:106] Creating Layer Sigmoid1 | |
| I1022 10:46:51.155088 8536 net.cpp:454] Sigmoid1 <- pool1 | |
| I1022 10:46:51.155092 8536 net.cpp:397] Sigmoid1 -> pool1 (in-place) | |
| I1022 10:46:51.155097 8536 net.cpp:150] Setting up Sigmoid1 | |
| I1022 10:46:51.155099 8536 net.cpp:157] Top shape: 100 32 16 16 (819200) | |
| I1022 10:46:51.155110 8536 net.cpp:165] Memory required for data: 20890000 | |
| I1022 10:46:51.155112 8536 layer_factory.hpp:76] Creating layer conv2 | |
| I1022 10:46:51.155118 8536 net.cpp:106] Creating Layer conv2 | |
| I1022 10:46:51.155122 8536 net.cpp:454] conv2 <- pool1 | |
| I1022 10:46:51.155125 8536 net.cpp:411] conv2 -> conv2 | |
| I1022 10:46:51.156155 8536 net.cpp:150] Setting up conv2 | |
| I1022 10:46:51.156163 8536 net.cpp:157] Top shape: 100 32 16 16 (819200) | |
| I1022 10:46:51.156165 8536 net.cpp:165] Memory required for data: 24166800 | |
| I1022 10:46:51.156182 8536 layer_factory.hpp:76] Creating layer Sigmoid2 | |
| I1022 10:46:51.156185 8536 net.cpp:106] Creating Layer Sigmoid2 | |
| I1022 10:46:51.156189 8536 net.cpp:454] Sigmoid2 <- conv2 | |
| I1022 10:46:51.156193 8536 net.cpp:397] Sigmoid2 -> conv2 (in-place) | |
| I1022 10:46:51.156196 8536 net.cpp:150] Setting up Sigmoid2 | |
| I1022 10:46:51.156199 8536 net.cpp:157] Top shape: 100 32 16 16 (819200) | |
| I1022 10:46:51.156201 8536 net.cpp:165] Memory required for data: 27443600 | |
| I1022 10:46:51.156203 8536 layer_factory.hpp:76] Creating layer pool2 | |
| I1022 10:46:51.156206 8536 net.cpp:106] Creating Layer pool2 | |
| I1022 10:46:51.156208 8536 net.cpp:454] pool2 <- conv2 | |
| I1022 10:46:51.156211 8536 net.cpp:411] pool2 -> pool2 | |
| I1022 10:46:51.156224 8536 net.cpp:150] Setting up pool2 | |
| I1022 10:46:51.156229 8536 net.cpp:157] Top shape: 100 32 8 8 (204800) | |
| I1022 10:46:51.156229 8536 net.cpp:165] Memory required for data: 28262800 | |
| I1022 10:46:51.156231 8536 layer_factory.hpp:76] Creating layer conv3 | |
| I1022 10:46:51.156237 8536 net.cpp:106] Creating Layer conv3 | |
| I1022 10:46:51.156240 8536 net.cpp:454] conv3 <- pool2 | |
| I1022 10:46:51.156244 8536 net.cpp:411] conv3 -> conv3 | |
| I1022 10:46:51.157595 8536 net.cpp:150] Setting up conv3 | |
| I1022 10:46:51.157603 8536 net.cpp:157] Top shape: 100 64 8 8 (409600) | |
| I1022 10:46:51.157604 8536 net.cpp:165] Memory required for data: 29901200 | |
| I1022 10:46:51.157610 8536 layer_factory.hpp:76] Creating layer Sigmoid3 | |
| I1022 10:46:51.157616 8536 net.cpp:106] Creating Layer Sigmoid3 | |
| I1022 10:46:51.157634 8536 net.cpp:454] Sigmoid3 <- conv3 | |
| I1022 10:46:51.157649 8536 net.cpp:397] Sigmoid3 -> conv3 (in-place) | |
| I1022 10:46:51.157660 8536 net.cpp:150] Setting up Sigmoid3 | |
| I1022 10:46:51.157665 8536 net.cpp:157] Top shape: 100 64 8 8 (409600) | |
| I1022 10:46:51.157666 8536 net.cpp:165] Memory required for data: 31539600 | |
| I1022 10:46:51.157668 8536 layer_factory.hpp:76] Creating layer pool3 | |
| I1022 10:46:51.157673 8536 net.cpp:106] Creating Layer pool3 | |
| I1022 10:46:51.157675 8536 net.cpp:454] pool3 <- conv3 | |
| I1022 10:46:51.157678 8536 net.cpp:411] pool3 -> pool3 | |
| I1022 10:46:51.157698 8536 net.cpp:150] Setting up pool3 | |
| I1022 10:46:51.157712 8536 net.cpp:157] Top shape: 100 64 4 4 (102400) | |
| I1022 10:46:51.157716 8536 net.cpp:165] Memory required for data: 31949200 | |
| I1022 10:46:51.157716 8536 layer_factory.hpp:76] Creating layer ip1 | |
| I1022 10:46:51.157724 8536 net.cpp:106] Creating Layer ip1 | |
| I1022 10:46:51.157727 8536 net.cpp:454] ip1 <- pool3 | |
| I1022 10:46:51.157730 8536 net.cpp:411] ip1 -> ip1 | |
| I1022 10:46:51.158514 8536 net.cpp:150] Setting up ip1 | |
| I1022 10:46:51.158522 8536 net.cpp:157] Top shape: 100 10 (1000) | |
| I1022 10:46:51.158524 8536 net.cpp:165] Memory required for data: 31953200 | |
| I1022 10:46:51.158529 8536 layer_factory.hpp:76] Creating layer loss | |
| I1022 10:46:51.158535 8536 net.cpp:106] Creating Layer loss | |
| I1022 10:46:51.158537 8536 net.cpp:454] loss <- ip1 | |
| I1022 10:46:51.158540 8536 net.cpp:454] loss <- label | |
| I1022 10:46:51.158545 8536 net.cpp:411] loss -> loss | |
| I1022 10:46:51.158552 8536 layer_factory.hpp:76] Creating layer loss | |
| I1022 10:46:51.158612 8536 net.cpp:150] Setting up loss | |
| I1022 10:46:51.158617 8536 net.cpp:157] Top shape: (1) | |
| I1022 10:46:51.158618 8536 net.cpp:160] with loss weight 1 | |
| I1022 10:46:51.158638 8536 net.cpp:165] Memory required for data: 31953204 | |
| I1022 10:46:51.158639 8536 net.cpp:226] loss needs backward computation. | |
| I1022 10:46:51.158642 8536 net.cpp:226] ip1 needs backward computation. | |
| I1022 10:46:51.158643 8536 net.cpp:226] pool3 needs backward computation. | |
| I1022 10:46:51.158645 8536 net.cpp:226] Sigmoid3 needs backward computation. | |
| I1022 10:46:51.158646 8536 net.cpp:226] conv3 needs backward computation. | |
| I1022 10:46:51.158648 8536 net.cpp:226] pool2 needs backward computation. | |
| I1022 10:46:51.158650 8536 net.cpp:226] Sigmoid2 needs backward computation. | |
| I1022 10:46:51.158653 8536 net.cpp:226] conv2 needs backward computation. | |
| I1022 10:46:51.158654 8536 net.cpp:226] Sigmoid1 needs backward computation. | |
| I1022 10:46:51.158656 8536 net.cpp:226] pool1 needs backward computation. | |
| I1022 10:46:51.158658 8536 net.cpp:226] conv1 needs backward computation. | |
| I1022 10:46:51.158660 8536 net.cpp:228] cifar does not need backward computation. | |
| I1022 10:46:51.158661 8536 net.cpp:270] This network produces output loss | |
| I1022 10:46:51.158669 8536 net.cpp:283] Network initialization done. | |
| I1022 10:46:51.158933 8536 solver.cpp:180] Creating test net (#0) specified by net file: examples/cifar10/cifar10_full_sigmoid_train_test.prototxt | |
| I1022 10:46:51.158953 8536 net.cpp:322] The NetState phase (1) differed from the phase (0) specified by a rule in layer cifar | |
| I1022 10:46:51.159047 8536 net.cpp:49] Initializing net from parameters: | |
| name: "CIFAR10_full" | |
| state { | |
| phase: TEST | |
| } | |
| layer { | |
| name: "cifar" | |
| type: "Data" | |
| top: "data" | |
| top: "label" | |
| include { | |
| phase: TEST | |
| } | |
| transform_param { | |
| mean_file: "examples/cifar10/mean.binaryproto" | |
| } | |
| data_param { | |
| source: "examples/cifar10/cifar10_test_lmdb" | |
| batch_size: 100 | |
| 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: 32 | |
| pad: 2 | |
| kernel_size: 5 | |
| stride: 1 | |
| weight_filler { | |
| type: "gaussian" | |
| std: 0.0001 | |
| } | |
| bias_filler { | |
| type: "constant" | |
| } | |
| } | |
| } | |
| layer { | |
| name: "pool1" | |
| type: "Pooling" | |
| bottom: "conv1" | |
| top: "pool1" | |
| pooling_param { | |
| pool: MAX | |
| kernel_size: 3 | |
| stride: 2 | |
| } | |
| } | |
| layer { | |
| name: "Sigmoid1" | |
| type: "Sigmoid" | |
| bottom: "pool1" | |
| top: "pool1" | |
| } | |
| 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: 32 | |
| pad: 2 | |
| kernel_size: 5 | |
| stride: 1 | |
| weight_filler { | |
| type: "gaussian" | |
| std: 0.01 | |
| } | |
| bias_filler { | |
| type: "constant" | |
| } | |
| } | |
| } | |
| layer { | |
| name: "Sigmoid2" | |
| type: "Sigmoid" | |
| bottom: "conv2" | |
| top: "conv2" | |
| } | |
| layer { | |
| name: "pool2" | |
| type: "Pooling" | |
| bottom: "conv2" | |
| top: "pool2" | |
| pooling_param { | |
| pool: AVE | |
| 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: 64 | |
| pad: 2 | |
| kernel_size: 5 | |
| stride: 1 | |
| weight_filler { | |
| type: "gaussian" | |
| std: 0.01 | |
| } | |
| bias_filler { | |
| type: "constant" | |
| } | |
| } | |
| } | |
| layer { | |
| name: "Sigmoid3" | |
| type: "Sigmoid" | |
| bottom: "conv3" | |
| top: "conv3" | |
| } | |
| layer { | |
| name: "pool3" | |
| type: "Pooling" | |
| bottom: "conv3" | |
| top: "pool3" | |
| pooling_param { | |
| pool: AVE | |
| kernel_size: 3 | |
| stride: 2 | |
| } | |
| } | |
| layer { | |
| name: "ip1" | |
| type: "InnerProduct" | |
| bottom: "pool3" | |
| top: "ip1" | |
| param { | |
| lr_mult: 1 | |
| decay_mult: 1 | |
| } | |
| param { | |
| lr_mult: 2 | |
| decay_mult: 0 | |
| } | |
| inner_product_param { | |
| num_output: 10 | |
| weight_filler { | |
| type: "gaussian" | |
| std: 0.01 | |
| } | |
| bias_filler { | |
| type: "constant" | |
| } | |
| } | |
| } | |
| layer { | |
| name: "accuracy" | |
| type: "Accuracy" | |
| bottom: "ip1" | |
| bottom: "label" | |
| top: "accuracy" | |
| include { | |
| phase: TEST | |
| } | |
| } | |
| layer { | |
| name: "loss" | |
| type: "SoftmaxWithLoss" | |
| bottom: "ip1" | |
| bottom: "label" | |
| top: "loss" | |
| } | |
| I1022 10:46:51.159111 8536 layer_factory.hpp:76] Creating layer cifar | |
| I1022 10:46:51.159279 8536 net.cpp:106] Creating Layer cifar | |
| I1022 10:46:51.159299 8536 net.cpp:411] cifar -> data | |
| I1022 10:46:51.159314 8536 net.cpp:411] cifar -> label | |
| I1022 10:46:51.159328 8536 data_transformer.cpp:25] Loading mean file from: examples/cifar10/mean.binaryproto | |
| I1022 10:46:51.159919 8540 db_lmdb.cpp:38] Opened lmdb examples/cifar10/cifar10_test_lmdb | |
| I1022 10:46:51.159982 8536 data_layer.cpp:45] output data size: 100,3,32,32 | |
| I1022 10:46:51.162274 8536 net.cpp:150] Setting up cifar | |
| I1022 10:46:51.162292 8536 net.cpp:157] Top shape: 100 3 32 32 (307200) | |
| I1022 10:46:51.162297 8536 net.cpp:157] Top shape: 100 (100) | |
| I1022 10:46:51.162298 8536 net.cpp:165] Memory required for data: 1229200 | |
| I1022 10:46:51.162303 8536 layer_factory.hpp:76] Creating layer label_cifar_1_split | |
| I1022 10:46:51.162314 8536 net.cpp:106] Creating Layer label_cifar_1_split | |
| I1022 10:46:51.162317 8536 net.cpp:454] label_cifar_1_split <- label | |
| I1022 10:46:51.162323 8536 net.cpp:411] label_cifar_1_split -> label_cifar_1_split_0 | |
| I1022 10:46:51.162330 8536 net.cpp:411] label_cifar_1_split -> label_cifar_1_split_1 | |
| I1022 10:46:51.162369 8536 net.cpp:150] Setting up label_cifar_1_split | |
| I1022 10:46:51.162374 8536 net.cpp:157] Top shape: 100 (100) | |
| I1022 10:46:51.162376 8536 net.cpp:157] Top shape: 100 (100) | |
| I1022 10:46:51.162377 8536 net.cpp:165] Memory required for data: 1230000 | |
| I1022 10:46:51.162380 8536 layer_factory.hpp:76] Creating layer conv1 | |
| I1022 10:46:51.162389 8536 net.cpp:106] Creating Layer conv1 | |
| I1022 10:46:51.162391 8536 net.cpp:454] conv1 <- data | |
| I1022 10:46:51.162396 8536 net.cpp:411] conv1 -> conv1 | |
| I1022 10:46:51.162612 8536 net.cpp:150] Setting up conv1 | |
| I1022 10:46:51.162618 8536 net.cpp:157] Top shape: 100 32 32 32 (3276800) | |
| I1022 10:46:51.162621 8536 net.cpp:165] Memory required for data: 14337200 | |
| I1022 10:46:51.162628 8536 layer_factory.hpp:76] Creating layer pool1 | |
| I1022 10:46:51.162634 8536 net.cpp:106] Creating Layer pool1 | |
| I1022 10:46:51.162636 8536 net.cpp:454] pool1 <- conv1 | |
| I1022 10:46:51.162641 8536 net.cpp:411] pool1 -> pool1 | |
| I1022 10:46:51.162667 8536 net.cpp:150] Setting up pool1 | |
| I1022 10:46:51.162672 8536 net.cpp:157] Top shape: 100 32 16 16 (819200) | |
| I1022 10:46:51.162673 8536 net.cpp:165] Memory required for data: 17614000 | |
| I1022 10:46:51.162675 8536 layer_factory.hpp:76] Creating layer Sigmoid1 | |
| I1022 10:46:51.162679 8536 net.cpp:106] Creating Layer Sigmoid1 | |
| I1022 10:46:51.162681 8536 net.cpp:454] Sigmoid1 <- pool1 | |
| I1022 10:46:51.162684 8536 net.cpp:397] Sigmoid1 -> pool1 (in-place) | |
| I1022 10:46:51.162688 8536 net.cpp:150] Setting up Sigmoid1 | |
| I1022 10:46:51.162691 8536 net.cpp:157] Top shape: 100 32 16 16 (819200) | |
| I1022 10:46:51.162693 8536 net.cpp:165] Memory required for data: 20890800 | |
| I1022 10:46:51.162710 8536 layer_factory.hpp:76] Creating layer conv2 | |
| I1022 10:46:51.162716 8536 net.cpp:106] Creating Layer conv2 | |
| I1022 10:46:51.162719 8536 net.cpp:454] conv2 <- pool1 | |
| I1022 10:46:51.162724 8536 net.cpp:411] conv2 -> conv2 | |
| I1022 10:46:51.163525 8536 net.cpp:150] Setting up conv2 | |
| I1022 10:46:51.163532 8536 net.cpp:157] Top shape: 100 32 16 16 (819200) | |
| I1022 10:46:51.163533 8536 net.cpp:165] Memory required for data: 24167600 | |
| I1022 10:46:51.163539 8536 layer_factory.hpp:76] Creating layer Sigmoid2 | |
| I1022 10:46:51.163543 8536 net.cpp:106] Creating Layer Sigmoid2 | |
| I1022 10:46:51.163545 8536 net.cpp:454] Sigmoid2 <- conv2 | |
| I1022 10:46:51.163548 8536 net.cpp:397] Sigmoid2 -> conv2 (in-place) | |
| I1022 10:46:51.163552 8536 net.cpp:150] Setting up Sigmoid2 | |
| I1022 10:46:51.163555 8536 net.cpp:157] Top shape: 100 32 16 16 (819200) | |
| I1022 10:46:51.163558 8536 net.cpp:165] Memory required for data: 27444400 | |
| I1022 10:46:51.163558 8536 layer_factory.hpp:76] Creating layer pool2 | |
| I1022 10:46:51.163563 8536 net.cpp:106] Creating Layer pool2 | |
| I1022 10:46:51.163564 8536 net.cpp:454] pool2 <- conv2 | |
| I1022 10:46:51.163568 8536 net.cpp:411] pool2 -> pool2 | |
| I1022 10:46:51.163580 8536 net.cpp:150] Setting up pool2 | |
| I1022 10:46:51.163586 8536 net.cpp:157] Top shape: 100 32 8 8 (204800) | |
| I1022 10:46:51.163588 8536 net.cpp:165] Memory required for data: 28263600 | |
| I1022 10:46:51.163589 8536 layer_factory.hpp:76] Creating layer conv3 | |
| I1022 10:46:51.163596 8536 net.cpp:106] Creating Layer conv3 | |
| I1022 10:46:51.163599 8536 net.cpp:454] conv3 <- pool2 | |
| I1022 10:46:51.163602 8536 net.cpp:411] conv3 -> conv3 | |
| I1022 10:46:51.165074 8536 net.cpp:150] Setting up conv3 | |
| I1022 10:46:51.165084 8536 net.cpp:157] Top shape: 100 64 8 8 (409600) | |
| I1022 10:46:51.165086 8536 net.cpp:165] Memory required for data: 29902000 | |
| I1022 10:46:51.165092 8536 layer_factory.hpp:76] Creating layer Sigmoid3 | |
| I1022 10:46:51.165096 8536 net.cpp:106] Creating Layer Sigmoid3 | |
| I1022 10:46:51.165098 8536 net.cpp:454] Sigmoid3 <- conv3 | |
| I1022 10:46:51.165102 8536 net.cpp:397] Sigmoid3 -> conv3 (in-place) | |
| I1022 10:46:51.165105 8536 net.cpp:150] Setting up Sigmoid3 | |
| I1022 10:46:51.165108 8536 net.cpp:157] Top shape: 100 64 8 8 (409600) | |
| I1022 10:46:51.165110 8536 net.cpp:165] Memory required for data: 31540400 | |
| I1022 10:46:51.165112 8536 layer_factory.hpp:76] Creating layer pool3 | |
| I1022 10:46:51.165117 8536 net.cpp:106] Creating Layer pool3 | |
| I1022 10:46:51.165117 8536 net.cpp:454] pool3 <- conv3 | |
| I1022 10:46:51.165122 8536 net.cpp:411] pool3 -> pool3 | |
| I1022 10:46:51.165137 8536 net.cpp:150] Setting up pool3 | |
| I1022 10:46:51.165158 8536 net.cpp:157] Top shape: 100 64 4 4 (102400) | |
| I1022 10:46:51.165169 8536 net.cpp:165] Memory required for data: 31950000 | |
| I1022 10:46:51.165179 8536 layer_factory.hpp:76] Creating layer ip1 | |
| I1022 10:46:51.165191 8536 net.cpp:106] Creating Layer ip1 | |
| I1022 10:46:51.165202 8536 net.cpp:454] ip1 <- pool3 | |
| I1022 10:46:51.165215 8536 net.cpp:411] ip1 -> ip1 | |
| I1022 10:46:51.165566 8536 net.cpp:150] Setting up ip1 | |
| I1022 10:46:51.165573 8536 net.cpp:157] Top shape: 100 10 (1000) | |
| I1022 10:46:51.165575 8536 net.cpp:165] Memory required for data: 31954000 | |
| I1022 10:46:51.165580 8536 layer_factory.hpp:76] Creating layer ip1_ip1_0_split | |
| I1022 10:46:51.165585 8536 net.cpp:106] Creating Layer ip1_ip1_0_split | |
| I1022 10:46:51.165587 8536 net.cpp:454] ip1_ip1_0_split <- ip1 | |
| I1022 10:46:51.165591 8536 net.cpp:411] ip1_ip1_0_split -> ip1_ip1_0_split_0 | |
| I1022 10:46:51.165594 8536 net.cpp:411] ip1_ip1_0_split -> ip1_ip1_0_split_1 | |
| I1022 10:46:51.165618 8536 net.cpp:150] Setting up ip1_ip1_0_split | |
| I1022 10:46:51.165623 8536 net.cpp:157] Top shape: 100 10 (1000) | |
| I1022 10:46:51.165627 8536 net.cpp:157] Top shape: 100 10 (1000) | |
| I1022 10:46:51.165628 8536 net.cpp:165] Memory required for data: 31962000 | |
| I1022 10:46:51.165630 8536 layer_factory.hpp:76] Creating layer accuracy | |
| I1022 10:46:51.165637 8536 net.cpp:106] Creating Layer accuracy | |
| I1022 10:46:51.165640 8536 net.cpp:454] accuracy <- ip1_ip1_0_split_0 | |
| I1022 10:46:51.165643 8536 net.cpp:454] accuracy <- label_cifar_1_split_0 | |
| I1022 10:46:51.165668 8536 net.cpp:411] accuracy -> accuracy | |
| I1022 10:46:51.165678 8536 net.cpp:150] Setting up accuracy | |
| I1022 10:46:51.165690 8536 net.cpp:157] Top shape: (1) | |
| I1022 10:46:51.165701 8536 net.cpp:165] Memory required for data: 31962004 | |
| I1022 10:46:51.165710 8536 layer_factory.hpp:76] Creating layer loss | |
| I1022 10:46:51.165722 8536 net.cpp:106] Creating Layer loss | |
| I1022 10:46:51.165732 8536 net.cpp:454] loss <- ip1_ip1_0_split_1 | |
| I1022 10:46:51.165743 8536 net.cpp:454] loss <- label_cifar_1_split_1 | |
| I1022 10:46:51.165756 8536 net.cpp:411] loss -> loss | |
| I1022 10:46:51.165771 8536 layer_factory.hpp:76] Creating layer loss | |
| I1022 10:46:51.165832 8536 net.cpp:150] Setting up loss | |
| I1022 10:46:51.165848 8536 net.cpp:157] Top shape: (1) | |
| I1022 10:46:51.165856 8536 net.cpp:160] with loss weight 1 | |
| I1022 10:46:51.165873 8536 net.cpp:165] Memory required for data: 31962008 | |
| I1022 10:46:51.165881 8536 net.cpp:226] loss needs backward computation. | |
| I1022 10:46:51.165890 8536 net.cpp:228] accuracy does not need backward computation. | |
| I1022 10:46:51.165900 8536 net.cpp:226] ip1_ip1_0_split needs backward computation. | |
| I1022 10:46:51.165910 8536 net.cpp:226] ip1 needs backward computation. | |
| I1022 10:46:51.165918 8536 net.cpp:226] pool3 needs backward computation. | |
| I1022 10:46:51.165927 8536 net.cpp:226] Sigmoid3 needs backward computation. | |
| I1022 10:46:51.165935 8536 net.cpp:226] conv3 needs backward computation. | |
| I1022 10:46:51.165945 8536 net.cpp:226] pool2 needs backward computation. | |
| I1022 10:46:51.165953 8536 net.cpp:226] Sigmoid2 needs backward computation. | |
| I1022 10:46:51.165962 8536 net.cpp:226] conv2 needs backward computation. | |
| I1022 10:46:51.165971 8536 net.cpp:226] Sigmoid1 needs backward computation. | |
| I1022 10:46:51.165979 8536 net.cpp:226] pool1 needs backward computation. | |
| I1022 10:46:51.165988 8536 net.cpp:226] conv1 needs backward computation. | |
| I1022 10:46:51.165998 8536 net.cpp:228] label_cifar_1_split does not need backward computation. | |
| I1022 10:46:51.166007 8536 net.cpp:228] cifar does not need backward computation. | |
| I1022 10:46:51.166016 8536 net.cpp:270] This network produces output accuracy | |
| I1022 10:46:51.166025 8536 net.cpp:270] This network produces output loss | |
| I1022 10:46:51.166043 8536 net.cpp:283] Network initialization done. | |
| I1022 10:46:51.166105 8536 solver.cpp:59] Solver scaffolding done. | |
| I1022 10:46:51.166326 8536 caffe.cpp:212] Starting Optimization | |
| I1022 10:46:51.166334 8536 solver.cpp:287] Solving CIFAR10_full | |
| I1022 10:46:51.166337 8536 solver.cpp:288] Learning Rate Policy: multistep | |
| I1022 10:46:51.166899 8536 solver.cpp:340] Iteration 0, Testing net (#0) | |
| I1022 10:46:53.488729 8536 solver.cpp:408] Test net output #0: accuracy = 0.1 | |
| I1022 10:46:53.488755 8536 solver.cpp:408] Test net output #1: loss = 2.31366 (* 1 = 2.31366 loss) | |
| I1022 10:46:53.521021 8536 solver.cpp:236] Iteration 0, loss = 2.294 | |
| I1022 10:46:53.521047 8536 solver.cpp:252] Train net output #0: loss = 2.294 (* 1 = 2.294 loss) | |
| I1022 10:46:53.521064 8536 sgd_solver.cpp:106] Iteration 0, lr = 0.001 | |
| I1022 10:46:58.813424 8536 solver.cpp:236] Iteration 100, loss = 2.30219 | |
| I1022 10:46:58.813460 8536 solver.cpp:252] Train net output #0: loss = 2.30219 (* 1 = 2.30219 loss) | |
| I1022 10:46:58.813465 8536 sgd_solver.cpp:106] Iteration 100, lr = 0.001 | |
| I1022 10:47:04.035897 8536 solver.cpp:236] Iteration 200, loss = 2.29944 | |
| I1022 10:47:04.035923 8536 solver.cpp:252] Train net output #0: loss = 2.29944 (* 1 = 2.29944 loss) | |
| I1022 10:47:04.035928 8536 sgd_solver.cpp:106] Iteration 200, lr = 0.001 | |
| I1022 10:47:09.264737 8536 solver.cpp:236] Iteration 300, loss = 2.29272 | |
| I1022 10:47:09.264763 8536 solver.cpp:252] Train net output #0: loss = 2.29272 (* 1 = 2.29272 loss) | |
| I1022 10:47:09.264770 8536 sgd_solver.cpp:106] Iteration 300, lr = 0.001 | |
| I1022 10:47:16.553303 8536 solver.cpp:236] Iteration 400, loss = 2.31916 | |
| I1022 10:47:16.553328 8536 solver.cpp:252] Train net output #0: loss = 2.31916 (* 1 = 2.31916 loss) | |
| I1022 10:47:16.553334 8536 sgd_solver.cpp:106] Iteration 400, lr = 0.001 | |
| I1022 10:47:27.018857 8536 solver.cpp:236] Iteration 500, loss = 2.30496 | |
| I1022 10:47:27.018935 8536 solver.cpp:252] Train net output #0: loss = 2.30496 (* 1 = 2.30496 loss) | |
| I1022 10:47:27.018942 8536 sgd_solver.cpp:106] Iteration 500, lr = 0.001 | |
| I1022 10:47:37.502616 8536 solver.cpp:236] Iteration 600, loss = 2.30169 | |
| I1022 10:47:37.502653 8536 solver.cpp:252] Train net output #0: loss = 2.30169 (* 1 = 2.30169 loss) | |
| I1022 10:47:37.502658 8536 sgd_solver.cpp:106] Iteration 600, lr = 0.001 | |
| I1022 10:47:48.146152 8536 solver.cpp:236] Iteration 700, loss = 2.29867 | |
| I1022 10:47:48.146188 8536 solver.cpp:252] Train net output #0: loss = 2.29867 (* 1 = 2.29867 loss) | |
| I1022 10:47:48.146194 8536 sgd_solver.cpp:106] Iteration 700, lr = 0.001 | |
| I1022 10:47:58.747323 8536 solver.cpp:236] Iteration 800, loss = 2.29204 | |
| I1022 10:47:58.747630 8536 solver.cpp:252] Train net output #0: loss = 2.29204 (* 1 = 2.29204 loss) | |
| I1022 10:47:58.747637 8536 sgd_solver.cpp:106] Iteration 800, lr = 0.001 | |
| I1022 10:48:09.425886 8536 solver.cpp:236] Iteration 900, loss = 2.31886 | |
| I1022 10:48:09.425925 8536 solver.cpp:252] Train net output #0: loss = 2.31886 (* 1 = 2.31886 loss) | |
| I1022 10:48:09.425930 8536 sgd_solver.cpp:106] Iteration 900, lr = 0.001 | |
| I1022 10:48:19.959596 8536 solver.cpp:340] Iteration 1000, Testing net (#0) | |
| I1022 10:48:25.035497 8536 solver.cpp:408] Test net output #0: accuracy = 0.1 | |
| I1022 10:48:25.035532 8536 solver.cpp:408] Test net output #1: loss = 2.30539 (* 1 = 2.30539 loss) | |
| I1022 10:48:25.095721 8536 solver.cpp:236] Iteration 1000, loss = 2.30494 | |
| I1022 10:48:25.095741 8536 solver.cpp:252] Train net output #0: loss = 2.30494 (* 1 = 2.30494 loss) | |
| I1022 10:48:25.095747 8536 sgd_solver.cpp:106] Iteration 1000, lr = 0.001 | |
| I1022 10:48:35.566923 8536 solver.cpp:236] Iteration 1100, loss = 2.29925 | |
| I1022 10:48:35.567010 8536 solver.cpp:252] Train net output #0: loss = 2.29925 (* 1 = 2.29925 loss) | |
| I1022 10:48:35.567015 8536 sgd_solver.cpp:106] Iteration 1100, lr = 0.001 | |
| I1022 10:48:46.141048 8536 solver.cpp:236] Iteration 1200, loss = 2.29594 | |
| I1022 10:48:46.141077 8536 solver.cpp:252] Train net output #0: loss = 2.29594 (* 1 = 2.29594 loss) | |
| I1022 10:48:46.141083 8536 sgd_solver.cpp:106] Iteration 1200, lr = 0.001 | |
| I1022 10:48:56.712587 8536 solver.cpp:236] Iteration 1300, loss = 2.28786 | |
| I1022 10:48:56.712626 8536 solver.cpp:252] Train net output #0: loss = 2.28786 (* 1 = 2.28786 loss) | |
| I1022 10:48:56.712631 8536 sgd_solver.cpp:106] Iteration 1300, lr = 0.001 | |
| I1022 10:49:07.135187 8536 solver.cpp:236] Iteration 1400, loss = 2.31282 | |
| I1022 10:49:07.135274 8536 solver.cpp:252] Train net output #0: loss = 2.31282 (* 1 = 2.31282 loss) | |
| I1022 10:49:07.135280 8536 sgd_solver.cpp:106] Iteration 1400, lr = 0.001 | |
| I1022 10:49:17.569702 8536 solver.cpp:236] Iteration 1500, loss = 2.29396 | |
| I1022 10:49:17.569737 8536 solver.cpp:252] Train net output #0: loss = 2.29396 (* 1 = 2.29396 loss) | |
| I1022 10:49:17.569743 8536 sgd_solver.cpp:106] Iteration 1500, lr = 0.001 | |
| I1022 10:49:28.073379 8536 solver.cpp:236] Iteration 1600, loss = 2.24384 | |
| I1022 10:49:28.073416 8536 solver.cpp:252] Train net output #0: loss = 2.24384 (* 1 = 2.24384 loss) | |
| I1022 10:49:28.073421 8536 sgd_solver.cpp:106] Iteration 1600, lr = 0.001 | |
| I1022 10:49:38.646904 8536 solver.cpp:236] Iteration 1700, loss = 2.20429 | |
| I1022 10:49:38.646970 8536 solver.cpp:252] Train net output #0: loss = 2.20429 (* 1 = 2.20429 loss) | |
| I1022 10:49:38.646976 8536 sgd_solver.cpp:106] Iteration 1700, lr = 0.001 | |
| I1022 10:49:49.163851 8536 solver.cpp:236] Iteration 1800, loss = 2.08868 | |
| I1022 10:49:49.163887 8536 solver.cpp:252] Train net output #0: loss = 2.08868 (* 1 = 2.08868 loss) | |
| I1022 10:49:49.163892 8536 sgd_solver.cpp:106] Iteration 1800, lr = 0.001 | |
| I1022 10:49:59.671581 8536 solver.cpp:236] Iteration 1900, loss = 2.08125 | |
| I1022 10:49:59.671609 8536 solver.cpp:252] Train net output #0: loss = 2.08125 (* 1 = 2.08125 loss) | |
| I1022 10:49:59.671617 8536 sgd_solver.cpp:106] Iteration 1900, lr = 0.001 | |
| I1022 10:50:10.055730 8536 solver.cpp:340] Iteration 2000, Testing net (#0) | |
| I1022 10:50:15.141084 8536 solver.cpp:408] Test net output #0: accuracy = 0.2355 | |
| I1022 10:50:15.141109 8536 solver.cpp:408] Test net output #1: loss = 2.05127 (* 1 = 2.05127 loss) | |
| I1022 10:50:15.208761 8536 solver.cpp:236] Iteration 2000, loss = 2.11215 | |
| I1022 10:50:15.208791 8536 solver.cpp:252] Train net output #0: loss = 2.11215 (* 1 = 2.11215 loss) | |
| I1022 10:50:15.208796 8536 sgd_solver.cpp:106] Iteration 2000, lr = 0.001 | |
| I1022 10:50:25.757208 8536 solver.cpp:236] Iteration 2100, loss = 2.0152 | |
| I1022 10:50:25.757263 8536 solver.cpp:252] Train net output #0: loss = 2.0152 (* 1 = 2.0152 loss) | |
| I1022 10:50:25.757274 8536 sgd_solver.cpp:106] Iteration 2100, lr = 0.001 | |
| I1022 10:50:36.272904 8536 solver.cpp:236] Iteration 2200, loss = 2.06809 | |
| I1022 10:50:36.272940 8536 solver.cpp:252] Train net output #0: loss = 2.06809 (* 1 = 2.06809 loss) | |
| I1022 10:50:36.272946 8536 sgd_solver.cpp:106] Iteration 2200, lr = 0.001 | |
| I1022 10:50:46.837977 8536 solver.cpp:236] Iteration 2300, loss = 1.93628 | |
| I1022 10:50:46.838059 8536 solver.cpp:252] Train net output #0: loss = 1.93628 (* 1 = 1.93628 loss) | |
| I1022 10:50:46.838064 8536 sgd_solver.cpp:106] Iteration 2300, lr = 0.001 | |
| I1022 10:50:57.433101 8536 solver.cpp:236] Iteration 2400, loss = 1.99968 | |
| I1022 10:50:57.433138 8536 solver.cpp:252] Train net output #0: loss = 1.99968 (* 1 = 1.99968 loss) | |
| I1022 10:50:57.433145 8536 sgd_solver.cpp:106] Iteration 2400, lr = 0.001 | |
| I1022 10:51:07.859019 8536 solver.cpp:236] Iteration 2500, loss = 2.02649 | |
| I1022 10:51:07.859055 8536 solver.cpp:252] Train net output #0: loss = 2.02649 (* 1 = 2.02649 loss) | |
| I1022 10:51:07.859061 8536 sgd_solver.cpp:106] Iteration 2500, lr = 0.001 | |
| I1022 10:51:18.373309 8536 solver.cpp:236] Iteration 2600, loss = 1.94685 | |
| I1022 10:51:18.373383 8536 solver.cpp:252] Train net output #0: loss = 1.94685 (* 1 = 1.94685 loss) | |
| I1022 10:51:18.373389 8536 sgd_solver.cpp:106] Iteration 2600, lr = 0.001 | |
| I1022 10:51:28.929222 8536 solver.cpp:236] Iteration 2700, loss = 2.01653 | |
| I1022 10:51:28.929262 8536 solver.cpp:252] Train net output #0: loss = 2.01653 (* 1 = 2.01653 loss) | |
| I1022 10:51:28.929267 8536 sgd_solver.cpp:106] Iteration 2700, lr = 0.001 | |
| I1022 10:51:39.529273 8536 solver.cpp:236] Iteration 2800, loss = 1.87566 | |
| I1022 10:51:39.529299 8536 solver.cpp:252] Train net output #0: loss = 1.87566 (* 1 = 1.87566 loss) | |
| I1022 10:51:39.529304 8536 sgd_solver.cpp:106] Iteration 2800, lr = 0.001 | |
| I1022 10:51:50.083822 8536 solver.cpp:236] Iteration 2900, loss = 1.92571 | |
| I1022 10:51:50.083925 8536 solver.cpp:252] Train net output #0: loss = 1.92571 (* 1 = 1.92571 loss) | |
| I1022 10:51:50.083938 8536 sgd_solver.cpp:106] Iteration 2900, lr = 0.001 | |
| I1022 10:52:00.507814 8536 solver.cpp:340] Iteration 3000, Testing net (#0) | |
| I1022 10:52:05.738812 8536 solver.cpp:408] Test net output #0: accuracy = 0.2823 | |
| I1022 10:52:05.738847 8536 solver.cpp:408] Test net output #1: loss = 1.94049 (* 1 = 1.94049 loss) | |
| I1022 10:52:05.809937 8536 solver.cpp:236] Iteration 3000, loss = 2.00196 | |
| I1022 10:52:05.809964 8536 solver.cpp:252] Train net output #0: loss = 2.00196 (* 1 = 2.00196 loss) | |
| I1022 10:52:05.809972 8536 sgd_solver.cpp:106] Iteration 3000, lr = 0.001 | |
| I1022 10:52:16.504825 8536 solver.cpp:236] Iteration 3100, loss = 1.88657 | |
| I1022 10:52:16.504851 8536 solver.cpp:252] Train net output #0: loss = 1.88657 (* 1 = 1.88657 loss) | |
| I1022 10:52:16.504856 8536 sgd_solver.cpp:106] Iteration 3100, lr = 0.001 | |
| I1022 10:52:27.067658 8536 solver.cpp:236] Iteration 3200, loss = 1.96299 | |
| I1022 10:52:27.067760 8536 solver.cpp:252] Train net output #0: loss = 1.96299 (* 1 = 1.96299 loss) | |
| I1022 10:52:27.067766 8536 sgd_solver.cpp:106] Iteration 3200, lr = 0.001 | |
| I1022 10:52:37.625941 8536 solver.cpp:236] Iteration 3300, loss = 1.8326 | |
| I1022 10:52:37.625978 8536 solver.cpp:252] Train net output #0: loss = 1.8326 (* 1 = 1.8326 loss) | |
| I1022 10:52:37.625984 8536 sgd_solver.cpp:106] Iteration 3300, lr = 0.001 | |
| I1022 10:52:48.165839 8536 solver.cpp:236] Iteration 3400, loss = 1.8503 | |
| I1022 10:52:48.165876 8536 solver.cpp:252] Train net output #0: loss = 1.8503 (* 1 = 1.8503 loss) | |
| I1022 10:52:48.165881 8536 sgd_solver.cpp:106] Iteration 3400, lr = 0.001 | |
| I1022 10:52:58.601848 8536 solver.cpp:236] Iteration 3500, loss = 1.97493 | |
| I1022 10:52:58.601970 8536 solver.cpp:252] Train net output #0: loss = 1.97493 (* 1 = 1.97493 loss) | |
| I1022 10:52:58.601986 8536 sgd_solver.cpp:106] Iteration 3500, lr = 0.001 | |
| I1022 10:53:08.896368 8536 solver.cpp:236] Iteration 3600, loss = 1.85193 | |
| I1022 10:53:08.896407 8536 solver.cpp:252] Train net output #0: loss = 1.85193 (* 1 = 1.85193 loss) | |
| I1022 10:53:08.896414 8536 sgd_solver.cpp:106] Iteration 3600, lr = 0.001 | |
| I1022 10:53:19.378051 8536 solver.cpp:236] Iteration 3700, loss = 1.94166 | |
| I1022 10:53:19.378079 8536 solver.cpp:252] Train net output #0: loss = 1.94166 (* 1 = 1.94166 loss) | |
| I1022 10:53:19.378085 8536 sgd_solver.cpp:106] Iteration 3700, lr = 0.001 | |
| I1022 10:53:29.866140 8536 solver.cpp:236] Iteration 3800, loss = 1.76446 | |
| I1022 10:53:29.866227 8536 solver.cpp:252] Train net output #0: loss = 1.76446 (* 1 = 1.76446 loss) | |
| I1022 10:53:29.866233 8536 sgd_solver.cpp:106] Iteration 3800, lr = 0.001 | |
| I1022 10:53:40.353912 8536 solver.cpp:236] Iteration 3900, loss = 1.81656 | |
| I1022 10:53:40.353950 8536 solver.cpp:252] Train net output #0: loss = 1.81656 (* 1 = 1.81656 loss) | |
| I1022 10:53:40.353956 8536 sgd_solver.cpp:106] Iteration 3900, lr = 0.001 | |
| I1022 10:53:50.744300 8536 solver.cpp:340] Iteration 4000, Testing net (#0) | |
| I1022 10:53:55.826562 8536 solver.cpp:408] Test net output #0: accuracy = 0.3026 | |
| I1022 10:53:55.826597 8536 solver.cpp:408] Test net output #1: loss = 1.8486 (* 1 = 1.8486 loss) | |
| I1022 10:53:55.895066 8536 solver.cpp:236] Iteration 4000, loss = 1.93899 | |
| I1022 10:53:55.895099 8536 solver.cpp:252] Train net output #0: loss = 1.93899 (* 1 = 1.93899 loss) | |
| I1022 10:53:55.895105 8536 sgd_solver.cpp:106] Iteration 4000, lr = 0.001 | |
| I1022 10:54:06.391891 8536 solver.cpp:236] Iteration 4100, loss = 1.77812 | |
| I1022 10:54:06.391999 8536 solver.cpp:252] Train net output #0: loss = 1.77812 (* 1 = 1.77812 loss) | |
| I1022 10:54:06.392014 8536 sgd_solver.cpp:106] Iteration 4100, lr = 0.001 | |
| I1022 10:54:16.879597 8536 solver.cpp:236] Iteration 4200, loss = 1.90906 | |
| I1022 10:54:16.879633 8536 solver.cpp:252] Train net output #0: loss = 1.90906 (* 1 = 1.90906 loss) | |
| I1022 10:54:16.879639 8536 sgd_solver.cpp:106] Iteration 4200, lr = 0.001 | |
| I1022 10:54:27.365455 8536 solver.cpp:236] Iteration 4300, loss = 1.70922 | |
| I1022 10:54:27.365483 8536 solver.cpp:252] Train net output #0: loss = 1.70922 (* 1 = 1.70922 loss) | |
| I1022 10:54:27.365488 8536 sgd_solver.cpp:106] Iteration 4300, lr = 0.001 | |
| I1022 10:54:37.859992 8536 solver.cpp:236] Iteration 4400, loss = 1.79718 | |
| I1022 10:54:37.860100 8536 solver.cpp:252] Train net output #0: loss = 1.79718 (* 1 = 1.79718 loss) | |
| I1022 10:54:37.860105 8536 sgd_solver.cpp:106] Iteration 4400, lr = 0.001 | |
| I1022 10:54:48.347522 8536 solver.cpp:236] Iteration 4500, loss = 1.95083 | |
| I1022 10:54:48.347558 8536 solver.cpp:252] Train net output #0: loss = 1.95083 (* 1 = 1.95083 loss) | |
| I1022 10:54:48.347563 8536 sgd_solver.cpp:106] Iteration 4500, lr = 0.001 | |
| I1022 10:54:58.794389 8536 solver.cpp:236] Iteration 4600, loss = 1.76111 | |
| I1022 10:54:58.794441 8536 solver.cpp:252] Train net output #0: loss = 1.76111 (* 1 = 1.76111 loss) | |
| I1022 10:54:58.794453 8536 sgd_solver.cpp:106] Iteration 4600, lr = 0.001 | |
| I1022 10:55:09.395520 8536 solver.cpp:236] Iteration 4700, loss = 1.85389 | |
| I1022 10:55:09.395634 8536 solver.cpp:252] Train net output #0: loss = 1.85389 (* 1 = 1.85389 loss) | |
| I1022 10:55:09.395640 8536 sgd_solver.cpp:106] Iteration 4700, lr = 0.001 | |
| I1022 10:55:19.962901 8536 solver.cpp:236] Iteration 4800, loss = 1.69763 | |
| I1022 10:55:19.962941 8536 solver.cpp:252] Train net output #0: loss = 1.69763 (* 1 = 1.69763 loss) | |
| I1022 10:55:19.962945 8536 sgd_solver.cpp:106] Iteration 4800, lr = 0.001 | |
| I1022 10:55:30.501989 8536 solver.cpp:236] Iteration 4900, loss = 1.7594 | |
| I1022 10:55:30.502025 8536 solver.cpp:252] Train net output #0: loss = 1.7594 (* 1 = 1.7594 loss) | |
| I1022 10:55:30.502030 8536 sgd_solver.cpp:106] Iteration 4900, lr = 0.001 | |
| I1022 10:55:40.928983 8536 solver.cpp:340] Iteration 5000, Testing net (#0) | |
| I1022 10:55:45.998831 8536 solver.cpp:408] Test net output #0: accuracy = 0.308 | |
| I1022 10:55:45.998867 8536 solver.cpp:408] Test net output #1: loss = 1.80463 (* 1 = 1.80463 loss) | |
| I1022 10:55:46.066536 8536 solver.cpp:236] Iteration 5000, loss = 1.93135 | |
| I1022 10:55:46.066562 8536 solver.cpp:252] Train net output #0: loss = 1.93135 (* 1 = 1.93135 loss) | |
| I1022 10:55:46.066568 8536 sgd_solver.cpp:106] Iteration 5000, lr = 0.001 | |
| I1022 10:55:56.585947 8536 solver.cpp:236] Iteration 5100, loss = 1.69818 | |
| I1022 10:55:56.585983 8536 solver.cpp:252] Train net output #0: loss = 1.69818 (* 1 = 1.69818 loss) | |
| I1022 10:55:56.585988 8536 sgd_solver.cpp:106] Iteration 5100, lr = 0.001 | |
| I1022 10:56:07.120666 8536 solver.cpp:236] Iteration 5200, loss = 1.86811 | |
| I1022 10:56:07.120702 8536 solver.cpp:252] Train net output #0: loss = 1.86811 (* 1 = 1.86811 loss) | |
| I1022 10:56:07.120707 8536 sgd_solver.cpp:106] Iteration 5200, lr = 0.001 | |
| I1022 10:56:17.630897 8536 solver.cpp:236] Iteration 5300, loss = 1.6767 | |
| I1022 10:56:17.630985 8536 solver.cpp:252] Train net output #0: loss = 1.6767 (* 1 = 1.6767 loss) | |
| I1022 10:56:17.630997 8536 sgd_solver.cpp:106] Iteration 5300, lr = 0.001 | |
| I1022 10:56:28.141518 8536 solver.cpp:236] Iteration 5400, loss = 1.74997 | |
| I1022 10:56:28.141554 8536 solver.cpp:252] Train net output #0: loss = 1.74997 (* 1 = 1.74997 loss) | |
| I1022 10:56:28.141561 8536 sgd_solver.cpp:106] Iteration 5400, lr = 0.001 | |
| I1022 10:56:38.661532 8536 solver.cpp:236] Iteration 5500, loss = 1.92415 | |
| I1022 10:56:38.661571 8536 solver.cpp:252] Train net output #0: loss = 1.92415 (* 1 = 1.92415 loss) | |
| I1022 10:56:38.661576 8536 sgd_solver.cpp:106] Iteration 5500, lr = 0.001 | |
| I1022 10:56:49.181596 8536 solver.cpp:236] Iteration 5600, loss = 1.69164 | |
| I1022 10:56:49.181702 8536 solver.cpp:252] Train net output #0: loss = 1.69164 (* 1 = 1.69164 loss) | |
| I1022 10:56:49.181707 8536 sgd_solver.cpp:106] Iteration 5600, lr = 0.001 | |
| I1022 10:56:59.446550 8536 solver.cpp:236] Iteration 5700, loss = 1.87641 | |
| I1022 10:56:59.446586 8536 solver.cpp:252] Train net output #0: loss = 1.87641 (* 1 = 1.87641 loss) | |
| I1022 10:56:59.446593 8536 sgd_solver.cpp:106] Iteration 5700, lr = 0.001 | |
| I1022 10:57:10.012836 8536 solver.cpp:236] Iteration 5800, loss = 1.66923 | |
| I1022 10:57:10.012872 8536 solver.cpp:252] Train net output #0: loss = 1.66923 (* 1 = 1.66923 loss) | |
| I1022 10:57:10.012878 8536 sgd_solver.cpp:106] Iteration 5800, lr = 0.001 | |
