caffe手写数字mnist训练测试使用教程

一、确保caffe环境已经搭建好，由于没有N卡GPU，只能使用CPU，反正有的是时间。

二、准备相关的配置文件：

lenet_solver.prototxt ——

# The train/test net protocol buffer definition
net: "D:/MyWorks/caffe-windows-master/3rdparty/bin/lenet_train_test.prototxt"
# test_iter specifies how many forward passes the test should carry out.
# In the case of MNIST, we have test batch size 100 and 100 test iterations,
# covering the full 10,000 testing images.
test_iter: 100
# Carry out testing every 500 training iterations.
test_interval: 500
# The base learning rate, momentum and the weight decay of the network.
base_lr: 0.01
momentum: 0.9
weight_decay: 0.0005
# The learning rate policy
lr_policy: "inv"
gamma: 0.0001
power: 0.75
# Display every 100 iterations
display: 100
# The maximum number of iterations
max_iter: 20000
# snapshot intermediate results
snapshot: 5000
snapshot_prefix: "lenet"
# solver mode: 0 for CPU and 1 for GPU
solver_mode: 0

lenet_train_test.prototxt ——

name: "LeNet"
layers {
  name: "mnist"
  type: DATA
  top: "data"
  top: "label"
  data_param {
    source: "D:/MyWorks/caffe-windows-master/3rdparty/bin/mnist-train-leveldb"
    backend: LEVELDB
    batch_size: 128
  }
  transform_param {
    scale: 0.00390625
  }
  include: { phase: TRAIN }
}
layers {
  name: "mnist"
  type: DATA
  top: "data"
  top: "label"
  data_param {
    source: "D:/MyWorks/caffe-windows-master/3rdparty/bin/mnist-test-leveldb"
    backend: LEVELDB
    batch_size: 100
  }
  transform_param {
    scale: 0.00390625
  }
  include: { phase: TEST }
}


layers {
  name: "conv1"
  type: CONVOLUTION
  bottom: "data"
  top: "conv1"
  blobs_lr: 1
  blobs_lr: 2
  convolution_param {
    num_output: 20
    kernel_size: 5
    stride: 1
    weight_filler {
      type: "xavier"
    }
    bias_filler {
      type: "constant"
    }
  }
}
layers {
  name: "pool1"
  type: POOLING
  bottom: "conv1"
  top: "pool1"
  pooling_param {
    pool: MAX
    kernel_size: 2
    stride: 2
  }
}
layers {
  name: "conv2"
  type: CONVOLUTION
  bottom: "pool1"
  top: "conv2"
  blobs_lr: 1
  blobs_lr: 2
  convolution_param {
    num_output: 50
    kernel_size: 5
    stride: 1
    weight_filler {
      type: "xavier"
    }
    bias_filler {
      type: "constant"
    }
  }
}
layers {
  name: "pool2"
  type: POOLING
  bottom: "conv2"
  top: "pool2"
  pooling_param {
    pool: MAX
    kernel_size: 2
    stride: 2
  }
}
layers {
  name: "ip1"
  type: INNER_PRODUCT
  bottom: "pool2"
  top: "ip1"
  blobs_lr: 1
  blobs_lr: 2
  inner_product_param {
    num_output: 500
    weight_filler {
      type: "xavier"
    }
    bias_filler {
      type: "constant"
    }
  }
}
layers {
  name: "relu1"
  type: RELU
  bottom: "ip1"
  top: "ip1"
}
layers {
  name: "ip2"
  type: INNER_PRODUCT
  bottom: "ip1"
  top: "ip2"
  blobs_lr: 1
  blobs_lr: 2
  inner_product_param {
    num_output: 10
    weight_filler {
      type: "xavier"
    }
    bias_filler {
      type: "constant"
    }
  }
}
layers {
  name: "accuracy"
  type: ACCURACY
  bottom: "ip2"
  bottom: "label"
  top: "accuracy"
  include: { phase: TEST }
}
layers {
  name: "loss"
  type: SOFTMAX_LOSS
  bottom: "ip2"
  bottom: "label"
  top: "loss"
}

lenet_test.prototxt ——

name: "LeNet-test"
layers {
  layer {
    name: "mnist"
    type: "data"
    source: "D:/MyWorks/caffe-windows-master/3rdparty/bin/mnist-test-leveldb"
    batchsize: 100
    scale: 0.00390625
  }
  top: "data"
  top: "label"
}
layers {
  layer {
    name: "conv1"
    type: "conv"
    num_output: 20
    kernelsize: 5
    stride: 1
    weight_filler {
      type: "xavier"
    }
    bias_filler {
      type: "constant"
    }
  }
  bottom: "data"
  top: "conv1"
}
layers {
  layer {
    name: "pool1"
    type: "pool"
    kernelsize: 2
    stride: 2
    pool: MAX
  }
  bottom: "conv1"
  top: "pool1"
}
layers {
  layer {
    name: "conv2"
    type: "conv"
    num_output: 50
    kernelsize: 5
    stride: 1
    weight_filler {
      type: "xavier"
    }
    bias_filler {
      type: "constant"
    }
  }
  bottom: "pool1"
  top: "conv2"
}
layers {
  layer {
    name: "pool2"
    type: "pool"
    kernelsize: 2
    stride: 2
    pool: MAX
  }
  bottom: "conv2"
  top: "pool2"
}
layers {
  layer {
    name: "ip1"
    type: "innerproduct"
    num_output: 500
    weight_filler {
      type: "xavier"
    }
    bias_filler {
      type: "constant"
    }
  }
  bottom: "pool2"
  top: "ip1"
}
layers {
  layer {
    name: "relu1"
    type: "relu"
  }
  bottom: "ip1"
  top: "ip1"
}
layers {
  layer {
    name: "ip2"
    type: "innerproduct"
    num_output: 10
    weight_filler {
      type: "xavier"
    }
    bias_filler {
      type: "constant"
    }
  }
  bottom: "ip1"
  top: "ip2"
}
layers {
  layer {
    name: "prob"
    type: "softmax"
  }
  bottom: "ip2"
  top: "prob"
}
layers {
  layer {
    name: "accuracy"
    type: "accuracy"
  }
  bottom: "prob"
  bottom: "label"
  top: "accuracy"
}

三、开始训练，去干点别的事情吧！

四、回来发现已经训练好了，测试！