20161114——LeNet-5预测数据-第一次跑

最新推荐文章于 2024-06-13 15:15:05 发布
最新推荐文章于 2024-06-13 15:15:05 发布
阅读量485 收藏
点赞数 1
分类专栏： 我的机器学习之路 文章标签： 人工智能 机器学习
版权声明：本文为博主原创文章，遵循 CC 4.0 BY-SA 版权协议，转载请附上原文出处链接和本声明。
本文链接：https://blog.csdn.net/luwei6896/article/details/53967904
版权
root@ubuntu:/home/caffe/caffe# ./build/tools/caffe.bin test \
> -model examples/mnist/lenet_train_test.prototxt \> -weights examples/mnist/lenet_iter_10000.caffemodel \
> -iterations 100I1113 22:23:18.152611  9264 caffe.cpp:279] Use CPU.
I1113 22:23:18.173339  9264 net.cpp:322] The NetState phase (1) differed from the phase (0) specified by a rule in layer mnist
I1113 22:23:18.175729  9264 net.cpp:58] Initializing net from parameters:
name: "LeNet"
state {
  phase: TEST
  level: 0
  stage: ""
}
layer {
  name: "mnist"
  type: "Data"
  top: "data"
  top: "label"
  include {
    phase: TEST
  }
  transform_param {
    scale: 0.00390625
  }
  data_param {
    source: "examples/mnist/mnist_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: 20
    kernel_size: 5
    stride: 1
    weight_filler {
      type: "xavier"
    }
    bias_filler {
      type: "constant"
    }
  }
}
layer {
  name: "pool1"
  type: "Pooling"
  bottom: "conv1"
  top: "pool1"
  pooling_param {
    pool: MAX
    kernel_size: 2
    stride: 2
  }
}
layer {
  name: "conv2"
  type: "Convolution"
  bottom: "pool1"
  top: "conv2"
  param {
    lr_mult: 1
  }
  param {
    lr_mult: 2
  }
  convolution_param {
    num_output: 50
    kernel_size: 5
    stride: 1
    weight_filler {
      type: "xavier"
    }
    bias_filler {
      type: "constant"
    }
  }
}
layer {
  name: "pool2"
  type: "Pooling"
  bottom: "conv2"
  top: "pool2"
  pooling_param {
    pool: MAX
    kernel_size: 2
    stride: 2
  }
}
layer {
  name: "ip1"
  type: "InnerProduct"
  bottom: "pool2"
  top: "ip1"
  param {
    lr_mult: 1
  }
  param {
    lr_mult: 2
  }
  inner_product_param {
    num_output: 500
    weight_filler {
      type: "xavier"
    }
    bias_filler {
      type: "constant"
    }
  }
}
layer {
  name: "relu1"
  type: "ReLU"
  bottom: "ip1"
  top: "ip1"
}
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: "xavier"
    }
    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"
}
I1113 22:23:18.176378  9264 layer_factory.hpp:77] Creating layer mnist
I1113 22:23:18.178973  9264 net.cpp:100] Creating Layer mnist
I1113 22:23:18.179025  9264 net.cpp:408] mnist -> data
I1113 22:23:18.179074  9264 net.cpp:408] mnist -> label
I1113 22:23:18.179217  9267 db_lmdb.cpp:35] Opened lmdb examples/mnist/mnist_test_lmdb
I1113 22:23:18.180079  9264 data_layer.cpp:41] output data size: 100,1,28,28
I1113 22:23:18.180670  9264 net.cpp:150] Setting up mnist
I1113 22:23:18.180714  9264 net.cpp:157] Top shape: 100 1 28 28 (78400)
I1113 22:23:18.180728  9264 net.cpp:157] Top shape: 100 (100)
I1113 22:23:18.180737  9264 net.cpp:165] Memory required for data: 314000
I1113 22:23:18.180752  9264 layer_factory.hpp:77] Creating layer label_mnist_1_split
I1113 22:23:18.180765  9264 net.cpp:100] Creating Layer label_mnist_1_split
I1113 22:23:18.180774  9264 net.cpp:434] label_mnist_1_split <- label
I1113 22:23:18.180790  9264 net.cpp:408] label_mnist_1_split -> label_mnist_1_split_0
I1113 22:23:18.180807  9264 net.cpp:408] label_mnist_1_split -> label_mnist_1_split_1
I1113 22:23:18.181561  9264 net.cpp:150] Setting up label_mnist_1_split
I1113 22:23:18.181596  9264 net.cpp:157] Top shape: 100 (100)
I1113 22:23:18.181610  9264 net.cpp:157] Top shape: 100 (100)
I1113 22:23:18.181619  9264 net.cpp:165] Memory required for data: 314800
I1113 22:23:18.181630  9264 layer_factory.hpp:77] Creating layer conv1
I1113 22:23:18.181660  9264 net.cpp:100] Creating Layer conv1
I1113 22:23:18.181671  9264 net.cpp:434] conv1 <- data
I1113 22:23:18.181685  9264 net.cpp:408] conv1 -> conv1
I1113 22:23:18.182523  9264 net.cpp:150] Setting up conv1
I1113 22:23:18.182587  9264 net.cpp:157] Top shape: 100 20 24 24 (1152000)
I1113 22:23:18.182601  9264 net.cpp:165] Memory required for data: 4922800
I1113 22:23:18.183148  9264 layer_factory.hpp:77] Creating layer pool1
I1113 22:23:18.183276  9264 net.cpp:100] Creating Layer pool1
I1113 22:23:18.183292  9264 net.cpp:434] pool1 <- conv1
I1113 22:23:18.183305  9264 net.cpp:408] pool1 -> pool1
I1113 22:23:18.184011  9264 net.cpp:150] Setting up pool1