| I1022 10:57:20.575670 8536 solver.cpp:236] Iteration 5900, loss = 1.72808 | |
| I1022 10:57:20.575757 8536 solver.cpp:252] Train net output #0: loss = 1.72808 (* 1 = 1.72808 loss) | |
| I1022 10:57:20.575763 8536 sgd_solver.cpp:106] Iteration 5900, lr = 0.001 | |
| I1022 10:57:31.017033 8536 solver.cpp:340] Iteration 6000, Testing net (#0) | |
| I1022 10:57:36.112232 8536 solver.cpp:408] Test net output #0: accuracy = 0.3242 | |
| I1022 10:57:36.112277 8536 solver.cpp:408] Test net output #1: loss = 1.77918 (* 1 = 1.77918 loss) | |
| I1022 10:57:36.181268 8536 solver.cpp:236] Iteration 6000, loss = 1.90746 | |
| I1022 10:57:36.181306 8536 solver.cpp:252] Train net output #0: loss = 1.90746 (* 1 = 1.90746 loss) | |
| I1022 10:57:36.181313 8536 sgd_solver.cpp:106] Iteration 6000, lr = 0.001 | |
| I1022 10:57:46.716547 8536 solver.cpp:236] Iteration 6100, loss = 1.68754 | |
| I1022 10:57:46.716584 8536 solver.cpp:252] Train net output #0: loss = 1.68754 (* 1 = 1.68754 loss) | |
| I1022 10:57:46.716590 8536 sgd_solver.cpp:106] Iteration 6100, lr = 0.001 | |
| I1022 10:57:57.256057 8536 solver.cpp:236] Iteration 6200, loss = 1.8501 | |
| I1022 10:57:57.256147 8536 solver.cpp:252] Train net output #0: loss = 1.8501 (* 1 = 1.8501 loss) | |
| I1022 10:57:57.256153 8536 sgd_solver.cpp:106] Iteration 6200, lr = 0.001 | |
| I1022 10:58:07.789499 8536 solver.cpp:236] Iteration 6300, loss = 1.66586 | |
| I1022 10:58:07.789525 8536 solver.cpp:252] Train net output #0: loss = 1.66586 (* 1 = 1.66586 loss) | |
| I1022 10:58:07.789531 8536 sgd_solver.cpp:106] Iteration 6300, lr = 0.001 | |
| I1022 10:58:18.372783 8536 solver.cpp:236] Iteration 6400, loss = 1.72565 | |
| I1022 10:58:18.372819 8536 solver.cpp:252] Train net output #0: loss = 1.72565 (* 1 = 1.72565 loss) | |
| I1022 10:58:18.372825 8536 sgd_solver.cpp:106] Iteration 6400, lr = 0.001 | |
| I1022 10:58:28.895227 8536 solver.cpp:236] Iteration 6500, loss = 1.91026 | |
| I1022 10:58:28.895318 8536 solver.cpp:252] Train net output #0: loss = 1.91026 (* 1 = 1.91026 loss) | |
| I1022 10:58:28.895333 8536 sgd_solver.cpp:106] Iteration 6500, lr = 0.001 | |
| I1022 10:58:39.382961 8536 solver.cpp:236] Iteration 6600, loss = 1.66892 | |
| I1022 10:58:39.382988 8536 solver.cpp:252] Train net output #0: loss = 1.66892 (* 1 = 1.66892 loss) | |
| I1022 10:58:39.382994 8536 sgd_solver.cpp:106] Iteration 6600, lr = 0.001 | |
| I1022 10:58:49.741792 8536 solver.cpp:236] Iteration 6700, loss = 1.86964 | |
| I1022 10:58:49.741818 8536 solver.cpp:252] Train net output #0: loss = 1.86964 (* 1 = 1.86964 loss) | |
| I1022 10:58:49.741824 8536 sgd_solver.cpp:106] Iteration 6700, lr = 0.001 | |
| I1022 10:59:00.210350 8536 solver.cpp:236] Iteration 6800, loss = 1.65025 | |
| I1022 10:59:00.210438 8536 solver.cpp:252] Train net output #0: loss = 1.65025 (* 1 = 1.65025 loss) | |
| I1022 10:59:00.210445 8536 sgd_solver.cpp:106] Iteration 6800, lr = 0.001 | |
| I1022 10:59:10.731395 8536 solver.cpp:236] Iteration 6900, loss = 1.68919 | |
| I1022 10:59:10.731431 8536 solver.cpp:252] Train net output #0: loss = 1.68919 (* 1 = 1.68919 loss) | |
| I1022 10:59:10.731437 8536 sgd_solver.cpp:106] Iteration 6900, lr = 0.001 | |
| I1022 10:59:21.131471 8536 solver.cpp:340] Iteration 7000, Testing net (#0) | |
| I1022 10:59:26.214035 8536 solver.cpp:408] Test net output #0: accuracy = 0.3266 | |
| I1022 10:59:26.214061 8536 solver.cpp:408] Test net output #1: loss = 1.76517 (* 1 = 1.76517 loss) | |
| I1022 10:59:26.280478 8536 solver.cpp:236] Iteration 7000, loss = 1.8829 | |
| I1022 10:59:26.280509 8536 solver.cpp:252] Train net output #0: loss = 1.8829 (* 1 = 1.8829 loss) | |
| I1022 10:59:26.280515 8536 sgd_solver.cpp:106] Iteration 7000, lr = 0.001 | |
| I1022 10:59:36.885742 8536 solver.cpp:236] Iteration 7100, loss = 1.6448 | |
| I1022 10:59:36.885826 8536 solver.cpp:252] Train net output #0: loss = 1.6448 (* 1 = 1.6448 loss) | |
| I1022 10:59:36.885838 8536 sgd_solver.cpp:106] Iteration 7100, lr = 0.001 | |
| I1022 10:59:47.461748 8536 solver.cpp:236] Iteration 7200, loss = 1.87966 | |
| I1022 10:59:47.461779 8536 solver.cpp:252] Train net output #0: loss = 1.87966 (* 1 = 1.87966 loss) | |
| I1022 10:59:47.461784 8536 sgd_solver.cpp:106] Iteration 7200, lr = 0.001 | |
| I1022 10:59:57.976933 8536 solver.cpp:236] Iteration 7300, loss = 1.65232 | |
| I1022 10:59:57.976960 8536 solver.cpp:252] Train net output #0: loss = 1.65232 (* 1 = 1.65232 loss) | |
| I1022 10:59:57.976965 8536 sgd_solver.cpp:106] Iteration 7300, lr = 0.001 | |
| I1022 11:00:08.474916 8536 solver.cpp:236] Iteration 7400, loss = 1.69676 | |
| I1022 11:00:08.475021 8536 solver.cpp:252] Train net output #0: loss = 1.69676 (* 1 = 1.69676 loss) | |
| I1022 11:00:08.475033 8536 sgd_solver.cpp:106] Iteration 7400, lr = 0.001 | |
| I1022 11:00:19.075892 8536 solver.cpp:236] Iteration 7500, loss = 1.85232 | |
| I1022 11:00:19.075929 8536 solver.cpp:252] Train net output #0: loss = 1.85232 (* 1 = 1.85232 loss) | |
| I1022 11:00:19.075934 8536 sgd_solver.cpp:106] Iteration 7500, lr = 0.001 | |
| I1022 11:00:29.613118 8536 solver.cpp:236] Iteration 7600, loss = 1.6578 | |
| I1022 11:00:29.613145 8536 solver.cpp:252] Train net output #0: loss = 1.6578 (* 1 = 1.6578 loss) | |
| I1022 11:00:29.613150 8536 sgd_solver.cpp:106] Iteration 7600, lr = 0.001 | |
| I1022 11:00:40.122006 8536 solver.cpp:236] Iteration 7700, loss = 1.81921 | |
| I1022 11:00:40.122114 8536 solver.cpp:252] Train net output #0: loss = 1.81921 (* 1 = 1.81921 loss) | |
| I1022 11:00:40.122129 8536 sgd_solver.cpp:106] Iteration 7700, lr = 0.001 | |
| I1022 11:00:50.393288 8536 solver.cpp:236] Iteration 7800, loss = 1.63087 | |
| I1022 11:00:50.393326 8536 solver.cpp:252] Train net output #0: loss = 1.63087 (* 1 = 1.63087 loss) | |
| I1022 11:00:50.393331 8536 sgd_solver.cpp:106] Iteration 7800, lr = 0.001 | |
| I1022 11:01:00.896307 8536 solver.cpp:236] Iteration 7900, loss = 1.68231 | |
| I1022 11:01:00.896335 8536 solver.cpp:252] Train net output #0: loss = 1.68231 (* 1 = 1.68231 loss) | |
| I1022 11:01:00.896340 8536 sgd_solver.cpp:106] Iteration 7900, lr = 0.001 | |
| I1022 11:01:11.294525 8536 solver.cpp:340] Iteration 8000, Testing net (#0) | |
| I1022 11:01:16.378751 8536 solver.cpp:408] Test net output #0: accuracy = 0.3361 | |
| I1022 11:01:16.378777 8536 solver.cpp:408] Test net output #1: loss = 1.75532 (* 1 = 1.75532 loss) | |
| I1022 11:01:16.446626 8536 solver.cpp:236] Iteration 8000, loss = 1.87086 | |
| I1022 11:01:16.446662 8536 solver.cpp:252] Train net output #0: loss = 1.87086 (* 1 = 1.87086 loss) | |
| I1022 11:01:16.446668 8536 sgd_solver.cpp:106] Iteration 8000, lr = 0.001 | |
| I1022 11:01:26.943320 8536 solver.cpp:236] Iteration 8100, loss = 1.64373 | |
| I1022 11:01:26.943356 8536 solver.cpp:252] Train net output #0: loss = 1.64373 (* 1 = 1.64373 loss) | |
| I1022 11:01:26.943361 8536 sgd_solver.cpp:106] Iteration 8100, lr = 0.001 | |
| I1022 11:01:37.441681 8536 solver.cpp:236] Iteration 8200, loss = 1.81089 | |
| I1022 11:01:37.441720 8536 solver.cpp:252] Train net output #0: loss = 1.81089 (* 1 = 1.81089 loss) | |
| I1022 11:01:37.441725 8536 sgd_solver.cpp:106] Iteration 8200, lr = 0.001 | |
| I1022 11:01:47.950942 8536 solver.cpp:236] Iteration 8300, loss = 1.62866 | |
| I1022 11:01:47.951035 8536 solver.cpp:252] Train net output #0: loss = 1.62866 (* 1 = 1.62866 loss) | |
| I1022 11:01:47.951040 8536 sgd_solver.cpp:106] Iteration 8300, lr = 0.001 | |
| I1022 11:01:58.455128 8536 solver.cpp:236] Iteration 8400, loss = 1.66818 | |
| I1022 11:01:58.455165 8536 solver.cpp:252] Train net output #0: loss = 1.66818 (* 1 = 1.66818 loss) | |
| I1022 11:01:58.455171 8536 sgd_solver.cpp:106] Iteration 8400, lr = 0.001 | |
| I1022 11:02:08.955467 8536 solver.cpp:236] Iteration 8500, loss = 1.87937 | |
| I1022 11:02:08.955505 8536 solver.cpp:252] Train net output #0: loss = 1.87937 (* 1 = 1.87937 loss) | |
| I1022 11:02:08.955510 8536 sgd_solver.cpp:106] Iteration 8500, lr = 0.001 | |
| I1022 11:02:19.468601 8536 solver.cpp:236] Iteration 8600, loss = 1.65065 | |
| I1022 11:02:19.468689 8536 solver.cpp:252] Train net output #0: loss = 1.65065 (* 1 = 1.65065 loss) | |
| I1022 11:02:19.468694 8536 sgd_solver.cpp:106] Iteration 8600, lr = 0.001 | |
| I1022 11:02:29.970166 8536 solver.cpp:236] Iteration 8700, loss = 1.83875 | |
| I1022 11:02:29.970202 8536 solver.cpp:252] Train net output #0: loss = 1.83875 (* 1 = 1.83875 loss) | |
| I1022 11:02:29.970208 8536 sgd_solver.cpp:106] Iteration 8700, lr = 0.001 | |
| I1022 11:02:40.421144 8536 solver.cpp:236] Iteration 8800, loss = 1.61695 | |
| I1022 11:02:40.421172 8536 solver.cpp:252] Train net output #0: loss = 1.61695 (* 1 = 1.61695 loss) | |
| I1022 11:02:40.421178 8536 sgd_solver.cpp:106] Iteration 8800, lr = 0.001 | |
| I1022 11:02:50.732916 8536 solver.cpp:236] Iteration 8900, loss = 1.66214 | |
| I1022 11:02:50.733006 8536 solver.cpp:252] Train net output #0: loss = 1.66214 (* 1 = 1.66214 loss) | |
| I1022 11:02:50.733011 8536 sgd_solver.cpp:106] Iteration 8900, lr = 0.001 | |
| I1022 11:03:01.170263 8536 solver.cpp:340] Iteration 9000, Testing net (#0) | |
| I1022 11:03:06.284602 8536 solver.cpp:408] Test net output #0: accuracy = 0.3435 | |
| I1022 11:03:06.284641 8536 solver.cpp:408] Test net output #1: loss = 1.74518 (* 1 = 1.74518 loss) | |
| I1022 11:03:06.351274 8536 solver.cpp:236] Iteration 9000, loss = 1.86458 | |
| I1022 11:03:06.351312 8536 solver.cpp:252] Train net output #0: loss = 1.86458 (* 1 = 1.86458 loss) | |
| I1022 11:03:06.351318 8536 sgd_solver.cpp:106] Iteration 9000, lr = 0.001 | |
| I1022 11:03:16.893996 8536 solver.cpp:236] Iteration 9100, loss = 1.63709 | |
| I1022 11:03:16.894035 8536 solver.cpp:252] Train net output #0: loss = 1.63709 (* 1 = 1.63709 loss) | |
| I1022 11:03:16.894042 8536 sgd_solver.cpp:106] Iteration 9100, lr = 0.001 | |
| I1022 11:03:27.532582 8536 solver.cpp:236] Iteration 9200, loss = 1.81429 | |
| I1022 11:03:27.532681 8536 solver.cpp:252] Train net output #0: loss = 1.81429 (* 1 = 1.81429 loss) | |
| I1022 11:03:27.532687 8536 sgd_solver.cpp:106] Iteration 9200, lr = 0.001 | |
| I1022 11:03:38.066777 8536 solver.cpp:236] Iteration 9300, loss = 1.60886 | |
| I1022 11:03:38.066812 8536 solver.cpp:252] Train net output #0: loss = 1.60886 (* 1 = 1.60886 loss) | |
| I1022 11:03:38.066818 8536 sgd_solver.cpp:106] Iteration 9300, lr = 0.001 | |
| I1022 11:03:48.565462 8536 solver.cpp:236] Iteration 9400, loss = 1.65687 | |
| I1022 11:03:48.565500 8536 solver.cpp:252] Train net output #0: loss = 1.65687 (* 1 = 1.65687 loss) | |
| I1022 11:03:48.565505 8536 sgd_solver.cpp:106] Iteration 9400, lr = 0.001 | |
| I1022 11:03:59.042943 8536 solver.cpp:236] Iteration 9500, loss = 1.87088 | |
| I1022 11:03:59.043051 8536 solver.cpp:252] Train net output #0: loss = 1.87088 (* 1 = 1.87088 loss) | |
| I1022 11:03:59.043056 8536 sgd_solver.cpp:106] Iteration 9500, lr = 0.001 | |
| I1022 11:04:09.542875 8536 solver.cpp:236] Iteration 9600, loss = 1.62396 | |
| I1022 11:04:09.542901 8536 solver.cpp:252] Train net output #0: loss = 1.62396 (* 1 = 1.62396 loss) | |
| I1022 11:04:09.542906 8536 sgd_solver.cpp:106] Iteration 9600, lr = 0.001 | |
| I1022 11:04:20.051120 8536 solver.cpp:236] Iteration 9700, loss = 1.82969 | |
| I1022 11:04:20.051158 8536 solver.cpp:252] Train net output #0: loss = 1.82969 (* 1 = 1.82969 loss) | |
| I1022 11:04:20.051164 8536 sgd_solver.cpp:106] Iteration 9700, lr = 0.001 | |
| I1022 11:04:30.550485 8536 solver.cpp:236] Iteration 9800, loss = 1.61907 | |
| I1022 11:04:30.550571 8536 solver.cpp:252] Train net output #0: loss = 1.61907 (* 1 = 1.61907 loss) | |
| I1022 11:04:30.550576 8536 sgd_solver.cpp:106] Iteration 9800, lr = 0.001 | |
| I1022 11:04:40.839673 8536 solver.cpp:236] Iteration 9900, loss = 1.64604 | |
| I1022 11:04:40.839700 8536 solver.cpp:252] Train net output #0: loss = 1.64604 (* 1 = 1.64604 loss) | |
| I1022 11:04:40.839706 8536 sgd_solver.cpp:106] Iteration 9900, lr = 0.001 | |
| I1022 11:04:51.266861 8536 solver.cpp:340] Iteration 10000, Testing net (#0) | |
| I1022 11:04:56.391234 8536 solver.cpp:408] Test net output #0: accuracy = 0.3516 | |
| I1022 11:04:56.391273 8536 solver.cpp:408] Test net output #1: loss = 1.72936 (* 1 = 1.72936 loss) | |
| I1022 11:04:56.461269 8536 solver.cpp:236] Iteration 10000, loss = 1.86576 | |
| I1022 11:04:56.461308 8536 solver.cpp:252] Train net output #0: loss = 1.86576 (* 1 = 1.86576 loss) | |
| I1022 11:04:56.461313 8536 sgd_solver.cpp:106] Iteration 10000, lr = 0.001 | |
| I1022 11:05:06.995796 8536 solver.cpp:236] Iteration 10100, loss = 1.61587 | |
| I1022 11:05:06.995857 8536 solver.cpp:252] Train net output #0: loss = 1.61587 (* 1 = 1.61587 loss) | |
| I1022 11:05:06.995862 8536 sgd_solver.cpp:106] Iteration 10100, lr = 0.001 | |
| I1022 11:05:17.510552 8536 solver.cpp:236] Iteration 10200, loss = 1.79931 | |
| I1022 11:05:17.510587 8536 solver.cpp:252] Train net output #0: loss = 1.79931 (* 1 = 1.79931 loss) | |
| I1022 11:05:17.510592 8536 sgd_solver.cpp:106] Iteration 10200, lr = 0.001 | |
| I1022 11:05:28.022467 8536 solver.cpp:236] Iteration 10300, loss = 1.5957 | |
| I1022 11:05:28.022495 8536 solver.cpp:252] Train net output #0: loss = 1.5957 (* 1 = 1.5957 loss) | |
| I1022 11:05:28.022500 8536 sgd_solver.cpp:106] Iteration 10300, lr = 0.001 | |
| I1022 11:05:38.533931 8536 solver.cpp:236] Iteration 10400, loss = 1.6241 | |
| I1022 11:05:38.534000 8536 solver.cpp:252] Train net output #0: loss = 1.6241 (* 1 = 1.6241 loss) | |
| I1022 11:05:38.534006 8536 sgd_solver.cpp:106] Iteration 10400, lr = 0.001 | |
| I1022 11:05:49.132987 8536 solver.cpp:236] Iteration 10500, loss = 1.84335 | |
| I1022 11:05:49.133025 8536 solver.cpp:252] Train net output #0: loss = 1.84335 (* 1 = 1.84335 loss) | |
| I1022 11:05:49.133030 8536 sgd_solver.cpp:106] Iteration 10500, lr = 0.001 | |
| I1022 11:05:59.760514 8536 solver.cpp:236] Iteration 10600, loss = 1.59885 | |
| I1022 11:05:59.760542 8536 solver.cpp:252] Train net output #0: loss = 1.59885 (* 1 = 1.59885 loss) | |
| I1022 11:05:59.760547 8536 sgd_solver.cpp:106] Iteration 10600, lr = 0.001 | |
| I1022 11:06:10.231766 8536 solver.cpp:236] Iteration 10700, loss = 1.8319 | |
| I1022 11:06:10.231876 8536 solver.cpp:252] Train net output #0: loss = 1.8319 (* 1 = 1.8319 loss) | |
| I1022 11:06:10.231884 8536 sgd_solver.cpp:106] Iteration 10700, lr = 0.001 | |
| I1022 11:06:20.699128 8536 solver.cpp:236] Iteration 10800, loss = 1.57596 | |
| I1022 11:06:20.699163 8536 solver.cpp:252] Train net output #0: loss = 1.57596 (* 1 = 1.57596 loss) | |
| I1022 11:06:20.699168 8536 sgd_solver.cpp:106] Iteration 10800, lr = 0.001 | |
| I1022 11:06:31.177182 8536 solver.cpp:236] Iteration 10900, loss = 1.61976 | |
| I1022 11:06:31.177217 8536 solver.cpp:252] Train net output #0: loss = 1.61976 (* 1 = 1.61976 loss) | |
| I1022 11:06:31.177222 8536 sgd_solver.cpp:106] Iteration 10900, lr = 0.001 | |
| I1022 11:06:41.333619 8536 solver.cpp:340] Iteration 11000, Testing net (#0) | |
| I1022 11:06:46.432960 8536 solver.cpp:408] Test net output #0: accuracy = 0.3564 | |
| I1022 11:06:46.432986 8536 solver.cpp:408] Test net output #1: loss = 1.71685 (* 1 = 1.71685 loss) | |
| I1022 11:06:46.501219 8536 solver.cpp:236] Iteration 11000, loss = 1.85086 | |
| I1022 11:06:46.501255 8536 solver.cpp:252] Train net output #0: loss = 1.85086 (* 1 = 1.85086 loss) | |
| I1022 11:06:46.501260 8536 sgd_solver.cpp:106] Iteration 11000, lr = 0.001 | |
| I1022 11:06:57.333284 8536 solver.cpp:236] Iteration 11100, loss = 1.60589 | |
| I1022 11:06:57.333320 8536 solver.cpp:252] Train net output #0: loss = 1.60589 (* 1 = 1.60589 loss) | |
| I1022 11:06:57.333325 8536 sgd_solver.cpp:106] Iteration 11100, lr = 0.001 | |
| I1022 11:07:08.047196 8536 solver.cpp:236] Iteration 11200, loss = 1.83182 | |
| I1022 11:07:08.047233 8536 solver.cpp:252] Train net output #0: loss = 1.83182 (* 1 = 1.83182 loss) | |
| I1022 11:07:08.047240 8536 sgd_solver.cpp:106] Iteration 11200, lr = 0.001 | |
| I1022 11:07:18.816196 8536 solver.cpp:236] Iteration 11300, loss = 1.61551 | |
| I1022 11:07:18.816288 8536 solver.cpp:252] Train net output #0: loss = 1.61551 (* 1 = 1.61551 loss) | |
| I1022 11:07:18.816294 8536 sgd_solver.cpp:106] Iteration 11300, lr = 0.001 | |
| I1022 11:07:29.498409 8536 solver.cpp:236] Iteration 11400, loss = 1.59523 | |
| I1022 11:07:29.498445 8536 solver.cpp:252] Train net output #0: loss = 1.59523 (* 1 = 1.59523 loss) | |
| I1022 11:07:29.498451 8536 sgd_solver.cpp:106] Iteration 11400, lr = 0.001 | |
| I1022 11:07:40.003278 8536 solver.cpp:236] Iteration 11500, loss = 1.84488 | |
| I1022 11:07:40.003315 8536 solver.cpp:252] Train net output #0: loss = 1.84488 (* 1 = 1.84488 loss) | |
| I1022 11:07:40.003320 8536 sgd_solver.cpp:106] Iteration 11500, lr = 0.001 | |
| I1022 11:07:50.503192 8536 solver.cpp:236] Iteration 11600, loss = 1.58921 | |
| I1022 11:07:50.503299 8536 solver.cpp:252] Train net output #0: loss = 1.58921 (* 1 = 1.58921 loss) | |
| I1022 11:07:50.503306 8536 sgd_solver.cpp:106] Iteration 11600, lr = 0.001 | |
| I1022 11:08:00.999313 8536 solver.cpp:236] Iteration 11700, loss = 1.79409 | |
| I1022 11:08:00.999351 8536 solver.cpp:252] Train net output #0: loss = 1.79409 (* 1 = 1.79409 loss) | |
| I1022 11:08:00.999357 8536 sgd_solver.cpp:106] Iteration 11700, lr = 0.001 | |
| I1022 11:08:11.489598 8536 solver.cpp:236] Iteration 11800, loss = 1.60815 | |
| I1022 11:08:11.489636 8536 solver.cpp:252] Train net output #0: loss = 1.60815 (* 1 = 1.60815 loss) | |
| I1022 11:08:11.489641 8536 sgd_solver.cpp:106] Iteration 11800, lr = 0.001 | |
| I1022 11:08:21.969847 8536 solver.cpp:236] Iteration 11900, loss = 1.59865 | |
| I1022 11:08:21.969949 8536 solver.cpp:252] Train net output #0: loss = 1.59865 (* 1 = 1.59865 loss) | |
| I1022 11:08:21.969955 8536 sgd_solver.cpp:106] Iteration 11900, lr = 0.001 | |
| I1022 11:08:32.275290 8536 solver.cpp:340] Iteration 12000, Testing net (#0) | |
| I1022 11:08:37.402657 8536 solver.cpp:408] Test net output #0: accuracy = 0.3667 | |
| I1022 11:08:37.402709 8536 solver.cpp:408] Test net output #1: loss = 1.69912 (* 1 = 1.69912 loss) | |
| I1022 11:08:37.464241 8536 solver.cpp:236] Iteration 12000, loss = 1.84505 | |
| I1022 11:08:37.464270 8536 solver.cpp:252] Train net output #0: loss = 1.84505 (* 1 = 1.84505 loss) | |
| I1022 11:08:37.464277 8536 sgd_solver.cpp:106] Iteration 12000, lr = 0.001 | |
| I1022 11:08:48.033339 8536 solver.cpp:236] Iteration 12100, loss = 1.59986 | |
| I1022 11:08:48.033368 8536 solver.cpp:252] Train net output #0: loss = 1.59986 (* 1 = 1.59986 loss) | |
| I1022 11:08:48.033373 8536 sgd_solver.cpp:106] Iteration 12100, lr = 0.001 | |
| I1022 11:08:58.613881 8536 solver.cpp:236] Iteration 12200, loss = 1.80453 | |
| I1022 11:08:58.613993 8536 solver.cpp:252] Train net output #0: loss = 1.80453 (* 1 = 1.80453 loss) | |
| I1022 11:08:58.614008 8536 sgd_solver.cpp:106] Iteration 12200, lr = 0.001 | |
| I1022 11:09:09.184016 8536 solver.cpp:236] Iteration 12300, loss = 1.5932 | |
| I1022 11:09:09.184044 8536 solver.cpp:252] Train net output #0: loss = 1.5932 (* 1 = 1.5932 loss) | |
| I1022 11:09:09.184049 8536 sgd_solver.cpp:106] Iteration 12300, lr = 0.001 | |
| I1022 11:09:19.737473 8536 solver.cpp:236] Iteration 12400, loss = 1.592 | |
| I1022 11:09:19.737510 8536 solver.cpp:252] Train net output #0: loss = 1.592 (* 1 = 1.592 loss) | |
| I1022 11:09:19.737515 8536 sgd_solver.cpp:106] Iteration 12400, lr = 0.001 | |
| I1022 11:09:30.287920 8536 solver.cpp:236] Iteration 12500, loss = 1.83297 | |
| I1022 11:09:30.288010 8536 solver.cpp:252] Train net output #0: loss = 1.83297 (* 1 = 1.83297 loss) | |
| I1022 11:09:30.288017 8536 sgd_solver.cpp:106] Iteration 12500, lr = 0.001 | |
| I1022 11:09:40.788943 8536 solver.cpp:236] Iteration 12600, loss = 1.58538 | |
| I1022 11:09:40.788976 8536 solver.cpp:252] Train net output #0: loss = 1.58538 (* 1 = 1.58538 loss) | |
| I1022 11:09:40.788981 8536 sgd_solver.cpp:106] Iteration 12600, lr = 0.001 | |
| I1022 11:09:51.276540 8536 solver.cpp:236] Iteration 12700, loss = 1.75173 | |
| I1022 11:09:51.276576 8536 solver.cpp:252] Train net output #0: loss = 1.75173 (* 1 = 1.75173 loss) | |
| I1022 11:09:51.276581 8536 sgd_solver.cpp:106] Iteration 12700, lr = 0.001 | |
| I1022 11:10:01.787467 8536 solver.cpp:236] Iteration 12800, loss = 1.58274 | |
| I1022 11:10:01.787518 8536 solver.cpp:252] Train net output #0: loss = 1.58274 (* 1 = 1.58274 loss) | |
| I1022 11:10:01.787523 8536 sgd_solver.cpp:106] Iteration 12800, lr = 0.001 | |
| I1022 11:10:12.301200 8536 solver.cpp:236] Iteration 12900, loss = 1.56434 | |
| I1022 11:10:12.301235 8536 solver.cpp:252] Train net output #0: loss = 1.56434 (* 1 = 1.56434 loss) | |
| I1022 11:10:12.301240 8536 sgd_solver.cpp:106] Iteration 12900, lr = 0.001 | |
| I1022 11:10:22.702301 8536 solver.cpp:340] Iteration 13000, Testing net (#0) | |
| I1022 11:10:27.768751 8536 solver.cpp:408] Test net output #0: accuracy = 0.3836 | |
| I1022 11:10:27.768797 8536 solver.cpp:408] Test net output #1: loss = 1.67065 (* 1 = 1.67065 loss) | |
| I1022 11:10:27.828917 8536 solver.cpp:236] Iteration 13000, loss = 1.785 | |
| I1022 11:10:27.828958 8536 solver.cpp:252] Train net output #0: loss = 1.785 (* 1 = 1.785 loss) | |
| I1022 11:10:27.828971 8536 sgd_solver.cpp:106] Iteration 13000, lr = 0.001 | |
| I1022 11:10:38.977177 8536 solver.cpp:236] Iteration 13100, loss = 1.55043 | |
| I1022 11:10:38.977267 8536 solver.cpp:252] Train net output #0: loss = 1.55043 (* 1 = 1.55043 loss) | |
| I1022 11:10:38.977272 8536 sgd_solver.cpp:106] Iteration 13100, lr = 0.001 | |
| I1022 11:10:49.519579 8536 solver.cpp:236] Iteration 13200, loss = 1.75727 | |
| I1022 11:10:49.519615 8536 solver.cpp:252] Train net output #0: loss = 1.75727 (* 1 = 1.75727 loss) | |
| I1022 11:10:49.519621 8536 sgd_solver.cpp:106] Iteration 13200, lr = 0.001 | |
| I1022 11:10:59.970903 8536 solver.cpp:236] Iteration 13300, loss = 1.55308 | |
| I1022 11:10:59.970929 8536 solver.cpp:252] Train net output #0: loss = 1.55308 (* 1 = 1.55308 loss) | |
| I1022 11:10:59.970935 8536 sgd_solver.cpp:106] Iteration 13300, lr = 0.001 | |
| I1022 11:11:10.417140 8536 solver.cpp:236] Iteration 13400, loss = 1.53396 | |
| I1022 11:11:10.417229 8536 solver.cpp:252] Train net output #0: loss = 1.53396 (* 1 = 1.53396 loss) | |
| I1022 11:11:10.417244 8536 sgd_solver.cpp:106] Iteration 13400, lr = 0.001 | |
| I1022 11:11:20.927251 8536 solver.cpp:236] Iteration 13500, loss = 1.77538 | |
| I1022 11:11:20.927286 8536 solver.cpp:252] Train net output #0: loss = 1.77538 (* 1 = 1.77538 loss) | |
| I1022 11:11:20.927292 8536 sgd_solver.cpp:106] Iteration 13500, lr = 0.001 | |
| I1022 11:11:31.486376 8536 solver.cpp:236] Iteration 13600, loss = 1.52669 | |
| I1022 11:11:31.486413 8536 solver.cpp:252] Train net output #0: loss = 1.52669 (* 1 = 1.52669 loss) | |
| I1022 11:11:31.486420 8536 sgd_solver.cpp:106] Iteration 13600, lr = 0.001 | |
| I1022 11:11:42.005573 8536 solver.cpp:236] Iteration 13700, loss = 1.77915 | |
| I1022 11:11:42.005667 8536 solver.cpp:252] Train net output #0: loss = 1.77915 (* 1 = 1.77915 loss) | |
| I1022 11:11:42.005672 8536 sgd_solver.cpp:106] Iteration 13700, lr = 0.001 | |
| I1022 11:11:52.551605 8536 solver.cpp:236] Iteration 13800, loss = 1.53946 | |
| I1022 11:11:52.551642 8536 solver.cpp:252] Train net output #0: loss = 1.53946 (* 1 = 1.53946 loss) | |
| I1022 11:11:52.551650 8536 sgd_solver.cpp:106] Iteration 13800, lr = 0.001 | |
| I1022 11:12:03.124747 8536 solver.cpp:236] Iteration 13900, loss = 1.52055 | |
| I1022 11:12:03.124783 8536 solver.cpp:252] Train net output #0: loss = 1.52055 (* 1 = 1.52055 loss) | |
| I1022 11:12:03.124789 8536 sgd_solver.cpp:106] Iteration 13900, lr = 0.001 | |
| I1022 11:12:13.656524 8536 solver.cpp:340] Iteration 14000, Testing net (#0) | |
| I1022 11:12:18.744698 8536 solver.cpp:408] Test net output #0: accuracy = 0.3958 | |
| I1022 11:12:18.744724 8536 solver.cpp:408] Test net output #1: loss = 1.64517 (* 1 = 1.64517 loss) | |
| I1022 11:12:18.813886 8536 solver.cpp:236] Iteration 14000, loss = 1.7507 | |
| I1022 11:12:18.813925 8536 solver.cpp:252] Train net output #0: loss = 1.7507 (* 1 = 1.7507 loss) | |
| I1022 11:12:18.813930 8536 sgd_solver.cpp:106] Iteration 14000, lr = 0.001 | |
| I1022 11:12:29.110478 8536 solver.cpp:236] Iteration 14100, loss = 1.50475 | |
| I1022 11:12:29.110515 8536 solver.cpp:252] Train net output #0: loss = 1.50475 (* 1 = 1.50475 loss) | |
| I1022 11:12:29.110520 8536 sgd_solver.cpp:106] Iteration 14100, lr = 0.001 | |
| I1022 11:12:39.656483 8536 solver.cpp:236] Iteration 14200, loss = 1.72815 | |
| I1022 11:12:39.656520 8536 solver.cpp:252] Train net output #0: loss = 1.72815 (* 1 = 1.72815 loss) | |
| I1022 11:12:39.656525 8536 sgd_solver.cpp:106] Iteration 14200, lr = 0.001 | |
| I1022 11:12:50.243865 8536 solver.cpp:236] Iteration 14300, loss = 1.54549 | |
| I1022 11:12:50.243968 8536 solver.cpp:252] Train net output #0: loss = 1.54549 (* 1 = 1.54549 loss) | |
| I1022 11:12:50.243973 8536 sgd_solver.cpp:106] Iteration 14300, lr = 0.001 | |
| I1022 11:13:00.794697 8536 solver.cpp:236] Iteration 14400, loss = 1.49902 | |
| I1022 11:13:00.794734 8536 solver.cpp:252] Train net output #0: loss = 1.49902 (* 1 = 1.49902 loss) | |
| I1022 11:13:00.794739 8536 sgd_solver.cpp:106] Iteration 14400, lr = 0.001 | |
| I1022 11:13:11.538136 8536 solver.cpp:236] Iteration 14500, loss = 1.73908 | |
| I1022 11:13:11.538175 8536 solver.cpp:252] Train net output #0: loss = 1.73908 (* 1 = 1.73908 loss) | |
| I1022 11:13:11.538180 8536 sgd_solver.cpp:106] Iteration 14500, lr = 0.001 | |
| I1022 11:13:22.039759 8536 solver.cpp:236] Iteration 14600, loss = 1.52042 | |
| I1022 11:13:22.039891 8536 solver.cpp:252] Train net output #0: loss = 1.52042 (* 1 = 1.52042 loss) | |
| I1022 11:13:22.039897 8536 sgd_solver.cpp:106] Iteration 14600, lr = 0.001 | |
| I1022 11:13:32.508842 8536 solver.cpp:236] Iteration 14700, loss = 1.72301 | |
| I1022 11:13:32.508872 8536 solver.cpp:252] Train net output #0: loss = 1.72301 (* 1 = 1.72301 loss) | |
| I1022 11:13:32.508877 8536 sgd_solver.cpp:106] Iteration 14700, lr = 0.001 | |
| I1022 11:13:42.999177 8536 solver.cpp:236] Iteration 14800, loss = 1.54298 | |
| I1022 11:13:42.999207 8536 solver.cpp:252] Train net output #0: loss = 1.54298 (* 1 = 1.54298 loss) | |
| I1022 11:13:42.999212 8536 sgd_solver.cpp:106] Iteration 14800, lr = 0.001 | |
| I1022 11:13:53.459342 8536 solver.cpp:236] Iteration 14900, loss = 1.50204 | |
| I1022 11:13:53.459427 8536 solver.cpp:252] Train net output #0: loss = 1.50204 (* 1 = 1.50204 loss) | |
| I1022 11:13:53.459434 8536 sgd_solver.cpp:106] Iteration 14900, lr = 0.001 | |
| I1022 11:14:03.834810 8536 solver.cpp:340] Iteration 15000, Testing net (#0) | |
| I1022 11:14:09.060551 8536 solver.cpp:408] Test net output #0: accuracy = 0.4042 | |
| I1022 11:14:09.060587 8536 solver.cpp:408] Test net output #1: loss = 1.63019 (* 1 = 1.63019 loss) | |
| I1022 11:14:09.127522 8536 solver.cpp:236] Iteration 15000, loss = 1.73408 | |
| I1022 11:14:09.127557 8536 solver.cpp:252] Train net output #0: loss = 1.73408 (* 1 = 1.73408 loss) | |
| I1022 11:14:09.127563 8536 sgd_solver.cpp:106] Iteration 15000, lr = 0.001 | |
| I1022 11:14:19.661502 8536 solver.cpp:236] Iteration 15100, loss = 1.50655 | |
| I1022 11:14:19.661540 8536 solver.cpp:252] Train net output #0: loss = 1.50655 (* 1 = 1.50655 loss) | |
| I1022 11:14:19.661545 8536 sgd_solver.cpp:106] Iteration 15100, lr = 0.001 | |
| I1022 11:14:29.982547 8536 solver.cpp:236] Iteration 15200, loss = 1.70413 | |
| I1022 11:14:29.982632 8536 solver.cpp:252] Train net output #0: loss = 1.70413 (* 1 = 1.70413 loss) | |
| I1022 11:14:29.982638 8536 sgd_solver.cpp:106] Iteration 15200, lr = 0.001 | |
| I1022 11:14:40.694537 8536 solver.cpp:236] Iteration 15300, loss = 1.5455 | |
| I1022 11:14:40.694574 8536 solver.cpp:252] Train net output #0: loss = 1.5455 (* 1 = 1.5455 loss) | |
| I1022 11:14:40.694581 8536 sgd_solver.cpp:106] Iteration 15300, lr = 0.001 | |
| I1022 11:14:51.381824 8536 solver.cpp:236] Iteration 15400, loss = 1.48214 | |
| I1022 11:14:51.381875 8536 solver.cpp:252] Train net output #0: loss = 1.48214 (* 1 = 1.48214 loss) | |
| I1022 11:14:51.381886 8536 sgd_solver.cpp:106] Iteration 15400, lr = 0.001 | |
| I1022 11:15:01.925511 8536 solver.cpp:236] Iteration 15500, loss = 1.72135 | |
| I1022 11:15:01.925597 8536 solver.cpp:252] Train net output #0: loss = 1.72135 (* 1 = 1.72135 loss) | |
| I1022 11:15:01.925604 8536 sgd_solver.cpp:106] Iteration 15500, lr = 0.001 | |
| I1022 11:15:12.441215 8536 solver.cpp:236] Iteration 15600, loss = 1.50004 | |
| I1022 11:15:12.441249 8536 solver.cpp:252] Train net output #0: loss = 1.50004 (* 1 = 1.50004 loss) | |
| I1022 11:15:12.441256 8536 sgd_solver.cpp:106] Iteration 15600, lr = 0.001 | |
| I1022 11:15:22.982206 8536 solver.cpp:236] Iteration 15700, loss = 1.69016 | |
| I1022 11:15:22.982230 8536 solver.cpp:252] Train net output #0: loss = 1.69016 (* 1 = 1.69016 loss) | |
| I1022 11:15:22.982236 8536 sgd_solver.cpp:106] Iteration 15700, lr = 0.001 | |
| I1022 11:15:33.495342 8536 solver.cpp:236] Iteration 15800, loss = 1.51237 | |
| I1022 11:15:33.495446 8536 solver.cpp:252] Train net output #0: loss = 1.51237 (* 1 = 1.51237 loss) | |
| I1022 11:15:33.495461 8536 sgd_solver.cpp:106] Iteration 15800, lr = 0.001 | |
| I1022 11:15:44.020524 8536 solver.cpp:236] Iteration 15900, loss = 1.46484 | |
| I1022 11:15:44.020562 8536 solver.cpp:252] Train net output #0: loss = 1.46484 (* 1 = 1.46484 loss) | |
| I1022 11:15:44.020567 8536 sgd_solver.cpp:106] Iteration 15900, lr = 0.001 | |
| I1022 11:15:54.478747 8536 solver.cpp:340] Iteration 16000, Testing net (#0) | |
| I1022 11:15:59.627290 8536 solver.cpp:408] Test net output #0: accuracy = 0.4152 | |
| I1022 11:15:59.627323 8536 solver.cpp:408] Test net output #1: loss = 1.61079 (* 1 = 1.61079 loss) | |
| I1022 11:15:59.693795 8536 solver.cpp:236] Iteration 16000, loss = 1.70227 | |
| I1022 11:15:59.693831 8536 solver.cpp:252] Train net output #0: loss = 1.70227 (* 1 = 1.70227 loss) | |
| I1022 11:15:59.693838 8536 sgd_solver.cpp:106] Iteration 16000, lr = 0.001 | |
| I1022 11:16:10.220052 8536 solver.cpp:236] Iteration 16100, loss = 1.48295 | |
| I1022 11:16:10.220165 8536 solver.cpp:252] Train net output #0: loss = 1.48295 (* 1 = 1.48295 loss) | |
| I1022 11:16:10.220170 8536 sgd_solver.cpp:106] Iteration 16100, lr = 0.001 | |
| I1022 11:16:20.602689 8536 solver.cpp:236] Iteration 16200, loss = 1.67739 | |
| I1022 11:16:20.602717 8536 solver.cpp:252] Train net output #0: loss = 1.67739 (* 1 = 1.67739 loss) | |
| I1022 11:16:20.602725 8536 sgd_solver.cpp:106] Iteration 16200, lr = 0.001 | |
| I1022 11:16:31.125738 8536 solver.cpp:236] Iteration 16300, loss = 1.52506 | |
| I1022 11:16:31.125776 8536 solver.cpp:252] Train net output #0: loss = 1.52506 (* 1 = 1.52506 loss) | |
| I1022 11:16:31.125780 8536 sgd_solver.cpp:106] Iteration 16300, lr = 0.001 | |
| I1022 11:16:41.658015 8536 solver.cpp:236] Iteration 16400, loss = 1.44382 | |
| I1022 11:16:41.658126 8536 solver.cpp:252] Train net output #0: loss = 1.44382 (* 1 = 1.44382 loss) | |
| I1022 11:16:41.658141 8536 sgd_solver.cpp:106] Iteration 16400, lr = 0.001 | |
| I1022 11:16:52.156821 8536 solver.cpp:236] Iteration 16500, loss = 1.70314 | |
| I1022 11:16:52.156857 8536 solver.cpp:252] Train net output #0: loss = 1.70314 (* 1 = 1.70314 loss) | |
| I1022 11:16:52.156862 8536 sgd_solver.cpp:106] Iteration 16500, lr = 0.001 | |
| I1022 11:17:04.051683 8536 solver.cpp:236] Iteration 16600, loss = 1.45752 | |
| I1022 11:17:04.051722 8536 solver.cpp:252] Train net output #0: loss = 1.45752 (* 1 = 1.45752 loss) | |
| I1022 11:17:04.051728 8536 sgd_solver.cpp:106] Iteration 16600, lr = 0.001 | |
| I1022 11:17:19.876646 8536 solver.cpp:236] Iteration 16700, loss = 1.67541 | |
| I1022 11:17:19.876736 8536 solver.cpp:252] Train net output #0: loss = 1.67541 (* 1 = 1.67541 loss) | |
| I1022 11:17:19.876749 8536 sgd_solver.cpp:106] Iteration 16700, lr = 0.001 | |
| I1022 11:17:35.541496 8536 solver.cpp:236] Iteration 16800, loss = 1.50925 | |
| I1022 11:17:35.541533 8536 solver.cpp:252] Train net output #0: loss = 1.50925 (* 1 = 1.50925 loss) | |
| I1022 11:17:35.541538 8536 sgd_solver.cpp:106] Iteration 16800, lr = 0.001 | |
| I1022 11:17:51.122148 8536 solver.cpp:236] Iteration 16900, loss = 1.45147 | |
| I1022 11:17:51.122239 8536 solver.cpp:252] Train net output #0: loss = 1.45147 (* 1 = 1.45147 loss) | |
| I1022 11:17:51.122246 8536 sgd_solver.cpp:106] Iteration 16900, lr = 0.001 | |
| I1022 11:18:06.762167 8536 solver.cpp:340] Iteration 17000, Testing net (#0) | |
| I1022 11:18:14.622251 8536 solver.cpp:408] Test net output #0: accuracy = 0.4195 | |
| I1022 11:18:14.622300 8536 solver.cpp:408] Test net output #1: loss = 1.5997 (* 1 = 1.5997 loss) | |
| I1022 11:18:14.727629 8536 solver.cpp:236] Iteration 17000, loss = 1.69983 | |
| I1022 11:18:14.727658 8536 solver.cpp:252] Train net output #0: loss = 1.69983 (* 1 = 1.69983 loss) | |
| I1022 11:18:14.727663 8536 sgd_solver.cpp:106] Iteration 17000, lr = 0.001 | |
| I1022 11:18:30.554291 8536 solver.cpp:236] Iteration 17100, loss = 1.48101 | |
| I1022 11:18:30.554368 8536 solver.cpp:252] Train net output #0: loss = 1.48101 (* 1 = 1.48101 loss) | |
| I1022 11:18:30.554379 8536 sgd_solver.cpp:106] Iteration 17100, lr = 0.001 | |
| I1022 11:18:46.366008 8536 solver.cpp:236] Iteration 17200, loss = 1.66837 | |
| I1022 11:18:46.366046 8536 solver.cpp:252] Train net output #0: loss = 1.66837 (* 1 = 1.66837 loss) | |
| I1022 11:18:46.366051 8536 sgd_solver.cpp:106] Iteration 17200, lr = 0.001 | |
| I1022 11:19:01.853924 8536 solver.cpp:236] Iteration 17300, loss = 1.5098 | |
| I1022 11:19:01.854136 8536 solver.cpp:252] Train net output #0: loss = 1.5098 (* 1 = 1.5098 loss) | |
| I1022 11:19:01.854148 8536 sgd_solver.cpp:106] Iteration 17300, lr = 0.001 | |
| I1022 11:19:17.490357 8536 solver.cpp:236] Iteration 17400, loss = 1.43694 | |
| I1022 11:19:17.490387 8536 solver.cpp:252] Train net output #0: loss = 1.43694 (* 1 = 1.43694 loss) | |
| I1022 11:19:17.490394 8536 sgd_solver.cpp:106] Iteration 17400, lr = 0.001 | |
| I1022 11:19:33.085923 8536 solver.cpp:236] Iteration 17500, loss = 1.69321 | |
| I1022 11:19:33.086015 8536 solver.cpp:252] Train net output #0: loss = 1.69321 (* 1 = 1.69321 loss) | |
| I1022 11:19:33.086021 8536 sgd_solver.cpp:106] Iteration 17500, lr = 0.001 | |
| I1022 11:19:48.616554 8536 solver.cpp:236] Iteration 17600, loss = 1.47778 | |
| I1022 11:19:48.616591 8536 solver.cpp:252] Train net output #0: loss = 1.47778 (* 1 = 1.47778 loss) | |
| I1022 11:19:48.616598 8536 sgd_solver.cpp:106] Iteration 17600, lr = 0.001 | |
| I1022 11:20:04.112418 8536 solver.cpp:236] Iteration 17700, loss = 1.66035 | |
| I1022 11:20:04.112510 8536 solver.cpp:252] Train net output #0: loss = 1.66035 (* 1 = 1.66035 loss) | |
| I1022 11:20:04.112516 8536 sgd_solver.cpp:106] Iteration 17700, lr = 0.001 | |
| I1022 11:20:19.785087 8536 solver.cpp:236] Iteration 17800, loss = 1.48203 | |
| I1022 11:20:19.785143 8536 solver.cpp:252] Train net output #0: loss = 1.48203 (* 1 = 1.48203 loss) | |
| I1022 11:20:19.785154 8536 sgd_solver.cpp:106] Iteration 17800, lr = 0.001 | |
| I1022 11:20:35.435070 8536 solver.cpp:236] Iteration 17900, loss = 1.43121 | |
| I1022 11:20:35.435169 8536 solver.cpp:252] Train net output #0: loss = 1.43121 (* 1 = 1.43121 loss) | |
| I1022 11:20:35.435175 8536 sgd_solver.cpp:106] Iteration 17900, lr = 0.001 | |
| I1022 11:20:51.206642 8536 solver.cpp:340] Iteration 18000, Testing net (#0) | |
| I1022 11:20:59.075762 8536 solver.cpp:408] Test net output #0: accuracy = 0.4216 | |
| I1022 11:20:59.075790 8536 solver.cpp:408] Test net output #1: loss = 1.59268 (* 1 = 1.59268 loss) | |
| I1022 11:20:59.178589 8536 solver.cpp:236] Iteration 18000, loss = 1.68358 | |
| I1022 11:20:59.178624 8536 solver.cpp:252] Train net output #0: loss = 1.68358 (* 1 = 1.68358 loss) | |
| I1022 11:20:59.178629 8536 sgd_solver.cpp:106] Iteration 18000, lr = 0.001 | |
