Created
July 24, 2015 07:47
-
-
Save amiltonwong/76d5ba61e69f2e89a896 to your computer and use it in GitHub Desktop.
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
I0724 15:20:24.137687 26781 caffe.cpp:113] Use GPU with device ID 0 | |
I0724 15:20:24.351411 26781 caffe.cpp:121] Starting Optimization | |
I0724 15:20:24.351496 26781 solver.cpp:32] Initializing solver from parameters: | |
test_iter: 100 | |
test_interval: 500 | |
base_lr: 0.001 | |
display: 100 | |
max_iter: 4000 | |
lr_policy: "fixed" | |
momentum: 0.9 | |
weight_decay: 0.004 | |
snapshot: 4000 | |
snapshot_prefix: "examples/cifar10/cifar10_quick" | |
solver_mode: GPU | |
net: "examples/cifar10/cifar10_quick_train_test.prototxt" | |
I0724 15:20:24.351536 26781 solver.cpp:70] Creating training net from net file: examples/cifar10/cifar10_quick_train_test.prototxt | |
I0724 15:20:24.351855 26781 net.cpp:257] The NetState phase (0) differed from the phase (1) specified by a rule in layer cifar | |
I0724 15:20:24.351873 26781 net.cpp:257] The NetState phase (0) differed from the phase (1) specified by a rule in layer accuracy | |
I0724 15:20:24.351958 26781 net.cpp:42] Initializing net from parameters: | |
name: "CIFAR10_quick" | |
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 | |
} | |
param { | |
lr_mult: 2 | |
} | |
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: "relu1" | |
type: "ReLU" | |
bottom: "pool1" | |
top: "pool1" | |
} | |
layer { | |
name: "conv2" | |
type: "Convolution" | |
bottom: "pool1" | |
top: "conv2" | |
param { | |
lr_mult: 1 | |
} | |
param { | |
lr_mult: 2 | |
} | |
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: "relu2" | |
type: "ReLU" | |
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 | |
} | |
param { | |
lr_mult: 2 | |
} | |
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: "relu3" | |
type: "ReLU" | |
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 | |
} | |
param { | |
lr_mult: 2 | |
} | |
inner_product_param { | |
num_output: 64 | |
weight_filler { | |
type: "gaussian" | |
std: 0.1 | |
} | |
bias_filler { | |
type: "constant" | |
} | |
} | |
} | |
layer { | |
name: "ip2" | |
type: "InnerProduct" | |
bottom: "ip1" | |
top: "ip2" | |
param { | |
lr_mult: 1 | |
} | |
param { | |
lr_mult: 2 | |
} | |
inner_product_param { | |
num_output: 10 | |
weight_filler { | |
type: "gaussian" | |
std: 0.1 | |
} | |
bias_filler { | |
type: "constant" | |
} | |
} | |
} | |
layer { | |
name: "loss" | |
type: "SoftmaxWithLoss" | |
bottom: "ip2" | |
bottom: "label" | |
top: "loss" | |
} | |
I0724 15:20:24.352023 26781 layer_factory.hpp:74] Creating layer cifar | |
I0724 15:20:24.352041 26781 net.cpp:84] Creating Layer cifar | |
I0724 15:20:24.352046 26781 net.cpp:338] cifar -> data | |
I0724 15:20:24.352074 26781 net.cpp:338] cifar -> label | |
I0724 15:20:24.352087 26781 net.cpp:113] Setting up cifar | |
I0724 15:20:24.352169 26781 db.cpp:34] Opened lmdb examples/cifar10/cifar10_train_lmdb | |
I0724 15:20:24.352215 26781 data_layer.cpp:67] output data size: 100,3,32,32 | |
I0724 15:20:24.352226 26781 data_transformer.cpp:22] Loading mean file from: examples/cifar10/mean.binaryproto | |
I0724 15:20:24.352715 26781 net.cpp:120] Top shape: 100 3 32 32 (307200) | |
I0724 15:20:24.352723 26781 net.cpp:120] Top shape: 100 (100) | |
I0724 15:20:24.352730 26781 layer_factory.hpp:74] Creating layer conv1 | |
I0724 15:20:24.352751 26781 net.cpp:84] Creating Layer conv1 | |
I0724 15:20:24.352780 26781 net.cpp:380] conv1 <- data | |
I0724 15:20:24.352792 26781 net.cpp:338] conv1 -> conv1 | |
I0724 15:20:24.352802 26781 net.cpp:113] Setting up conv1 | |
I0724 15:20:24.415385 26781 net.cpp:120] Top shape: 100 32 32 32 (3276800) | |
I0724 15:20:24.415421 26781 layer_factory.hpp:74] Creating layer pool1 | |
I0724 15:20:24.415436 26781 net.cpp:84] Creating Layer pool1 | |
I0724 15:20:24.415441 26781 net.cpp:380] pool1 <- conv1 | |
I0724 15:20:24.415451 26781 net.cpp:338] pool1 -> pool1 | |
I0724 15:20:24.415459 26781 net.cpp:113] Setting up pool1 | |
I0724 15:20:24.415549 26781 net.cpp:120] Top shape: 100 32 16 16 (819200) | |
I0724 15:20:24.415565 26781 layer_factory.hpp:74] Creating layer relu1 | |
I0724 15:20:24.415580 26781 net.cpp:84] Creating Layer relu1 | |
I0724 15:20:24.415585 26781 net.cpp:380] relu1 <- pool1 | |
I0724 15:20:24.415590 26781 net.cpp:327] relu1 -> pool1 (in-place) | |
I0724 15:20:24.415594 26781 net.cpp:113] Setting up relu1 | |
I0724 15:20:24.415668 26781 net.cpp:120] Top shape: 100 32 16 16 (819200) | |
I0724 15:20:24.415675 26781 layer_factory.hpp:74] Creating layer conv2 | |
I0724 15:20:24.415702 26781 net.cpp:84] Creating Layer conv2 | |
I0724 15:20:24.415706 26781 net.cpp:380] conv2 <- pool1 | |
I0724 15:20:24.415711 26781 net.cpp:338] conv2 -> conv2 | |
I0724 15:20:24.415719 26781 net.cpp:113] Setting up conv2 | |
I0724 15:20:24.416695 26781 net.cpp:120] Top shape: 100 32 16 16 (819200) | |
I0724 15:20:24.416708 26781 layer_factory.hpp:74] Creating layer relu2 | |
I0724 15:20:24.416713 26781 net.cpp:84] Creating Layer relu2 | |
I0724 15:20:24.416718 26781 net.cpp:380] relu2 <- conv2 | |
I0724 15:20:24.416723 26781 net.cpp:327] relu2 -> conv2 (in-place) | |
I0724 15:20:24.416728 26781 net.cpp:113] Setting up relu2 | |
I0724 15:20:24.416787 26781 net.cpp:120] Top shape: 100 32 16 16 (819200) | |
I0724 15:20:24.416795 26781 layer_factory.hpp:74] Creating layer pool2 | |
I0724 15:20:24.416810 26781 net.cpp:84] Creating Layer pool2 | |
I0724 15:20:24.416813 26781 net.cpp:380] pool2 <- conv2 | |
I0724 15:20:24.416818 26781 net.cpp:338] pool2 -> pool2 | |
I0724 15:20:24.416832 26781 net.cpp:113] Setting up pool2 | |
I0724 15:20:24.416990 26781 net.cpp:120] Top shape: 100 32 8 8 (204800) | |
I0724 15:20:24.416997 26781 layer_factory.hpp:74] Creating layer conv3 | |
I0724 15:20:24.417004 26781 net.cpp:84] Creating Layer conv3 | |
I0724 15:20:24.417007 26781 net.cpp:380] conv3 <- pool2 | |
I0724 15:20:24.417013 26781 net.cpp:338] conv3 -> conv3 | |
I0724 15:20:24.417019 26781 net.cpp:113] Setting up conv3 | |
I0724 15:20:24.418752 26781 net.cpp:120] Top shape: 100 64 8 8 (409600) | |
I0724 15:20:24.418766 26781 layer_factory.hpp:74] Creating layer relu3 | |
I0724 15:20:24.418772 26781 net.cpp:84] Creating Layer relu3 | |
I0724 15:20:24.418776 26781 net.cpp:380] relu3 <- conv3 | |
I0724 15:20:24.418781 26781 net.cpp:327] relu3 -> conv3 (in-place) | |
I0724 15:20:24.418787 26781 net.cpp:113] Setting up relu3 | |
I0724 15:20:24.418829 26781 net.cpp:120] Top shape: 100 64 8 8 (409600) | |
I0724 15:20:24.418836 26781 layer_factory.hpp:74] Creating layer pool3 | |
I0724 15:20:24.418841 26781 net.cpp:84] Creating Layer pool3 | |
I0724 15:20:24.418845 26781 net.cpp:380] pool3 <- conv3 | |
I0724 15:20:24.418850 26781 net.cpp:338] pool3 -> pool3 | |
I0724 15:20:24.418855 26781 net.cpp:113] Setting up pool3 | |
I0724 15:20:24.418905 26781 net.cpp:120] Top shape: 100 64 4 4 (102400) | |
I0724 15:20:24.418912 26781 layer_factory.hpp:74] Creating layer ip1 | |
I0724 15:20:24.418921 26781 net.cpp:84] Creating Layer ip1 | |
I0724 15:20:24.418926 26781 net.cpp:380] ip1 <- pool3 | |
I0724 15:20:24.418932 26781 net.cpp:338] ip1 -> ip1 | |
I0724 15:20:24.418941 26781 net.cpp:113] Setting up ip1 | |
I0724 15:20:24.420871 26781 net.cpp:120] Top shape: 100 64 (6400) | |
I0724 15:20:24.420881 26781 layer_factory.hpp:74] Creating layer ip2 | |