I1113 22:23:18.184041  9264 net.cpp:157] Top shape: 100 20 12 12 (288000)
I1113 22:23:18.184052  9264 net.cpp:165] Memory required for data: 6074800
I1113 22:23:18.184065  9264 layer_factory.hpp:77] Creating layer conv2
I1113 22:23:18.184089  9264 net.cpp:100] Creating Layer conv2
I1113 22:23:18.184118  9264 net.cpp:434] conv2 <- pool1
I1113 22:23:18.184135  9264 net.cpp:408] conv2 -> conv2
I1113 22:23:18.184834  9264 net.cpp:150] Setting up conv2
I1113 22:23:18.185140  9264 net.cpp:157] Top shape: 100 50 8 8 (320000)
I1113 22:23:18.185185  9264 net.cpp:165] Memory required for data: 7354800
I1113 22:23:18.185240  9264 layer_factory.hpp:77] Creating layer pool2
I1113 22:23:18.185333  9264 net.cpp:100] Creating Layer pool2
I1113 22:23:18.185431  9264 net.cpp:434] pool2 <- conv2
I1113 22:23:18.185504  9264 net.cpp:408] pool2 -> pool2
I1113 22:23:18.185549  9264 net.cpp:150] Setting up pool2
I1113 22:23:18.185569  9264 net.cpp:157] Top shape: 100 50 4 4 (80000)
I1113 22:23:18.185578  9264 net.cpp:165] Memory required for data: 7674800
I1113 22:23:18.185588  9264 layer_factory.hpp:77] Creating layer ip1
I1113 22:23:18.186364  9264 net.cpp:100] Creating Layer ip1
I1113 22:23:18.186419  9264 net.cpp:434] ip1 <- pool2
I1113 22:23:18.186444  9264 net.cpp:408] ip1 -> ip1
I1113 22:23:18.195292  9264 net.cpp:150] Setting up ip1
I1113 22:23:18.195338  9264 net.cpp:157] Top shape: 100 500 (50000)
I1113 22:23:18.195349  9264 net.cpp:165] Memory required for data: 7874800
I1113 22:23:18.195384  9264 layer_factory.hpp:77] Creating layer relu1
I1113 22:23:18.195401  9264 net.cpp:100] Creating Layer relu1
I1113 22:23:18.195411  9264 net.cpp:434] relu1 <- ip1
I1113 22:23:18.195423  9264 net.cpp:395] relu1 -> ip1 (in-place)
I1113 22:23:18.196239  9264 net.cpp:150] Setting up relu1
I1113 22:23:18.196269  9264 net.cpp:157] Top shape: 100 500 (50000)
I1113 22:23:18.196280  9264 net.cpp:165] Memory required for data: 8074800
I1113 22:23:18.196292  9264 layer_factory.hpp:77] Creating layer ip2
I1113 22:23:18.196313  9264 net.cpp:100] Creating Layer ip2
I1113 22:23:18.196326  9264 net.cpp:434] ip2 <- ip1
I1113 22:23:18.196346  9264 net.cpp:408] ip2 -> ip2
I1113 22:23:18.196473  9264 net.cpp:150] Setting up ip2
I1113 22:23:18.196491  9264 net.cpp:157] Top shape: 100 10 (1000)
I1113 22:23:18.196501  9264 net.cpp:165] Memory required for data: 8078800
I1113 22:23:18.196516  9264 layer_factory.hpp:77] Creating layer ip2_ip2_0_split
I1113 22:23:18.196532  9264 net.cpp:100] Creating Layer ip2_ip2_0_split
I1113 22:23:18.196542  9264 net.cpp:434] ip2_ip2_0_split <- ip2
I1113 22:23:18.196554  9264 net.cpp:408] ip2_ip2_0_split -> ip2_ip2_0_split_0
I1113 22:23:18.196568  9264 net.cpp:408] ip2_ip2_0_split -> ip2_ip2_0_split_1
I1113 22:23:18.196585  9264 net.cpp:150] Setting up ip2_ip2_0_split
I1113 22:23:18.196599  9264 net.cpp:157] Top shape: 100 10 (1000)
I1113 22:23:18.196610  9264 net.cpp:157] Top shape: 100 10 (1000)
I1113 22:23:18.196620  9264 net.cpp:165] Memory required for data: 8086800
I1113 22:23:18.196630  9264 layer_factory.hpp:77] Creating layer accuracy
I1113 22:23:18.197474  9264 net.cpp:100] Creating Layer accuracy
I1113 22:23:18.197502  9264 net.cpp:434] accuracy <- ip2_ip2_0_split_0
I1113 22:23:18.197517  9264 net.cpp:434] accuracy <- label_mnist_1_split_0
I1113 22:23:18.197532  9264 net.cpp:408] accuracy -> accuracy
I1113 22:23:18.197580  9264 net.cpp:150] Setting up accuracy
I1113 22:23:18.197602  9264 net.cpp:157] Top shape: (1)
I1113 22:23:18.197614  9264 net.cpp:165] Memory required for data: 8086804
I1113 22:23:18.197626  9264 layer_factory.hpp:77] Creating layer loss
I1113 22:23:18.198261  9264 net.cpp:100] Creating Layer loss
I1113 22:23:18.198290  9264 net.cpp:434] loss <- ip2_ip2_0_split_1
I1113 22:23:18.198304  9264 net.cpp:434] loss <- label_mnist_1_split_1
I1113 22:23:18.198345  9264 net.cpp:408] loss -> loss
I1113 22:23:18.199340  9264 layer_factory.hpp:77] Creating layer loss
I1113 22:23:18.199407  9264 net.cpp:150] Setting up loss
I1113 22:23:18.199426  9264 net.cpp:157] Top shape: (1)
I1113 22:23:18.199436  9264 net.cpp:160]     with loss weight 1
I1113 22:23:18.200038  9264 net.cpp:165] Memory required for data: 8086808
I1113 22:23:18.200050  9264 net.cpp:226] loss needs backward computation.