| I1022 11:21:14.756212 8536 solver.cpp:236] Iteration 18100, loss = 1.47416 | |
| I1022 11:21:14.756289 8536 solver.cpp:252] Train net output #0: loss = 1.47416 (* 1 = 1.47416 loss) | |
| I1022 11:21:14.756294 8536 sgd_solver.cpp:106] Iteration 18100, lr = 0.001 | |
| I1022 11:21:30.346837 8536 solver.cpp:236] Iteration 18200, loss = 1.66951 | |
| I1022 11:21:30.346864 8536 solver.cpp:252] Train net output #0: loss = 1.66951 (* 1 = 1.66951 loss) | |
| I1022 11:21:30.346869 8536 sgd_solver.cpp:106] Iteration 18200, lr = 0.001 | |
| I1022 11:21:45.791546 8536 solver.cpp:236] Iteration 18300, loss = 1.47203 | |
| I1022 11:21:45.791656 8536 solver.cpp:252] Train net output #0: loss = 1.47203 (* 1 = 1.47203 loss) | |
| I1022 11:21:45.791661 8536 sgd_solver.cpp:106] Iteration 18300, lr = 0.001 | |
| I1022 11:22:01.326298 8536 solver.cpp:236] Iteration 18400, loss = 1.41797 | |
| I1022 11:22:01.326336 8536 solver.cpp:252] Train net output #0: loss = 1.41797 (* 1 = 1.41797 loss) | |
| I1022 11:22:01.326342 8536 sgd_solver.cpp:106] Iteration 18400, lr = 0.001 | |
| I1022 11:22:16.918109 8536 solver.cpp:236] Iteration 18500, loss = 1.68132 | |
| I1022 11:22:16.918206 8536 solver.cpp:252] Train net output #0: loss = 1.68132 (* 1 = 1.68132 loss) | |
| I1022 11:22:16.918218 8536 sgd_solver.cpp:106] Iteration 18500, lr = 0.001 | |
| I1022 11:22:32.644271 8536 solver.cpp:236] Iteration 18600, loss = 1.48547 | |
| I1022 11:22:32.644326 8536 solver.cpp:252] Train net output #0: loss = 1.48547 (* 1 = 1.48547 loss) | |
| I1022 11:22:32.644338 8536 sgd_solver.cpp:106] Iteration 18600, lr = 0.001 | |
| I1022 11:22:48.072569 8536 solver.cpp:236] Iteration 18700, loss = 1.65187 | |
| I1022 11:22:48.072660 8536 solver.cpp:252] Train net output #0: loss = 1.65187 (* 1 = 1.65187 loss) | |
| I1022 11:22:48.072665 8536 sgd_solver.cpp:106] Iteration 18700, lr = 0.001 | |
| I1022 11:23:03.644287 8536 solver.cpp:236] Iteration 18800, loss = 1.46749 | |
| I1022 11:23:03.644325 8536 solver.cpp:252] Train net output #0: loss = 1.46749 (* 1 = 1.46749 loss) | |
| I1022 11:23:03.644330 8536 sgd_solver.cpp:106] Iteration 18800, lr = 0.001 | |
| I1022 11:23:19.428458 8536 solver.cpp:236] Iteration 18900, loss = 1.40564 | |
| I1022 11:23:19.428541 8536 solver.cpp:252] Train net output #0: loss = 1.40564 (* 1 = 1.40564 loss) | |
| I1022 11:23:19.428552 8536 sgd_solver.cpp:106] Iteration 18900, lr = 0.001 | |
| I1022 11:23:35.089431 8536 solver.cpp:340] Iteration 19000, Testing net (#0) | |
| I1022 11:23:42.964900 8536 solver.cpp:408] Test net output #0: accuracy = 0.4217 | |
| I1022 11:23:42.964949 8536 solver.cpp:408] Test net output #1: loss = 1.58359 (* 1 = 1.58359 loss) | |
| I1022 11:23:43.056345 8536 solver.cpp:236] Iteration 19000, loss = 1.68481 | |
| I1022 11:23:43.056387 8536 solver.cpp:252] Train net output #0: loss = 1.68481 (* 1 = 1.68481 loss) | |
| I1022 11:23:43.056393 8536 sgd_solver.cpp:106] Iteration 19000, lr = 0.001 | |
| I1022 11:23:58.878793 8536 solver.cpp:236] Iteration 19100, loss = 1.48403 | |
| I1022 11:23:58.878886 8536 solver.cpp:252] Train net output #0: loss = 1.48403 (* 1 = 1.48403 loss) | |
| I1022 11:23:58.878895 8536 sgd_solver.cpp:106] Iteration 19100, lr = 0.001 | |
| I1022 11:24:14.850080 8536 solver.cpp:236] Iteration 19200, loss = 1.62435 | |
| I1022 11:24:14.850108 8536 solver.cpp:252] Train net output #0: loss = 1.62435 (* 1 = 1.62435 loss) | |
| I1022 11:24:14.850114 8536 sgd_solver.cpp:106] Iteration 19200, lr = 0.001 | |
| I1022 11:24:30.725255 8536 solver.cpp:236] Iteration 19300, loss = 1.46326 | |
| I1022 11:24:30.725335 8536 solver.cpp:252] Train net output #0: loss = 1.46326 (* 1 = 1.46326 loss) | |
| I1022 11:24:30.725342 8536 sgd_solver.cpp:106] Iteration 19300, lr = 0.001 | |
| I1022 11:24:46.118641 8536 solver.cpp:236] Iteration 19400, loss = 1.39482 | |
| I1022 11:24:46.118669 8536 solver.cpp:252] Train net output #0: loss = 1.39482 (* 1 = 1.39482 loss) | |
| I1022 11:24:46.118674 8536 sgd_solver.cpp:106] Iteration 19400, lr = 0.001 | |
| I1022 11:24:58.404811 8536 solver.cpp:236] Iteration 19500, loss = 1.67783 | |
| I1022 11:24:58.404847 8536 solver.cpp:252] Train net output #0: loss = 1.67783 (* 1 = 1.67783 loss) | |
| I1022 11:24:58.404853 8536 sgd_solver.cpp:106] Iteration 19500, lr = 0.001 | |
| I1022 11:25:09.095026 8536 solver.cpp:236] Iteration 19600, loss = 1.4618 | |
| I1022 11:25:09.095100 8536 solver.cpp:252] Train net output #0: loss = 1.4618 (* 1 = 1.4618 loss) | |
| I1022 11:25:09.095106 8536 sgd_solver.cpp:106] Iteration 19600, lr = 0.001 | |
| I1022 11:25:19.825944 8536 solver.cpp:236] Iteration 19700, loss = 1.65221 | |
| I1022 11:25:19.825980 8536 solver.cpp:252] Train net output #0: loss = 1.65221 (* 1 = 1.65221 loss) | |
| I1022 11:25:19.825986 8536 sgd_solver.cpp:106] Iteration 19700, lr = 0.001 | |
| I1022 11:25:30.354027 8536 solver.cpp:236] Iteration 19800, loss = 1.47043 | |
| I1022 11:25:30.354063 8536 solver.cpp:252] Train net output #0: loss = 1.47043 (* 1 = 1.47043 loss) | |
| I1022 11:25:30.354068 8536 sgd_solver.cpp:106] Iteration 19800, lr = 0.001 | |
| I1022 11:25:40.996464 8536 solver.cpp:236] Iteration 19900, loss = 1.4104 | |
| I1022 11:25:40.996557 8536 solver.cpp:252] Train net output #0: loss = 1.4104 (* 1 = 1.4104 loss) | |
| I1022 11:25:40.996569 8536 sgd_solver.cpp:106] Iteration 19900, lr = 0.001 | |
| I1022 11:25:51.598611 8536 solver.cpp:340] Iteration 20000, Testing net (#0) | |
| I1022 11:25:56.685449 8536 solver.cpp:408] Test net output #0: accuracy = 0.4218 | |
| I1022 11:25:56.685497 8536 solver.cpp:408] Test net output #1: loss = 1.57866 (* 1 = 1.57866 loss) | |
| I1022 11:25:56.754575 8536 solver.cpp:236] Iteration 20000, loss = 1.68525 | |
| I1022 11:25:56.754611 8536 solver.cpp:252] Train net output #0: loss = 1.68525 (* 1 = 1.68525 loss) | |
| I1022 11:25:56.754616 8536 sgd_solver.cpp:106] Iteration 20000, lr = 0.001 | |
| I1022 11:26:07.276060 8536 solver.cpp:236] Iteration 20100, loss = 1.45834 | |
| I1022 11:26:07.276096 8536 solver.cpp:252] Train net output #0: loss = 1.45834 (* 1 = 1.45834 loss) | |
| I1022 11:26:07.276101 8536 sgd_solver.cpp:106] Iteration 20100, lr = 0.001 | |
| I1022 11:26:17.806164 8536 solver.cpp:236] Iteration 20200, loss = 1.63545 | |
| I1022 11:26:17.806272 8536 solver.cpp:252] Train net output #0: loss = 1.63545 (* 1 = 1.63545 loss) | |
| I1022 11:26:17.806288 8536 sgd_solver.cpp:106] Iteration 20200, lr = 0.001 | |
| I1022 11:26:28.304659 8536 solver.cpp:236] Iteration 20300, loss = 1.45305 | |
| I1022 11:26:28.304697 8536 solver.cpp:252] Train net output #0: loss = 1.45305 (* 1 = 1.45305 loss) | |
| I1022 11:26:28.304702 8536 sgd_solver.cpp:106] Iteration 20300, lr = 0.001 | |
| I1022 11:26:38.699551 8536 solver.cpp:236] Iteration 20400, loss = 1.39302 | |
| I1022 11:26:38.699579 8536 solver.cpp:252] Train net output #0: loss = 1.39302 (* 1 = 1.39302 loss) | |
| I1022 11:26:38.699585 8536 sgd_solver.cpp:106] Iteration 20400, lr = 0.001 | |
| I1022 11:26:49.082860 8536 solver.cpp:236] Iteration 20500, loss = 1.67335 | |
| I1022 11:26:49.082969 8536 solver.cpp:252] Train net output #0: loss = 1.67335 (* 1 = 1.67335 loss) | |
| I1022 11:26:49.082974 8536 sgd_solver.cpp:106] Iteration 20500, lr = 0.001 | |
| I1022 11:26:59.632510 8536 solver.cpp:236] Iteration 20600, loss = 1.45115 | |
| I1022 11:26:59.632539 8536 solver.cpp:252] Train net output #0: loss = 1.45115 (* 1 = 1.45115 loss) | |
| I1022 11:26:59.632544 8536 sgd_solver.cpp:106] Iteration 20600, lr = 0.001 | |
| I1022 11:27:10.128350 8536 solver.cpp:236] Iteration 20700, loss = 1.61431 | |
| I1022 11:27:10.128391 8536 solver.cpp:252] Train net output #0: loss = 1.61431 (* 1 = 1.61431 loss) | |
| I1022 11:27:10.128396 8536 sgd_solver.cpp:106] Iteration 20700, lr = 0.001 | |
| I1022 11:27:20.647275 8536 solver.cpp:236] Iteration 20800, loss = 1.45903 | |
| I1022 11:27:20.647382 8536 solver.cpp:252] Train net output #0: loss = 1.45903 (* 1 = 1.45903 loss) | |
| I1022 11:27:20.647388 8536 sgd_solver.cpp:106] Iteration 20800, lr = 0.001 | |
| I1022 11:27:31.165225 8536 solver.cpp:236] Iteration 20900, loss = 1.39597 | |
| I1022 11:27:31.165261 8536 solver.cpp:252] Train net output #0: loss = 1.39597 (* 1 = 1.39597 loss) | |
| I1022 11:27:31.165267 8536 sgd_solver.cpp:106] Iteration 20900, lr = 0.001 | |
| I1022 11:27:41.585317 8536 solver.cpp:340] Iteration 21000, Testing net (#0) | |
| I1022 11:27:46.819574 8536 solver.cpp:408] Test net output #0: accuracy = 0.4259 | |
| I1022 11:27:46.819612 8536 solver.cpp:408] Test net output #1: loss = 1.56778 (* 1 = 1.56778 loss) | |
| I1022 11:27:46.890826 8536 solver.cpp:236] Iteration 21000, loss = 1.6642 | |
| I1022 11:27:46.890862 8536 solver.cpp:252] Train net output #0: loss = 1.6642 (* 1 = 1.6642 loss) | |
| I1022 11:27:46.890867 8536 sgd_solver.cpp:106] Iteration 21000, lr = 0.001 | |
| I1022 11:27:57.555428 8536 solver.cpp:236] Iteration 21100, loss = 1.44994 | |
| I1022 11:27:57.555532 8536 solver.cpp:252] Train net output #0: loss = 1.44994 (* 1 = 1.44994 loss) | |
| I1022 11:27:57.555539 8536 sgd_solver.cpp:106] Iteration 21100, lr = 0.001 | |
| I1022 11:28:08.379308 8536 solver.cpp:236] Iteration 21200, loss = 1.60958 | |
| I1022 11:28:08.379346 8536 solver.cpp:252] Train net output #0: loss = 1.60958 (* 1 = 1.60958 loss) | |
| I1022 11:28:08.379353 8536 sgd_solver.cpp:106] Iteration 21200, lr = 0.001 | |
| I1022 11:28:19.174197 8536 solver.cpp:236] Iteration 21300, loss = 1.45987 | |
| I1022 11:28:19.174235 8536 solver.cpp:252] Train net output #0: loss = 1.45987 (* 1 = 1.45987 loss) | |
| I1022 11:28:19.174240 8536 sgd_solver.cpp:106] Iteration 21300, lr = 0.001 | |
| I1022 11:28:29.767763 8536 solver.cpp:236] Iteration 21400, loss = 1.39451 | |
| I1022 11:28:29.767822 8536 solver.cpp:252] Train net output #0: loss = 1.39451 (* 1 = 1.39451 loss) | |
| I1022 11:28:29.767828 8536 sgd_solver.cpp:106] Iteration 21400, lr = 0.001 | |
| I1022 11:28:40.006217 8536 solver.cpp:236] Iteration 21500, loss = 1.66202 | |
| I1022 11:28:40.006244 8536 solver.cpp:252] Train net output #0: loss = 1.66202 (* 1 = 1.66202 loss) | |
| I1022 11:28:40.006249 8536 sgd_solver.cpp:106] Iteration 21500, lr = 0.001 | |
| I1022 11:28:50.555627 8536 solver.cpp:236] Iteration 21600, loss = 1.434 | |
| I1022 11:28:50.555654 8536 solver.cpp:252] Train net output #0: loss = 1.434 (* 1 = 1.434 loss) | |
| I1022 11:28:50.555658 8536 sgd_solver.cpp:106] Iteration 21600, lr = 0.001 | |
| I1022 11:29:01.152758 8536 solver.cpp:236] Iteration 21700, loss = 1.58727 | |
| I1022 11:29:01.152865 8536 solver.cpp:252] Train net output #0: loss = 1.58727 (* 1 = 1.58727 loss) | |
| I1022 11:29:01.152880 8536 sgd_solver.cpp:106] Iteration 21700, lr = 0.001 | |
| I1022 11:29:11.751288 8536 solver.cpp:236] Iteration 21800, loss = 1.43126 | |
| I1022 11:29:11.751315 8536 solver.cpp:252] Train net output #0: loss = 1.43126 (* 1 = 1.43126 loss) | |
| I1022 11:29:11.751322 8536 sgd_solver.cpp:106] Iteration 21800, lr = 0.001 | |
| I1022 11:29:22.510365 8536 solver.cpp:236] Iteration 21900, loss = 1.37895 | |
| I1022 11:29:22.510401 8536 solver.cpp:252] Train net output #0: loss = 1.37895 (* 1 = 1.37895 loss) | |
| I1022 11:29:22.510406 8536 sgd_solver.cpp:106] Iteration 21900, lr = 0.001 | |
| I1022 11:29:32.994801 8536 solver.cpp:340] Iteration 22000, Testing net (#0) | |
| I1022 11:29:38.177765 8536 solver.cpp:408] Test net output #0: accuracy = 0.4308 | |
| I1022 11:29:38.177800 8536 solver.cpp:408] Test net output #1: loss = 1.55757 (* 1 = 1.55757 loss) | |
| I1022 11:29:38.244222 8536 solver.cpp:236] Iteration 22000, loss = 1.66062 | |
| I1022 11:29:38.244257 8536 solver.cpp:252] Train net output #0: loss = 1.66062 (* 1 = 1.66062 loss) | |
| I1022 11:29:38.244262 8536 sgd_solver.cpp:106] Iteration 22000, lr = 0.001 | |
| I1022 11:29:49.013748 8536 solver.cpp:236] Iteration 22100, loss = 1.43977 | |
| I1022 11:29:49.013784 8536 solver.cpp:252] Train net output #0: loss = 1.43977 (* 1 = 1.43977 loss) | |
| I1022 11:29:49.013792 8536 sgd_solver.cpp:106] Iteration 22100, lr = 0.001 | |
| I1022 11:29:59.681447 8536 solver.cpp:236] Iteration 22200, loss = 1.6062 | |
| I1022 11:29:59.681485 8536 solver.cpp:252] Train net output #0: loss = 1.6062 (* 1 = 1.6062 loss) | |
| I1022 11:29:59.681490 8536 sgd_solver.cpp:106] Iteration 22200, lr = 0.001 | |
| I1022 11:30:10.171463 8536 solver.cpp:236] Iteration 22300, loss = 1.43489 | |
| I1022 11:30:10.171563 8536 solver.cpp:252] Train net output #0: loss = 1.43489 (* 1 = 1.43489 loss) | |
| I1022 11:30:10.171569 8536 sgd_solver.cpp:106] Iteration 22300, lr = 0.001 | |
| I1022 11:30:20.638829 8536 solver.cpp:236] Iteration 22400, loss = 1.37533 | |
| I1022 11:30:20.638866 8536 solver.cpp:252] Train net output #0: loss = 1.37533 (* 1 = 1.37533 loss) | |
| I1022 11:30:20.638871 8536 sgd_solver.cpp:106] Iteration 22400, lr = 0.001 | |
| I1022 11:30:31.072971 8536 solver.cpp:236] Iteration 22500, loss = 1.6607 | |
| I1022 11:30:31.073007 8536 solver.cpp:252] Train net output #0: loss = 1.6607 (* 1 = 1.6607 loss) | |
| I1022 11:30:31.073014 8536 sgd_solver.cpp:106] Iteration 22500, lr = 0.001 | |
| I1022 11:30:41.480780 8536 solver.cpp:236] Iteration 22600, loss = 1.43862 | |
| I1022 11:30:41.480878 8536 solver.cpp:252] Train net output #0: loss = 1.43862 (* 1 = 1.43862 loss) | |
| I1022 11:30:41.480883 8536 sgd_solver.cpp:106] Iteration 22600, lr = 0.001 | |
| I1022 11:30:51.959641 8536 solver.cpp:236] Iteration 22700, loss = 1.57797 | |
| I1022 11:30:51.959671 8536 solver.cpp:252] Train net output #0: loss = 1.57797 (* 1 = 1.57797 loss) | |
| I1022 11:30:51.959676 8536 sgd_solver.cpp:106] Iteration 22700, lr = 0.001 | |
| I1022 11:31:02.416637 8536 solver.cpp:236] Iteration 22800, loss = 1.43455 | |
| I1022 11:31:02.416674 8536 solver.cpp:252] Train net output #0: loss = 1.43455 (* 1 = 1.43455 loss) | |
| I1022 11:31:02.416681 8536 sgd_solver.cpp:106] Iteration 22800, lr = 0.001 | |
| I1022 11:31:12.869185 8536 solver.cpp:236] Iteration 22900, loss = 1.36973 | |
| I1022 11:31:12.869288 8536 solver.cpp:252] Train net output #0: loss = 1.36973 (* 1 = 1.36973 loss) | |
| I1022 11:31:12.869294 8536 sgd_solver.cpp:106] Iteration 22900, lr = 0.001 | |
| I1022 11:31:23.270099 8536 solver.cpp:340] Iteration 23000, Testing net (#0) | |
| I1022 11:31:28.469693 8536 solver.cpp:408] Test net output #0: accuracy = 0.4314 | |
| I1022 11:31:28.469728 8536 solver.cpp:408] Test net output #1: loss = 1.55868 (* 1 = 1.55868 loss) | |
| I1022 11:31:28.536613 8536 solver.cpp:236] Iteration 23000, loss = 1.64803 | |
| I1022 11:31:28.536639 8536 solver.cpp:252] Train net output #0: loss = 1.64803 (* 1 = 1.64803 loss) | |
| I1022 11:31:28.536644 8536 sgd_solver.cpp:106] Iteration 23000, lr = 0.001 | |
| I1022 11:31:39.238369 8536 solver.cpp:236] Iteration 23100, loss = 1.42929 | |
| I1022 11:31:39.238409 8536 solver.cpp:252] Train net output #0: loss = 1.42929 (* 1 = 1.42929 loss) | |
| I1022 11:31:39.238414 8536 sgd_solver.cpp:106] Iteration 23100, lr = 0.001 | |
| I1022 11:31:49.765833 8536 solver.cpp:236] Iteration 23200, loss = 1.58734 | |
| I1022 11:31:49.765919 8536 solver.cpp:252] Train net output #0: loss = 1.58734 (* 1 = 1.58734 loss) | |
| I1022 11:31:49.765924 8536 sgd_solver.cpp:106] Iteration 23200, lr = 0.001 | |
| I1022 11:32:00.358640 8536 solver.cpp:236] Iteration 23300, loss = 1.43082 | |
| I1022 11:32:00.358676 8536 solver.cpp:252] Train net output #0: loss = 1.43082 (* 1 = 1.43082 loss) | |
| I1022 11:32:00.358682 8536 sgd_solver.cpp:106] Iteration 23300, lr = 0.001 | |
| I1022 11:32:10.823578 8536 solver.cpp:236] Iteration 23400, loss = 1.37993 | |
| I1022 11:32:10.823603 8536 solver.cpp:252] Train net output #0: loss = 1.37993 (* 1 = 1.37993 loss) | |
| I1022 11:32:10.823608 8536 sgd_solver.cpp:106] Iteration 23400, lr = 0.001 | |
| I1022 11:32:21.264716 8536 solver.cpp:236] Iteration 23500, loss = 1.63362 | |
| I1022 11:32:21.264787 8536 solver.cpp:252] Train net output #0: loss = 1.63362 (* 1 = 1.63362 loss) | |
| I1022 11:32:21.264793 8536 sgd_solver.cpp:106] Iteration 23500, lr = 0.001 | |
| I1022 11:32:31.528718 8536 solver.cpp:236] Iteration 23600, loss = 1.43366 | |
| I1022 11:32:31.528754 8536 solver.cpp:252] Train net output #0: loss = 1.43366 (* 1 = 1.43366 loss) | |
| I1022 11:32:31.528760 8536 sgd_solver.cpp:106] Iteration 23600, lr = 0.001 | |
| I1022 11:32:41.933521 8536 solver.cpp:236] Iteration 23700, loss = 1.58674 | |
| I1022 11:32:41.933560 8536 solver.cpp:252] Train net output #0: loss = 1.58674 (* 1 = 1.58674 loss) | |
| I1022 11:32:41.933567 8536 sgd_solver.cpp:106] Iteration 23700, lr = 0.001 | |
| I1022 11:32:52.656209 8536 solver.cpp:236] Iteration 23800, loss = 1.41597 | |
| I1022 11:32:52.656311 8536 solver.cpp:252] Train net output #0: loss = 1.41597 (* 1 = 1.41597 loss) | |
| I1022 11:32:52.656316 8536 sgd_solver.cpp:106] Iteration 23800, lr = 0.001 | |
| I1022 11:33:03.166453 8536 solver.cpp:236] Iteration 23900, loss = 1.36292 | |
| I1022 11:33:03.166481 8536 solver.cpp:252] Train net output #0: loss = 1.36292 (* 1 = 1.36292 loss) | |
| I1022 11:33:03.166486 8536 sgd_solver.cpp:106] Iteration 23900, lr = 0.001 | |
| I1022 11:33:13.531685 8536 solver.cpp:340] Iteration 24000, Testing net (#0) | |
| I1022 11:33:18.598912 8536 solver.cpp:408] Test net output #0: accuracy = 0.431 | |
| I1022 11:33:18.598944 8536 solver.cpp:408] Test net output #1: loss = 1.54621 (* 1 = 1.54621 loss) | |
| I1022 11:33:18.665500 8536 solver.cpp:236] Iteration 24000, loss = 1.63982 | |
| I1022 11:33:18.665534 8536 solver.cpp:252] Train net output #0: loss = 1.63982 (* 1 = 1.63982 loss) | |
| I1022 11:33:18.665539 8536 sgd_solver.cpp:106] Iteration 24000, lr = 0.001 | |
| I1022 11:33:29.116564 8536 solver.cpp:236] Iteration 24100, loss = 1.41556 | |
| I1022 11:33:29.116669 8536 solver.cpp:252] Train net output #0: loss = 1.41556 (* 1 = 1.41556 loss) | |
| I1022 11:33:29.116684 8536 sgd_solver.cpp:106] Iteration 24100, lr = 0.001 | |
| I1022 11:33:39.566375 8536 solver.cpp:236] Iteration 24200, loss = 1.56235 | |
| I1022 11:33:39.566411 8536 solver.cpp:252] Train net output #0: loss = 1.56235 (* 1 = 1.56235 loss) | |
| I1022 11:33:39.566416 8536 sgd_solver.cpp:106] Iteration 24200, lr = 0.001 | |
| I1022 11:33:50.017849 8536 solver.cpp:236] Iteration 24300, loss = 1.41285 | |
| I1022 11:33:50.017886 8536 solver.cpp:252] Train net output #0: loss = 1.41285 (* 1 = 1.41285 loss) | |
| I1022 11:33:50.017891 8536 sgd_solver.cpp:106] Iteration 24300, lr = 0.001 | |
| I1022 11:34:00.466984 8536 solver.cpp:236] Iteration 24400, loss = 1.37083 | |
| I1022 11:34:00.467092 8536 solver.cpp:252] Train net output #0: loss = 1.37083 (* 1 = 1.37083 loss) | |
| I1022 11:34:00.467108 8536 sgd_solver.cpp:106] Iteration 24400, lr = 0.001 | |
| I1022 11:34:10.928581 8536 solver.cpp:236] Iteration 24500, loss = 1.63523 | |
| I1022 11:34:10.928616 8536 solver.cpp:252] Train net output #0: loss = 1.63523 (* 1 = 1.63523 loss) | |
| I1022 11:34:10.928621 8536 sgd_solver.cpp:106] Iteration 24500, lr = 0.001 | |
| I1022 11:34:21.387420 8536 solver.cpp:236] Iteration 24600, loss = 1.42009 | |
| I1022 11:34:21.387447 8536 solver.cpp:252] Train net output #0: loss = 1.42009 (* 1 = 1.42009 loss) | |
| I1022 11:34:21.387452 8536 sgd_solver.cpp:106] Iteration 24600, lr = 0.001 | |
| I1022 11:34:31.623276 8536 solver.cpp:236] Iteration 24700, loss = 1.56551 | |
| I1022 11:34:31.623383 8536 solver.cpp:252] Train net output #0: loss = 1.56551 (* 1 = 1.56551 loss) | |
| I1022 11:34:31.623389 8536 sgd_solver.cpp:106] Iteration 24700, lr = 0.001 | |
| I1022 11:34:42.077611 8536 solver.cpp:236] Iteration 24800, loss = 1.43489 | |
| I1022 11:34:42.077648 8536 solver.cpp:252] Train net output #0: loss = 1.43489 (* 1 = 1.43489 loss) | |
| I1022 11:34:42.077653 8536 sgd_solver.cpp:106] Iteration 24800, lr = 0.001 | |
| I1022 11:34:52.528301 8536 solver.cpp:236] Iteration 24900, loss = 1.36453 | |
| I1022 11:34:52.528327 8536 solver.cpp:252] Train net output #0: loss = 1.36453 (* 1 = 1.36453 loss) | |
| I1022 11:34:52.528333 8536 sgd_solver.cpp:106] Iteration 24900, lr = 0.001 | |
| I1022 11:35:02.889945 8536 solver.cpp:340] Iteration 25000, Testing net (#0) | |
| I1022 11:35:07.998065 8536 solver.cpp:408] Test net output #0: accuracy = 0.4329 | |
| I1022 11:35:07.998100 8536 solver.cpp:408] Test net output #1: loss = 1.54336 (* 1 = 1.54336 loss) | |
| I1022 11:35:08.064112 8536 solver.cpp:236] Iteration 25000, loss = 1.62938 | |
| I1022 11:35:08.064149 8536 solver.cpp:252] Train net output #0: loss = 1.62938 (* 1 = 1.62938 loss) | |
| I1022 11:35:08.064154 8536 sgd_solver.cpp:106] Iteration 25000, lr = 0.001 | |
| I1022 11:35:18.592597 8536 solver.cpp:236] Iteration 25100, loss = 1.41808 | |
| I1022 11:35:18.592625 8536 solver.cpp:252] Train net output #0: loss = 1.41808 (* 1 = 1.41808 loss) | |
| I1022 11:35:18.592630 8536 sgd_solver.cpp:106] Iteration 25100, lr = 0.001 | |
| I1022 11:35:29.145010 8536 solver.cpp:236] Iteration 25200, loss = 1.55364 | |
| I1022 11:35:29.145056 8536 solver.cpp:252] Train net output #0: loss = 1.55364 (* 1 = 1.55364 loss) | |
| I1022 11:35:29.145066 8536 sgd_solver.cpp:106] Iteration 25200, lr = 0.001 | |
| I1022 11:35:39.651571 8536 solver.cpp:236] Iteration 25300, loss = 1.43037 | |
| I1022 11:35:39.651674 8536 solver.cpp:252] Train net output #0: loss = 1.43037 (* 1 = 1.43037 loss) | |
| I1022 11:35:39.651680 8536 sgd_solver.cpp:106] Iteration 25300, lr = 0.001 | |
| I1022 11:35:50.193931 8536 solver.cpp:236] Iteration 25400, loss = 1.36059 | |
| I1022 11:35:50.193969 8536 solver.cpp:252] Train net output #0: loss = 1.36059 (* 1 = 1.36059 loss) | |
| I1022 11:35:50.193974 8536 sgd_solver.cpp:106] Iteration 25400, lr = 0.001 | |
| I1022 11:36:00.654006 8536 solver.cpp:236] Iteration 25500, loss = 1.64038 | |
| I1022 11:36:00.654042 8536 solver.cpp:252] Train net output #0: loss = 1.64038 (* 1 = 1.64038 loss) | |
| I1022 11:36:00.654047 8536 sgd_solver.cpp:106] Iteration 25500, lr = 0.001 | |
| I1022 11:36:11.120630 8536 solver.cpp:236] Iteration 25600, loss = 1.43291 | |
| I1022 11:36:11.120718 8536 solver.cpp:252] Train net output #0: loss = 1.43291 (* 1 = 1.43291 loss) | |
| I1022 11:36:11.120723 8536 sgd_solver.cpp:106] Iteration 25600, lr = 0.001 | |
| I1022 11:36:21.485081 8536 solver.cpp:236] Iteration 25700, loss = 1.55469 | |
| I1022 11:36:21.485117 8536 solver.cpp:252] Train net output #0: loss = 1.55469 (* 1 = 1.55469 loss) | |
| I1022 11:36:21.485124 8536 sgd_solver.cpp:106] Iteration 25700, lr = 0.001 | |
| I1022 11:36:31.832706 8536 solver.cpp:236] Iteration 25800, loss = 1.40826 | |
| I1022 11:36:31.832743 8536 solver.cpp:252] Train net output #0: loss = 1.40826 (* 1 = 1.40826 loss) | |
| I1022 11:36:31.832748 8536 sgd_solver.cpp:106] Iteration 25800, lr = 0.001 | |
| I1022 11:36:42.356111 8536 solver.cpp:236] Iteration 25900, loss = 1.3593 | |
| I1022 11:36:42.356178 8536 solver.cpp:252] Train net output #0: loss = 1.3593 (* 1 = 1.3593 loss) | |
| I1022 11:36:42.356183 8536 sgd_solver.cpp:106] Iteration 25900, lr = 0.001 | |
| I1022 11:36:52.779764 8536 solver.cpp:340] Iteration 26000, Testing net (#0) | |
| I1022 11:36:57.839565 8536 solver.cpp:408] Test net output #0: accuracy = 0.435 | |
| I1022 11:36:57.839587 8536 solver.cpp:408] Test net output #1: loss = 1.53226 (* 1 = 1.53226 loss) | |
| I1022 11:36:57.899875 8536 solver.cpp:236] Iteration 26000, loss = 1.61618 | |
| I1022 11:36:57.899909 8536 solver.cpp:252] Train net output #0: loss = 1.61618 (* 1 = 1.61618 loss) | |
| I1022 11:36:57.899915 8536 sgd_solver.cpp:106] Iteration 26000, lr = 0.001 | |
| I1022 11:37:08.415971 8536 solver.cpp:236] Iteration 26100, loss = 1.41137 | |
| I1022 11:37:08.416005 8536 solver.cpp:252] Train net output #0: loss = 1.41137 (* 1 = 1.41137 loss) | |
| I1022 11:37:08.416010 8536 sgd_solver.cpp:106] Iteration 26100, lr = 0.001 | |
| I1022 11:37:18.888247 8536 solver.cpp:236] Iteration 26200, loss = 1.5667 | |
| I1022 11:37:18.888304 8536 solver.cpp:252] Train net output #0: loss = 1.5667 (* 1 = 1.5667 loss) | |
| I1022 11:37:18.888310 8536 sgd_solver.cpp:106] Iteration 26200, lr = 0.001 | |
| I1022 11:37:29.355510 8536 solver.cpp:236] Iteration 26300, loss = 1.42579 | |
| I1022 11:37:29.355546 8536 solver.cpp:252] Train net output #0: loss = 1.42579 (* 1 = 1.42579 loss) | |
| I1022 11:37:29.355551 8536 sgd_solver.cpp:106] Iteration 26300, lr = 0.001 | |
| I1022 11:37:39.817487 8536 solver.cpp:236] Iteration 26400, loss = 1.36754 | |
| I1022 11:37:39.817513 8536 solver.cpp:252] Train net output #0: loss = 1.36754 (* 1 = 1.36754 loss) | |
| I1022 11:37:39.817519 8536 sgd_solver.cpp:106] Iteration 26400, lr = 0.001 | |
| I1022 11:37:50.302403 8536 solver.cpp:236] Iteration 26500, loss = 1.60775 | |
| I1022 11:37:50.302498 8536 solver.cpp:252] Train net output #0: loss = 1.60775 (* 1 = 1.60775 loss) | |
| I1022 11:37:50.302505 8536 sgd_solver.cpp:106] Iteration 26500, lr = 0.001 | |
| I1022 11:38:00.894369 8536 solver.cpp:236] Iteration 26600, loss = 1.40808 | |
| I1022 11:38:00.894407 8536 solver.cpp:252] Train net output #0: loss = 1.40808 (* 1 = 1.40808 loss) | |
| I1022 11:38:00.894412 8536 sgd_solver.cpp:106] Iteration 26600, lr = 0.001 | |
| I1022 11:38:11.447166 8536 solver.cpp:236] Iteration 26700, loss = 1.54969 | |
| I1022 11:38:11.447195 8536 solver.cpp:252] Train net output #0: loss = 1.54969 (* 1 = 1.54969 loss) | |
| I1022 11:38:11.447201 8536 sgd_solver.cpp:106] Iteration 26700, lr = 0.001 | |
| I1022 11:38:21.740226 8536 solver.cpp:236] Iteration 26800, loss = 1.41622 | |
| I1022 11:38:21.740335 8536 solver.cpp:252] Train net output #0: loss = 1.41622 (* 1 = 1.41622 loss) | |
| I1022 11:38:21.740340 8536 sgd_solver.cpp:106] Iteration 26800, lr = 0.001 | |
| I1022 11:38:32.202797 8536 solver.cpp:236] Iteration 26900, loss = 1.35619 | |
| I1022 11:38:32.202834 8536 solver.cpp:252] Train net output #0: loss = 1.35619 (* 1 = 1.35619 loss) | |
| I1022 11:38:32.202839 8536 sgd_solver.cpp:106] Iteration 26900, lr = 0.001 | |
| I1022 11:38:42.690114 8536 solver.cpp:340] Iteration 27000, Testing net (#0) | |
| I1022 11:38:47.855507 8536 solver.cpp:408] Test net output #0: accuracy = 0.4396 | |
| I1022 11:38:47.855541 8536 solver.cpp:408] Test net output #1: loss = 1.52835 (* 1 = 1.52835 loss) | |
| I1022 11:38:47.921905 8536 solver.cpp:236] Iteration 27000, loss = 1.5928 | |
| I1022 11:38:47.921938 8536 solver.cpp:252] Train net output #0: loss = 1.5928 (* 1 = 1.5928 loss) | |
| I1022 11:38:47.921943 8536 sgd_solver.cpp:106] Iteration 27000, lr = 0.001 | |
| I1022 11:38:58.413758 8536 solver.cpp:236] Iteration 27100, loss = 1.41937 | |
| I1022 11:38:58.413866 8536 solver.cpp:252] Train net output #0: loss = 1.41937 (* 1 = 1.41937 loss) | |
| I1022 11:38:58.413871 8536 sgd_solver.cpp:106] Iteration 27100, lr = 0.001 | |
| I1022 11:39:09.149554 8536 solver.cpp:236] Iteration 27200, loss = 1.54478 | |
| I1022 11:39:09.149590 8536 solver.cpp:252] Train net output #0: loss = 1.54478 (* 1 = 1.54478 loss) | |
| I1022 11:39:09.149596 8536 sgd_solver.cpp:106] Iteration 27200, lr = 0.001 | |
| I1022 11:39:19.607967 8536 solver.cpp:236] Iteration 27300, loss = 1.39288 | |
| I1022 11:39:19.608003 8536 solver.cpp:252] Train net output #0: loss = 1.39288 (* 1 = 1.39288 loss) | |
| I1022 11:39:19.608010 8536 sgd_solver.cpp:106] Iteration 27300, lr = 0.001 | |
| I1022 11:39:30.068370 8536 solver.cpp:236] Iteration 27400, loss = 1.34193 | |
| I1022 11:39:30.068480 8536 solver.cpp:252] Train net output #0: loss = 1.34193 (* 1 = 1.34193 loss) | |
| I1022 11:39:30.068485 8536 sgd_solver.cpp:106] Iteration 27400, lr = 0.001 | |
| I1022 11:39:40.523309 8536 solver.cpp:236] Iteration 27500, loss = 1.59458 | |
| I1022 11:39:40.523345 8536 solver.cpp:252] Train net output #0: loss = 1.59458 (* 1 = 1.59458 loss) | |
| I1022 11:39:40.523351 8536 sgd_solver.cpp:106] Iteration 27500, lr = 0.001 | |
| I1022 11:39:51.256079 8536 solver.cpp:236] Iteration 27600, loss = 1.40049 | |
| I1022 11:39:51.256115 8536 solver.cpp:252] Train net output #0: loss = 1.40049 (* 1 = 1.40049 loss) | |
| I1022 11:39:51.256121 8536 sgd_solver.cpp:106] Iteration 27600, lr = 0.001 | |
| I1022 11:40:01.740494 8536 solver.cpp:236] Iteration 27700, loss = 1.52313 | |
| I1022 11:40:01.740583 8536 solver.cpp:252] Train net output #0: loss = 1.52313 (* 1 = 1.52313 loss) | |
| I1022 11:40:01.740589 8536 sgd_solver.cpp:106] Iteration 27700, lr = 0.001 | |
| I1022 11:40:12.170728 8536 solver.cpp:236] Iteration 27800, loss = 1.42821 | |
| I1022 11:40:12.170765 8536 solver.cpp:252] Train net output #0: loss = 1.42821 (* 1 = 1.42821 loss) | |
| I1022 11:40:12.170771 8536 sgd_solver.cpp:106] Iteration 27800, lr = 0.001 | |
| I1022 11:40:22.424669 8536 solver.cpp:236] Iteration 27900, loss = 1.33793 | |
| I1022 11:40:22.424705 8536 solver.cpp:252] Train net output #0: loss = 1.33793 (* 1 = 1.33793 loss) | |
| I1022 11:40:22.424710 8536 sgd_solver.cpp:106] Iteration 27900, lr = 0.001 | |
| I1022 11:40:32.779554 8536 solver.cpp:340] Iteration 28000, Testing net (#0) | |
| I1022 11:40:37.841409 8536 solver.cpp:408] Test net output #0: accuracy = 0.4415 | |
| I1022 11:40:37.841436 8536 solver.cpp:408] Test net output #1: loss = 1.52014 (* 1 = 1.52014 loss) | |
| I1022 11:40:37.907819 8536 solver.cpp:236] Iteration 28000, loss = 1.60771 | |
| I1022 11:40:37.907845 8536 solver.cpp:252] Train net output #0: loss = 1.60771 (* 1 = 1.60771 loss) | |
| I1022 11:40:37.907850 8536 sgd_solver.cpp:106] Iteration 28000, lr = 0.001 | |
| I1022 11:40:48.367655 8536 solver.cpp:236] Iteration 28100, loss = 1.40457 | |
| I1022 11:40:48.367691 8536 solver.cpp:252] Train net output #0: loss = 1.40457 (* 1 = 1.40457 loss) | |
| I1022 11:40:48.367696 8536 sgd_solver.cpp:106] Iteration 28100, lr = 0.001 | |
| I1022 11:40:58.821410 8536 solver.cpp:236] Iteration 28200, loss = 1.5407 | |
| I1022 11:40:58.821439 8536 solver.cpp:252] Train net output #0: loss = 1.5407 (* 1 = 1.5407 loss) | |
| I1022 11:40:58.821444 8536 sgd_solver.cpp:106] Iteration 28200, lr = 0.001 | |
| I1022 11:41:09.278826 8536 solver.cpp:236] Iteration 28300, loss = 1.42615 | |
| I1022 11:41:09.278933 8536 solver.cpp:252] Train net output #0: loss = 1.42615 (* 1 = 1.42615 loss) | |
| I1022 11:41:09.278937 8536 sgd_solver.cpp:106] Iteration 28300, lr = 0.001 | |
| I1022 11:41:19.734524 8536 solver.cpp:236] Iteration 28400, loss = 1.34755 | |
| I1022 11:41:19.734550 8536 solver.cpp:252] Train net output #0: loss = 1.34755 (* 1 = 1.34755 loss) | |
| I1022 11:41:19.734555 8536 sgd_solver.cpp:106] Iteration 28400, lr = 0.001 | |
| I1022 11:41:30.195405 8536 solver.cpp:236] Iteration 28500, loss = 1.61975 | |
| I1022 11:41:30.195442 8536 solver.cpp:252] Train net output #0: loss = 1.61975 (* 1 = 1.61975 loss) | |
| I1022 11:41:30.195447 8536 sgd_solver.cpp:106] Iteration 28500, lr = 0.001 | |
| I1022 11:41:40.655810 8536 solver.cpp:236] Iteration 28600, loss = 1.39329 | |
| I1022 11:41:40.655881 8536 solver.cpp:252] Train net output #0: loss = 1.39329 (* 1 = 1.39329 loss) | |
| I1022 11:41:40.655887 8536 sgd_solver.cpp:106] Iteration 28600, lr = 0.001 | |
| I1022 11:41:51.121827 8536 solver.cpp:236] Iteration 28700, loss = 1.535 | |
| I1022 11:41:51.121863 8536 solver.cpp:252] Train net output #0: loss = 1.535 (* 1 = 1.535 loss) | |
| I1022 11:41:51.121868 8536 sgd_solver.cpp:106] Iteration 28700, lr = 0.001 | |
| I1022 11:42:01.590451 8536 solver.cpp:236] Iteration 28800, loss = 1.45852 | |
| I1022 11:42:01.590488 8536 solver.cpp:252] Train net output #0: loss = 1.45852 (* 1 = 1.45852 loss) | |
| I1022 11:42:01.590493 8536 sgd_solver.cpp:106] Iteration 28800, lr = 0.001 | |
| I1022 11:42:11.847002 8536 solver.cpp:236] Iteration 28900, loss = 1.34046 | |
| I1022 11:42:11.847098 8536 solver.cpp:252] Train net output #0: loss = 1.34046 (* 1 = 1.34046 loss) | |
| I1022 11:42:11.847113 8536 sgd_solver.cpp:106] Iteration 28900, lr = 0.001 | |
| I1022 11:42:22.172972 8536 solver.cpp:340] Iteration 29000, Testing net (#0) | |
| I1022 11:42:27.223444 8536 solver.cpp:408] Test net output #0: accuracy = 0.4431 | |
| I1022 11:42:27.223471 8536 solver.cpp:408] Test net output #1: loss = 1.51207 (* 1 = 1.51207 loss) | |
| I1022 11:42:27.290298 8536 solver.cpp:236] Iteration 29000, loss = 1.59525 | |
| I1022 11:42:27.290333 8536 solver.cpp:252] Train net output #0: loss = 1.59525 (* 1 = 1.59525 loss) | |
| I1022 11:42:27.290338 8536 sgd_solver.cpp:106] Iteration 29000, lr = 0.001 | |
| I1022 11:42:37.746861 8536 solver.cpp:236] Iteration 29100, loss = 1.38651 | |
| I1022 11:42:37.746891 8536 solver.cpp:252] Train net output #0: loss = 1.38651 (* 1 = 1.38651 loss) | |
| I1022 11:42:37.746897 8536 sgd_solver.cpp:106] Iteration 29100, lr = 0.001 | |
| I1022 11:42:48.209677 8536 solver.cpp:236] Iteration 29200, loss = 1.53256 | |
| I1022 11:42:48.209781 8536 solver.cpp:252] Train net output #0: loss = 1.53256 (* 1 = 1.53256 loss) | |
| I1022 11:42:48.209786 8536 sgd_solver.cpp:106] Iteration 29200, lr = 0.001 | |
| I1022 11:42:58.672240 8536 solver.cpp:236] Iteration 29300, loss = 1.40044 | |
| I1022 11:42:58.672268 8536 solver.cpp:252] Train net output #0: loss = 1.40044 (* 1 = 1.40044 loss) | |
| I1022 11:42:58.672274 8536 sgd_solver.cpp:106] Iteration 29300, lr = 0.001 | |
| I1022 11:43:09.128767 8536 solver.cpp:236] Iteration 29400, loss = 1.33163 | |
| I1022 11:43:09.128803 8536 solver.cpp:252] Train net output #0: loss = 1.33163 (* 1 = 1.33163 loss) | |
| I1022 11:43:09.128808 8536 sgd_solver.cpp:106] Iteration 29400, lr = 0.001 | |
| I1022 11:43:19.593798 8536 solver.cpp:236] Iteration 29500, loss = 1.59855 | |
| I1022 11:43:19.593894 8536 solver.cpp:252] Train net output #0: loss = 1.59855 (* 1 = 1.59855 loss) | |
| I1022 11:43:19.593900 8536 sgd_solver.cpp:106] Iteration 29500, lr = 0.001 | |
| I1022 11:43:30.049505 8536 solver.cpp:236] Iteration 29600, loss = 1.39279 | |
| I1022 11:43:30.049542 8536 solver.cpp:252] Train net output #0: loss = 1.39279 (* 1 = 1.39279 loss) | |
| I1022 11:43:30.049548 8536 sgd_solver.cpp:106] Iteration 29600, lr = 0.001 | |
| I1022 11:43:40.500507 8536 solver.cpp:236] Iteration 29700, loss = 1.52737 | |
| I1022 11:43:40.500543 8536 solver.cpp:252] Train net output #0: loss = 1.52737 (* 1 = 1.52737 loss) | |
| I1022 11:43:40.500548 8536 sgd_solver.cpp:106] Iteration 29700, lr = 0.001 | |
| I1022 11:43:50.961921 8536 solver.cpp:236] Iteration 29800, loss = 1.41553 | |
| I1022 11:43:50.961999 8536 solver.cpp:252] Train net output #0: loss = 1.41553 (* 1 = 1.41553 loss) | |
| I1022 11:43:50.962004 8536 sgd_solver.cpp:106] Iteration 29800, lr = 0.001 | |
| I1022 11:44:01.424275 8536 solver.cpp:236] Iteration 29900, loss = 1.3335 | |
| I1022 11:44:01.424310 8536 solver.cpp:252] Train net output #0: loss = 1.3335 (* 1 = 1.3335 loss) | |
| I1022 11:44:01.424315 8536 sgd_solver.cpp:106] Iteration 29900, lr = 0.001 | |
| I1022 11:44:11.556589 8536 solver.cpp:340] Iteration 30000, Testing net (#0) | |
| I1022 11:44:16.622351 8536 solver.cpp:408] Test net output #0: accuracy = 0.4487 | |
| I1022 11:44:16.622377 8536 solver.cpp:408] Test net output #1: loss = 1.50874 (* 1 = 1.50874 loss) | |
| I1022 11:44:16.689335 8536 solver.cpp:236] Iteration 30000, loss = 1.60222 | |
| I1022 11:44:16.689368 8536 solver.cpp:252] Train net output #0: loss = 1.60222 (* 1 = 1.60222 loss) | |
| I1022 11:44:16.689373 8536 sgd_solver.cpp:46] MultiStep Status: Iteration 30000, step = 1 | |
| I1022 11:44:16.689374 8536 sgd_solver.cpp:106] Iteration 30000, lr = 0.0001 | |
| I1022 11:44:27.153178 8536 solver.cpp:236] Iteration 30100, loss = 1.35916 | |
| I1022 11:44:27.153288 8536 solver.cpp:252] Train net output #0: loss = 1.35916 (* 1 = 1.35916 loss) | |
| I1022 11:44:27.153293 8536 sgd_solver.cpp:106] Iteration 30100, lr = 0.0001 | |
| I1022 11:44:37.619060 8536 solver.cpp:236] Iteration 30200, loss = 1.53312 | |
| I1022 11:44:37.619094 8536 solver.cpp:252] Train net output #0: loss = 1.53312 (* 1 = 1.53312 loss) | |
| I1022 11:44:37.619101 8536 sgd_solver.cpp:106] Iteration 30200, lr = 0.0001 | |
| I1022 11:44:48.081648 8536 solver.cpp:236] Iteration 30300, loss = 1.39462 | |
| I1022 11:44:48.081684 8536 solver.cpp:252] Train net output #0: loss = 1.39462 (* 1 = 1.39462 loss) | |
