All test examples converge to one embedding point with siamese network model...

[YeoChan Yoon]

I'm tring to learn siamese network model for cifar10 data.

I built paired data using cifar10 data and trained with revised cifar10 model definition (original: https://github.com/ycyoon/caffe/blob/master/examples/cifar10/cifar10_full_train_test.prototxt) for siamese network.

I just appended feat layers for 2D embedding and changed weight filler from gausian to xavier.

Here's the revised model definition

name: "cifar10_siamese_train_test"
layer {
  name: "pair_data"
  type: "Data"
  top: "pair_data"
  top: "sim"
  include {
    phase: TRAIN
  }
  transform_param {
    scale: 0.00390625
  }
  data_param {
    source: "examples/siamese/mnist_siamese_train_leveldb/"
    batch_size: 64
  }
}
layer {
  name: "pair_data"
  type: "Data"
  top: "pair_data"
  top: "sim"
  include {
    phase: TEST
  }
  transform_param {
    scale: 0.00390625
  }
  data_param {
    source: "examples/siamese/mnist_siamese_test_leveldb/"
    batch_size: 100
  }
}
layer {
  name: "slice_pair"
  type: "Slice"
  bottom: "pair_data"
  top: "data"
  top: "data_p"
  slice_param {
    slice_dim: 1
    slice_point: 1
  }
}
layer {
  name: "conv1"
  type: "Convolution"
  bottom: "data"
  top: "conv1"
  param {
    lr_mult: 1
  }
  param {
    lr_mult: 2
  }
  convolution_param {
    num_output: 20
    pad: 0
    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: 3
    stride: 2
  }
}
layer {
  name: "relu1"
  type: "ReLU"
  bottom: "pool1"
  top: "pool1"
}
layer {
  name: "norm1"
  type: "LRN"
  bottom: "pool1"
  top: "norm1"
  lrn_param {
    local_size: 3
    alpha: 5e-05
    beta: 0.75
    norm_region: WITHIN_CHANNEL
  }
}
layer {
  name: "conv2"
  type: "Convolution"
  bottom: "norm1"
  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: "xavier"
    }
    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: "norm2"
  type: "LRN"
  bottom: "pool2"
  top: "norm2"
  lrn_param {
    local_size: 3
    alpha: 5e-05
    beta: 0.75
    norm_region: WITHIN_CHANNEL
  }
}
layer {
  name: "conv3"
  type: "Convolution"
  bottom: "norm2"
  top: "conv3"
  convolution_param {
    num_output: 64
    pad: 2
    kernel_size: 5
    stride: 1
    weight_filler {
      type: "xavier"
    }
    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
    decay_mult: 250
  }
  param {
    lr_mult: 2
    decay_mult: 0
  }
  inner_product_param {
    num_output: 10
    weight_filler {
      type: "xavier"
    }
    bias_filler {
      type: "constant"
    }
  }
}
layer {
  name: "feat"
  type: "InnerProduct"
  bottom: "ip1"
  top: "feat"
  param {
    lr_mult: 1
  }
  param {
    lr_mult: 2
  }
  inner_product_param {
    num_output: 2
    weight_filler {
      type: "xavier"
    }
    bias_filler {
      type: "constant"
    }
  }
}
layer {
  name: "conv1_p"
  type: "Convolution"
  bottom: "data_p"
  top: "conv1_p"
  param {
    lr_mult: 1
  }
  param {
    lr_mult: 2
  }
  convolution_param {
    num_output: 20
    pad: 0
    kernel_size: 5
    stride: 1
    weight_filler {
      type: "xavier"
    }
    bias_filler {
      type: "constant"
    }
  }
}
layer {
  name: "pool1_p"
  type: "Pooling"
  bottom: "conv1_p"
  top: "pool1_p"
  pooling_param {
    pool: MAX
    kernel_size: 3
    stride: 2
  }
}
layer {
  name: "relu1_p"
  type: "ReLU"
  bottom: "pool1_p"
  top: "pool1_p"
}
layer {
  name: "norm1_p"
  type: "LRN"
  bottom: "pool1_p"
  top: "norm1_p"
  lrn_param {
    local_size: 3
    alpha: 5e-05
    beta: 0.75
    norm_region: WITHIN_CHANNEL
  }
}
layer {
  name: "conv2_p"
  type: "Convolution"
  bottom: "norm1_p"
  top: "conv2_p"
  param {
    lr_mult: 1
  }
  param {
    lr_mult: 2
  }
  convolution_param {
    num_output: 32
    pad: 2
    kernel_size: 5
    stride: 1
    weight_filler {
      type: "xavier"
    }
    bias_filler {
      type: "constant"
    }
  }
}
layer {
  name: "relu2_p"
  type: "ReLU"
  bottom: "conv2_p"
  top: "conv2_p"
}
layer {
  name: "pool2_p"
  type: "Pooling"
  bottom: "conv2_p"
  top: "pool2_p"
  pooling_param {
    pool: AVE
    kernel_size: 3
    stride: 2
  }
}
layer {
  name: "norm2_p"
  type: "LRN"
  bottom: "pool2_p"
  top: "norm2_p"
  lrn_param {
    local_size: 3
    alpha: 5e-05
    beta: 0.75
    norm_region: WITHIN_CHANNEL
  }
}
layer {
  name: "conv3_p"
  type: "Convolution"
  bottom: "norm2_p"
  top: "conv3_p"
  convolution_param {
    num_output: 64
    pad: 2
    kernel_size: 5
    stride: 1
    weight_filler {
      type: "xavier"
    }
    bias_filler {
      type: "constant"
    }
  }
}
layer {
  name: "relu3_p"
  type: "ReLU"
  bottom: "conv3_p"
  top: "conv3_p"
}
layer {
  name: "pool3_p"
  type: "Pooling"
  bottom: "conv3_p"
  top: "pool3_p"
  pooling_param {
    pool: AVE
    kernel_size: 3
    stride: 2
  }
}
layer {
  name: "ip1_p"
  type: "InnerProduct"
  bottom: "pool3_p"
  top: "ip1_p"
  param {
    lr_mult: 1
    decay_mult: 250
  }
  param {
    lr_mult: 2
    decay_mult: 0
  }
  inner_product_param {
    num_output: 10
    weight_filler {
      type: "xavier"
    }
    bias_filler {
      type: "constant"
    }
  }
}
layer {
  name: "feat_p"
  type: "InnerProduct"
  bottom: "ip1_p"
  top: "feat_p"
  param {
    lr_mult: 1
  }
  param {
    lr_mult: 2
  }
  inner_product_param {
    num_output: 2
    weight_filler {
      type: "xavier"
    }
    bias_filler {
      type: "constant"
    }
  }
}
layer {
  name: "loss"
  type: "ContrastiveLoss"
  bottom: "feat"
  bottom: "feat_p"
  bottom: "sim"
  top: "loss"
  contrastive_loss_param {
    margin: 1
  }
}

However, when I tried to get embeding points from test examples, all outputs are converged to one same point. (for example, 0.38288754 -0.08905841)

I changed the cifar10 data to mnist data, however it also gives the same result.

Is there anything wrong with the definition file?

[Subhabrata Debnath]

I am experiencing the same problem. Were you able to solve it ?