I1113 22:23:18.200058  9264 net.cpp:228] accuracy does not need backward computation.
I1113 22:23:18.200067  9264 net.cpp:226] ip2_ip2_0_split needs backward computation.
I1113 22:23:18.200073  9264 net.cpp:226] ip2 needs backward computation.
I1113 22:23:18.200079  9264 net.cpp:226] relu1 needs backward computation.
I1113 22:23:18.200086  9264 net.cpp:226] ip1 needs backward computation.
I1113 22:23:18.200093  9264 net.cpp:226] pool2 needs backward computation.
I1113 22:23:18.200099  9264 net.cpp:226] conv2 needs backward computation.
I1113 22:23:18.200105  9264 net.cpp:226] pool1 needs backward computation.
I1113 22:23:18.200112  9264 net.cpp:226] conv1 needs backward computation.
I1113 22:23:18.200119  9264 net.cpp:228] label_mnist_1_split does not need backward computation.
I1113 22:23:18.200126  9264 net.cpp:228] mnist does not need backward computation.
I1113 22:23:18.200135  9264 net.cpp:270] This network produces output accuracy
I1113 22:23:18.200147  9264 net.cpp:270] This network produces output loss
I1113 22:23:18.200181  9264 net.cpp:283] Network initialization done.
I1113 22:23:18.213127  9264 caffe.cpp:285] Running for 100 iterations.
I1113 22:23:18.716567  9264 caffe.cpp:308] Batch 0, accuracy = 1
I1113 22:23:18.716647  9264 caffe.cpp:308] Batch 0, loss = 0.00938862
I1113 22:23:19.142791  9264 caffe.cpp:308] Batch 1, accuracy = 1
I1113 22:23:19.142884  9264 caffe.cpp:308] Batch 1, loss = 0.00550929
I1113 22:23:19.606091  9264 caffe.cpp:308] Batch 2, accuracy = 0.99
I1113 22:23:19.606170  9264 caffe.cpp:308] Batch 2, loss = 0.0508872
I1113 22:23:20.054841  9264 caffe.cpp:308] Batch 3, accuracy = 0.99
I1113 22:23:20.054941  9264 caffe.cpp:308] Batch 3, loss = 0.0376309
I1113 22:23:20.485735  9264 caffe.cpp:308] Batch 4, accuracy = 0.99
I1113 22:23:20.485807  9264 caffe.cpp:308] Batch 4, loss = 0.077192
I1113 22:23:20.962743  9264 caffe.cpp:308] Batch 5, accuracy = 0.99
I1113 22:23:20.962816  9264 caffe.cpp:308] Batch 5, loss = 0.0515025
I1113 22:23:21.388486  9264 caffe.cpp:308] Batch 6, accuracy = 0.98
I1113 22:23:21.388574  9264 caffe.cpp:308] Batch 6, loss = 0.0931183
I1113 22:23:21.815683  9264 caffe.cpp:308] Batch 7, accuracy = 0.99
I1113 22:23:21.815806  9264 caffe.cpp:308] Batch 7, loss = 0.041062
I1113 22:23:22.237282  9264 caffe.cpp:308] Batch 8, accuracy = 1
I1113 22:23:22.237368  9264 caffe.cpp:308] Batch 8, loss = 0.00694906
I1113 22:23:22.686130  9264 caffe.cpp:308] Batch 9, accuracy = 0.99
I1113 22:23:22.686210  9264 caffe.cpp:308] Batch 9, loss = 0.0272126
I1113 22:23:23.110630  9264 caffe.cpp:308] Batch 10, accuracy = 0.98
I1113 22:23:23.110712  9264 caffe.cpp:308] Batch 10, loss = 0.0507764
I1113 22:23:23.545310  9264 caffe.cpp:308] Batch 11, accuracy = 0.98
I1113 22:23:23.545372  9264 caffe.cpp:308] Batch 11, loss = 0.0363619
I1113 22:23:23.970454  9264 caffe.cpp:308] Batch 12, accuracy = 0.95
I1113 22:23:23.970535  9264 caffe.cpp:308] Batch 12, loss = 0.139559
I1113 22:23:24.413075  9264 caffe.cpp:308] Batch 13, accuracy = 0.98
I1113 22:23:24.413163  9264 caffe.cpp:308] Batch 13, loss = 0.0643891
I1113 22:23:24.836765  9264 caffe.cpp:308] Batch 14, accuracy = 0.99
I1113 22:23:24.836839  9264 caffe.cpp:308] Batch 14, loss = 0.0208035
I1113 22:23:25.255944  9264 caffe.cpp:308] Batch 15, accuracy = 0.99
I1113 22:23:25.256047  9264 caffe.cpp:308] Batch 15, loss = 0.0302129
I1113 22:23:25.695009  9264 caffe.cpp:308] Batch 16, accuracy = 0.98