I0724 15:20:24.420888 26781 net.cpp:84] Creating Layer ip2 | |
I0724 15:20:24.420892 26781 net.cpp:380] ip2 <- ip1 | |
I0724 15:20:24.420913 26781 net.cpp:338] ip2 -> ip2 | |
I0724 15:20:24.420920 26781 net.cpp:113] Setting up ip2 | |
I0724 15:20:24.420950 26781 net.cpp:120] Top shape: 100 10 (1000) | |
I0724 15:20:24.420959 26781 layer_factory.hpp:74] Creating layer loss | |
I0724 15:20:24.420967 26781 net.cpp:84] Creating Layer loss | |
I0724 15:20:24.420970 26781 net.cpp:380] loss <- ip2 | |
I0724 15:20:24.420975 26781 net.cpp:380] loss <- label | |
I0724 15:20:24.420982 26781 net.cpp:338] loss -> loss | |
I0724 15:20:24.420989 26781 net.cpp:113] Setting up loss | |
I0724 15:20:24.420994 26781 layer_factory.hpp:74] Creating layer loss | |
I0724 15:20:24.421064 26781 net.cpp:120] Top shape: (1) | |
I0724 15:20:24.421072 26781 net.cpp:122] with loss weight 1 | |
I0724 15:20:24.421088 26781 net.cpp:167] loss needs backward computation. | |
I0724 15:20:24.421092 26781 net.cpp:167] ip2 needs backward computation. | |
I0724 15:20:24.421097 26781 net.cpp:167] ip1 needs backward computation. | |
I0724 15:20:24.421099 26781 net.cpp:167] pool3 needs backward computation. | |
I0724 15:20:24.421103 26781 net.cpp:167] relu3 needs backward computation. | |
I0724 15:20:24.421105 26781 net.cpp:167] conv3 needs backward computation. | |
I0724 15:20:24.421110 26781 net.cpp:167] pool2 needs backward computation. | |
I0724 15:20:24.421113 26781 net.cpp:167] relu2 needs backward computation. | |
I0724 15:20:24.421116 26781 net.cpp:167] conv2 needs backward computation. | |
I0724 15:20:24.421119 26781 net.cpp:167] relu1 needs backward computation. | |
I0724 15:20:24.421123 26781 net.cpp:167] pool1 needs backward computation. | |
I0724 15:20:24.421126 26781 net.cpp:167] conv1 needs backward computation. | |
I0724 15:20:24.421129 26781 net.cpp:169] cifar does not need backward computation. | |
I0724 15:20:24.421133 26781 net.cpp:205] This network produces output loss | |
I0724 15:20:24.421142 26781 net.cpp:447] Collecting Learning Rate and Weight Decay. | |
I0724 15:20:24.421150 26781 net.cpp:217] Network initialization done. | |
I0724 15:20:24.421154 26781 net.cpp:218] Memory required for data: 31978804 | |
I0724 15:20:24.421432 26781 solver.cpp:154] Creating test net (#0) specified by net file: examples/cifar10/cifar10_quick_train_test.prototxt | |
I0724 15:20:24.421455 26781 net.cpp:257] The NetState phase (1) differed from the phase (0) specified by a rule in layer cifar | |
I0724 15:20:24.421550 26781 net.cpp:42] Initializing net from parameters: | |
name: "CIFAR10_quick" | |
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 | |
} | |
param { | |
lr_mult: 2 | |
} | |
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: "relu1" | |
type: "ReLU" | |
bottom: "pool1" | |
top: "pool1" | |
} | |
layer { | |
name: "conv2" | |
type: "Convolution" | |
bottom: "pool1" | |
top: "conv2" | |
param { | |
lr_mult: 1 | |
} | |
param { | |
lr_mult: 2 | |
} | |
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: "relu2" | |
type: "ReLU" | |
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 | |
} | |
param { | |
lr_mult: 2 | |
} | |
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: "relu3" | |
type: "ReLU" | |
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 | |
} | |
param { | |
lr_mult: 2 | |
} | |
inner_product_param { | |
num_output: 64 | |
weight_filler { | |
type: "gaussian" | |
std: 0.1 | |
} | |
bias_filler { | |
type: "constant" | |
} | |
} | |
} | |
layer { | |
name: "ip2" | |
type: "InnerProduct" | |
bottom: "ip1" | |
top: "ip2" | |
param { | |
lr_mult: 1 | |
} | |
param { | |
lr_mult: 2 | |
} | |
inner_product_param { | |
num_output: 10 | |
weight_filler { | |
type: "gaussian" | |
std: 0.1 | |
} | |
bias_filler { | |
type: "constant" | |
} | |
} | |
} | |
layer { | |
name: "accuracy" | |
type: "Accuracy" | |
bottom: "ip2" | |
bottom: "label" | |
top: "accuracy" | |
include { | |
phase: TEST | |
} | |
} | |
layer { | |
name: "loss" | |
type: "SoftmaxWithLoss" | |
bottom: "ip2" | |
bottom: "label" | |
top: "loss" | |
} | |
I0724 15:20:24.421622 26781 layer_factory.hpp:74] Creating layer cifar | |
I0724 15:20:24.421632 26781 net.cpp:84] Creating Layer cifar | |
I0724 15:20:24.421636 26781 net.cpp:338] cifar -> data | |
I0724 15:20:24.421644 26781 net.cpp:338] cifar -> label | |
I0724 15:20:24.421650 26781 net.cpp:113] Setting up cifar | |
I0724 15:20:24.421689 26781 db.cpp:34] Opened lmdb examples/cifar10/cifar10_test_lmdb | |
I0724 15:20:24.421721 26781 data_layer.cpp:67] output data size: 100,3,32,32 | |
I0724 15:20:24.421728 26781 data_transformer.cpp:22] Loading mean file from: examples/cifar10/mean.binaryproto | |
I0724 15:20:24.422838 26781 net.cpp:120] Top shape: 100 3 32 32 (307200) | |
I0724 15:20:24.422873 26781 net.cpp:120] Top shape: 100 (100) | |
I0724 15:20:24.422884 26781 layer_factory.hpp:74] Creating layer label_cifar_1_split | |
I0724 15:20:24.422909 26781 net.cpp:84] Creating Layer label_cifar_1_split | |
I0724 15:20:24.422924 26781 net.cpp:380] label_cifar_1_split <- label | |
I0724 15:20:24.422937 26781 net.cpp:338] label_cifar_1_split -> label_cifar_1_split_0 | |
I0724 15:20:24.422950 26781 net.cpp:338] label_cifar_1_split -> label_cifar_1_split_1 | |
I0724 15:20:24.422962 26781 net.cpp:113] Setting up label_cifar_1_split | |
I0724 15:20:24.423040 26781 net.cpp:120] Top shape: 100 (100) | |
I0724 15:20:24.423063 26781 net.cpp:120] Top shape: 100 (100) | |
I0724 15:20:24.423074 26781 layer_factory.hpp:74] Creating layer conv1 | |
I0724 15:20:24.423099 26781 net.cpp:84] Creating Layer conv1 | |
I0724 15:20:24.423111 26781 net.cpp:380] conv1 <- data | |
I0724 15:20:24.423126 26781 net.cpp:338] conv1 -> conv1 | |
I0724 15:20:24.423151 26781 net.cpp:113] Setting up conv1 | |
I0724 15:20:24.424074 26781 net.cpp:120] Top shape: 100 32 32 32 (3276800) | |
I0724 15:20:24.424105 26781 layer_factory.hpp:74] Creating layer pool1 | |
I0724 15:20:24.424116 26781 net.cpp:84] Creating Layer pool1 | |
I0724 15:20:24.424123 26781 net.cpp:380] pool1 <- conv1 | |
I0724 15:20:24.424186 26781 net.cpp:338] pool1 -> pool1 | |
I0724 15:20:24.424202 26781 net.cpp:113] Setting up pool1 | |
I0724 15:20:24.424496 26781 net.cpp:120] Top shape: 100 32 16 16 (819200) | |
I0724 15:20:24.424515 26781 layer_factory.hpp:74] Creating layer relu1 | |
I0724 15:20:24.424533 26781 net.cpp:84] Creating Layer relu1 | |
I0724 15:20:24.424541 26781 net.cpp:380] relu1 <- pool1 | |
I0724 15:20:24.424551 26781 net.cpp:327] relu1 -> pool1 (in-place) | |
I0724 15:20:24.424561 26781 net.cpp:113] Setting up relu1 | |
I0724 15:20:24.424649 26781 net.cpp:120] Top shape: 100 32 16 16 (819200) | |
I0724 15:20:24.424661 26781 layer_factory.hpp:74] Creating layer conv2 | |
I0724 15:20:24.424679 26781 net.cpp:84] Creating Layer conv2 | |
I0724 15:20:24.424686 26781 net.cpp:380] conv2 <- pool1 | |
I0724 15:20:24.424700 26781 net.cpp:338] conv2 -> conv2 | |
I0724 15:20:24.424715 26781 net.cpp:113] Setting up conv2 | |
I0724 15:20:24.426895 26781 net.cpp:120] Top shape: 100 32 16 16 (819200) | |
I0724 15:20:24.426990 26781 layer_factory.hpp:74] Creating layer relu2 | |
I0724 15:20:24.427023 26781 net.cpp:84] Creating Layer relu2 | |
I0724 15:20:24.427033 26781 net.cpp:380] relu2 <- conv2 | |
I0724 15:20:24.427085 26781 net.cpp:327] relu2 -> conv2 (in-place) | |
I0724 15:20:24.427103 26781 net.cpp:113] Setting up relu2 | |