| I1022 11:44:48.081689 8536 sgd_solver.cpp:106] Iteration 30300, lr = 0.0001 | |
| I1022 11:44:58.642339 8536 solver.cpp:236] Iteration 30400, loss = 1.30419 | |
| I1022 11:44:58.642408 8536 solver.cpp:252] Train net output #0: loss = 1.30419 (* 1 = 1.30419 loss) | |
| I1022 11:44:58.642412 8536 sgd_solver.cpp:106] Iteration 30400, lr = 0.0001 | |
| I1022 11:45:09.102916 8536 solver.cpp:236] Iteration 30500, loss = 1.56812 | |
| I1022 11:45:09.102943 8536 solver.cpp:252] Train net output #0: loss = 1.56812 (* 1 = 1.56812 loss) | |
| I1022 11:45:09.102947 8536 sgd_solver.cpp:106] Iteration 30500, lr = 0.0001 | |
| I1022 11:45:19.749362 8536 solver.cpp:236] Iteration 30600, loss = 1.35886 | |
| I1022 11:45:19.749402 8536 solver.cpp:252] Train net output #0: loss = 1.35886 (* 1 = 1.35886 loss) | |
| I1022 11:45:19.749407 8536 sgd_solver.cpp:106] Iteration 30600, lr = 0.0001 | |
| I1022 11:45:30.359005 8536 solver.cpp:236] Iteration 30700, loss = 1.51926 | |
| I1022 11:45:30.359102 8536 solver.cpp:252] Train net output #0: loss = 1.51926 (* 1 = 1.51926 loss) | |
| I1022 11:45:30.359107 8536 sgd_solver.cpp:106] Iteration 30700, lr = 0.0001 | |
| I1022 11:45:40.980216 8536 solver.cpp:236] Iteration 30800, loss = 1.38614 | |
| I1022 11:45:40.980244 8536 solver.cpp:252] Train net output #0: loss = 1.38614 (* 1 = 1.38614 loss) | |
| I1022 11:45:40.980249 8536 sgd_solver.cpp:106] Iteration 30800, lr = 0.0001 | |
| I1022 11:45:51.586606 8536 solver.cpp:236] Iteration 30900, loss = 1.30159 | |
| I1022 11:45:51.586642 8536 solver.cpp:252] Train net output #0: loss = 1.30159 (* 1 = 1.30159 loss) | |
| I1022 11:45:51.586647 8536 sgd_solver.cpp:106] Iteration 30900, lr = 0.0001 | |
| I1022 11:46:01.836460 8536 solver.cpp:340] Iteration 31000, Testing net (#0) | |
| I1022 11:46:06.811063 8536 solver.cpp:408] Test net output #0: accuracy = 0.4635 | |
| I1022 11:46:06.811087 8536 solver.cpp:408] Test net output #1: loss = 1.4685 (* 1 = 1.4685 loss) | |
| I1022 11:46:06.870163 8536 solver.cpp:236] Iteration 31000, loss = 1.56621 | |
| I1022 11:46:06.870183 8536 solver.cpp:252] Train net output #0: loss = 1.56621 (* 1 = 1.56621 loss) | |
| I1022 11:46:06.870188 8536 sgd_solver.cpp:106] Iteration 31000, lr = 0.0001 | |
| I1022 11:46:17.354161 8536 solver.cpp:236] Iteration 31100, loss = 1.35175 | |
| I1022 11:46:17.354197 8536 solver.cpp:252] Train net output #0: loss = 1.35175 (* 1 = 1.35175 loss) | |
| I1022 11:46:17.354202 8536 sgd_solver.cpp:106] Iteration 31100, lr = 0.0001 | |
| I1022 11:46:27.861910 8536 solver.cpp:236] Iteration 31200, loss = 1.51541 | |
| I1022 11:46:27.861963 8536 solver.cpp:252] Train net output #0: loss = 1.51541 (* 1 = 1.51541 loss) | |
| I1022 11:46:27.861973 8536 sgd_solver.cpp:106] Iteration 31200, lr = 0.0001 | |
| I1022 11:46:38.359853 8536 solver.cpp:236] Iteration 31300, loss = 1.384 | |
| I1022 11:46:38.359959 8536 solver.cpp:252] Train net output #0: loss = 1.384 (* 1 = 1.384 loss) | |
| I1022 11:46:38.359964 8536 sgd_solver.cpp:106] Iteration 31300, lr = 0.0001 | |
| I1022 11:46:48.870753 8536 solver.cpp:236] Iteration 31400, loss = 1.30238 | |
| I1022 11:46:48.870789 8536 solver.cpp:252] Train net output #0: loss = 1.30238 (* 1 = 1.30238 loss) | |
| I1022 11:46:48.870795 8536 sgd_solver.cpp:106] Iteration 31400, lr = 0.0001 | |
| I1022 11:46:59.352591 8536 solver.cpp:236] Iteration 31500, loss = 1.56438 | |
| I1022 11:46:59.352627 8536 solver.cpp:252] Train net output #0: loss = 1.56438 (* 1 = 1.56438 loss) | |
| I1022 11:46:59.352632 8536 sgd_solver.cpp:106] Iteration 31500, lr = 0.0001 | |
| I1022 11:47:09.941921 8536 solver.cpp:236] Iteration 31600, loss = 1.34864 | |
| I1022 11:47:09.942008 8536 solver.cpp:252] Train net output #0: loss = 1.34864 (* 1 = 1.34864 loss) | |
| I1022 11:47:09.942014 8536 sgd_solver.cpp:106] Iteration 31600, lr = 0.0001 | |
| I1022 11:47:20.485656 8536 solver.cpp:236] Iteration 31700, loss = 1.51057 | |
| I1022 11:47:20.485682 8536 solver.cpp:252] Train net output #0: loss = 1.51057 (* 1 = 1.51057 loss) | |
| I1022 11:47:20.485687 8536 sgd_solver.cpp:106] Iteration 31700, lr = 0.0001 | |
| I1022 11:47:30.989984 8536 solver.cpp:236] Iteration 31800, loss = 1.38325 | |
| I1022 11:47:30.990020 8536 solver.cpp:252] Train net output #0: loss = 1.38325 (* 1 = 1.38325 loss) | |
| I1022 11:47:30.990025 8536 sgd_solver.cpp:106] Iteration 31800, lr = 0.0001 | |
| I1022 11:47:41.506654 8536 solver.cpp:236] Iteration 31900, loss = 1.30192 | |
| I1022 11:47:41.506739 8536 solver.cpp:252] Train net output #0: loss = 1.30192 (* 1 = 1.30192 loss) | |
| I1022 11:47:41.506744 8536 sgd_solver.cpp:106] Iteration 31900, lr = 0.0001 | |
| I1022 11:47:51.971623 8536 solver.cpp:340] Iteration 32000, Testing net (#0) | |
| I1022 11:47:56.997818 8536 solver.cpp:408] Test net output #0: accuracy = 0.4626 | |
| I1022 11:47:56.997854 8536 solver.cpp:408] Test net output #1: loss = 1.46698 (* 1 = 1.46698 loss) | |
| I1022 11:47:57.055127 8536 solver.cpp:236] Iteration 32000, loss = 1.56312 | |
| I1022 11:47:57.055163 8536 solver.cpp:252] Train net output #0: loss = 1.56312 (* 1 = 1.56312 loss) | |
| I1022 11:47:57.055168 8536 sgd_solver.cpp:106] Iteration 32000, lr = 0.0001 | |
| I1022 11:48:07.438297 8536 solver.cpp:236] Iteration 32100, loss = 1.34697 | |
| I1022 11:48:07.438334 8536 solver.cpp:252] Train net output #0: loss = 1.34697 (* 1 = 1.34697 loss) | |
| I1022 11:48:07.438340 8536 sgd_solver.cpp:106] Iteration 32100, lr = 0.0001 | |
| I1022 11:48:18.006460 8536 solver.cpp:236] Iteration 32200, loss = 1.50827 | |
| I1022 11:48:18.006554 8536 solver.cpp:252] Train net output #0: loss = 1.50827 (* 1 = 1.50827 loss) | |
| I1022 11:48:18.006561 8536 sgd_solver.cpp:106] Iteration 32200, lr = 0.0001 | |
| I1022 11:48:28.571450 8536 solver.cpp:236] Iteration 32300, loss = 1.38206 | |
| I1022 11:48:28.571477 8536 solver.cpp:252] Train net output #0: loss = 1.38206 (* 1 = 1.38206 loss) | |
| I1022 11:48:28.571483 8536 sgd_solver.cpp:106] Iteration 32300, lr = 0.0001 | |
| I1022 11:48:39.103261 8536 solver.cpp:236] Iteration 32400, loss = 1.30185 | |
| I1022 11:48:39.103297 8536 solver.cpp:252] Train net output #0: loss = 1.30185 (* 1 = 1.30185 loss) | |
| I1022 11:48:39.103302 8536 sgd_solver.cpp:106] Iteration 32400, lr = 0.0001 | |
| I1022 11:48:49.596376 8536 solver.cpp:236] Iteration 32500, loss = 1.56153 | |
| I1022 11:48:49.596485 8536 solver.cpp:252] Train net output #0: loss = 1.56153 (* 1 = 1.56153 loss) | |
| I1022 11:48:49.596492 8536 sgd_solver.cpp:106] Iteration 32500, lr = 0.0001 | |
| I1022 11:49:00.156345 8536 solver.cpp:236] Iteration 32600, loss = 1.34566 | |
| I1022 11:49:00.156373 8536 solver.cpp:252] Train net output #0: loss = 1.34566 (* 1 = 1.34566 loss) | |
| I1022 11:49:00.156380 8536 sgd_solver.cpp:106] Iteration 32600, lr = 0.0001 | |
| I1022 11:49:10.681100 8536 solver.cpp:236] Iteration 32700, loss = 1.5073 | |
| I1022 11:49:10.681138 8536 solver.cpp:252] Train net output #0: loss = 1.5073 (* 1 = 1.5073 loss) | |
| I1022 11:49:10.681143 8536 sgd_solver.cpp:106] Iteration 32700, lr = 0.0001 | |
| I1022 11:49:21.214411 8536 solver.cpp:236] Iteration 32800, loss = 1.38062 | |
| I1022 11:49:21.214481 8536 solver.cpp:252] Train net output #0: loss = 1.38062 (* 1 = 1.38062 loss) | |
| I1022 11:49:21.214486 8536 sgd_solver.cpp:106] Iteration 32800, lr = 0.0001 | |
| I1022 11:49:31.773218 8536 solver.cpp:236] Iteration 32900, loss = 1.30129 | |
| I1022 11:49:31.773253 8536 solver.cpp:252] Train net output #0: loss = 1.30129 (* 1 = 1.30129 loss) | |
| I1022 11:49:31.773258 8536 sgd_solver.cpp:106] Iteration 32900, lr = 0.0001 | |
| I1022 11:49:42.182524 8536 solver.cpp:340] Iteration 33000, Testing net (#0) | |
| I1022 11:49:47.386404 8536 solver.cpp:408] Test net output #0: accuracy = 0.4632 | |
| I1022 11:49:47.386440 8536 solver.cpp:408] Test net output #1: loss = 1.46573 (* 1 = 1.46573 loss) | |
| I1022 11:49:47.449969 8536 solver.cpp:236] Iteration 33000, loss = 1.55962 | |
| I1022 11:49:47.450004 8536 solver.cpp:252] Train net output #0: loss = 1.55962 (* 1 = 1.55962 loss) | |
| I1022 11:49:47.450009 8536 sgd_solver.cpp:106] Iteration 33000, lr = 0.0001 | |
| I1022 11:49:57.774360 8536 solver.cpp:236] Iteration 33100, loss = 1.34456 | |
| I1022 11:49:57.774466 8536 solver.cpp:252] Train net output #0: loss = 1.34456 (* 1 = 1.34456 loss) | |
| I1022 11:49:57.774471 8536 sgd_solver.cpp:106] Iteration 33100, lr = 0.0001 | |
| I1022 11:50:08.294697 8536 solver.cpp:236] Iteration 33200, loss = 1.50617 | |
| I1022 11:50:08.294734 8536 solver.cpp:252] Train net output #0: loss = 1.50617 (* 1 = 1.50617 loss) | |
| I1022 11:50:08.294739 8536 sgd_solver.cpp:106] Iteration 33200, lr = 0.0001 | |
| I1022 11:50:18.818411 8536 solver.cpp:236] Iteration 33300, loss = 1.3794 | |
| I1022 11:50:18.818452 8536 solver.cpp:252] Train net output #0: loss = 1.3794 (* 1 = 1.3794 loss) | |
| I1022 11:50:18.818459 8536 sgd_solver.cpp:106] Iteration 33300, lr = 0.0001 | |
| I1022 11:50:29.380736 8536 solver.cpp:236] Iteration 33400, loss = 1.30068 | |
| I1022 11:50:29.380823 8536 solver.cpp:252] Train net output #0: loss = 1.30068 (* 1 = 1.30068 loss) | |
| I1022 11:50:29.380828 8536 sgd_solver.cpp:106] Iteration 33400, lr = 0.0001 | |
| I1022 11:50:39.882395 8536 solver.cpp:236] Iteration 33500, loss = 1.55791 | |
| I1022 11:50:39.882432 8536 solver.cpp:252] Train net output #0: loss = 1.55791 (* 1 = 1.55791 loss) | |
| I1022 11:50:39.882437 8536 sgd_solver.cpp:106] Iteration 33500, lr = 0.0001 | |
| I1022 11:50:50.369966 8536 solver.cpp:236] Iteration 33600, loss = 1.34354 | |
| I1022 11:50:50.370002 8536 solver.cpp:252] Train net output #0: loss = 1.34354 (* 1 = 1.34354 loss) | |
| I1022 11:50:50.370007 8536 sgd_solver.cpp:106] Iteration 33600, lr = 0.0001 | |
| I1022 11:51:00.860945 8536 solver.cpp:236] Iteration 33700, loss = 1.50495 | |
| I1022 11:51:00.861033 8536 solver.cpp:252] Train net output #0: loss = 1.50495 (* 1 = 1.50495 loss) | |
| I1022 11:51:00.861039 8536 sgd_solver.cpp:106] Iteration 33700, lr = 0.0001 | |
| I1022 11:51:11.364581 8536 solver.cpp:236] Iteration 33800, loss = 1.37846 | |
| I1022 11:51:11.364619 8536 solver.cpp:252] Train net output #0: loss = 1.37846 (* 1 = 1.37846 loss) | |
| I1022 11:51:11.364624 8536 sgd_solver.cpp:106] Iteration 33800, lr = 0.0001 | |
| I1022 11:51:21.898196 8536 solver.cpp:236] Iteration 33900, loss = 1.30004 | |
| I1022 11:51:21.898232 8536 solver.cpp:252] Train net output #0: loss = 1.30004 (* 1 = 1.30004 loss) | |
| I1022 11:51:21.898237 8536 sgd_solver.cpp:106] Iteration 33900, lr = 0.0001 | |
| I1022 11:51:32.306247 8536 solver.cpp:340] Iteration 34000, Testing net (#0) | |
| I1022 11:51:37.394461 8536 solver.cpp:408] Test net output #0: accuracy = 0.463 | |
| I1022 11:51:37.394496 8536 solver.cpp:408] Test net output #1: loss = 1.46451 (* 1 = 1.46451 loss) | |
| I1022 11:51:37.462360 8536 solver.cpp:236] Iteration 34000, loss = 1.55647 | |
| I1022 11:51:37.462395 8536 solver.cpp:252] Train net output #0: loss = 1.55647 (* 1 = 1.55647 loss) | |
| I1022 11:51:37.462400 8536 sgd_solver.cpp:106] Iteration 34000, lr = 0.0001 | |
| I1022 11:51:47.964891 8536 solver.cpp:236] Iteration 34100, loss = 1.34263 | |
| I1022 11:51:47.964916 8536 solver.cpp:252] Train net output #0: loss = 1.34263 (* 1 = 1.34263 loss) | |
| I1022 11:51:47.964922 8536 sgd_solver.cpp:106] Iteration 34100, lr = 0.0001 | |
| I1022 11:51:58.268203 8536 solver.cpp:236] Iteration 34200, loss = 1.50366 | |
| I1022 11:51:58.268229 8536 solver.cpp:252] Train net output #0: loss = 1.50366 (* 1 = 1.50366 loss) | |
| I1022 11:51:58.268234 8536 sgd_solver.cpp:106] Iteration 34200, lr = 0.0001 | |
| I1022 11:52:08.780068 8536 solver.cpp:236] Iteration 34300, loss = 1.3776 | |
| I1022 11:52:08.780158 8536 solver.cpp:252] Train net output #0: loss = 1.3776 (* 1 = 1.3776 loss) | |
| I1022 11:52:08.780164 8536 sgd_solver.cpp:106] Iteration 34300, lr = 0.0001 | |
| I1022 11:52:19.272028 8536 solver.cpp:236] Iteration 34400, loss = 1.29945 | |
| I1022 11:52:19.272063 8536 solver.cpp:252] Train net output #0: loss = 1.29945 (* 1 = 1.29945 loss) | |
| I1022 11:52:19.272069 8536 sgd_solver.cpp:106] Iteration 34400, lr = 0.0001 | |
| I1022 11:52:29.783839 8536 solver.cpp:236] Iteration 34500, loss = 1.55522 | |
| I1022 11:52:29.783877 8536 solver.cpp:252] Train net output #0: loss = 1.55522 (* 1 = 1.55522 loss) | |
| I1022 11:52:29.783882 8536 sgd_solver.cpp:106] Iteration 34500, lr = 0.0001 | |
| I1022 11:52:40.297365 8536 solver.cpp:236] Iteration 34600, loss = 1.34177 | |
| I1022 11:52:40.297511 8536 solver.cpp:252] Train net output #0: loss = 1.34177 (* 1 = 1.34177 loss) | |
| I1022 11:52:40.297518 8536 sgd_solver.cpp:106] Iteration 34600, lr = 0.0001 | |
| I1022 11:52:50.825037 8536 solver.cpp:236] Iteration 34700, loss = 1.50237 | |
| I1022 11:52:50.825076 8536 solver.cpp:252] Train net output #0: loss = 1.50237 (* 1 = 1.50237 loss) | |
| I1022 11:52:50.825083 8536 sgd_solver.cpp:106] Iteration 34700, lr = 0.0001 | |
| I1022 11:53:01.321588 8536 solver.cpp:236] Iteration 34800, loss = 1.37683 | |
| I1022 11:53:01.321624 8536 solver.cpp:252] Train net output #0: loss = 1.37683 (* 1 = 1.37683 loss) | |
| I1022 11:53:01.321629 8536 sgd_solver.cpp:106] Iteration 34800, lr = 0.0001 | |
| I1022 11:53:11.829751 8536 solver.cpp:236] Iteration 34900, loss = 1.29893 | |
| I1022 11:53:11.829828 8536 solver.cpp:252] Train net output #0: loss = 1.29893 (* 1 = 1.29893 loss) | |
| I1022 11:53:11.829834 8536 sgd_solver.cpp:106] Iteration 34900, lr = 0.0001 | |
| I1022 11:53:22.262332 8536 solver.cpp:340] Iteration 35000, Testing net (#0) | |
| I1022 11:53:27.346029 8536 solver.cpp:408] Test net output #0: accuracy = 0.4635 | |
| I1022 11:53:27.346063 8536 solver.cpp:408] Test net output #1: loss = 1.46337 (* 1 = 1.46337 loss) | |
| I1022 11:53:27.414964 8536 solver.cpp:236] Iteration 35000, loss = 1.55403 | |
| I1022 11:53:27.415000 8536 solver.cpp:252] Train net output #0: loss = 1.55403 (* 1 = 1.55403 loss) | |
| I1022 11:53:27.415006 8536 sgd_solver.cpp:46] MultiStep Status: Iteration 35000, step = 2 | |
| I1022 11:53:27.415009 8536 sgd_solver.cpp:106] Iteration 35000, lr = 1e-05 | |
| I1022 11:53:37.923060 8536 solver.cpp:236] Iteration 35100, loss = 1.34269 | |
| I1022 11:53:37.923089 8536 solver.cpp:252] Train net output #0: loss = 1.34269 (* 1 = 1.34269 loss) | |
| I1022 11:53:37.923095 8536 sgd_solver.cpp:106] Iteration 35100, lr = 1e-05 | |
| I1022 11:53:48.417335 8536 solver.cpp:236] Iteration 35200, loss = 1.50586 | |
| I1022 11:53:48.417438 8536 solver.cpp:252] Train net output #0: loss = 1.50586 (* 1 = 1.50586 loss) | |
| I1022 11:53:48.417443 8536 sgd_solver.cpp:106] Iteration 35200, lr = 1e-05 | |
| I1022 11:53:58.963354 8536 solver.cpp:236] Iteration 35300, loss = 1.38995 | |
| I1022 11:53:58.963392 8536 solver.cpp:252] Train net output #0: loss = 1.38995 (* 1 = 1.38995 loss) | |
| I1022 11:53:58.963397 8536 sgd_solver.cpp:106] Iteration 35300, lr = 1e-05 | |
| I1022 11:54:09.442404 8536 solver.cpp:236] Iteration 35400, loss = 1.29111 | |
| I1022 11:54:09.442440 8536 solver.cpp:252] Train net output #0: loss = 1.29111 (* 1 = 1.29111 loss) | |
| I1022 11:54:09.442445 8536 sgd_solver.cpp:106] Iteration 35400, lr = 1e-05 | |
| I1022 11:54:19.973907 8536 solver.cpp:236] Iteration 35500, loss = 1.54932 | |
| I1022 11:54:19.973978 8536 solver.cpp:252] Train net output #0: loss = 1.54932 (* 1 = 1.54932 loss) | |
| I1022 11:54:19.973984 8536 sgd_solver.cpp:106] Iteration 35500, lr = 1e-05 | |
| I1022 11:54:30.503561 8536 solver.cpp:236] Iteration 35600, loss = 1.34406 | |
| I1022 11:54:30.503600 8536 solver.cpp:252] Train net output #0: loss = 1.34406 (* 1 = 1.34406 loss) | |
| I1022 11:54:30.503605 8536 sgd_solver.cpp:106] Iteration 35600, lr = 1e-05 | |
| I1022 11:54:41.000613 8536 solver.cpp:236] Iteration 35700, loss = 1.50176 | |
| I1022 11:54:41.000650 8536 solver.cpp:252] Train net output #0: loss = 1.50176 (* 1 = 1.50176 loss) | |
| I1022 11:54:41.000654 8536 sgd_solver.cpp:106] Iteration 35700, lr = 1e-05 | |
| I1022 11:54:51.552323 8536 solver.cpp:236] Iteration 35800, loss = 1.38695 | |
| I1022 11:54:51.552412 8536 solver.cpp:252] Train net output #0: loss = 1.38695 (* 1 = 1.38695 loss) | |
| I1022 11:54:51.552417 8536 sgd_solver.cpp:106] Iteration 35800, lr = 1e-05 | |
| I1022 11:55:02.025522 8536 solver.cpp:236] Iteration 35900, loss = 1.28913 | |
| I1022 11:55:02.025559 8536 solver.cpp:252] Train net output #0: loss = 1.28913 (* 1 = 1.28913 loss) | |
| I1022 11:55:02.025564 8536 sgd_solver.cpp:106] Iteration 35900, lr = 1e-05 | |
| I1022 11:55:12.383492 8536 solver.cpp:340] Iteration 36000, Testing net (#0) | |
| I1022 11:55:17.454381 8536 solver.cpp:408] Test net output #0: accuracy = 0.4624 | |
| I1022 11:55:17.454416 8536 solver.cpp:408] Test net output #1: loss = 1.46146 (* 1 = 1.46146 loss) | |
| I1022 11:55:17.519373 8536 solver.cpp:236] Iteration 36000, loss = 1.54857 | |
| I1022 11:55:17.519407 8536 solver.cpp:252] Train net output #0: loss = 1.54857 (* 1 = 1.54857 loss) | |
| I1022 11:55:17.519412 8536 sgd_solver.cpp:106] Iteration 36000, lr = 1e-05 | |
| I1022 11:55:28.058266 8536 solver.cpp:236] Iteration 36100, loss = 1.34408 | |
| I1022 11:55:28.058336 8536 solver.cpp:252] Train net output #0: loss = 1.34408 (* 1 = 1.34408 loss) | |
| I1022 11:55:28.058342 8536 sgd_solver.cpp:106] Iteration 36100, lr = 1e-05 | |
| I1022 11:55:38.576894 8536 solver.cpp:236] Iteration 36200, loss = 1.50114 | |
| I1022 11:55:38.576936 8536 solver.cpp:252] Train net output #0: loss = 1.50114 (* 1 = 1.50114 loss) | |
| I1022 11:55:38.576942 8536 sgd_solver.cpp:106] Iteration 36200, lr = 1e-05 | |
| I1022 11:55:48.832025 8536 solver.cpp:236] Iteration 36300, loss = 1.38623 | |
| I1022 11:55:48.832062 8536 solver.cpp:252] Train net output #0: loss = 1.38623 (* 1 = 1.38623 loss) | |
| I1022 11:55:48.832067 8536 sgd_solver.cpp:106] Iteration 36300, lr = 1e-05 | |
| I1022 11:55:59.301328 8536 solver.cpp:236] Iteration 36400, loss = 1.28841 | |
| I1022 11:55:59.301399 8536 solver.cpp:252] Train net output #0: loss = 1.28841 (* 1 = 1.28841 loss) | |
| I1022 11:55:59.301405 8536 sgd_solver.cpp:106] Iteration 36400, lr = 1e-05 | |
| I1022 11:56:09.872306 8536 solver.cpp:236] Iteration 36500, loss = 1.54851 | |
| I1022 11:56:09.872337 8536 solver.cpp:252] Train net output #0: loss = 1.54851 (* 1 = 1.54851 loss) | |
| I1022 11:56:09.872342 8536 sgd_solver.cpp:106] Iteration 36500, lr = 1e-05 | |
| I1022 11:56:20.589830 8536 solver.cpp:236] Iteration 36600, loss = 1.34401 | |
| I1022 11:56:20.589867 8536 solver.cpp:252] Train net output #0: loss = 1.34401 (* 1 = 1.34401 loss) | |
| I1022 11:56:20.589872 8536 sgd_solver.cpp:106] Iteration 36600, lr = 1e-05 | |
| I1022 11:56:32.078441 8536 solver.cpp:236] Iteration 36700, loss = 1.50103 | |
| I1022 11:56:32.078553 8536 solver.cpp:252] Train net output #0: loss = 1.50103 (* 1 = 1.50103 loss) | |
| I1022 11:56:32.078568 8536 sgd_solver.cpp:106] Iteration 36700, lr = 1e-05 | |
| I1022 11:56:42.593332 8536 solver.cpp:236] Iteration 36800, loss = 1.38598 | |
| I1022 11:56:42.593369 8536 solver.cpp:252] Train net output #0: loss = 1.38598 (* 1 = 1.38598 loss) | |
| I1022 11:56:42.593372 8536 sgd_solver.cpp:106] Iteration 36800, lr = 1e-05 | |
| I1022 11:56:53.084534 8536 solver.cpp:236] Iteration 36900, loss = 1.28808 | |
| I1022 11:56:53.084561 8536 solver.cpp:252] Train net output #0: loss = 1.28808 (* 1 = 1.28808 loss) | |
| I1022 11:56:53.084565 8536 sgd_solver.cpp:106] Iteration 36900, lr = 1e-05 | |
| I1022 11:57:03.611953 8536 solver.cpp:340] Iteration 37000, Testing net (#0) | |
| I1022 11:57:09.567453 8536 solver.cpp:408] Test net output #0: accuracy = 0.4627 | |
| I1022 11:57:09.567499 8536 solver.cpp:408] Test net output #1: loss = 1.46136 (* 1 = 1.46136 loss) | |
| I1022 11:57:09.636031 8536 solver.cpp:236] Iteration 37000, loss = 1.54852 | |
| I1022 11:57:09.636056 8536 solver.cpp:252] Train net output #0: loss = 1.54852 (* 1 = 1.54852 loss) | |
| I1022 11:57:09.636061 8536 sgd_solver.cpp:106] Iteration 37000, lr = 1e-05 | |
| I1022 11:57:20.137596 8536 solver.cpp:236] Iteration 37100, loss = 1.3439 | |
| I1022 11:57:20.137624 8536 solver.cpp:252] Train net output #0: loss = 1.3439 (* 1 = 1.3439 loss) | |
| I1022 11:57:20.137629 8536 sgd_solver.cpp:106] Iteration 37100, lr = 1e-05 | |
| I1022 11:57:30.688383 8536 solver.cpp:236] Iteration 37200, loss = 1.50103 | |
| I1022 11:57:30.688410 8536 solver.cpp:252] Train net output #0: loss = 1.50103 (* 1 = 1.50103 loss) | |
| I1022 11:57:30.688416 8536 sgd_solver.cpp:106] Iteration 37200, lr = 1e-05 | |
| I1022 11:57:44.473042 8536 solver.cpp:236] Iteration 37300, loss = 1.38583 | |
| I1022 11:57:44.473134 8536 solver.cpp:252] Train net output #0: loss = 1.38583 (* 1 = 1.38583 loss) | |
| I1022 11:57:44.473140 8536 sgd_solver.cpp:106] Iteration 37300, lr = 1e-05 | |
| I1022 11:57:59.218261 8536 solver.cpp:236] Iteration 37400, loss = 1.28786 | |
| I1022 11:57:59.218289 8536 solver.cpp:252] Train net output #0: loss = 1.28786 (* 1 = 1.28786 loss) | |
| I1022 11:57:59.218294 8536 sgd_solver.cpp:106] Iteration 37400, lr = 1e-05 | |
| I1022 11:58:09.733634 8536 solver.cpp:236] Iteration 37500, loss = 1.54853 | |
| I1022 11:58:09.733669 8536 solver.cpp:252] Train net output #0: loss = 1.54853 (* 1 = 1.54853 loss) | |
| I1022 11:58:09.733675 8536 sgd_solver.cpp:106] Iteration 37500, lr = 1e-05 | |
| I1022 11:58:21.074187 8536 solver.cpp:236] Iteration 37600, loss = 1.34378 | |
| I1022 11:58:21.074277 8536 solver.cpp:252] Train net output #0: loss = 1.34378 (* 1 = 1.34378 loss) | |
| I1022 11:58:21.074283 8536 sgd_solver.cpp:106] Iteration 37600, lr = 1e-05 | |
| I1022 11:58:36.137024 8536 solver.cpp:236] Iteration 37700, loss = 1.50102 | |
| I1022 11:58:36.137074 8536 solver.cpp:252] Train net output #0: loss = 1.50102 (* 1 = 1.50102 loss) | |
| I1022 11:58:36.137084 8536 sgd_solver.cpp:106] Iteration 37700, lr = 1e-05 | |
| I1022 11:58:51.222218 8536 solver.cpp:236] Iteration 37800, loss = 1.38572 | |
| I1022 11:58:51.222304 8536 solver.cpp:252] Train net output #0: loss = 1.38572 (* 1 = 1.38572 loss) | |
| I1022 11:58:51.222309 8536 sgd_solver.cpp:106] Iteration 37800, lr = 1e-05 | |
| I1022 11:59:06.287335 8536 solver.cpp:236] Iteration 37900, loss = 1.2877 | |
| I1022 11:59:06.287374 8536 solver.cpp:252] Train net output #0: loss = 1.2877 (* 1 = 1.2877 loss) | |
| I1022 11:59:06.287379 8536 sgd_solver.cpp:106] Iteration 37900, lr = 1e-05 | |
| I1022 11:59:17.783056 8536 solver.cpp:340] Iteration 38000, Testing net (#0) | |
| I1022 11:59:22.865303 8536 solver.cpp:408] Test net output #0: accuracy = 0.4628 | |
| I1022 11:59:22.865396 8536 solver.cpp:408] Test net output #1: loss = 1.46126 (* 1 = 1.46126 loss) | |
| I1022 11:59:22.928535 8536 solver.cpp:236] Iteration 38000, loss = 1.54853 | |
| I1022 11:59:22.928562 8536 solver.cpp:252] Train net output #0: loss = 1.54853 (* 1 = 1.54853 loss) | |
| I1022 11:59:22.928566 8536 sgd_solver.cpp:106] Iteration 38000, lr = 1e-05 | |
| I1022 11:59:33.442332 8536 solver.cpp:236] Iteration 38100, loss = 1.34365 | |
| I1022 11:59:33.442369 8536 solver.cpp:252] Train net output #0: loss = 1.34365 (* 1 = 1.34365 loss) | |
| I1022 11:59:33.442374 8536 sgd_solver.cpp:106] Iteration 38100, lr = 1e-05 | |
| I1022 11:59:43.953290 8536 solver.cpp:236] Iteration 38200, loss = 1.50101 | |
| I1022 11:59:43.953317 8536 solver.cpp:252] Train net output #0: loss = 1.50101 (* 1 = 1.50101 loss) | |
| I1022 11:59:43.953321 8536 sgd_solver.cpp:106] Iteration 38200, lr = 1e-05 | |
| I1022 11:59:57.230319 8536 solver.cpp:236] Iteration 38300, loss = 1.38562 | |
| I1022 11:59:57.230406 8536 solver.cpp:252] Train net output #0: loss = 1.38562 (* 1 = 1.38562 loss) | |
| I1022 11:59:57.230412 8536 sgd_solver.cpp:106] Iteration 38300, lr = 1e-05 | |
| I1022 12:00:12.137794 8536 solver.cpp:236] Iteration 38400, loss = 1.28757 | |
| I1022 12:00:12.137830 8536 solver.cpp:252] Train net output #0: loss = 1.28757 (* 1 = 1.28757 loss) | |
| I1022 12:00:12.137836 8536 sgd_solver.cpp:106] Iteration 38400, lr = 1e-05 | |
| I1022 12:00:27.207715 8536 solver.cpp:236] Iteration 38500, loss = 1.54851 | |
| I1022 12:00:27.207752 8536 solver.cpp:252] Train net output #0: loss = 1.54851 (* 1 = 1.54851 loss) | |
| I1022 12:00:27.207758 8536 sgd_solver.cpp:106] Iteration 38500, lr = 1e-05 | |
| I1022 12:00:38.647505 8536 solver.cpp:236] Iteration 38600, loss = 1.34354 | |
| I1022 12:00:38.647543 8536 solver.cpp:252] Train net output #0: loss = 1.34354 (* 1 = 1.34354 loss) | |
| I1022 12:00:38.647558 8536 sgd_solver.cpp:106] Iteration 38600, lr = 1e-05 | |
| I1022 12:00:49.226148 8536 solver.cpp:236] Iteration 38700, loss = 1.50098 | |
| I1022 12:00:49.226184 8536 solver.cpp:252] Train net output #0: loss = 1.50098 (* 1 = 1.50098 loss) | |
| I1022 12:00:49.226189 8536 sgd_solver.cpp:106] Iteration 38700, lr = 1e-05 | |
| I1022 12:00:59.797153 8536 solver.cpp:236] Iteration 38800, loss = 1.38552 | |
| I1022 12:00:59.797193 8536 solver.cpp:252] Train net output #0: loss = 1.38552 (* 1 = 1.38552 loss) | |
| I1022 12:00:59.797199 8536 sgd_solver.cpp:106] Iteration 38800, lr = 1e-05 | |
| I1022 12:01:12.047413 8536 solver.cpp:236] Iteration 38900, loss = 1.28745 | |
| I1022 12:01:12.047502 8536 solver.cpp:252] Train net output #0: loss = 1.28745 (* 1 = 1.28745 loss) | |
| I1022 12:01:12.047508 8536 sgd_solver.cpp:106] Iteration 38900, lr = 1e-05 | |
| I1022 12:01:22.470515 8536 solver.cpp:340] Iteration 39000, Testing net (#0) | |
| I1022 12:01:27.582578 8536 solver.cpp:408] Test net output #0: accuracy = 0.4627 | |
| I1022 12:01:27.582612 8536 solver.cpp:408] Test net output #1: loss = 1.46116 (* 1 = 1.46116 loss) | |
| I1022 12:01:27.647591 8536 solver.cpp:236] Iteration 39000, loss = 1.54848 | |
| I1022 12:01:27.647629 8536 solver.cpp:252] Train net output #0: loss = 1.54848 (* 1 = 1.54848 loss) | |
| I1022 12:01:27.647634 8536 sgd_solver.cpp:106] Iteration 39000, lr = 1e-05 | |
| I1022 12:01:38.190942 8536 solver.cpp:236] Iteration 39100, loss = 1.34343 | |
| I1022 12:01:38.190976 8536 solver.cpp:252] Train net output #0: loss = 1.34343 (* 1 = 1.34343 loss) | |
| I1022 12:01:38.190982 8536 sgd_solver.cpp:106] Iteration 39100, lr = 1e-05 | |
| I1022 12:01:48.655378 8536 solver.cpp:236] Iteration 39200, loss = 1.50094 | |
| I1022 12:01:48.655448 8536 solver.cpp:252] Train net output #0: loss = 1.50094 (* 1 = 1.50094 loss) | |
| I1022 12:01:48.655454 8536 sgd_solver.cpp:106] Iteration 39200, lr = 1e-05 | |
| I1022 12:01:59.202416 8536 solver.cpp:236] Iteration 39300, loss = 1.38543 | |
| I1022 12:01:59.202452 8536 solver.cpp:252] Train net output #0: loss = 1.38543 (* 1 = 1.38543 loss) | |
| I1022 12:01:59.202458 8536 sgd_solver.cpp:106] Iteration 39300, lr = 1e-05 | |
| I1022 12:02:09.744837 8536 solver.cpp:236] Iteration 39400, loss = 1.28733 | |
| I1022 12:02:09.744874 8536 solver.cpp:252] Train net output #0: loss = 1.28733 (* 1 = 1.28733 loss) | |
| I1022 12:02:09.744879 8536 sgd_solver.cpp:106] Iteration 39400, lr = 1e-05 | |
| I1022 12:02:20.084470 8536 solver.cpp:236] Iteration 39500, loss = 1.54845 | |
| I1022 12:02:20.084569 8536 solver.cpp:252] Train net output #0: loss = 1.54845 (* 1 = 1.54845 loss) | |
| I1022 12:02:20.084575 8536 sgd_solver.cpp:106] Iteration 39500, lr = 1e-05 | |
| I1022 12:02:30.590031 8536 solver.cpp:236] Iteration 39600, loss = 1.34332 | |
| I1022 12:02:30.590059 8536 solver.cpp:252] Train net output #0: loss = 1.34332 (* 1 = 1.34332 loss) | |
| I1022 12:02:30.590064 8536 sgd_solver.cpp:106] Iteration 39600, lr = 1e-05 | |
| I1022 12:02:41.090946 8536 solver.cpp:236] Iteration 39700, loss = 1.50089 | |
| I1022 12:02:41.090981 8536 solver.cpp:252] Train net output #0: loss = 1.50089 (* 1 = 1.50089 loss) | |
| I1022 12:02:41.090986 8536 sgd_solver.cpp:106] Iteration 39700, lr = 1e-05 | |
| I1022 12:02:51.588835 8536 solver.cpp:236] Iteration 39800, loss = 1.38534 | |
| I1022 12:02:51.588923 8536 solver.cpp:252] Train net output #0: loss = 1.38534 (* 1 = 1.38534 loss) | |
| I1022 12:02:51.588930 8536 sgd_solver.cpp:106] Iteration 39800, lr = 1e-05 | |
| I1022 12:03:02.116503 8536 solver.cpp:236] Iteration 39900, loss = 1.28723 | |
| I1022 12:03:02.116530 8536 solver.cpp:252] Train net output #0: loss = 1.28723 (* 1 = 1.28723 loss) | |
| I1022 12:03:02.116535 8536 sgd_solver.cpp:106] Iteration 39900, lr = 1e-05 | |
| I1022 12:03:12.569393 8536 solver.cpp:461] Snapshotting to binary proto file examples/cifar10_full_sigmoid_iter_40000.caffemodel | |
| I1022 12:03:12.607450 8536 sgd_solver.cpp:269] Snapshotting solver state to binary proto file examples/cifar10_full_sigmoid_iter_40000.solverstate | |
| I1022 12:03:12.655246 8536 solver.cpp:320] Iteration 40000, loss = 1.54841 | |
| I1022 12:03:12.655266 8536 solver.cpp:340] Iteration 40000, Testing net (#0) | |
| I1022 12:03:17.699828 8536 solver.cpp:408] Test net output #0: accuracy = 0.463 | |
| I1022 12:03:17.699864 8536 solver.cpp:408] Test net output #1: loss = 1.46104 (* 1 = 1.46104 loss) | |
| I1022 12:03:17.699867 8536 solver.cpp:325] Optimization Done. | |
| I1022 12:03:17.699869 8536 caffe.cpp:215] Optimization Done. |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| I1022 10:47:12.151319 8564 caffe.cpp:184] Using GPUs 0 | |
| I1022 10:47:12.435086 8564 solver.cpp:47] Initializing solver from parameters: | |
| test_iter: 100 | |
| test_interval: 1000 | |
| base_lr: 0.001 | |
| display: 100 | |
| max_iter: 40000 | |
| lr_policy: "multistep" | |
| gamma: 0.1 | |
| momentum: 0.9 | |
| weight_decay: 0.004 | |
| snapshot_prefix: "examples/cifar10_full_sigmoid_bn" | |
| solver_mode: GPU | |
| device_id: 0 | |
| net: "examples/cifar10/cifar10_full_sigmoid_train_test_bn.prototxt" | |
| test_initialization: false | |
| stepvalue: 30000 | |
| stepvalue: 35000 | |
| I1022 10:47:12.435207 8564 solver.cpp:90] Creating training net from net file: examples/cifar10/cifar10_full_sigmoid_train_test_bn.prototxt | |
| I1022 10:47:12.435619 8564 net.cpp:322] The NetState phase (0) differed from the phase (1) specified by a rule in layer cifar | |
| I1022 10:47:12.435631 8564 net.cpp:322] The NetState phase (0) differed from the phase (1) specified by a rule in layer bn1 | |
| I1022 10:47:12.435644 8564 net.cpp:322] The NetState phase (0) differed from the phase (1) specified by a rule in layer bn2 | |
| I1022 10:47:12.435648 8564 net.cpp:322] The NetState phase (0) differed from the phase (1) specified by a rule in layer bn3 | |
| I1022 10:47:12.435652 8564 net.cpp:322] The NetState phase (0) differed from the phase (1) specified by a rule in layer accuracy | |
| I1022 10:47:12.435748 8564 net.cpp:49] Initializing net from parameters: | |
| name: "CIFAR10_full" | |
| state { | |
| phase: TRAIN | |
| } | |
| layer { | |
| name: "cifar" | |
| type: "Data" | |
| top: "data" | |
| top: "label" | |
| include { | |
| phase: TRAIN | |
| } | |
| transform_param { | |
| mean_file: "examples/cifar10/mean.binaryproto" | |
| } | |
| data_param { | |
| source: "examples/cifar10/cifar10_train_lmdb" | |
| batch_size: 100 | |
| backend: LMDB | |
| } | |
| } | |
| layer { | |
| name: "conv1" | |
| type: "Convolution" | |
| bottom: "data" | |
| top: "conv1" | |
| param { | |
| lr_mult: 1 | |
| decay_mult: 1 | |
| } | |
| param { | |
| lr_mult: 0 | |
| decay_mult: 0 | |
| } | |
| convolution_param { | |
| num_output: 32 | |
| pad: 2 | |
| kernel_size: 5 | |
| stride: 1 | |
| weight_filler { | |
| type: "gaussian" | |
| std: 0.0001 | |
| } | |
| bias_filler { | |
| type: "constant" | |
| } | |
| } | |
| } | |
| layer { | |
| name: "pool1" | |
| type: "Pooling" | |
| bottom: "conv1" | |
| top: "pool1" | |
| pooling_param { | |
| pool: MAX | |
| kernel_size: 3 | |
| stride: 2 | |
| } | |
| } | |
| layer { | |
| name: "bn1" | |
| type: "BatchNorm" | |
| bottom: "pool1" | |
| top: "bn1" | |
| param { | |
| lr_mult: 0 | |
| } | |
| param { | |
| lr_mult: 0 | |
| } | |
| param { | |
| lr_mult: 0 | |
| } | |
| include { | |
| phase: TRAIN | |
| } | |
| batch_norm_param { | |
| use_global_stats: false | |
| } | |
| } | |
| layer { | |
| name: "Sigmoid1" | |
| type: "Sigmoid" | |
| bottom: "bn1" | |
| top: "bn1" | |
| } | |
| layer { | |
| name: "conv2" | |
| type: "Convolution" | |
| bottom: "bn1" | |
| top: "conv2" | |
| param { | |
| lr_mult: 1 | |
| decay_mult: 1 | |
| } | |
| param { | |
| lr_mult: 0 | |
| decay_mult: 0 | |
| } | |
| convolution_param { | |
| num_output: 32 | |
| pad: 2 | |
| kernel_size: 5 | |
| stride: 1 | |
| weight_filler { | |
| type: "gaussian" | |
| std: 0.01 | |
| } | |
| bias_filler { | |
| type: "constant" | |
| } | |
| } | |
| } | |
| layer { | |
| name: "bn2" | |
| type: "BatchNorm" | |
| bottom: "conv2" | |
| top: "conv2" | |
| param { | |
| lr_mult: 0 | |
| } | |
| param { | |
| lr_mult: 0 | |
| } | |
| param { | |
| lr_mult: 0 | |
| } | |
| include { | |
| phase: TRAIN | |
| } | |
| batch_norm_param { | |
| use_global_stats: false | |
| } | |
| } | |
| layer { | |
| name: "Sigmoid2" | |
| type: "Sigmoid" | |
| bottom: "conv2" | |
| top: "conv2" | |
| } | |
| layer { | |
| name: "pool2" | |
| type: "Pooling" | |
| bottom: "conv2" | |
| top: "pool2" | |
| pooling_param { | |
| pool: AVE | |
| kernel_size: 3 | |
| stride: 2 | |
| } | |
| } | |
| layer { | |
| name: "conv3" | |
| type: "Convolution" | |
| bottom: "pool2" | |
| top: "conv3" | |
| param { | |
| lr_mult: 1 | |
| decay_mult: 1 | |
| } | |
| param { | |
| lr_mult: 0 | |
| decay_mult: 0 | |
| } | |
| convolution_param { | |
| num_output: 64 | |
| pad: 2 | |
| kernel_size: 5 | |
| stride: 1 | |
| weight_filler { | |
| type: "gaussian" | |
| std: 0.01 | |
| } | |
| bias_filler { | |
| type: "constant" | |
| } | |
| } | |
| } | |
| layer { | |
| name: "bn3" | |
| type: "BatchNorm" | |
| bottom: "conv3" | |
| top: "conv3" | |
| param { | |
| lr_mult: 0 | |
| } | |
| param { | |
| lr_mult: 0 | |
| } | |
| param { | |
| lr_mult: 0 | |
| } | |
| include { | |
| phase: TRAIN | |
| } | |
| batch_norm_param { | |
| use_global_stats: false | |
| } | |
| } | |
| layer { | |
| name: "Sigmoid3" | |
| type: "Sigmoid" | |
| bottom: "conv3" | |
| top: "conv3" | |
| } | |
| layer { | |
| name: "pool3" | |
| type: "Pooling" | |
| bottom: "conv3" | |
| top: "pool3" | |
| pooling_param { | |
| pool: AVE | |
| kernel_size: 3 | |
| stride: 2 | |
| } | |
| } | |
| layer { | |
| name: "ip1" | |
| type: "InnerProduct" | |
| bottom: "pool3" | |
| top: "ip1" | |
| param { | |
| lr_mult: 1 | |
| decay_mult: 1 | |
| } | |
| param { | |
| lr_mult: 2 | |
| decay_mult: 0 | |
| } | |
| inner_product_param { | |
| num_output: 10 | |
| weight_filler { | |
| type: "gaussian" | |
| std: 0.01 | |
| } | |
| bias_filler { | |
| type: "constant" | |
| } | |
| } | |
| } | |
| layer { | |
| name: "loss" | |
| type: "SoftmaxWithLoss" | |
| bottom: "ip1" | |
| bottom: "label" | |
| top: "loss" | |
| } | |
| I1022 10:47:12.436312 8564 layer_factory.hpp:76] Creating layer cifar | |
| I1022 10:47:12.436770 8564 net.cpp:106] Creating Layer cifar | |
| I1022 10:47:12.436794 8564 net.cpp:411] cifar -> data | |
| I1022 10:47:12.436823 8564 net.cpp:411] cifar -> label | |
| I1022 10:47:12.436842 8564 data_transformer.cpp:25] Loading mean file from: examples/cifar10/mean.binaryproto | |
| I1022 10:47:12.437497 8566 db_lmdb.cpp:38] Opened lmdb examples/cifar10/cifar10_train_lmdb | |
| I1022 10:47:12.475785 8564 data_layer.cpp:45] output data size: 100,3,32,32 | |
| I1022 10:47:12.481014 8564 net.cpp:150] Setting up cifar | |
| I1022 10:47:12.481081 8564 net.cpp:157] Top shape: 100 3 32 32 (307200) | |
| I1022 10:47:12.481094 8564 net.cpp:157] Top shape: 100 (100) | |
| I1022 10:47:12.481103 8564 net.cpp:165] Memory required for data: 1229200 | |
| I1022 10:47:12.481117 8564 layer_factory.hpp:76] Creating layer conv1 | |
| I1022 10:47:12.481139 8564 net.cpp:106] Creating Layer conv1 | |
| I1022 10:47:12.481151 8564 net.cpp:454] conv1 <- data | |
| I1022 10:47:12.481169 8564 net.cpp:411] conv1 -> conv1 | |
| I1022 10:47:12.481775 8564 net.cpp:150] Setting up conv1 | |
| I1022 10:47:12.481782 8564 net.cpp:157] Top shape: 100 32 32 32 (3276800) | |