I1113 22:23:25.695098  9264 caffe.cpp:308] Batch 16, loss = 0.0369838
I1113 22:23:26.124233  9264 caffe.cpp:308] Batch 17, accuracy = 1
I1113 22:23:26.124357  9264 caffe.cpp:308] Batch 17, loss = 0.0165045
I1113 22:23:26.555377  9264 caffe.cpp:308] Batch 18, accuracy = 1
I1113 22:23:26.555450  9264 caffe.cpp:308] Batch 18, loss = 0.0110177
I1113 22:23:26.963515  9264 caffe.cpp:308] Batch 19, accuracy = 0.99
I1113 22:23:26.963590  9264 caffe.cpp:308] Batch 19, loss = 0.0634679
I1113 22:23:27.390830  9264 caffe.cpp:308] Batch 20, accuracy = 0.98
I1113 22:23:27.390894  9264 caffe.cpp:308] Batch 20, loss = 0.08095
I1113 22:23:27.818614  9264 caffe.cpp:308] Batch 21, accuracy = 0.97
I1113 22:23:27.818688  9264 caffe.cpp:308] Batch 21, loss = 0.0524366
I1113 22:23:28.243798  9264 caffe.cpp:308] Batch 22, accuracy = 0.99
I1113 22:23:28.243883  9264 caffe.cpp:308] Batch 22, loss = 0.041327
I1113 22:23:28.676961  9264 caffe.cpp:308] Batch 23, accuracy = 0.99
I1113 22:23:28.677062  9264 caffe.cpp:308] Batch 23, loss = 0.0261502
I1113 22:23:29.112872  9264 caffe.cpp:308] Batch 24, accuracy = 0.98
I1113 22:23:29.112973  9264 caffe.cpp:308] Batch 24, loss = 0.0512742
I1113 22:23:29.541849  9264 caffe.cpp:308] Batch 25, accuracy = 0.99
I1113 22:23:29.541923  9264 caffe.cpp:308] Batch 25, loss = 0.0888764
I1113 22:23:29.961591  9264 caffe.cpp:308] Batch 26, accuracy = 0.99
I1113 22:23:29.961663  9264 caffe.cpp:308] Batch 26, loss = 0.108548
I1113 22:23:30.394340  9264 caffe.cpp:308] Batch 27, accuracy = 1
I1113 22:23:30.394412  9264 caffe.cpp:308] Batch 27, loss = 0.0103665
I1113 22:23:30.818648  9264 caffe.cpp:308] Batch 28, accuracy = 0.99
I1113 22:23:30.818737  9264 caffe.cpp:308] Batch 28, loss = 0.0305791
I1113 22:23:31.259657  9264 caffe.cpp:308] Batch 29, accuracy = 0.97
I1113 22:23:31.259748  9264 caffe.cpp:308] Batch 29, loss = 0.108499
I1113 22:23:31.703888  9264 caffe.cpp:308] Batch 30, accuracy = 1
I1113 22:23:31.703981  9264 caffe.cpp:308] Batch 30, loss = 0.0138441
I1113 22:23:32.149801  9264 caffe.cpp:308] Batch 31, accuracy = 1
I1113 22:23:32.149922  9264 caffe.cpp:308] Batch 31, loss = 0.00274972
I1113 22:23:32.585639  9264 caffe.cpp:308] Batch 32, accuracy = 0.99
I1113 22:23:32.585721  9264 caffe.cpp:308] Batch 32, loss = 0.014872
I1113 22:23:33.010319  9264 caffe.cpp:308] Batch 33, accuracy = 1
I1113 22:23:33.010391  9264 caffe.cpp:308] Batch 33, loss = 0.00557933
I1113 22:23:33.434504  9264 caffe.cpp:308] Batch 34, accuracy = 0.99
I1113 22:23:33.434592  9264 caffe.cpp:308] Batch 34, loss = 0.0435118
I1113 22:23:33.862126  9264 caffe.cpp:308] Batch 35, accuracy = 0.95
I1113 22:23:33.862200  9264 caffe.cpp:308] Batch 35, loss = 0.141068
I1113 22:23:34.296036  9264 caffe.cpp:308] Batch 36, accuracy = 1
I1113 22:23:34.296131  9264 caffe.cpp:308] Batch 36, loss = 0.00427433
I1113 22:23:34.745066  9264 caffe.cpp:308] Batch 37, accuracy = 0.98
I1113 22:23:34.745139  9264 caffe.cpp:308] Batch 37, loss = 0.0834443
I1113 22:23:35.169420  9264 caffe.cpp:308] Batch 38, accuracy = 0.99
I1113 22:23:35.169481  9264 caffe.cpp:308] Batch 38, loss = 0.0193586
I1113 22:23:35.584350  9264 caffe.cpp:308] Batch 39, accuracy = 0.98
I1113 22:23:35.584434  9264 caffe.cpp:308] Batch 39, loss = 0.03763
I1113 22:23:36.018128  9264 caffe.cpp:308] Batch 40, accuracy = 0.99
I1113 22:23:36.018245  9264 caffe.cpp:308] Batch 40, loss = 0.0272643
I1113 22:23:36.435228  9264 caffe.cpp:308] Batch 41, accuracy = 0.97
I1113 22:23:36.435304  9264 caffe.cpp:308] Batch 41, loss = 0.0821054
I1113 22:23:36.854449  9264 caffe.cpp:308] Batch 42, accuracy = 0.99
I1113 22:23:36.854549  9264 caffe.cpp:308] Batch 42, loss = 0.0260138
I1113 22:23:37.285959  9264 caffe.cpp:308] Batch 43, accuracy = 1
I1113 22:23:37.286029  9264 caffe.cpp:308] Batch 43, loss = 0.0122153
I1113 22:23:37.721631  9264 caffe.cpp:308] Batch 44, accuracy = 0.99
I1113 22:23:37.721696  9264 caffe.cpp:308] Batch 44, loss = 0.0179641
I1113 22:23:38.153909  9264 caffe.cpp:308] Batch 45, accuracy = 0.99
I1113 22:23:38.153973  9264 caffe.cpp:308] Batch 45, loss = 0.0411753
I1113 22:23:38.592052  9264 caffe.cpp:308] Batch 46, accuracy = 1
I1113 22:23:38.592134  9264 caffe.cpp:308] Batch 46, loss = 0.00871828
I1113 22:23:39.017051  9264 caffe.cpp:308] Batch 47, accuracy = 0.99
I1113 22:23:39.017124  9264 caffe.cpp:308] Batch 47, loss = 0.019142
I1113 22:23:39.444370  9264 caffe.cpp:308] Batch 48, accuracy = 0.97
I1113 22:23:39.444437  9264 caffe.cpp:308] Batch 48, loss = 0.113587
I1113 22:23:39.861088  9264 caffe.cpp:308] Batch 49, accuracy = 1
I1113 22:23:39.861165  9264 caffe.cpp:308] Batch 49, loss = 0.00563929
I1113 22:23:40.282407  9264 caffe.cpp:308] Batch 50, accuracy = 1
I1113 22:23:40.282487  9264 caffe.cpp:308] Batch 50, loss = 0.000420446
I1113 22:23:40.722947  9264 caffe.cpp:308] Batch 51, accuracy = 0.99
I1113 22:23:40.723053  9264 caffe.cpp:308] Batch 51, loss = 0.00879608
I1113 22:23:41.158830  9264 caffe.cpp:308] Batch 52, accuracy = 0.99
I1113 22:23:41.158905  9264 caffe.cpp:308] Batch 52, loss = 0.0259591
I1113 22:23:41.591497  9264 caffe.cpp:308] Batch 53, accuracy = 1
I1113 22:23:41.591567  9264 caffe.cpp:308] Batch 53, loss = 0.00159477
I1113 22:23:42.023552  9264 caffe.cpp:308] Batch 54, accuracy = 1
I1113 22:23:42.023633  9264 caffe.cpp:308] Batch 54, loss = 0.00324901
I1113 22:23:42.444705  9264 caffe.cpp:308] Batch 55, accuracy = 1
I1113 22:23:42.444790  9264 caffe.cpp:308] Batch 55, loss = 0.000148405
I1113 22:23:42.876534  9264 caffe.cpp:308] Batch 56, accuracy = 1
I1113 22:23:42.876618  9264 caffe.cpp:308] Batch 56, loss = 0.0105164
I1113 22:23:43.315635  9264 caffe.cpp:308] Batch 57, accuracy = 1
I1113 22:23:43.315773  9264 caffe.cpp:308] Batch 57, loss = 0.00503844
I1113 22:23:43.763896  9264 caffe.cpp:308] Batch 58, accuracy = 0.99
I1113 22:23:43.763972  9264 caffe.cpp:308] Batch 58, loss = 0.0108849
I1113 22:23:44.192829  9264 caffe.cpp:308] Batch 59, accuracy = 0.98
I1113 22:23:44.192906  9264 caffe.cpp:308] Batch 59, loss = 0.0836057
I1113 22:23:44.622547  9264 caffe.cpp:308] Batch 60, accuracy = 1
I1113 22:23:44.622627  9264 caffe.cpp:308] Batch 60, loss = 0.00554896
I1113 22:23:45.061022  9264 caffe.cpp:308] Batch 61, accuracy = 1
I1113 22:23:45.061094  9264 caffe.cpp:308] Batch 61, loss = 0.00197286
I1113 22:23:45.490329  9264 caffe.cpp:308] Batch 62, accuracy = 1
I1113 22:23:45.490404  9264 caffe.cpp:308] Batch 62, loss = 4.45838e-05
I1113 22:23:45.924926  9264 caffe.cpp:308] Batch 63, accuracy = 1
I1113 22:23:45.925246  9264 caffe.cpp:308] Batch 63, loss = 8.86665e-05
I1113 22:23:46.351568  9264 caffe.cpp:308] Batch 64, accuracy = 1
I1113 22:23:46.351652  9264 caffe.cpp:308] Batch 64, loss = 0.00062603
I1113 22:23:46.793261  9264 caffe.cpp:308] Batch 65, accuracy = 0.94
I1113 22:23:46.793334  9264 caffe.cpp:308] Batch 65, loss = 0.113657
I1113 22:23:47.218242  9264 caffe.cpp:308] Batch 66, accuracy = 0.98
I1113 22:23:47.218310  9264 caffe.cpp:308] Batch 66, loss = 0.0446646
I1113 22:23:47.645225  9264 caffe.cpp:308] Batch 67, accuracy = 0.99
I1113 22:23:47.645298  9264 caffe.cpp:308] Batch 67, loss = 0.0358956
I1113 22:23:48.068356  9264 caffe.cpp:308] Batch 68, accuracy = 0.99
I1113 22:23:48.068570  9264 caffe.cpp:308] Batch 68, loss = 0.0121253
I1113 22:23:48.509443  9264 caffe.cpp:308] Batch 69, accuracy = 1
I1113 22:23:48.509609  9264 caffe.cpp:308] Batch 69, loss = 0.00187437
I1113 22:23:48.938766  9264 caffe.cpp:308] Batch 70, accuracy = 1
I1113 22:23:48.938841  9264 caffe.cpp:308] Batch 70, loss = 0.000640704
I1113 22:23:49.364161  9264 caffe.cpp:308] Batch 71, accuracy = 1
I1113 22:23:49.364253  9264 caffe.cpp:308] Batch 71, loss = 0.000420613
I1113 22:23:49.821877  9264 caffe.cpp:308] Batch 72, accuracy = 1
I1113 22:23:49.821961  9264 caffe.cpp:308] Batch 72, loss = 0.00922661
I1113 22:23:50.275329  9264 caffe.cpp:308] Batch 73, accuracy = 1
I1113 22:23:50.275413  9264 caffe.cpp:308] Batch 73, loss = 0.000654965
I1113 22:23:50.710718  9264 caffe.cpp:308] Batch 74, accuracy = 1
I1113 22:23:50.710805  9264 caffe.cpp:308] Batch 74, loss = 0.00321759
I1113 22:23:51.151079  9264 caffe.cpp:308] Batch 75, accuracy = 1
I1113 22:23:51.151149  9264 caffe.cpp:308] Batch 75, loss = 0.00241048
I1113 22:23:51.577095  9264 caffe.cpp:308] Batch 76, accuracy = 1
I1113 22:23:51.577168  9264 caffe.cpp:308] Batch 76, loss = 0.000133293
I1113 22:23:52.017668  9264 caffe.cpp:308] Batch 77, accuracy = 1
I1113 22:23:52.017741  9264 caffe.cpp:308] Batch 77, loss = 0.000138202
I1113 22:23:52.448783  9264 caffe.cpp:308] Batch 78, accuracy = 1
I1113 22:23:52.448870  9264 caffe.cpp:308] Batch 78, loss = 0.00147909
I1113 22:23:52.893913  9264 caffe.cpp:308] Batch 79, accuracy = 1
I1113 22:23:52.893987  9264 caffe.cpp:308] Batch 79, loss = 0.00162909
I1113 22:23:53.331879  9264 caffe.cpp:308] Batch 80, accuracy = 0.99
I1113 22:23:53.332003  9264 caffe.cpp:308] Batch 80, loss = 0.0239026
I1113 22:23:53.772898  9264 caffe.cpp:308] Batch 81, accuracy = 1
I1113 22:23:53.772971  9264 caffe.cpp:308] Batch 81, loss = 0.00128124
I1113 22:23:54.223368  9264 caffe.cpp:308] Batch 82, accuracy = 1
I1113 22:23:54.223448  9264 caffe.cpp:308] Batch 82, loss = 0.0012403
I1113 22:23:54.652359  9264 caffe.cpp:308] Batch 83, accuracy = 1
I1113 22:23:54.652432  9264 caffe.cpp:308] Batch 83, loss = 0.0126122
I1113 22:23:55.076644  9264 caffe.cpp:308] Batch 84, accuracy = 0.99
I1113 22:23:55.076748  9264 caffe.cpp:308] Batch 84, loss = 0.0249088
I1113 22:23:55.525315  9264 caffe.cpp:308] Batch 85, accuracy = 0.99
I1113 22:23:55.525436  9264 caffe.cpp:308] Batch 85, loss = 0.0145371
I1113 22:23:55.963310  9264 caffe.cpp:308] Batch 86, accuracy = 1
I1113 22:23:55.963379  9264 caffe.cpp:308] Batch 86, loss = 8.19215e-05
I1113 22:23:56.396705  9264 caffe.cpp:308] Batch 87, accuracy = 1
I1113 22:23:56.396802  9264 caffe.cpp:308] Batch 87, loss = 0.000117623
I1113 22:23:56.826771  9264 caffe.cpp:308] Batch 88, accuracy = 1
I1113 22:23:56.826851  9264 caffe.cpp:308] Batch 88, loss = 5.30659e-05
I1113 22:23:57.260174  9264 caffe.cpp:308] Batch 89, accuracy = 1
I1113 22:23:57.260283  9264 caffe.cpp:308] Batch 89, loss = 4.73997e-05
I1113 22:23:57.692572  9264 caffe.cpp:308] Batch 90, accuracy = 0.97
I1113 22:23:57.692658  9264 caffe.cpp:308] Batch 90, loss = 0.082495
I1113 22:23:58.121317  9264 caffe.cpp:308] Batch 91, accuracy = 1
I1113 22:23:58.121409  9264 caffe.cpp:308] Batch 91, loss = 3.68753e-05
I1113 22:23:58.558925  9264 caffe.cpp:308] Batch 92, accuracy = 1
I1113 22:23:58.559033  9264 caffe.cpp:308] Batch 92, loss = 9.49426e-05
I1113 22:23:58.984719  9264 caffe.cpp:308] Batch 93, accuracy = 1
I1113 22:23:58.984807  9264 caffe.cpp:308] Batch 93, loss = 0.00213211
I1113 22:23:59.403184  9264 caffe.cpp:308] Batch 94, accuracy = 1
I1113 22:23:59.403254  9264 caffe.cpp:308] Batch 94, loss = 0.000319923
I1113 22:23:59.828168  9264 caffe.cpp:308] Batch 95, accuracy = 1
I1113 22:23:59.828251  9264 caffe.cpp:308] Batch 95, loss = 0.00170309
I1113 22:24:00.258958  9264 caffe.cpp:308] Batch 96, accuracy = 0.98
I1113 22:24:00.259034  9264 caffe.cpp:308] Batch 96, loss = 0.0682994
I1113 22:24:00.676637  9264 caffe.cpp:308] Batch 97, accuracy = 0.97
I1113 22:24:00.676724  9264 caffe.cpp:308] Batch 97, loss = 0.0928666
I1113 22:24:01.116845  9264 caffe.cpp:308] Batch 98, accuracy = 1
I1113 22:24:01.116920  9264 caffe.cpp:308] Batch 98, loss = 0.00344261
I1113 22:24:01.559696  9264 caffe.cpp:308] Batch 99, accuracy = 0.99
I1113 22:24:01.559839  9264 caffe.cpp:308] Batch 99, loss = 0.0130951
I1113 22:24:01.559859  9264 caffe.cpp:313] Loss: 0.0291962
I1113 22:24:01.559878  9264 caffe.cpp:325] accuracy = 0.9911
I1113 22:24:01.559896  9264 caffe.cpp:325] loss = 0.0291962 (* 1 = 0.0291962 loss)
root@ubuntu:/home/caffe/caffe# 
Wendell易
关注
  • 1
    点赞
  • 踩
  • 0
    收藏
    觉得还不错? 一键收藏
  • 0
    评论
专栏目录
学习笔记04·预测测试数据集、LeNet…
rj496631097的博客
04-01 1822
（1）用训练好的LeNet-5模型对数据进行预测，在git中敲入命令./Build/x64/Release/caffe.exe test -model examples/mnist/lenet_train_test.prototxt -weights examples/mnist/lenet_iter_10000.caffemodel -iterations 100 。 （这里是我们第一次使用c
手写体数字识别例程——LeNet-5模型
2Know的博客
10-17 6800
上一篇博客中介绍了Caffe环境的搭建，本片博客中介绍一下，在caffe中训练的第一个CNN模型LeNet-5。如果存在不正确的地方欢迎指正。
深度学习手记（五）之LeNet-5预测模型
星影L的博客
04-18 1764
直接进入主题，大名鼎鼎的LeNet诞生于1994年，是最早的深层卷积神经网络之一，并且推动了深度学习的发展。LeNet模型是由Yann LeCun所完成的，他一直认为直接将每个像素作为多层神经网络的输入是不能利用图像本身所含有的空间相关性，效率大打折扣。 LeNet5当时的特征有如下几点： 每一个卷积层包含三个部分：卷积、降采样和非线性激活函数 使用卷积提取空间特征 降采样的最大池化层 双曲...
CV预测：快速使用LeNet-5卷积神经网络
最新发布
AI相关从业分享
06-13 539
LeNet网络逐层结构:图片输入:32x32x1第一层:卷积核(6个，5x5x1，步长:1)，输出:28x28x6最大池化层:卷积核(2x2，步长:2)，输出:14x14x6第二层:卷积核(16个，5x5x6，步长:1)，输出:10x10x16最大池化层:卷积核(2x2，步长:2)，输出:5x5x16拉直后输出:5 x5x16=400全连接层1:120个节点全连接层2:84个节点输出层:10个节点注:在进行卷积运算时，没有对输入进行填充。
matlab精度检验代码-Lenet_5_scratch_tensorflow:该代码包含从零开始以及在tensorflow中对lenet5的
05-22
matlab精度检验代码LeNet-5从零开始并使用tensorflow 该存储库包含Matlab从头开始使用Lenet-5的代码，也使用tensorflow。 这是我的课程CS698U中的一项作业。 作业的问题陈述 实施和培训LeNet-5 CNN以进行MNIST数字分类任务 从零开始，例如仅使用Numpy，Matlab。 使用深度学习库； Caffe，火炬，Tensorflow，Keras之一 相互比较上面两个实现的结果。 在报告中： 比较转换层与功能层所花费的时间 比较conv层和fc层中的参数数量 可视化从每个数字类的随机选择的测试示例中提取的特征，并显示t-SNE图。 绘制训练和验证错误率与迭代次数的关系图 探索不同批次大小（16、32、64、128）对训练的影响 迭代是一个小批量，一个时期是整个训练数据的传递 项目详情 matlab：-该文件夹包含来自matlab中strach的Le Net 5的实现。 README.pdf包含此实现的许多详细信息。 同样要运行此命令，您必须安装matconvnet，因为该卷积使用convnn.m进行卷积。此文件中包含所有参数和其他驱动程序
【CNTK】CNTK学习笔记之图像识别-树叶识别Train and Test
努力奋斗
08-26 8974
CNTK，OpenCV，卷积神经网络的图像识别实战。
TensorFlow学习笔记——LeNet-5（训练自己的数据集）
Python博客
12-07 1996
　　在之前的TensorFlow学习笔记——图像识别与卷积神经网络（链接：请点击我）中了解了一下经典的卷积神经网络模型LeNet模型。那其实之前学习了别人的代码实现了LeNet网络对MNIST数据集的训练。而这篇文章是想自己完成LeNet网络来训练自己的数据集。LeNet主要用来进行手写字符的识别与分类，下面记录一下自己学习的过程。 　　我的学习步骤分为以下四步： 1，温习LeNet-5的网络...
Lenet-5卷积神经网络结构详解（一）——原理解析
qq_42805721的博客
11-07 1万+
Lenet-5卷积神经网络结构详解（一）——原理解释 前言 本系列博文将阐释深度学习网络Lenet-5的具体原理，网络结构及其代码实现。本文不适用任何高级框架，只使用numpy从底层手撸lenet-5。本系列博文共计3篇，本文是第一篇。系列中代码参考地址： link. link Lenet-5 神经网络出自论文 Gradient-Based Learning Applied to Docume...
LeNet-5源代码（详注）、mnist_data及论文
Jeff zhao
05-07 3127
LeNet-5是论文Gradient-Based Learning Applied to Document Recognition（IEEE-1998）提出的一种用于手写体字符识别的卷积神经网络模型。 源代码 '''利用tensorflow实现的卷积神经网络来进行MNIST手写数字图像的分类。''' import numpy as np import tensorflow as tf ...
LeNet-5模型之MNIST手写数字数据集训练-验证-测试三部曲（tesorflow）
weixin_44403952的博客
12-01 3674
LeNet5简介 LeNet-5模型是Yann Lecun教授在1998年提出的，成功应用于数字识别问题的卷积神经神经网络。在MNIST数据集上，达到了99.27%的正确率。 图像输入：28×28像素 CONV1——卷积模板5×5×32，步长为1，全0填充，输出28×28×32图像 Pool1——池化层模板是2×2，步长为2，全0填充，输出14×14×32图像 CONV2——卷积模板5×5×6...
卷积神经网络的网络结构——以LeNet-5为例
strint的专栏
03-09 4万+
卷积神经网络是一种特殊的多层神经网络，像其它的神经网络一样，卷积神经网络也使用一种反向传播算法来进行训练，不同之处在于网络的结构。卷积神经网络的网络连接具有局部连接、参数共享的特点。局部连接是相对于普通神经网络的全连接而言的，是指这一层的某个节点只与上一层的部分节点相连。参数共享是指一层中多个节点的连接共享相同的一组参数。
复现resnet和lenet进行训练、验证和预测，含模型和数据集，有详细注释。
why的博客
12-13 534
该项目是本人之前参与华为某项目商用的小部分代码，公开出来供大家入门学习CNN使用。github地址：https://github.com/HuiyanWen/resnet_tensorflow
内涵：caffe学习之Lenet-5详解
算法工程师进阶之路
05-31 1469
from pylab import * %matplotlib inline caffe_root = '../' # this file should be run from {caffe_root}/examples (otherwise change this line) import sys sys.path.insert(0, caffe_root + 'python') impo
Caffe中Mnist的训练日志解读
xiaoyezi_1834的博客
02-23 3612
Lenet的网络结构图如下： Mnist的训练日志如下（其中max_iter设为3，使用CPU模式） I0223 14:55:42.870805  6406 caffe.cpp:178] Use CPU.//使用CPU模式 I0223 14:55:42.871371  6406 solver.cpp:48] Initializing solver from param
LeNet5的深入解析
AI小作坊 的博客
12-09 3万+
论文：Gradient-based-learning-applied-to-document-recognition参考：http://blog.csdn.net/strint/article/details/44163869LeNet5 这个网络虽然很小，但是它包含了深度学习的基本模块：卷积层，池化层，全链接层。是其他深度学习模型的基础， 这里我们对LeNet5进行深入分析，并由此看看 Caf
深度学习 --- 卷积神经网络CNN（LeNet-5网络详解）
热门推荐
进击的菜鸟
11-26 23万+
卷积神经网络（Convolutional Neural Network，CNN）是一种前馈型的神经网络，其在大型图像处理方面有出色的表现，目前已经被大范围使用到图像分类、定位等领域中。相比于其他神经网络结构，卷积神经网络需要的参数相对较少，使的其能够广泛应用。 本节打算先介绍背景和简单的基本概念术语，然后详细介绍LeNet-5网络，其工作流程大家需要先搞清楚，学习原理其实是基于BP学习原理的，只...
浅谈LeNet-5
江小举的博客
04-07 2万+
浅谈LeNet-5 一、LetNet是什么？ LetNet是一种入门级的神经网络模型，是一个简单的卷积神经网络，可以用来做手写体识别。 下面我将以上图为例简单介绍一下LetNet神经网络的工作流程。 输入层 读入需要被网络处理的图片，这里图片的大小为32*32的黑白图片 Conv1层 图片大小:32x32，通道数为1 filter(卷积核)：5x5x6(大小5x5,个数6) 步长：1 输...
"深度学习经典网络——LeNet-5详解与推荐
深度学习经典网络推荐.pdf中提到了一种经典的用于识别数字的卷积网络LeNet-5。LeNet-5共有7层，不包括输入层，每层都包含可训练参数。每个层都有多个特征图（Feature Map），每个Feature Map通过卷积滤波器提取输入...
评论
添加红包

请填写红包祝福语或标题

个

红包个数最小为10个

元

红包金额最低5元

当前余额3.43元 前往充值 >
需支付：10.00元
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付元
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明：

1.余额是钱包充值的虚拟货币，按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载，可以购买VIP、付费专栏及课程。

余额充值