I0724 15:20:24.427176 26781 net.cpp:120] Top shape: 100 32 16 16 (819200) | |
I0724 15:20:24.427186 26781 layer_factory.hpp:74] Creating layer pool2 | |
I0724 15:20:24.427203 26781 net.cpp:84] Creating Layer pool2 | |
I0724 15:20:24.427208 26781 net.cpp:380] pool2 <- conv2 | |
I0724 15:20:24.427213 26781 net.cpp:338] pool2 -> pool2 | |
I0724 15:20:24.427232 26781 net.cpp:113] Setting up pool2 | |
I0724 15:20:24.427284 26781 net.cpp:120] Top shape: 100 32 8 8 (204800) | |
I0724 15:20:24.427291 26781 layer_factory.hpp:74] Creating layer conv3 | |
I0724 15:20:24.427302 26781 net.cpp:84] Creating Layer conv3 | |
I0724 15:20:24.427307 26781 net.cpp:380] conv3 <- pool2 | |
I0724 15:20:24.427314 26781 net.cpp:338] conv3 -> conv3 | |
I0724 15:20:24.427325 26781 net.cpp:113] Setting up conv3 | |
I0724 15:20:24.429347 26781 net.cpp:120] Top shape: 100 64 8 8 (409600) | |
I0724 15:20:24.429378 26781 layer_factory.hpp:74] Creating layer relu3 | |
I0724 15:20:24.429389 26781 net.cpp:84] Creating Layer relu3 | |
I0724 15:20:24.429394 26781 net.cpp:380] relu3 <- conv3 | |
I0724 15:20:24.429404 26781 net.cpp:327] relu3 -> conv3 (in-place) | |
I0724 15:20:24.429414 26781 net.cpp:113] Setting up relu3 | |
I0724 15:20:24.429577 26781 net.cpp:120] Top shape: 100 64 8 8 (409600) | |
I0724 15:20:24.429586 26781 layer_factory.hpp:74] Creating layer pool3 | |
I0724 15:20:24.429594 26781 net.cpp:84] Creating Layer pool3 | |
I0724 15:20:24.429599 26781 net.cpp:380] pool3 <- conv3 | |
I0724 15:20:24.429605 26781 net.cpp:338] pool3 -> pool3 | |
I0724 15:20:24.429612 26781 net.cpp:113] Setting up pool3 | |
I0724 15:20:24.429666 26781 net.cpp:120] Top shape: 100 64 4 4 (102400) | |
I0724 15:20:24.429673 26781 layer_factory.hpp:74] Creating layer ip1 | |
I0724 15:20:24.429682 26781 net.cpp:84] Creating Layer ip1 | |
I0724 15:20:24.429685 26781 net.cpp:380] ip1 <- pool3 | |
I0724 15:20:24.429692 26781 net.cpp:338] ip1 -> ip1 | |
I0724 15:20:24.429702 26781 net.cpp:113] Setting up ip1 | |
I0724 15:20:24.431738 26781 net.cpp:120] Top shape: 100 64 (6400) | |
I0724 15:20:24.431764 26781 layer_factory.hpp:74] Creating layer ip2 | |
I0724 15:20:24.431777 26781 net.cpp:84] Creating Layer ip2 | |
I0724 15:20:24.431784 26781 net.cpp:380] ip2 <- ip1 | |
I0724 15:20:24.431797 26781 net.cpp:338] ip2 -> ip2 | |
I0724 15:20:24.431829 26781 net.cpp:113] Setting up ip2 | |
I0724 15:20:24.431875 26781 net.cpp:120] Top shape: 100 10 (1000) | |
I0724 15:20:24.431890 26781 layer_factory.hpp:74] Creating layer ip2_ip2_0_split | |
I0724 15:20:24.431906 26781 net.cpp:84] Creating Layer ip2_ip2_0_split | |
I0724 15:20:24.431916 26781 net.cpp:380] ip2_ip2_0_split <- ip2 | |
I0724 15:20:24.431927 26781 net.cpp:338] ip2_ip2_0_split -> ip2_ip2_0_split_0 | |
I0724 15:20:24.431941 26781 net.cpp:338] ip2_ip2_0_split -> ip2_ip2_0_split_1 | |
I0724 15:20:24.431947 26781 net.cpp:113] Setting up ip2_ip2_0_split | |
I0724 15:20:24.431957 26781 net.cpp:120] Top shape: 100 10 (1000) | |
I0724 15:20:24.431962 26781 net.cpp:120] Top shape: 100 10 (1000) | |
I0724 15:20:24.431967 26781 layer_factory.hpp:74] Creating layer accuracy | |
I0724 15:20:24.431978 26781 net.cpp:84] Creating Layer accuracy | |
I0724 15:20:24.431982 26781 net.cpp:380] accuracy <- ip2_ip2_0_split_0 | |
I0724 15:20:24.431988 26781 net.cpp:380] accuracy <- label_cifar_1_split_0 | |
I0724 15:20:24.431993 26781 net.cpp:338] accuracy -> accuracy | |
I0724 15:20:24.432003 26781 net.cpp:113] Setting up accuracy | |
I0724 15:20:24.432018 26781 net.cpp:120] Top shape: (1) | |
I0724 15:20:24.432023 26781 layer_factory.hpp:74] Creating layer loss | |
I0724 15:20:24.432030 26781 net.cpp:84] Creating Layer loss | |
I0724 15:20:24.432035 26781 net.cpp:380] loss <- ip2_ip2_0_split_1 | |
I0724 15:20:24.432040 26781 net.cpp:380] loss <- label_cifar_1_split_1 | |
I0724 15:20:24.432046 26781 net.cpp:338] loss -> loss | |
I0724 15:20:24.432054 26781 net.cpp:113] Setting up loss | |
I0724 15:20:24.432060 26781 layer_factory.hpp:74] Creating layer loss | |
I0724 15:20:24.432152 26781 net.cpp:120] Top shape: (1) | |
I0724 15:20:24.432159 26781 net.cpp:122] with loss weight 1 | |
I0724 15:20:24.432173 26781 net.cpp:167] loss needs backward computation. | |
I0724 15:20:24.432188 26781 net.cpp:169] accuracy does not need backward computation. | |
I0724 15:20:24.432193 26781 net.cpp:167] ip2_ip2_0_split needs backward computation. | |
I0724 15:20:24.432200 26781 net.cpp:167] ip2 needs backward computation. | |
I0724 15:20:24.432206 26781 net.cpp:167] ip1 needs backward computation. | |
I0724 15:20:24.432214 26781 net.cpp:167] pool3 needs backward computation. | |
I0724 15:20:24.432224 26781 net.cpp:167] relu3 needs backward computation. | |
I0724 15:20:24.432230 26781 net.cpp:167] conv3 needs backward computation. | |
I0724 15:20:24.432235 26781 net.cpp:167] pool2 needs backward computation. | |
I0724 15:20:24.432240 26781 net.cpp:167] relu2 needs backward computation. | |
I0724 15:20:24.432250 26781 net.cpp:167] conv2 needs backward computation. | |
I0724 15:20:24.432255 26781 net.cpp:167] relu1 needs backward computation. | |
I0724 15:20:24.432368 26781 net.cpp:167] pool1 needs backward computation. | |
I0724 15:20:24.432376 26781 net.cpp:167] conv1 needs backward computation. | |
I0724 15:20:24.432386 26781 net.cpp:169] label_cifar_1_split does not need backward computation. | |
I0724 15:20:24.432395 26781 net.cpp:169] cifar does not need backward computation. | |
I0724 15:20:24.432417 26781 net.cpp:205] This network produces output accuracy | |
I0724 15:20:24.432425 26781 net.cpp:205] This network produces output loss | |
I0724 15:20:24.432443 26781 net.cpp:447] Collecting Learning Rate and Weight Decay. | |
I0724 15:20:24.432459 26781 net.cpp:217] Network initialization done. | |
I0724 15:20:24.432468 26781 net.cpp:218] Memory required for data: 31987608 | |
I0724 15:20:24.432510 26781 solver.cpp:42] Solver scaffolding done. | |
I0724 15:20:24.432535 26781 solver.cpp:222] Solving CIFAR10_quick | |
I0724 15:20:24.432540 26781 solver.cpp:223] Learning Rate Policy: fixed | |
I0724 15:20:24.432546 26781 solver.cpp:266] Iteration 0, Testing net (#0) | |
I0724 15:20:24.876049 26781 solver.cpp:315] Test net output #0: accuracy = 0.1409 | |
I0724 15:20:24.876102 26781 solver.cpp:315] Test net output #1: loss = 2.30229 (* 1 = 2.30229 loss) | |
I0724 15:20:24.882349 26781 solver.cpp:189] Iteration 0, loss = 2.30285 | |
I0724 15:20:24.882375 26781 solver.cpp:204] Train net output #0: loss = 2.30285 (* 1 = 2.30285 loss) | |
I0724 15:20:24.882385 26781 solver.cpp:464] Iteration 0, lr = 0.001 | |
I0724 15:20:26.019618 26781 solver.cpp:189] Iteration 100, loss = 1.71262 | |
I0724 15:20:26.019659 26781 solver.cpp:204] Train net output #0: loss = 1.71262 (* 1 = 1.71262 loss) | |
I0724 15:20:26.019667 26781 solver.cpp:464] Iteration 100, lr = 0.001 | |
I0724 15:20:27.161996 26781 solver.cpp:189] Iteration 200, loss = 1.61527 | |
I0724 15:20:27.162050 26781 solver.cpp:204] Train net output #0: loss = 1.61527 (* 1 = 1.61527 loss) | |
I0724 15:20:27.162058 26781 solver.cpp:464] Iteration 200, lr = 0.001 | |
I0724 15:20:28.303331 26781 solver.cpp:189] Iteration 300, loss = 1.32747 | |
I0724 15:20:28.303442 26781 solver.cpp:204] Train net output #0: loss = 1.32747 (* 1 = 1.32747 loss) | |
I0724 15:20:28.303455 26781 solver.cpp:464] Iteration 300, lr = 0.001 | |