| I1022 10:47:12.481796 8564 net.cpp:165] Memory required for data: 14336400 | |
| I1022 10:47:12.481807 8564 layer_factory.hpp:76] Creating layer pool1 | |
| I1022 10:47:12.481813 8564 net.cpp:106] Creating Layer pool1 | |
| I1022 10:47:12.481830 8564 net.cpp:454] pool1 <- conv1 | |
| I1022 10:47:12.481843 8564 net.cpp:411] pool1 -> pool1 | |
| I1022 10:47:12.482364 8564 net.cpp:150] Setting up pool1 | |
| I1022 10:47:12.482389 8564 net.cpp:157] Top shape: 100 32 16 16 (819200) | |
| I1022 10:47:12.482399 8564 net.cpp:165] Memory required for data: 17613200 | |
| I1022 10:47:12.482409 8564 layer_factory.hpp:76] Creating layer bn1 | |
| I1022 10:47:12.482426 8564 net.cpp:106] Creating Layer bn1 | |
| I1022 10:47:12.482437 8564 net.cpp:454] bn1 <- pool1 | |
| I1022 10:47:12.482450 8564 net.cpp:411] bn1 -> bn1 | |
| I1022 10:47:12.482601 8564 net.cpp:150] Setting up bn1 | |
| I1022 10:47:12.482616 8564 net.cpp:157] Top shape: 100 32 16 16 (819200) | |
| I1022 10:47:12.482626 8564 net.cpp:165] Memory required for data: 20890000 | |
| I1022 10:47:12.482642 8564 layer_factory.hpp:76] Creating layer Sigmoid1 | |
| I1022 10:47:12.482655 8564 net.cpp:106] Creating Layer Sigmoid1 | |
| I1022 10:47:12.482664 8564 net.cpp:454] Sigmoid1 <- bn1 | |
| I1022 10:47:12.482674 8564 net.cpp:397] Sigmoid1 -> bn1 (in-place) | |
| I1022 10:47:12.482686 8564 net.cpp:150] Setting up Sigmoid1 | |
| I1022 10:47:12.482695 8564 net.cpp:157] Top shape: 100 32 16 16 (819200) | |
| I1022 10:47:12.482703 8564 net.cpp:165] Memory required for data: 24166800 | |
| I1022 10:47:12.482712 8564 layer_factory.hpp:76] Creating layer conv2 | |
| I1022 10:47:12.482728 8564 net.cpp:106] Creating Layer conv2 | |
| I1022 10:47:12.482738 8564 net.cpp:454] conv2 <- bn1 | |
| I1022 10:47:12.482749 8564 net.cpp:411] conv2 -> conv2 | |
| I1022 10:47:12.484033 8564 net.cpp:150] Setting up conv2 | |
| I1022 10:47:12.484043 8564 net.cpp:157] Top shape: 100 32 16 16 (819200) | |
| I1022 10:47:12.484045 8564 net.cpp:165] Memory required for data: 27443600 | |
| I1022 10:47:12.484050 8564 layer_factory.hpp:76] Creating layer bn2 | |
| I1022 10:47:12.484055 8564 net.cpp:106] Creating Layer bn2 | |
| I1022 10:47:12.484071 8564 net.cpp:454] bn2 <- conv2 | |
| I1022 10:47:12.484086 8564 net.cpp:397] bn2 -> conv2 (in-place) | |
| I1022 10:47:12.484208 8564 net.cpp:150] Setting up bn2 | |
| I1022 10:47:12.484215 8564 net.cpp:157] Top shape: 100 32 16 16 (819200) | |
| I1022 10:47:12.484216 8564 net.cpp:165] Memory required for data: 30720400 | |
| I1022 10:47:12.484223 8564 layer_factory.hpp:76] Creating layer Sigmoid2 | |
| I1022 10:47:12.484241 8564 net.cpp:106] Creating Layer Sigmoid2 | |
| I1022 10:47:12.484263 8564 net.cpp:454] Sigmoid2 <- conv2 | |
| I1022 10:47:12.484274 8564 net.cpp:397] Sigmoid2 -> conv2 (in-place) | |
| I1022 10:47:12.484287 8564 net.cpp:150] Setting up Sigmoid2 | |
| I1022 10:47:12.484297 8564 net.cpp:157] Top shape: 100 32 16 16 (819200) | |
| I1022 10:47:12.484305 8564 net.cpp:165] Memory required for data: 33997200 | |
| I1022 10:47:12.484314 8564 layer_factory.hpp:76] Creating layer pool2 | |
| I1022 10:47:12.484328 8564 net.cpp:106] Creating Layer pool2 | |
| I1022 10:47:12.484336 8564 net.cpp:454] pool2 <- conv2 | |
| I1022 10:47:12.484350 8564 net.cpp:411] pool2 -> pool2 | |
| I1022 10:47:12.484395 8564 net.cpp:150] Setting up pool2 | |
| I1022 10:47:12.484411 8564 net.cpp:157] Top shape: 100 32 8 8 (204800) | |
| I1022 10:47:12.484419 8564 net.cpp:165] Memory required for data: 34816400 | |
| I1022 10:47:12.484428 8564 layer_factory.hpp:76] Creating layer conv3 | |
| I1022 10:47:12.484442 8564 net.cpp:106] Creating Layer conv3 | |
| I1022 10:47:12.484452 8564 net.cpp:454] conv3 <- pool2 | |
| I1022 10:47:12.484463 8564 net.cpp:411] conv3 -> conv3 | |
| I1022 10:47:12.485782 8564 net.cpp:150] Setting up conv3 | |
| I1022 10:47:12.485788 8564 net.cpp:157] Top shape: 100 64 8 8 (409600) | |
| I1022 10:47:12.485791 8564 net.cpp:165] Memory required for data: 36454800 | |
| I1022 10:47:12.485795 8564 layer_factory.hpp:76] Creating layer bn3 | |
| I1022 10:47:12.485800 8564 net.cpp:106] Creating Layer bn3 | |
| I1022 10:47:12.485802 8564 net.cpp:454] bn3 <- conv3 | |
| I1022 10:47:12.485805 8564 net.cpp:397] bn3 -> conv3 (in-place) | |
| I1022 10:47:12.485916 8564 net.cpp:150] Setting up bn3 | |
| I1022 10:47:12.485932 8564 net.cpp:157] Top shape: 100 64 8 8 (409600) | |
| I1022 10:47:12.485941 8564 net.cpp:165] Memory required for data: 38093200 | |
| I1022 10:47:12.485954 8564 layer_factory.hpp:76] Creating layer Sigmoid3 | |
| I1022 10:47:12.485966 8564 net.cpp:106] Creating Layer Sigmoid3 | |
| I1022 10:47:12.485976 8564 net.cpp:454] Sigmoid3 <- conv3 | |
| I1022 10:47:12.485986 8564 net.cpp:397] Sigmoid3 -> conv3 (in-place) | |
| I1022 10:47:12.485997 8564 net.cpp:150] Setting up Sigmoid3 | |
| I1022 10:47:12.486008 8564 net.cpp:157] Top shape: 100 64 8 8 (409600) | |
| I1022 10:47:12.486016 8564 net.cpp:165] Memory required for data: 39731600 | |
| I1022 10:47:12.486027 8564 layer_factory.hpp:76] Creating layer pool3 | |
| I1022 10:47:12.486037 8564 net.cpp:106] Creating Layer pool3 | |
| I1022 10:47:12.486045 8564 net.cpp:454] pool3 <- conv3 | |
| I1022 10:47:12.486055 8564 net.cpp:411] pool3 -> pool3 | |
| I1022 10:47:12.486081 8564 net.cpp:150] Setting up pool3 | |
| I1022 10:47:12.486088 8564 net.cpp:157] Top shape: 100 64 4 4 (102400) | |
| I1022 10:47:12.486089 8564 net.cpp:165] Memory required for data: 40141200 | |
| I1022 10:47:12.486091 8564 layer_factory.hpp:76] Creating layer ip1 | |
| I1022 10:47:12.486099 8564 net.cpp:106] Creating Layer ip1 | |
| I1022 10:47:12.486101 8564 net.cpp:454] ip1 <- pool3 | |
| I1022 10:47:12.486104 8564 net.cpp:411] ip1 -> ip1 | |
| I1022 10:47:12.486748 8564 net.cpp:150] Setting up ip1 | |
| I1022 10:47:12.486757 8564 net.cpp:157] Top shape: 100 10 (1000) | |
| I1022 10:47:12.486758 8564 net.cpp:165] Memory required for data: 40145200 | |
| I1022 10:47:12.486765 8564 layer_factory.hpp:76] Creating layer loss | |
| I1022 10:47:12.486773 8564 net.cpp:106] Creating Layer loss | |
| I1022 10:47:12.486789 8564 net.cpp:454] loss <- ip1 | |
| I1022 10:47:12.486800 8564 net.cpp:454] loss <- label | |
| I1022 10:47:12.486814 8564 net.cpp:411] loss -> loss | |
| I1022 10:47:12.486829 8564 layer_factory.hpp:76] Creating layer loss | |
| I1022 10:47:12.486891 8564 net.cpp:150] Setting up loss | |
| I1022 10:47:12.486896 8564 net.cpp:157] Top shape: (1) | |
| I1022 10:47:12.486897 8564 net.cpp:160] with loss weight 1 | |
| I1022 10:47:12.486915 8564 net.cpp:165] Memory required for data: 40145204 | |
| I1022 10:47:12.486918 8564 net.cpp:226] loss needs backward computation. | |
| I1022 10:47:12.486920 8564 net.cpp:226] ip1 needs backward computation. | |
| I1022 10:47:12.486922 8564 net.cpp:226] pool3 needs backward computation. | |
| I1022 10:47:12.486924 8564 net.cpp:226] Sigmoid3 needs backward computation. | |
| I1022 10:47:12.486925 8564 net.cpp:226] bn3 needs backward computation. | |
| I1022 10:47:12.486938 8564 net.cpp:226] conv3 needs backward computation. | |
| I1022 10:47:12.486940 8564 net.cpp:226] pool2 needs backward computation. | |
| I1022 10:47:12.486943 8564 net.cpp:226] Sigmoid2 needs backward computation. | |
| I1022 10:47:12.486944 8564 net.cpp:226] bn2 needs backward computation. | |
| I1022 10:47:12.486945 8564 net.cpp:226] conv2 needs backward computation. | |
| I1022 10:47:12.486948 8564 net.cpp:226] Sigmoid1 needs backward computation. | |
| I1022 10:47:12.486949 8564 net.cpp:226] bn1 needs backward computation. | |
| I1022 10:47:12.486951 8564 net.cpp:226] pool1 needs backward computation. | |
| I1022 10:47:12.486953 8564 net.cpp:226] conv1 needs backward computation. | |
| I1022 10:47:12.486955 8564 net.cpp:228] cifar does not need backward computation. | |
| I1022 10:47:12.486958 8564 net.cpp:270] This network produces output loss | |
| I1022 10:47:12.486964 8564 net.cpp:283] Network initialization done. | |
| I1022 10:47:12.487342 8564 solver.cpp:180] Creating test net (#0) specified by net file: examples/cifar10/cifar10_full_sigmoid_train_test_bn.prototxt | |
| I1022 10:47:12.487367 8564 net.cpp:322] The NetState phase (1) differed from the phase (0) specified by a rule in layer cifar | |
| I1022 10:47:12.487373 8564 net.cpp:322] The NetState phase (1) differed from the phase (0) specified by a rule in layer bn1 | |
| I1022 10:47:12.487377 8564 net.cpp:322] The NetState phase (1) differed from the phase (0) specified by a rule in layer bn2 | |
| I1022 10:47:12.487380 8564 net.cpp:322] The NetState phase (1) differed from the phase (0) specified by a rule in layer bn3 | |
| I1022 10:47:12.487475 8564 net.cpp:49] Initializing net from parameters: | |
| name: "CIFAR10_full" | |
| state { | |
| phase: TEST | |
| } | |
| layer { | |
| name: "cifar" | |
| type: "Data" | |
| top: "data" | |
| top: "label" | |
| include { | |
| phase: TEST | |
| } | |
| transform_param { | |
| mean_file: "examples/cifar10/mean.binaryproto" | |
| } | |
| data_param { | |
| source: "examples/cifar10/cifar10_test_lmdb" | |
| batch_size: 100 | |
| backend: LMDB | |
| } | |
| } | |
| layer { | |
| name: "conv1" | |
| type: "Convolution" | |
| bottom: "data" | |
| top: "conv1" | |
| param { | |
| lr_mult: 1 | |
| decay_mult: 1 | |
| } | |
| param { | |
| lr_mult: 0 | |
| decay_mult: 0 | |
| } | |
| convolution_param { | |
| num_output: 32 | |
| pad: 2 | |
| kernel_size: 5 | |
| stride: 1 | |
| weight_filler { | |
| type: "gaussian" | |
| std: 0.0001 | |
| } | |
| bias_filler { | |
| type: "constant" | |
| } | |
| } | |
| } | |
| layer { | |
| name: "pool1" | |
| type: "Pooling" | |
| bottom: "conv1" | |
| top: "pool1" | |
| pooling_param { | |
| pool: MAX | |
| kernel_size: 3 | |
| stride: 2 | |
| } | |
| } | |
| layer { | |
| name: "bn1" | |
| type: "BatchNorm" | |
| bottom: "pool1" | |
| top: "bn1" | |
| param { | |
| lr_mult: 0 | |
| } | |
| param { | |
| lr_mult: 0 | |
| } | |
| param { | |
| lr_mult: 0 | |
| } | |
| include { | |
| phase: TEST | |
| } | |
| batch_norm_param { | |
| use_global_stats: true | |
| } | |
| } | |
| layer { | |
| name: "Sigmoid1" | |
| type: "Sigmoid" | |
| bottom: "bn1" | |
| top: "bn1" | |
| } | |
| layer { | |
| name: "conv2" | |
| type: "Convolution" | |
| bottom: "bn1" | |
| top: "conv2" | |
| param { | |
| lr_mult: 1 | |
| decay_mult: 1 | |
| } | |
| param { | |
| lr_mult: 0 | |
| decay_mult: 0 | |
| } | |
| convolution_param { | |
| num_output: 32 | |
| pad: 2 | |
| kernel_size: 5 | |
| stride: 1 | |
| weight_filler { | |
| type: "gaussian" | |
| std: 0.01 | |
| } | |
| bias_filler { | |
| type: "constant" | |
| } | |
| } | |
| } | |
| layer { | |
| name: "bn2" | |
| type: "BatchNorm" | |
| bottom: "conv2" | |
| top: "conv2" | |
| param { | |
| lr_mult: 0 | |
| } | |
| param { | |
| lr_mult: 0 | |
| } | |
| param { | |
| lr_mult: 0 | |
| } | |
| include { | |
| phase: TEST | |
| } | |
| batch_norm_param { | |
| use_global_stats: true | |
| } | |
| } | |
| layer { | |
| name: "Sigmoid2" | |
| type: "Sigmoid" | |
| bottom: "conv2" | |
| top: "conv2" | |
| } | |
| layer { | |
| name: "pool2" | |
| type: "Pooling" | |
| bottom: "conv2" | |
| top: "pool2" | |
| pooling_param { | |
| pool: AVE | |
| kernel_size: 3 | |
| stride: 2 | |
| } | |
| } | |
| layer { | |
| name: "conv3" | |
| type: "Convolution" | |
| bottom: "pool2" | |
| top: "conv3" | |
| param { | |
| lr_mult: 1 | |
| decay_mult: 1 | |
| } | |
| param { | |
| lr_mult: 0 | |
| decay_mult: 0 | |
| } | |
| convolution_param { | |
| num_output: 64 | |
| pad: 2 | |
| kernel_size: 5 | |
| stride: 1 | |
| weight_filler { | |
| type: "gaussian" | |
| std: 0.01 | |
| } | |
| bias_filler { | |
| type: "constant" | |
| } | |
| } | |
| } | |
| layer { | |
| name: "bn3" | |
| type: "BatchNorm" | |
| bottom: "conv3" | |
| top: "conv3" | |
| param { | |
| lr_mult: 0 | |
| } | |
| param { | |
| lr_mult: 0 | |
| } | |
| param { | |
| lr_mult: 0 | |
| } | |
| include { | |
| phase: TEST | |
| } | |
| batch_norm_param { | |
| use_global_stats: true | |
| } | |
| } | |
| layer { | |
| name: "Sigmoid3" | |
| type: "Sigmoid" | |
| bottom: "conv3" | |
| top: "conv3" | |
| } | |
| layer { | |
| name: "pool3" | |
| type: "Pooling" | |
| bottom: "conv3" | |
| top: "pool3" | |
| pooling_param { | |
| pool: AVE | |
| kernel_size: 3 | |
| stride: 2 | |
| } | |
| } | |
| layer { | |
| name: "ip1" | |
| type: "InnerProduct" | |
| bottom: "pool3" | |
| top: "ip1" | |
| param { | |
| lr_mult: 1 | |
| decay_mult: 1 | |
| } | |
| param { | |
| lr_mult: 2 | |
| decay_mult: 0 | |
| } | |
| inner_product_param { | |
| num_output: 10 | |
| weight_filler { | |
| type: "gaussian" | |
| std: 0.01 | |
| } | |
| bias_filler { | |
| type: "constant" | |
| } | |
| } | |
| } | |
| layer { | |
| name: "accuracy" | |
| type: "Accuracy" | |
| bottom: "ip1" | |
| bottom: "label" | |
| top: "accuracy" | |
| include { | |
| phase: TEST | |
| } | |
| } | |
| layer { | |
| name: "loss" | |
| type: "SoftmaxWithLoss" | |
| bottom: "ip1" | |
| bottom: "label" | |
| top: "loss" | |
| } | |
| I1022 10:47:12.487540 8564 layer_factory.hpp:76] Creating layer cifar | |
| I1022 10:47:12.487623 8564 net.cpp:106] Creating Layer cifar | |
| I1022 10:47:12.487627 8564 net.cpp:411] cifar -> data | |
| I1022 10:47:12.487640 8564 net.cpp:411] cifar -> label | |
| I1022 10:47:12.487645 8564 data_transformer.cpp:25] Loading mean file from: examples/cifar10/mean.binaryproto | |
| I1022 10:47:12.488327 8568 db_lmdb.cpp:38] Opened lmdb examples/cifar10/cifar10_test_lmdb | |
| I1022 10:47:12.488445 8564 data_layer.cpp:45] output data size: 100,3,32,32 | |
| I1022 10:47:12.493044 8564 net.cpp:150] Setting up cifar | |
| I1022 10:47:12.493069 8564 net.cpp:157] Top shape: 100 3 32 32 (307200) | |
| I1022 10:47:12.493072 8564 net.cpp:157] Top shape: 100 (100) | |
| I1022 10:47:12.493075 8564 net.cpp:165] Memory required for data: 1229200 | |
| I1022 10:47:12.493079 8564 layer_factory.hpp:76] Creating layer label_cifar_1_split | |
| I1022 10:47:12.493089 8564 net.cpp:106] Creating Layer label_cifar_1_split | |
| I1022 10:47:12.493091 8564 net.cpp:454] label_cifar_1_split <- label | |
| I1022 10:47:12.493096 8564 net.cpp:411] label_cifar_1_split -> label_cifar_1_split_0 | |
| I1022 10:47:12.493104 8564 net.cpp:411] label_cifar_1_split -> label_cifar_1_split_1 | |
| I1022 10:47:12.493218 8564 net.cpp:150] Setting up label_cifar_1_split | |
| I1022 10:47:12.493223 8564 net.cpp:157] Top shape: 100 (100) | |
| I1022 10:47:12.493226 8564 net.cpp:157] Top shape: 100 (100) | |
| I1022 10:47:12.493237 8564 net.cpp:165] Memory required for data: 1230000 | |
| I1022 10:47:12.493238 8564 layer_factory.hpp:76] Creating layer conv1 | |
| I1022 10:47:12.493247 8564 net.cpp:106] Creating Layer conv1 | |
| I1022 10:47:12.493248 8564 net.cpp:454] conv1 <- data | |
| I1022 10:47:12.493252 8564 net.cpp:411] conv1 -> conv1 | |
| I1022 10:47:12.493455 8564 net.cpp:150] Setting up conv1 | |
| I1022 10:47:12.493461 8564 net.cpp:157] Top shape: 100 32 32 32 (3276800) | |
| I1022 10:47:12.493473 8564 net.cpp:165] Memory required for data: 14337200 | |
| I1022 10:47:12.493479 8564 layer_factory.hpp:76] Creating layer pool1 | |
| I1022 10:47:12.493484 8564 net.cpp:106] Creating Layer pool1 | |
| I1022 10:47:12.493485 8564 net.cpp:454] pool1 <- conv1 | |
| I1022 10:47:12.493489 8564 net.cpp:411] pool1 -> pool1 | |
| I1022 10:47:12.493526 8564 net.cpp:150] Setting up pool1 | |
| I1022 10:47:12.493530 8564 net.cpp:157] Top shape: 100 32 16 16 (819200) | |
| I1022 10:47:12.493532 8564 net.cpp:165] Memory required for data: 17614000 | |
| I1022 10:47:12.493543 8564 layer_factory.hpp:76] Creating layer bn1 | |
| I1022 10:47:12.493549 8564 net.cpp:106] Creating Layer bn1 | |
| I1022 10:47:12.493551 8564 net.cpp:454] bn1 <- pool1 | |
| I1022 10:47:12.493556 8564 net.cpp:411] bn1 -> bn1 | |
| I1022 10:47:12.493680 8564 net.cpp:150] Setting up bn1 | |
| I1022 10:47:12.493685 8564 net.cpp:157] Top shape: 100 32 16 16 (819200) | |
| I1022 10:47:12.493686 8564 net.cpp:165] Memory required for data: 20890800 | |
| I1022 10:47:12.493702 8564 layer_factory.hpp:76] Creating layer Sigmoid1 | |
| I1022 10:47:12.493707 8564 net.cpp:106] Creating Layer Sigmoid1 | |
| I1022 10:47:12.493710 8564 net.cpp:454] Sigmoid1 <- bn1 | |
| I1022 10:47:12.493712 8564 net.cpp:397] Sigmoid1 -> bn1 (in-place) | |
| I1022 10:47:12.493716 8564 net.cpp:150] Setting up Sigmoid1 | |
| I1022 10:47:12.493732 8564 net.cpp:157] Top shape: 100 32 16 16 (819200) | |
| I1022 10:47:12.493734 8564 net.cpp:165] Memory required for data: 24167600 | |
| I1022 10:47:12.493736 8564 layer_factory.hpp:76] Creating layer conv2 | |
| I1022 10:47:12.493741 8564 net.cpp:106] Creating Layer conv2 | |
| I1022 10:47:12.493743 8564 net.cpp:454] conv2 <- bn1 | |
| I1022 10:47:12.493747 8564 net.cpp:411] conv2 -> conv2 | |
| I1022 10:47:12.494468 8564 net.cpp:150] Setting up conv2 | |
| I1022 10:47:12.494490 8564 net.cpp:157] Top shape: 100 32 16 16 (819200) | |
| I1022 10:47:12.494500 8564 net.cpp:165] Memory required for data: 27444400 | |
| I1022 10:47:12.495412 8564 layer_factory.hpp:76] Creating layer bn2 | |
| I1022 10:47:12.495437 8564 net.cpp:106] Creating Layer bn2 | |
| I1022 10:47:12.495450 8564 net.cpp:454] bn2 <- conv2 | |
| I1022 10:47:12.495460 8564 net.cpp:397] bn2 -> conv2 (in-place) | |
| I1022 10:47:12.495592 8564 net.cpp:150] Setting up bn2 | |
| I1022 10:47:12.495607 8564 net.cpp:157] Top shape: 100 32 16 16 (819200) | |
| I1022 10:47:12.495616 8564 net.cpp:165] Memory required for data: 30721200 | |
| I1022 10:47:12.495630 8564 layer_factory.hpp:76] Creating layer Sigmoid2 | |
| I1022 10:47:12.495642 8564 net.cpp:106] Creating Layer Sigmoid2 | |
| I1022 10:47:12.495652 8564 net.cpp:454] Sigmoid2 <- conv2 | |
| I1022 10:47:12.495661 8564 net.cpp:397] Sigmoid2 -> conv2 (in-place) | |
| I1022 10:47:12.495672 8564 net.cpp:150] Setting up Sigmoid2 | |
| I1022 10:47:12.495682 8564 net.cpp:157] Top shape: 100 32 16 16 (819200) | |
| I1022 10:47:12.495690 8564 net.cpp:165] Memory required for data: 33998000 | |
| I1022 10:47:12.495699 8564 layer_factory.hpp:76] Creating layer pool2 | |
| I1022 10:47:12.495712 8564 net.cpp:106] Creating Layer pool2 | |
| I1022 10:47:12.495720 8564 net.cpp:454] pool2 <- conv2 | |
| I1022 10:47:12.495729 8564 net.cpp:411] pool2 -> pool2 | |
| I1022 10:47:12.495950 8564 net.cpp:150] Setting up pool2 | |
| I1022 10:47:12.495965 8564 net.cpp:157] Top shape: 100 32 8 8 (204800) | |
| I1022 10:47:12.495972 8564 net.cpp:165] Memory required for data: 34817200 | |
| I1022 10:47:12.495982 8564 layer_factory.hpp:76] Creating layer conv3 | |
| I1022 10:47:12.495996 8564 net.cpp:106] Creating Layer conv3 | |
| I1022 10:47:12.496006 8564 net.cpp:454] conv3 <- pool2 | |
| I1022 10:47:12.496016 8564 net.cpp:411] conv3 -> conv3 | |
| I1022 10:47:12.497537 8564 net.cpp:150] Setting up conv3 | |
| I1022 10:47:12.497558 8564 net.cpp:157] Top shape: 100 64 8 8 (409600) | |
| I1022 10:47:12.497567 8564 net.cpp:165] Memory required for data: 36455600 | |
| I1022 10:47:12.497580 8564 layer_factory.hpp:76] Creating layer bn3 | |
| I1022 10:47:12.497591 8564 net.cpp:106] Creating Layer bn3 | |
| I1022 10:47:12.497601 8564 net.cpp:454] bn3 <- conv3 | |
| I1022 10:47:12.497611 8564 net.cpp:397] bn3 -> conv3 (in-place) | |
| I1022 10:47:12.497735 8564 net.cpp:150] Setting up bn3 | |
| I1022 10:47:12.497748 8564 net.cpp:157] Top shape: 100 64 8 8 (409600) | |
| I1022 10:47:12.497757 8564 net.cpp:165] Memory required for data: 38094000 | |
| I1022 10:47:12.497769 8564 layer_factory.hpp:76] Creating layer Sigmoid3 | |
| I1022 10:47:12.497781 8564 net.cpp:106] Creating Layer Sigmoid3 | |
| I1022 10:47:12.497788 8564 net.cpp:454] Sigmoid3 <- conv3 | |
| I1022 10:47:12.497798 8564 net.cpp:397] Sigmoid3 -> conv3 (in-place) | |
| I1022 10:47:12.497809 8564 net.cpp:150] Setting up Sigmoid3 | |
| I1022 10:47:12.497818 8564 net.cpp:157] Top shape: 100 64 8 8 (409600) | |
| I1022 10:47:12.497828 8564 net.cpp:165] Memory required for data: 39732400 | |
| I1022 10:47:12.497836 8564 layer_factory.hpp:76] Creating layer pool3 | |
| I1022 10:47:12.497848 8564 net.cpp:106] Creating Layer pool3 | |
| I1022 10:47:12.497855 8564 net.cpp:454] pool3 <- conv3 | |
| I1022 10:47:12.497866 8564 net.cpp:411] pool3 -> pool3 | |
| I1022 10:47:12.497891 8564 net.cpp:150] Setting up pool3 | |
| I1022 10:47:12.497903 8564 net.cpp:157] Top shape: 100 64 4 4 (102400) | |
| I1022 10:47:12.497911 8564 net.cpp:165] Memory required for data: 40142000 | |
| I1022 10:47:12.497921 8564 layer_factory.hpp:76] Creating layer ip1 | |
| I1022 10:47:12.497934 8564 net.cpp:106] Creating Layer ip1 | |
| I1022 10:47:12.497943 8564 net.cpp:454] ip1 <- pool3 | |
| I1022 10:47:12.497953 8564 net.cpp:411] ip1 -> ip1 | |
| I1022 10:47:12.498276 8564 net.cpp:150] Setting up ip1 | |
| I1022 10:47:12.498301 8564 net.cpp:157] Top shape: 100 10 (1000) | |
| I1022 10:47:12.498309 8564 net.cpp:165] Memory required for data: 40146000 | |
| I1022 10:47:12.498323 8564 layer_factory.hpp:76] Creating layer ip1_ip1_0_split | |
| I1022 10:47:12.498337 8564 net.cpp:106] Creating Layer ip1_ip1_0_split | |
| I1022 10:47:12.498347 8564 net.cpp:454] ip1_ip1_0_split <- ip1 | |
| I1022 10:47:12.498358 8564 net.cpp:411] ip1_ip1_0_split -> ip1_ip1_0_split_0 | |
| I1022 10:47:12.498370 8564 net.cpp:411] ip1_ip1_0_split -> ip1_ip1_0_split_1 | |
| I1022 10:47:12.498404 8564 net.cpp:150] Setting up ip1_ip1_0_split | |
| I1022 10:47:12.498416 8564 net.cpp:157] Top shape: 100 10 (1000) | |
| I1022 10:47:12.498426 8564 net.cpp:157] Top shape: 100 10 (1000) | |
| I1022 10:47:12.498435 8564 net.cpp:165] Memory required for data: 40154000 | |
| I1022 10:47:12.498443 8564 layer_factory.hpp:76] Creating layer accuracy | |
| I1022 10:47:12.498461 8564 net.cpp:106] Creating Layer accuracy | |
| I1022 10:47:12.498471 8564 net.cpp:454] accuracy <- ip1_ip1_0_split_0 | |
| I1022 10:47:12.498479 8564 net.cpp:454] accuracy <- label_cifar_1_split_0 | |
| I1022 10:47:12.498491 8564 net.cpp:411] accuracy -> accuracy | |
| I1022 10:47:12.498505 8564 net.cpp:150] Setting up accuracy | |
| I1022 10:47:12.498517 8564 net.cpp:157] Top shape: (1) | |
| I1022 10:47:12.498525 8564 net.cpp:165] Memory required for data: 40154004 | |
| I1022 10:47:12.498534 8564 layer_factory.hpp:76] Creating layer loss | |
| I1022 10:47:12.498545 8564 net.cpp:106] Creating Layer loss | |
| I1022 10:47:12.498554 8564 net.cpp:454] loss <- ip1_ip1_0_split_1 | |
| I1022 10:47:12.498564 8564 net.cpp:454] loss <- label_cifar_1_split_1 | |
| I1022 10:47:12.498576 8564 net.cpp:411] loss -> loss | |
| I1022 10:47:12.498589 8564 layer_factory.hpp:76] Creating layer loss | |
| I1022 10:47:12.498657 8564 net.cpp:150] Setting up loss | |
| I1022 10:47:12.498670 8564 net.cpp:157] Top shape: (1) | |
| I1022 10:47:12.498679 8564 net.cpp:160] with loss weight 1 | |
| I1022 10:47:12.498697 8564 net.cpp:165] Memory required for data: 40154008 | |
| I1022 10:47:12.498704 8564 net.cpp:226] loss needs backward computation. | |
| I1022 10:47:12.498714 8564 net.cpp:228] accuracy does not need backward computation. | |
| I1022 10:47:12.498724 8564 net.cpp:226] ip1_ip1_0_split needs backward computation. | |
| I1022 10:47:12.498728 8564 net.cpp:226] ip1 needs backward computation. | |
| I1022 10:47:12.498730 8564 net.cpp:226] pool3 needs backward computation. | |
| I1022 10:47:12.498733 8564 net.cpp:226] Sigmoid3 needs backward computation. | |
| I1022 10:47:12.498734 8564 net.cpp:226] bn3 needs backward computation. | |
| I1022 10:47:12.498736 8564 net.cpp:226] conv3 needs backward computation. | |
| I1022 10:47:12.498738 8564 net.cpp:226] pool2 needs backward computation. | |
| I1022 10:47:12.498739 8564 net.cpp:226] Sigmoid2 needs backward computation. | |
| I1022 10:47:12.498741 8564 net.cpp:226] bn2 needs backward computation. | |
| I1022 10:47:12.498744 8564 net.cpp:226] conv2 needs backward computation. | |
| I1022 10:47:12.498745 8564 net.cpp:226] Sigmoid1 needs backward computation. | |
| I1022 10:47:12.498746 8564 net.cpp:226] bn1 needs backward computation. | |
| I1022 10:47:12.498749 8564 net.cpp:226] pool1 needs backward computation. | |
| I1022 10:47:12.498750 8564 net.cpp:226] conv1 needs backward computation. | |
| I1022 10:47:12.498754 8564 net.cpp:228] label_cifar_1_split does not need backward computation. | |
| I1022 10:47:12.498761 8564 net.cpp:228] cifar does not need backward computation. | |
| I1022 10:47:12.498764 8564 net.cpp:270] This network produces output accuracy | |
| I1022 10:47:12.498765 8564 net.cpp:270] This network produces output loss | |
| I1022 10:47:12.498775 8564 net.cpp:283] Network initialization done. | |
| I1022 10:47:12.498841 8564 solver.cpp:59] Solver scaffolding done. | |
| I1022 10:47:12.499248 8564 caffe.cpp:212] Starting Optimization | |
| I1022 10:47:12.499254 8564 solver.cpp:287] Solving CIFAR10_full | |
| I1022 10:47:12.499256 8564 solver.cpp:288] Learning Rate Policy: multistep | |
| I1022 10:47:12.585510 8564 solver.cpp:236] Iteration 0, loss = 2.32097 | |
| I1022 10:47:12.585546 8564 solver.cpp:252] Train net output #0: loss = 2.32097 (* 1 = 2.32097 loss) | |
| I1022 10:47:12.585561 8564 sgd_solver.cpp:106] Iteration 0, lr = 0.001 | |
| I1022 10:47:23.672806 8564 solver.cpp:236] Iteration 100, loss = 2.00252 | |
| I1022 10:47:23.672839 8564 solver.cpp:252] Train net output #0: loss = 2.00252 (* 1 = 2.00252 loss) | |
| I1022 10:47:23.672844 8564 sgd_solver.cpp:106] Iteration 100, lr = 0.001 | |
| I1022 10:47:34.785794 8564 solver.cpp:236] Iteration 200, loss = 1.91384 | |
| I1022 10:47:34.785820 8564 solver.cpp:252] Train net output #0: loss = 1.91384 (* 1 = 1.91384 loss) | |
| I1022 10:47:34.785825 8564 sgd_solver.cpp:106] Iteration 200, lr = 0.001 | |
| I1022 10:47:46.001471 8564 solver.cpp:236] Iteration 300, loss = 1.67202 | |
| I1022 10:47:46.001549 8564 solver.cpp:252] Train net output #0: loss = 1.67202 (* 1 = 1.67202 loss) | |
| I1022 10:47:46.001554 8564 sgd_solver.cpp:106] Iteration 300, lr = 0.001 | |
| I1022 10:47:57.287842 8564 solver.cpp:236] Iteration 400, loss = 1.54156 | |
| I1022 10:47:57.287878 8564 solver.cpp:252] Train net output #0: loss = 1.54156 (* 1 = 1.54156 loss) | |
| I1022 10:47:57.287883 8564 sgd_solver.cpp:106] Iteration 400, lr = 0.001 | |
| I1022 10:48:08.542645 8564 solver.cpp:236] Iteration 500, loss = 1.74834 | |
| I1022 10:48:08.542682 8564 solver.cpp:252] Train net output #0: loss = 1.74834 (* 1 = 1.74834 loss) | |
| I1022 10:48:08.542690 8564 sgd_solver.cpp:106] Iteration 500, lr = 0.001 | |
| I1022 10:48:19.854049 8564 solver.cpp:236] Iteration 600, loss = 1.54865 | |
| I1022 10:48:19.854112 8564 solver.cpp:252] Train net output #0: loss = 1.54865 (* 1 = 1.54865 loss) | |
| I1022 10:48:19.854118 8564 sgd_solver.cpp:106] Iteration 600, lr = 0.001 | |
| I1022 10:48:30.431449 8564 solver.cpp:236] Iteration 700, loss = 1.63565 | |
| I1022 10:48:30.431483 8564 solver.cpp:252] Train net output #0: loss = 1.63565 (* 1 = 1.63565 loss) | |
| I1022 10:48:30.431488 8564 sgd_solver.cpp:106] Iteration 700, lr = 0.001 | |
| I1022 10:48:41.589982 8564 solver.cpp:236] Iteration 800, loss = 1.46557 | |
| I1022 10:48:41.590018 8564 solver.cpp:252] Train net output #0: loss = 1.46557 (* 1 = 1.46557 loss) | |
| I1022 10:48:41.590024 8564 sgd_solver.cpp:106] Iteration 800, lr = 0.001 | |
| I1022 10:48:52.775574 8564 solver.cpp:236] Iteration 900, loss = 1.28691 | |
| I1022 10:48:52.775684 8564 solver.cpp:252] Train net output #0: loss = 1.28691 (* 1 = 1.28691 loss) | |
| I1022 10:48:52.775691 8564 sgd_solver.cpp:106] Iteration 900, lr = 0.001 | |
| I1022 10:49:03.904574 8564 solver.cpp:340] Iteration 1000, Testing net (#0) | |
| I1022 10:49:09.307780 8564 solver.cpp:408] Test net output #0: accuracy = 0.2611 | |
| I1022 10:49:09.307806 8564 solver.cpp:408] Test net output #1: loss = 2.35121 (* 1 = 2.35121 loss) | |
| I1022 10:49:09.368461 8564 solver.cpp:236] Iteration 1000, loss = 1.56718 | |
| I1022 10:49:09.368495 8564 solver.cpp:252] Train net output #0: loss = 1.56718 (* 1 = 1.56718 loss) | |
| I1022 10:49:09.368501 8564 sgd_solver.cpp:106] Iteration 1000, lr = 0.001 | |
| I1022 10:49:20.522966 8564 solver.cpp:236] Iteration 1100, loss = 1.40465 | |
| I1022 10:49:20.523023 8564 solver.cpp:252] Train net output #0: loss = 1.40465 (* 1 = 1.40465 loss) | |
| I1022 10:49:20.523033 8564 sgd_solver.cpp:106] Iteration 1100, lr = 0.001 | |
| I1022 10:49:31.700487 8564 solver.cpp:236] Iteration 1200, loss = 1.51424 | |
| I1022 10:49:31.700564 8564 solver.cpp:252] Train net output #0: loss = 1.51424 (* 1 = 1.51424 loss) | |
| I1022 10:49:31.700572 8564 sgd_solver.cpp:106] Iteration 1200, lr = 0.001 | |
| I1022 10:49:42.873749 8564 solver.cpp:236] Iteration 1300, loss = 1.32168 | |
| I1022 10:49:42.873785 8564 solver.cpp:252] Train net output #0: loss = 1.32168 (* 1 = 1.32168 loss) | |
| I1022 10:49:42.873790 8564 sgd_solver.cpp:106] Iteration 1300, lr = 0.001 | |
| I1022 10:49:54.031334 8564 solver.cpp:236] Iteration 1400, loss = 1.10562 | |
| I1022 10:49:54.031373 8564 solver.cpp:252] Train net output #0: loss = 1.10562 (* 1 = 1.10562 loss) | |
| I1022 10:49:54.031379 8564 sgd_solver.cpp:106] Iteration 1400, lr = 0.001 | |
| I1022 10:50:05.151456 8564 solver.cpp:236] Iteration 1500, loss = 1.4976 | |
| I1022 10:50:05.151557 8564 solver.cpp:252] Train net output #0: loss = 1.4976 (* 1 = 1.4976 loss) | |
| I1022 10:50:05.151562 8564 sgd_solver.cpp:106] Iteration 1500, lr = 0.001 | |
| I1022 10:50:15.751418 8564 solver.cpp:236] Iteration 1600, loss = 1.26973 | |
| I1022 10:50:15.751456 8564 solver.cpp:252] Train net output #0: loss = 1.26973 (* 1 = 1.26973 loss) | |
| I1022 10:50:15.751461 8564 sgd_solver.cpp:106] Iteration 1600, lr = 0.001 | |
| I1022 10:50:26.941982 8564 solver.cpp:236] Iteration 1700, loss = 1.37544 | |
| I1022 10:50:26.942031 8564 solver.cpp:252] Train net output #0: loss = 1.37544 (* 1 = 1.37544 loss) | |
| I1022 10:50:26.942044 8564 sgd_solver.cpp:106] Iteration 1700, lr = 0.001 | |
| I1022 10:50:38.099313 8564 solver.cpp:236] Iteration 1800, loss = 1.21621 | |
| I1022 10:50:38.099403 8564 solver.cpp:252] Train net output #0: loss = 1.21621 (* 1 = 1.21621 loss) | |
| I1022 10:50:38.099417 8564 sgd_solver.cpp:106] Iteration 1800, lr = 0.001 | |
| I1022 10:50:49.302039 8564 solver.cpp:236] Iteration 1900, loss = 1.06617 | |
| I1022 10:50:49.302084 8564 solver.cpp:252] Train net output #0: loss = 1.06617 (* 1 = 1.06617 loss) | |
| I1022 10:50:49.302098 8564 sgd_solver.cpp:106] Iteration 1900, lr = 0.001 | |
| I1022 10:51:00.498013 8564 solver.cpp:340] Iteration 2000, Testing net (#0) | |
| I1022 10:51:05.954262 8564 solver.cpp:408] Test net output #0: accuracy = 0.433 | |
| I1022 10:51:05.954301 8564 solver.cpp:408] Test net output #1: loss = 1.59692 (* 1 = 1.59692 loss) | |
| I1022 10:51:06.020668 8564 solver.cpp:236] Iteration 2000, loss = 1.40223 | |
| I1022 10:51:06.020705 8564 solver.cpp:252] Train net output #0: loss = 1.40223 (* 1 = 1.40223 loss) | |
| I1022 10:51:06.020714 8564 sgd_solver.cpp:106] Iteration 2000, lr = 0.001 | |
| I1022 10:51:17.194422 8564 solver.cpp:236] Iteration 2100, loss = 1.20316 | |
| I1022 10:51:17.194530 8564 solver.cpp:252] Train net output #0: loss = 1.20316 (* 1 = 1.20316 loss) | |
| I1022 10:51:17.194536 8564 sgd_solver.cpp:106] Iteration 2100, lr = 0.001 | |
| I1022 10:51:28.343837 8564 solver.cpp:236] Iteration 2200, loss = 1.28942 | |
| I1022 10:51:28.343873 8564 solver.cpp:252] Train net output #0: loss = 1.28942 (* 1 = 1.28942 loss) | |
| I1022 10:51:28.343879 8564 sgd_solver.cpp:106] Iteration 2200, lr = 0.001 | |
| I1022 10:51:39.587674 8564 solver.cpp:236] Iteration 2300, loss = 1.1539 | |
| I1022 10:51:39.587702 8564 solver.cpp:252] Train net output #0: loss = 1.1539 (* 1 = 1.1539 loss) | |
| I1022 10:51:39.587708 8564 sgd_solver.cpp:106] Iteration 2300, lr = 0.001 | |
| I1022 10:51:50.779501 8564 solver.cpp:236] Iteration 2400, loss = 1.02162 | |
| I1022 10:51:50.779579 8564 solver.cpp:252] Train net output #0: loss = 1.02162 (* 1 = 1.02162 loss) | |
| I1022 10:51:50.779587 8564 sgd_solver.cpp:106] Iteration 2400, lr = 0.001 | |
| I1022 10:52:01.780865 8564 solver.cpp:236] Iteration 2500, loss = 1.34051 | |
| I1022 10:52:01.780903 8564 solver.cpp:252] Train net output #0: loss = 1.34051 (* 1 = 1.34051 loss) | |
| I1022 10:52:01.780910 8564 sgd_solver.cpp:106] Iteration 2500, lr = 0.001 | |
| I1022 10:52:12.687367 8564 solver.cpp:236] Iteration 2600, loss = 1.14407 | |
| I1022 10:52:12.687405 8564 solver.cpp:252] Train net output #0: loss = 1.14407 (* 1 = 1.14407 loss) | |
| I1022 10:52:12.687412 8564 sgd_solver.cpp:106] Iteration 2600, lr = 0.001 | |
| I1022 10:52:24.006764 8564 solver.cpp:236] Iteration 2700, loss = 1.22465 | |
| I1022 10:52:24.006851 8564 solver.cpp:252] Train net output #0: loss = 1.22465 (* 1 = 1.22465 loss) | |
| I1022 10:52:24.006857 8564 sgd_solver.cpp:106] Iteration 2700, lr = 0.001 | |
| I1022 10:52:35.180367 8564 solver.cpp:236] Iteration 2800, loss = 1.11153 | |
| I1022 10:52:35.180408 8564 solver.cpp:252] Train net output #0: loss = 1.11153 (* 1 = 1.11153 loss) | |
| I1022 10:52:35.180413 8564 sgd_solver.cpp:106] Iteration 2800, lr = 0.001 | |
| I1022 10:52:46.357211 8564 solver.cpp:236] Iteration 2900, loss = 0.991401 | |
| I1022 10:52:46.357241 8564 solver.cpp:252] Train net output #0: loss = 0.991401 (* 1 = 0.991401 loss) | |
| I1022 10:52:46.357249 8564 sgd_solver.cpp:106] Iteration 2900, lr = 0.001 | |
| I1022 10:52:57.411651 8564 solver.cpp:340] Iteration 3000, Testing net (#0) | |
| I1022 10:53:02.781502 8564 solver.cpp:408] Test net output #0: accuracy = 0.4682 | |
| I1022 10:53:02.781536 8564 solver.cpp:408] Test net output #1: loss = 1.44699 (* 1 = 1.44699 loss) | |
| I1022 10:53:02.842921 8564 solver.cpp:236] Iteration 3000, loss = 1.31119 | |
| I1022 10:53:02.842955 8564 solver.cpp:252] Train net output #0: loss = 1.31119 (* 1 = 1.31119 loss) | |
| I1022 10:53:02.842960 8564 sgd_solver.cpp:106] Iteration 3000, lr = 0.001 | |
| I1022 10:53:13.969511 8564 solver.cpp:236] Iteration 3100, loss = 1.10167 | |
| I1022 10:53:13.969535 8564 solver.cpp:252] Train net output #0: loss = 1.10167 (* 1 = 1.10167 loss) | |
| I1022 10:53:13.969540 8564 sgd_solver.cpp:106] Iteration 3100, lr = 0.001 | |
| I1022 10:53:25.092070 8564 solver.cpp:236] Iteration 3200, loss = 1.17905 | |
| I1022 10:53:25.092097 8564 solver.cpp:252] Train net output #0: loss = 1.17905 (* 1 = 1.17905 loss) | |
| I1022 10:53:25.092103 8564 sgd_solver.cpp:106] Iteration 3200, lr = 0.001 | |
| I1022 10:53:36.228162 8564 solver.cpp:236] Iteration 3300, loss = 1.07725 | |
| I1022 10:53:36.228250 8564 solver.cpp:252] Train net output #0: loss = 1.07725 (* 1 = 1.07725 loss) | |
| I1022 10:53:36.228255 8564 sgd_solver.cpp:106] Iteration 3300, lr = 0.001 | |
| I1022 10:53:47.350854 8564 solver.cpp:236] Iteration 3400, loss = 0.958348 | |
| I1022 10:53:47.350880 8564 solver.cpp:252] Train net output #0: loss = 0.958348 (* 1 = 0.958348 loss) | |
| I1022 10:53:47.350886 8564 sgd_solver.cpp:106] Iteration 3400, lr = 0.001 | |
| I1022 10:53:57.929698 8564 solver.cpp:236] Iteration 3500, loss = 1.28172 | |
| I1022 10:53:57.929733 8564 solver.cpp:252] Train net output #0: loss = 1.28172 (* 1 = 1.28172 loss) | |
| I1022 10:53:57.929736 8564 sgd_solver.cpp:106] Iteration 3500, lr = 0.001 | |
| I1022 10:54:09.044430 8564 solver.cpp:236] Iteration 3600, loss = 1.06781 | |
| I1022 10:54:09.044541 8564 solver.cpp:252] Train net output #0: loss = 1.06781 (* 1 = 1.06781 loss) | |
| I1022 10:54:09.044548 8564 sgd_solver.cpp:106] Iteration 3600, lr = 0.001 | |
| I1022 10:54:20.169020 8564 solver.cpp:236] Iteration 3700, loss = 1.14847 | |
| I1022 10:54:20.169054 8564 solver.cpp:252] Train net output #0: loss = 1.14847 (* 1 = 1.14847 loss) | |
| I1022 10:54:20.169057 8564 sgd_solver.cpp:106] Iteration 3700, lr = 0.001 | |
| I1022 10:54:31.299976 8564 solver.cpp:236] Iteration 3800, loss = 1.0454 | |
| I1022 10:54:31.300004 8564 solver.cpp:252] Train net output #0: loss = 1.0454 (* 1 = 1.0454 loss) | |
| I1022 10:54:31.300009 8564 sgd_solver.cpp:106] Iteration 3800, lr = 0.001 | |
| I1022 10:54:42.414695 8564 solver.cpp:236] Iteration 3900, loss = 0.932119 | |
| I1022 10:54:42.414767 8564 solver.cpp:252] Train net output #0: loss = 0.932119 (* 1 = 0.932119 loss) | |
| I1022 10:54:42.414773 8564 sgd_solver.cpp:106] Iteration 3900, lr = 0.001 | |
| I1022 10:54:53.471880 8564 solver.cpp:340] Iteration 4000, Testing net (#0) | |
| I1022 10:54:59.028641 8564 solver.cpp:408] Test net output #0: accuracy = 0.5081 | |
| I1022 10:54:59.028719 8564 solver.cpp:408] Test net output #1: loss = 1.34308 (* 1 = 1.34308 loss) | |
| I1022 10:54:59.091145 8564 solver.cpp:236] Iteration 4000, loss = 1.25277 | |
| I1022 10:54:59.091172 8564 solver.cpp:252] Train net output #0: loss = 1.25277 (* 1 = 1.25277 loss) | |
| I1022 10:54:59.091177 8564 sgd_solver.cpp:106] Iteration 4000, lr = 0.001 | |
| I1022 10:55:10.318908 8564 solver.cpp:236] Iteration 4100, loss = 1.0367 | |
| I1022 10:55:10.318945 8564 solver.cpp:252] Train net output #0: loss = 1.0367 (* 1 = 1.0367 loss) | |
| I1022 10:55:10.318951 8564 sgd_solver.cpp:106] Iteration 4100, lr = 0.001 | |
| I1022 10:55:21.530535 8564 solver.cpp:236] Iteration 4200, loss = 1.12556 | |
| I1022 10:55:21.530627 8564 solver.cpp:252] Train net output #0: loss = 1.12556 (* 1 = 1.12556 loss) | |
| I1022 10:55:21.530642 8564 sgd_solver.cpp:106] Iteration 4200, lr = 0.001 | |
| I1022 10:55:32.687075 8564 solver.cpp:236] Iteration 4300, loss = 1.01821 | |
| I1022 10:55:32.687113 8564 solver.cpp:252] Train net output #0: loss = 1.01821 (* 1 = 1.01821 loss) | |
| I1022 10:55:32.687119 8564 sgd_solver.cpp:106] Iteration 4300, lr = 0.001 | |
| I1022 10:55:43.568413 8564 solver.cpp:236] Iteration 4400, loss = 0.908041 | |
| I1022 10:55:43.568449 8564 solver.cpp:252] Train net output #0: loss = 0.908041 (* 1 = 0.908041 loss) | |
| I1022 10:55:43.568455 8564 sgd_solver.cpp:106] Iteration 4400, lr = 0.001 | |
| I1022 10:55:54.459111 8564 solver.cpp:236] Iteration 4500, loss = 1.2231 | |
| I1022 10:55:54.459228 8564 solver.cpp:252] Train net output #0: loss = 1.2231 (* 1 = 1.2231 loss) | |
| I1022 10:55:54.459234 8564 sgd_solver.cpp:106] Iteration 4500, lr = 0.001 | |
| I1022 10:56:05.629084 8564 solver.cpp:236] Iteration 4600, loss = 1.00847 | |
| I1022 10:56:05.629120 8564 solver.cpp:252] Train net output #0: loss = 1.00847 (* 1 = 1.00847 loss) | |
| I1022 10:56:05.629127 8564 sgd_solver.cpp:106] Iteration 4600, lr = 0.001 | |
| I1022 10:56:16.782022 8564 solver.cpp:236] Iteration 4700, loss = 1.10768 | |
| I1022 10:56:16.782059 8564 solver.cpp:252] Train net output #0: loss = 1.10768 (* 1 = 1.10768 loss) | |
| I1022 10:56:16.782064 8564 sgd_solver.cpp:106] Iteration 4700, lr = 0.001 | |
| I1022 10:56:27.917898 8564 solver.cpp:236] Iteration 4800, loss = 0.993115 | |
| I1022 10:56:27.917973 8564 solver.cpp:252] Train net output #0: loss = 0.993115 (* 1 = 0.993115 loss) | |
| I1022 10:56:27.917979 8564 sgd_solver.cpp:106] Iteration 4800, lr = 0.001 | |
| I1022 10:56:39.055897 8564 solver.cpp:236] Iteration 4900, loss = 0.88821 | |
| I1022 10:56:39.055954 8564 solver.cpp:252] Train net output #0: loss = 0.88821 (* 1 = 0.88821 loss) | |
| I1022 10:56:39.055968 8564 sgd_solver.cpp:106] Iteration 4900, lr = 0.001 | |
| I1022 10:56:50.117326 8564 solver.cpp:340] Iteration 5000, Testing net (#0) | |
| I1022 10:56:55.494091 8564 solver.cpp:408] Test net output #0: accuracy = 0.5277 | |
| I1022 10:56:55.494128 8564 solver.cpp:408] Test net output #1: loss = 1.28907 (* 1 = 1.28907 loss) | |
| I1022 10:56:55.553964 8564 solver.cpp:236] Iteration 5000, loss = 1.19346 | |
| I1022 10:56:55.553992 8564 solver.cpp:252] Train net output #0: loss = 1.19346 (* 1 = 1.19346 loss) | |
| I1022 10:56:55.553997 8564 sgd_solver.cpp:106] Iteration 5000, lr = 0.001 | |
| I1022 10:57:06.736750 8564 solver.cpp:236] Iteration 5100, loss = 0.981668 | |
| I1022 10:57:06.736826 8564 solver.cpp:252] Train net output #0: loss = 0.981668 (* 1 = 0.981668 loss) | |
| I1022 10:57:06.736840 8564 sgd_solver.cpp:106] Iteration 5100, lr = 0.001 | |
| I1022 10:57:17.931282 8564 solver.cpp:236] Iteration 5200, loss = 1.09377 | |
| I1022 10:57:17.931318 8564 solver.cpp:252] Train net output #0: loss = 1.09377 (* 1 = 1.09377 loss) | |
| I1022 10:57:17.931324 8564 sgd_solver.cpp:106] Iteration 5200, lr = 0.001 | |
| I1022 10:57:29.089121 8564 solver.cpp:236] Iteration 5300, loss = 0.970221 | |
| I1022 10:57:29.089157 8564 solver.cpp:252] Train net output #0: loss = 0.970221 (* 1 = 0.970221 loss) | |
| I1022 10:57:29.089164 8564 sgd_solver.cpp:106] Iteration 5300, lr = 0.001 | |
| I1022 10:57:39.725975 8564 solver.cpp:236] Iteration 5400, loss = 0.870625 | |
| I1022 10:57:39.726053 8564 solver.cpp:252] Train net output #0: loss = 0.870625 (* 1 = 0.870625 loss) | |
| I1022 10:57:39.726068 8564 sgd_solver.cpp:106] Iteration 5400, lr = 0.001 | |
| I1022 10:57:50.877043 8564 solver.cpp:236] Iteration 5500, loss = 1.16349 | |
| I1022 10:57:50.877079 8564 solver.cpp:252] Train net output #0: loss = 1.16349 (* 1 = 1.16349 loss) | |
| I1022 10:57:50.877084 8564 sgd_solver.cpp:106] Iteration 5500, lr = 0.001 | |
| I1022 10:58:02.045374 8564 solver.cpp:236] Iteration 5600, loss = 0.957349 | |
| I1022 10:58:02.045411 8564 solver.cpp:252] Train net output #0: loss = 0.957349 (* 1 = 0.957349 loss) | |
| I1022 10:58:02.045418 8564 sgd_solver.cpp:106] Iteration 5600, lr = 0.001 | |
| I1022 10:58:13.218036 8564 solver.cpp:236] Iteration 5700, loss = 1.07998 | |
| I1022 10:58:13.218133 8564 solver.cpp:252] Train net output #0: loss = 1.07998 (* 1 = 1.07998 loss) | |
| I1022 10:58:13.218139 8564 sgd_solver.cpp:106] Iteration 5700, lr = 0.001 | |
| I1022 10:58:24.429282 8564 solver.cpp:236] Iteration 5800, loss = 0.94905 | |
| I1022 10:58:24.429318 8564 solver.cpp:252] Train net output #0: loss = 0.94905 (* 1 = 0.94905 loss) | |
| I1022 10:58:24.429324 8564 sgd_solver.cpp:106] Iteration 5800, lr = 0.001 | |
| I1022 10:58:35.542251 8564 solver.cpp:236] Iteration 5900, loss = 0.854363 | |
| I1022 10:58:35.542279 8564 solver.cpp:252] Train net output #0: loss = 0.854363 (* 1 = 0.854363 loss) | |
| I1022 10:58:35.542284 8564 sgd_solver.cpp:106] Iteration 5900, lr = 0.001 | |
| I1022 10:58:46.627142 8564 solver.cpp:340] Iteration 6000, Testing net (#0) | |
| I1022 10:58:51.998172 8564 solver.cpp:408] Test net output #0: accuracy = 0.5365 | |
| I1022 10:58:51.998198 8564 solver.cpp:408] Test net output #1: loss = 1.28651 (* 1 = 1.28651 loss) | |
| I1022 10:58:52.059741 8564 solver.cpp:236] Iteration 6000, loss = 1.13377 | |
| I1022 10:58:52.059770 8564 solver.cpp:252] Train net output #0: loss = 1.13377 (* 1 = 1.13377 loss) | |
| I1022 10:58:52.059775 8564 sgd_solver.cpp:106] Iteration 6000, lr = 0.001 | |
| I1022 10:59:03.276649 8564 solver.cpp:236] Iteration 6100, loss = 0.935908 | |
| I1022 10:59:03.276685 8564 solver.cpp:252] Train net output #0: loss = 0.935908 (* 1 = 0.935908 loss) | |
| I1022 10:59:03.276690 8564 sgd_solver.cpp:106] Iteration 6100, lr = 0.001 | |
| I1022 10:59:14.417157 8564 solver.cpp:236] Iteration 6200, loss = 1.06605 | |
| I1022 10:59:14.417193 8564 solver.cpp:252] Train net output #0: loss = 1.06605 (* 1 = 1.06605 loss) | |
| I1022 10:59:14.417198 8564 sgd_solver.cpp:106] Iteration 6200, lr = 0.001 | |
| I1022 10:59:25.146883 8564 solver.cpp:236] Iteration 6300, loss = 0.930281 | |
| I1022 10:59:25.146986 8564 solver.cpp:252] Train net output #0: loss = 0.930281 (* 1 = 0.930281 loss) | |
| I1022 10:59:25.146991 8564 sgd_solver.cpp:106] Iteration 6300, lr = 0.001 | |
| I1022 10:59:36.180222 8564 solver.cpp:236] Iteration 6400, loss = 0.837334 | |
| I1022 10:59:36.180259 8564 solver.cpp:252] Train net output #0: loss = 0.837334 (* 1 = 0.837334 loss) | |
| I1022 10:59:36.180265 8564 sgd_solver.cpp:106] Iteration 6400, lr = 0.001 | |
| I1022 10:59:47.442986 8564 solver.cpp:236] Iteration 6500, loss = 1.10648 | |
| I1022 10:59:47.443020 8564 solver.cpp:252] Train net output #0: loss = 1.10648 (* 1 = 1.10648 loss) | |
| I1022 10:59:47.443027 8564 sgd_solver.cpp:106] Iteration 6500, lr = 0.001 | |
| I1022 10:59:58.569824 8564 solver.cpp:236] Iteration 6600, loss = 0.917222 | |
| I1022 10:59:58.569929 8564 solver.cpp:252] Train net output #0: loss = 0.917222 (* 1 = 0.917222 loss) | |
| I1022 10:59:58.569936 8564 sgd_solver.cpp:106] Iteration 6600, lr = 0.001 | |
| I1022 11:00:09.693030 8564 solver.cpp:236] Iteration 6700, loss = 1.05244 | |
| I1022 11:00:09.693065 8564 solver.cpp:252] Train net output #0: loss = 1.05244 (* 1 = 1.05244 loss) | |
| I1022 11:00:09.693071 8564 sgd_solver.cpp:106] Iteration 6700, lr = 0.001 | |
| I1022 11:00:20.911839 8564 solver.cpp:236] Iteration 6800, loss = 0.912773 | |
| I1022 11:00:20.911865 8564 solver.cpp:252] Train net output #0: loss = 0.912773 (* 1 = 0.912773 loss) | |
| I1022 11:00:20.911870 8564 sgd_solver.cpp:106] Iteration 6800, lr = 0.001 | |
| I1022 11:00:32.065438 8564 solver.cpp:236] Iteration 6900, loss = 0.821233 | |
| I1022 11:00:32.065534 8564 solver.cpp:252] Train net output #0: loss = 0.821233 (* 1 = 0.821233 loss) | |
| I1022 11:00:32.065548 8564 sgd_solver.cpp:106] Iteration 6900, lr = 0.001 | |
| I1022 11:00:43.133543 8564 solver.cpp:340] Iteration 7000, Testing net (#0) | |
| I1022 11:00:48.502912 8564 solver.cpp:408] Test net output #0: accuracy = 0.5371 | |
| I1022 11:00:48.502943 8564 solver.cpp:408] Test net output #1: loss = 1.31288 (* 1 = 1.31288 loss) | |
| I1022 11:00:48.566998 8564 solver.cpp:236] Iteration 7000, loss = 1.08283 | |
| I1022 11:00:48.567031 8564 solver.cpp:252] Train net output #0: loss = 1.08283 (* 1 = 1.08283 loss) | |
| I1022 11:00:48.567037 8564 sgd_solver.cpp:106] Iteration 7000, lr = 0.001 | |
| I1022 11:00:59.689734 8564 solver.cpp:236] Iteration 7100, loss = 0.900495 | |
| I1022 11:00:59.689760 8564 solver.cpp:252] Train net output #0: loss = 0.900495 (* 1 = 0.900495 loss) | |
| I1022 11:00:59.689766 8564 sgd_solver.cpp:106] Iteration 7100, lr = 0.001 | |
| I1022 11:01:10.820148 8564 solver.cpp:236] Iteration 7200, loss = 1.04202 | |
| I1022 11:01:10.820230 8564 solver.cpp:252] Train net output #0: loss = 1.04202 (* 1 = 1.04202 loss) | |
| I1022 11:01:10.820243 8564 sgd_solver.cpp:106] Iteration 7200, lr = 0.001 | |
| I1022 11:01:21.419250 8564 solver.cpp:236] Iteration 7300, loss = 0.897834 | |
| I1022 11:01:21.419277 8564 solver.cpp:252] Train net output #0: loss = 0.897834 (* 1 = 0.897834 loss) | |
| I1022 11:01:21.419283 8564 sgd_solver.cpp:106] Iteration 7300, lr = 0.001 | |
| I1022 11:01:32.553884 8564 solver.cpp:236] Iteration 7400, loss = 0.807126 | |
| I1022 11:01:32.553920 8564 solver.cpp:252] Train net output #0: loss = 0.807126 (* 1 = 0.807126 loss) | |
| I1022 11:01:32.553926 8564 sgd_solver.cpp:106] Iteration 7400, lr = 0.001 | |
| I1022 11:01:43.689129 8564 solver.cpp:236] Iteration 7500, loss = 1.06264 | |
| I1022 11:01:43.689229 8564 solver.cpp:252] Train net output #0: loss = 1.06264 (* 1 = 1.06264 loss) | |
| I1022 11:01:43.689244 8564 sgd_solver.cpp:106] Iteration 7500, lr = 0.001 | |
| I1022 11:01:54.822784 8564 solver.cpp:236] Iteration 7600, loss = 0.886083 | |
| I1022 11:01:54.822820 8564 solver.cpp:252] Train net output #0: loss = 0.886083 (* 1 = 0.886083 loss) | |
| I1022 11:01:54.822826 8564 sgd_solver.cpp:106] Iteration 7600, lr = 0.001 | |
| I1022 11:02:05.959888 8564 solver.cpp:236] Iteration 7700, loss = 1.03317 | |
| I1022 11:02:05.959926 8564 solver.cpp:252] Train net output #0: loss = 1.03317 (* 1 = 1.03317 loss) | |
| I1022 11:02:05.959931 8564 sgd_solver.cpp:106] Iteration 7700, lr = 0.001 | |
| I1022 11:02:17.088526 8564 solver.cpp:236] Iteration 7800, loss = 0.88481 | |
| I1022 11:02:17.088672 8564 solver.cpp:252] Train net output #0: loss = 0.88481 (* 1 = 0.88481 loss) | |
| I1022 11:02:17.088678 8564 sgd_solver.cpp:106] Iteration 7800, lr = 0.001 | |
| I1022 11:02:28.228469 8564 solver.cpp:236] Iteration 7900, loss = 0.794562 | |
| I1022 11:02:28.228497 8564 solver.cpp:252] Train net output #0: loss = 0.794562 (* 1 = 0.794562 loss) | |
| I1022 11:02:28.228502 8564 sgd_solver.cpp:106] Iteration 7900, lr = 0.001 | |
| I1022 11:02:39.296113 8564 solver.cpp:340] Iteration 8000, Testing net (#0) | |
| I1022 11:02:44.658010 8564 solver.cpp:408] Test net output #0: accuracy = 0.5319 | |
| I1022 11:02:44.658043 8564 solver.cpp:408] Test net output #1: loss = 1.34801 (* 1 = 1.34801 loss) | |
| I1022 11:02:44.722715 8564 solver.cpp:236] Iteration 8000, loss = 1.0467 | |
| I1022 11:02:44.722740 8564 solver.cpp:252] Train net output #0: loss = 1.0467 (* 1 = 1.0467 loss) | |
| I1022 11:02:44.722746 8564 sgd_solver.cpp:106] Iteration 8000, lr = 0.001 | |
| I1022 11:02:55.872843 8564 solver.cpp:236] Iteration 8100, loss = 0.87316 | |
| I1022 11:02:55.872930 8564 solver.cpp:252] Train net output #0: loss = 0.87316 (* 1 = 0.87316 loss) | |
| I1022 11:02:55.872936 8564 sgd_solver.cpp:106] Iteration 8100, lr = 0.001 | |
| I1022 11:03:06.519069 8564 solver.cpp:236] Iteration 8200, loss = 1.027 | |
| I1022 11:03:06.519096 8564 solver.cpp:252] Train net output #0: loss = 1.027 (* 1 = 1.027 loss) | |
| I1022 11:03:06.519103 8564 sgd_solver.cpp:106] Iteration 8200, lr = 0.001 | |
| I1022 11:03:17.697319 8564 solver.cpp:236] Iteration 8300, loss = 0.875373 | |
| I1022 11:03:17.697355 8564 solver.cpp:252] Train net output #0: loss = 0.875373 (* 1 = 0.875373 loss) | |
| I1022 11:03:17.697361 8564 sgd_solver.cpp:106] Iteration 8300, lr = 0.001 | |
| I1022 11:03:28.966516 8564 solver.cpp:236] Iteration 8400, loss = 0.784034 | |
| I1022 11:03:28.966579 8564 solver.cpp:252] Train net output #0: loss = 0.784034 (* 1 = 0.784034 loss) | |
| I1022 11:03:28.966585 8564 sgd_solver.cpp:106] Iteration 8400, lr = 0.001 | |
| I1022 11:03:40.141294 8564 solver.cpp:236] Iteration 8500, loss = 1.03405 | |
| I1022 11:03:40.141330 8564 solver.cpp:252] Train net output #0: loss = 1.03405 (* 1 = 1.03405 loss) | |
| I1022 11:03:40.141336 8564 sgd_solver.cpp:106] Iteration 8500, lr = 0.001 | |
| I1022 11:03:51.266589 8564 solver.cpp:236] Iteration 8600, loss = 0.861715 | |
| I1022 11:03:51.266628 8564 solver.cpp:252] Train net output #0: loss = 0.861715 (* 1 = 0.861715 loss) | |
| I1022 11:03:51.266633 8564 sgd_solver.cpp:106] Iteration 8600, lr = 0.001 | |
| I1022 11:04:02.381254 8564 solver.cpp:236] Iteration 8700, loss = 1.02292 | |
| I1022 11:04:02.381361 8564 solver.cpp:252] Train net output #0: loss = 1.02292 (* 1 = 1.02292 loss) | |
| I1022 11:04:02.381366 8564 sgd_solver.cpp:106] Iteration 8700, lr = 0.001 | |
| I1022 11:04:13.497334 8564 solver.cpp:236] Iteration 8800, loss = 0.867321 | |
| I1022 11:04:13.497371 8564 solver.cpp:252] Train net output #0: loss = 0.867321 (* 1 = 0.867321 loss) | |
| I1022 11:04:13.497377 8564 sgd_solver.cpp:106] Iteration 8800, lr = 0.001 | |
| I1022 11:04:24.643878 8564 solver.cpp:236] Iteration 8900, loss = 0.774756 | |
| I1022 11:04:24.643906 8564 solver.cpp:252] Train net output #0: loss = 0.774756 (* 1 = 0.774756 loss) | |
| I1022 11:04:24.643911 8564 sgd_solver.cpp:106] Iteration 8900, lr = 0.001 | |
| I1022 11:04:35.720197 8564 solver.cpp:340] Iteration 9000, Testing net (#0) | |
| I1022 11:04:41.095862 8564 solver.cpp:408] Test net output #0: accuracy = 0.5283 | |
| I1022 11:04:41.095897 8564 solver.cpp:408] Test net output #1: loss = 1.37123 (* 1 = 1.37123 loss) | |
| I1022 11:04:41.155951 8564 solver.cpp:236] Iteration 9000, loss = 1.02307 | |
| I1022 11:04:41.155977 8564 solver.cpp:252] Train net output #0: loss = 1.02307 (* 1 = 1.02307 loss) | |
| I1022 11:04:41.155982 8564 sgd_solver.cpp:106] Iteration 9000, lr = 0.001 | |
| I1022 11:04:52.246351 8564 solver.cpp:236] Iteration 9100, loss = 0.851682 | |
| I1022 11:04:52.246389 8564 solver.cpp:252] Train net output #0: loss = 0.851682 (* 1 = 0.851682 loss) | |
| I1022 11:04:52.246397 8564 sgd_solver.cpp:106] Iteration 9100, lr = 0.001 | |
| I1022 11:05:02.972261 8564 solver.cpp:236] Iteration 9200, loss = 1.02035 | |
| I1022 11:05:02.972296 8564 solver.cpp:252] Train net output #0: loss = 1.02035 (* 1 = 1.02035 loss) | |
| I1022 11:05:02.972302 8564 sgd_solver.cpp:106] Iteration 9200, lr = 0.001 | |
| I1022 11:05:14.125594 8564 solver.cpp:236] Iteration 9300, loss = 0.859746 | |
| I1022 11:05:14.125676 8564 solver.cpp:252] Train net output #0: loss = 0.859746 (* 1 = 0.859746 loss) | |
| I1022 11:05:14.125691 8564 sgd_solver.cpp:106] Iteration 9300, lr = 0.001 | |
| I1022 11:05:25.269296 8564 solver.cpp:236] Iteration 9400, loss = 0.766938 | |
| I1022 11:05:25.269353 8564 solver.cpp:252] Train net output #0: loss = 0.766938 (* 1 = 0.766938 loss) | |
| I1022 11:05:25.269366 8564 sgd_solver.cpp:106] Iteration 9400, lr = 0.001 | |
| I1022 11:05:36.419596 8564 solver.cpp:236] Iteration 9500, loss = 1.01155 | |
| I1022 11:05:36.419631 8564 solver.cpp:252] Train net output #0: loss = 1.01155 (* 1 = 1.01155 loss) | |
| I1022 11:05:36.419637 8564 sgd_solver.cpp:106] Iteration 9500, lr = 0.001 | |
| I1022 11:05:47.633718 8564 solver.cpp:236] Iteration 9600, loss = 0.843507 | |
| I1022 11:05:47.633806 8564 solver.cpp:252] Train net output #0: loss = 0.843507 (* 1 = 0.843507 loss) | |
| I1022 11:05:47.633812 8564 sgd_solver.cpp:106] Iteration 9600, lr = 0.001 | |
| I1022 11:05:58.928779 8564 solver.cpp:236] Iteration 9700, loss = 1.01803 | |
| I1022 11:05:58.928814 8564 solver.cpp:252] Train net output #0: loss = 1.01803 (* 1 = 1.01803 loss) | |
| I1022 11:05:58.928819 8564 sgd_solver.cpp:106] Iteration 9700, lr = 0.001 | |
| I1022 11:06:10.038100 8564 solver.cpp:236] Iteration 9800, loss = 0.852406 | |
| I1022 11:06:10.038135 8564 solver.cpp:252] Train net output #0: loss = 0.852406 (* 1 = 0.852406 loss) | |
| I1022 11:06:10.038141 8564 sgd_solver.cpp:106] Iteration 9800, lr = 0.001 | |
| I1022 11:06:21.140257 8564 solver.cpp:236] Iteration 9900, loss = 0.7607 | |
| I1022 11:06:21.140364 8564 solver.cpp:252] Train net output #0: loss = 0.7607 (* 1 = 0.7607 loss) | |
| I1022 11:06:21.140370 8564 sgd_solver.cpp:106] Iteration 9900, lr = 0.001 | |
| I1022 11:06:32.189060 8564 solver.cpp:340] Iteration 10000, Testing net (#0) | |
| I1022 11:06:37.536795 8564 solver.cpp:408] Test net output #0: accuracy = 0.5362 | |
| I1022 11:06:37.536830 8564 solver.cpp:408] Test net output #1: loss = 1.35428 (* 1 = 1.35428 loss) | |
| I1022 11:06:37.596839 8564 solver.cpp:236] Iteration 10000, loss = 0.999949 | |
| I1022 11:06:37.596855 8564 solver.cpp:252] Train net output #0: loss = 0.999949 (* 1 = 0.999949 loss) | |
| I1022 11:06:37.596860 8564 sgd_solver.cpp:106] Iteration 10000, lr = 0.001 | |
| I1022 11:06:48.214331 8564 solver.cpp:236] Iteration 10100, loss = 0.836307 | |
| I1022 11:06:48.214370 8564 solver.cpp:252] Train net output #0: loss = 0.836307 (* 1 = 0.836307 loss) | |
| I1022 11:06:48.214375 8564 sgd_solver.cpp:106] Iteration 10100, lr = 0.001 | |
| I1022 11:06:59.703713 8564 solver.cpp:236] Iteration 10200, loss = 1.01446 | |
| I1022 11:06:59.703809 8564 solver.cpp:252] Train net output #0: loss = 1.01446 (* 1 = 1.01446 loss) | |
| I1022 11:06:59.703815 8564 sgd_solver.cpp:106] Iteration 10200, lr = 0.001 | |
| I1022 11:07:10.990681 8564 solver.cpp:236] Iteration 10300, loss = 0.844063 | |
| I1022 11:07:10.990718 8564 solver.cpp:252] Train net output #0: loss = 0.844063 (* 1 = 0.844063 loss) | |
| I1022 11:07:10.990725 8564 sgd_solver.cpp:106] Iteration 10300, lr = 0.001 | |
| I1022 11:07:22.325598 8564 solver.cpp:236] Iteration 10400, loss = 0.756001 | |
| I1022 11:07:22.325623 8564 solver.cpp:252] Train net output #0: loss = 0.756001 (* 1 = 0.756001 loss) | |
| I1022 11:07:22.325629 8564 sgd_solver.cpp:106] Iteration 10400, lr = 0.001 | |
| I1022 11:07:33.638324 8564 solver.cpp:236] Iteration 10500, loss = 0.989357 | |
| I1022 11:07:33.638408 8564 solver.cpp:252] Train net output #0: loss = 0.989357 (* 1 = 0.989357 loss) | |
| I1022 11:07:33.638413 8564 sgd_solver.cpp:106] Iteration 10500, lr = 0.001 | |
| I1022 11:07:44.758133 8564 solver.cpp:236] Iteration 10600, loss = 0.830034 | |
| I1022 11:07:44.758170 8564 solver.cpp:252] Train net output #0: loss = 0.830034 (* 1 = 0.830034 loss) | |
| I1022 11:07:44.758175 8564 sgd_solver.cpp:106] Iteration 10600, lr = 0.001 | |
| I1022 11:07:55.887380 8564 solver.cpp:236] Iteration 10700, loss = 1.01092 | |
| I1022 11:07:55.887404 8564 solver.cpp:252] Train net output #0: loss = 1.01092 (* 1 = 1.01092 loss) | |
| I1022 11:07:55.887409 8564 sgd_solver.cpp:106] Iteration 10700, lr = 0.001 | |
| I1022 11:08:07.015717 8564 solver.cpp:236] Iteration 10800, loss = 0.836887 | |
| I1022 11:08:07.015804 8564 solver.cpp:252] Train net output #0: loss = 0.836887 (* 1 = 0.836887 loss) | |
| I1022 11:08:07.015810 8564 sgd_solver.cpp:106] Iteration 10800, lr = 0.001 | |
| I1022 11:08:18.133378 8564 solver.cpp:236] Iteration 10900, loss = 0.750266 | |
| I1022 11:08:18.133406 8564 solver.cpp:252] Train net output #0: loss = 0.750266 (* 1 = 0.750266 loss) | |
| I1022 11:08:18.133412 8564 sgd_solver.cpp:106] Iteration 10900, lr = 0.001 | |
| I1022 11:08:29.220507 8564 solver.cpp:340] Iteration 11000, Testing net (#0) | |
| I1022 11:08:34.411219 8564 solver.cpp:408] Test net output #0: accuracy = 0.5407 | |
| I1022 11:08:34.411245 8564 solver.cpp:408] Test net output #1: loss = 1.3382 (* 1 = 1.3382 loss) | |
| I1022 11:08:34.469854 8564 solver.cpp:236] Iteration 11000, loss = 0.979585 | |
| I1022 11:08:34.469892 8564 solver.cpp:252] Train net output #0: loss = 0.979585 (* 1 = 0.979585 loss) | |
| I1022 11:08:34.469898 8564 sgd_solver.cpp:106] Iteration 11000, lr = 0.001 | |
| I1022 11:08:45.416860 8564 solver.cpp:236] Iteration 11100, loss = 0.825167 | |
| I1022 11:08:45.416954 8564 solver.cpp:252] Train net output #0: loss = 0.825167 (* 1 = 0.825167 loss) | |
| I1022 11:08:45.416968 8564 sgd_solver.cpp:106] Iteration 11100, lr = 0.001 | |
| I1022 11:08:56.604718 8564 solver.cpp:236] Iteration 11200, loss = 1.00726 | |
| I1022 11:08:56.604755 8564 solver.cpp:252] Train net output #0: loss = 1.00726 (* 1 = 1.00726 loss) | |
| I1022 11:08:56.604761 8564 sgd_solver.cpp:106] Iteration 11200, lr = 0.001 | |
| I1022 11:09:07.828991 8564 solver.cpp:236] Iteration 11300, loss = 0.829028 | |
| I1022 11:09:07.829046 8564 solver.cpp:252] Train net output #0: loss = 0.829028 (* 1 = 0.829028 loss) | |
| I1022 11:09:07.829058 8564 sgd_solver.cpp:106] Iteration 11300, lr = 0.001 | |
| I1022 11:09:19.032860 8564 solver.cpp:236] Iteration 11400, loss = 0.746158 | |
| I1022 11:09:19.032960 8564 solver.cpp:252] Train net output #0: loss = 0.746158 (* 1 = 0.746158 loss) | |
| I1022 11:09:19.032966 8564 sgd_solver.cpp:106] Iteration 11400, lr = 0.001 | |
| I1022 11:09:30.211771 8564 solver.cpp:236] Iteration 11500, loss = 0.970272 | |
| I1022 11:09:30.211805 8564 solver.cpp:252] Train net output #0: loss = 0.970272 (* 1 = 0.970272 loss) | |
| I1022 11:09:30.211810 8564 sgd_solver.cpp:106] Iteration 11500, lr = 0.001 | |
| I1022 11:09:41.342051 8564 solver.cpp:236] Iteration 11600, loss = 0.821736 | |
| I1022 11:09:41.342087 8564 solver.cpp:252] Train net output #0: loss = 0.821736 (* 1 = 0.821736 loss) | |
| I1022 11:09:41.342092 8564 sgd_solver.cpp:106] Iteration 11600, lr = 0.001 | |
| I1022 11:09:52.452190 8564 solver.cpp:236] Iteration 11700, loss = 1.00449 | |
| I1022 11:09:52.452307 8564 solver.cpp:252] Train net output #0: loss = 1.00449 (* 1 = 1.00449 loss) | |
| I1022 11:09:52.452313 8564 sgd_solver.cpp:106] Iteration 11700, lr = 0.001 | |
| I1022 11:10:03.584844 8564 solver.cpp:236] Iteration 11800, loss = 0.822382 | |
| I1022 11:10:03.584882 8564 solver.cpp:252] Train net output #0: loss = 0.822382 (* 1 = 0.822382 loss) | |
| I1022 11:10:03.584887 8564 sgd_solver.cpp:106] Iteration 11800, lr = 0.001 | |
| I1022 11:10:14.722473 8564 solver.cpp:236] Iteration 11900, loss = 0.742509 | |
| I1022 11:10:14.722512 8564 solver.cpp:252] Train net output #0: loss = 0.742509 (* 1 = 0.742509 loss) | |
| I1022 11:10:14.722517 8564 sgd_solver.cpp:106] Iteration 11900, lr = 0.001 | |
| I1022 11:10:25.526278 8564 solver.cpp:340] Iteration 12000, Testing net (#0) | |
| I1022 11:10:31.367588 8564 solver.cpp:408] Test net output #0: accuracy = 0.5481 | |
| I1022 11:10:31.367638 8564 solver.cpp:408] Test net output #1: loss = 1.31082 (* 1 = 1.31082 loss) | |
| I1022 11:10:31.427549 8564 solver.cpp:236] Iteration 12000, loss = 0.961934 | |
| I1022 11:10:31.427590 8564 solver.cpp:252] Train net output #0: loss = 0.961934 (* 1 = 0.961934 loss) | |
| I1022 11:10:31.427603 8564 sgd_solver.cpp:106] Iteration 12000, lr = 0.001 | |
| I1022 11:10:42.697722 8564 solver.cpp:236] Iteration 12100, loss = 0.818081 | |
| I1022 11:10:42.697759 8564 solver.cpp:252] Train net output #0: loss = 0.818081 (* 1 = 0.818081 loss) | |
| I1022 11:10:42.697765 8564 sgd_solver.cpp:106] Iteration 12100, lr = 0.001 | |
| I1022 11:10:53.799393 8564 solver.cpp:236] Iteration 12200, loss = 1.0012 | |
| I1022 11:10:53.799429 8564 solver.cpp:252] Train net output #0: loss = 1.0012 (* 1 = 1.0012 loss) | |
| I1022 11:10:53.799435 8564 sgd_solver.cpp:106] Iteration 12200, lr = 0.001 | |
| I1022 11:11:04.892750 8564 solver.cpp:236] Iteration 12300, loss = 0.815408 | |
| I1022 11:11:04.892848 8564 solver.cpp:252] Train net output #0: loss = 0.815408 (* 1 = 0.815408 loss) | |
| I1022 11:11:04.892863 8564 sgd_solver.cpp:106] Iteration 12300, lr = 0.001 | |
| I1022 11:11:15.972353 8564 solver.cpp:236] Iteration 12400, loss = 0.738728 | |
| I1022 11:11:15.972391 8564 solver.cpp:252] Train net output #0: loss = 0.738728 (* 1 = 0.738728 loss) | |
| I1022 11:11:15.972396 8564 sgd_solver.cpp:106] Iteration 12400, lr = 0.001 | |
| I1022 11:11:27.199014 8564 solver.cpp:236] Iteration 12500, loss = 0.954308 | |
| I1022 11:11:27.199050 8564 solver.cpp:252] Train net output #0: loss = 0.954308 (* 1 = 0.954308 loss) | |
| I1022 11:11:27.199056 8564 sgd_solver.cpp:106] Iteration 12500, lr = 0.001 | |
| I1022 11:11:38.343364 8564 solver.cpp:236] Iteration 12600, loss = 0.814748 | |
| I1022 11:11:38.343442 8564 solver.cpp:252] Train net output #0: loss = 0.814748 (* 1 = 0.814748 loss) | |
| I1022 11:11:38.343448 8564 sgd_solver.cpp:106] Iteration 12600, lr = 0.001 | |
| I1022 11:11:49.477916 8564 solver.cpp:236] Iteration 12700, loss = 0.998423 | |
| I1022 11:11:49.477952 8564 solver.cpp:252] Train net output #0: loss = 0.998423 (* 1 = 0.998423 loss) | |
| I1022 11:11:49.477958 8564 sgd_solver.cpp:106] Iteration 12700, lr = 0.001 | |
| I1022 11:12:00.666256 8564 solver.cpp:236] Iteration 12800, loss = 0.808999 | |
| I1022 11:12:00.666290 8564 solver.cpp:252] Train net output #0: loss = 0.808999 (* 1 = 0.808999 loss) | |
| I1022 11:12:00.666296 8564 sgd_solver.cpp:106] Iteration 12800, lr = 0.001 | |
| I1022 11:12:11.909124 8564 solver.cpp:236] Iteration 12900, loss = 0.735762 | |
| I1022 11:12:11.909216 8564 solver.cpp:252] Train net output #0: loss = 0.735762 (* 1 = 0.735762 loss) | |
| I1022 11:12:11.909224 8564 sgd_solver.cpp:106] Iteration 12900, lr = 0.001 | |
| I1022 11:12:22.464344 8564 solver.cpp:340] Iteration 13000, Testing net (#0) | |
| I1022 11:12:27.836658 8564 solver.cpp:408] Test net output #0: accuracy = 0.5545 | |
| I1022 11:12:27.836694 8564 solver.cpp:408] Test net output #1: loss = 1.29025 (* 1 = 1.29025 loss) | |
| I1022 11:12:27.902407 8564 solver.cpp:236] Iteration 13000, loss = 0.947423 | |
| I1022 11:12:27.902444 8564 solver.cpp:252] Train net output #0: loss = 0.947423 (* 1 = 0.947423 loss) | |
| I1022 11:12:27.902451 8564 sgd_solver.cpp:106] Iteration 13000, lr = 0.001 | |
| I1022 11:12:39.058715 8564 solver.cpp:236] Iteration 13100, loss = 0.811838 | |
| I1022 11:12:39.058750 8564 solver.cpp:252] Train net output #0: loss = 0.811838 (* 1 = 0.811838 loss) | |
| I1022 11:12:39.058756 8564 sgd_solver.cpp:106] Iteration 13100, lr = 0.001 | |
| I1022 11:12:50.252737 8564 solver.cpp:236] Iteration 13200, loss = 0.996014 | |
| I1022 11:12:50.252817 8564 solver.cpp:252] Train net output #0: loss = 0.996014 (* 1 = 0.996014 loss) | |
| I1022 11:12:50.252825 8564 sgd_solver.cpp:106] Iteration 13200, lr = 0.001 | |
| I1022 11:13:01.459527 8564 solver.cpp:236] Iteration 13300, loss = 0.803881 | |
| I1022 11:13:01.459558 8564 solver.cpp:252] Train net output #0: loss = 0.803881 (* 1 = 0.803881 loss) | |
| I1022 11:13:01.459563 8564 sgd_solver.cpp:106] Iteration 13300, lr = 0.001 | |
| I1022 11:13:12.829118 8564 solver.cpp:236] Iteration 13400, loss = 0.733281 | |
| I1022 11:13:12.829152 8564 solver.cpp:252] Train net output #0: loss = 0.733281 (* 1 = 0.733281 loss) | |
| I1022 11:13:12.829157 8564 sgd_solver.cpp:106] Iteration 13400, lr = 0.001 | |
| I1022 11:13:23.933081 8564 solver.cpp:236] Iteration 13500, loss = 0.940187 | |
| I1022 11:13:23.933150 8564 solver.cpp:252] Train net output #0: loss = 0.940187 (* 1 = 0.940187 loss) | |
| I1022 11:13:23.933156 8564 sgd_solver.cpp:106] Iteration 13500, lr = 0.001 | |
| I1022 11:13:35.036816 8564 solver.cpp:236] Iteration 13600, loss = 0.808968 | |
| I1022 11:13:35.036855 8564 solver.cpp:252] Train net output #0: loss = 0.808968 (* 1 = 0.808968 loss) | |
| I1022 11:13:35.036860 8564 sgd_solver.cpp:106] Iteration 13600, lr = 0.001 | |
| I1022 11:13:46.142202 8564 solver.cpp:236] Iteration 13700, loss = 0.9926 | |
| I1022 11:13:46.142241 8564 solver.cpp:252] Train net output #0: loss = 0.9926 (* 1 = 0.9926 loss) | |
| I1022 11:13:46.142246 8564 sgd_solver.cpp:106] Iteration 13700, lr = 0.001 | |
| I1022 11:13:57.239224 8564 solver.cpp:236] Iteration 13800, loss = 0.798586 | |
| I1022 11:13:57.239305 8564 solver.cpp:252] Train net output #0: loss = 0.798586 (* 1 = 0.798586 loss) | |
| I1022 11:13:57.239311 8564 sgd_solver.cpp:106] Iteration 13800, lr = 0.001 | |
| I1022 11:14:08.067500 8564 solver.cpp:236] Iteration 13900, loss = 0.731287 | |
| I1022 11:14:08.067539 8564 solver.cpp:252] Train net output #0: loss = 0.731287 (* 1 = 0.731287 loss) | |
| I1022 11:14:08.067545 8564 sgd_solver.cpp:106] Iteration 13900, lr = 0.001 | |
| I1022 11:14:19.069638 8564 solver.cpp:340] Iteration 14000, Testing net (#0) | |
| I1022 11:14:24.406167 8564 solver.cpp:408] Test net output #0: accuracy = 0.5616 | |
| I1022 11:14:24.406201 8564 solver.cpp:408] Test net output #1: loss = 1.26744 (* 1 = 1.26744 loss) | |
| I1022 11:14:24.467852 8564 solver.cpp:236] Iteration 14000, loss = 0.933317 | |
| I1022 11:14:24.467885 8564 solver.cpp:252] Train net output #0: loss = 0.933317 (* 1 = 0.933317 loss) | |
| I1022 11:14:24.467892 8564 sgd_solver.cpp:106] Iteration 14000, lr = 0.001 | |
| I1022 11:14:35.680349 8564 solver.cpp:236] Iteration 14100, loss = 0.80619 | |
| I1022 11:14:35.680428 8564 solver.cpp:252] Train net output #0: loss = 0.80619 (* 1 = 0.80619 loss) | |
| I1022 11:14:35.680444 8564 sgd_solver.cpp:106] Iteration 14100, lr = 0.001 | |
| I1022 11:14:47.013340 8564 solver.cpp:236] Iteration 14200, loss = 0.989585 | |
| I1022 11:14:47.013378 8564 solver.cpp:252] Train net output #0: loss = 0.989585 (* 1 = 0.989585 loss) | |
| I1022 11:14:47.013384 8564 sgd_solver.cpp:106] Iteration 14200, lr = 0.001 | |
| I1022 11:14:58.274118 8564 solver.cpp:236] Iteration 14300, loss = 0.793235 | |
| I1022 11:14:58.274148 8564 solver.cpp:252] Train net output #0: loss = 0.793235 (* 1 = 0.793235 loss) | |
| I1022 11:14:58.274154 8564 sgd_solver.cpp:106] Iteration 14300, lr = 0.001 | |
| I1022 11:15:09.438632 8564 solver.cpp:236] Iteration 14400, loss = 0.729674 | |
| I1022 11:15:09.438715 8564 solver.cpp:252] Train net output #0: loss = 0.729674 (* 1 = 0.729674 loss) | |
| I1022 11:15:09.438720 8564 sgd_solver.cpp:106] Iteration 14400, lr = 0.001 | |
| I1022 11:15:20.577165 8564 solver.cpp:236] Iteration 14500, loss = 0.926367 | |
| I1022 11:15:20.577200 8564 solver.cpp:252] Train net output #0: loss = 0.926367 (* 1 = 0.926367 loss) | |
| I1022 11:15:20.577204 8564 sgd_solver.cpp:106] Iteration 14500, lr = 0.001 | |
| I1022 11:15:31.728301 8564 solver.cpp:236] Iteration 14600, loss = 0.804492 | |
| I1022 11:15:31.728334 8564 solver.cpp:252] Train net output #0: loss = 0.804492 (* 1 = 0.804492 loss) | |
| I1022 11:15:31.728339 8564 sgd_solver.cpp:106] Iteration 14600, lr = 0.001 | |
| I1022 11:15:42.894465 8564 solver.cpp:236] Iteration 14700, loss = 0.985793 | |
| I1022 11:15:42.894574 8564 solver.cpp:252] Train net output #0: loss = 0.985793 (* 1 = 0.985793 loss) | |
| I1022 11:15:42.894588 8564 sgd_solver.cpp:106] Iteration 14700, lr = 0.001 | |
| I1022 11:15:54.076133 8564 solver.cpp:236] Iteration 14800, loss = 0.788194 | |
| I1022 11:15:54.076169 8564 solver.cpp:252] Train net output #0: loss = 0.788194 (* 1 = 0.788194 loss) | |
| I1022 11:15:54.076174 8564 sgd_solver.cpp:106] Iteration 14800, lr = 0.001 | |
| I1022 11:16:04.727046 8564 solver.cpp:236] Iteration 14900, loss = 0.727778 | |
| I1022 11:16:04.727082 8564 solver.cpp:252] Train net output #0: loss = 0.727778 (* 1 = 0.727778 loss) | |
| I1022 11:16:04.727089 8564 sgd_solver.cpp:106] Iteration 14900, lr = 0.001 | |
| I1022 11:16:15.877264 8564 solver.cpp:340] Iteration 15000, Testing net (#0) | |
| I1022 11:16:21.270452 8564 solver.cpp:408] Test net output #0: accuracy = 0.5694 | |
| I1022 11:16:21.270490 8564 solver.cpp:408] Test net output #1: loss = 1.24658 (* 1 = 1.24658 loss) | |
| I1022 11:16:21.331501 8564 solver.cpp:236] Iteration 15000, loss = 0.91994 | |
| I1022 11:16:21.331537 8564 solver.cpp:252] Train net output #0: loss = 0.91994 (* 1 = 0.91994 loss) | |
| I1022 11:16:21.331542 8564 sgd_solver.cpp:106] Iteration 15000, lr = 0.001 | |
| I1022 11:16:32.508613 8564 solver.cpp:236] Iteration 15100, loss = 0.802954 | |
| I1022 11:16:32.508647 8564 solver.cpp:252] Train net output #0: loss = 0.802954 (* 1 = 0.802954 loss) | |
| I1022 11:16:32.508652 8564 sgd_solver.cpp:106] Iteration 15100, lr = 0.001 | |
| I1022 11:16:43.667011 8564 solver.cpp:236] Iteration 15200, loss = 0.982127 | |
| I1022 11:16:43.667048 8564 solver.cpp:252] Train net output #0: loss = 0.982127 (* 1 = 0.982127 loss) | |
| I1022 11:16:43.667053 8564 sgd_solver.cpp:106] Iteration 15200, lr = 0.001 | |
| I1022 11:16:54.787158 8564 solver.cpp:236] Iteration 15300, loss = 0.784026 | |
| I1022 11:16:54.787225 8564 solver.cpp:252] Train net output #0: loss = 0.784026 (* 1 = 0.784026 loss) | |
| I1022 11:16:54.787230 8564 sgd_solver.cpp:106] Iteration 15300, lr = 0.001 | |
| I1022 11:17:08.938073 8564 solver.cpp:236] Iteration 15400, loss = 0.726201 | |
| I1022 11:17:08.938112 8564 solver.cpp:252] Train net output #0: loss = 0.726201 (* 1 = 0.726201 loss) | |
| I1022 11:17:08.938118 8564 sgd_solver.cpp:106] Iteration 15400, lr = 0.001 | |
| I1022 11:17:25.518004 8564 solver.cpp:236] Iteration 15500, loss = 0.914209 | |
| I1022 11:17:25.518091 8564 solver.cpp:252] Train net output #0: loss = 0.914209 (* 1 = 0.914209 loss) | |
| I1022 11:17:25.518105 8564 sgd_solver.cpp:106] Iteration 15500, lr = 0.001 | |
| I1022 11:17:41.969053 8564 solver.cpp:236] Iteration 15600, loss = 0.800772 | |
| I1022 11:17:41.969089 8564 solver.cpp:252] Train net output #0: loss = 0.800772 (* 1 = 0.800772 loss) | |
| I1022 11:17:41.969094 8564 sgd_solver.cpp:106] Iteration 15600, lr = 0.001 | |
| I1022 11:17:58.458911 8564 solver.cpp:236] Iteration 15700, loss = 0.979371 | |
| I1022 11:17:58.458995 8564 solver.cpp:252] Train net output #0: loss = 0.979371 (* 1 = 0.979371 loss) | |
| I1022 11:17:58.459012 8564 sgd_solver.cpp:106] Iteration 15700, lr = 0.001 | |
| I1022 11:18:14.624419 8564 solver.cpp:236] Iteration 15800, loss = 0.780081 | |
| I1022 11:18:14.624439 8564 solver.cpp:252] Train net output #0: loss = 0.780081 (* 1 = 0.780081 loss) | |
| I1022 11:18:14.624446 8564 sgd_solver.cpp:106] Iteration 15800, lr = 0.001 | |
| I1022 11:18:31.274682 8564 solver.cpp:236] Iteration 15900, loss = 0.724637 | |
| I1022 11:18:31.274773 8564 solver.cpp:252] Train net output #0: loss = 0.724637 (* 1 = 0.724637 loss) | |
| I1022 11:18:31.274785 8564 sgd_solver.cpp:106] Iteration 15900, lr = 0.001 | |
| I1022 11:18:47.855376 8564 solver.cpp:340] Iteration 16000, Testing net (#0) | |
| I1022 11:18:55.940703 8564 solver.cpp:408] Test net output #0: accuracy = 0.5739 | |
| I1022 11:18:55.940752 8564 solver.cpp:408] Test net output #1: loss = 1.23572 (* 1 = 1.23572 loss) | |
| I1022 11:18:56.042127 8564 solver.cpp:236] Iteration 16000, loss = 0.907968 | |
| I1022 11:18:56.042177 8564 solver.cpp:252] Train net output #0: loss = 0.907968 (* 1 = 0.907968 loss) | |
| I1022 11:18:56.042189 8564 sgd_solver.cpp:106] Iteration 16000, lr = 0.001 | |
| I1022 11:19:12.616708 8564 solver.cpp:236] Iteration 16100, loss = 0.798735 | |
| I1022 11:19:12.616796 8564 solver.cpp:252] Train net output #0: loss = 0.798735 (* 1 = 0.798735 loss) | |
| I1022 11:19:12.616802 8564 sgd_solver.cpp:106] Iteration 16100, lr = 0.001 | |
| I1022 11:19:29.076385 8564 solver.cpp:236] Iteration 16200, loss = 0.975644 | |
| I1022 11:19:29.076413 8564 solver.cpp:252] Train net output #0: loss = 0.975644 (* 1 = 0.975644 loss) | |
| I1022 11:19:29.076418 8564 sgd_solver.cpp:106] Iteration 16200, lr = 0.001 | |
| I1022 11:19:45.614449 8564 solver.cpp:236] Iteration 16300, loss = 0.776063 | |
| I1022 11:19:45.614531 8564 solver.cpp:252] Train net output #0: loss = 0.776063 (* 1 = 0.776063 loss) | |
| I1022 11:19:45.614544 8564 sgd_solver.cpp:106] Iteration 16300, lr = 0.001 | |
| I1022 11:20:01.939510 8564 solver.cpp:236] Iteration 16400, loss = 0.722728 | |
| I1022 11:20:01.939538 8564 solver.cpp:252] Train net output #0: loss = 0.722728 (* 1 = 0.722728 loss) | |
| I1022 11:20:01.939543 8564 sgd_solver.cpp:106] Iteration 16400, lr = 0.001 | |
| I1022 11:20:18.522455 8564 solver.cpp:236] Iteration 16500, loss = 0.902115 | |
| I1022 11:20:18.522598 8564 solver.cpp:252] Train net output #0: loss = 0.902115 (* 1 = 0.902115 loss) | |
| I1022 11:20:18.522627 8564 sgd_solver.cpp:106] Iteration 16500, lr = 0.001 | |
| I1022 11:20:35.039793 8564 solver.cpp:236] Iteration 16600, loss = 0.796508 | |
| I1022 11:20:35.039821 8564 solver.cpp:252] Train net output #0: loss = 0.796508 (* 1 = 0.796508 loss) | |
| I1022 11:20:35.039827 8564 sgd_solver.cpp:106] Iteration 16600, lr = 0.001 | |
| I1022 11:20:51.746162 8564 solver.cpp:236] Iteration 16700, loss = 0.972412 | |
| I1022 11:20:51.746352 8564 solver.cpp:252] Train net output #0: loss = 0.972412 (* 1 = 0.972412 loss) | |
| I1022 11:20:51.746366 8564 sgd_solver.cpp:106] Iteration 16700, lr = 0.001 | |
| I1022 11:21:07.782186 8564 solver.cpp:236] Iteration 16800, loss = 0.772334 | |
| I1022 11:21:07.782222 8564 solver.cpp:252] Train net output #0: loss = 0.772334 (* 1 = 0.772334 loss) | |
| I1022 11:21:07.782228 8564 sgd_solver.cpp:106] Iteration 16800, lr = 0.001 | |
| I1022 11:21:24.191629 8564 solver.cpp:236] Iteration 16900, loss = 0.721072 | |
| I1022 11:21:24.191701 8564 solver.cpp:252] Train net output #0: loss = 0.721072 (* 1 = 0.721072 loss) | |
| I1022 11:21:24.191707 8564 sgd_solver.cpp:106] Iteration 16900, lr = 0.001 | |
| I1022 11:21:40.489735 8564 solver.cpp:340] Iteration 17000, Testing net (#0) | |
| I1022 11:21:48.526864 8564 solver.cpp:408] Test net output #0: accuracy = 0.5768 | |
| I1022 11:21:48.526900 8564 solver.cpp:408] Test net output #1: loss = 1.22559 (* 1 = 1.22559 loss) | |
| I1022 11:21:48.621413 8564 solver.cpp:236] Iteration 17000, loss = 0.895788 | |
| I1022 11:21:48.621464 8564 solver.cpp:252] Train net output #0: loss = 0.895788 (* 1 = 0.895788 loss) | |
| I1022 11:21:48.621474 8564 sgd_solver.cpp:106] Iteration 17000, lr = 0.001 | |
| I1022 11:22:05.027693 8564 solver.cpp:236] Iteration 17100, loss = 0.794307 | |
| I1022 11:22:05.027760 8564 solver.cpp:252] Train net output #0: loss = 0.794307 (* 1 = 0.794307 loss) | |
| I1022 11:22:05.027773 8564 sgd_solver.cpp:106] Iteration 17100, lr = 0.001 | |
| I1022 11:22:21.424247 8564 solver.cpp:236] Iteration 17200, loss = 0.969369 | |
| I1022 11:22:21.424275 8564 solver.cpp:252] Train net output #0: loss = 0.969369 (* 1 = 0.969369 loss) | |
| I1022 11:22:21.424280 8564 sgd_solver.cpp:106] Iteration 17200, lr = 0.001 | |
| I1022 11:22:38.082953 8564 solver.cpp:236] Iteration 17300, loss = 0.768728 | |
| I1022 11:22:38.083216 8564 solver.cpp:252] Train net output #0: loss = 0.768728 (* 1 = 0.768728 loss) | |
| I1022 11:22:38.083228 8564 sgd_solver.cpp:106] Iteration 17300, lr = 0.001 | |
| I1022 11:22:54.353318 8564 solver.cpp:236] Iteration 17400, loss = 0.719676 | |
| I1022 11:22:54.353346 8564 solver.cpp:252] Train net output #0: loss = 0.719676 (* 1 = 0.719676 loss) | |
| I1022 11:22:54.353351 8564 sgd_solver.cpp:106] Iteration 17400, lr = 0.001 | |
| I1022 11:23:10.783258 8564 solver.cpp:236] Iteration 17500, loss = 0.890551 | |
| I1022 11:23:10.783357 8564 solver.cpp:252] Train net output #0: loss = 0.890551 (* 1 = 0.890551 loss) | |
| I1022 11:23:10.783365 8564 sgd_solver.cpp:106] Iteration 17500, lr = 0.001 | |
| I1022 11:23:27.494442 8564 solver.cpp:236] Iteration 17600, loss = 0.791581 | |
| I1022 11:23:27.494472 8564 solver.cpp:252] Train net output #0: loss = 0.791581 (* 1 = 0.791581 loss) | |
| I1022 11:23:27.494477 8564 sgd_solver.cpp:106] Iteration 17600, lr = 0.001 | |
| I1022 11:23:43.656267 8564 solver.cpp:236] Iteration 17700, loss = 0.966551 | |
| I1022 11:23:43.656338 8564 solver.cpp:252] Train net output #0: loss = 0.966551 (* 1 = 0.966551 loss) | |
| I1022 11:23:43.656349 8564 sgd_solver.cpp:106] Iteration 17700, lr = 0.001 | |
| I1022 11:24:00.333573 8564 solver.cpp:236] Iteration 17800, loss = 0.765147 | |
| I1022 11:24:00.333632 8564 solver.cpp:252] Train net output #0: loss = 0.765147 (* 1 = 0.765147 loss) | |
| I1022 11:24:00.333641 8564 sgd_solver.cpp:106] Iteration 17800, lr = 0.001 | |
| I1022 11:24:17.216095 8564 solver.cpp:236] Iteration 17900, loss = 0.718045 | |
| I1022 11:24:17.216167 8564 solver.cpp:252] Train net output #0: loss = 0.718045 (* 1 = 0.718045 loss) | |
| I1022 11:24:17.216176 8564 sgd_solver.cpp:106] Iteration 17900, lr = 0.001 | |
| I1022 11:24:33.808125 8564 solver.cpp:340] Iteration 18000, Testing net (#0) | |
| I1022 11:24:41.843014 8564 solver.cpp:408] Test net output #0: accuracy = 0.5824 | |
| I1022 11:24:41.843049 8564 solver.cpp:408] Test net output #1: loss = 1.20896 (* 1 = 1.20896 loss) | |
| I1022 11:24:41.935667 8564 solver.cpp:236] Iteration 18000, loss = 0.885244 | |
| I1022 11:24:41.935703 8564 solver.cpp:252] Train net output #0: loss = 0.885244 (* 1 = 0.885244 loss) | |
| I1022 11:24:41.935706 8564 sgd_solver.cpp:106] Iteration 18000, lr = 0.001 | |
| I1022 11:24:56.184695 8564 solver.cpp:236] Iteration 18100, loss = 0.788617 | |
| I1022 11:24:56.184777 8564 solver.cpp:252] Train net output #0: loss = 0.788617 (* 1 = 0.788617 loss) | |
| I1022 11:24:56.184783 8564 sgd_solver.cpp:106] Iteration 18100, lr = 0.001 | |
| I1022 11:25:07.470768 8564 solver.cpp:236] Iteration 18200, loss = 0.964112 | |
| I1022 11:25:07.470803 8564 solver.cpp:252] Train net output #0: loss = 0.964112 (* 1 = 0.964112 loss) | |
| I1022 11:25:07.470809 8564 sgd_solver.cpp:106] Iteration 18200, lr = 0.001 | |
| I1022 11:25:18.818291 8564 solver.cpp:236] Iteration 18300, loss = 0.761697 | |
| I1022 11:25:18.818331 8564 solver.cpp:252] Train net output #0: loss = 0.761697 (* 1 = 0.761697 loss) | |
| I1022 11:25:18.818338 8564 sgd_solver.cpp:106] Iteration 18300, lr = 0.001 | |
| I1022 11:25:29.977766 8564 solver.cpp:236] Iteration 18400, loss = 0.71644 | |
| I1022 11:25:29.977860 8564 solver.cpp:252] Train net output #0: loss = 0.71644 (* 1 = 0.71644 loss) | |
| I1022 11:25:29.977866 8564 sgd_solver.cpp:106] Iteration 18400, lr = 0.001 | |
| I1022 11:25:41.235028 8564 solver.cpp:236] Iteration 18500, loss = 0.879611 | |
| I1022 11:25:41.235064 8564 solver.cpp:252] Train net output #0: loss = 0.879611 (* 1 = 0.879611 loss) | |
| I1022 11:25:41.235069 8564 sgd_solver.cpp:106] Iteration 18500, lr = 0.001 | |
| I1022 11:25:52.447937 8564 solver.cpp:236] Iteration 18600, loss = 0.785579 | |
| I1022 11:25:52.447963 8564 solver.cpp:252] Train net output #0: loss = 0.785579 (* 1 = 0.785579 loss) | |
| I1022 11:25:52.447969 8564 sgd_solver.cpp:106] Iteration 18600, lr = 0.001 | |
| I1022 11:26:03.143340 8564 solver.cpp:236] Iteration 18700, loss = 0.96214 | |
| I1022 11:26:03.143430 8564 solver.cpp:252] Train net output #0: loss = 0.96214 (* 1 = 0.96214 loss) | |
| I1022 11:26:03.143436 8564 sgd_solver.cpp:106] Iteration 18700, lr = 0.001 | |
| I1022 11:26:14.272738 8564 solver.cpp:236] Iteration 18800, loss = 0.758737 | |
| I1022 11:26:14.272775 8564 solver.cpp:252] Train net output #0: loss = 0.758737 (* 1 = 0.758737 loss) | |
| I1022 11:26:14.272780 8564 sgd_solver.cpp:106] Iteration 18800, lr = 0.001 | |
| I1022 11:26:25.401067 8564 solver.cpp:236] Iteration 18900, loss = 0.714998 | |
| I1022 11:26:25.401103 8564 solver.cpp:252] Train net output #0: loss = 0.714998 (* 1 = 0.714998 loss) | |
| I1022 11:26:25.401108 8564 sgd_solver.cpp:106] Iteration 18900, lr = 0.001 | |
| I1022 11:26:36.441985 8564 solver.cpp:340] Iteration 19000, Testing net (#0) | |
| I1022 11:26:41.808882 8564 solver.cpp:408] Test net output #0: accuracy = 0.5863 | |
| I1022 11:26:41.808917 8564 solver.cpp:408] Test net output #1: loss = 1.19419 (* 1 = 1.19419 loss) | |
| I1022 11:26:41.871855 8564 solver.cpp:236] Iteration 19000, loss = 0.874707 | |
| I1022 11:26:41.871889 8564 solver.cpp:252] Train net output #0: loss = 0.874707 (* 1 = 0.874707 loss) | |
| I1022 11:26:41.871896 8564 sgd_solver.cpp:106] Iteration 19000, lr = 0.001 | |
| I1022 11:26:53.007905 8564 solver.cpp:236] Iteration 19100, loss = 0.783215 | |
| I1022 11:26:53.007941 8564 solver.cpp:252] Train net output #0: loss = 0.783215 (* 1 = 0.783215 loss) | |
| I1022 11:26:53.007946 8564 sgd_solver.cpp:106] Iteration 19100, lr = 0.001 | |
| I1022 11:27:04.149597 8564 solver.cpp:236] Iteration 19200, loss = 0.95961 | |
| I1022 11:27:04.149632 8564 solver.cpp:252] Train net output #0: loss = 0.95961 (* 1 = 0.95961 loss) | |
| I1022 11:27:04.149637 8564 sgd_solver.cpp:106] Iteration 19200, lr = 0.001 | |
| I1022 11:27:15.280393 8564 solver.cpp:236] Iteration 19300, loss = 0.756417 | |
| I1022 11:27:15.280501 8564 solver.cpp:252] Train net output #0: loss = 0.756417 (* 1 = 0.756417 loss) | |
| I1022 11:27:15.280506 8564 sgd_solver.cpp:106] Iteration 19300, lr = 0.001 | |
| I1022 11:27:26.421196 8564 solver.cpp:236] Iteration 19400, loss = 0.713458 | |
| I1022 11:27:26.421247 8564 solver.cpp:252] Train net output #0: loss = 0.713458 (* 1 = 0.713458 loss) | |
| I1022 11:27:26.421257 8564 sgd_solver.cpp:106] Iteration 19400, lr = 0.001 | |
| I1022 11:27:37.574345 8564 solver.cpp:236] Iteration 19500, loss = 0.87001 | |
| I1022 11:27:37.574383 8564 solver.cpp:252] Train net output #0: loss = 0.87001 (* 1 = 0.87001 loss) | |
| I1022 11:27:37.574388 8564 sgd_solver.cpp:106] Iteration 19500, lr = 0.001 | |
| I1022 11:27:48.366206 8564 solver.cpp:236] Iteration 19600, loss = 0.779889 | |
| I1022 11:27:48.366286 8564 solver.cpp:252] Train net output #0: loss = 0.779889 (* 1 = 0.779889 loss) | |
| I1022 11:27:48.366291 8564 sgd_solver.cpp:106] Iteration 19600, lr = 0.001 | |
| I1022 11:27:59.622241 8564 solver.cpp:236] Iteration 19700, loss = 0.958023 | |
| I1022 11:27:59.622279 8564 solver.cpp:252] Train net output #0: loss = 0.958023 (* 1 = 0.958023 loss) | |
| I1022 11:27:59.622285 8564 sgd_solver.cpp:106] Iteration 19700, lr = 0.001 | |
| I1022 11:28:11.075382 8564 solver.cpp:236] Iteration 19800, loss = 0.753502 | |
| I1022 11:28:11.075420 8564 solver.cpp:252] Train net output #0: loss = 0.753502 (* 1 = 0.753502 loss) | |
| I1022 11:28:11.075426 8564 sgd_solver.cpp:106] Iteration 19800, lr = 0.001 | |
| I1022 11:28:22.427258 8564 solver.cpp:236] Iteration 19900, loss = 0.711883 | |
| I1022 11:28:22.427316 8564 solver.cpp:252] Train net output #0: loss = 0.711883 (* 1 = 0.711883 loss) | |
| I1022 11:28:22.427323 8564 sgd_solver.cpp:106] Iteration 19900, lr = 0.001 | |
| I1022 11:28:33.530458 8564 solver.cpp:340] Iteration 20000, Testing net (#0) | |
| I1022 11:28:38.890738 8564 solver.cpp:408] Test net output #0: accuracy = 0.5903 | |
| I1022 11:28:38.890771 8564 solver.cpp:408] Test net output #1: loss = 1.18231 (* 1 = 1.18231 loss) | |
| I1022 11:28:38.950404 8564 solver.cpp:236] Iteration 20000, loss = 0.865208 | |
| I1022 11:28:38.950429 8564 solver.cpp:252] Train net output #0: loss = 0.865208 (* 1 = 0.865208 loss) | |
| I1022 11:28:38.950435 8564 sgd_solver.cpp:106] Iteration 20000, lr = 0.001 | |
| I1022 11:28:50.120688 8564 solver.cpp:236] Iteration 20100, loss = 0.776528 | |
| I1022 11:28:50.120715 8564 solver.cpp:252] Train net output #0: loss = 0.776528 (* 1 = 0.776528 loss) | |
| I1022 11:28:50.120721 8564 sgd_solver.cpp:106] Iteration 20100, lr = 0.001 | |
| I1022 11:29:01.329221 8564 solver.cpp:236] Iteration 20200, loss = 0.955798 | |
| I1022 11:29:01.329303 8564 solver.cpp:252] Train net output #0: loss = 0.955798 (* 1 = 0.955798 loss) | |
| I1022 11:29:01.329310 8564 sgd_solver.cpp:106] Iteration 20200, lr = 0.001 | |
| I1022 11:29:12.538288 8564 solver.cpp:236] Iteration 20300, loss = 0.751241 | |
| I1022 11:29:12.538323 8564 solver.cpp:252] Train net output #0: loss = 0.751241 (* 1 = 0.751241 loss) | |
| I1022 11:29:12.538331 8564 sgd_solver.cpp:106] Iteration 20300, lr = 0.001 | |
| I1022 11:29:23.880916 8564 solver.cpp:236] Iteration 20400, loss = 0.710354 | |
| I1022 11:29:23.880952 8564 solver.cpp:252] Train net output #0: loss = 0.710354 (* 1 = 0.710354 loss) | |
| I1022 11:29:23.880959 8564 sgd_solver.cpp:106] Iteration 20400, lr = 0.001 | |
| I1022 11:29:34.894707 8564 solver.cpp:236] Iteration 20500, loss = 0.860603 | |
| I1022 11:29:34.894789 8564 solver.cpp:252] Train net output #0: loss = 0.860603 (* 1 = 0.860603 loss) | |
| I1022 11:29:34.894795 8564 sgd_solver.cpp:106] Iteration 20500, lr = 0.001 | |
| I1022 11:29:46.037618 8564 solver.cpp:236] Iteration 20600, loss = 0.773506 | |
| I1022 11:29:46.037654 8564 solver.cpp:252] Train net output #0: loss = 0.773506 (* 1 = 0.773506 loss) | |
| I1022 11:29:46.037660 8564 sgd_solver.cpp:106] Iteration 20600, lr = 0.001 | |
| I1022 11:29:57.307436 8564 solver.cpp:236] Iteration 20700, loss = 0.953711 | |
| I1022 11:29:57.307464 8564 solver.cpp:252] Train net output #0: loss = 0.953711 (* 1 = 0.953711 loss) | |
| I1022 11:29:57.307471 8564 sgd_solver.cpp:106] Iteration 20700, lr = 0.001 | |
| I1022 11:30:08.457167 8564 solver.cpp:236] Iteration 20800, loss = 0.748636 | |
| I1022 11:30:08.457299 8564 solver.cpp:252] Train net output #0: loss = 0.748636 (* 1 = 0.748636 loss) | |
| I1022 11:30:08.457305 8564 sgd_solver.cpp:106] Iteration 20800, lr = 0.001 | |
| I1022 11:30:19.559459 8564 solver.cpp:236] Iteration 20900, loss = 0.708837 | |
| I1022 11:30:19.559486 8564 solver.cpp:252] Train net output #0: loss = 0.708837 (* 1 = 0.708837 loss) | |
| I1022 11:30:19.559491 8564 sgd_solver.cpp:106] Iteration 20900, lr = 0.001 | |
| I1022 11:30:30.583124 8564 solver.cpp:340] Iteration 21000, Testing net (#0) | |
| I1022 11:30:35.926051 8564 solver.cpp:408] Test net output #0: accuracy = 0.5928 | |
| I1022 11:30:35.926100 8564 solver.cpp:408] Test net output #1: loss = 1.17083 (* 1 = 1.17083 loss) | |
| I1022 11:30:35.986152 8564 solver.cpp:236] Iteration 21000, loss = 0.855822 | |
| I1022 11:30:35.986181 8564 solver.cpp:252] Train net output #0: loss = 0.855822 (* 1 = 0.855822 loss) | |
| I1022 11:30:35.986186 8564 sgd_solver.cpp:106] Iteration 21000, lr = 0.001 | |
| I1022 11:30:47.237026 8564 solver.cpp:236] Iteration 21100, loss = 0.770431 | |
| I1022 11:30:47.237118 8564 solver.cpp:252] Train net output #0: loss = 0.770431 (* 1 = 0.770431 loss) | |
| I1022 11:30:47.237123 8564 sgd_solver.cpp:106] Iteration 21100, lr = 0.001 | |
| I1022 11:30:58.322841 8564 solver.cpp:236] Iteration 21200, loss = 0.951486 | |
| I1022 11:30:58.322876 8564 solver.cpp:252] Train net output #0: loss = 0.951486 (* 1 = 0.951486 loss) | |
| I1022 11:30:58.322882 8564 sgd_solver.cpp:106] Iteration 21200, lr = 0.001 | |
| I1022 11:31:09.406991 8564 solver.cpp:236] Iteration 21300, loss = 0.74614 | |
| I1022 11:31:09.407028 8564 solver.cpp:252] Train net output #0: loss = 0.74614 (* 1 = 0.74614 loss) | |
| I1022 11:31:09.407034 8564 sgd_solver.cpp:106] Iteration 21300, lr = 0.001 | |
| I1022 11:31:20.486799 8564 solver.cpp:236] Iteration 21400, loss = 0.707738 | |
| I1022 11:31:20.486889 8564 solver.cpp:252] Train net output #0: loss = 0.707738 (* 1 = 0.707738 loss) | |
| I1022 11:31:20.486894 8564 sgd_solver.cpp:106] Iteration 21400, lr = 0.001 | |
| I1022 11:31:31.223381 8564 solver.cpp:236] Iteration 21500, loss = 0.851042 | |
| I1022 11:31:31.223417 8564 solver.cpp:252] Train net output #0: loss = 0.851042 (* 1 = 0.851042 loss) | |
| I1022 11:31:31.223424 8564 sgd_solver.cpp:106] Iteration 21500, lr = 0.001 | |
| I1022 11:31:42.617928 8564 solver.cpp:236] Iteration 21600, loss = 0.767019 | |
| I1022 11:31:42.617962 8564 solver.cpp:252] Train net output #0: loss = 0.767019 (* 1 = 0.767019 loss) | |
| I1022 11:31:42.617967 8564 sgd_solver.cpp:106] Iteration 21600, lr = 0.001 | |
| I1022 11:31:53.699035 8564 solver.cpp:236] Iteration 21700, loss = 0.949351 | |
| I1022 11:31:53.699142 8564 solver.cpp:252] Train net output #0: loss = 0.949351 (* 1 = 0.949351 loss) | |
| I1022 11:31:53.699149 8564 sgd_solver.cpp:106] Iteration 21700, lr = 0.001 | |
| I1022 11:32:04.918752 8564 solver.cpp:236] Iteration 21800, loss = 0.743844 | |
| I1022 11:32:04.918788 8564 solver.cpp:252] Train net output #0: loss = 0.743844 (* 1 = 0.743844 loss) | |
| I1022 11:32:04.918794 8564 sgd_solver.cpp:106] Iteration 21800, lr = 0.001 | |
| I1022 11:32:16.003837 8564 solver.cpp:236] Iteration 21900, loss = 0.706585 | |
| I1022 11:32:16.003865 8564 solver.cpp:252] Train net output #0: loss = 0.706585 (* 1 = 0.706585 loss) | |
| I1022 11:32:16.003871 8564 sgd_solver.cpp:106] Iteration 21900, lr = 0.001 | |
| I1022 11:32:27.019603 8564 solver.cpp:340] Iteration 22000, Testing net (#0) | |
| I1022 11:32:32.357235 8564 solver.cpp:408] Test net output #0: accuracy = 0.5964 | |
| I1022 11:32:32.357269 8564 solver.cpp:408] Test net output #1: loss = 1.15447 (* 1 = 1.15447 loss) | |
| I1022 11:32:32.418025 8564 solver.cpp:236] Iteration 22000, loss = 0.846732 | |
| I1022 11:32:32.418059 8564 solver.cpp:252] Train net output #0: loss = 0.846732 (* 1 = 0.846732 loss) | |
| I1022 11:32:32.418066 8564 sgd_solver.cpp:106] Iteration 22000, lr = 0.001 | |
| I1022 11:32:43.519179 8564 solver.cpp:236] Iteration 22100, loss = 0.764138 | |
| I1022 11:32:43.519207 8564 solver.cpp:252] Train net output #0: loss = 0.764138 (* 1 = 0.764138 loss) | |
| I1022 11:32:43.519213 8564 sgd_solver.cpp:106] Iteration 22100, lr = 0.001 | |
| I1022 11:32:54.859508 8564 solver.cpp:236] Iteration 22200, loss = 0.947374 | |
| I1022 11:32:54.859544 8564 solver.cpp:252] Train net output #0: loss = 0.947374 (* 1 = 0.947374 loss) | |
| I1022 11:32:54.859549 8564 sgd_solver.cpp:106] Iteration 22200, lr = 0.001 | |
| I1022 11:33:05.993993 8564 solver.cpp:236] Iteration 22300, loss = 0.741793 | |
| I1022 11:33:05.994052 8564 solver.cpp:252] Train net output #0: loss = 0.741793 (* 1 = 0.741793 loss) | |
| I1022 11:33:05.994057 8564 sgd_solver.cpp:106] Iteration 22300, lr = 0.001 | |
| I1022 11:33:16.766129 8564 solver.cpp:236] Iteration 22400, loss = 0.705549 | |
| I1022 11:33:16.766155 8564 solver.cpp:252] Train net output #0: loss = 0.705549 (* 1 = 0.705549 loss) | |
| I1022 11:33:16.766161 8564 sgd_solver.cpp:106] Iteration 22400, lr = 0.001 | |
| I1022 11:33:27.659243 8564 solver.cpp:236] Iteration 22500, loss = 0.842922 | |
| I1022 11:33:27.659278 8564 solver.cpp:252] Train net output #0: loss = 0.842922 (* 1 = 0.842922 loss) | |
| I1022 11:33:27.659284 8564 sgd_solver.cpp:106] Iteration 22500, lr = 0.001 | |
| I1022 11:33:38.746275 8564 solver.cpp:236] Iteration 22600, loss = 0.761138 | |
| I1022 11:33:38.746358 8564 solver.cpp:252] Train net output #0: loss = 0.761138 (* 1 = 0.761138 loss) | |
| I1022 11:33:38.746364 8564 sgd_solver.cpp:106] Iteration 22600, lr = 0.001 | |
| I1022 11:33:49.833434 8564 solver.cpp:236] Iteration 22700, loss = 0.945344 | |
| I1022 11:33:49.833472 8564 solver.cpp:252] Train net output #0: loss = 0.945344 (* 1 = 0.945344 loss) | |
| I1022 11:33:49.833477 8564 sgd_solver.cpp:106] Iteration 22700, lr = 0.001 | |
| I1022 11:34:00.923348 8564 solver.cpp:236] Iteration 22800, loss = 0.739821 | |
| I1022 11:34:00.923383 8564 solver.cpp:252] Train net output #0: loss = 0.739821 (* 1 = 0.739821 loss) | |
| I1022 11:34:00.923388 8564 sgd_solver.cpp:106] Iteration 22800, lr = 0.001 | |
| I1022 11:34:12.000700 8564 solver.cpp:236] Iteration 22900, loss = 0.704914 | |
| I1022 11:34:12.000787 8564 solver.cpp:252] Train net output #0: loss = 0.704914 (* 1 = 0.704914 loss) | |
| I1022 11:34:12.000793 8564 sgd_solver.cpp:106] Iteration 22900, lr = 0.001 | |
| I1022 11:34:23.007793 8564 solver.cpp:340] Iteration 23000, Testing net (#0) | |
| I1022 11:34:28.350219 8564 solver.cpp:408] Test net output #0: accuracy = 0.5997 | |
| I1022 11:34:28.350253 8564 solver.cpp:408] Test net output #1: loss = 1.14273 (* 1 = 1.14273 loss) | |
| I1022 11:34:28.411208 8564 solver.cpp:236] Iteration 23000, loss = 0.838659 | |
| I1022 11:34:28.411240 8564 solver.cpp:252] Train net output #0: loss = 0.838659 (* 1 = 0.838659 loss) | |
| I1022 11:34:28.411245 8564 sgd_solver.cpp:106] Iteration 23000, lr = 0.001 | |
| I1022 11:34:39.491907 8564 solver.cpp:236] Iteration 23100, loss = 0.757982 | |
| I1022 11:34:39.491943 8564 solver.cpp:252] Train net output #0: loss = 0.757982 (* 1 = 0.757982 loss) | |
| I1022 11:34:39.491950 8564 sgd_solver.cpp:106] Iteration 23100, lr = 0.001 | |
| I1022 11:34:50.578348 8564 solver.cpp:236] Iteration 23200, loss = 0.943172 | |
| I1022 11:34:50.578438 8564 solver.cpp:252] Train net output #0: loss = 0.943172 (* 1 = 0.943172 loss) | |
| I1022 11:34:50.578444 8564 sgd_solver.cpp:106] Iteration 23200, lr = 0.001 | |
| I1022 11:35:01.657322 8564 solver.cpp:236] Iteration 23300, loss = 0.737682 | |
| I1022 11:35:01.657358 8564 solver.cpp:252] Train net output #0: loss = 0.737682 (* 1 = 0.737682 loss) | |
| I1022 11:35:01.657362 8564 sgd_solver.cpp:106] Iteration 23300, lr = 0.001 | |
| I1022 11:35:12.263994 8564 solver.cpp:236] Iteration 23400, loss = 0.704163 | |
| I1022 11:35:12.264029 8564 solver.cpp:252] Train net output #0: loss = 0.704163 (* 1 = 0.704163 loss) | |
| I1022 11:35:12.264035 8564 sgd_solver.cpp:106] Iteration 23400, lr = 0.001 | |
| I1022 11:35:23.451151 8564 solver.cpp:236] Iteration 23500, loss = 0.834991 | |
| I1022 11:35:23.451241 8564 solver.cpp:252] Train net output #0: loss = 0.834991 (* 1 = 0.834991 loss) | |
| I1022 11:35:23.451246 8564 sgd_solver.cpp:106] Iteration 23500, lr = 0.001 | |
| I1022 11:35:34.569589 8564 solver.cpp:236] Iteration 23600, loss = 0.755007 | |
| I1022 11:35:34.569615 8564 solver.cpp:252] Train net output #0: loss = 0.755007 (* 1 = 0.755007 loss) | |
| I1022 11:35:34.569622 8564 sgd_solver.cpp:106] Iteration 23600, lr = 0.001 | |
| I1022 11:35:45.756487 8564 solver.cpp:236] Iteration 23700, loss = 0.940553 | |
| I1022 11:35:45.756525 8564 solver.cpp:252] Train net output #0: loss = 0.940553 (* 1 = 0.940553 loss) | |
| I1022 11:35:45.756531 8564 sgd_solver.cpp:106] Iteration 23700, lr = 0.001 | |
| I1022 11:35:56.835918 8564 solver.cpp:236] Iteration 23800, loss = 0.736149 | |
| I1022 11:35:56.836004 8564 solver.cpp:252] Train net output #0: loss = 0.736149 (* 1 = 0.736149 loss) | |
| I1022 11:35:56.836009 8564 sgd_solver.cpp:106] Iteration 23800, lr = 0.001 | |
| I1022 11:36:07.916920 8564 solver.cpp:236] Iteration 23900, loss = 0.703207 | |
| I1022 11:36:07.916954 8564 solver.cpp:252] Train net output #0: loss = 0.703207 (* 1 = 0.703207 loss) | |
| I1022 11:36:07.916959 8564 sgd_solver.cpp:106] Iteration 23900, lr = 0.001 | |
| I1022 11:36:18.925489 8564 solver.cpp:340] Iteration 24000, Testing net (#0) | |
| I1022 11:36:24.265583 8564 solver.cpp:408] Test net output #0: accuracy = 0.6033 | |
| I1022 11:36:24.265617 8564 solver.cpp:408] Test net output #1: loss = 1.13033 (* 1 = 1.13033 loss) | |
| I1022 11:36:24.326654 8564 solver.cpp:236] Iteration 24000, loss = 0.831267 | |
| I1022 11:36:24.326686 8564 solver.cpp:252] Train net output #0: loss = 0.831267 (* 1 = 0.831267 loss) | |
| I1022 11:36:24.326691 8564 sgd_solver.cpp:106] Iteration 24000, lr = 0.001 | |
| I1022 11:36:35.457151 8564 solver.cpp:236] Iteration 24100, loss = 0.752061 | |
| I1022 11:36:35.457238 8564 solver.cpp:252] Train net output #0: loss = 0.752061 (* 1 = 0.752061 loss) | |
| I1022 11:36:35.457244 8564 sgd_solver.cpp:106] Iteration 24100, lr = 0.001 | |
| I1022 11:36:46.583427 8564 solver.cpp:236] Iteration 24200, loss = 0.938287 | |
| I1022 11:36:46.583462 8564 solver.cpp:252] Train net output #0: loss = 0.938287 (* 1 = 0.938287 loss) | |
| I1022 11:36:46.583468 8564 sgd_solver.cpp:106] Iteration 24200, lr = 0.001 | |
| I1022 11:36:57.236263 8564 solver.cpp:236] Iteration 24300, loss = 0.734244 | |
| I1022 11:36:57.236289 8564 solver.cpp:252] Train net output #0: loss = 0.734244 (* 1 = 0.734244 loss) | |
| I1022 11:36:57.236295 8564 sgd_solver.cpp:106] Iteration 24300, lr = 0.001 | |
| I1022 11:37:08.289324 8564 solver.cpp:236] Iteration 24400, loss = 0.70257 | |
| I1022 11:37:08.289400 8564 solver.cpp:252] Train net output #0: loss = 0.70257 (* 1 = 0.70257 loss) | |
| I1022 11:37:08.289407 8564 sgd_solver.cpp:106] Iteration 24400, lr = 0.001 | |
| I1022 11:37:19.383482 8564 solver.cpp:236] Iteration 24500, loss = 0.827955 | |
| I1022 11:37:19.383508 8564 solver.cpp:252] Train net output #0: loss = 0.827955 (* 1 = 0.827955 loss) | |
| I1022 11:37:19.383514 8564 sgd_solver.cpp:106] Iteration 24500, lr = 0.001 | |
| I1022 11:37:30.461596 8564 solver.cpp:236] Iteration 24600, loss = 0.748821 | |
| I1022 11:37:30.461632 8564 solver.cpp:252] Train net output #0: loss = 0.748821 (* 1 = 0.748821 loss) | |
| I1022 11:37:30.461637 8564 sgd_solver.cpp:106] Iteration 24600, lr = 0.001 | |
| I1022 11:37:41.538147 8564 solver.cpp:236] Iteration 24700, loss = 0.93621 | |
| I1022 11:37:41.538231 8564 solver.cpp:252] Train net output #0: loss = 0.93621 (* 1 = 0.93621 loss) | |
| I1022 11:37:41.538238 8564 sgd_solver.cpp:106] Iteration 24700, lr = 0.001 | |
| I1022 11:37:52.664242 8564 solver.cpp:236] Iteration 24800, loss = 0.732527 | |
| I1022 11:37:52.664278 8564 solver.cpp:252] Train net output #0: loss = 0.732527 (* 1 = 0.732527 loss) | |
| I1022 11:37:52.664284 8564 sgd_solver.cpp:106] Iteration 24800, lr = 0.001 | |
| I1022 11:38:03.848577 8564 solver.cpp:236] Iteration 24900, loss = 0.701903 | |
| I1022 11:38:03.848613 8564 solver.cpp:252] Train net output #0: loss = 0.701903 (* 1 = 0.701903 loss) | |
| I1022 11:38:03.848618 8564 sgd_solver.cpp:106] Iteration 24900, lr = 0.001 | |
| I1022 11:38:14.962733 8564 solver.cpp:340] Iteration 25000, Testing net (#0) | |
| I1022 11:38:20.346894 8564 solver.cpp:408] Test net output #0: accuracy = 0.6041 | |
| I1022 11:38:20.346930 8564 solver.cpp:408] Test net output #1: loss = 1.12697 (* 1 = 1.12697 loss) | |
| I1022 11:38:20.407196 8564 solver.cpp:236] Iteration 25000, loss = 0.825223 | |
| I1022 11:38:20.407218 8564 solver.cpp:252] Train net output #0: loss = 0.825223 (* 1 = 0.825223 loss) | |
| I1022 11:38:20.407223 8564 sgd_solver.cpp:106] Iteration 25000, lr = 0.001 | |
| I1022 11:38:31.484741 8564 solver.cpp:236] Iteration 25100, loss = 0.746047 | |
| I1022 11:38:31.484776 8564 solver.cpp:252] Train net output #0: loss = 0.746047 (* 1 = 0.746047 loss) | |
| I1022 11:38:31.484782 8564 sgd_solver.cpp:106] Iteration 25100, lr = 0.001 | |
| I1022 11:38:42.700664 8564 solver.cpp:236] Iteration 25200, loss = 0.933658 | |
| I1022 11:38:42.700695 8564 solver.cpp:252] Train net output #0: loss = 0.933658 (* 1 = 0.933658 loss) | |
| I1022 11:38:42.700700 8564 sgd_solver.cpp:106] Iteration 25200, lr = 0.001 | |
| I1022 11:38:53.403936 8564 solver.cpp:236] Iteration 25300, loss = 0.730973 | |
| I1022 11:38:53.404041 8564 solver.cpp:252] Train net output #0: loss = 0.730973 (* 1 = 0.730973 loss) | |
| I1022 11:38:53.404047 8564 sgd_solver.cpp:106] Iteration 25300, lr = 0.001 | |
| I1022 11:39:04.652273 8564 solver.cpp:236] Iteration 25400, loss = 0.701237 | |
| I1022 11:39:04.652310 8564 solver.cpp:252] Train net output #0: loss = 0.701237 (* 1 = 0.701237 loss) | |
| I1022 11:39:04.652315 8564 sgd_solver.cpp:106] Iteration 25400, lr = 0.001 | |
| I1022 11:39:15.846827 8564 solver.cpp:236] Iteration 25500, loss = 0.822009 | |
| I1022 11:39:15.846861 8564 solver.cpp:252] Train net output #0: loss = 0.822009 (* 1 = 0.822009 loss) | |
| I1022 11:39:15.846868 8564 sgd_solver.cpp:106] Iteration 25500, lr = 0.001 | |
| I1022 11:39:26.928272 8564 solver.cpp:236] Iteration 25600, loss = 0.743156 | |
| I1022 11:39:26.928349 8564 solver.cpp:252] Train net output #0: loss = 0.743156 (* 1 = 0.743156 loss) | |
| I1022 11:39:26.928356 8564 sgd_solver.cpp:106] Iteration 25600, lr = 0.001 | |
| I1022 11:39:38.012732 8564 solver.cpp:236] Iteration 25700, loss = 0.931252 | |
| I1022 11:39:38.012768 8564 solver.cpp:252] Train net output #0: loss = 0.931252 (* 1 = 0.931252 loss) | |
| I1022 11:39:38.012773 8564 sgd_solver.cpp:106] Iteration 25700, lr = 0.001 | |
| I1022 11:39:49.334939 8564 solver.cpp:236] Iteration 25800, loss = 0.729736 | |
| I1022 11:39:49.334977 8564 solver.cpp:252] Train net output #0: loss = 0.729736 (* 1 = 0.729736 loss) | |
| I1022 11:39:49.334983 8564 sgd_solver.cpp:106] Iteration 25800, lr = 0.001 | |
| I1022 11:40:00.471437 8564 solver.cpp:236] Iteration 25900, loss = 0.700592 | |
| I1022 11:40:00.471545 8564 solver.cpp:252] Train net output #0: loss = 0.700592 (* 1 = 0.700592 loss) | |
| I1022 11:40:00.471560 8564 sgd_solver.cpp:106] Iteration 25900, lr = 0.001 | |
| I1022 11:40:11.500015 8564 solver.cpp:340] Iteration 26000, Testing net (#0) | |
| I1022 11:40:16.842622 8564 solver.cpp:408] Test net output #0: accuracy = 0.6068 | |
| I1022 11:40:16.842646 8564 solver.cpp:408] Test net output #1: loss = 1.12279 (* 1 = 1.12279 loss) | |
| I1022 11:40:16.903074 8564 solver.cpp:236] Iteration 26000, loss = 0.819177 | |
| I1022 11:40:16.903106 8564 solver.cpp:252] Train net output #0: loss = 0.819177 (* 1 = 0.819177 loss) | |
| I1022 11:40:16.903111 8564 sgd_solver.cpp:106] Iteration 26000, lr = 0.001 | |
| I1022 11:40:27.986927 8564 solver.cpp:236] Iteration 26100, loss = 0.740474 | |
| I1022 11:40:27.986963 8564 solver.cpp:252] Train net output #0: loss = 0.740474 (* 1 = 0.740474 loss) | |
| I1022 11:40:27.986968 8564 sgd_solver.cpp:106] Iteration 26100, lr = 0.001 | |
| I1022 11:40:38.552723 8564 solver.cpp:236] Iteration 26200, loss = 0.928696 | |
| I1022 11:40:38.552841 8564 solver.cpp:252] Train net output #0: loss = 0.928696 (* 1 = 0.928696 loss) | |
| I1022 11:40:38.552846 8564 sgd_solver.cpp:106] Iteration 26200, lr = 0.001 | |
| I1022 11:40:49.637251 8564 solver.cpp:236] Iteration 26300, loss = 0.728425 | |
| I1022 11:40:49.637286 8564 solver.cpp:252] Train net output #0: loss = 0.728425 (* 1 = 0.728425 loss) | |
| I1022 11:40:49.637292 8564 sgd_solver.cpp:106] Iteration 26300, lr = 0.001 | |
| I1022 11:41:00.718488 8564 solver.cpp:236] Iteration 26400, loss = 0.700227 | |
| I1022 11:41:00.718523 8564 solver.cpp:252] Train net output #0: loss = 0.700227 (* 1 = 0.700227 loss) | |
| I1022 11:41:00.718529 8564 sgd_solver.cpp:106] Iteration 26400, lr = 0.001 | |
| I1022 11:41:11.802881 8564 solver.cpp:236] Iteration 26500, loss = 0.816485 | |
| I1022 11:41:11.802988 8564 solver.cpp:252] Train net output #0: loss = 0.816485 (* 1 = 0.816485 loss) | |
| I1022 11:41:11.803002 8564 sgd_solver.cpp:106] Iteration 26500, lr = 0.001 | |
| I1022 11:41:22.868739 8564 solver.cpp:236] Iteration 26600, loss = 0.738439 | |
| I1022 11:41:22.868774 8564 solver.cpp:252] Train net output #0: loss = 0.738439 (* 1 = 0.738439 loss) | |
| I1022 11:41:22.868780 8564 sgd_solver.cpp:106] Iteration 26600, lr = 0.001 | |
| I1022 11:41:33.944108 8564 solver.cpp:236] Iteration 26700, loss = 0.926261 | |
| I1022 11:41:33.944144 8564 solver.cpp:252] Train net output #0: loss = 0.926261 (* 1 = 0.926261 loss) | |
| I1022 11:41:33.944150 8564 sgd_solver.cpp:106] Iteration 26700, lr = 0.001 | |
| I1022 11:41:45.022961 8564 solver.cpp:236] Iteration 26800, loss = 0.726731 | |
| I1022 11:41:45.023020 8564 solver.cpp:252] Train net output #0: loss = 0.726731 (* 1 = 0.726731 loss) | |
| I1022 11:41:45.023026 8564 sgd_solver.cpp:106] Iteration 26800, lr = 0.001 | |
| I1022 11:41:56.095005 8564 solver.cpp:236] Iteration 26900, loss = 0.699937 | |
| I1022 11:41:56.095031 8564 solver.cpp:252] Train net output #0: loss = 0.699937 (* 1 = 0.699937 loss) | |
| I1022 11:41:56.095037 8564 sgd_solver.cpp:106] Iteration 26900, lr = 0.001 | |
| I1022 11:42:07.107566 8564 solver.cpp:340] Iteration 27000, Testing net (#0) | |
| I1022 11:42:12.445190 8564 solver.cpp:408] Test net output #0: accuracy = 0.6064 | |
| I1022 11:42:12.445225 8564 solver.cpp:408] Test net output #1: loss = 1.1241 (* 1 = 1.1241 loss) | |
| I1022 11:42:12.509949 8564 solver.cpp:236] Iteration 27000, loss = 0.813889 | |
| I1022 11:42:12.509969 8564 solver.cpp:252] Train net output #0: loss = 0.813889 (* 1 = 0.813889 loss) | |
| I1022 11:42:12.509975 8564 sgd_solver.cpp:106] Iteration 27000, lr = 0.001 | |
| I1022 11:42:23.462714 8564 solver.cpp:236] Iteration 27100, loss = 0.736272 | |
| I1022 11:42:23.462770 8564 solver.cpp:252] Train net output #0: loss = 0.736272 (* 1 = 0.736272 loss) | |
| I1022 11:42:23.462776 8564 sgd_solver.cpp:106] Iteration 27100, lr = 0.001 | |
| I1022 11:42:34.140887 8564 solver.cpp:236] Iteration 27200, loss = 0.923642 | |
| I1022 11:42:34.140914 8564 solver.cpp:252] Train net output #0: loss = 0.923642 (* 1 = 0.923642 loss) | |
| I1022 11:42:34.140919 8564 sgd_solver.cpp:106] Iteration 27200, lr = 0.001 | |
| I1022 11:42:45.218492 8564 solver.cpp:236] Iteration 27300, loss = 0.725376 | |
| I1022 11:42:45.218525 8564 solver.cpp:252] Train net output #0: loss = 0.725376 (* 1 = 0.725376 loss) | |
| I1022 11:42:45.218530 8564 sgd_solver.cpp:106] Iteration 27300, lr = 0.001 | |
| I1022 11:42:56.288832 8564 solver.cpp:236] Iteration 27400, loss = 0.699479 | |
| I1022 11:42:56.288938 8564 solver.cpp:252] Train net output #0: loss = 0.699479 (* 1 = 0.699479 loss) | |
| I1022 11:42:56.288952 8564 sgd_solver.cpp:106] Iteration 27400, lr = 0.001 | |
| I1022 11:43:07.361426 8564 solver.cpp:236] Iteration 27500, loss = 0.811632 | |
| I1022 11:43:07.361460 8564 solver.cpp:252] Train net output #0: loss = 0.811632 (* 1 = 0.811632 loss) | |
| I1022 11:43:07.361464 8564 sgd_solver.cpp:106] Iteration 27500, lr = 0.001 | |
| I1022 11:43:18.437834 8564 solver.cpp:236] Iteration 27600, loss = 0.733876 | |
| I1022 11:43:18.437870 8564 solver.cpp:252] Train net output #0: loss = 0.733876 (* 1 = 0.733876 loss) | |
| I1022 11:43:18.437875 8564 sgd_solver.cpp:106] Iteration 27600, lr = 0.001 | |
| I1022 11:43:29.509420 8564 solver.cpp:236] Iteration 27700, loss = 0.921256 | |
| I1022 11:43:29.509500 8564 solver.cpp:252] Train net output #0: loss = 0.921256 (* 1 = 0.921256 loss) | |
| I1022 11:43:29.509506 8564 sgd_solver.cpp:106] Iteration 27700, lr = 0.001 | |
| I1022 11:43:40.585213 8564 solver.cpp:236] Iteration 27800, loss = 0.724267 | |
| I1022 11:43:40.585239 8564 solver.cpp:252] Train net output #0: loss = 0.724267 (* 1 = 0.724267 loss) | |
| I1022 11:43:40.585245 8564 sgd_solver.cpp:106] Iteration 27800, lr = 0.001 | |
| I1022 11:43:51.663720 8564 solver.cpp:236] Iteration 27900, loss = 0.699296 | |
| I1022 11:43:51.663745 8564 solver.cpp:252] Train net output #0: loss = 0.699296 (* 1 = 0.699296 loss) | |
| I1022 11:43:51.663750 8564 sgd_solver.cpp:106] Iteration 27900, lr = 0.001 | |
| I1022 11:44:02.674868 8564 solver.cpp:340] Iteration 28000, Testing net (#0) | |
| I1022 11:44:08.014457 8564 solver.cpp:408] Test net output #0: accuracy = 0.607 | |
| I1022 11:44:08.014492 8564 solver.cpp:408] Test net output #1: loss = 1.12152 (* 1 = 1.12152 loss) | |
| I1022 11:44:08.077186 8564 solver.cpp:236] Iteration 28000, loss = 0.809138 | |
| I1022 11:44:08.077211 8564 solver.cpp:252] Train net output #0: loss = 0.809138 (* 1 = 0.809138 loss) | |
| I1022 11:44:08.077217 8564 sgd_solver.cpp:106] Iteration 28000, lr = 0.001 | |
| I1022 11:44:18.640517 8564 solver.cpp:236] Iteration 28100, loss = 0.731815 | |
| I1022 11:44:18.640543 8564 solver.cpp:252] Train net output #0: loss = 0.731815 (* 1 = 0.731815 loss) | |
| I1022 11:44:18.640549 8564 sgd_solver.cpp:106] Iteration 28100, lr = 0.001 | |
| I1022 11:44:29.718418 8564 solver.cpp:236] Iteration 28200, loss = 0.919272 | |
| I1022 11:44:29.718454 8564 solver.cpp:252] Train net output #0: loss = 0.919272 (* 1 = 0.919272 loss) | |
| I1022 11:44:29.718461 8564 sgd_solver.cpp:106] Iteration 28200, lr = 0.001 | |
| I1022 11:44:40.796396 8564 solver.cpp:236] Iteration 28300, loss = 0.723178 | |
| I1022 11:44:40.796490 8564 solver.cpp:252] Train net output #0: loss = 0.723178 (* 1 = 0.723178 loss) | |
| I1022 11:44:40.796496 8564 sgd_solver.cpp:106] Iteration 28300, lr = 0.001 | |
| I1022 11:44:51.901971 8564 solver.cpp:236] Iteration 28400, loss = 0.698887 | |
| I1022 11:44:51.902007 8564 solver.cpp:252] Train net output #0: loss = 0.698887 (* 1 = 0.698887 loss) | |
| I1022 11:44:51.902012 8564 sgd_solver.cpp:106] Iteration 28400, lr = 0.001 | |
| I1022 11:45:03.053115 8564 solver.cpp:236] Iteration 28500, loss = 0.807208 | |
| I1022 11:45:03.053153 8564 solver.cpp:252] Train net output #0: loss = 0.807208 (* 1 = 0.807208 loss) | |
| I1022 11:45:03.053158 8564 sgd_solver.cpp:106] Iteration 28500, lr = 0.001 | |
| I1022 11:45:14.136247 8564 solver.cpp:236] Iteration 28600, loss = 0.729844 | |
| I1022 11:45:14.136340 8564 solver.cpp:252] Train net output #0: loss = 0.729844 (* 1 = 0.729844 loss) | |
| I1022 11:45:14.136355 8564 sgd_solver.cpp:106] Iteration 28600, lr = 0.001 | |
| I1022 11:45:25.517149 8564 solver.cpp:236] Iteration 28700, loss = 0.917208 | |
| I1022 11:45:25.517184 8564 solver.cpp:252] Train net output #0: loss = 0.917208 (* 1 = 0.917208 loss) | |
| I1022 11:45:25.517189 8564 sgd_solver.cpp:106] Iteration 28700, lr = 0.001 | |
| I1022 11:45:36.650207 8564 solver.cpp:236] Iteration 28800, loss = 0.722189 | |
| I1022 11:45:36.650243 8564 solver.cpp:252] Train net output #0: loss = 0.722189 (* 1 = 0.722189 loss) | |
| I1022 11:45:36.650249 8564 sgd_solver.cpp:106] Iteration 28800, lr = 0.001 | |
| I1022 11:45:47.943670 8564 solver.cpp:236] Iteration 28900, loss = 0.698848 | |
| I1022 11:45:47.943789 8564 solver.cpp:252] Train net output #0: loss = 0.698848 (* 1 = 0.698848 loss) | |
| I1022 11:45:47.943802 8564 sgd_solver.cpp:106] Iteration 28900, lr = 0.001 | |
| I1022 11:45:59.067703 8564 solver.cpp:340] Iteration 29000, Testing net (#0) | |
| I1022 11:46:04.197167 8564 solver.cpp:408] Test net output #0: accuracy = 0.6066 | |
| I1022 11:46:04.197201 8564 solver.cpp:408] Test net output #1: loss = 1.12055 (* 1 = 1.12055 loss) | |
| I1022 11:46:04.253844 8564 solver.cpp:236] Iteration 29000, loss = 0.805276 | |
| I1022 11:46:04.253864 8564 solver.cpp:252] Train net output #0: loss = 0.805276 (* 1 = 0.805276 loss) | |
| I1022 11:46:04.253870 8564 sgd_solver.cpp:106] Iteration 29000, lr = 0.001 | |
| I1022 11:46:15.100793 8564 solver.cpp:236] Iteration 29100, loss = 0.727829 | |
| I1022 11:46:15.100821 8564 solver.cpp:252] Train net output #0: loss = 0.727829 (* 1 = 0.727829 loss) | |
| I1022 11:46:15.100826 8564 sgd_solver.cpp:106] Iteration 29100, lr = 0.001 | |
| I1022 11:46:26.232477 8564 solver.cpp:236] Iteration 29200, loss = 0.914706 | |
| I1022 11:46:26.232568 8564 solver.cpp:252] Train net output #0: loss = 0.914706 (* 1 = 0.914706 loss) | |
| I1022 11:46:26.232573 8564 sgd_solver.cpp:106] Iteration 29200, lr = 0.001 | |
| I1022 11:46:37.359396 8564 solver.cpp:236] Iteration 29300, loss = 0.721268 | |
| I1022 11:46:37.359431 8564 solver.cpp:252] Train net output #0: loss = 0.721268 (* 1 = 0.721268 loss) | |
| I1022 11:46:37.359437 8564 sgd_solver.cpp:106] Iteration 29300, lr = 0.001 | |
| I1022 11:46:48.493046 8564 solver.cpp:236] Iteration 29400, loss = 0.698388 | |
| I1022 11:46:48.493074 8564 solver.cpp:252] Train net output #0: loss = 0.698388 (* 1 = 0.698388 loss) | |
| I1022 11:46:48.493080 8564 sgd_solver.cpp:106] Iteration 29400, lr = 0.001 | |
| I1022 11:46:59.621747 8564 solver.cpp:236] Iteration 29500, loss = 0.803337 | |
| I1022 11:46:59.621840 8564 solver.cpp:252] Train net output #0: loss = 0.803337 (* 1 = 0.803337 loss) | |
| I1022 11:46:59.621846 8564 sgd_solver.cpp:106] Iteration 29500, lr = 0.001 | |
| I1022 11:47:10.850951 8564 solver.cpp:236] Iteration 29600, loss = 0.725892 | |
| I1022 11:47:10.850978 8564 solver.cpp:252] Train net output #0: loss = 0.725892 (* 1 = 0.725892 loss) | |
| I1022 11:47:10.850985 8564 sgd_solver.cpp:106] Iteration 29600, lr = 0.001 | |
| I1022 11:47:21.993129 8564 solver.cpp:236] Iteration 29700, loss = 0.912513 | |
| I1022 11:47:21.993155 8564 solver.cpp:252] Train net output #0: loss = 0.912513 (* 1 = 0.912513 loss) | |
| I1022 11:47:21.993161 8564 sgd_solver.cpp:106] Iteration 29700, lr = 0.001 | |
| I1022 11:47:33.122323 8564 solver.cpp:236] Iteration 29800, loss = 0.72026 | |
| I1022 11:47:33.122427 8564 solver.cpp:252] Train net output #0: loss = 0.72026 (* 1 = 0.72026 loss) | |
| I1022 11:47:33.122433 8564 sgd_solver.cpp:106] Iteration 29800, lr = 0.001 | |
| I1022 11:47:44.294265 8564 solver.cpp:236] Iteration 29900, loss = 0.698303 | |
| I1022 11:47:44.294301 8564 solver.cpp:252] Train net output #0: loss = 0.698303 (* 1 = 0.698303 loss) | |
| I1022 11:47:44.294306 8564 sgd_solver.cpp:106] Iteration 29900, lr = 0.001 | |
| I1022 11:47:55.107378 8564 solver.cpp:340] Iteration 30000, Testing net (#0) | |
| I1022 11:48:00.266275 8564 solver.cpp:408] Test net output #0: accuracy = 0.6067 | |
| I1022 11:48:00.266300 8564 solver.cpp:408] Test net output #1: loss = 1.11859 (* 1 = 1.11859 loss) | |
| I1022 11:48:00.326344 8564 solver.cpp:236] Iteration 30000, loss = 0.801821 | |
| I1022 11:48:00.326373 8564 solver.cpp:252] Train net output #0: loss = 0.801821 (* 1 = 0.801821 loss) | |
| I1022 11:48:00.326377 8564 sgd_solver.cpp:46] MultiStep Status: Iteration 30000, step = 1 | |
| I1022 11:48:00.326380 8564 sgd_solver.cpp:106] Iteration 30000, lr = 0.0001 | |
| I1022 11:48:11.492182 8564 solver.cpp:236] Iteration 30100, loss = 0.622436 | |
| I1022 11:48:11.492238 8564 solver.cpp:252] Train net output #0: loss = 0.622436 (* 1 = 0.622436 loss) | |
| I1022 11:48:11.492244 8564 sgd_solver.cpp:106] Iteration 30100, lr = 0.0001 | |
| I1022 11:48:22.675348 8564 solver.cpp:236] Iteration 30200, loss = 0.835799 | |
| I1022 11:48:22.675382 8564 solver.cpp:252] Train net output #0: loss = 0.835799 (* 1 = 0.835799 loss) | |
| I1022 11:48:22.675389 8564 sgd_solver.cpp:106] Iteration 30200, lr = 0.0001 | |
| I1022 11:48:33.874954 8564 solver.cpp:236] Iteration 30300, loss = 0.701128 | |
| I1022 11:48:33.874990 8564 solver.cpp:252] Train net output #0: loss = 0.701128 (* 1 = 0.701128 loss) | |
| I1022 11:48:33.874995 8564 sgd_solver.cpp:106] Iteration 30300, lr = 0.0001 | |
| I1022 11:48:44.991941 8564 solver.cpp:236] Iteration 30400, loss = 0.640916 | |
| I1022 11:48:44.992039 8564 solver.cpp:252] Train net output #0: loss = 0.640916 (* 1 = 0.640916 loss) | |
| I1022 11:48:44.992045 8564 sgd_solver.cpp:106] Iteration 30400, lr = 0.0001 | |
| I1022 11:48:56.155804 8564 solver.cpp:236] Iteration 30500, loss = 0.7662 | |
| I1022 11:48:56.155830 8564 solver.cpp:252] Train net output #0: loss = 0.7662 (* 1 = 0.7662 loss) | |
| I1022 11:48:56.155835 8564 sgd_solver.cpp:106] Iteration 30500, lr = 0.0001 | |
| I1022 11:49:07.312994 8564 solver.cpp:236] Iteration 30600, loss = 0.613073 | |
| I1022 11:49:07.313030 8564 solver.cpp:252] Train net output #0: loss = 0.613073 (* 1 = 0.613073 loss) | |
| I1022 11:49:07.313037 8564 sgd_solver.cpp:106] Iteration 30600, lr = 0.0001 | |
| I1022 11:49:18.473657 8564 solver.cpp:236] Iteration 30700, loss = 0.82724 | |
| I1022 11:49:18.473747 8564 solver.cpp:252] Train net output #0: loss = 0.82724 (* 1 = 0.82724 loss) | |
| I1022 11:49:18.473753 8564 sgd_solver.cpp:106] Iteration 30700, lr = 0.0001 | |
| I1022 11:49:29.647655 8564 solver.cpp:236] Iteration 30800, loss = 0.697973 | |
| I1022 11:49:29.647689 8564 solver.cpp:252] Train net output #0: loss = 0.697973 (* 1 = 0.697973 loss) | |
| I1022 11:49:29.647694 8564 sgd_solver.cpp:106] Iteration 30800, lr = 0.0001 | |
| I1022 11:49:40.797430 8564 solver.cpp:236] Iteration 30900, loss = 0.640463 | |
| I1022 11:49:40.797466 8564 solver.cpp:252] Train net output #0: loss = 0.640463 (* 1 = 0.640463 loss) | |
| I1022 11:49:40.797472 8564 sgd_solver.cpp:106] Iteration 30900, lr = 0.0001 | |
| I1022 11:49:51.474588 8564 solver.cpp:340] Iteration 31000, Testing net (#0) | |
| I1022 11:49:56.904474 8564 solver.cpp:408] Test net output #0: accuracy = 0.698 | |
| I1022 11:49:56.904500 8564 solver.cpp:408] Test net output #1: loss = 0.89011 (* 1 = 0.89011 loss) | |
| I1022 11:49:56.965456 8564 solver.cpp:236] Iteration 31000, loss = 0.760433 | |
| I1022 11:49:56.965490 8564 solver.cpp:252] Train net output #0: loss = 0.760433 (* 1 = 0.760433 loss) | |
| I1022 11:49:56.965497 8564 sgd_solver.cpp:106] Iteration 31000, lr = 0.0001 | |
| I1022 11:50:08.120568 8564 solver.cpp:236] Iteration 31100, loss = 0.610041 | |
| I1022 11:50:08.120604 8564 solver.cpp:252] Train net output #0: loss = 0.610041 (* 1 = 0.610041 loss) | |
| I1022 11:50:08.120609 8564 sgd_solver.cpp:106] Iteration 31100, lr = 0.0001 | |
| I1022 11:50:19.286280 8564 solver.cpp:236] Iteration 31200, loss = 0.824354 | |
| I1022 11:50:19.286329 8564 solver.cpp:252] Train net output #0: loss = 0.824354 (* 1 = 0.824354 loss) | |
| I1022 11:50:19.286336 8564 sgd_solver.cpp:106] Iteration 31200, lr = 0.0001 | |
| I1022 11:50:30.454028 8564 solver.cpp:236] Iteration 31300, loss = 0.695509 | |
| I1022 11:50:30.454115 8564 solver.cpp:252] Train net output #0: loss = 0.695509 (* 1 = 0.695509 loss) | |
| I1022 11:50:30.454121 8564 sgd_solver.cpp:106] Iteration 31300, lr = 0.0001 | |
| I1022 11:50:41.572471 8564 solver.cpp:236] Iteration 31400, loss = 0.640334 | |
| I1022 11:50:41.572505 8564 solver.cpp:252] Train net output #0: loss = 0.640334 (* 1 = 0.640334 loss) | |
| I1022 11:50:41.572510 8564 sgd_solver.cpp:106] Iteration 31400, lr = 0.0001 | |
| I1022 11:50:52.692456 8564 solver.cpp:236] Iteration 31500, loss = 0.755575 | |
| I1022 11:50:52.692490 8564 solver.cpp:252] Train net output #0: loss = 0.755575 (* 1 = 0.755575 loss) | |
| I1022 11:50:52.692497 8564 sgd_solver.cpp:106] Iteration 31500, lr = 0.0001 | |
| I1022 11:51:03.806608 8564 solver.cpp:236] Iteration 31600, loss = 0.607259 | |
| I1022 11:51:03.806701 8564 solver.cpp:252] Train net output #0: loss = 0.607259 (* 1 = 0.607259 loss) | |
| I1022 11:51:03.806707 8564 sgd_solver.cpp:106] Iteration 31600, lr = 0.0001 | |
| I1022 11:51:14.951393 8564 solver.cpp:236] Iteration 31700, loss = 0.822136 | |
| I1022 11:51:14.951429 8564 solver.cpp:252] Train net output #0: loss = 0.822136 (* 1 = 0.822136 loss) | |
| I1022 11:51:14.951434 8564 sgd_solver.cpp:106] Iteration 31700, lr = 0.0001 | |
| I1022 11:51:26.088932 8564 solver.cpp:236] Iteration 31800, loss = 0.693714 | |
| I1022 11:51:26.088958 8564 solver.cpp:252] Train net output #0: loss = 0.693714 (* 1 = 0.693714 loss) | |
| I1022 11:51:26.088963 8564 sgd_solver.cpp:106] Iteration 31800, lr = 0.0001 | |
| I1022 11:51:36.769866 8564 solver.cpp:236] Iteration 31900, loss = 0.640403 | |
| I1022 11:51:36.769934 8564 solver.cpp:252] Train net output #0: loss = 0.640403 (* 1 = 0.640403 loss) | |
| I1022 11:51:36.769940 8564 sgd_solver.cpp:106] Iteration 31900, lr = 0.0001 | |
| I1022 11:51:47.784500 8564 solver.cpp:340] Iteration 32000, Testing net (#0) | |
| I1022 11:51:53.157312 8564 solver.cpp:408] Test net output #0: accuracy = 0.6981 | |
| I1022 11:51:53.157338 8564 solver.cpp:408] Test net output #1: loss = 0.886714 (* 1 = 0.886714 loss) | |
| I1022 11:51:53.217130 8564 solver.cpp:236] Iteration 32000, loss = 0.751232 | |
| I1022 11:51:53.217154 8564 solver.cpp:252] Train net output #0: loss = 0.751232 (* 1 = 0.751232 loss) | |
| I1022 11:51:53.217157 8564 sgd_solver.cpp:106] Iteration 32000, lr = 0.0001 | |
| I1022 11:52:04.356755 8564 solver.cpp:236] Iteration 32100, loss = 0.604974 | |
| I1022 11:52:04.356791 8564 solver.cpp:252] Train net output #0: loss = 0.604974 (* 1 = 0.604974 loss) | |
| I1022 11:52:04.356797 8564 sgd_solver.cpp:106] Iteration 32100, lr = 0.0001 | |
| I1022 11:52:15.493125 8564 solver.cpp:236] Iteration 32200, loss = 0.820321 | |
| I1022 11:52:15.493232 8564 solver.cpp:252] Train net output #0: loss = 0.820321 (* 1 = 0.820321 loss) | |
| I1022 11:52:15.493248 8564 sgd_solver.cpp:106] Iteration 32200, lr = 0.0001 | |
| I1022 11:52:26.634297 8564 solver.cpp:236] Iteration 32300, loss = 0.692454 | |
| I1022 11:52:26.634323 8564 solver.cpp:252] Train net output #0: loss = 0.692454 (* 1 = 0.692454 loss) | |
| I1022 11:52:26.634328 8564 sgd_solver.cpp:106] Iteration 32300, lr = 0.0001 | |
| I1022 11:52:37.799424 8564 solver.cpp:236] Iteration 32400, loss = 0.640601 | |
| I1022 11:52:37.799459 8564 solver.cpp:252] Train net output #0: loss = 0.640601 (* 1 = 0.640601 loss) | |
| I1022 11:52:37.799464 8564 sgd_solver.cpp:106] Iteration 32400, lr = 0.0001 | |
| I1022 11:52:48.946633 8564 solver.cpp:236] Iteration 32500, loss = 0.747267 | |
| I1022 11:52:48.946722 8564 solver.cpp:252] Train net output #0: loss = 0.747267 (* 1 = 0.747267 loss) | |
| I1022 11:52:48.946727 8564 sgd_solver.cpp:106] Iteration 32500, lr = 0.0001 | |
| I1022 11:53:00.080404 8564 solver.cpp:236] Iteration 32600, loss = 0.602863 | |
| I1022 11:53:00.080440 8564 solver.cpp:252] Train net output #0: loss = 0.602863 (* 1 = 0.602863 loss) | |
| I1022 11:53:00.080446 8564 sgd_solver.cpp:106] Iteration 32600, lr = 0.0001 | |
| I1022 11:53:11.218451 8564 solver.cpp:236] Iteration 32700, loss = 0.818763 | |
| I1022 11:53:11.218485 8564 solver.cpp:252] Train net output #0: loss = 0.818763 (* 1 = 0.818763 loss) | |
| I1022 11:53:11.218492 8564 sgd_solver.cpp:106] Iteration 32700, lr = 0.0001 | |
| I1022 11:53:22.386447 8564 solver.cpp:236] Iteration 32800, loss = 0.691561 | |
| I1022 11:53:22.386544 8564 solver.cpp:252] Train net output #0: loss = 0.691561 (* 1 = 0.691561 loss) | |
| I1022 11:53:22.386559 8564 sgd_solver.cpp:106] Iteration 32800, lr = 0.0001 | |
| I1022 11:53:32.999651 8564 solver.cpp:236] Iteration 32900, loss = 0.640797 | |
| I1022 11:53:32.999677 8564 solver.cpp:252] Train net output #0: loss = 0.640797 (* 1 = 0.640797 loss) | |
| I1022 11:53:32.999681 8564 sgd_solver.cpp:106] Iteration 32900, lr = 0.0001 | |
| I1022 11:53:44.102501 8564 solver.cpp:340] Iteration 33000, Testing net (#0) | |
| I1022 11:53:49.633316 8564 solver.cpp:408] Test net output #0: accuracy = 0.7001 | |
| I1022 11:53:49.633379 8564 solver.cpp:408] Test net output #1: loss = 0.874897 (* 1 = 0.874897 loss) | |
| I1022 11:53:49.697208 8564 solver.cpp:236] Iteration 33000, loss = 0.74367 | |
| I1022 11:53:49.697249 8564 solver.cpp:252] Train net output #0: loss = 0.74367 (* 1 = 0.74367 loss) | |
| I1022 11:53:49.697266 8564 sgd_solver.cpp:106] Iteration 33000, lr = 0.0001 | |
| I1022 11:54:00.877634 8564 solver.cpp:236] Iteration 33100, loss = 0.600944 | |
| I1022 11:54:00.877754 8564 solver.cpp:252] Train net output #0: loss = 0.600944 (* 1 = 0.600944 loss) | |
| I1022 11:54:00.877760 8564 sgd_solver.cpp:106] Iteration 33100, lr = 0.0001 | |
| I1022 11:54:12.002399 8564 solver.cpp:236] Iteration 33200, loss = 0.817458 | |
| I1022 11:54:12.002436 8564 solver.cpp:252] Train net output #0: loss = 0.817458 (* 1 = 0.817458 loss) | |
| I1022 11:54:12.002442 8564 sgd_solver.cpp:106] Iteration 33200, lr = 0.0001 | |
| I1022 11:54:23.172279 8564 solver.cpp:236] Iteration 33300, loss = 0.690838 | |
| I1022 11:54:23.172317 8564 solver.cpp:252] Train net output #0: loss = 0.690838 (* 1 = 0.690838 loss) | |
| I1022 11:54:23.172323 8564 sgd_solver.cpp:106] Iteration 33300, lr = 0.0001 | |
| I1022 11:54:34.295259 8564 solver.cpp:236] Iteration 33400, loss = 0.640957 | |
| I1022 11:54:34.295341 8564 solver.cpp:252] Train net output #0: loss = 0.640957 (* 1 = 0.640957 loss) | |
| I1022 11:54:34.295356 8564 sgd_solver.cpp:106] Iteration 33400, lr = 0.0001 | |
| I1022 11:54:45.470350 8564 solver.cpp:236] Iteration 33500, loss = 0.740471 | |
| I1022 11:54:45.470386 8564 solver.cpp:252] Train net output #0: loss = 0.740471 (* 1 = 0.740471 loss) | |
| I1022 11:54:45.470391 8564 sgd_solver.cpp:106] Iteration 33500, lr = 0.0001 | |
| I1022 11:54:56.598796 8564 solver.cpp:236] Iteration 33600, loss = 0.599203 | |
| I1022 11:54:56.598834 8564 solver.cpp:252] Train net output #0: loss = 0.599203 (* 1 = 0.599203 loss) | |
| I1022 11:54:56.598839 8564 sgd_solver.cpp:106] Iteration 33600, lr = 0.0001 | |
| I1022 11:55:07.674783 8564 solver.cpp:236] Iteration 33700, loss = 0.816411 | |
| I1022 11:55:07.674871 8564 solver.cpp:252] Train net output #0: loss = 0.816411 (* 1 = 0.816411 loss) | |
| I1022 11:55:07.674876 8564 sgd_solver.cpp:106] Iteration 33700, lr = 0.0001 | |
| I1022 11:55:18.246320 8564 solver.cpp:236] Iteration 33800, loss = 0.690261 | |
| I1022 11:55:18.246356 8564 solver.cpp:252] Train net output #0: loss = 0.690261 (* 1 = 0.690261 loss) | |
| I1022 11:55:18.246361 8564 sgd_solver.cpp:106] Iteration 33800, lr = 0.0001 | |
| I1022 11:55:29.413061 8564 solver.cpp:236] Iteration 33900, loss = 0.641011 | |
| I1022 11:55:29.413099 8564 solver.cpp:252] Train net output #0: loss = 0.641011 (* 1 = 0.641011 loss) | |
| I1022 11:55:29.413103 8564 sgd_solver.cpp:106] Iteration 33900, lr = 0.0001 | |
| I1022 11:55:40.484966 8564 solver.cpp:340] Iteration 34000, Testing net (#0) | |
| I1022 11:55:45.826992 8564 solver.cpp:408] Test net output #0: accuracy = 0.7038 | |
| I1022 11:55:45.827024 8564 solver.cpp:408] Test net output #1: loss = 0.864182 (* 1 = 0.864182 loss) | |
| I1022 11:55:45.888005 8564 solver.cpp:236] Iteration 34000, loss = 0.737492 | |
| I1022 11:55:45.888039 8564 solver.cpp:252] Train net output #0: loss = 0.737492 (* 1 = 0.737492 loss) | |
| I1022 11:55:45.888046 8564 sgd_solver.cpp:106] Iteration 34000, lr = 0.0001 | |
| I1022 11:55:56.987709 8564 solver.cpp:236] Iteration 34100, loss = 0.597436 | |
| I1022 11:55:56.987745 8564 solver.cpp:252] Train net output #0: loss = 0.597436 (* 1 = 0.597436 loss) | |
| I1022 11:55:56.987750 8564 sgd_solver.cpp:106] Iteration 34100, lr = 0.0001 | |
| I1022 11:56:08.133744 8564 solver.cpp:236] Iteration 34200, loss = 0.815482 | |
| I1022 11:56:08.133779 8564 solver.cpp:252] Train net output #0: loss = 0.815482 (* 1 = 0.815482 loss) | |
| I1022 11:56:08.133785 8564 sgd_solver.cpp:106] Iteration 34200, lr = 0.0001 | |
| I1022 11:56:19.492722 8564 solver.cpp:236] Iteration 34300, loss = 0.689794 | |
| I1022 11:56:19.492826 8564 solver.cpp:252] Train net output #0: loss = 0.689794 (* 1 = 0.689794 loss) | |
| I1022 11:56:19.492838 8564 sgd_solver.cpp:106] Iteration 34300, lr = 0.0001 | |
| I1022 11:56:31.645283 8564 solver.cpp:236] Iteration 34400, loss = 0.641044 | |
| I1022 11:56:31.645311 8564 solver.cpp:252] Train net output #0: loss = 0.641044 (* 1 = 0.641044 loss) | |
| I1022 11:56:31.645318 8564 sgd_solver.cpp:106] Iteration 34400, lr = 0.0001 | |
| I1022 11:56:42.801470 8564 solver.cpp:236] Iteration 34500, loss = 0.734743 | |
| I1022 11:56:42.801506 8564 solver.cpp:252] Train net output #0: loss = 0.734743 (* 1 = 0.734743 loss) | |
| I1022 11:56:42.801512 8564 sgd_solver.cpp:106] Iteration 34500, lr = 0.0001 | |
| I1022 11:56:53.931752 8564 solver.cpp:236] Iteration 34600, loss = 0.595749 | |
| I1022 11:56:53.931860 8564 solver.cpp:252] Train net output #0: loss = 0.595749 (* 1 = 0.595749 loss) | |
| I1022 11:56:53.931866 8564 sgd_solver.cpp:106] Iteration 34600, lr = 0.0001 | |
| I1022 11:57:05.137717 8564 solver.cpp:236] Iteration 34700, loss = 0.814748 | |
| I1022 11:57:05.137779 8564 solver.cpp:252] Train net output #0: loss = 0.814748 (* 1 = 0.814748 loss) | |
| I1022 11:57:05.137792 8564 sgd_solver.cpp:106] Iteration 34700, lr = 0.0001 | |
| I1022 11:57:16.654422 8564 solver.cpp:236] Iteration 34800, loss = 0.689472 | |
| I1022 11:57:16.654458 8564 solver.cpp:252] Train net output #0: loss = 0.689472 (* 1 = 0.689472 loss) | |
| I1022 11:57:16.654464 8564 sgd_solver.cpp:106] Iteration 34800, lr = 0.0001 | |
| I1022 11:57:27.834247 8564 solver.cpp:236] Iteration 34900, loss = 0.641037 | |
| I1022 11:57:27.834344 8564 solver.cpp:252] Train net output #0: loss = 0.641037 (* 1 = 0.641037 loss) | |
| I1022 11:57:27.834359 8564 sgd_solver.cpp:106] Iteration 34900, lr = 0.0001 | |
| I1022 11:57:41.276960 8564 solver.cpp:340] Iteration 35000, Testing net (#0) | |
| I1022 11:57:49.033498 8564 solver.cpp:408] Test net output #0: accuracy = 0.7061 | |
| I1022 11:57:49.033550 8564 solver.cpp:408] Test net output #1: loss = 0.855199 (* 1 = 0.855199 loss) | |
| I1022 11:57:49.124814 8564 solver.cpp:236] Iteration 35000, loss = 0.732233 | |
| I1022 11:57:49.124850 8564 solver.cpp:252] Train net output #0: loss = 0.732233 (* 1 = 0.732233 loss) | |
| I1022 11:57:49.124855 8564 sgd_solver.cpp:46] MultiStep Status: Iteration 35000, step = 2 | |
| I1022 11:57:49.124857 8564 sgd_solver.cpp:106] Iteration 35000, lr = 1e-05 | |
| I1022 11:58:03.152465 8564 solver.cpp:236] Iteration 35100, loss = 0.590509 | |
| I1022 11:58:03.152554 8564 solver.cpp:252] Train net output #0: loss = 0.590509 (* 1 = 0.590509 loss) | |
| I1022 11:58:03.152567 8564 sgd_solver.cpp:106] Iteration 35100, lr = 1e-05 | |
| I1022 11:58:14.294396 8564 solver.cpp:236] Iteration 35200, loss = 0.812733 | |
| I1022 11:58:14.294447 8564 solver.cpp:252] Train net output #0: loss = 0.812733 (* 1 = 0.812733 loss) | |
| I1022 11:58:14.294456 8564 sgd_solver.cpp:106] Iteration 35200, lr = 1e-05 | |
| I1022 11:58:28.432822 8564 solver.cpp:236] Iteration 35300, loss = 0.69186 | |
| I1022 11:58:28.432859 8564 solver.cpp:252] Train net output #0: loss = 0.69186 (* 1 = 0.69186 loss) | |
| I1022 11:58:28.432867 8564 sgd_solver.cpp:106] Iteration 35300, lr = 1e-05 | |
| I1022 11:58:44.333853 8564 solver.cpp:236] Iteration 35400, loss = 0.623849 | |
| I1022 11:58:44.333941 8564 solver.cpp:252] Train net output #0: loss = 0.623849 (* 1 = 0.623849 loss) | |
| I1022 11:58:44.333947 8564 sgd_solver.cpp:106] Iteration 35400, lr = 1e-05 | |
| I1022 11:59:00.238854 8564 solver.cpp:236] Iteration 35500, loss = 0.722968 | |
| I1022 11:59:00.238890 8564 solver.cpp:252] Train net output #0: loss = 0.722968 (* 1 = 0.722968 loss) | |
| I1022 11:59:00.238895 8564 sgd_solver.cpp:106] Iteration 35500, lr = 1e-05 | |
| I1022 11:59:14.286098 8564 solver.cpp:236] Iteration 35600, loss = 0.581258 | |
| I1022 11:59:14.286133 8564 solver.cpp:252] Train net output #0: loss = 0.581258 (* 1 = 0.581258 loss) | |
| I1022 11:59:14.286137 8564 sgd_solver.cpp:106] Iteration 35600, lr = 1e-05 | |
| I1022 11:59:24.880439 8564 solver.cpp:236] Iteration 35700, loss = 0.810592 | |
| I1022 11:59:24.880535 8564 solver.cpp:252] Train net output #0: loss = 0.810592 (* 1 = 0.810592 loss) | |
| I1022 11:59:24.880550 8564 sgd_solver.cpp:106] Iteration 35700, lr = 1e-05 | |
| I1022 11:59:36.006855 8564 solver.cpp:236] Iteration 35800, loss = 0.689966 | |
| I1022 11:59:36.006881 8564 solver.cpp:252] Train net output #0: loss = 0.689966 (* 1 = 0.689966 loss) | |
| I1022 11:59:36.006886 8564 sgd_solver.cpp:106] Iteration 35800, lr = 1e-05 | |
| I1022 11:59:47.159737 8564 solver.cpp:236] Iteration 35900, loss = 0.624944 | |
| I1022 11:59:47.159782 8564 solver.cpp:252] Train net output #0: loss = 0.624944 (* 1 = 0.624944 loss) | |
| I1022 11:59:47.159788 8564 sgd_solver.cpp:106] Iteration 35900, lr = 1e-05 | |
| I1022 12:00:02.516868 8564 solver.cpp:340] Iteration 36000, Testing net (#0) | |
| I1022 12:00:10.301590 8564 solver.cpp:408] Test net output #0: accuracy = 0.7322 | |
| I1022 12:00:10.301635 8564 solver.cpp:408] Test net output #1: loss = 0.787552 (* 1 = 0.787552 loss) | |
| I1022 12:00:10.400202 8564 solver.cpp:236] Iteration 36000, loss = 0.723015 | |
| I1022 12:00:10.400240 8564 solver.cpp:252] Train net output #0: loss = 0.723015 (* 1 = 0.723015 loss) | |
| I1022 12:00:10.400246 8564 sgd_solver.cpp:106] Iteration 36000, lr = 1e-05 | |
| I1022 12:00:26.318359 8564 solver.cpp:236] Iteration 36100, loss = 0.579465 | |
| I1022 12:00:26.318409 8564 solver.cpp:252] Train net output #0: loss = 0.579465 (* 1 = 0.579465 loss) | |
| I1022 12:00:26.318419 8564 sgd_solver.cpp:106] Iteration 36100, lr = 1e-05 | |
| I1022 12:00:38.633611 8564 solver.cpp:236] Iteration 36200, loss = 0.809783 | |
| I1022 12:00:38.633683 8564 solver.cpp:252] Train net output #0: loss = 0.809783 (* 1 = 0.809783 loss) | |
| I1022 12:00:38.633689 8564 sgd_solver.cpp:106] Iteration 36200, lr = 1e-05 | |
| I1022 12:00:49.836977 8564 solver.cpp:236] Iteration 36300, loss = 0.689634 | |
| I1022 12:00:49.837014 8564 solver.cpp:252] Train net output #0: loss = 0.689634 (* 1 = 0.689634 loss) | |
| I1022 12:00:49.837020 8564 sgd_solver.cpp:106] Iteration 36300, lr = 1e-05 | |
| I1022 12:01:01.212606 8564 solver.cpp:236] Iteration 36400, loss = 0.625275 | |
| I1022 12:01:01.212643 8564 solver.cpp:252] Train net output #0: loss = 0.625275 (* 1 = 0.625275 loss) | |
| I1022 12:01:01.212649 8564 sgd_solver.cpp:106] Iteration 36400, lr = 1e-05 | |
| I1022 12:01:13.876613 8564 solver.cpp:236] Iteration 36500, loss = 0.722823 | |
| I1022 12:01:13.876703 8564 solver.cpp:252] Train net output #0: loss = 0.722823 (* 1 = 0.722823 loss) | |
| I1022 12:01:13.876708 8564 sgd_solver.cpp:106] Iteration 36500, lr = 1e-05 | |
| I1022 12:01:24.787740 8564 solver.cpp:236] Iteration 36600, loss = 0.57876 | |
| I1022 12:01:24.787770 8564 solver.cpp:252] Train net output #0: loss = 0.57876 (* 1 = 0.57876 loss) | |
| I1022 12:01:24.787775 8564 sgd_solver.cpp:106] Iteration 36600, lr = 1e-05 | |
| I1022 12:01:35.699867 8564 solver.cpp:236] Iteration 36700, loss = 0.809348 | |
| I1022 12:01:35.699905 8564 solver.cpp:252] Train net output #0: loss = 0.809348 (* 1 = 0.809348 loss) | |
| I1022 12:01:35.699910 8564 sgd_solver.cpp:106] Iteration 36700, lr = 1e-05 | |
| I1022 12:01:46.821734 8564 solver.cpp:236] Iteration 36800, loss = 0.689604 | |
| I1022 12:01:46.821812 8564 solver.cpp:252] Train net output #0: loss = 0.689604 (* 1 = 0.689604 loss) | |
| I1022 12:01:46.821817 8564 sgd_solver.cpp:106] Iteration 36800, lr = 1e-05 | |
| I1022 12:01:57.975872 8564 solver.cpp:236] Iteration 36900, loss = 0.625462 | |
| I1022 12:01:57.975899 8564 solver.cpp:252] Train net output #0: loss = 0.625462 (* 1 = 0.625462 loss) | |
| I1022 12:01:57.975905 8564 sgd_solver.cpp:106] Iteration 36900, lr = 1e-05 | |
| I1022 12:02:09.113597 8564 solver.cpp:340] Iteration 37000, Testing net (#0) | |
| I1022 12:02:14.513370 8564 solver.cpp:408] Test net output #0: accuracy = 0.7335 | |
| I1022 12:02:14.513393 8564 solver.cpp:408] Test net output #1: loss = 0.784984 (* 1 = 0.784984 loss) | |
| I1022 12:02:14.577836 8564 solver.cpp:236] Iteration 37000, loss = 0.72254 | |
| I1022 12:02:14.577860 8564 solver.cpp:252] Train net output #0: loss = 0.72254 (* 1 = 0.72254 loss) | |
| I1022 12:02:14.577865 8564 sgd_solver.cpp:106] Iteration 37000, lr = 1e-05 | |
| I1022 12:02:25.719784 8564 solver.cpp:236] Iteration 37100, loss = 0.578365 | |
| I1022 12:02:25.719871 8564 solver.cpp:252] Train net output #0: loss = 0.578365 (* 1 = 0.578365 loss) | |
| I1022 12:02:25.719877 8564 sgd_solver.cpp:106] Iteration 37100, lr = 1e-05 | |
| I1022 12:02:36.856029 8564 solver.cpp:236] Iteration 37200, loss = 0.809059 | |
| I1022 12:02:36.856065 8564 solver.cpp:252] Train net output #0: loss = 0.809059 (* 1 = 0.809059 loss) | |
| I1022 12:02:36.856070 8564 sgd_solver.cpp:106] Iteration 37200, lr = 1e-05 | |
| I1022 12:02:47.976176 8564 solver.cpp:236] Iteration 37300, loss = 0.689619 | |
| I1022 12:02:47.976202 8564 solver.cpp:252] Train net output #0: loss = 0.689619 (* 1 = 0.689619 loss) | |
| I1022 12:02:47.976207 8564 sgd_solver.cpp:106] Iteration 37300, lr = 1e-05 | |
| I1022 12:02:59.134668 8564 solver.cpp:236] Iteration 37400, loss = 0.625574 | |
| I1022 12:02:59.134768 8564 solver.cpp:252] Train net output #0: loss = 0.625574 (* 1 = 0.625574 loss) | |
| I1022 12:02:59.134774 8564 sgd_solver.cpp:106] Iteration 37400, lr = 1e-05 | |
| I1022 12:03:10.289475 8564 solver.cpp:236] Iteration 37500, loss = 0.722203 | |
| I1022 12:03:10.289512 8564 solver.cpp:252] Train net output #0: loss = 0.722203 (* 1 = 0.722203 loss) | |
| I1022 12:03:10.289518 8564 sgd_solver.cpp:106] Iteration 37500, lr = 1e-05 | |
| I1022 12:03:19.174435 8564 solver.cpp:236] Iteration 37600, loss = 0.57808 | |
| I1022 12:03:19.174460 8564 solver.cpp:252] Train net output #0: loss = 0.57808 (* 1 = 0.57808 loss) | |
| I1022 12:03:19.174465 8564 sgd_solver.cpp:106] Iteration 37600, lr = 1e-05 | |
| I1022 12:03:26.015373 8564 solver.cpp:236] Iteration 37700, loss = 0.808847 | |
| I1022 12:03:26.015408 8564 solver.cpp:252] Train net output #0: loss = 0.808847 (* 1 = 0.808847 loss) | |
| I1022 12:03:26.015414 8564 sgd_solver.cpp:106] Iteration 37700, lr = 1e-05 | |
| I1022 12:03:35.227355 8564 solver.cpp:236] Iteration 37800, loss = 0.689611 | |
| I1022 12:03:35.227442 8564 solver.cpp:252] Train net output #0: loss = 0.689611 (* 1 = 0.689611 loss) | |
| I1022 12:03:35.227447 8564 sgd_solver.cpp:106] Iteration 37800, lr = 1e-05 | |
| I1022 12:03:44.473572 8564 solver.cpp:236] Iteration 37900, loss = 0.625647 | |
| I1022 12:03:44.473598 8564 solver.cpp:252] Train net output #0: loss = 0.625647 (* 1 = 0.625647 loss) | |
| I1022 12:03:44.473603 8564 sgd_solver.cpp:106] Iteration 37900, lr = 1e-05 | |
| I1022 12:03:50.886652 8564 solver.cpp:340] Iteration 38000, Testing net (#0) | |
| I1022 12:03:53.312328 8564 solver.cpp:408] Test net output #0: accuracy = 0.7346 | |
| I1022 12:03:53.312361 8564 solver.cpp:408] Test net output #1: loss = 0.784374 (* 1 = 0.784374 loss) | |
| I1022 12:03:53.341193 8564 solver.cpp:236] Iteration 38000, loss = 0.721833 | |
| I1022 12:03:53.341214 8564 solver.cpp:252] Train net output #0: loss = 0.721833 (* 1 = 0.721833 loss) | |
| I1022 12:03:53.341219 8564 sgd_solver.cpp:106] Iteration 38000, lr = 1e-05 | |
| I1022 12:03:58.522460 8564 solver.cpp:236] Iteration 38100, loss = 0.577827 | |
| I1022 12:03:58.522500 8564 solver.cpp:252] Train net output #0: loss = 0.577827 (* 1 = 0.577827 loss) | |
| I1022 12:03:58.522505 8564 sgd_solver.cpp:106] Iteration 38100, lr = 1e-05 | |
| I1022 12:04:06.463690 8564 solver.cpp:236] Iteration 38200, loss = 0.808664 | |
| I1022 12:04:06.463769 8564 solver.cpp:252] Train net output #0: loss = 0.808664 (* 1 = 0.808664 loss) | |
| I1022 12:04:06.463783 8564 sgd_solver.cpp:106] Iteration 38200, lr = 1e-05 | |
| I1022 12:04:15.694309 8564 solver.cpp:236] Iteration 38300, loss = 0.689592 | |
| I1022 12:04:15.694337 8564 solver.cpp:252] Train net output #0: loss = 0.689592 (* 1 = 0.689592 loss) | |
| I1022 12:04:15.694342 8564 sgd_solver.cpp:106] Iteration 38300, lr = 1e-05 | |
| I1022 12:04:24.921224 8564 solver.cpp:236] Iteration 38400, loss = 0.625694 | |
| I1022 12:04:24.921252 8564 solver.cpp:252] Train net output #0: loss = 0.625694 (* 1 = 0.625694 loss) | |
| I1022 12:04:24.921257 8564 sgd_solver.cpp:106] Iteration 38400, lr = 1e-05 | |
| I1022 12:04:34.151561 8564 solver.cpp:236] Iteration 38500, loss = 0.721449 | |
| I1022 12:04:34.151587 8564 solver.cpp:252] Train net output #0: loss = 0.721449 (* 1 = 0.721449 loss) | |
| I1022 12:04:34.151592 8564 sgd_solver.cpp:106] Iteration 38500, lr = 1e-05 | |
| I1022 12:04:39.363477 8564 solver.cpp:236] Iteration 38600, loss = 0.577588 | |
| I1022 12:04:39.363580 8564 solver.cpp:252] Train net output #0: loss = 0.577588 (* 1 = 0.577588 loss) | |
| I1022 12:04:39.363586 8564 sgd_solver.cpp:106] Iteration 38600, lr = 1e-05 | |
| I1022 12:04:44.483985 8564 solver.cpp:236] Iteration 38700, loss = 0.808506 | |
| I1022 12:04:44.484009 8564 solver.cpp:252] Train net output #0: loss = 0.808506 (* 1 = 0.808506 loss) | |
| I1022 12:04:44.484014 8564 sgd_solver.cpp:106] Iteration 38700, lr = 1e-05 | |
| I1022 12:04:49.612747 8564 solver.cpp:236] Iteration 38800, loss = 0.689549 | |
| I1022 12:04:49.612782 8564 solver.cpp:252] Train net output #0: loss = 0.689549 (* 1 = 0.689549 loss) | |
| I1022 12:04:49.612787 8564 sgd_solver.cpp:106] Iteration 38800, lr = 1e-05 | |
| I1022 12:04:54.751665 8564 solver.cpp:236] Iteration 38900, loss = 0.625724 | |
| I1022 12:04:54.751689 8564 solver.cpp:252] Train net output #0: loss = 0.625724 (* 1 = 0.625724 loss) | |
| I1022 12:04:54.751693 8564 sgd_solver.cpp:106] Iteration 38900, lr = 1e-05 | |
| I1022 12:05:02.868448 8564 solver.cpp:340] Iteration 39000, Testing net (#0) | |
| I1022 12:05:05.362264 8564 solver.cpp:408] Test net output #0: accuracy = 0.7345 | |
| I1022 12:05:05.362287 8564 solver.cpp:408] Test net output #1: loss = 0.783969 (* 1 = 0.783969 loss) | |
| I1022 12:05:05.390991 8564 solver.cpp:236] Iteration 39000, loss = 0.721051 | |
| I1022 12:05:05.391011 8564 solver.cpp:252] Train net output #0: loss = 0.721051 (* 1 = 0.721051 loss) | |
| I1022 12:05:05.391016 8564 sgd_solver.cpp:106] Iteration 39000, lr = 1e-05 | |
| I1022 12:05:10.499817 8564 solver.cpp:236] Iteration 39100, loss = 0.577356 | |
| I1022 12:05:10.499923 8564 solver.cpp:252] Train net output #0: loss = 0.577356 (* 1 = 0.577356 loss) | |
| I1022 12:05:10.499928 8564 sgd_solver.cpp:106] Iteration 39100, lr = 1e-05 | |
| I1022 12:05:15.609794 8564 solver.cpp:236] Iteration 39200, loss = 0.808365 | |
| I1022 12:05:15.609818 8564 solver.cpp:252] Train net output #0: loss = 0.808365 (* 1 = 0.808365 loss) | |
| I1022 12:05:15.609823 8564 sgd_solver.cpp:106] Iteration 39200, lr = 1e-05 | |
| I1022 12:05:20.723965 8564 solver.cpp:236] Iteration 39300, loss = 0.689493 | |
| I1022 12:05:20.724001 8564 solver.cpp:252] Train net output #0: loss = 0.689493 (* 1 = 0.689493 loss) | |
| I1022 12:05:20.724006 8564 sgd_solver.cpp:106] Iteration 39300, lr = 1e-05 | |
| I1022 12:05:25.915421 8564 solver.cpp:236] Iteration 39400, loss = 0.625735 | |
| I1022 12:05:25.915446 8564 solver.cpp:252] Train net output #0: loss = 0.625735 (* 1 = 0.625735 loss) | |
| I1022 12:05:25.915451 8564 sgd_solver.cpp:106] Iteration 39400, lr = 1e-05 | |
| I1022 12:05:31.056268 8564 solver.cpp:236] Iteration 39500, loss = 0.720644 | |
| I1022 12:05:31.056306 8564 solver.cpp:252] Train net output #0: loss = 0.720644 (* 1 = 0.720644 loss) | |
| I1022 12:05:31.056311 8564 sgd_solver.cpp:106] Iteration 39500, lr = 1e-05 | |
| I1022 12:05:36.230237 8564 solver.cpp:236] Iteration 39600, loss = 0.577124 | |
| I1022 12:05:36.230263 8564 solver.cpp:252] Train net output #0: loss = 0.577124 (* 1 = 0.577124 loss) | |
| I1022 12:05:36.230268 8564 sgd_solver.cpp:106] Iteration 39600, lr = 1e-05 | |
| I1022 12:05:41.367669 8564 solver.cpp:236] Iteration 39700, loss = 0.808229 | |
| I1022 12:05:41.367772 8564 solver.cpp:252] Train net output #0: loss = 0.808229 (* 1 = 0.808229 loss) | |
| I1022 12:05:41.367777 8564 sgd_solver.cpp:106] Iteration 39700, lr = 1e-05 | |
| I1022 12:05:47.132261 8564 solver.cpp:236] Iteration 39800, loss = 0.689421 | |
| I1022 12:05:47.132287 8564 solver.cpp:252] Train net output #0: loss = 0.689421 (* 1 = 0.689421 loss) | |
| I1022 12:05:47.132292 8564 sgd_solver.cpp:106] Iteration 39800, lr = 1e-05 | |
| I1022 12:05:52.247568 8564 solver.cpp:236] Iteration 39900, loss = 0.625732 | |
| I1022 12:05:52.247591 8564 solver.cpp:252] Train net output #0: loss = 0.625732 (* 1 = 0.625732 loss) | |
| I1022 12:05:52.247596 8564 sgd_solver.cpp:106] Iteration 39900, lr = 1e-05 | |
| I1022 12:05:57.339275 8564 solver.cpp:461] Snapshotting to binary proto file examples/cifar10_full_sigmoid_bn_iter_40000.caffemodel | |
| I1022 12:05:57.340836 8564 sgd_solver.cpp:269] Snapshotting solver state to binary proto file examples/cifar10_full_sigmoid_bn_iter_40000.solverstate | |
| I1022 12:05:57.365314 8564 solver.cpp:320] Iteration 40000, loss = 0.720238 | |
| I1022 12:05:57.365331 8564 solver.cpp:340] Iteration 40000, Testing net (#0) | |
| I1022 12:05:59.749449 8564 solver.cpp:408] Test net output #0: accuracy = 0.7344 | |
| I1022 12:05:59.749474 8564 solver.cpp:408] Test net output #1: loss = 0.783632 (* 1 = 0.783632 loss) | |
| I1022 12:05:59.749477 8564 solver.cpp:325] Optimization Done. | |
| I1022 12:05:59.749480 8564 caffe.cpp:215] Optimization Done. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment