LeHedge v1

lehedge_solver.prototxt

# The train/test net protocol buffer definition
net: "lehedge_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: 282
# Carry out testing every 500 training iterations.
test_interval: 564
# The base learning rate, momentum and the weight decay of the network.
base_lr: 0.0001
momentum: 0.9
weight_decay: 0.00001
# The learning rate policy
lr_policy: "inv"
gamma: 0.0001
power: 0.75
# Display every 100 iterations
display: 5
# The maximum number of iterations
max_iter: 25000
# snapshot intermediate results
snapshot: 282
snapshot_prefix: "snp"
# solver mode: CPU or GPU
solver_mode: GPU

lehedge_train_test.prototxt

name: "LeHedge"
layers {
  name: "mydata"
  type: DATA
  top: "data"
  top: "label"
  data_param {
    source: "lehedge_train_leveldb"
    batch_size: 64
  }
  include: { phase: TRAIN }
}
layers {
  name: "mydata"
  type: DATA
  top: "data"
  top: "label"
  data_param {
    source: "lehedge_val_leveldb"
    batch_size: 64
  }
  include: { phase: TEST }
}

layers {
  name: "conv1"
  type: CONVOLUTION
  bottom: "data"
  top: "conv1"
  convolution_param {
    num_output: 104 
    kernel_h: 51
    kernel_w: 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_h: 19
    kernel_w: 27
    stride: 1
  }
}
layers {
  name: "conv2"
  type: CONVOLUTION
  bottom: "pool1"
  top: "conv2"
  convolution_param {
    num_output: 104
    kernel_h: 3
    kernel_w: 65
    stride: 1
    weight_filler {
      type: "xavier"
    }
    bias_filler {
      type: "constant"
    }
  }
}
layers {
  name: "pool2"
  type: POOLING
  bottom: "conv2"
  top: "pool2"
  pooling_param {
    pool: MAX
    kernel_h: 18
    kernel_w: 6
    stride: 1
  }
}
layers {
  name: "ip1"
  type: INNER_PRODUCT
  bottom: "pool2"
  top: "ip1"
  inner_product_param {
    num_output: 19
    weight_filler {
      type: "xavier"
    }
    bias_filler {
      type: "constant"
    }
  }
}
layers {
  name: "accuracy"
  type: ACCURACY
  bottom: "ip1"
  bottom: "label"
  top: "accuracy"
  include: { phase: TEST }
}
layers {
  name: "loss"
  type: SOFTMAX_LOSS
  bottom: "ip1"
  bottom: "label"
  top: "loss"
}