I0724 15:20:29.442226 26781 solver.cpp:189] Iteration 400, loss = 1.27856 | |
I0724 15:20:29.442315 26781 solver.cpp:204] Train net output #0: loss = 1.27856 (* 1 = 1.27856 loss) | |
I0724 15:20:29.442324 26781 solver.cpp:464] Iteration 400, lr = 0.001 | |
I0724 15:20:30.591828 26781 solver.cpp:266] Iteration 500, Testing net (#0) | |
I0724 15:20:30.981670 26781 solver.cpp:315] Test net output #0: accuracy = 0.5677 | |
I0724 15:20:30.981708 26781 solver.cpp:315] Test net output #1: loss = 1.23664 (* 1 = 1.23664 loss) | |
I0724 15:20:30.985673 26781 solver.cpp:189] Iteration 500, loss = 1.21874 | |
I0724 15:20:30.985690 26781 solver.cpp:204] Train net output #0: loss = 1.21874 (* 1 = 1.21874 loss) | |
I0724 15:20:30.985697 26781 solver.cpp:464] Iteration 500, lr = 0.001 | |
I0724 15:20:32.136443 26781 solver.cpp:189] Iteration 600, loss = 1.37051 | |
I0724 15:20:32.136487 26781 solver.cpp:204] Train net output #0: loss = 1.37051 (* 1 = 1.37051 loss) | |
I0724 15:20:32.136493 26781 solver.cpp:464] Iteration 600, lr = 0.001 | |
I0724 15:20:33.293810 26781 solver.cpp:189] Iteration 700, loss = 1.10339 | |
I0724 15:20:33.294037 26781 solver.cpp:204] Train net output #0: loss = 1.10339 (* 1 = 1.10339 loss) | |
I0724 15:20:33.294054 26781 solver.cpp:464] Iteration 700, lr = 0.001 | |
I0724 15:20:34.435643 26781 solver.cpp:189] Iteration 800, loss = 0.98746 | |
I0724 15:20:34.435688 26781 solver.cpp:204] Train net output #0: loss = 0.98746 (* 1 = 0.98746 loss) | |
I0724 15:20:34.435695 26781 solver.cpp:464] Iteration 800, lr = 0.001 | |
I0724 15:20:35.587226 26781 solver.cpp:189] Iteration 900, loss = 1.03267 | |
I0724 15:20:35.587270 26781 solver.cpp:204] Train net output #0: loss = 1.03267 (* 1 = 1.03267 loss) | |
I0724 15:20:35.587277 26781 solver.cpp:464] Iteration 900, lr = 0.001 | |
I0724 15:20:36.719394 26781 solver.cpp:266] Iteration 1000, Testing net (#0) | |
I0724 15:20:37.105604 26781 solver.cpp:315] Test net output #0: accuracy = 0.6296 | |
I0724 15:20:37.105654 26781 solver.cpp:315] Test net output #1: loss = 1.0718 (* 1 = 1.0718 loss) | |
I0724 15:20:37.110713 26781 solver.cpp:189] Iteration 1000, loss = 1.02915 | |
I0724 15:20:37.110757 26781 solver.cpp:204] Train net output #0: loss = 1.02915 (* 1 = 1.02915 loss) | |
I0724 15:20:37.110765 26781 solver.cpp:464] Iteration 1000, lr = 0.001 | |
I0724 15:20:38.247428 26781 solver.cpp:189] Iteration 1100, loss = 0.954241 | |
I0724 15:20:38.247462 26781 solver.cpp:204] Train net output #0: loss = 0.954241 (* 1 = 0.954241 loss) | |
I0724 15:20:38.247468 26781 solver.cpp:464] Iteration 1100, lr = 0.001 | |
I0724 15:20:39.385828 26781 solver.cpp:189] Iteration 1200, loss = 0.907719 | |
I0724 15:20:39.385861 26781 solver.cpp:204] Train net output #0: loss = 0.907719 (* 1 = 0.907719 loss) | |
I0724 15:20:39.385869 26781 solver.cpp:464] Iteration 1200, lr = 0.001 | |
I0724 15:20:40.525348 26781 solver.cpp:189] Iteration 1300, loss = 0.839362 | |
I0724 15:20:40.525449 26781 solver.cpp:204] Train net output #0: loss = 0.839362 (* 1 = 0.839362 loss) | |
I0724 15:20:40.525457 26781 solver.cpp:464] Iteration 1300, lr = 0.001 | |
I0724 15:20:41.679199 26781 solver.cpp:189] Iteration 1400, loss = 0.838737 | |
I0724 15:20:41.679234 26781 solver.cpp:204] Train net output #0: loss = 0.838737 (* 1 = 0.838737 loss) | |
I0724 15:20:41.679240 26781 solver.cpp:464] Iteration 1400, lr = 0.001 | |
I0724 15:20:42.811251 26781 solver.cpp:266] Iteration 1500, Testing net (#0) | |
I0724 15:20:43.193924 26781 solver.cpp:315] Test net output #0: accuracy = 0.663 | |
I0724 15:20:43.194023 26781 solver.cpp:315] Test net output #1: loss = 0.981802 (* 1 = 0.981802 loss) | |
I0724 15:20:43.198870 26781 solver.cpp:189] Iteration 1500, loss = 0.907493 | |
I0724 15:20:43.198968 26781 solver.cpp:204] Train net output #0: loss = 0.907493 (* 1 = 0.907493 loss) | |
I0724 15:20:43.198981 26781 solver.cpp:464] Iteration 1500, lr = 0.001 | |
I0724 15:20:44.336916 26781 solver.cpp:189] Iteration 1600, loss = 0.904637 | |
I0724 15:20:44.336961 26781 solver.cpp:204] Train net output #0: loss = 0.904637 (* 1 = 0.904637 loss) | |
I0724 15:20:44.336968 26781 solver.cpp:464] Iteration 1600, lr = 0.001 | |
I0724 15:20:45.474933 26781 solver.cpp:189] Iteration 1700, loss = 0.809851 | |
I0724 15:20:45.474968 26781 solver.cpp:204] Train net output #0: loss = 0.809851 (* 1 = 0.809851 loss) | |
I0724 15:20:45.474977 26781 solver.cpp:464] Iteration 1700, lr = 0.001 | |
I0724 15:20:46.617063 26781 solver.cpp:189] Iteration 1800, loss = 0.770444 | |
I0724 15:20:46.617099 26781 solver.cpp:204] Train net output #0: loss = 0.770444 (* 1 = 0.770444 loss) | |
I0724 15:20:46.617105 26781 solver.cpp:464] Iteration 1800, lr = 0.001 | |
I0724 15:20:47.766064 26781 solver.cpp:189] Iteration 1900, loss = 0.786517 | |
I0724 15:20:47.766163 26781 solver.cpp:204] Train net output #0: loss = 0.786517 (* 1 = 0.786517 loss) | |
I0724 15:20:47.766177 26781 solver.cpp:464] Iteration 1900, lr = 0.001 | |
I0724 15:20:48.903381 26781 solver.cpp:266] Iteration 2000, Testing net (#0) | |
I0724 15:20:49.290642 26781 solver.cpp:315] Test net output #0: accuracy = 0.68 | |
I0724 15:20:49.290671 26781 solver.cpp:315] Test net output #1: loss = 0.91725 (* 1 = 0.91725 loss) | |
I0724 15:20:49.294699 26781 solver.cpp:189] Iteration 2000, loss = 0.77896 | |
I0724 15:20:49.294729 26781 solver.cpp:204] Train net output #0: loss = 0.77896 (* 1 = 0.77896 loss) | |
I0724 15:20:49.294736 26781 solver.cpp:464] Iteration 2000, lr = 0.001 | |
I0724 15:20:50.434298 26781 solver.cpp:189] Iteration 2100, loss = 0.845264 | |
I0724 15:20:50.434342 26781 solver.cpp:204] Train net output #0: loss = 0.845264 (* 1 = 0.845264 loss) | |
I0724 15:20:50.434350 26781 solver.cpp:464] Iteration 2100, lr = 0.001 | |
I0724 15:20:51.573161 26781 solver.cpp:189] Iteration 2200, loss = 0.752443 | |
I0724 15:20:51.573199 26781 solver.cpp:204] Train net output #0: loss = 0.752443 (* 1 = 0.752443 loss) | |
I0724 15:20:51.573205 26781 solver.cpp:464] Iteration 2200, lr = 0.001 | |
I0724 15:20:52.711212 26781 solver.cpp:189] Iteration 2300, loss = 0.664528 | |
I0724 15:20:52.711246 26781 solver.cpp:204] Train net output #0: loss = 0.664528 (* 1 = 0.664528 loss) | |
I0724 15:20:52.711251 26781 solver.cpp:464] Iteration 2300, lr = 0.001 | |
I0724 15:20:53.851538 26781 solver.cpp:189] Iteration 2400, loss = 0.700644 | |
I0724 15:20:53.851573 26781 solver.cpp:204] Train net output #0: loss = 0.700644 (* 1 = 0.700644 loss) | |
I0724 15:20:53.851589 26781 solver.cpp:464] Iteration 2400, lr = 0.001 | |
I0724 15:20:54.979451 26781 solver.cpp:266] Iteration 2500, Testing net (#0) | |
I0724 15:20:55.364724 26781 solver.cpp:315] Test net output #0: accuracy = 0.6927 | |
I0724 15:20:55.364758 26781 solver.cpp:315] Test net output #1: loss = 0.891418 (* 1 = 0.891418 loss) | |
I0724 15:20:55.368755 26781 solver.cpp:189] Iteration 2500, loss = 0.723417 | |
I0724 15:20:55.368772 26781 solver.cpp:204] Train net output #0: loss = 0.723417 (* 1 = 0.723417 loss) | |
I0724 15:20:55.368779 26781 solver.cpp:464] Iteration 2500, lr = 0.001 | |
I0724 15:20:56.509495 26781 solver.cpp:189] Iteration 2600, loss = 0.731871 | |
I0724 15:20:56.509529 26781 solver.cpp:204] Train net output #0: loss = 0.731871 (* 1 = 0.731871 loss) | |
I0724 15:20:56.509537 26781 solver.cpp:464] Iteration 2600, lr = 0.001 | |
I0724 15:20:57.650527 26781 solver.cpp:189] Iteration 2700, loss = 0.700396 | |
I0724 15:20:57.650605 26781 solver.cpp:204] Train net output #0: loss = 0.700396 (* 1 = 0.700396 loss) | |
I0724 15:20:57.650614 26781 solver.cpp:464] Iteration 2700, lr = 0.001 | |
I0724 15:20:58.790457 26781 solver.cpp:189] Iteration 2800, loss = 0.623085 | |
I0724 15:20:58.790501 26781 solver.cpp:204] Train net output #0: loss = 0.623085 (* 1 = 0.623085 loss) | |
I0724 15:20:58.790508 26781 solver.cpp:464] Iteration 2800, lr = 0.001 | |
I0724 15:20:59.931192 26781 solver.cpp:189] Iteration 2900, loss = 0.705873 | |
I0724 15:20:59.931236 26781 solver.cpp:204] Train net output #0: loss = 0.705873 (* 1 = 0.705873 loss) | |
I0724 15:20:59.931242 26781 solver.cpp:464] Iteration 2900, lr = 0.001 | |
I0724 15:21:01.062194 26781 solver.cpp:266] Iteration 3000, Testing net (#0) | |
I0724 15:21:01.447918 26781 solver.cpp:315] Test net output #0: accuracy = 0.6994 | |
I0724 15:21:01.448011 26781 solver.cpp:315] Test net output #1: loss = 0.880686 (* 1 = 0.880686 loss) | |
I0724 15:21:01.453021 26781 solver.cpp:189] Iteration 3000, loss = 0.691863 | |
I0724 15:21:01.453069 26781 solver.cpp:204] Train net output #0: loss = 0.691863 (* 1 = 0.691863 loss) | |
I0724 15:21:01.453078 26781 solver.cpp:464] Iteration 3000, lr = 0.001 | |
I0724 15:21:02.595587 26781 solver.cpp:189] Iteration 3100, loss = 0.692409 | |
I0724 15:21:02.595631 26781 solver.cpp:204] Train net output #0: loss = 0.692409 (* 1 = 0.692409 loss) | |
I0724 15:21:02.595638 26781 solver.cpp:464] Iteration 3100, lr = 0.001 | |
I0724 15:21:03.735291 26781 solver.cpp:189] Iteration 3200, loss = 0.689493 | |
I0724 15:21:03.735324 26781 solver.cpp:204] Train net output #0: loss = 0.689493 (* 1 = 0.689493 loss) | |
I0724 15:21:03.735332 26781 solver.cpp:464] Iteration 3200, lr = 0.001 | |
I0724 15:21:04.873982 26781 solver.cpp:189] Iteration 3300, loss = 0.581388 | |
I0724 15:21:04.874025 26781 solver.cpp:204] Train net output #0: loss = 0.581388 (* 1 = 0.581388 loss) | |
I0724 15:21:04.874033 26781 solver.cpp:464] Iteration 3300, lr = 0.001 | |
I0724 15:21:06.017020 26781 solver.cpp:189] Iteration 3400, loss = 0.67786 | |
I0724 15:21:06.017074 26781 solver.cpp:204] Train net output #0: loss = 0.67786 (* 1 = 0.67786 loss) | |
I0724 15:21:06.017083 26781 solver.cpp:464] Iteration 3400, lr = 0.001 | |
I0724 15:21:07.144655 26781 solver.cpp:266] Iteration 3500, Testing net (#0) | |
I0724 15:21:07.530313 26781 solver.cpp:315] Test net output #0: accuracy = 0.7069 | |
I0724 15:21:07.530344 26781 solver.cpp:315] Test net output #1: loss = 0.869211 (* 1 = 0.869211 loss) | |
I0724 15:21:07.534332 26781 solver.cpp:189] Iteration 3500, loss = 0.660259 | |
I0724 15:21:07.534349 26781 solver.cpp:204] Train net output #0: loss = 0.660259 (* 1 = 0.660259 loss) | |
I0724 15:21:07.534356 26781 solver.cpp:464] Iteration 3500, lr = 0.001 | |
I0724 15:21:08.672732 26781 solver.cpp:189] Iteration 3600, loss = 0.667279 | |
I0724 15:21:08.672802 26781 solver.cpp:204] Train net output #0: loss = 0.667279 (* 1 = 0.667279 loss) | |
I0724 15:21:08.672811 26781 solver.cpp:464] Iteration 3600, lr = 0.001 | |
I0724 15:21:09.810492 26781 solver.cpp:189] Iteration 3700, loss = 0.671383 | |
I0724 15:21:09.810528 26781 solver.cpp:204] Train net output #0: loss = 0.671383 (* 1 = 0.671383 loss) | |
I0724 15:21:09.810534 26781 solver.cpp:464] Iteration 3700, lr = 0.001 | |
I0724 15:21:10.951683 26781 solver.cpp:189] Iteration 3800, loss = 0.565697 | |
I0724 15:21:10.951776 26781 solver.cpp:204] Train net output #0: loss = 0.565697 (* 1 = 0.565697 loss) | |
I0724 15:21:10.951787 26781 solver.cpp:464] Iteration 3800, lr = 0.001 | |
I0724 15:21:12.089732 26781 solver.cpp:189] Iteration 3900, loss = 0.670047 | |
I0724 15:21:12.089776 26781 solver.cpp:204] Train net output #0: loss = 0.670047 (* 1 = 0.670047 loss) | |
I0724 15:21:12.089784 26781 solver.cpp:464] Iteration 3900, lr = 0.001 | |
I0724 15:21:13.224203 26781 solver.cpp:334] Snapshotting to examples/cifar10/cifar10_quick_iter_4000.caffemodel | |
I0724 15:21:13.227545 26781 solver.cpp:342] Snapshotting solver state to examples/cifar10/cifar10_quick_iter_4000.solverstate | |
I0724 15:21:13.233026 26781 solver.cpp:248] Iteration 4000, loss = 0.604699 | |
I0724 15:21:13.233064 26781 solver.cpp:266] Iteration 4000, Testing net (#0) | |
I0724 15:21:13.614272 26781 solver.cpp:315] Test net output #0: accuracy = 0.7148 | |
I0724 15:21:13.614313 26781 solver.cpp:315] Test net output #1: loss = 0.843019 (* 1 = 0.843019 loss) | |
I0724 15:21:13.614320 26781 solver.cpp:253] Optimization Done. | |
I0724 15:21:13.614323 26781 caffe.cpp:134] Optimization Done. | |
I0724 15:21:13.689227 31785 caffe.cpp:113] Use GPU with device ID 0 | |
I0724 15:21:13.846545 31785 caffe.cpp:121] Starting Optimization | |
I0724 15:21:13.846644 31785 solver.cpp:32] Initializing solver from parameters: | |
test_iter: 100 | |
test_interval: 500 | |
base_lr: 0.0001 | |
display: 100 | |
max_iter: 5000 | |
lr_policy: "fixed" | |
momentum: 0.9 | |
weight_decay: 0.004 | |
snapshot: 5000 | |
snapshot_prefix: "examples/cifar10/cifar10_quick" | |
solver_mode: GPU | |
net: "examples/cifar10/cifar10_quick_train_test.prototxt" | |
I0724 15:21:13.846679 31785 solver.cpp:70] Creating training net from net file: examples/cifar10/cifar10_quick_train_test.prototxt | |
I0724 15:21:13.846997 31785 net.cpp:257] The NetState phase (0) differed from the phase (1) specified by a rule in layer cifar | |
I0724 15:21:13.847019 31785 net.cpp:257] The NetState phase (0) differed from the phase (1) specified by a rule in layer accuracy | |
I0724 15:21:13.847129 31785 net.cpp:42] Initializing net from parameters: | |
name: "CIFAR10_quick" | |
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 | |
} | |
param { | |
lr_mult: 2 | |
} | |
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: "relu1" | |
type: "ReLU" | |
bottom: "pool1" | |
top: "pool1" | |
} | |
layer { | |
name: "conv2" | |
type: "Convolution" | |
bottom: "pool1" | |
top: "conv2" | |
param { | |
lr_mult: 1 | |
} | |
param { | |
lr_mult: 2 | |
} | |
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: "relu2" | |
type: "ReLU" | |
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 | |
} | |
param { | |
lr_mult: 2 | |
} | |
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: "relu3" | |
type: "ReLU" | |
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 | |
} | |
param { | |
lr_mult: 2 | |
} | |
inner_product_param { | |
num_output: 64 | |
weight_filler { | |
type: "gaussian" | |
std: 0.1 | |
} | |
bias_filler { | |
type: "constant" | |
} | |
} | |
} | |
layer { | |
name: "ip2" | |
type: "InnerProduct" | |
bottom: "ip1" | |
top: "ip2" | |
param { | |
lr_mult: 1 | |
} | |
param { | |
lr_mult: 2 | |
} | |
inner_product_param { | |
num_output: 10 | |
weight_filler { | |
type: "gaussian" | |
std: 0.1 | |
} | |
bias_filler { | |
type: "constant" | |
} | |
} | |
} | |
layer { | |
name: "loss" | |
type: "SoftmaxWithLoss" | |
bottom: "ip2" | |
bottom: "label" | |
top: "loss" | |
} | |
I0724 15:21:13.847203 31785 layer_factory.hpp:74] Creating layer cifar | |
I0724 15:21:13.847221 31785 net.cpp:84] Creating Layer cifar | |
I0724 15:21:13.847228 31785 net.cpp:338] cifar -> data | |
I0724 15:21:13.847255 31785 net.cpp:338] cifar -> label | |
I0724 15:21:13.847265 31785 net.cpp:113] Setting up cifar | |
I0724 15:21:13.847321 31785 db.cpp:34] Opened lmdb examples/cifar10/cifar10_train_lmdb | |
I0724 15:21:13.847365 31785 data_layer.cpp:67] output data size: 100,3,32,32 | |
I0724 15:21:13.847374 31785 data_transformer.cpp:22] Loading mean file from: examples/cifar10/mean.binaryproto | |
I0724 15:21:13.847839 31785 net.cpp:120] Top shape: 100 3 32 32 (307200) | |
I0724 15:21:13.847847 31785 net.cpp:120] Top shape: 100 (100) | |
I0724 15:21:13.847852 31785 layer_factory.hpp:74] Creating layer conv1 | |
I0724 15:21:13.847863 31785 net.cpp:84] Creating Layer conv1 | |
I0724 15:21:13.847868 31785 net.cpp:380] conv1 <- data | |
I0724 15:21:13.847877 31785 net.cpp:338] conv1 -> conv1 | |
I0724 15:21:13.847887 31785 net.cpp:113] Setting up conv1 | |
I0724 15:21:13.894073 31785 net.cpp:120] Top shape: 100 32 32 32 (3276800) | |
I0724 15:21:13.894109 31785 layer_factory.hpp:74] Creating layer pool1 | |
I0724 15:21:13.894125 31785 net.cpp:84] Creating Layer pool1 | |
I0724 15:21:13.894130 31785 net.cpp:380] pool1 <- conv1 | |
I0724 15:21:13.894148 31785 net.cpp:338] pool1 -> pool1 | |
I0724 15:21:13.894158 31785 net.cpp:113] Setting up pool1 | |
I0724 15:21:13.894232 31785 net.cpp:120] Top shape: 100 32 16 16 (819200) | |
I0724 15:21:13.894240 31785 layer_factory.hpp:74] Creating layer relu1 | |
I0724 15:21:13.894246 31785 net.cpp:84] Creating Layer relu1 | |
I0724 15:21:13.894250 31785 net.cpp:380] relu1 <- pool1 | |
I0724 15:21:13.894264 31785 net.cpp:327] relu1 -> pool1 (in-place) | |
I0724 15:21:13.894270 31785 net.cpp:113] Setting up relu1 | |
I0724 15:21:13.894337 31785 net.cpp:120] Top shape: 100 32 16 16 (819200) | |
I0724 15:21:13.894345 31785 layer_factory.hpp:74] Creating layer conv2 | |
I0724 15:21:13.894364 31785 net.cpp:84] Creating Layer conv2 | |
I0724 15:21:13.894368 31785 net.cpp:380] conv2 <- pool1 | |
I0724 15:21:13.894373 31785 net.cpp:338] conv2 -> conv2 | |
I0724 15:21:13.894381 31785 net.cpp:113] Setting up conv2 | |
I0724 15:21:13.895388 31785 net.cpp:120] Top shape: 100 32 16 16 (819200) | |
I0724 15:21:13.895411 31785 layer_factory.hpp:74] Creating layer relu2 | |
I0724 15:21:13.895416 31785 net.cpp:84] Creating Layer relu2 | |
I0724 15:21:13.895421 31785 net.cpp:380] relu2 <- conv2 | |
I0724 15:21:13.895426 31785 net.cpp:327] relu2 -> conv2 (in-place) | |
I0724 15:21:13.895432 31785 net.cpp:113] Setting up relu2 | |
I0724 15:21:13.895480 31785 net.cpp:120] Top shape: 100 32 16 16 (819200) | |
I0724 15:21:13.895488 31785 layer_factory.hpp:74] Creating layer pool2 | |
I0724 15:21:13.895493 31785 net.cpp:84] Creating Layer pool2 | |
I0724 15:21:13.895496 31785 net.cpp:380] pool2 <- conv2 | |
I0724 15:21:13.895503 31785 net.cpp:338] pool2 -> pool2 | |
I0724 15:21:13.895509 31785 net.cpp:113] Setting up pool2 | |
I0724 15:21:13.895668 31785 net.cpp:120] Top shape: 100 32 8 8 (204800) | |
I0724 15:21:13.895675 31785 layer_factory.hpp:74] Creating layer conv3 | |
I0724 15:21:13.895683 31785 net.cpp:84] Creating Layer conv3 | |
I0724 15:21:13.895685 31785 net.cpp:380] conv3 <- pool2 | |
I0724 15:21:13.895691 31785 net.cpp:338] conv3 -> conv3 | |
I0724 15:21:13.895697 31785 net.cpp:113] Setting up conv3 | |
I0724 15:21:13.897600 31785 net.cpp:120] Top shape: 100 64 8 8 (409600) | |
I0724 15:21:13.897620 31785 layer_factory.hpp:74] Creating layer relu3 | |
I0724 15:21:13.897629 31785 net.cpp:84] Creating Layer relu3 | |
I0724 15:21:13.897632 31785 net.cpp:380] relu3 <- conv3 | |
I0724 15:21:13.897639 31785 net.cpp:327] relu3 -> conv3 (in-place) | |
I0724 15:21:13.897644 31785 net.cpp:113] Setting up relu3 | |
I0724 15:21:13.897716 31785 net.cpp:120] Top shape: 100 64 8 8 (409600) | |
I0724 15:21:13.897722 31785 layer_factory.hpp:74] Creating layer pool3 | |
I0724 15:21:13.897732 31785 net.cpp:84] Creating Layer pool3 | |
I0724 15:21:13.897737 31785 net.cpp:380] pool3 <- conv3 | |
I0724 15:21:13.897744 31785 net.cpp:338] pool3 -> pool3 | |
I0724 15:21:13.897752 31785 net.cpp:113] Setting up pool3 | |
I0724 15:21:13.897816 31785 net.cpp:120] Top shape: 100 64 4 4 (102400) | |
I0724 15:21:13.897825 31785 layer_factory.hpp:74] Creating layer ip1 | |
I0724 15:21:13.897855 31785 net.cpp:84] Creating Layer ip1 | |
I0724 15:21:13.897858 31785 net.cpp:380] ip1 <- pool3 | |
I0724 15:21:13.897866 31785 net.cpp:338] ip1 -> ip1 | |
I0724 15:21:13.897881 31785 net.cpp:113] Setting up ip1 | |
I0724 15:21:13.899966 31785 net.cpp:120] Top shape: 100 64 (6400) | |
I0724 15:21:13.900015 31785 layer_factory.hpp:74] Creating layer ip2 | |
I0724 15:21:13.900027 31785 net.cpp:84] Creating Layer ip2 | |
I0724 15:21:13.900032 31785 net.cpp:380] ip2 <- ip1 | |
I0724 15:21:13.900070 31785 net.cpp:338] ip2 -> ip2 | |
I0724 15:21:13.900082 31785 net.cpp:113] Setting up ip2 | |
I0724 15:21:13.900115 31785 net.cpp:120] Top shape: 100 10 (1000) | |
I0724 15:21:13.900127 31785 layer_factory.hpp:74] Creating layer loss | |
I0724 15:21:13.900138 31785 net.cpp:84] Creating Layer loss | |
I0724 15:21:13.900142 31785 net.cpp:380] loss <- ip2 | |
I0724 15:21:13.900146 31785 net.cpp:380] loss <- label | |
I0724 15:21:13.900156 31785 net.cpp:338] loss -> loss | |
I0724 15:21:13.900164 31785 net.cpp:113] Setting up loss | |
I0724 15:21:13.900171 31785 layer_factory.hpp:74] Creating layer loss | |
I0724 15:21:13.900398 31785 net.cpp:120] Top shape: (1) | |
I0724 15:21:13.900408 31785 net.cpp:122] with loss weight 1 | |
I0724 15:21:13.900472 31785 net.cpp:167] loss needs backward computation. | |
I0724 15:21:13.900480 31785 net.cpp:167] ip2 needs backward computation. | |
I0724 15:21:13.900482 31785 net.cpp:167] ip1 needs backward computation. | |
I0724 15:21:13.900487 31785 net.cpp:167] pool3 needs backward computation. | |
I0724 15:21:13.900491 31785 net.cpp:167] relu3 needs backward computation. | |
I0724 15:21:13.900495 31785 net.cpp:167] conv3 needs backward computation. | |
I0724 15:21:13.900502 31785 net.cpp:167] pool2 needs backward computation. | |
I0724 15:21:13.900507 31785 net.cpp:167] relu2 needs backward computation. | |
I0724 15:21:13.900511 31785 net.cpp:167] conv2 needs backward computation. | |
I0724 15:21:13.900514 31785 net.cpp:167] relu1 needs backward computation. | |
I0724 15:21:13.900518 31785 net.cpp:167] pool1 needs backward computation. | |
I0724 15:21:13.900523 31785 net.cpp:167] conv1 needs backward computation. | |
I0724 15:21:13.900527 31785 net.cpp:169] cifar does not need backward computation. | |
I0724 15:21:13.900532 31785 net.cpp:205] This network produces output loss | |
I0724 15:21:13.900547 31785 net.cpp:447] Collecting Learning Rate and Weight Decay. | |
I0724 15:21:13.900558 31785 net.cpp:217] Network initialization done. | |
I0724 15:21:13.900563 31785 net.cpp:218] Memory required for data: 31978804 | |
I0724 15:21:13.901029 31785 solver.cpp:154] Creating test net (#0) specified by net file: examples/cifar10/cifar10_quick_train_test.prototxt | |
I0724 15:21:13.901087 31785 net.cpp:257] The NetState phase (1) differed from the phase (0) specified by a rule in layer cifar | |
I0724 15:21:13.901259 31785 net.cpp:42] Initializing net from parameters: | |
name: "CIFAR10_quick" | |
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 | |
} | |
param { | |
lr_mult: 2 | |
} | |
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: "relu1" | |
type: "ReLU" | |
bottom: "pool1" | |
top: "pool1" | |
} | |
layer { | |
name: "conv2" | |
type: "Convolution" | |
bottom: "pool1" | |
top: "conv2" | |
param { | |
lr_mult: 1 | |
} | |
param { | |
lr_mult: 2 | |
} | |
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: "relu2" | |
type: "ReLU" | |
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 | |
} | |
param { | |
lr_mult: 2 | |
} | |
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: "relu3" | |
type: "ReLU" | |
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 | |
} | |
param { | |
lr_mult: 2 | |
} | |
inner_product_param { | |
num_output: 64 | |
weight_filler { | |
type: "gaussian" | |
std: 0.1 | |
} | |
bias_filler { | |
type: "constant" | |
} | |
} | |
} | |
layer { | |
name: "ip2" | |
type: "InnerProduct" | |
bottom: "ip1" | |
top: "ip2" | |
param { | |
lr_mult: 1 | |
} | |
param { | |
lr_mult: 2 | |
} | |
inner_product_param { | |
num_output: 10 | |
weight_filler { | |
type: "gaussian" | |
std: 0.1 | |
} | |
bias_filler { | |
type: "constant" | |
} | |
} | |
} | |
layer { | |
name: "accuracy" | |
type: "Accuracy" | |
bottom: "ip2" | |
bottom: "label" | |
top: "accuracy" | |
include { | |
phase: TEST | |
} | |
} | |
layer { | |
name: "loss" | |
type: "SoftmaxWithLoss" | |
bottom: "ip2" | |
bottom: "label" | |
top: "loss" | |
} | |
I0724 15:21:13.901396 31785 layer_factory.hpp:74] Creating layer cifar | |
I0724 15:21:13.901414 31785 net.cpp:84] Creating Layer cifar | |
I0724 15:21:13.901424 31785 net.cpp:338] cifar -> data | |
I0724 15:21:13.901450 31785 net.cpp:338] cifar -> label | |
I0724 15:21:13.901465 31785 net.cpp:113] Setting up cifar | |
I0724 15:21:13.901571 31785 db.cpp:34] Opened lmdb examples/cifar10/cifar10_test_lmdb | |
I0724 15:21:13.901623 31785 data_layer.cpp:67] output data size: 100,3,32,32 | |
I0724 15:21:13.901631 31785 data_transformer.cpp:22] Loading mean file from: examples/cifar10/mean.binaryproto | |
I0724 15:21:13.902997 31785 net.cpp:120] Top shape: 100 3 32 32 (307200) | |
I0724 15:21:13.903058 31785 net.cpp:120] Top shape: 100 (100) | |
I0724 15:21:13.903071 31785 layer_factory.hpp:74] Creating layer label_cifar_1_split | |
I0724 15:21:13.903090 31785 net.cpp:84] Creating Layer label_cifar_1_split | |
I0724 15:21:13.903095 31785 net.cpp:380] label_cifar_1_split <- label | |
I0724 15:21:13.903103 31785 net.cpp:338] label_cifar_1_split -> label_cifar_1_split_0 | |
I0724 15:21:13.903116 31785 net.cpp:338] label_cifar_1_split -> label_cifar_1_split_1 | |
I0724 15:21:13.903126 31785 net.cpp:113] Setting up label_cifar_1_split | |
I0724 15:21:13.903142 31785 net.cpp:120] Top shape: 100 (100) | |
I0724 15:21:13.903172 31785 net.cpp:120] Top shape: 100 (100) | |
I0724 15:21:13.903178 31785 layer_factory.hpp:74] Creating layer conv1 | |
I0724 15:21:13.903192 31785 net.cpp:84] Creating Layer conv1 | |
I0724 15:21:13.903197 31785 net.cpp:380] conv1 <- data | |
I0724 15:21:13.903234 31785 net.cpp:338] conv1 -> conv1 | |
I0724 15:21:13.903277 31785 net.cpp:113] Setting up conv1 | |
I0724 15:21:13.903806 31785 net.cpp:120] Top shape: 100 32 32 32 (3276800) | |
I0724 15:21:13.903820 31785 layer_factory.hpp:74] Creating layer pool1 | |
I0724 15:21:13.903828 31785 net.cpp:84] Creating Layer pool1 | |
I0724 15:21:13.903832 31785 net.cpp:380] pool1 <- conv1 | |
I0724 15:21:13.903837 31785 net.cpp:338] pool1 -> pool1 | |
I0724 15:21:13.903843 31785 net.cpp:113] Setting up pool1 | |
I0724 15:21:13.904022 31785 net.cpp:120] Top shape: 100 32 16 16 (819200) | |
I0724 15:21:13.904031 31785 layer_factory.hpp:74] Creating layer relu1 | |
I0724 15:21:13.904039 31785 net.cpp:84] Creating Layer relu1 | |
I0724 15:21:13.904043 31785 net.cpp:380] relu1 <- pool1 | |
I0724 15:21:13.904057 31785 net.cpp:327] relu1 -> pool1 (in-place) | |
I0724 15:21:13.904065 31785 net.cpp:113] Setting up relu1 | |
I0724 15:21:13.904126 31785 net.cpp:120] Top shape: 100 32 16 16 (819200) | |
I0724 15:21:13.904139 31785 layer_factory.hpp:74] Creating layer conv2 | |
I0724 15:21:13.904153 31785 net.cpp:84] Creating Layer conv2 | |
I0724 15:21:13.904160 31785 net.cpp:380] conv2 <- pool1 | |
I0724 15:21:13.904170 31785 net.cpp:338] conv2 -> conv2 | |
I0724 15:21:13.904177 31785 net.cpp:113] Setting up conv2 | |
I0724 15:21:13.905239 31785 net.cpp:120] Top shape: 100 32 16 16 (819200) | |
I0724 15:21:13.905256 31785 layer_factory.hpp:74] Creating layer relu2 | |
I0724 15:21:13.905262 31785 net.cpp:84] Creating Layer relu2 | |
I0724 15:21:13.905266 31785 net.cpp:380] relu2 <- conv2 | |
I0724 15:21:13.905283 31785 net.cpp:327] relu2 -> conv2 (in-place) | |
I0724 15:21:13.905289 31785 net.cpp:113] Setting up relu2 | |
I0724 15:21:13.905339 31785 net.cpp:120] Top shape: 100 32 16 16 (819200) | |
I0724 15:21:13.905345 31785 layer_factory.hpp:74] Creating layer pool2 | |
I0724 15:21:13.905352 31785 net.cpp:84] Creating Layer pool2 | |
I0724 15:21:13.905356 31785 net.cpp:380] pool2 <- conv2 | |
I0724 15:21:13.905361 31785 net.cpp:338] pool2 -> pool2 | |
I0724 15:21:13.905366 31785 net.cpp:113] Setting up pool2 | |
I0724 15:21:13.905414 31785 net.cpp:120] Top shape: 100 32 8 8 (204800) | |
I0724 15:21:13.905421 31785 layer_factory.hpp:74] Creating layer conv3 | |
I0724 15:21:13.905428 31785 net.cpp:84] Creating Layer conv3 | |
I0724 15:21:13.905432 31785 net.cpp:380] conv3 <- pool2 | |
I0724 15:21:13.905437 31785 net.cpp:338] conv3 -> conv3 | |
I0724 15:21:13.905444 31785 net.cpp:113] Setting up conv3 | |
I0724 15:21:13.907307 31785 net.cpp:120] Top shape: 100 64 8 8 (409600) | |
I0724 15:21:13.907331 31785 layer_factory.hpp:74] Creating layer relu3 | |
I0724 15:21:13.907338 31785 net.cpp:84] Creating Layer relu3 | |
I0724 15:21:13.907342 31785 net.cpp:380] relu3 <- conv3 | |
I0724 15:21:13.907349 31785 net.cpp:327] relu3 -> conv3 (in-place) | |
I0724 15:21:13.907354 31785 net.cpp:113] Setting up relu3 | |
I0724 15:21:13.907536 31785 net.cpp:120] Top shape: 100 64 8 8 (409600) | |
I0724 15:21:13.907544 31785 layer_factory.hpp:74] Creating layer pool3 | |
I0724 15:21:13.907551 31785 net.cpp:84] Creating Layer pool3 | |
I0724 15:21:13.907554 31785 net.cpp:380] pool3 <- conv3 | |
I0724 15:21:13.907559 31785 net.cpp:338] pool3 -> pool3 | |
I0724 15:21:13.907565 31785 net.cpp:113] Setting up pool3 | |
I0724 15:21:13.907613 31785 net.cpp:120] Top shape: 100 64 4 4 (102400) | |
I0724 15:21:13.907620 31785 layer_factory.hpp:74] Creating layer ip1 | |
I0724 15:21:13.907629 31785 net.cpp:84] Creating Layer ip1 | |
I0724 15:21:13.907631 31785 net.cpp:380] ip1 <- pool3 | |
I0724 15:21:13.907637 31785 net.cpp:338] ip1 -> ip1 | |
I0724 15:21:13.907644 31785 net.cpp:113] Setting up ip1 | |
I0724 15:21:13.909636 31785 net.cpp:120] Top shape: 100 64 (6400) | |
I0724 15:21:13.909647 31785 layer_factory.hpp:74] Creating layer ip2 | |
I0724 15:21:13.909656 31785 net.cpp:84] Creating Layer ip2 | |
I0724 15:21:13.909659 31785 net.cpp:380] ip2 <- ip1 | |
I0724 15:21:13.909665 31785 net.cpp:338] ip2 -> ip2 | |
I0724 15:21:13.909672 31785 net.cpp:113] Setting up ip2 | |
I0724 15:21:13.909701 31785 net.cpp:120] Top shape: 100 10 (1000) | |
I0724 15:21:13.909711 31785 layer_factory.hpp:74] Creating layer ip2_ip2_0_split | |
I0724 15:21:13.909718 31785 net.cpp:84] Creating Layer ip2_ip2_0_split | |
I0724 15:21:13.909723 31785 net.cpp:380] ip2_ip2_0_split <- ip2 | |
I0724 15:21:13.909726 31785 net.cpp:338] ip2_ip2_0_split -> ip2_ip2_0_split_0 | |
I0724 15:21:13.909732 31785 net.cpp:338] ip2_ip2_0_split -> ip2_ip2_0_split_1 | |
I0724 15:21:13.909739 31785 net.cpp:113] Setting up ip2_ip2_0_split | |
I0724 15:21:13.909745 31785 net.cpp:120] Top shape: 100 10 (1000) | |
I0724 15:21:13.909749 31785 net.cpp:120] Top shape: 100 10 (1000) | |
I0724 15:21:13.909754 31785 layer_factory.hpp:74] Creating layer accuracy | |
I0724 15:21:13.909762 31785 net.cpp:84] Creating Layer accuracy | |
I0724 15:21:13.909766 31785 net.cpp:380] accuracy <- ip2_ip2_0_split_0 | |
I0724 15:21:13.909770 31785 net.cpp:380] accuracy <- label_cifar_1_split_0 | |
I0724 15:21:13.909775 31785 net.cpp:338] accuracy -> accuracy | |
I0724 15:21:13.909781 31785 net.cpp:113] Setting up accuracy | |
I0724 15:21:13.909790 31785 net.cpp:120] Top shape: (1) | |
I0724 15:21:13.909792 31785 layer_factory.hpp:74] Creating layer loss | |
I0724 15:21:13.909798 31785 net.cpp:84] Creating Layer loss | |
I0724 15:21:13.909801 31785 net.cpp:380] loss <- ip2_ip2_0_split_1 | |
I0724 15:21:13.909806 31785 net.cpp:380] loss <- label_cifar_1_split_1 | |
I0724 15:21:13.909812 31785 net.cpp:338] loss -> loss | |
I0724 15:21:13.909817 31785 net.cpp:113] Setting up loss | |
I0724 15:21:13.909822 31785 layer_factory.hpp:74] Creating layer loss | |
I0724 15:21:13.909888 31785 net.cpp:120] Top shape: (1) | |
I0724 15:21:13.909894 31785 net.cpp:122] with loss weight 1 | |
I0724 15:21:13.909905 31785 net.cpp:167] loss needs backward computation. | |
I0724 15:21:13.909920 31785 net.cpp:169] accuracy does not need backward computation. | |
I0724 15:21:13.909924 31785 net.cpp:167] ip2_ip2_0_split needs backward computation. | |
I0724 15:21:13.909929 31785 net.cpp:167] ip2 needs backward computation. | |
I0724 15:21:13.909931 31785 net.cpp:167] ip1 needs backward computation. | |
I0724 15:21:13.909934 31785 net.cpp:167] pool3 needs backward computation. | |
I0724 15:21:13.909937 31785 net.cpp:167] relu3 needs backward computation. | |
I0724 15:21:13.909941 31785 net.cpp:167] conv3 needs backward computation. | |
I0724 15:21:13.909945 31785 net.cpp:167] pool2 needs backward computation. | |
I0724 15:21:13.909947 31785 net.cpp:167] relu2 needs backward computation. | |
I0724 15:21:13.909951 31785 net.cpp:167] conv2 needs backward computation. | |
I0724 15:21:13.909955 31785 net.cpp:167] relu1 needs backward computation. | |
I0724 15:21:13.909957 31785 net.cpp:167] pool1 needs backward computation. | |
I0724 15:21:13.909961 31785 net.cpp:167] conv1 needs backward computation. | |
I0724 15:21:13.909965 31785 net.cpp:169] label_cifar_1_split does not need backward computation. | |
I0724 15:21:13.909981 31785 net.cpp:169] cifar does not need backward computation. | |
I0724 15:21:13.909988 31785 net.cpp:205] This network produces output accuracy | |
I0724 15:21:13.909996 31785 net.cpp:205] This network produces output loss | |
I0724 15:21:13.910007 31785 net.cpp:447] Collecting Learning Rate and Weight Decay. | |
I0724 15:21:13.910013 31785 net.cpp:217] Network initialization done. | |
I0724 15:21:13.910017 31785 net.cpp:218] Memory required for data: 31987608 | |
I0724 15:21:13.910058 31785 solver.cpp:42] Solver scaffolding done. | |
I0724 15:21:13.910080 31785 caffe.cpp:126] Resuming from examples/cifar10/cifar10_quick_iter_4000.solverstate | |
I0724 15:21:13.910086 31785 solver.cpp:222] Solving CIFAR10_quick | |
I0724 15:21:13.910090 31785 solver.cpp:223] Learning Rate Policy: fixed | |
I0724 15:21:13.910094 31785 solver.cpp:226] Restoring previous solver status from examples/cifar10/cifar10_quick_iter_4000.solverstate | |
I0724 15:21:13.911710 31785 solver.cpp:564] SGDSolver: restoring history | |
I0724 15:21:13.911932 31785 solver.cpp:266] Iteration 4000, Testing net (#0) | |
I0724 15:21:14.293946 31785 solver.cpp:315] Test net output #0: accuracy = 0.7148 | |
I0724 15:21:14.294061 31785 solver.cpp:315] Test net output #1: loss = 0.843019 (* 1 = 0.843019 loss) | |
I0724 15:21:14.300997 31785 solver.cpp:189] Iteration 4000, loss = 0.604699 | |
I0724 15:21:14.301045 31785 solver.cpp:204] Train net output #0: loss = 0.604699 (* 1 = 0.604699 loss) | |
I0724 15:21:14.301059 31785 solver.cpp:464] Iteration 4000, lr = 0.0001 | |
I0724 15:21:15.444993 31785 solver.cpp:189] Iteration 4100, loss = 0.560168 | |
I0724 15:21:15.445029 31785 solver.cpp:204] Train net output #0: loss = 0.560168 (* 1 = 0.560168 loss) | |
I0724 15:21:15.445036 31785 solver.cpp:464] Iteration 4100, lr = 0.0001 | |
I0724 15:21:16.581609 31785 solver.cpp:189] Iteration 4200, loss = 0.583019 | |
I0724 15:21:16.581660 31785 solver.cpp:204] Train net output #0: loss = 0.583019 (* 1 = 0.583019 loss) | |
I0724 15:21:16.581668 31785 solver.cpp:464] Iteration 4200, lr = 0.0001 | |
I0724 15:21:17.721596 31785 solver.cpp:189] Iteration 4300, loss = 0.458659 | |
I0724 15:21:17.721722 31785 solver.cpp:204] Train net output #0: loss = 0.458659 (* 1 = 0.458659 loss) | |
I0724 15:21:17.721736 31785 solver.cpp:464] Iteration 4300, lr = 0.0001 | |
I0724 15:21:18.860602 31785 solver.cpp:189] Iteration 4400, loss = 0.462953 | |
I0724 15:21:18.860636 31785 solver.cpp:204] Train net output #0: loss = 0.462953 (* 1 = 0.462953 loss) | |
I0724 15:21:18.860642 31785 solver.cpp:464] Iteration 4400, lr = 0.0001 | |
I0724 15:21:19.993888 31785 solver.cpp:266] Iteration 4500, Testing net (#0) | |
I0724 15:21:20.385629 31785 solver.cpp:315] Test net output #0: accuracy = 0.7502 | |
I0724 15:21:20.385663 31785 solver.cpp:315] Test net output #1: loss = 0.748229 (* 1 = 0.748229 loss) | |
I0724 15:21:20.389665 31785 solver.cpp:189] Iteration 4500, loss = 0.497331 | |
I0724 15:21:20.389683 31785 solver.cpp:204] Train net output #0: loss = 0.497331 (* 1 = 0.497331 loss) | |
I0724 15:21:20.389716 31785 solver.cpp:464] Iteration 4500, lr = 0.0001 | |
I0724 15:21:21.528735 31785 solver.cpp:189] Iteration 4600, loss = 0.476239 | |
I0724 15:21:21.528861 31785 solver.cpp:204] Train net output #0: loss = 0.476239 (* 1 = 0.476239 loss) | |
I0724 15:21:21.528877 31785 solver.cpp:464] Iteration 4600, lr = 0.0001 | |
I0724 15:21:22.665988 31785 solver.cpp:189] Iteration 4700, loss = 0.567064 | |
I0724 15:21:22.666031 31785 solver.cpp:204] Train net output #0: loss = 0.567064 (* 1 = 0.567064 loss) | |
I0724 15:21:22.666038 31785 solver.cpp:464] Iteration 4700, lr = 0.0001 | |
I0724 15:21:23.802747 31785 solver.cpp:189] Iteration 4800, loss = 0.439045 | |
I0724 15:21:23.802781 31785 solver.cpp:204] Train net output #0: loss = 0.439045 (* 1 = 0.439045 loss) | |
I0724 15:21:23.802788 31785 solver.cpp:464] Iteration 4800, lr = 0.0001 | |
I0724 15:21:24.940335 31785 solver.cpp:189] Iteration 4900, loss = 0.445887 | |
I0724 15:21:24.940371 31785 solver.cpp:204] Train net output #0: loss = 0.445887 (* 1 = 0.445887 loss) | |
I0724 15:21:24.940377 31785 solver.cpp:464] Iteration 4900, lr = 0.0001 | |
I0724 15:21:26.072255 31785 solver.cpp:334] Snapshotting to examples/cifar10/cifar10_quick_iter_5000.caffemodel | |
I0724 15:21:26.075196 31785 solver.cpp:342] Snapshotting solver state to examples/cifar10/cifar10_quick_iter_5000.solverstate | |
I0724 15:21:26.081387 31785 solver.cpp:248] Iteration 5000, loss = 0.478067 | |
I0724 15:21:26.081419 31785 solver.cpp:266] Iteration 5000, Testing net (#0) | |
I0724 15:21:26.460409 31785 solver.cpp:315] Test net output #0: accuracy = 0.7519 | |
I0724 15:21:26.460441 31785 solver.cpp:315] Test net output #1: loss = 0.743144 (* 1 = 0.743144 loss) | |
I0724 15:21:26.460448 31785 solver.cpp:253] Optimization Done. | |
I0724 15:21:26.460451 31785 caffe.cpp:134] Optimization Done. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment