szm-R

name: "DarkNet"
input: "data"
input_shape {
  dim: 1
  dim: 3
  dim: 224
  dim: 224
}

#####################################################################

szm-R

name: "Inception_v4"
input: "data"
input_shape {
  dim: 1
  dim: 3
  dim: 299
  dim: 299
}
layer {
  name: "conv1_3x3_s2"

szm-R

name: "Inception_v3"
input: "data"
input_shape {
  dim: 1
  dim: 3
  dim: 395 
  dim: 395
}
layer {
  name: "conv1_3x3_s2"

szm-R

name: resnet269_v2
input: "data"
input_shape {
  dim: 1
  dim: 3
  dim: 320
  dim: 320
}
layer {
  name: "conv1"

szm-R

name: SqueezeNet
input: "data"
input_shape {
  dim: 1
  dim: 3
  dim: 224 
  dim: 224
}
layer {
  name: "conv1"

szm-R

name: "DNetSqueezeNet"
layer {
  name: "train_data"
  type: "Data"
  top: "data"
  data_param {
    backend: LMDB
    source: "examples/kitti/kitti_train_images.lmdb"
    batch_size: 8
  }